2013 Annual Meeting Abstract Supplement

AN OFFICIAL JOURNAL OF THE AMERICAN COLLEGE OF RHEUMATOLOGY

Arthritis & Rheumatism VOLUME 65, NUMBER 10 (SUPPLEMENT)

ABSTRACT SUPPLEMENT 2013 ANNUAL MEETING October 25–30, 2013 San Diego, CA

OCTOBER 2013

Get On-Demand Annual Meeting Education with SessionSelect!

ATTENDEE BONUS!

zŽƵƌĂŶŶƵ Ă ƌĞŐŝƐƚƌĂƟŽ ůŵĞĞƟŶŐ ŶĨĞĞŐŝǀĞƐ ǇŽƵ &ZĂĐĐĞƐ ƐƚŽƚŚĞŽŶ ůŝŶĞ ĐŽŶƚĞŶƚŽĨ ƚŚĞϮϬϭϯĂ ŶŶƵĂů ŵĞĞƟŶŐĨŽ ƌŽŶĞǇĞĂƌ͊

SessionSelect is a comprehensive digital library of recorded presentations from the annual meeting. Within 24 hours of virtually every live session, you can view fully-synchronized presentations with speaker slides and audio—just as if you were actually in the room. Take advantage of the most convenient way to experience the education from the annual meeting at your own pace. With SessionSelect you can: t4UBZVQUPEBUFPOUIFNPTUSFMFWBOUJOGPSNBUJPOBOESFTFBSDIQFSUBJOJOHUPSIFVNBUJDEJTFBTFT t"QQMZTFTTJPOJOGPSNBUJPOUPZPVSQSBDUJDFUPFOIBODFZPVSLOPXMFEHF DPNQFUFODF and performance t7JFXDPOUFOUPOZPVSNPCJMFEFWJDFPSEPXOMPBE.1öMFTGPSMJTUFOJOHUPBVEJPPOUIFHP

SessionSelect Lounge

/P/FFEUP.JTTB4FTTJPO8IJMF:PVSF0OTJUF

Got some down time between sessions? Relax while you replay sessions—at your convenience—on dedicated computer terminals. Location: Hall D Lobby Hours: .POEBZ 0DUPCFS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7:00 AM – 6:00 PM 5VFTEBZ 0DUPCFS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7:00 AM – 6:00 PM 8FEOFTEBZ 0DUPCFS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7:00 AM – 1:00 PM

To learn more about SessionSelect or to view a demo, come by the Discovery Center, located in the Exhibit Hall F-G (Booth PSUIF$PODJFSHF$FOUFS MPDBUFEJOUIF4BJMT1BWJMJPO

ZĞĐŽƌĚŝŶŐƐŽĨŝŶĚŝǀŝĚƵĂůƐĞƐƐŝŽŶƐĂƌĞƐƵďũĞĐƚƚŽĐŚĂŶŐĞ͘dŝĐŬĞƚĞĚƐĞƐƐŝŽŶƐĂŶĚƉƌĞͲŵĞĞƟŶŐĐŽƵƌƐĞƐĂƌĞŶŽƚŝŶĐůƵĚĞĚŝŶĐŽŵƉůŝŵĞŶƚĂƌǇĂĐĐĞƐƐ͘

AMERICANCOLLEGEOFRHEUMATOLOGY ABSTRACTSUPPLEMENT

AMERICANCOLLEGEOFRHEUMATOLOGY 77thAnnualMeeting ASSOCIATIONOFRHEUMATOLOGYHEALTHPROFESSIONALS 48thAnnualMeeting October25Ͳ30,2013 SanDiego,CA Copyright©2013bytheAmericanCollegeofRheumatology,Atlanta,GA ThesupplementwasnotfinancedbyprofitͲmakingorganizationsorbyorganizationsrepresentingforͲprofitinterests.Theeditorialandpeer reviewprocesseswerehandledentirelybytheAmericanCollegeofRheumatology(ACR)accordingtoitspeerreviewprocessforabstracts submittedforpresentationattheACRAnnualMeeting.

ACR/ARHP2013AnnualMeetingOverall NeedsAssessment/PracticeGaps

dŚĞŵĞƌŝĐĂŶŽůůĞŐĞŽĨZŚĞƵŵĂƚŽůŽŐǇĂŶĚƚŚĞƐƐŽĐŝĂƚŝŽŶŽĨ ZŚĞƵŵĂƚŽůŽŐǇ,ĞĂůƚŚWƌŽĨĞƐƐŝŽŶĂůƐĂƌĞĐŽŵŵŝƚƚĞĚƚŽƉƌŽǀŝĚŝŶŐ ĐŽŵƉƌĞŚĞŶƐŝǀĞĞĚƵĐĂƚŝŽŶƚŽŝŵƉƌŽǀĞƚŚĞŬŶŽǁůĞĚŐĞĂŶĚ ƉĞƌĨŽƌŵĂŶĐĞŽĨƉŚǇƐŝĐŝĂŶƐ͕ŚĞĂůƚŚƉƌŽĨĞƐƐŝŽŶĂůƐĂŶĚƐĐŝĞŶƚŝƐƚƐ͘ dŚƌŽƵŐŚĞǀŝĚĞŶĐĞͲďĂƐĞĚĞĚƵĐĂƚŝŽŶĂůƉƌŽŐƌĂŵƐ͕ƚŚĞŽƌŐĂŶŝǌĂƚŝŽŶ ƐƚƌŝǀĞƐƚŽĞŶŚĂŶĐĞƉƌĂĐƚŝĐĞƉĞƌĨŽƌŵĂŶĐĞĂŶĚŝŵƉƌŽǀĞƚŚĞƋƵĂůŝƚǇŽĨ ĐĂƌĞŝŶƚŚŽƐĞǁŝƚŚŽƌĂƚƌŝƐŬĨŽƌĂƌƚŚƌŝƚŝƐ͕ƌŚĞƵŵĂƚŝĐĂŶĚ ŵƵƐĐƵůŽƐŬĞůĞƚĂůĚŝƐĞĂƐĞƐ͘dŚĞϮϬϭϯĂŶŶƵĂůŵĞĞƚŝŶŐƉƌŽŐƌĂŵŚĂƐ ďĞĞŶĚĞǀĞůŽƉĞĚŝŶĚĞƉĞŶĚĞŶƚŽĨĐŽŵŵĞƌĐŝĂůŝŶĨůƵĞŶĐĞ͘

dŚĞƉƌŽŐƌĂŵŝƐƚŚĞƌĞƐƵůƚŽĨĂƉůĂŶŶŝŶŐƉƌŽĐĞƐƐƚŚĂƚŝĚĞŶƚŝĨŝĞĚ ĞĚƵĐĂƚŝŽŶĂůŶĞĞĚƐƚŽĐŚĂŶŐĞŽƌĞŶŚĂŶĐĞƚŚĞŬŶŽǁůĞĚŐĞ͕ ĐŽŵƉĞƚĞŶĐĞŽƌƉĞƌĨŽƌŵĂŶĐĞŽĨƌŚĞƵŵĂƚŽůŽŐǇƉƌŽĨĞƐƐŝŽŶĂůƐ͘dŚĞ ƉƌŽŐƌĂŵ͛ƐĐŽŶƚĞŶƚǁĂƐĚĞƌŝǀĞĚĨƌŽŵďŽƚŚŶĞĞĚƐĂƐƐĞƐƐŵĞŶƚĂŶĚ ƉƌĂĐƚŝĐĞŐĂƉĂŶĂůǇƐŝƐďĂƐĞĚŽŶƉƌŽĨĞƐƐŝŽŶĂůĂĐƚŝǀŝƚŝĞƐ͕ƉƌĂĐƚŝĐĞ ƐĞƚƚŝŶŐ͕/DƌĞĐĞƌƚŝĨŝĐĂƚŝŽŶƌĞƋƵŝƌĞŵĞŶƚƐĂŶĚƉŚǇƐŝĐŝĂŶĂƚƚƌŝďƵƚĞƐ͘

ProgramHighlights x

x

x

x

x

x

x

x

x

ĚƵĐĂƚŝŽŶĂůƚƌĂĐŬƐƚŽŚĞůƉĂƚƚĞŶĚĞĞƐŝĚĞŶƚŝĨǇĐŽŶƚĞŶƚ ƚĂƌŐĞƚĞĚƚŽƚŚĞŵ͘dƌĂĐŬƐŝŶĐůƵĚĞ͗ďĂƐŝĐƐĐŝĞŶĐĞ͕ ďƵƐŝŶĞƐƐͬĂĚŵŝŶŝƐƚƌĂƚŝŽŶ͕ĐůŝŶŝĐĂůƉƌĂĐƚŝĐĞĂŶĚĐůŝŶŝĐĂů ƐĐŝĞŶĐĞ͘ >ĂƚĞƐƚƐĐŝĞŶĐĞĂŶĚďĞƐƚͲƉƌĂĐƚŝĐĞƐƉƌĞƐĞŶƚĞĚƚŚƌŽƵŐŚƉĞĞƌͲ ƌĞǀŝĞǁĞĚĂŶĚƐĞůĞĐƚĞĚĐůŝŶŝĐĂůĂŶĚƐĐŝĞŶƚŝĨŝĐĂďƐƚƌĂĐƚƐ͕ĂŶĚ ŝŶǀŝƚĞĚƐƉĞĂŬĞƌƐƉƌŽǀŝĚŝŶŐĐůŝŶŝĐĂů͕ĞǀŝĚĞŶĐĞͲďĂƐĞĚĂŶĚ ƋƵĂůŝƚǇĨŽĐƵƐĞĚĐŽŶƚĞŶƚ͘ ŝǀĞƌƐĞĨŽƌŵĂƚƐŽĨĞĚƵĐĂƚŝŽŶĚĞůŝǀĞƌǇ͕ŝŶĐůƵĚŝŶŐ͗ĚŝĚĂĐƚŝĐ ůĞĐƚƵƌĞƐ͕ĚĞďĂƚĞƐ͕ĂŶĚŝŶƚĞƌĂĐƚŝǀĞƐĞƐƐŝŽŶƐ͕ƐƵĐŚĂƐƉŽƐƚĞƌ ƚŽƵƌƐ͕DĞĞƚƚŚĞWƌŽĨĞƐƐŽƌƐĂŶĚtŽƌŬƐŚŽƉƐĞƐƐŝŽŶƐ͘ ůĂƌŐĞƌĨŽƌƵŵĨŽƌĚŝƐĐƵƐƐŝŽŶŽĨƉƌĂĐƚŝĐĂůŵĂŶĂŐĞŵĞŶƚ ŝƐƐƵĞƐƐƵĐŚĂƐƚŚĞƵƌďƐŝĚĞŽŶƐƵůƚƐʹƐŬƚŚĞWƌŽĨĞƐƐŽƌƐ ƐĞƐƐŝŽŶ͘ ǆƚĞŶƐŝǀĞůĞĂƌŶŝŶŐŽƉƉŽƌƚƵŶŝƚŝĞƐŝŶƚŚĞďĂƐŝĐƐĐŝĞŶĐĞŽĨ ƌŚĞƵŵĂƚŽůŽŐǇ͕ĂŶĂƌĞĂŽĨƚŚĞƉƌŽŐƌĂŵĚĞǀĞůŽƉĞĚďǇĂ ƐƵďĐŽŵŵŝƚƚĞĞŽĨh^ĂŶĚŝŶƚĞƌŶĂƚŝŽŶĂůůǇƉƌŽŵŝŶĞŶƚďĂƐŝĐ ƐĐŝĞŶƚŝƐƚƐ͘ ůŝŶŝĐĂůŵĂŶĂŐĞŵĞŶƚƐĞƐƐŝŽŶƐ͕ŝŶĐůƵĚŝŶŐƚŚĞdŚŝĞǀĞƐ͛ DĂƌŬĞƚ͕ƵƌďƐŝĚĞŽŶƐƵůƚƐʹƐŬƚŚĞWƌŽĨĞƐƐŽƌƐ͕dŚĞ'ƌĞĂƚ ĞďĂƚĞĂŶĚƚŚĞZƵƉƵƐƌǇƚŚĞŵĂƚŽƐƵƐͲůŝŶŝĐĂůƐƉĞĐƚƐĂŶĚdƌĞĂƚŵĞŶƚ͗ >ƵƉƵƐEĞƉŚƌŝƚŝƐĂŶĚ'ĞŶĞƚŝĐƐ ;ďƐƚƌĂĐƚƐηϴϴϬͲϴϴϱͿ • • • • • • • • • • • • • S377 SystemicLupusErythematosusͲHumanEtiologyandPathogenesisI ;ďƐƚƌĂĐƚƐηϴϴϲͲϴϵϬͿ • • • • • • • • • • • • • S380 VasculitisI ;ďƐƚƌĂĐƚƐηϴϵϭͲϴϵϲͿ • • • • • • • • • • • • • S382 4:30Ͳ6:00PM ARHPConcurrentAbstractSession EpidemiologyandPublicHealth ;ďƐƚƌĂĐƚƐηϴϵϳͲϵϬϮͿ • • • • • • • • • • • • • S385 Psychology/SocialSciences/Pediatrics ;ďƐƚƌĂĐƚƐηϵϬϯͲϵϬϴͿ • • • • • • • • • • • • • S388

MONDAY,OCTOBER28,2013

8:30AMͲ4:00PM ACR/ARHPPosterSessionB Posterpresenterswillbeavailablefrom9:00–11:00AM. (Abstracts#909Ͳ1689) • • • • • • • • • • • • S391 11:00AMͲ12:30PM ACRPlenarySessionII (Abstracts#1690Ͳ1695) • • • • • • • • • • • • S715 2:30Ͳ4:00PM ACRConcurrentAbstractSessions EpidemiologyandHealthServicesResearchII:Epidemiologyin SystemicLupusErythematosusandRheumatoidArthritis (Abstracts#1696Ͳ1701) • • • • • • • • • • • • S718 GeneticsandGenomicsofRheumaticDiseaseI (Abstracts#1702Ͳ1707)• • • • • • • • • • • • S721 HealthServicesResearch,QualityMeasuresandQualityofCareͲ RheumatoidArthritis (Abstracts#1708Ͳ1713) • • • • • • • • • • • • S724 ImaginginPediatricArthritis,Spondyloarthritis,andOsteoarthritis (Abstracts#1714Ͳ1719) • • • • • • • • • • • • S726 MetabolicandCrystalArthropathiesII (Abstracts#1720Ͳ1725) • • • • • • • • • • • • S729 RheumatoidArthritisͲAnimalModelsI (Abstracts#1726Ͳ1731) • • • • • • • • • • • • S731 RheumatoidArthritisTreatmentͲSmallMolecules,Biologicsand GeneTherapy:EfficacyandSafetyofNovelEntities (Abstracts#1732Ͳ1737) • • • • • • • • • • • • S733 SystemicLupusErythematosusͲClinicalAspectsandTreatment: BiologicTherapy (Abstracts#1738Ͳ1743) • • • • • • • • • • • • S737 SystemicSclerosis,FibrosingSyndromes,andRaynaud’sͲClinical AspectsandTherapeuticsI (Abstracts#1744Ͳ1749) • • • • • • • • • • • • S740 VasculitisII (Abstracts#1750Ͳ1755) • • • • • • • • • • • • S743 2:30Ͳ4:00PM ACR/ARHPCombinedAbstractSession ACR/ARHPCombinedEpidemiologyAbstractSession (Abstracts#1756Ͳ1761) • • • • • • • • • • • • S746 4:30Ͳ6:00PM ACRConcurrentAbstractSessions Cytokines,Mediators,CellͲcellAdhesion,CellTraffickingand AngiogenesisII:MechanismsthatContributetoAutoimmune Inflammation (Abstracts#1762Ͳ1767) • • • • • • • • • • • • S749

EpidemiologyandHealthServicesResearchIII:HealthcareCosts andMortalityinRheumaticDisease (Abstracts#1768Ͳ1773) • • • • • • • • • • • • S751 MuscleBiology,MyositisandMyopathiesI:InsightsInto MechanismsoftheIdiopathicInflammatoryMyopathies (Abstracts#1774Ͳ1779) • • • • • • • • • • • • S754 PediatricRheumatologyͲClinicalandTherapeuticAspectsII: AutoinflammatoryDiseaseandSystemicJuvenileIdiopathic Arthritis (Abstracts#1780Ͳ1785) • • • • • • • • • • • • S757 RheumatoidArthritisͲClinicalAspectsIII:PredictorsofDisease CourseinRheumatoidArthritis (Abstracts#1786Ͳ1791) • • • • • • • • • • • • S760 RheumatoidArthritisTreatmentͲSmallMolecules,Biologicsand GeneTherapy:Safety&EfficacyofSmallMoleculeAgents (Abstracts#1792Ͳ1797) • • • • • • • • • • • • S763 SpondylarthropathiesandPsoriaticArthritis:ClinicalAspectsand Treatment:TherapeuticsandOutcomesinSpondyloarthritis (Abstracts#1798Ͳ1803) • • • • • • • • • • • • S766 SystemicLupusErythematosusͲClinicalAspects:NonͲbiologic DiseaseͲmodifyingAntirheumaticDrugs (Abstracts#1804Ͳ1809) • • • • • • • • • • • • S769 SystemicSclerosis,FibrosingSyndromes,andRaynaud’sͲClinical AspectsandTherapeuticsII (Abstracts#1810Ͳ1815) • • • • • • • • • • • • S772 4:30Ͳ6:00PM ARHPConcurrentAbstractSession RehabilitationSciences (Abstracts#1816Ͳ1821) • • • • • • • • • • • • S775

TUESDAY,OCTOBER29,2013

8:30AMͲ4:00PM ACR/ARHPPosterSessionC Posterpresenterswillbeavailablefrom9:00–11:00AM. (Abstracts#1822Ͳ2643) • • • • • • • • • • • • S779 11:00AMͲ12:30PM ACRPlenarySessionIII Discovery2013 (Abstracts#2644Ͳ2649)• • • • • • • • • • • • S1126 2:30Ͳ4:00PM ACRConcurrentAbstractSessions AntiphospholipidSyndrome (Abstracts#2650Ͳ2655) • • • • • • • • • • • • S1129 EpidemiologyandHealthServicesResearchVI:RiskFactorsin RheumaticDiseaseSusceptibility (Abstracts#2656Ͳ2661) • • • • • • • • • • • • S1131

HealthServicesResearch,QualityMeasuresandQualityofCareͲ InnovationsinHealthCareDelivery (Abstracts#2662Ͳ2667)• • • • • • • • • • • • S1135 OsteoarthritisI:TherapeuticsinOsteoarthritis (Abstracts#2668Ͳ2673)• • • • • • • • • • • • S1138 PediatricRheumatologyͲClinicalandTherapeuticAspectsIII: SystemicLupusErythematosusandOtherDiseaseOutcomes (Abstracts#2674Ͳ2679)• • • • • • • • • • • • S1141 RheumatoidArthritisͲClinicalAspectsIV:Comorbiditiesin RheumatoidArthritis (Abstracts#2680Ͳ2685)• • • • • • • • • • • • S1144 RheumatoidArthritisTreatmentͲSmallMolecules,Biologicsand GeneTherapy:NovelTreatmentStrategiesinRheumatoidArthritis (Abstracts#2686Ͳ2691)• • • • • • • • • • • • S1147 SpondylarthropathiesandPsoriaticArthritisͲPathogenesis,Etiology (Abstracts#2692Ͳ2697)• • • • • • • • • • • • S1151 SystemicLupusErythematosusͲHumanEtiologyandPathogenesisII (Abstracts#2698Ͳ2703)• • • • • • • • • • • • S1153 SystemicSclerosis,FibrosingSyndromes,andRaynaud's: Pathogenesis,AnimalModels,Genetics:NovelSignalingPathways MediatingFibrosis (Abstracts#2704Ͳ2709)• • • • • • • • • • • • S1155 TͲcellBiologyandTargetsinAutoimmuneDisease (Abstracts#2710Ͳ2715)• • • • • • • • • • • • S1158 2:30Ͳ4:00PM ARHPConcurrentAbstractSession ClinicalPractice/PatientCare (Abstracts#2716Ͳ2721)• • • • • • • • • • • • S1160 4:30Ͳ6:00PM RheumatologyResearchFoundationSpecialSession 2013RheumatologyResearchFoundationEdmondL.Dubois,MD MemorialLectureship (Abstracts#2722Ͳ2727)• • • • • • • • • • • • S1162 4:30Ͳ6:00PM ACRConcurrentAbstractSession BiologyandPathologyofBoneandJointII:OsteoclastBiologyand Arthritis (Abstracts#2728Ͳ2733)• • • • • • • • • • • • S1164 GeneticsandGenomicsofRheumaticDiseaseII (Abstracts#2734Ͳ2739)• • • • • • • • • • • • S1167 InnateImmunityandRheumaticDisease (Abstracts#2740Ͳ2745)• • • • • • • • • • • • S1169 MiscellaneousRheumaticandInflammatoryDiseasesII: AutoinflammatorySyndromes (Abstracts#2746Ͳ2751)• • • • • • • • • • • • S1172

OsteoarthritisII:RiskFactorsandNaturalHistoryofOsteoarthritis (Abstracts#2752Ͳ2757)• • • • • • • • • • • • S1175 RheumatoidArthritisͲClinicalAspectsV:ObservationalStudiesin RheumatoidArthritis (Abstracts#2758Ͳ2763)• • • • • • • • • • • • S1177 RheumatoidArthritisTreatmentͲSmallMolecules,Biologicsand GeneTherapy:EfficacyofApprovedBiologicsI (Abstracts#2764Ͳ2769)• • • • • • • • • • • • S1181 Sjögren'sSyndrome:BasicScience (Abstracts#2770Ͳ2775)• • • • • • • • • • • • S1184 SpondylarthropathiesandPsoriaticArthritis:ClinicalAspectsand Treatment:ClinicalFeaturesofSpondyloarthritis (Abstracts#2776Ͳ2781)• • • • • • • • • • • • S1187 VasculitisIII (Abstracts#2782Ͳ2787)• • • • • • • • • • • • S1190 4:30Ͳ6:00PM ACR/ARHPCombinedAbstractSession ACR/ARHPCombinedPediatricsAbstractSession (Abstracts#2788Ͳ2793)• • • • • • • • • • • • S1193

WEDNESDAY,OCTOBER30,2013

9:00Ͳ10:30AM ACRConcurrentAbstractSessions BcellsinSystemicLupusErythematosus (Abstracts#2794Ͳ2799)• • • • • • • • • • • • S1196 ImaginginRheumatoidArthritis (Abstracts#2800Ͳ2805)• • • • • • • • • • • • S1198 InfectionRelatedRheumaticDiseases (Abstracts#2806Ͳ2811)• • • • • • • • • • • • S1201 MedicalEducation (Abstracts#2812Ͳ2817)• • • • • • • • • • • • S1204 RheumatoidArthritisͲAnimalModelsII (Abstracts#2818Ͳ2823)• • • • • • • • • • • • S1207 RheumatoidArthritisͲClinicalAspectsVI:CardiovascularDiseasein RheumatoidArthritis (Abstracts#2824Ͳ2829)• • • • • • • • • • • • S1209 SystemicLupusErythematosusClinicalAspects:Pregnancy (Abstracts2830Ͳ2835) • • • • • • • • • • • • S1212 SpondylarthropathiesandPsoriaticArthritis:ClinicalAspectsand Treatment:ClinicalandImagingAspectsofAxialSpondyloarthritis (Abstracts#2836Ͳ2841)• • • • • • • • • • • • S1215 ARHPConcurrentAbstractSession Education/CommunityPrograms (Abstracts2842Ͳ2847) • • • • • • • • • • • • S1218

11:00AM–12:30PM ACRConcurrentAbstractSessions Fibromyalgia,SoftTissueDisordersandPain:Treatmentand OutcomeAssessment (Abstracts#2848Ͳ2853)• • • • • • • • • • • • S1221 Orthopedics,LowBackPain,RehabilitationandMechanismsof PaininArthritis (Abstracts#2854Ͳ2859)• • • • • • • • • • • • S1223 PediatricRheumatologyͲPathogenesisandGenetics (Abstracts#2860Ͳ2865)• • • • • • • • • • • • S1226 RheumatoidArthritisͲClinicalAspectsVII:Remission,Flareand OutcomeMeasuresinRheumatoidArthritis (Abstracts#2866Ͳ2871)• • • • • • • • • • • • S1228 RheumatoidArthritisͲPathogeneticPathways (Abstracts#2872Ͳ2877)• • • • • • • • • • • • S1232 Sjögren'sSyndrome:ClinicalAdvances (Abstracts#2878Ͳ2883)• • • • • • • • • • • • S1234 SpondylarthropathiesandPsoriaticArthritis:ClinicalAspectsand Treatment:ImaginginAxialSpondylarthropathies:Challenges, Advances (Abstracts#2884Ͳ2889)• • • • • • • • • • • • S1237 SystemicLupusErythematosusͲAnimalModels (Abstracts#2890Ͳ2895)• • • • • • • • • • • • S1240 SystemicLupusErythematosusͲClinicalAspects:Cardiovascularand OtherComplicationsofLupus (Abstracts#2896Ͳ2901)• • • • • • • • • • • • S1242 SystemicSclerosis,FibrosingSyndromesandRaynaud’sͲ Pathogenesis,AnimalModelsandGenetics:Pathogenesisof SystemicSclerosis (Abstracts#2902Ͳ2907)• • • • • • • • • • • • S1245 11:00AM–12:30PM ARHPConcurrentAbstractSession ResearchandHealthServices (Abstracts#2908Ͳ2913)• • • • • • • • • • • • S1247 11:00AM–12:30PM ACRConcurrentAbstractSessions SystemicSclerosis,FibrosingSyndromes,andRaynaud’sͲClinical AspectsandTherapeuticsIII (Abstracts#2914Ͳ2919)• • • • • • • • • • • • S1250

Antiphospholipid Syndrome: Clinical Manifestations and New Biomarkers in Antiphospholipid Syndrome Sunday, October 27, 2013, 8:30 AM–4:00 PM

1 Anti-␤2GP1-Domain-1 Antibodies Are a Marker Of APS Severity. Nancy Agmon-Levin1, Luciana Seguro2, Cristina Rosa´rio3, Michael Mahler4, Mariele Gatto5, Nir Tomer6, Elaine P. Leon7, Andrea Doria8, La´szlo´ Kovaćs9, Nathalie Costedoat-Chalumeau10, Boris Gilburd11 and Yehuda Shoenfeld12. 1 The Zabludowicz Center for Autoimmune Diseases, Tel-Hashomer, Israel, 2 Division of Rheumatology, University of Saõ Paulo School of Medicine, Sao Paulo, Brazil, 3Internal Medicine Department, Hospital de Pedro Hispano, Rua Dr. Eduardo Torres Matosinhos, Porto, Portugal, 4INOVA Diagnostics, San Diego, CA, 5Department of Medicine, University of Padova, Padova, Italy, 6The Zabludowicz Center for Autoimmune Diseases, Sheba Medical Center, Tel-Hashomer, Ramat Gan, Israel, 7Faculdade de Medicina da Universidade de Saõ Paulo, Saõ Paulo, Brazil, 8University of Padova, Padova, Italy, 9University of Szeged, Szeged, Hungary, 10Hopital Cochin, Paris, France, 11The Zabludowicz Center for Autoimmune Diseases, Sheba Medical Center, Tel-Hashomer, Ramat-Gan, Israel, 12Sheba Medical Center, Ramat Gan, Israel. Background/Purpose: Autoantibodies targeting Domain-1(D1) of beta2-glycoprotein-1(␤2GP1) are common among patients with the antiphospholipid syndrome (APS)1. However, few studies assessed their clinical relevance and correlation with other anti-phospholipid antibodies (aPLA). Thus, in the current study we evaluated the correlations between anti-␤2GP1-D1 levels and APS clinical and serological parameters. Methods: Serum samples of 178 APS patients, 50 healthy subjects and 47 patients diagnosed with sepsis were utilized to determine levels of IgGantibodies to cardiolipin(aCL), ␤2GP1 and ␤2GP1-D1 (BIO-FLASH®, INOVA). Demographic, clinical and serological (i.e. Lupus anticoagulant (LAC)) data were analyzed. Results: Among APS patients aCL, anti-␤2GP1 and anti-␤2GP1-D1 were present in 60%, 70% and 49%, respectively. The presence of anti␤2GP1-D1 antibodies correlated with the presence of LAC (91 vs. 67%; p⬍0.001), aCL (98 vs. 23%; p⬍0.001). Triple aPLA positivity was present in 89% of anti-␤2GP1-D1 positive sample compared with 16% among anti␤2GP1-D1 negative samples (p⬍0.001). Anti-␤2GP1-D1 antibodies correlated with higher titers of aCL antibodies (p⬍0.001) and anti -␤2GP1 (p⫽0.03). Anti ␤2GP1 antibodies correlated with the occurrence of venous thrombosis (p⬍ 0.009). Anti-␤2GP1-D1 positivity (⬎20CU) related to the occurrence of any thrombotic event (91% vs. 79% OR 2.54; 95% CI 1.05–6.15; p⫽0.039). Medium levels of anti-␤2GP1-D1 (⬎30CU) correlated with arterial thrombosis (55% vs. 33% OR⫽2.5; 95%CI 1.37–4.67; p ⫽ 0.006). High levels of anti-␤2GP1-D1 (⬎64CU) were associated with multiple thrombotic events (62% vs. 31% OR 3.58; 95%CI 1.49–8.61; p⫽0.005) arterial thrombosis (60% vs. 33% OR 3.04; 95%CI 1.61–5.73; p⫽0.001) and CNS manifestations (45% vs. 27% OR 1.99; 95%CI 1.03–3.81; p⫽0.038). Conclusion: In our cohort of APS patients, anti-␤2GP1-D1 antibodies were a marker of high risk aPLA profile. Moreover their presence, particularly in med-high levels correlated with repeated thromboses, arterial occlusion and CNS-related manifestations. Thus, anti-␤2GP1-D1 may potentially serve as a biomarker of thrombosis in APS. 1

Disclosure: C. Lopez-Pedrera Sr., None; P. Ruiz-Limon, None; R. Teruel, None; Aguirre Zamorano, None; R. M. Carretero-Prieto, None; N. Barbarroja, None; A. Rodriguez-Ariza, None; E. Collantes-Estevez, None; R. gonzalez-Conejero, None; C. Martinez, None; M. J. Cuadrado, None; C. Perez-Sanchez, None.

3 The Global Antiphospholipid Syndrome Score (GAPSS) In Primary APS. Savino Sciascia1, Giovanni Sanna2, Veronica Murru3, Dario Roccatello1, Munther A. Khamashta3 and Maria Laura Bertolaccini3. 1Centro di Immunopatologia e Documentazione su Malattie rare, Torino, Italy, 2Louise Coote Lupus Unit, St. Thomas’ Hospital, London, United Kingdom, 3Lupus Research Unit, The Rayne Institute, Kings College London School of Medicine, London, United Kingdom. Background/Purpose: Antiphospholipid Syndrome (APS) is a heterogeneous entity with a wide variation in clinical course and laboratory profile. We aimed to evaluate the clinical utility of GAPSS (Global APS Score) in a cohort of primary APS patients. GAPSS is derived from the combination of independent risk for thrombosis and pregnancy loss (PL), taking into account the antiphospholipid antibodies (aPL) profile, the conventional cardiovascular risk factors, and the autoimmune antibodies profile. Methods: This study included 62 consecutive with primary APS. Data on clinical manifestations, conventional cardiovascular risk factors, and antiphospholipid antibodies profile were collected. GAPSS scoring system was calculated for each patient by adding together the points corresponding to the risk factors, based on a linear transformation of the corresponding ␤ regression coefficient as follows: 3 for hyperlipidemia, 1 for arterial hypertension, 5 for aCL IgG/IgM, 4 for anti-␤2GPI IgG/IgM, 3 for aPS/PT IgG/IgM and 4 for LA. Results: Higher values of GAPSS were showed in patients who experienced thrombosis alone compared to those with pregnancy loss (PL) alone (11.5⫾4.6 and 8.7⫾3.2,p⫽0.0378). Patients with both thrombosis and pregnancy loss showed higher GAPSS compared to those pregnancy loss alone (12.5⫾4.6 vs 8.7⫾3.2,p⫽0.0152). Higher values were also shown sub-grouping for the site of thrombosis, compared to pregnancy loss alone

Mahler M, et al Autoimmun Rev. 2012;12(2):313–7.

Disclosure: N. Agmon-Levin, None; L. Seguro, None; C. Rosa´rio, None; M. Mahler, Inova Diagnostics, Inc., 3; M. Gatto, None; N. Tomer, None; E. P. Leon, None; A. Doria, None; L. Kovaćs, None; N. Costedoat-Chalumeau, None; B. Gilburd, None; Y. Shoenfeld, None.

2 Charaterization Of Micrornas Involved In The Regulation Of Atherotrhombosis In Antiphospholipid Syndrome and Systemic Lupus Erythematosus. Chary Lopez-Pedrera Sr.1, Patricia Ruiz-Limon1, Raul Teruel2, Ańgeles Aguirre Zamorano1, Rosario M. Carretero-Prieto1, Nuria Barbarroja1, Antonio Rodriguez-Ariza1, Eduardo Collantes-Estevez1, Rocio gonzalez-Conejero2, Constantino Martinez2, Ma Jose Cuadrado3 and Carlos Perez-Sanchez1. 1IMIBIC-Reina Sofia Hospital, Cordoba, Spain, 2Regional Centre for Blood Donation, University of Murcia, Murcia, Spain, 3The Rayne Institute, London, United Kingdom.

S1

Sunday, October 27

Background/Purpose: miRNAs are key players in pathophysiological processes, but no previous studies have investigated their asscociation with the cardiovascular and atherothrombotic risks observed in primary antiphospholipid syndrome (APS) and systemic lupus erythematosus (SLE). The aim was to characterize the miRNAs involved in pro-inflammatory, prothrombotic and pro-oxidative status in SLE and APS patients. Methods: In silico search was performed and six putative miRNAs regulating factors involved in the proinflammatory and pro-oxidative status of APS and SLE patients (miR-124a, -125a, -125b, -146a, -155, and -222) were selected and quantified by RT-PCR in neutrophils and monocytes from 11 APS and 17 SLE patients, and 26 healthy donors. Pro-inflammatory and prothrombotic proteins and oxidative stress markers were evaluated by flow cytometry. Proteins related to the biogenesis of miRNAs were quantified by RT-PCR and Western Blot. Results: Expression of selected miRNAs was significantly decreased in neutrophils from SLE and APS patients compared to healthy donors. However, only miR-124a was found significantly reduced in monocytes from SLE and APS patients. The expression of those miRNAs negatively correlated with autoimmunity-related parameters (anti-dsDNA and aCL-IgG), disease activity (SLEDAI), inflammation (IL-2, -6, -8 and MCP-1) and oxidative stress (peroxides). Low levels of specific miRNAs in neutrophils and monocytes from both SLE and APS were found further associated with thrombotic events and pathological carotid intima media thickness. This selective decrease miRNA expression was related to a significant reduction in the expression of miRNAs biogenesis genes (Xportin-5, Drosha, Dicer, Ago-1 and Ago-2) in neutrophils from APS and SLE patients compared to healthy donors. In vitro treatment of healthy monocytes with purified antiocardiolipin antibodies from APS patients caused a significant decrease in the levels of miR-124a. Notably, when this miRNA was transfected into monocytic THP-1 cells a significant decrease (⬃30%) in MCP-1, STAT3 and p38 expression was observed, pointing at these proteins as potential targets of miR-124a. Conclusion: 1. Decreased expression of a number of miRNAs in monocytes and neutrophils from APS and SLE patients correlates with autoimmunity, inflammation, thrombosis and oxidative stress and atherothrombotic markers. 2. A down-regulation of miRNA processing might explain the low expression of evaluated miRNAs in APS and SLE neutrophils. 3. Antiocardiolipin antibodies directly regulate miR-124a expression, which is directly involved in the proatherosclerotic profile of APS and SLE patients.

ACR Poster Session A

Sunday, October 27

(12.2⫾5.2, for arterial thrombosis; 12.⫾4.0 for venous and 8.7⫾3.2, p⫽0.0214 and p⫽0.0472, respectively). Patients with thrombotic recurrences showed higher values of GAPSS compared to those without (13.7⫾3.1Vs 9.4⫾3.9,p⫽0.0205, respectively). Higher values were also seen when comparing recurrences Vs no recurrences according to the site of thrombotic event (13.9⫾3.6 and 11.0⫾4.3,p⫽0.0143 for arterial; 13.6⫾2.18 Vs 8.91⫾3.6,p⫽0.001, for venous events, respectively). GAPSS values higher or equal to 11 were strongly associated with higher risk of recurrences (OR 18.27 [95%IC 3.74–114.5] for cut off 11, OR 20.64[3.92–185.92] for cut off 12, 21.64[3.89–189.56] for cut off 15, respectively. GAPSS values higher or equal to 11 seemed to have the best accuracy, in terms of sensitivity and specificity. Conclusion: GAPSS is demonstrated to be a valid tool to a substantial improvement in risk stratification for thrombosis in primary APS, also in terms of recurrences.

5 Plasma Levels Of PF-4var/CXCL4L1, A Non Allelic Variant Of Platelet Factor-4 (PF-4/CXCL4), Are Elevated In Patients With Antiphospholipid Syndrome (APS). Marina Sikara1, Markos Patsouras1, Elias Eliopoulos2 and Panayiotis Vlachoyiannopoulos1. 1School of Medicine, National University of Athens, Athens, Greece, 2Agricultural University of Athens, Athens, Greece. Background/Purpose: Platelet derived chemokines, such as PF-4 and a recently isolated protein product of its nonallelic variant gene PF-4var, are implicated in several aspects of vascular thrombosis and inflammation. The above chemokines present only 4.3% aminoacid divergence in the mature proteins; however they exhibit distinct platelet secretion mode and function. The precise role of PF-4var regarding the haemostatic balance is not yet studied. Previous study from our group demonstrates a novel interaction between ␤2-glycoprotein I (␤2GPI), the major autoantigen in APS, and PF-4 or PF-4var. This complex formation leads in the stabilization of ␤2GPI dimeric structure which facilitates the antibody recognition and platelet activation, as indicated by p38MAPK phosphorylation and thromboxane production. To determine PF-4var plasma levels and platelet PF-4var expression (RNA level) in patients with APS and evaluate the correlation with clinical and laboratory parameters of the disease. Methods: From 70 patients, who fulfill the revised diagnostic criteria for APS, blood samples were taken and separate samples of serum, plasma and platelets were isolated. Complete blood count, C-reactive protein (CRP), erythrocyte sedimentation rate (ESR), aPTT, anti-cardiolipin (anti-CL) and anti-␤2GPI antibodies were measured. A healthy control (n⫽30) and a disease control group (SLE, n⫽26) were included in the study. Plasma levels of PF-4var were determined using a commercially available ELISA, which absolutely discriminates PF-4var from PF-4 protein. Statistical analyses were evaluated by Mann-Whitney t-test and Kruskal-Wallis test. Platelet PF-4var and PF4 RNA levels were determined in RNA isolated from donor’s and patient’s platelet preparations using quantitative Real-Time PCR. Results: APS patients showed higher levels of plasma PF-4var compared to healthy individuals (median 137 pg/ml; intarquartile range 66.4–200.5 pg/ml versus 79.03 [40.3–99.2] pg/ml, p⫽0.0052). In addition, RT-PCR revealed significantly higher PF-4var expression in platelets derived from patients comparing to healthy donors (2⫺⌬⌬CT: 1.540; [1.020–1.943] versus 0.89 [0.705–1.203], p⬍0.001). PF-4var levels were significantly elevated in patients suffering from primary APS (PAPS) than those with APS secondary to SLE (SAPS), (197.7 [113.3–304.8] pg/ml versus 126.6 [49.94–170] pg/ml, p⫽0.0086). Regarding the clinical presentation of the disease, patients who experienced thrombotic events versus pregnancy morbidity or arterial versus venous thrombotic events do not show statistically significant difference in PF-4var levels. A positive correlation was also revealed between the presence of thrombocytopenia and the elongation of aPTT with the higher PF-4var levels (p⫽ 0.0048 and p⫽0.0195, respectively). Conclusion: Preliminary results suggest that higher PF-4var levels are present in plasma of APS patients and especially in those with PAPS and these are associated with laboratory characteristics indicative for higher risk of thrombotic complications.

Disclosure: S. Sciascia, None; G. Sanna, None; V. Murru, None; D. Roccatello, None; M. A. Khamashta, None; M. L. Bertolaccini, None.

4 Proteomic Analyses Of Monocyte Responses To IgG From Patients With Antiphospholipid Syndrome. Vera M. Ripoll, Anastasia Lambrianides, Wendy Heywood, Anisur Rahman, Yiannis Ioannou and Ian Giles. University College London, London, United Kingdom. Background/Purpose: Antiphospholipid syndrome (APS) is an autoimmune disorder characterised by recurrent vascular thrombosis (VT) and/or pregnancy morbidity (PM) caused by antiphospholipid antibodies (aPL). Whilst aPL are known to activate monocytes by inducing production of tissue factor and pro-inflammatory cytokines, upstream mechanisms leading to this pattern of cellular activation are not fully understood. There is evidence that aPL from VT and PM patients stimulate different monocyte intracellular pathways but these have not been characterised in detail. We therefore carried out proteomic analysis of human monocytes treated with IgG from VT, PM and healthy controls (HC) using two different techniques. Methods: Human monocytic cell line U937 and healthy monocytes were treated with 200 ␮g/ml of IgG purified from patients with VT, PM or HC. Proteomic analysis was performed using two dimensional difference gel electrophoresis (2D DiGE) and Label Free Quantitative LC-MS/MS. Differentially upregulated proteins were identified by mass spectrometry and validated using quantitative PCR (qPCR) and western blotting. Results: The table shows the numbers of proteins whose expression in 2D DiGE analysis was regulated by at least two-fold (up or down) in cells exposed to VT IgG or PM IgG compared to HC IgG. VT IgG PM IgG

U 937

Healthy monocytes

41 upregulated, 11 downregulated 22 upregulated, 1 downregulated

119 upregulated, 21 downregulated 105 upregulated, 22 downregulated

Monocytes showed more regulated proteins than U937 cells. Far more proteins were upregulated than downregulated in both cell types. 11 proteins showing the most significant regulation were identified by mass spectrometry analysis. Of these, vimentin and zinc finger CCCH domain-containing protein 18 (ZCH18) were the most significantly upregulated in both U937 cells and monocytes. Additionally, Myeloperoxidase (MPO) and CAP Gly domaincontaining linker protein 2 (CLIP2) were significantly induced in monocytes. Upregulation of vimentin, ZCH18, MPO and CLIP2 was confirmed by qPCR. Induction of vimentin and MPO was also validated by western blotting. Anti-cardiolipin/vimentin antibodies have been reported in APS. We found anti-vimentin in serum of 35% of patients with APS but no correlation between anti-vimentin level and ability of IgG to up-regulate monocyte vimentin. LC-MS/MS of healthy monocytes exposed to VT IgG revealed over 100 proteins that were differentially regulated compared to monocytes exposed to HC IgG. Pathway analysis showed that these proteins were involved in cytoskeletal, coagulation and integrin-signaling functions; all relevant to APS. S100 A11 and POTE ankyrin domain family member E were significantly upregulated whereas Plasminogen activator inhibitor 2 and Coronin 1 were downregulated. Targets were validated using qPCR. Conclusion: Two different proteomic techniques were used to identify novel proteins involved in the actions of aPL on monocytes. Several of these proteins are strongly associated with autoimmune disease and coagulation and could become potential new therapeutic targets for APS.

Disclosure: M. Sikara, None; M. Patsouras, None; E. Eliopoulos, None; P. Vlachoyiannopoulos, None.

6 Researchers’ Own Experience In Managing Obstetric Complications In Pregnant Women With Antiphospholipid Syndrome. Lidia Ostanek1, Danuta Bobrowska - Snarska1, Barbara Nestorowicz2 and Marek Brzosko3. 1 Pomeranian Medical University, Szczecin, Poland, Szczecin, Poland, 2Pomeranian Medical University, Szczecin, Poland, Szczecin, Poland, 3Department of Rheumatology and Internal Diseases Pomeranian Medical University in Szczecin, Szczecin, Poland. Background/Purpose: 1. To analyse risk factors for obstetric pathology in patients with antiphospholipid syndrome (APS). 2. To assess the effectiveness of treatment depending on pharmacotherapy. Methods: The courses of 275 pregnancies and deliveries in patients with diagnosed APS (PAPS-68, SAPS-41 patients) were analysed. 101

Disclosure: V. M. Ripoll, None; A. Lambrianides, None; W. Heywood, None; A. Rahman, None; Y. Ioannou, None; I. Giles, None.

S2

Sunday, October 27

pregnancies were treated according to current recommendations. The treatment was consistently followed up with both a rheumatologist and an obstetrician. A retrospective analysis of previous pregnancies and deliveries, both treated and untreated was conducted. The immunological profile of the patients (ANA, SS-A, SS-B, Ro52, lupus anticoagulant (LA), anticardiolipin antibodies (aCL), anti ␤2GPI (a␤2GPI), antiprothrombin antibodies (aPTR), anti-phosphatidylserine (aPS), complement component 3 and complement component 4) was investigated. Assessment of metabolic and environmental factors influencing pregnancy (smoking, arterial hypertension, lipid profile, hyperglycaemia, uric acid concentration) was carried out. Yate’s chi-sqared test, Pearson’s ch-sqared test, Spearman’s rank correlation and logistic regression analysis were used to conduct the statistical analysis. Results: Risk factors for unsuccessful pregnancy included: young age at the diagnosis and during pregnancy (p⬍0,001), presence of LA (p⫽0,0046), aPTR (p⫽0,034), arterial hypertension (p⫽0,012), hyperuricemia (p⫽0,005). Miscarriages were more frequent in patients with PAPS than with SAPS (p⫽0,007), in patients with aCL IgM (p⫽0,042) and in patients treated with glucocorticosteroid (GS) (p⫽0,027). Premature births were more frequent in young patients (p⫽0,002), with SAPS (p⫽0,005), with presence of ANA (p⫽0,0026), LA (p⫽0,045), low concentration of C3 (p⫽0,006), patients with smoking history (p⫽0,018) and those treated with GS (p⫽0,000). The presence of LA involved the risk of fetal death (p⫽0,027). Chances of successful pregnancy were enhanced by: the application of low molecular weight heparin (LMWH)(p⫽0,000), aspirin (ASA) (p⫽0,001), combined treatment: LMWH⫹ASA (p⫽0,002), and immunoglobulins iv (p⫽0,002). Conclusion: 1. The occurrence of aPL and diagnosis of APS increase the risk of obstetric pathology. 2. Apart from immunologic factors (aCL, LA, aPTR,2C3), there are other factors which determine the risk of pregnancy failure, i.e. demographic factors (age), environmental factors (smoking, hyperuricemia) and associated treatment (GS). 3. PAPS increases the risk of miscarriage, SAPS increases the risk of premature birth. 4. Treating APS patients with ASA, LMWH, and in justified cases with immunoglobulins iv, increases the chances of successful pregnancy.

Conclusion: We have prospectively demonstrated that GAPSS is a valid tool for accurate prediction of thrombosis in SLE patients with aPL. 1. GAPSS: the Global Anti-Phospholipid Syndrome Score. Rheumatology (Oxford). 2013 Jan 12. [Epub ahead of print] Disclosure: S. Sciascia, None; M. J. Cuadrado, None; G. Sanna, None; V. Murru, None; D. Roccatello, None; O. Ateka, None; M. A. Khamashta, None; M. L. Bertolaccini, None.

8 Clinical Evaluation Of Two Anti-Beta2glycoprotein I Domain 1 Autoantibody Assays To Aid In The Diagnosis and Risk Assessment Of The Antiphospholipid Syndrome. Rohan Willis1, Michael Mahler2, Francesca Pregnolato3, Charis Pericleous4, Anisur Rahman4, John Ioannou5, Ian Giles4, Gabriella Lakos2, Roger Albesa2, Navid Zohoury2, Pier-Luigi Meroni6 and Silvia S. Pierangeli1. 1University of Texas Medical Branch, Galveston, TX, 2INOVA Diagnostics, San Diego, CA, 3Lab of Immunology, IRCCS Istituto Auxologico Italiano, Milan, Italy, 4University College London, London, United Kingdom, 5Rayne Institute, University College London, London, UK, London, United Kingdom, 6 University of Milan, Milano, Italy.

Disclosure: L. Ostanek, None; D. Bobrowska - Snarska, None; B. Nestorowicz, None; M. Brzosko, None.

Background/Purpose: Antiphospholipid Syndrome (APS) is characterized by the presence of antibodies to phospholipids (aPL) and to ␤2glycoprotein I (␤2GPI) in patients with thrombosis or pregnancy morbidity. Several studies provided evidence that antibodies to domain 1 of ␤2GPI (␤2GPI-D1) represent a promising biomarker for the diagnosis and risk assessment of APS. Several immunoassays have been described to detect those antibodies, but harmonization of those tests as well as correlation of those assays has not been properly addressed. Here we aimed to compare two assays for the detection of anti-␤2GPI-D1 antibodies using clinically defined patients samples. Methods: A total of 72 APS patients and 45 controls (including healthy individuals, patients with infectious or autoimmune diseases) were tested for anti-␤2-GPI antibodies by ELISA (in-house, London, UK and INOVA Diagnostics, San Diego, US) and for anti-␤2GPI-D1 by QUANTA Flash CIA (INOVA) and by ELISA (in-house, London, UK). History of thrombosis was known for 94 patients (39 with and 55 without history of thrombosis). Results: Both anti-␤2GPI-D1 assays showed good qualitative (79.5%, 95% CI 71.0–86.4; kappa⫽0.60, 95% CI 0.46–0.74) and quantitative agreements (spearman‘s rho⫽0.76, 95% CI 0.67–0.83) as well as similar discrimination between APS patients and controls as shown by ROC analysis.

7 Prospective Validation Of The Global Antiphospholipid Syndrome Score (GAPSS). Savino Sciascia1, Ma Jose Cuadrado2, Giovanni Sanna3, Veronica Murru4, Dario Roccatello1, Oier Ateka5, Munther A. Khamashta4 and Maria Laura Bertolaccini4. 1Centro di Immunopatologia e Documentazione su Malattie rare, Torino, Italy, 2The Rayne Institute, London, United Kingdom, 3 Louise Coote Lupus Unit, St. Thomas’ Hospital, London, United Kingdom, 4 Lupus Research Unit, The Rayne Institute, Kings College London School of Medicine, London, United Kingdom, 5Lupus Unit, London, United Kingdom. Prospective Validation of the Global Antiphospholipid Syndrome Score (GAPSS) Background/Purpose: Backgrounds: This study was performed to prospectively and independently validate the GAPSS (1) (Global APS Score), a score system derived from the combination of independent risk factors for thrombosis, taking into account the antiphospholipid antibodies (aPL) profile and the conventional cardiovascular risk factors. Methods: GAPSS were prospectively applied to 52 consecutive SLE patients with positive aPL (51 female, mean age 39.1⫾10.6, mean follow-up 47.53⫾19.15 months). GAPSS scoring system was calculated yearly for each patient by adding together the points corresponding to the risk factors. Results: An increase in the GAPSS (entry Vs. last visit) was seen in patients who experienced thrombosis (n⫽5, 6⫾5.05 Vs 9.4⫾4.93, p⫽0.0388). No changes were observed in those without thrombotic event (n⫽47, 8.28⫾4.88 Vs 7.13⫾5.75, p⫽ NS). An increase of more than 3 points in GAPSS during the follow-up was strongly associated with higher risk of thrombosis (HR 25.00 [95%IC 3.74–189.1]). The risk of thrombosis was also evaluated by Kaplan-Meier analysis (fig.1; IE: ischemic event) and the cumulative proportion of thrombosis-free individuals was higher (p⫽0.002) in the patients whose GAPSS was not increased by 3 or more points.

Comparison between assays Qualitative (Total Agreement %/kappa) Quantitative (Spearman) B2 ELISA (UK) B2 ELISA (INOVA) D1 ELISA (UK)

S3

B2 ELISA (INOVA)

D1 ELISA (UK)

88.9 (0.77) 0.82 (0.75–0.87) /

86.3 (0.73) 0.83 (0.76–0.88) 82.1 (0.65) 0.68 (0.57–0.77) /

/

D1 CIA 89.7 (0.79) 0.84 (0.78–0.89) 85.5 (0.71) 0.76 (0.67–0.83) 79.5 (0.60) 0.76 (0.67–0.83)

Conclusion: A-␤2GPI-D1 is the prevalent specificity of a-␤2GPI not only in PAPS with thrombosis, but also in PAPS with pure obstetric disease and in SLE/UCTD patients with no APS manifestations. In the latter group, the pathogenic potential of a-␤2GPI-D1 might be mitigated by the absence of additional risk factors and/or to the presence of adequate prophylaxis. aPL carriers appear not to have a polarized profile, suggesting that a-␤2GPI-D1 may be a fingerprint of systemic autoimmune diseases.

Sunday, October 27

Comparison between APS patients and controls

Assay D1 CIA D1 ELISA B2 ELISA (UK) B2 ELISA (INOVA)

Area under the curve (95% Confidence interval) 0.89 (0.83–0.96) 0.83 (0.75–0.90) 0.90 (0.84–0.96)

LRⴙ/LRⴚ at cut-off 9.69/0.15 4.02/0.44 6.25/0.19

LRⴙ/LRⴚ at LRⴙ max 13.8/0.71 11.25/0.77 9.69/0.60

Sensitivity (95% Confidence interval) 86.1 (75.9–93.1) 62.5 (50.3–73.6) 83.3 (72.7–91.1)

Specificity (95% Confidence interval) 91.1 (78.8–97.5) 84.4 (70.5–93.5) 86.7 (73.2–94.9)

0.93 (0.88–0.97)

7.5/0.19

24.4/0.47

83.3 (72.7–91.1)

88.9 (75.9–96.3)

Disclosure: L. Andreoli, None; C. Nalli, None; M. O. Borghi, None; F. Pregnolato, None; A. Zanola, None; C. Grossi, None; M. Gerosa, None; F. Allegri, None; M. Mahler, Inova Diagnostics, Inc., 3; G. L. Norman, Inova Diagnostics, Inc., 3; P. L. Meroni, None; A. Tincani, None.

Discrimination between patients with and without history of thrombosis D1 CIA 89.7 76.4 3.80 0.13

Sensitivity Specificity Likelihood ratio (⫹) Likelihood ratio (⫺)

D1 ELISA 64.1 74.5 2.52 0.48

B2 ELISA (UK) 87.2 72.7 3.20 0.18

B2 ELISA (INOVA) 79.5 80.0 3.97 0.26

10 Autoantibodies Targeting Domain 1 Of Beta 2 Glycoprotein I As Promising Marker In The Diagnosis and Risk Stratification Of The Antiphospholipid Syndrome. Navid Zohoury1, Munther A. Khamashta2, Tatsuya Atsumi3, Jacek Musial4, Toshiyuki Watanabe5, Maria Papp6, Concepcioń Gonza´lez-Rodrı´guez7, Roger Albesa1, Gary L. Norman1, Pier-Luigi Meroni8 and Michael Mahler1. 1INOVA Diagnostics, San Diego, CA, 2 Lupus Research Unit, The Rayne Institute, Kings College London School of Medicine, London, United Kingdom, 3Hokkaido University Graduate School of Medicine, Sapporo, Japan, 4Jagiellonian University Medical College, Krakow, Poland, 5Hokkaido University Graduate School of Medicine, Hokkaido, Japan, 6University of Debrecen, Debrecen, Hungary, 7Virgen Macarena University Hospital, Seville, Spain, 8University of Milan, Milano, Italy.

Conclusion: Anti-␤2GPI-D1 antibodies are a promising biomarker to aid in the diagnosis and risk assessment of APS patients. Confirmatory studies using multi-centric setups and large patient cohorts are necessary to confirm the data. Disclosure: R. Willis, None; M. Mahler, Inova Diagnostics, Inc., 3; F. Pregnolato, None; C. Pericleous, None; A. Rahman, None; J. Ioannou, None; I. Giles, None; G. Lakos, Inova Diagnostics, Inc., 3; R. Albesa, Inova Diagnostics, Inc., 3; N. Zohoury, Inova Diagnostics, Inc., 3; P. L. Meroni, Inova Diagnostics, Inc., 5; S. S. Pierangeli, None.

Background/Purpose: Antiphospholipid Syndrome (APS) is characterized by the presence of antibodies to phospholipids (aPL) and to ␤2glycoprotein I (␤2GPI) in patients with thrombosis or pregnancy morbidity. Several studies provided evidence that antibodies to Domain 1 of ␤2GPI (␤2GPI-D1) represent a promising biomarker for the diagnosis and risk assessment of APS patients. Here we aimed to analyze the diagnostic potential of ␤2GPI-D1 antibodies and the value in risk stratification of APS patients using a large multi-centric cohort of patients. Methods: A total of 273 APS patients and 1096 controls (including healthy individuals, patients with infectious or autoimmune diseases) were tested for anti-␤2GPI-D1 by QUANTA Flash CIA (INOVA). A reduced number of samples (n⫽622) were also tested on anti-␤2GPI by QUANTA Flash CIA (INOVA). Clinical data including the history of thrombosis was known in 89 APS patients (58 had a history of thrombosis). Results: In the entire cohort, anti-␤2GPI-D1 antibodies differentiated APS and controls with a sensitivity and specificity of 49.8% and 99.6%, respectively. The likelihood ratios were LR⫹ 136.5 and LR- 0.5. The prevalence of anti-␤2GPI antibodies was higher in primary than in secondary APS and reach significance for anti-␤2GPI-D1 (p⫽0.0029), but not for anti-␤2GPI antibodies (p⫽0.06). When compared with the anti-␤2GPI assay, both assays showed good qualitative (79.8%, 95% CI 69.1–86.5; kappa⫽0.60, 95% CI 0.44–0.76) and quantitative agreements (spearman‘s rho⫽0.81, 95% CI 0.73–0.87) as well as similar discrimination between APS patients and controls as shown by ROC analysis. In the smaller sub-cohort (n⫽622), sensitivity/specificity were 50.2%/99.2% (␤2GPI-D1) and 72.8%/ 87.3% (␤2GPI), respectively.

9 Domain 1 Is The Main Specificity Of Anti-␤2glycoprotein I Antibodies In Systemic Autoimmune Diseases. Laura Andreoli1, Cecilia Nalli1, MariaOrietta Borghi2, Francesca Pregnolato3, Alessandra Zanola1, Claudia Grossi4, Maria Gerosa5, Flavio Allegri1, Michael Mahler6, Gary L. Norman6, Pier Luigi Meroni7 and Angela Tincani1. 1Rheumatology Unit, University of Brescia, Brescia, Italy, 2Lab of immunology, IRCCS Istituto Auxologico Italiano, Department of Clinical Sciences and Community Health, University of Milan, Milano, Italy, 3Lab of Immunology, IRCCS Istituto Auxologico Italiano, Milan, Italy, 4Laboratory of Immuno-rheumatology, IRCCS Istituto Auxologico Italiano, Milan, Italy, 5Division of Rheumatology, Gaetano Pini Institute, Department of Clinical Sciences and Community Health, University of Milan, Milan, Italy, 6INOVA Diagnostics, San Diego, CA, 7Division of Rheumatology, Gaetano Pini Institute, Milano, Italy. Background/Purpose: Anti-␤2glycoprotein I antibodies (a-␤2GPI) are involved in the pathogenesis of the Antiphospholipid Syndrome (APS). Antibodies to the domain 1 of ␤2GP1 (a-␤2GPI-D1) have been suggested as a marker of thrombosis, while antibodies to domain 4/5 (a-␤2GPI-D4/5) have been described in non-thrombotic, non-autoimmune conditions. The clinical significance of these biomarkers is still evolving. We analyze different groups of patients affected by systemic autoimmune diseases and asymptomatic carriers of anti-phospholipid antibodies (aPL). Methods: 154 patients with positive IgG a-␤2GPI at routine assay were enrolled in this multicenter study: A) 55 thrombotic (⫹/⫺ obstetric) primary APS (PAPS); B) 31 women with pure obstetric PAPS; C) 42 Systemic Lupus Erythematosus (SLE)/Undifferentiated Connective Tissue Disease (UCTD) with positive aPL but without any APS event in their medical history; D) 14 asymptomatic aPL carriers; E) 12 Rheumatoid Arthritis (RA). As controls, 99 adult normal healthy donors (NHD) were included. IgG a-␤2GPI-D1 and IgG a-␤2GPI-D4/5 were tested on research ELISAs containing recombinant ␤2GPI domain antigens. The results were considered as optical density (OD) values. Results: Table shows median OD values in each group. To define the preferential specificity of a-␤2GPI, the ratio between a-␤2GPI-D1 and a-␤2GPI-D4/5 was calculated. A p value less than 0.05 was considered as significant (ns⫽not significant).

Thrombotic PAPS (n⫽55) Obstetric PAPS (n⫽31) SLE/UCTD with aPL (n⫽42) aPL carriers (n⫽14) RA (n⫽12) NHD (n⫽99)

IgG a-␤2GPI (routine assay) 1.324

IgG a-␤2GPID1 0.458

IgG a-␤2GPID4/5 0.171

Mann-Whitney Test a-D1 vs a-D4/5 p⬍0.0001

D1–D4/5 RATIO (median) 2.3

1.058

0.312

0.154

p⫽0.0003

2.4

0.953

0.497

0.204

p⬍0.0001

2.1

1.030

0.317

0.228

p⫽ns

1.2

0.917 0.041

0.202 0.109

0.519 0.080

p⫽ns p⫽ns

1.0 1.4

Assay ␤2GPI ␤2GPI-D1 ␤2GPI ␤2GPI-D1 ␤2GPI ␤2GPI-D1

Cut-off 20 CU 20 CU 50.9 CU 7.3 CU 91.1 CU 11.1 CU

Sensitivity (95% Confidence interval) 72.8 (66.6–78.4) 50.2 (43.6–56.8) 62.1 (55.6–68.4) 60.0 (53.4–66.3) 57.4 (50.9–63.9) 56.2 (49.6–62.6)

Specificity (95% Confidence interval) 83.7 (79.7–87.3) 99.2 (97.8–99.8) 96.9 (94.6–98.4) 96.9 (94.6–98.4) 99.0 (97.4–99.7) 99.0 (97.4–99.7)

LRⴙ/LRⴚ at cut-off 4.5/0.33 64.8/0.50 20.0/0.39 19.4/0.41 55.6/0.43 54.3/0.44

When APS patients were stratified according to the history of thrombosis, anti-␤2GPI-D1 and anti-␤2GPI antibodies showed sensitivities/specificities of 37.9%/93.5% and 60.3%/67.7% respectively. At high specificity (96.8%), the sensitivity for thrombosis was significantly higher for ␤2GPI-D1 than for ␤2GPI (22.4% vs. 12.1%; p⬍0.05). Assay

␤2GPI ␤2GPI-D1 ␤2GPI ␤2GPI-D1

S4

Cut-off

Sensitivity (95% Confidence interval)

Specificity (95% Confidence interval)

LRⴙ/LRⴚ at cut-off

20 CU 20 CU 4868 CU 286.9 CU

60.3 (46.6–73.0) 37.9 (25.5–51.6) 12.1 (5.0–23.3) 22.4 (12.5–35.3)

67.7 (48.6–83.3) 93.5 (78.6–99.2) 96.8 (83.3–99.9) 96.8 (83.3–99.9)

1.87/0.59 5.88/0.66 3.74/0.91 6.95/0.80

Background/Purpose: Criteria laboratory tests for antiphospholipid syndrome (APS) include lupus anticoagulant (LAC) as well as IgG and IgM antibodies for cardiolipin and beta2 glycoprotein. Of these tests, only the LAC strongly predicts thrombosis. A recent study (PROMISSE) also showed that LAC was the primary predictor of adverse pregnancy outcome after 12 weeks gestation in aPL(antiphospholipid antibody)-associated pregnancies. Antibodies to phosphatidylserine-prothrombin complex (aPS/PT) have been suggested to strongly correlate with the presence of LAC. Anti-PS/PT antibodies are detected with immunoassay, which is more likely to be amenable to standardization than LAC testing, and can be performed on patients during anticoagulation therapy. Using the PROMISSE cohort we investigated if aPS/PT, similarly to LAC, can predict pregnancy outcomes. Methods: A total of 97 samples from the PROMISSE study were analyzed at the APLS Laboratory using QUANTA Lite® PS/PT IgG and IgM ELISA kits (INOVA Diagnostics). LAC test results were obtained from PROMISSE investigators. Qualitative agreements, kappa scores, and univariate and multivariable logistic regression analyses were done with SPSS software. Results: The prevalence of LAC positivity was 53/97 (54.6%), aPS/PT IgG was 35/97 (36.1%), aPS/PT IgM was 51/97 (52.6%) and any aPS/PT was 64/97 (66.0%). The inter-assay agreement between the aPS/PT and LAC is shown in Table 1. Both aPS/PT IgG and IgM showed significant correlation to LAC (p⬍0.001 and p⫽0.015). Anti-PS/PT IgG was significantly associated with any pregnancy morbidity (p⫽0.002; OR⫽6.2, 95% CI 1.7–22.8) and any pregnancy losses (p⫽0.006; OR⫽4.3, 95% CI 1.5–12.7). The correlation with preeclampsia did not reach significance (p⫽0.061; OR⫽3.4, 95% CI 1.0–11.3). Anti-PS/PT IgM was not correlated with adverse pregnancy outcomes. In a multivariable model, anti-PS/PT IgG remained a significant predictor of pregnancy morbidity and pregnancy losses.

Disclosure: N. Zohoury, Inova Diagnostics, Inc., 3; M. A. Khamashta, Inova Diagnostics, Inc., 5; T. Atsumi, None; J. Musial, None; T. Watanabe, None; M. Papp, None; C. Gonza´lez-Rodrı´guez, None; R. Albesa, Inova Diagnostics, Inc., 3; G. L. Norman, Inova Diagnostics, Inc., 3; P. L. Meroni, Inova Diagnostics, Inc., 5; M. Mahler, Inova Diagnostics, Inc., 3.

11 Catastrophic Antiphospholipid Syndrome (CAPS): Clinical, Immunologic Features, and Outcome Of 441 Patients From The “CAPS Registry”. Ignasi Rodriguez1, Gerard Espinosa2 and Ricard Cervera3. 1Fellow, Barcelona, Spain, 2Department of Autoimmune Diseases, Hospital Clıńic, Barcelona, Barcelona, Spain, 3Hospital Clinic, Barcelona, Spain. Background/Purpose: To analyze the clinical and laboratory features as well as the precipitating factors, treatment and outcome of patients with catastrophic antiphospholipid syndrome (CAPS). Methods: We analyzed the clinical and serologic features of the patients included on May 31, 2013, in the “CAPS Registry”, a web-site based international registry of patients with this condition. Categorical variables are presented as frequencies and continuous variables are presented as mean ⫾ standard deviation. Comparative analysis was performed with chi-square test for categorical variables and t-test for quantitative variables using nonparametric tests when parametric tests were not applicable. Results: The entire series includes 441 patients (456 episodes). Thirteen (2.9%) patients recurred twice and 2 recurred three times. Three hundred (68%) patients were female and 136 (30.8%) were male. They had a mean age of 38.4 ⫾ 17.0 years (range, 0–85). Male patients were older than female (43.4 years vs. 36.2 years; p⬍0.001). The main underlying conditions were primary APS (59.4%), systemic lupus erythematosus (SLE) (30.4%), and lupus-like disease (3.4%). The catastrophic episode was the first manifestation of the APS in 223 (50.6 %) patients. A precipitating factor was reported in 286 (62.7%) episodes, including infections (30.3%), malignancy (10.7%), surgery (10.5), oral contraceptive pills (7.2%), drugs (4.4%), cesarean section (4.2%), lupus flare (2.0%) and trauma (1.1%). A variety of thrombotic manifestations involving the majority of organs were recorded, including renal (73.6%), lung (58.4%), cerebral (56.3%), heart (49.3%), hepatic (36.3%), gastrointestinal (24%), splenic (16.3%), adrenal (10.4%), and pancreatic (7.1%) manifestations. Patients with APS associated with SLE had more episodes of livedo reticularis (29.1% vs. 15% p⫽0.001), heart valve lesions (22.1% vs. 10.4%; p⫽0.01), seizures (12.8% vs. 6.9%; p⫽0.042), pancreatitis (13.4% vs. 4.4%; p⫽0.001) and more frequently exhibited thrombocytopenia (72.4% vs. 62.5%; p⫽0.05). No differences in clinical manifestations were found between male and female patients. Regarding immunologic features, IgG anticardiolipin antibodies (aCL) were positive in 81.6% of the patients, lupus anticoagulant in 81.3%, IgM aCL in 48.6%, IgM anti-␤2GPI in 3.2% and IgG anti-␤2GPI in 11.1%. Female patients had a higher frequency of hemolysis and schistocytes. One hundred and seventy-five (38.6%) patients died at the time of the CAPS event. Conclusion: The CAPS is an uncommon but potentially life-threatening condition that needs a high clinical awareness. CAPS may affect any organ of the body and display a broad spectrum of manifestations.

Table 1. Inter-assay agreement and association of aPS/PT with LAC Total/Positive/Negative agreement (Kappa) PS/PT IgG PS/PT IgM ANY PS/PT

67.0/63.6/69.8 (0.357) 62.9/65.4/60.0 (0.254) 68.0/73.5/56.7 (0.341)

Chi squareP value OR (95% CI) P⫽0.000 5.920 (2.241–15.637) P⫽0.015 2.842 (1.243–6.496) P⫽0.001 4.710 (1.901–11.670)

Conclusion: Anti-PS/PT IgG antibodies show good correlation to LAC and association with pregnancy complications in APS patients. This suggests that aPS/PT may be a promising biomarker for obstetric risk assessment in APS given its advantages with respect to clinical applicability and standardization. Disclosure: R. Willis, None; A. E. Tebo, None; G. Lakos, Inova Diagnostics, Inc., 3; M. Mahler, Inova Diagnostics, Inc., 3; G. L. Norman, Inova Diagnostics, Inc., 3; W. D. Branch, UCB, 5; J. Salmon, None; M. M. Guerra, None; S. S. Pierangeli, None.

13 A MORE Specific Immunoassay For The Diagnosis Of APS. Claudia Grossi1, Maria Borghi1, Elizabeth Papalardo2, Silvia S. Pierangeli3 and Pier Luigi Meroni4. 1Lab of immunology, IRCCS Istituto Auxologico Italiano, Milano, Italy, 2Louisville APL Diagnostics, Inc, Seabrook, TX, 3University of Texas Medical Branch, Galveston, TX, 4Division of Rheumatology, Gaetano Pini Institute, Milan, Italy, Milan, Italy. Background/Purpose: Anticardiolipin (aCL) antibody assays are sensitive for the detection of antiphospholipid antibodies (aPL) in patients with the Antiphospholipid syndrome (APS) but are known to have suboptimal clinical specificity due to cross-reaction with infectious diseases and other non-APS sera. The APhL ELISA® is an immunoassay that detects antibodies to negatively charged phospholipids in the presence of ␤2GPI and has been shown to be more specific than the aCL in the diagnosis of APS. The objective of this study was to evaluate and compare the clinical sensitivity and specificity of the APhL ELISA® with two aCL Methods: Serum samples from confirmed primary and secondary APS patients (n⫽54), syphilis (n⫽20), infectious diseases (n⫽30), hyperglobulinemias (n⫽15), paraproteinemias (n⫽15) and healthy controls (n⫽54) were evaluated for IgG and IgM aPL antibodies in the APhL ELISA® (Louisville APL Diagnostics) and in two aCL assays (Louisville APL Diagnostics, Inc; LAPL) and in a laboratory developed test (LDT). Samples were obtained from Prof Meroni’s laboratory and kits were provided at no cost by the

Disclosure: I. Rodriguez, None; G. Espinosa, None; R. Cervera, None.

12 Phosphatidiyserine-Prothrombin Complex (aPS/PT) IgG Antibodies Correlate With Lupus Anticoagulant and Specific Pregnancy Complications In Patients With Antiphospholipid Syndrome. Rohan Willis1, Anne E. Tebo2, Gabriella Lakos3, Michael Mahler3, Gary L. Norman3, Ware D. Branch4, Jane Salmon5, Marta M. Guerra5 and Silvia S. Pierangeli1. 1University of Texas Medical Branch, Galveston, TX, 2University of Utah, Salt Lake City, UT, 3INOVA Diagnostics, San Diego, CA, 4Univ of Utah, Salt Lake City, UT, 5Hospital for Special Surgery, New York, NY.

S5

Sunday, October 27

Conclusion: Anti-␤2GPI-D1 and anti-␤2GPI antibodies show similar performance characteristics in differentiation of APS patients and controls. The correlation with history of thrombosis appears to be stronger for anti-␤2GPI-D1 compared to anti-␤2GPI-antibodies. Therefore, anti␤2GPI-D1 antibodies are a promising biomarker to aid in the diagnosis and risk assessment of APS patients, but further studies are needed to verify those preliminary findings.

Sunday, October 27

manufacturer. Assays were considered positive when above the stated cut-off for each test. Clinical sensitivities, specificities, positive predictive (PPV) and negative predictive values (NPV) were calculated individually by isotype or combined (IgG and IgM) for each assay. Results:

Clin Sens % Clin Spec % PPV % NPV %

APhL ELISA® IgG

APhL ELISA® IgM

APhL ELISA® IgG and IgM

aCL LAPL IgG

aCL LAPL IgM

aCL LAPL (IgG and IgM)

aCL LDT IgG

aCL LDT IgM

aCL LDT IgG and IgM

83.0 95.5 88.0 92.0

92.5 95.5 89.0 96.9

94 91 81.0 97.6

83.3 77.6 56.9 91.7

90.7 59.7 52.6 94.1

96 52.0 44.0 97.2

73.5 73.1 52.0 87.5

39.6 77.6 42.0 76.4

81.0 61.0 45.2 91.1

15 Establishment Of Standardized International Units For IgM ANTI␤2glycoprotein Antibody Measurement. Rohan Willis1, Claudia Grossi2, Maria Borghi2, Pier Luigi Meroni3, Gabriella Lakos4, Tammy Buckner5, Fernando S. Cavalcanti6, Maria Crisostomo7, Corina Dima8, Kerrie Jaskal9, Matthias Kast10, Luis R. Lopez11, Nina Olschowka10, Sarah Paul12, Tony Prestigiacomo7, Josep Puig6, Wendy Vandam7, Alfredo Villarreal12, Roger Walker7, Mike Watkins12 and Silvia S. Pierangeli1. 1University of Texas Medical Branch, Galveston, TX, 2Lab of immunology, IRCCS Istituto Auxologico Italiano, Milano, Italy, 3Division of Rheumatology, Gaetano Pini Institute, Milan, Italy, Milan, Italy, 4INOVA Diagnostics, San Diego, CA, 5 Corgenix, Broomfield, CO, 6Biokit, Barcelona, Spain, 7Bio-Rad Laboratories, Hercules, CA, 8Theratest Laboratories Inc, Lombard, IL, 9Instrumentation Laboratories, Bedford, MA, 10Phadia Thermofisher, Freiburg, Germany, 11 Corgenix Inc, Broomfield, CO, 12Bio-Rad Laboratories, Benicia, CA.

Conclusion: The APhL ELISA® showed better PPV, NPV, combined sensitivities and specificities whether analyzed for IgG or IgM independently or combined, when compared to the two aCL assays. The APhL ELISA® may be used as a first line of testing and certainly could be recommended as a confirmatory assay for the diagnosis of APS.

Background/Purpose: Recurrent IgG or IgM anti-␤2glycoprotein (ab2GPI) antibody positivity is a key laboratory indicator for classification of antiphospholipid syndrome (APS). Considerable inter-laboratory variation still exists for both isotyopes hindering efforts at standardization. At the 13th International Congress on Antiphospholipid Antibodies in 2010, the “Criteria aPL Testing” task force recommended both the development of international units (IU) and of suitable polyclonal and monoclonal reference material (RM) for a␤2GPI measurement. As such, we sought to prepare polyclonal RM and establish IU for the measurement for IgM a␤2GPI antibodies. Methods: IgM a␤2GPI fractions were affinity-purified (AP) from the sera of 3 APS patients using a combination of ion exchange and affinity chromatography. Purity was confirmed using SDS-PAGE; high-activity fractions were pooled, concentrated and sterile-filtered. The Bradford assay was used to assign IU values (1 IU/ml ⫽ 1␮g/ml AP- a␤2GPI). The RM serum was assigned an IU value using the AP- ab2GPI material and sent with 30 samples to six commercial companies (INOVA, Bio-Rad, Corgenix, Phadia, Human and Instrumentation Laboratory) for testing in their kits (eight total) according to an approved protocol to evaluate linearity, unit equivalency and commutability (CLSI guidelines EP14-A2 and C53-A). Results: The pooled AP material had a protein concentration of 15.125 ␮g/ml and was assigned a value of 15 IgM a␤2GPI IU/ml. The RM had a value of 220.3 IgM a␤2GPI IU/ml. The linearity (R2) of the RP curve for the various assays ranged from 0.9649 to 0.9996. The value of the RM in the various arbitrary kit units ranged from 71.6 to 568 units. Commutability among the different assays was confirmed as demonstrated in figure 1 (representative of all assays) in which dilutions of the IgM RM fell within the 95% prediction interval limits of the regression curve created by the 30 native serum samples. Commutability was also demonstrated using principal components analysis.

Disclosure: C. Grossi, None; M. Borghi, None; E. Papalardo, Louisville APL Diagnostics, 3; S. S. Pierangeli, Louisville APL Diagnostics, Inc, 4; P. L. Meroni, None.

14 Isotype Dependent Performance Of Beta2glycoprotein I Immunoassays In Two Diverse Patient Cohorts: Implications For Assay Harmonization and Standardization. Anne E. Tebo1, Rohan Willis2, Troy Jaskowski3, Jane E. Salmon4, Michelle Petri5, Ware D. Branch6 and Silvia S. Pierangeli2. 1 University of Utah, Salt Lake City, UT, 2University of Texas Medical Branch, Galveston, TX, 3ARUP Laboratories, Salt Lake City, UT, 4Hospital for Special Surgery, New York, NY, 5Johns Hopkins University School of Medicine, Baltimore, MD, 6Univ of Utah, Salt Lake City, UT. Background/Purpose: The presence IgG and/or IgM beta2 glycoprotein I (␤2GPI) antibodies are associated with thrombosis and/or pregnancy-related morbidity in antiphospholipid syndrome (APS) and/or systemic lupus erythematosus (SLE). Some reports show that IgA ␤2GPI antibodies may be of clinical relevance in certain patient groups. There are, however, concerns regarding the diagnostic performance of these assays probably due to lack of international calibration materials amongst other factors. This study was designed to determine isotype-specific prevalence and correlation between four different a␤2GPI immunoassays in two cohorts of patients. Methods: IgG, IgM, and IgA ␤2GPI antibodies were determined in 97 pregnant women with positive antiphospholipid (aPL) antibodies (PROMISSE cohort) and 204 SLE patients (‘HOPKINS’ cohort) with 4 commercial kits (Bio-Rad, Corgenix, INOVA, Phadia). Results were expressed in kitspecific arbitrary units (AU) for IgG, IgM, and IgA as well as in the recently established international consensus units (IU) for IgG and IgM isotypes. Isotype-specific prevalence, Kappa and Spearman’s Rho correlation coefficients were calculated using Analyse-it™. Results: The positivity rates of the a␤2GPI tests ranged from 1.0–10.2% and 40.6–63.2% (IgG); 3.9–6.3% and 32.3–47.4% (IgM); and 10.7–20.0% and 18.8–85.3% (IgA) in the HOPKINS and PROMISSE cohorts, respectively. The overall agreement between any two assays ranged from 92.2– 99.6% (IgG), 95.4–98.8% (IgM) and 77.6–92.2% (IgA) in both cohorts. While the Kappa coefficients (k) showed moderate-to-almost-perfect agreement for IgG and IgM (0.54–0.98), the analysis revealed fair-to-substantial correlations for IgA ␤2GPI tests (0.24–0.75). Despite differences in the positivity rates and varying agreements, good quantitative Rho Spearman’s correlation was observed for the IgG and IgM in all 4 kits. Rho correlations were significantly improved when results of the IgG and IgM ␤2GPI determinations were expressed in IU. However, suboptimal correlations were obtained for the IgA ␤2GPI assays, with better agreement observed between the Phadia and INOVA kits and between the Bio-Rad and Corgenix assays, respectively. Conclusion: Overall, our study demonstrates good qualitative agreements between these immunoassays for the determination of IgG and IgM antibody isotypes. The use of consensus IU clearly indicates an improvement in the harmonization of the IgG and IgM ␤2GPI. Further standardization of the IgA ␤2GPI assays is warranted.

Table. Performance of IgM RM in various IgM a␤2GPI assays Kit

Method

Cut-off

Linearity

RM in kit units

Bio-Rad Bio-plex Bio-Rad ELISA Corgenix ELISA IL ACUSTAR INOVA ELISA INOVA Bio-Flash Phadia Human GmbH

ELISA Multiplex ELISA Chemiluminescence ELISA Chemiluminescence Immunofluoroescence ELISA

⬍20 M units ⬍20 M units ⬍20 M units ⬍10 U/ml ⬍20 SMU ⬍10 U/ml ⬍10 U/ml ⬍7 U/ml

R2: 0.99 R2: 0.96 R2: 0. 98 R2: 0.99 R2: 0.99 R2: 0.99 R2: 0.99 R2: 0.99

93.8 M Units 96.4 M Units 77.4 M Units 69.1 U/ml 98.4 SMU 72.5 CU 568 U/ml 71.6 U/ml

Conclusion: The results of linearity and commutability studies suggest that this material can be used as a calibrant/reference in a wide array of assays of different formats and new international consensus are now available for IgM-a␤2GPI measurement. These studies contribute significantly to the much-needed standardization of a␤2GPI immunoassays and further characterization by international bodies (IRMM/IFCC) will be done Performance of IgM RM in various IgM ab2GPI assays Disclosure: R. Willis, None; C. Grossi, None; M. Borghi, None; P. L. Meroni, None; G. Lakos, Inova Diagnostics, Inc., 3; T. Buckner, Corgenix, 3; F. S. Cavalcanti, Biokit, 3; M. Crisostomo, Bio-Rad, 3; C. Dima, TheraTest Laboratories, 3; K. Jaskal, Instrumentation Laboratories, 3; M. Kast, Phadia, 3; L. R. Lopez, None; N. Olschowka, Phadia, 3; S. Paul, Bio-Iad, 3; T. Prestigiacomo, Bio-Rad, 3; J. Puig, Bio-Kit, 3; W. Vandam, Bio-Rad, 3; A. Villarreal, Bio-Rad, 3; R. Walker, Bio-Rad, 3; M. Watkins, Bio-Rad, 3; S. S. Pierangeli, None.

Disclosure: A. E. Tebo, None; R. Willis, None; T. Jaskowski, None; J. E. Salmon, Biogen Idec, BMS, J and J, Merck, Quest, 1, Roche Pharmaceuticals, 2, Alexion, Regeneron, Movartis, BMS, 5, Alliance for Lupus Research, RRF, Kunkel Society, 6; M. Petri, None; W. D. Branch, UCB, 5; S. S. Pierangeli, None.

S6

IgM

Only IgG and IgA ␤2glycoprotein I Antibody Isotypes Are Associated With Venous Thrombosis In Systemic Lupus Erythematosus. Anne Tebo1, Rohan Willis2, Troy Jaskowski3, Laurence S. Magder4, Silvia S. Pierangeli2, Ware Branch5 and Michelle Petri6. 1University of Utah, Salt Lake City, UT, 2University of Texas Medical Branch, Galveston, TX, 3 ARUP Laboratories, Salt Lake City, UT, 4University of Maryland, Baltimore, MD, 5Univ of Utah, Salt Lake City, UT, 6Johns Hopkins University School of Medicine, Baltimore, MD.

IgA

Any Assay

Background/Purpose: The current APS Classification Criteria recommend testing for IgG and IgM antibodies for ␤2glycoprotein I (␤2GPI), and do not differentiate between primary and secondary APS. This study was designed to investigate the clinical performance of different ␤2GPI IgG, IgM and IgA antibody tests in a SLE with arterial and/or venous thrombosis. Methods: IgG, IgM, and IgA anti-␤2GPI antibodies were determined in 200 patients from a lupus cohort with 3 ELISA (Bio-Rad, Corgenix and INOVA) and 1 fluoro-enzyme immunoassay commercial kits (Phadia), following the manufacturers’ instructions. Kits were provided at no cost by the manufacturers, and assays were performed at ARUP laboratories. The degree of agreement between the different kits was quantified using a Kappa statistic. Univariate and multivariate analyses were performed to assess the association between kit results and lifetime history of venous or arterial thrombosis. Results: The pairwise agreement between the different kits was high for IgG and IgM with Kappa coefficients ranging from 0.64 to 0.98 and 0.74 to 0.93 respectively. This measure was somewhat lower for IgA with Kappa coefficients ranging from 0.24 to 0.78. Twenty-seven (14%) had a history of venous thrombosis. For all assays and all kits, those with a positive assay result were more likely to have a history of venous thrombosis than those with a negative assay result. These differences reached statistical significance in univariate and multivariate analyses of IgG and IgA, adjusted for age, race, and anti-htn medications. The observed differences were not statistically significant for IgM, in part due to the relatively small number of patients positive for IgM. Thirty-four patients (17 %) had a history of arterial thrombosis. There was not strong evidence of an association between the assay results and arterial thrombosis.

Assay IgG

IgM

IgA

Company Phadia Bio-Rad Corgenix INOVA Any Phadia Bio-Rad Corgenix INOVA Any Phadia Bio-Rad Corgenix INOVA Any

Any Assay

6/20 (30%) 4/9 (44%) 4/10 (40%) 2/2 (100%) 6/22 (27%) 3/10 (30%) 4/10 (40%) 4/13 (31%) 3/8 (38%) 5/18 (28%) 9/36 (25%) 8/29 (28%) 9/41 (22%) 7/21 (33%) 12/56 (21%) 13/65 (20%)

21/180 (12%) 23/191 (12%) 23/190 (12%) 25/198 (13%) 21/178 (12%) 24/190 (13%) 23/190 (12%) 23/187 (12%) 24/192 (13%) 22/182 (12%) 18/164 (11%) 19/171 (11%) 18/159 (11%) 20/179 (11%) 15/144 (10%) 14/135 (10%)

P-value1

Adjusted P-value2

0.04 0.02 0.03 0.02 0.09 0.10 0.03 0.080 0.08 0.08 0.03 0.03 0.10 0.01 0.06 0.08

0.015 0.016 0.019 1.0 0.025 0.29 0.097 0.14 0.23 0.17 0.0055 0.031 0.037 0.0049 0.0080 0.021

Disclosure: A. Tebo, None; R. Willis, None; T. Jaskowski, None; L. S. Magder, None; S. S. Pierangeli, None; W. Branch, None; M. Petri, None.

17 Autoantibodies Against Component Of Complement One Contribute To The Complement Activation and Clinical Manifestation Of Antiphospholipid Syndrome (APS) Especially In Refractory Cases. Kenji Oku1, Olga Amengual1, Ikuma Nakagawa1, Toshiyuki Watanabe2, Yusaku Kanetsuka1, Toshiyuki Bohgaki1, Tetsuya Horita1, Shinsuke Yasuda1 and Tatsuya Atsumi1. 1Hokkaido University Graduate School of Medicine, Sapporo, Japan, 2Hokkaido University Graduate School of Medicine, Hokkaido, Japan. Background/Purpose: In the pathogenic mechanisms of antiphospholipid syndrome (APS), it is recognized that pathogenicity of antiphospholipid antibodies (aPL) has dominant effects. Complement is the part of the innate immune system and is one of the main effector mechanisms of antibodymediated immunity. We have previously reported that complement activation prevalently coexists in sera of antiphospholipid syndrome (APS) patients and functions as source of procoagulant cells activation. Recently, autoantibodies against C1q, the component of complement 1, were reported to correlate with complement activation in patients with systemic lupus erythematosus (SLE). They are not neutralizing antibodies but target the neoepitopes of deformed C1q bound to various molecules such as anionic phospholipids. The binding of anti-C1q antibodies to C1q induces accelerated activation of complement pathway. There are no previous studies discussing the involvement of anti-C1q antibodies in APS patients. The purpose of this study is to investigate the existence and significance of anti-C1q antibodies in APS patients. Methods: This study was comprised of 40 consecutive primary APS patients that visited Hokkaido University Hospital rheumatology clinic from 2002 to 2013 that had more than 2 years history of APS. Informed consent was obtained from every patients and the study was approved by ethics committee of the Hokkaido University Hospital. All the patients were retrospectively analyzed of their clinical manifestations and laboratory data. Ten patients had refractory APS defined as a clinical status of relapsing thrombosis or pregnancy morbidity during adequate secondary prophylaxis. Twenty non-SLE connective tissue disease patients without APS and 20 healthy control subjects were also included. An enzyme-linked immunosorbent assays (ELISA) were used to measure serum levels of anti-C1q antibody titers and anaphylatoxins (C3a,C4a). Results: Anti-C1q antibodies were more frequently detected in primary APS patients (14/40) than in non-SLE connective tissue disease patients (2/20) (p⬍0.01). In APS patients, anti-C1q antibodies titers were significantly correlated with serum C4a levels (p⫽0.013) and showed tendency of inverse correlation with serum C3a levels (p⫽0.07). The prevalence or the titers of anti-C1q antibodies were not associated with any of the specific clinical manifestations of APS or titers of aPLs. However, refractory APS patients, compared with APS patients without flare, had higher positivity (9/10 vs 3/20, p⫽0.01) and higher titers of anti-C1q antibodies (32.9⫹/⫺ 6.95 vs 10.9⫹/ ⫺1.89, p⫽0.01). Using a cut-off level of 20, the hazard ratio of the anti-C1q positivity for the reoccurrences of the APS manifestation during adequate secondary prophylaxis were 80.0 (95%CI 7.45–880, p⫽0.000007).

Table 2. Proportion (%) with a history of Arterial Thrombosis by antibody status

IgG

Phadia Bio-Rad Corgenix INOVA Any

2/20 (10%) 2/9 (22%) 2/10 (20%) 0/2 (0%) 3/22 (14%)

32/180 (18%) 32/191 (17%) 32/190 (17%) 34/198 (17%) 31/178 (17%)

P-value1

Adjusted P-value2

0.50 0.70 0.70 1.0 1.0

0.40 0.70 0.80 Too few 0.70

0.90 0.20 0.30 0.30 0.90 0.20 0.40 0.50 0.20 0.80 0.80

Conclusion: Only IgG and IgA isotypes were associated with venous thrombosis in SLE, using the p-value adjusted for age and ethnicity. Although the IgA kits showed variable agreements, our data indicate that testing for ␤2 GPI IgA isotype may be of clinical relevance in the evaluation of venous thrombosis in APS associated with SLE.

Fisher’s Exact Test Based on Logistic Regression, controlling for age and race, and anti-htn medications.

Company

0.70 0.070 0.20 0.10 0.50 0.20 0.80 0.80 0.20 0.84 0.70

Fisher’s Exact Test. Based on Logistic Regression, controlling for age, race, and anti-htn medications.

2

Assay

32/190 (17%) 30/190 (16%) 30/187 (16%) 31/192 (16%) 30/182 (16%) 25/164 (15%) 30/171 (18%) 28/159 (18%) 28/179 (16%) 24/144 (17%) 22/135 (16%)

2

1

Proportion(%) with history of Arterial Thrombosis (nⴝ34) AntibodyAntibodypositive negative

2/10 (20%) 4/10 (40%) 4/13 (31%) 3/8 (38%) 4/18 (22%) 9/36 (25%) 4/29 (14%) 6/41 (15%) 6/21 (29%) 10/56 (18%) 12/65 (18%)

1

Table 1. Proportion(%) with a history of Venous Thrombosis by antibody status. Proportion(%) with history of Venous Thrombosis (nⴝ27) AntibodyAntibodypositive negative

Phadia Bio-Rad Corgenix INOVA Any Phadia Bio-Rad Corgenix INOVA Any

S7

Sunday, October 27

16

Sunday, October 27

Conclusion: These findings indicate that anti-C1q antibodies are associated with complement activation in APS and may contribute to the manifestation especially in the refractory cases.

examined the association aPL antibodies detected in the APhL ELISA with clinical manifestations of APS in two lupus cohorts. Methods: A total of 590 serum samples from patients with lupus (diagnosed according to ACR criteria) from the HOPKINS cohort (n⫽543) and from the University of the West Indies Rheumatology Department Jamaica cohort (n⫽47) were analyzed for IgG and IgM aPL antibodies using the APhL ELISA (Louisville APL Diagnostics), following the manufacturer’s instructions. Samples were considered positive when above 15 GPL or MPL units for IgG and IgM, respectively. Association of venous thrombois, arterial thrombosis, any thrombosis, diagnosis of APS (Sapporo revised criteria), any miscarriage, any pregnancy morbidity or toxemia with a positive APhL test (either IgG or IgM) was assessed by univariate analysis, using the SPSS® v20.0 software. Results:

Disclosure: K. Oku, None; O. Amengual, None; I. Nakagawa, None; T. Watanabe, None; Y. Kanetsuka, None; T. Bohgaki, None; T. Horita, None; S. Yasuda, None; T. Atsumi, None.

18 Clinical Correlates Of Positive Anti-Cardiolipin and ␤-2 Glycoprotein 1 Antibodies In a Cohort Of 110 Patients At Mayo Clinic. Uma Thanarajasingam, Cynthia S. Crowson, Melissa R. Snyder, Rajeev Pruthi, Harvinder S. Luthra and Kevin G. Moder. Mayo Clinic, Rochester, MN. Background/Purpose: Testing for antiphospholipid antibodies is typically indicated for diagnosis and prognosis of antiphospholipid syndromes, however, selected patients undergo testing for other reasons. The clinical significance of positive antibodies in this subgroup of patients is unclear. In this study, we sought to identify patient characteristics and events associated with positive anti-cardiolipin (aCL) and anti-beta-2 glycoprotein 1 (antiB2GP1) antibodies in a large cohort of patients. The patients with positive results were compared to age and gender -matched controls who had tested negative for the same antiphospholipid antibodies. Methods: We included 110 adults and 110 age and gender-matched controls tested for aCL and anti-B2GP1 antibodies within the same calendar year (2009) at a large academic medical institution. Clinical and laboratory data was abstracted from 1998 through 2013. Results: Of the 110 cases, the mean age was 54.1 years and 76 (69.1%) were female. The majority of patients (84) were positive for aCL-IgG (41), IgM (25) or both (18); 29 were additionally tested for anti-B2GP1-IgG (14 positive)and IgM (15 positive). ANA was positive in 52 patients (61%). Thrombotic events were seen in 64 patients (59%); these included 31 (29%) with DVT/PE, 29 (27%) with stroke/TIA, 18 (22%) with any fetal loss, and 6 (6%) with any peripheral arterial events. Thirty-six patients (33%) met criteria for APS and 23 (21%) had SLE. Compared to controls, those with positive antiphospholipid antibodies were significantly more likely to be ANA positive (p⬍0.003), experience a thrombotic event (p⬍0.018), suffer fetal loss (p⬍0.022) and have SLE (p⬍0.001). There were non-statistically significant trends towards increased prevalence for DVT/PE (p⫽0.067) and peripheral arterial events (p⫽0.055). No statistically significant differences were seen between the cases and controls with respect to total number of pregnancies, oral contraceptive and stroke/TIA. Conclusion: Over one-half of patients who tested positive for antiphospholipid antibodies experienced thrombotic events. When compared to controls, those with positive antiphospholipid antibodies had a statistically significant increase in ANA positivity, any thrombotic event, fetal loss and a diagnosis of SLE. To our knowledge, this is the first study examining the frequency of clinically significant conditions and associated events in a large cohort of patients with positive anti-cardiolipin and anti-beta-2 glycoprotein antibodies as compared to age and sex matched controls with negative antibodies.

IgG APhL

P

OR

95% CI

Venous thrombosis Arterial thrombosis Any thrombosis Diagnosis of APS Any miscarriage Any pregnancy morbidity Toxemia

0.000 0.070 0.000 0.000 0.022 0.007 0.007

3.5 1.8 3.1 5.8 2.2 2.5 3.1

1.9–6.3 1.0–3.5 1.7–5.8 2.7–12.1 1.1–4.1 1.3–4.7 1.4–6.8

Any APhL positive test was significantly associated with venous, arterial, any thrombosis and with diagnosis of APS, but no association of IgM APhL positivity with clinical manifestations was observed. There was no difference in the results when data were calculated combined or individually in each cohort. Conclusion: The APhL ELISA is an excellent test to detect aPL associated with thrombosis and pregnancy morbidity in patients with lupus. Disclosure: S. S. Pierangeli, Louisville APL Diagnostics, Inc, 4; R. Willis, None; M. Petri, Anthera, Gaxo, Medimmune, TEVA, UCB, Pfizer, 2, Anthera, UCB, Genentech,Glaxo,Medimmune, Pfizer, 5; H. Fang, None; M. Smikle, None; K. de Ceulaer, None; E. N. Harris, Louisville APL Diagnostics, Inc, 5.

20 Complement Deposition On Platelets Is Associated To Venous Thrombosis In Systemic Lupus Erythematosus. Christian Lood, Helena Tydeń, Birgitta Gullstrand, Gunnar Sturfelt, Andreas Jo¨nsen, Lennart Truedsson and Anders A. Bengtsson. Lund University, Lund, Sweden. Background/Purpose: Anti-phospholipid (aPL) antibodies are associated with development of venous thrombosis and stroke in the autoimmune disease systemic lupus erythematosus (SLE). The underlying mechanism for aPL-mediated thrombosis is not known but may include activation of platelets and the complement system. The aim of this study was to investigate if aPL antibodies could interact with platelets and support complement activation. Furthermore, we investigated if this mechanism was operating in SLE patients and whether complement deposition on platelets was associated to venous thrombosis and specific for SLE. Methods: Complement deposition was measured by flow cytometry in patients with SLE, rheumatoid arthritis, systemic sclerosis, myocardial infarction and healthy individuals. Associations to cardiovascular disease were analyzed with logistic regression analysis and adjusted for traditional risk factors. Anti-cardiolipin antibodies were used to investigate if aPL antibodies supported platelet activation and complement deposition Results: Platelet deposition of C1q and C4d was increased in SLE as compared to healthy individuals (p⬍0.0001), but high levels were also seen in some patients with rheumatoid arthritis and systemic sclerosis. Complement deposition was clearly associated to venous thrombosis and aPL antibodies in SLE. In vitro, anti-cardiolipin antibodies increased platelet activation and supported C4d deposition on platelets. Conclusion: Platelet C1q and C4d deposition are identified as markers of venous thrombosis in SLE patients. Several mechanisms are operating in SLE to amplify platelet complement deposition, of which anti-cardiolipin antibodies was identified as one important contributor. Further studies are needed to elucidate the role of platelet complement deposition in development of venous thrombosis.

Disclosure: U. Thanarajasingam, None; C. S. Crowson, Genentech, Inc., 2; M. R. Snyder, None; R. Pruthi, None; H. S. Luthra, None; K. G. Moder, None.

19 Association Of Antiphospholipid Antibodied Detected In The Aphl ELISA With Clinical Manifestations Of The Antiphospholipid Syndrome In Two Lupus Cohorts. Silvia S. Pierangeli1, Rohan Willis1, Michelle Petri2, Hong Fang2, Monica Smikle3, Karel de Ceulaer4 and E. Nigel Harris5. 1University of Texas Medical Branch, Galveston, TX, 2Johns Hopkins University School of Medicine, Baltimore, MD, 3University of the West Indies, Kgn 7, Jamaica, 4University of the West Indies, KIngston, Jamaica, 5The University of the West Ind, Kingston, Jamaica. Background/Purpose: Antiphospholipid Syndrome (APS) is characterized by the presence of antibodies to phospholipid (aPL) and to ␤2glycoprotein I (␤2GPI) in patients with thrombosis or pregnancy morbidity. Anticardiolipin (aCL) assays are used to confirm the diagnosis of APS. Although a sensitive assay, aCL is often found positive in a number of infectious, drug-induced and non APS related autoimmune diseases. The APhL ELISA is an immunoassay that detects antibodies to negatively charged phospholipids in the presence of ␤2GPI and has been shown to be more specific and as sensitive as the aCL in the diagnosis of APS. Here we

Disclosure: C. Lood, None; H. Tydeń, None; B. Gullstrand, None; G. Sturfelt, None; A. Jo¨nsen, None; L. Truedsson, None; A. A. Bengtsson, None.

S8

Factor Xa Reactive Antibodies In Patients With Systemic LUPUS Erythematosus and Antiphospholipid Syndrome Have Differential Effects Upon Coagulation Assays and Endothelial CELLS. Bahar ArtimEsen, Natalia Smoktunowicz, Charis Pericleous, Vera M. Ripoll, Ian Mackie, Rachel Chambers, David A. Isenberg, Anisur Rahman, Yiannis Ioannou and Ian Giles. University College London, London, United Kingdom. Background/Purpose: Antiphospholipid antibodies have been shown to bind serine proteases (SP) involved in the coagulation cascade. Previously, we found that IgG anti-Factor(F)Xa antibody levels were higher in patients with the antiphospholipid syndrome (APS) and patients with systemic lupus erythematosus (SLE) but no APS compared with disease and healthy control (HC) groups. Given that FXa has important haemostatic and cellular effects we hypothesized that anti-FXa antibodies may be important in the pathogenesis of APS and SLE. Therefore, we investigated whether differences exist in the avidity and functional effects of APS and SLE anti-FXa IgG upon the coagulant and cellular functions of FXa. Methods: IgG was purified from patients with APS (n⫽15) and SLE (n⫽15) who had medium or high serum levels of anti-FXa binding and HC (n⫽10) negative for anti-FXa IgG. The avidity of IgG-FXa binding was measured under chaotropic conditions using a NaCl gradient. We measured effects of anti-FXa IgG on FXa-activated clotting time (ACT) and on FXa enzymatic activity in a chromogenic assay in the absence and presence of antithrombin (AT)III. Cellular effects of IgG on FXa-protease-activated receptor (PAR) mediated intracellular calcium release in human umbilical vein endothelial cells (HUVEC) were measured using the fluorescent image plate reader (FLIPR). Results: All SLE-IgG displayed significantly lower (less than 25 % binding) avidity compared to APS-IgG (25–70% binding) to FXa at 0.13 to 1M concentrations of NaCl (p⬍0.05). The mean residual binding of APS-IgG to FXa was significantly higher than that of SLE-IgG below 2M NaCl (26 vs 13 %; p⬍0.05 at 1 M). FXa enzymatic activity was significantly reduced by APS-IgG (90%) and SLE-IgG (92 %) compared to HC-IgG (98 %) (APS vs HC p⬍0.05, SLE vs HC p⬍0.05, APS vs SLE p⫽0.04). ATIII mediated inhibition of FXa however, was significantly reduced by APS-IgG (62%) compared with HC (79%) and SLE (81 %) (p⬍0.05). The greatest prolongation of FXa-ACT was observed with APS-IgG followed by SLE-IgG and HC-IgG (74, 63 and 26 sec respectively). In cultured HUVEC, APS IgG caused significantly greater enhancement of FXa-mediated Ca2⫹ flux compared to SLE and HC-IgG at the same concentration (400 ␮g/ml) (APS vs HC p⬍0.05, APS vs SLE p⫽0.03) and this effect was reduced at lower IgG concentration (200 ␮g/ml). Conclusion: FXa reactive IgG isolated from patients with APS displayed higher avidity binding to FXa and had greater functional effects upon FXa activity, ATIII mediated inhibition of FXa and FXa-PAR mediated signalling in cultured endothelial cells than SLE-IgG. Further work is now underway to further characterise the cellular effects of these anti-FXa IgG.

Disclosure: V. Rodriguez-Garcia, None; Y. Ioannou, None; D. A. Isenberg, None; I. Giles, None.

23 Seizures In Systemic Lupus Erythematosus Are Not An Inflammatory Manifestation Of Lupus Anticoagulant. Michelle Petri and Hong Fang. Johns Hopkins University School of Medicine, Baltimore, MD. Background/Purpose: About 50% of SLE patients make an antiphospholipid antibody and 10% develop classic antiphospholipid syndrome. Seizures are one of the components of the SLICC classification criteria, and can occur due to SLE itself or secondary to stroke. Because seizures have been proposed as an inflammatory (non-thrombotic) manifestation of antiphospholipid antibodies, we determined whether seizures in SLE were associated with the lupus anticoagulant. Methods: 2206 SLE patients were enrolled in a prospective cohort. Their mean age was 43 years; 93% were female, 55% Caucasian, 38% African-American and 7% other ethnicity. 27% had a history of lupus anticoagulant, 9% stroke and 10% seizures. The association of seizure with lupus anticoagulant and then excluding those with stroke was determined using chi-squared analyses. The lupus anticoagulant was determined by dRVVT with confirmatory testing. Results: Table One shows the association of seizure with the lupus anticoagulant (p ⫽ 0.0017).

Disclosure: B. Artim-Esen, None; N. Smoktunowicz, None; C. Pericleous, None; V. M. Ripoll, None; I. Mackie, None; R. Chambers, None; D. A. Isenberg, None; A. Rahman, None; Y. Ioannou, None; I. Giles, None.

22

Table 1. Association between Lupus Anticoagulant and Seizure (All Patient Analysis)

What Is The Prevalence Of Non-Criteria Antiphospholipid Antibodies in Patients With Antiphospholipid Syndrome? Veronica Rodriguez-Garcia1, Yiannis Ioannou2, D.A. Isenberg2 and Ian Giles2. 1 Hospital Regional Universitario Carlos Haya, Malaga, Spain, 2University College London, London, United Kingdom.

Seizure

LA Pos (Nⴝ591)

LA Neg (Nⴝ1615)

P-value

13.03%

8.54%

0.0017

Table 2 shows the association of seizure with the lupus anticoagulant, but excluding those with stroke. The p-value was 0.22.

Background/Purpose: Increasing interest has focused upon assays which are not currently included in the Antiphospholipid syndrome (APS) classification criteria to detect antibodies directed against other phospholipids (PL), PL binding proteins, coagulation factors and a mechanistic test for resistance of Annexin A5 (AA5) anticoagulant activity. Therefore, we carried out a systematic review to try and establish the prevalence of each non-criteria assay in APS and control populations. Methods: We searched PubMed and EMBASE using the keywords APS, antiphospholipid antibodies (aPL), non criteria, new assays, aCL, LA, anti-domain (aD)I, Ig A antiB2 GPI, antiphosphatidylserine (aPS), antiphosphatidylethanolamine (aPE), antiphosphatidic acid (aPA), antiprothrombin (aPT), antiphosphatidylserine-prothtombin (aPS-PT), anticardiolipin/ vimentin (CL/Vm) and AA5 resistance. Each publication was systematically examined.

Table 2. Association between Lupus Anticoagulant and Seizure (Excluding Those with Stroke) Seizure

LA Pos (Nⴝ493)

LA Neg (Nⴝ1515)

P-value

9.33%

7.59%

0.22

Conclusion: The apparent association of the lupus anticoagulant is explained by the association with stroke. Thus, seizures in the absence of stroke should not be attributed to the lupus anticoagulant. Seizures should not be added to the classification criteria for antiphospholipid syndrome. Disclosure: M. Petri, None; H. Fang, None.

S9

Sunday, October 27

Results: We selected 15 (6 retrospective, 1 case-control and 8 crosssectional studies from which we were able to extract original data on prevalence of non-criteria aPL in 1535 patients with APS and 1184 healthy and disease controls. We found the largest APS sample size from 3 studies of aDI, which found a prevalence of 34.26% of aDI in 645 patients with APS patients versus 3.3% in 30 healthy controls (HC). In reducing order of samples size. Three studies found an overall prevalence of 7.8% of IgM and 4.3% IgG aPE in 337 patients with APS compared with 3.9% IgM and 0.9% IgG aPE in 340 HC. IgA aCL were found in 37.46% of APS patients (n⫽262), 37.8% autoimmune disease (n⫽37) controls and 3.4% in healthy and atherosclerosis (n⫽527) controls. In contrast, 55% of 196 patients with APS were positive for IgA anti-B2GPI, versus 32.9% in 382 atherosclerosis and 13% in 145 healthy controls. The prevalence of AA5 resistance from 3 studies was 66.83% in 163 patients with APS compared with 0% in 80 HC. In 132 APS patients, 27.56% were positive for IgM aPT and 36% positive for IgG aPT versus 5% IgG and 0% IgM in HC. All of the remaining studies contained less than 100 patients with APS. Studies of other non-criteria aPL identified: 24% IgM and 65.5% IgG aPA in 67 APS patients versus 0% IgM/G aPA in 104 HC; 36.6% IgM and 50.8% IgG aPS in 89 APS patients versus 1% IgM and 0% IgG aPS in 104 HC; 33.8% IgM and 38.8% IgG aPI in 89 APS patients versus 0% in HC; 25% IgM and 27.5% IgG aPS/PT in 44 APS patients versus 3% IgM and 0% IgG in 138 disease controls; and 80% IgM and 92.4% IgG anti-CL/Vm in 40 APS patients versus 0% IgM/G in 32 HC. Conclusion: We found the highest prevalence of non-criteria aPL in the largest number of patients with APS in studies of AA5 resistance and aDI. Further prospective studies however are urgently required to confirm these findings.

21

25

Sunday, October 27

ACR Poster Session A B cell Function and Targeting in Systemic Lupus Erythematosus Sunday, October 27, 2013, 8:30 AM–4:00 PM

SLE Flares Are Characterized By Generalized Polyclonal Expansions Of Antibody Secreting Cells Without Preference For Autoimmune Responses. H. Travis Ichikawa Emory University, Atlanta, GA. Background/Purpose: Increased circulating antibody secreting cells (ASC), including both CD138- plasmablasts and CD138⫹ plasma cells (PB/PC), correlate with SLE activity and are prominent during Lupus flares. We tested whether this expansion is predominantly driven by typical SLE autoreactivities including anti-dsDNA, -Sm and -Ro and 9G4⫹ antibodies. Methods: Peripheral blood lymphocytes (PBL) were collected from SLE patients (n⫽12; 4 mild-severe flare patients defined by SELENA-SLEDAI) with high serum levels of ⱖ1 autoantibody (anti-dsDNA, anti-Sm or anti-Ro). PB/PC frequency was calculated as % of total B-cells by flow cytometry. Anti-dsDNA, anti-Sm and anti-Ro60, 9G4⫹ and anti-microbial (anti-tetanus toxoid and anti-influenza) IgG ASC were detected by ELISPOT and expressed as % of total IgG ASC. Memory cells were activated by stimulation with R848 and IL-2 for 7 days. Frequencies of antigen specific IgG memory cells were determined by ELISPOT as before. Results: PB/PC frequencies were increased for up to 25-fold in SLE compared to HC and comprised up to 38 % of total B-cells. Of total IgG ASC, frequencies of each SLE specific IgG ASC responses (anti-dsDNA, anti-Sm and anti-Ro60) never exceeded 3.4% (mean ⫾ SD, 0.25 ⫾ 0.65%) in patients with high serum antibody levels. 9G4⫹ ASC, which include several SLEspecific autoreactivities including anti-dsDNA and anti-apoptotic cells, were the most abundant autoreactivity but did not exceed 6% (mean ⫾ SD, 2.2 ⫾ 1.7%). Combined, all the lupus–related autoreactivities accounted for ⬍10% of all IgG ASC. No correlation was found between PB/PC frequencies and SLE specific or 9G4⫹ IgG ASC frequencies. Furthermore, anti-Tetanus and/or anti-influenza ASC, usually not found in healthy PBL, were found in SLE PBL in frequency similar to autoreactive responses (0.06 ⫾ 0.09% and 0.37 ⫾ 0.12%, respectively). Anti-Sm, anti-Ro and 9G4 as well as antitetanus and anti-influenza reactive IgG memory cells were present in frequencies typically higher than that found for ASC of the same antigenic reactivity. Notably, in 5 out of 6 patients with high anti-Ro serum titers, anti-Ro IgG memory cells represented ⬎ 2% of all IgG memory cells. In contrast, anti-dsDNA IgG memory cell frequencies (0.01 ⫾ 0.02%) were as low as anti-tetanus memory cell frequencies (0.12 ⫾ 0.31%). Conclusion: Combined, conventional lupus autoreactivities only account for ⬍10% of the greatly expanded numbers of circulating ASC characteristic of lupus flares. In addition, these frequencies are lower than the frequency of memory cells for the corresponding autoantigens. Moreover, antimicrobial responses are present in circulating ASC with frequencies similar to autoimmune responses. While our studies did not test for other potential SLE autoreactivities, our results are consistent with a polyclonal expansion of ASC during lupus flares that is not predominantly driven by conventional autoantigens even in patients with high titers of serum antibodies. These studies have important implications for our understanding of the mechanisms underlying lupus flares and the contribution of different cellular compartments to the generation of serum autoantibodies at different times in the course of the disease.

24 Plasma Cells In Acute Systemic Lupus Erythematosus Flares Are Characterized By a Highly Diversified Repertoire Accentuated By Clonal Expansions Of VH4-34 Antibodies. Christopher Tipton1, Christopher Fucile2, Alex Rosenberg2, Scott Jenks3, Jennifer Hom1, F. Eun-Hyung Lee1 and Inaki Sanz1. 1Emory University, Atlanta, GA, 2University of Rochester, Rochester, NY, 3Emory University School of Medicine, Atlanta, GA. Background/Purpose: Systemic Lupus Erythematosus (SLE) is an autoimmune disease in which faulty B cell tolerance promotes the generation of multiple autoantibodies of which anti-ds DNA, anti-Sm and VH4–34encoded 9G4 antibodies, are highly specific for this disease. Acute SLE flares are typically accompanied by a substantial expansion of circulating antibodysecreting plasmablasts (PB) and plasma cells (PC). Similar expansions of PB/PC populations in healthy subjects post-immunization are comprised of large clonal expansions of antigen-specific cells derived from pre-existing memory cells. To determine if SLE PB/PC expansions follow the same model, and to understand the diversity, origin and antigenic specificity of SLE PB/PC expansions, we used immunoglobulin heavy chain (IGH) deep sequencing to analyze sorted cells obtained from patients experiencing acute flares. Methods: Peripheral blood lymphocytes were sorted into various B cell populations. Multiple VH family-specific primer sets were used to amplify the IGH gene from extracted RNA and then sequenced using the Illumina MiSeq platform. Sequences were analyzed using internally designed analysis software. Our software conducts quality filtering, detailed mutational analysis and identifies inter-population relationships between samples. Results: PB/PC expansions in SLE acute flares are a highly diversified repertoire with lower VH mutation rates than memory cells. However, within this diverse pool significant clonal expansions accounting for 0.5% or more of the entire repertoire were regularly detected. Strikingly, VH4–34 sequences comprised the largest clones in all SLE samples. Also of interest, a substantial fraction of PB/PC in SLE were highly evolutionarily related to a pool of activated naı¨ve precursors. This was in stark contrast to results obtained in PB/PC samples from healthy, vaccinated subjects, which contained large, clonal expansions, high VH mutation rates, were largely related to memory cells, and had an absence of autoreactive VH4–34 clonal expansions. Figure 1 shows clonal relationships between the top 20% of PB and PC sequences and Naive and Memory clones. Clones are arranged in increasing size from center to outer edge, and lines link matching clones.

Disclosure: H. T. Ichikawa, None;

26 Autoantibodies Directed Against Cell Surface Components In Autoimmune Disease Patients: Proposal Of a Novel ELISA For The Detection Of Anti-Endothelial Cell Antibodies. Keiji Miura1, Ayako Kondo2, Kazuo Takahashi2, Daisuke Hirano2, Yoshiyuki Hiki1, Shunji Yoshida2, Yukio Yuzawa2 and Yoshikazu Kurosawa1. 1Fujita Health Universtiy, Toyoake, Japan, 2Fujita Health Universtiy School of Medicine, Toyoake, Japan. Background/Purpose: Sera from patients with systemic vasculitis or inflammatory conditions have been reported to contain antibodies that bind to endothelial cells (EC), i.e., AECA (anti-endothelial cell antibodies). AECA are known to play immunogenic effects by triggering EC activation and vascular damage, but the immunopathological role of AECA is not clear. SDS-PAGE and Western blotting have previously been used for detecting target antigens of AECA. However, we assumed that these methods are not appropriate for searching genuine target antigens on cell surface, and developed a novel solubilized cell surface protein-capture ELISA (CSPELISA). Methods: Antigens were obtained as cell surface proteins from the plasma membrane of human umbilical vein endothelial cells (HUVEC); these

Conclusion: Combined, our data support a model of generalized naı¨ve and memory activation underlying the activation phase of human SLE. This polyclonal activation is accentuated, and possibly promotes, antigen-selected clonal expansions dominated by VH4–34-encoded autoreactive 9G4 antibodies. Ongoing monoclonal antibody studies will clarify the nature of the selecting antigens. Disclosure: C. Tipton, None; C. Fucile, None; A. Rosenberg, None; S. Jenks, None; J. Hom, None; F. E. H. Lee, None; I. Sanz, Pfizer Inc, 5, Biogen Idec, 9.

S10

bodies (2 cases) pattern. By IP, one had anti-U1RNP, 2 had anti-Su/Ago2 (both had GW body staining in ANA) and 3 had anti-Ro60. Two were positive by anti-Ro52 ELISA and one was positive in anti-U1RNP-70kD ELISA (positive for U1RNP by IP). Among 6 cases with these autoantibodies (2 had more than one), a case with anti-Su⫹Ro52 had a diagnosis of SLE, however, other 5 cases had non-specific rheumatological symptoms only. It is of interest that anti-U1RNP and –Su/Ago2 that are the main autoantibody specificity induced by adjuvant mineral oils in animal models, was also seen in mineral oil injected human subjects. Prevalences of these antibodies was not as high as pristane-treated mice but similar to those in adjuvant mineral oil-injected mice. Conclusion: Patients with rheumatologic inflammatory syndrome after mineral oil injections have autoantibody specificity similar to those in mice with adjuvant mineral oil-induced autoimmunity (anti-U1RNP and Su/Ago2). In addition, they also developed anti-Ro60 and Ro52. Table. Prevalence of autoantibodies by immunoprecipitation Human Chemical injected

Mineral oil

Site of injection

buttocks, breast, thigh, leg, face rheumatologic symptoms 21 4% 10% 14% 14%

Features N⫽ Anti-U1RNP Anti-Su/Ago2 Anti-U1RNP or Su Anti-Ro60

BALB/cByJ mice Adjuvant mineral oil (incomplete Freund’s adjuvant) intraperitoneal

20 10% 10% 20% 0%

BALB/cByJ mice pristane

intraperitoneal immune complex glomerulonephritis 20 55% 45% 85% 0%

Disclosure: M. Satoh, None; O. Vera-Lastra, None; C. Martıńez, None; J. Sepulveda- Delgado, None; L. J. Jara, None; R. Vargas-Ramı´rez, None; B. T. Martin-Marquez, None; S. J. Calise, None; E. K. L. Chan, None; M. Va´zquez-Del Mercado, None.

Disclosure: K. Miura, None; A. Kondo, None; K. Takahashi, None; D. Hirano, None; Y. Hiki, None; S. Yoshida, None; Y. Yuzawa, None; Y. Kurosawa, None.

28

27

Differentiation, Activation, and Autoreactivity Of CD11cⴙ B Cells (ABCs). Alice E. Wiedeman, Natalia V. Giltiay, Lena Tanaka and Keith B. Elkon. University of Washington, Seattle, WA.

Specificity Of Autoantibodies In Patients With Rheumatologic Inflammatory Syndrome Following Mineral Oil Injections Is Similar To Those In Mice With Adjuvant Mineral Oil-Induced Autoimmunity. Minoru Satoh1, Olga Vera-Lastra2, Claudia Martıńez3, Jesu´s Sepulveda- Delgado3, Luis J. Jara4, Rau´l Vargas-Ramı´rez5, Beatriz Teresita Martin-Marquez6, S. John Calise1, Edward K.L. Chan1 and Monica Va´zquez-Del Mercado7. 1 University of Florida, Gainesville, FL, 2MD, Mexico City, Mexico, 3Hospital de Especialidades Centro Me´dico Nacional La Raza, IMSS, Mexico City, Mexico, 4Hospital de Especialidades Centro Medico La Raza, Me´xico City, Mexico, 5Instituto de Investigacioń en Reumatologıá y del Sistema Mu´sculo Esquele´tico, Universidad de Guadalajara, Guadalajara, Jalisco, Me´xico, Mexico, 6Instituto de Investigacioń en Reumatologıá y del Sistema Mu´sculo Esquele´tico, Guadalajara, Mexico, 7Universidad de Guadalajara, Guadalajara, Jalisco, Me´xico, Mexico.

Background/Purpose: Recently, a population of CD11c⫹ ageassociated B cells (ABCs) was identified in normal aged female mice. These cells could be expanded following activation by TLR7 agonists and their presence was associated with autoantibody production. The goals of the current study were to determine: a) when ABCs develop during B cell maturation in the spleen; b) how B cell intrinsic and extrinsic factors influence ABC generation; and c) how TLR7 expression influences ABC generation in a B cell receptor (BCR) transgenic (Tg) mouse with specificity for RNA (the TLR7 ligand). Methods: Knock-in mice that overexpressed TLR7 on a B6 background (TLR7 Tg) were crossed with either IFNaR knockout (KO), RNAse Tg, HEL-specific BCR Tg, or RNA specific BCR Tg (H564) mice. Mixed chimeras were generated with combinations of wildtype B6 and TLR7 Tg bone marrow. Flow cytometry was used to identify splenic B cell subsets and analyze CD11c expression. CD11c⫹ (ABCs) and CD11c- (non-ABCs) B cells were sorted using a BD FACS Aria, and cultured for 6d with or without TLRs 4, 7, or 9 agonists. Ig production was quantified by ELISA. Results: While in wildtype mice, CD11c⫹ ABCs were increased only in females greater than 6 months of age, TLR7 Tg mice had significantly expanded ABCs as young as 3 months in both males and females (1.1 ⫾ 0.2 % in B6 versus 3.4 ⫾ 1.1 % in TLR7 Tg, p ⬍ 0.003). ABC accumulation was greatest in the transitional B cell stages but the relative proportion of ABCs was also higher in marginal zone than follicular B cells. To address whether generation of ABCs was intrinsic or extrinsic we used WT:TLR7 Tg mixed bone marrow chimeras, and observed that ABC expansion of TLR7 Tg B cells had a definite cell-intrinsic component. However, when TLR7 Tg mice were crossed with RNAse Tg or IFNaR KO mice to remove the RNA ligand or response to IFN-I respectively, ABCs were modestly but statistically significantly reduced. In TLR7 Tg mice, serum levels of anti-RNA autoantibody correlated with % ABCs in spleen and we also observed that ABCs produced more antibody (including anti-RNA autoantibodies) in response to innate stimulation in vitro. Consistent with these findings, crossing TLR7 Tg and autoreactive H564 BCR mice further increased ABC accumulation, while

Background/Purpose: Intraperitoneal injection of pristane or adjuvant mineral oil (incomplete Freund’s adjuvant) in non-autoimmune strains of mice mainly induces autoantibodies to U1RNP and Su/argonaute2 (Ago2). Injection of mineral oil as a cosmetic procedure has been commonly performed in certain countries including Mexico, however, inflammatory syndrome among these subjects have been reported. In the present study, autoantibody specificity in patients who had mineral oil injection and inflammatory syndrome were examined and compared with data from animal models Methods: Twenty-one cases of patients, who had mineral oil injections (17 buttocks, 9 breast, 5 thigh, 2 legs, 1 face) and developed rheumatologic symptoms after mineral oil injections were studied. Autoantibodies were tested by immunofluorescence antinuclear antibodies (ANA) using HEp-2 slide, immunoprecipitation (IP) of 35S-methionine-labeled cell extract and anti-Ro52 and U1RNP-70kD ELISA. Autoantibody specificities were compared with those in BALB/cByJ mice that received a single 0.5 ml intraperitoneal injection of adjuvant mineral oils (incomplete Freund’s adjuvant or pristane). Clinical information was from medical record Results: By immunofluorescence ANA, 62% (13/21) were positive in nuclear (5 cases), nucleolar (3 cases), mitochondria-like (3 cases), or GW

S11

Sunday, October 27

cell surface proteins were biotinylated, solubilized with detergent, and captured on ELISA wells coated with NeutrAvidin biotin binding protein. AECA titers in serum from 126 autoimmune disease patients and 122 healthy controls (HC) were tested. Additionally, sera from 52 patients with biopsyproven lupus nephritis (LN), 25 with systemic lupus erythematosus (SLE) without renal involvement (non-LN SLE), 10 disease controls (DC) and 81 healthy controls were tested for IgG- and IgA-AECA to human glomerular EC (HGEC) by CSP-ELISA. Results: IgG-AECA were detected in 28 of 36 (78%) of SLE patients; in 13 of 16 (81%) of mixed connective tissue disease (MCTD) patients; in 5 of 9 (56%) of systemic sclerosis (SSc); and in 4 of 122 (3%) of healthy controls. Relatively weak denaturation of antigens on ELISA wells caused loss of binding of these autoantibodies. Titers of IgG- and IgA- AECA to HGEC were significantly higher in LN and non-LN SLE patients than in the combined DC and HC (P ⬍ 0.001) groups. The level of IgG-AECA did not correlate with active lesions, but the level of IgA-AECA to HGEC did correlate with histological evidence of active lesions in LN patients (P ⬍ 0.001). Immunocytochemical analysis showed AECA recognized membrane proteins on HGEC. The significant correlation of titer of AECA to both HGEC and HUVEC (R ⫽ 0.95 for IgG-, 0.93 for IgA-AECA, respectively) indicated AECA in LN patients recognize membrane proteins expressed on HGEC and HUVEC. To identify specific antigens against AECA, biotinylated CSPs were incubated with sera from LN patients with high titers of IgG-AECA, immunoprecipitated with immobilized protein G followed by immobilized avidin, and blotted with NeutrAvidin. A 150-kDa protein band that shifted to a 55-kDa protein band under reducing conditions was detected in patients with LN, but not in HC. Conclusion: This newly developed CSP-ELISA method enables the detection of antibodies to the labile epitopes of autoantigens such as membrane proteins, and this method is generally applicable to various kinds of membrane proteins and the antibodies against them. IgA-AECA was observed to be associated with pathological activity in LN. These EC membrane components recognized by AECA may be linked with the pathogenesis of LN. We propose CSP-ELISA for measuring AECA in serum samples for routine laboratory testing.

Sunday, October 27

crossing to an HEL mouse resulted in expansion of ABCs in the polyclonal but not in the HEL-specific subset. Conclusion: CD11c⫹ ABCs are B cells that appear early during B cell maturation, and are increased under conditions of TLR7 hyperreactivity. ABC generation is, in part, cell intrinsic but is also influenced by exposure to self antigen and to IFN-I. ABCs can be directly implicated in autoantibody production by comparing antibody concentrations following stimulation of ABCs and non-ABCs in vitro as well as by enhanced ABC expansion in a TLR7 Tg mice that has specificity to self antigen (RNA) but not to non-self (HEL). Since these results suggested that ABCs, at least in part, contribute to autoantibody production, CD11c⫹ B cells may be a good target for amelioration of autoimmunity without broadly impairing host defense.

Methods: B cell subsets were flow sorted from 12 healthy controls (HC) and 13 SLE patients with low disease activity (SLEDAI⬍6): naı¨ve (IgD⫹CD27⫺), unswitched memory (IgD⫹CD27⫹), switched memory (IgD⫺ CD27⫹ ), DN memory (IgD⫺ CD27⫺ ) and plasmablasts (IgD⫺CD27⫹⫹CD38⫹⫹). Total RNA was hybridized to Human Genome U133 Plus 2.0 GeneChip (Affymetrix). The scans were subjected to quality control measures and the resulting CEL files normalized with Robust Multi-array Average (RMA). All calculations and analyses were carried out using R/Bioconductor computational tools. To identify differentially expressed genes (DEGs) between groups of samples, a two-way ANOVA model using Method of Moments was applied. Gene that exhibited a FDR ⬍0.05 and fold change ⬎2 or ⬍-2 were considered significantly different. Results: Overall, there are fewer DEGs among the three memory B cell subsets than between naı¨ve and each of the three memory subsets in both healthy subjects and SLE patients, suggesting that the gene expression program is quite similar among all the memory B cells in HC and SLE. IL4R and IL21R are upregulated in naı¨ve cells whereas IL6R is over-expressed in memory cells. Also upregulated in naı¨ve cells is TCL1A, which promotes Akt-mediated survival. TACI is upregulated in the switched and unswitched memory subsets compared to naı¨ve cells in HC and SLE, and is higher in SLE DN B cells compared to the same population in HC. SLE DN cells had also higher levels of AICDA and FcRL4. Overall, most DEGs differentials between the two cohorts were identified in the DN subset. Within either cohort, thousands of DEGs were observed between plasmablasts and the other four B cell subsets. Conclusion: The transcriptome of multiple B cell subsets was remarkably similar between HC and SLE patients with low disease activity suggesting that most differences in active disease may be due to extrinsic differences. Our results are consistent with the important role of IL-4 and IL-21 as growth factors for naı¨ve cells and the known activity of IL-6 in memory differentiation into plasma cells. Of great interest is the upregulation of TACI, AICDA and FcRL4 in DN B cells, a population expanded in SLE. The implications for the potential germinal center origin and activation status of these cells, reported to represent exhausted cells in HIV infection, will be discussed.

Disclosure: A. E. Wiedeman, None; N. V. Giltiay, None; L. Tanaka, None; K. B. Elkon, None.

29 A Novel CD27(ⴚ) B-Cell Subset Identified Based On Intracellular Characteristics Is Expanded In SLE. S.J. Fleischer1, Capucine Daridon2 and Thomas Do¨rner3. 1Charite´ University Medicine Berlin, Berlin, Germany, 2Charite´ University Medicine / German Rheumatism Research Centre Berlin (DRFZ), Berlin, Germany, 3Charite´ university medecine/ German Rheumatism Research Centre Berlin (DRFZ), Berlin, Germany. Background/Purpose: Several studies linked the emergence of autoimmunity to abnormalities of the B-cell receptor (BCR) due to disturbances of signaling molecules or its co-receptors. Therefore this study focused on spleen tyrosine kinase (Syk), a key molecule of early BCR signaling in autoimmunity. Our model is systemic lupus erythematosus (SLE) which is an autoimmune disease known to be associated with a breakdown of selftolerance resulting in auto-antibody production, B-cell hyper-reactivity and disturbed B-cell homeostasis of peripheral B-cell subsets. Methods: Peripheral blood was taken from 31 healthy, 61 SLE patients, 15 patients with rheumatoid arthritis (RA) and 16 primary Sjo¨gren’s syndrome (pSS) patients. The expression of Syk and basal / BCR-induced phosphorylation of Syk by B-cells were studied by phosphoflow analysis. To assess the localization of Syk within B-cells, immunofluorescence was performed and analyzed by confocal microscopy. In addition, characterization of B-cells was evaluated by flow cytometry using the following markers CD19, CD20, CD27, CD38, IgD, CD95 and Ki67. Finally, the capacity of B-cells to differentiate into antibody producing cells was evaluated by flow cytometry and Elispot after 5 days of in vitro culture. Results: In this study, two different subsets according their expression of Syk (Sykbright and Sykdim) within the CD20⫹CD27(⫺) population have been identified. The frequency of CD27(⫺)Sykbright B-cells were significantly increased in SLE compare to HD, however the disease activity (SLEDAI) do not correlate with the frequency of this subset. No significant increase in the frequency of this population was observed in others autoimmune diseases (i.e. RA and pSS). This subset showed an accumulation of Syk within the cytoplasm, a superior response to the BCR and plasma cell differentiation compared to CD27(⫺) Sykdim B-cells. Finally, this subset exhibited a memory-like phenotype. Conclusion: A novel subset of B-cells defined as CD27(⫺)Sykbright B-cells and carrying memory features has been identified in SLE patients. This population represents a prime candidate how peripheral check point control of conventional B cells can be possibly bypassed in SLE.

Disclosure: C. Wei, Biogen Idec, 2; E. Ramos, None; N. Allaire, Biogen Idec, 1, Biogen Idec, 3; S. Szak, Biogen Idec, 1, Biogen Idec, 3; S. Kalled, Biogen Idec, 1, Biogen Idec, 3; A. Ranger, Biogen Idec, 1, Biogen Idec, 3; I. Sanz, Biogen Idec, 2, Pfizer Inc, 9.

31 IgD- CD27- B Cells From Systemic Lupus Erythematous Patients Have Increased Expression Of Genes Involved In RNA Sensing and Toll-Like Receptor 3 Signaling Pathways. Scott Jenks1, Edward Ramos2 and Ignacio Sanz1. 1Emory University School of Medicine, Atlanta, GA, 2Emory University, Atlanta, GA. Background/Purpose: SLE patients have perturbations in B cell subsets including a large expansion of IgD-CD27- B cells (DN) in patients with active disease. Patients also have changes in PBMC gene expression, particularly increased expression of interferon regulated genes (IRG). However, data on gene expression in SLE B cells is limited. The purpose of this study was to understand the function and potential dysregulation of DN in SLE by comparing differences in gene transcription between B cell subsets from SLE patients and healthy control donors (HCD). Methods: RNA was isolated from B cells from 3 HCD and 3 SLE patients with elevated DN sorted into naive (IgD⫹CD27-CXCR5⫹), switched memory (IgD-CD27⫹CXCR5⫹), and DN (IgD-CD27-CXCR5-) subsets. After amplification, high throughput sequencing of cDNA was used for transcriptional expression profiling. Sequencing data was processed and analyzed using the Tuxedo suite and EdgeR software package. Genes were tested for differential expression using a three way ANOVA model. Results: 115 of the 249 genes differentially expressed in B cells between SLE patients and HCD were IRG. In addition to previously defined IRG, pathway analysis found several genes involved in TLR-3 and RNA sensing signaling pathways. These included RNA sensing molecules RIG1 and MDA5, downstream kinases IKBKE and TBK1, kinase substrate and transcription factor IRF-7, and HERC5 and TRIM21, molecules that regulate IRF stability. Of these only IRF-7 has been previously described as differentially expressed in SLE B cells. While, expression for most genes was increased in all B cell subsets, the DN subset showed the largest increase in expression for all genes except IKBKE and TRIM21. The DN subset also had increased expression and differential splicing of TOLLIP, an inhibitor of MyD88 mediated TLR signaling.

Disclosure: S. J. Fleischer, None; C. Daridon, None; T. Do¨rner, None.

30 Gene Expression Profiling Analysis Of Human B-Cell Subsets In Health and Systemic Lupus Erythematosus. Chungwen Wei1, Edward Ramos1, Norm Allaire2, Suzanne Szak2, Susan Kalled2, Ann Ranger2 and Ignacio Sanz1. 1Emory University, Atlanta, GA, 2Biogen Idec Inc, Cambridge, MA. Background/Purpose: SLE is an autoimmune disease, which exhibits multiple B cell abnormalities including expanded populations of plasmablasts and DN memory cells as well as a contracted unswitched memory subset. The transcriptional profiles that underlie these homeostatic changes are poorly understood. To remedy this knowledge gap and generate insight into the disease pathogenesis, we carried out transcriptome analysis of sorted B cell subsets.

S12

ities, including the activation of naı¨ve B cells. However, gaps in terms of our knowledge regarding the extent and mechanisms of activation remain. We therefore investigated the expression of activation markers on activated naı¨ve (aN) and resting naı¨ve (rN) B cells of SLE patients as well as the stimulatory pathways responsible for their expression. Methods: Multidimensional flow analysis was utilized to determine levels of B cell activation markers including CD21, CD23, CD24, CD25, CD69, CD80, CD83, CD86, and IgM on aN and rN B cells of SLE patients with active disease versus healthy controls (HC). In addition, total PBMCs from SLE and HC patients were stimulated ex vivo with B cell receptor (anti-kappa or anti-lambda), T cell (CD40L), or cytokine stimulation for 16 h, 48 h, and 4 days and compared via flow analysis. Lastly, we compared the epigenome of SLE aN and rN by analyzing their genome-wide DNA methylation status using MeDIP-Seq. Results: Compared to HC, aN B cells, globally defined by Mitotracker green retention, from SLE patients exhibited low levels of CD21, CD24, CD69, and CD83, while CD86 levels were up-regulated. Following stimulation, aN B cells of SLE patients exhibited increased levels of CD21, CD25, CD69, CD80, CD83, and CD86 and decreased levels of CD24 at 16 h post-stimulation. Levels of CD21 and CD25 decreased at 48 h, followed by CD83 levels on day 4 post-stimulation. Expression of CD69, CD80, and CD86 remained high at all time points. However, aN B cells of SLE patients exhibited differential expression of CD23 and IgM based on stimulation type. Global DNA methylation analysis of aN B cells from an SLE patient revealed several genes, including interferonregulated genes, were hypomethylated when compared to the HC. Interestingly, the CD83 gene was hypermethylated in aN B cells of the SLE patient; a result consistent with low CD83 levels observed via flow staining and RNASeq transcriptional analysis performed on several SLE patients. Conclusion: Activated naı¨ve B cells of SLE patients differentially express activation markers including CD21, CD24, CD83, and CD86. The stimulation of total PBMCs through distinct pathways reveals that activation markers are temporally expressed on aN B cells of SLE patients only to decrease with prolonged stimulation, as would be expected in vivo. Epigenetic analysis indicated hypermethylation of CD83, a result consistent with flow and transcriptional studies. Decreased expression of CD83 could help explain several phenotypic characteristics of lupus B cells including: decreased marginal zone maturation, decreased IL-10 production, and increased Ig secretion [1, 2]. Our results provide important clues regarding abnormal B cell function, highlighting the power of integrated experimental approaches to address this problem. 1. Kretschmer, B., et al., CD83 modulates B cell function in vitro: increased IL-10 and reduced Ig secretion by CD83Tg B cells. PLoS One, 2007. 2(8): p. e755. 2. Luthje, K., et al., CD83 regulates splenic B cell maturation and peripheral B cell homeostasis. Int Immunol, 2008. 20(8): p. 949–60. Disclosure: E. Blalock, None; C. Scharer, None; S. Jenks, None; J. Boss, None; I. Sanz, Pfizer Inc, 5, Biogen Idec, 9.

33 B Cell Receptor Signaling As A Potential Clinical Parameter In Lupus. Michael Faludi, Christian A. Pineau, Evelyne Vinet, Ann E. Clarke, Sasha Bernatsky, Joyce Rauch and Emil P. Nashi. McGill University Health Center, Montreal, QC. Background/Purpose: B cells are central to SLE. Signaling through the B cell receptor (BCR) controls critical processes at various stages of B cell development. Murine BCR manipulation and human genomic studies have implicated the BCR pathway in SLE, but few studies have assessed this pathway in patients. Methods: Ethics committee approval was obtained. Blood lymphocytes were isolated from patients with SLE and other autoimmunities, and normal controls; cells were frozen in liquid nitrogen. Detailed clinical parameters were obtained. 100 000 B cells were stimulated with 20 ug/mL anti-IgM and anti-IgG for 0 or 5 minutes, and a 10-color flow cytometry panel was used to identify the following B cell subsets: naı¨ve mature, transitional, IgM memory, IgG memory, IgG B1 and IgM B1. We assayed surface BCR levels, pSyk (early signaling), pPLC-␥2 (mid-point), pERK1/2 (late) and carboxy-pLyn (regulation). Triplicates of each assay, daily bead-based voltage adjustment, and replication of all experiments on separate days ensured signal stability.

Disclosure: S. Jenks, None; E. Ramos, None; I. Sanz, Pfizer Inc, 5, Biogen Idec, 9.

32 Understanding The Stimulatory Pathways Responsible For Naı¨ve B Cell Activation In Systemic Lupus Erythematosus. Emily Blalock, Chris Scharer, Scott Jenks, Jeremy Boss and Ignacio Sanz. Emory University School of Medicine, Atlanta, GA. Background/Purpose: Systemic lupus erythematosus (SLE) is a recurrent autoimmune disease characterized by multiple B cell abnormal-

S13

Sunday, October 27

Conclusion: IKBKE and TBK1 are key signaling kinases for detecting RNA through the TLR3 and MDA5/RIG1 pathways. These pathways detect viral RNA but are also activated by endogenous RNA from apoptotic cells. Increased expression of IKBKE and TBK1 in SLE B cells is accompanied by the expression of both the upstream receptors and downstream kinase substrate. Furthermore, while TRIM21 mediates degradation of IRF-7, expression of TRIM21 was lowest in the DN and these cells expressed high levels of HERC5, which prevents IRF degradation. Coordinated expression of these genes strongly suggests these pathways are more active in SLE B cells and particularly in the DN population. In SLE, RNA from apoptotic cells can be internalized in RNP and Ro60 immune complexes through Fc receptors(FCR) and are thus available to MDA5/RIG1 and TLR-3. DN express high levels of FCR. This combined with increased expression of these signaling molecules likely results in RNA antigens having a powerful pro-inflammatory influence on DN.

Conclusion: We have optimized a flow cytometry-based technique that allows simultaneous study of BCR signaling in B cell subsets, at various stages of signaling, in a manner that yields stable results over time. We found that diminished BCR signaling in transitional B cells correlates with increased diversity of autoantibodies in SLE patients. We will study 150 patients. Clinical BCR measurement may inform SLE pathogenesis and guide emerging BCR-targeted therapies.

Sunday, October 27

Results:

Disclosure: M. Faludi, None; C. A. Pineau, None; E. Vinet, None; A. E. Clarke, None; S. Bernatsky, None; J. Rauch, None; E. P. Nashi, None.

34 Dichotomous Responses Of Human Systemic Lupus Erythematosus B Cell Subsets To B Cell Receptor Stimulation. Franziska Matzkies, Anthony DeFranco, Andrew J. Gross, Maria Dall’Era and Michelle Hermiston. University of California, San Francisco, San Francisco, CA. Background/Purpose: Autoantibody production is a hallmark of Systemic Lupus Erythematosus (SLE), supporting a central role for B cells in disease pathogenesis. Prior studies have demonstrated augmented B cell receptor (BCR) signaling in a subset of SLE patients relative to healthy controls (HC). Both B cell intrinsic factors such as aberrant signal network wiring and B cell extrinsic factors such as excessive cytokines have been implicated in loss of B cell tolerance and B cell hyperactivity. However, previous studies have generally examined these parameters in total peripheral blood B cells rather than in well defined subpopulations. This is an important distinction because BCR signaling in mature naive B-cells leads to proliferation and survival, while these same signals lead to receptor editing, depletion or anergy in immature cells. We hypothesize that dysregulation of these signaling networks could contribute to SLE pathogenesis. Methods: Peripheral blood was obtained from 12 patients satisfying ACR SLE criteria and 12 HC that lacked a family history of autoimmune disease. To minimize confounding influences of cytokines, patients had low disease activity (SLEADI⬍5). B cells were enriched with RosettespTM and ficoll, rested at 37°C for 1 hr, stimulated with 10 ␮g/ml (low dose) or 50 ␮g/ml (high dose) of F(ab’)2 goat antihuman IgM or phorbal 12-myristate 13-acetae (PMA), immediately fixed, and processed for flow cytometry. Phosphorylation of intracellular signaling proteins (pSyk, pPLCg, pERK, pAKT, pS6) in transitional (CD20⫹CD27⫺CD24HiCD38Hi) and naı¨ve (CD20⫹CD27⫺CD24Int CD38Int) B cells was measured using FlowJo 8.8.7 and Cytobank software. Results: The basal phosphorylation states of Syk, ERK, PLCg, and S6 were increased in both transitional and naı¨ve B cells of SLE patients relative to HCs. Upon BCR ligation, both transitional and naı¨ve SLE B cells demonstrated augmented phosphorylation of SYK, ERK, and PLCg relative to HC. Surprisingly, these same patients displayed hypo-phosphorylation of S6 in response to the same stimuli. Interestingly, hypo-activation of S6 was observed in mature naı¨ve but not transitional B cells. The response to stimulation with PMA, which bypasses proximal signaling machinery, was equivalent in SLE and HC B cell subsets, indicating that the difference in signaling was upstream of S6. Conclusion: We find that both transitional and naı¨ve B cells obtained from SLE patients with well-controlled disease have increased activation of the MAPK pathway in the basal state and in response to BCR stimulation. Phosphorylation of S6, which is downstream of the PI3K/ AKT/mTOR pathway, is also elevated in the basal state in SLE B cell subsets, but surprisingly hypophosphorylated in naı¨ve mature but not transitional B cells in response to BCR ligation in these same patients. While S6 is a target of the PI3K pathways after BCR ligation, it also receives inputs through growth hormone and cytokine receptors as well as the mTOR complex, which plays a key role in sensing energy and nutritional levels. Current investigations are under way to identify the mechanisms mediating these observations and their relation to SLE pathogenesis.

Fig 1. Stability of BCR signal. We found consistently high correlation between assays run on separate days (1a), between frozen and unfrozen cells (1b) and cells obtained at various times from the same person (1c). Each dot represents the fluorescence of one signaling parameter in a specific B cell subset.

Fig 2. The importance of studying different B cell subsets and BCR stages. Some patients had increased signal in one subset but decreased or normal in other subsets (2a). Signaling deviation can occur at different stages (2b, yellow highlights).

Fig 3. Signaling in 15 patients. Diminished IgM Memory BCR signal was found in SLE and other autoimmunities versus controls (3a). Among SLE patients, diminished signal in transitional B cells correlated with increased number of autoantibody specificities (3b).

Disclosure: F. Matzkies, None; A. DeFranco, None; A. J. Gross, None; M. Dall’Era, None; M. Hermiston, None.

S14

Defective Regulatory Function Of Granzyme B-Producing B Cells In Patients With Systemic Lupus Erythematosus. Naoko Ueki, Hiroaki Niiro, Shun-ichiro Ota, Hirofumi Tsuzuki, Siamak Jabbarzadeh-Tabrizi, Yuri Hirosaki, Kumiko Noda, Naoyasu Ueda, Atsushi Tanaka, Masahiro Ayano, Sho Ueda, Satomi Hisamoto, Daisuke Oryoji, Mitsuteru Akahoshi, Yojiro Arinobu, Hiroshi Tsukamoto, Takahiko Horiuchi and Koichi Akashi. Department of Medicine and Biosystemic Science, Kyushu University Graduate School of Medical Sciences, Fukuoka, Japan. Background/Purpose: The efficacy of B-cell depletion therapy highlights a pathogenic role of B cells in autoimmune diseases. In certain conditions, elimination of B cells can lead to exacerbation of these diseases, suggesting the existence of regulatory B cells (Bregs). Bregs are often referred to as IL-10producing B cells, however it is possible that Bregs can exert a regulatory function by an IL-10-independent mechanism. Granzyme B (GzmB) is known to exert both cytotoxic and non-cytotoxic effects on various cell types. In this study, we have determined whether granzyme B (GzmB)-producing B cells could function as another type of Bregs in humans, and also have tested their functions in patients with systemic lupus erythematosus. Methods: Levels of GzmB mRNA and protein in B cells were assessed using quantitative real-time PCR and intracellular staining, respectively. To evaluate the function of GzmB-producting B cells, they were co-cultured with activated T cells, and growth, survival and cytokine production of T cells were then assessed using flow cytometry. Results: Among the stimulators tested, IL-21 was the potent inducer of GzmB in normal B cells and it acted synergistically with antigen receptor stimulation. Naive B cells produced higher levels of GzmB as compared with memory B cells. In addition, GzmB-producing B cells inhibited the growth, survival and cytokine production of T cells, which is in line with the idea that these cells function as Bregs in normal subjects. In SLE patients, naive B cells similarly produced more GzmB than memory B cells, however levels of its production in both subsets were apparently higher than those in normal subjects. Intriguingly, however, GzmB-producing B cells in SLE patients were without regulatory effects on T cell functions. A molecular explanation for these findings is now in progress. Conclusion: Our current findings could help to better understand a role of Bregs in the pathogenesis of autoimmune diseases as well as to provide a novel clue to manipulate the generation of Bregs for therapeutic application in the future.

Disclosure: J. E. Wither, None; N. H. Chang, None; T. Li, None; J. Kim, None; C. Landolt-Marticorena, None; P. R. Fortin, None; D. D. Gladman, Janssen, Abbvie, Amgen, Pfizer, UCB, Novartis, Lily, Celgene, 2, Janssen, Abbvie, Amgen, Pfizer, UCB, Novartis, Lily, Celgene, 5; M. B. Urowitz, None.

37 Follicular Entry Of Lymphotoxin-Expressing B Cells Via Type I Interferon Disrupts Marginal Zone Barrier Integrity and Exacerbates Systemic Autoimmunity. Hao Li1, Hui-Chen Hsu2, Qi Wu1, Jun Li1, PingAr Yang1, Yang-Xin Fu3 and John D. Mountz1. 1University of Alabama at Birmingham, Birmingham, AL, 2Birmingham VA Medical Center, Birmingham, AL, 3The University of Chicago, Chicago, IL. Background/Purpose: Marginal zone macrophages (MZMs), a small subset of specialized splenic macrophages located in the MZ, act as final follicular entry barrier to clear apoptotic cells (ACs) to prevent AC antigens (Ags) entering to the spleen follicles. We recently reported reduced tolerogenic function, frequency and numbers of MZMs in the spleen of lupus prone BXD2 mice. Loss of MZMs were further confirmed in B6 Sle1.Sle2.Sle3 mice and lupus patients. Expression of lymphotoxin (LT) by B cells has been implicated as a critical factor to maintain MZMs. The present study investigated how type I interferons (IFNs) dissociated this interaction and induced MZM loss in lupus. Methods: Confocal microscope and FACS analysis were carried out to determine the percentages and location of of MZMs, B cells and the distribution of LT and its receptor (LT␤R) in the spleen. A mixed bone marrow (BXD2-GFP⫹Ifn␣r⫹/⫹:BXD2-GFP⫺Ifn␣r⫺/⫺ ⫽1:1) reconstitution in BXD2 Rag2⫺/⫺ mice was carried out to determine if type I IFNs affect MZ integrity through direct action on MZMs or indirectly by affecting B cells. LT-LT␤R interaction in vivo was disrupted by repeated injection of mice with a low dose of LTR fusion protein (LTR:Fc, 15 ug/mouse per week, 4 weeks) or by administrations of CpG (2.5 ug/mouse, every other day for 2 weeks). Results: Repeated injections of CpG to B6, but not B6-Ifn␣r⫺/⫺ mice, induced loss of MZMs. Consistent with this, the spontaneous loss of MZMs and the elevated autoantibodies against MZM Ags, Marco and scavenger receptor A, in BXD2 mice were prevented in the absence of type I IFN signaling. Marginal zone B cells (MZs) are the major LT expressing cells in the marginal zone. Although there was comparable expression of LT, MZs were mainly localized in the follicles in BXD2 mice but in the marginal zone in both B6 and BXD2-Ifn␣r⫺/⫺ mice. BM reconstitution experiment further shows that the absence of type I IFNR did not directly affect MZM repopulation, but rather affected MZ B cell distribution. GFP⫹Ifn␣r⫹/⫹ B cells mainly migrated to the follicles whereas GFP⫺Ifn␣r⫺/⫺B cells remained in the marginal zone and became the source of LT to support MZMs. Administration of a low dose of LTR:Fc to BXD2 mice selectively depleted MZMs and enhanced the spontaneous germinal center (GC) response without affecting follicular dendritic cell networks. Conclusion: Our present study suggests a novel mechanism associated with type I IFN-promoted immune response against apoptotic self-Ags as a result of follicular migration of LT expressing and self-antigen carrier B cells. The shift of LT expressing B cells from the MZ to the follicles by type I IFNs

Disclosure: N. Ueki, None; H. Niiro, None; S. I. Ota, None; H. Tsuzuki, None; S. Jabbarzadeh-Tabrizi, None; Y. Hirosaki, None; K. Noda, None; N. Ueda, None; A. Tanaka, None; M. Ayano, None; S. Ueda, None; S. Hisamoto, None; D. Oryoji, None; M. Akahoshi, None; Y. Arinobu, None; H. Tsukamoto, None; T. Horiuchi, None; K. Akashi, None.

36 IFN-␣ Induces Altered Transitional B Cell Signaling and Function In Systemic Lupus Erythematosus. Joan E. Wither1, Nan-Hua Chang2, Timothy Li2, Julie Kim2, Carolina Landolt-Marticorena3, Paul R. Fortin4, Dafna D. Gladman5 and Murray B. Urowitz5. 1Toronto Western Research Institute, Toronto Western Hospital, University of Toronto, Toronto, ON, 2Toronto Western Hospital, Toronto, ON, 3Toronto Western Hospital, University of Toronto, Toronto, ON, 4Centre de Recherche du Chu de Que´bec et Universite´ Laval, Quebec City, QC, 5University of Toronto, Toronto Western Hospital, Toronto, ON. Background/Purpose: Previous experiments suggest that the B cells of lupus patients are hyper-responsive to B cell receptor engagement resulting in increased tyrosine phosphorylation and Ca2⫹mobilization. However the precise B cell populations that are affected and the mechanisms leading to this hyper-responsiveness have yet to be determined. In this study we have used Phosflow to address these questions. Methods: PBMC were isolated from 27 healthy controls and 39 SLE patients with ⱖ 4 ACR criteria. Phosflow was used to assess the levels of p-SYK, p-PLC␥2, or p-ERK following Ig receptor engagement with goat anti-human IgM F(ab’)2in distinct B cell subsets defined by anti-CD19,CD27, -IgD, -IgM and -CD38. B cell proliferation and apoptosis following anti-IgM stimulation were assessed by flow cytometry, using CFSE and annexin V staining, respectively. For some experiments, healthy control B cells were incubated with IFN-␣, or 50% plasma ⫾ anti-IFN or irrelevant Ab. Lupus associated SNPs were determined by TaqMan genotyping. Results: There were increased basal levels of p-SYK and p-ERK in naı¨ve B cells (CD19⫹CD27⫺IgD⫹) from lupus patients as compared to controls.

S15

Sunday, October 27

The levels of basal p-SYK correlated with CD86 expression suggesting that these cells had already been activated in-vivo. Following crosslinking with anti-IgM, there was a significant increase in the proportion of p-SYK⫹ cells above basal levels in the naı¨ve B cell population of lupus patients as compared to controls. Similar trends were seen for the proportion of p-PLC␥2⫹ and p-ERK⫹ cells. The increases seen in p-SYK⫹ cells were most marked for the transitional B cell subset (CD19⫹CD27⫺IgD⫹CD38hiIgMhi), where the levels of p-SYK correlated with enhanced proliferation and survival. There was no correlation between lupus associated SNPs in BLK, LYN, PTPN22, and CSK, and the proportion of p-SYK⫹ cells following IgM crosslinking. The proportion of p-SYK⫹cells in the transitional B cell subset fluctuated between visits, suggesting a possible role for pro-inflammatory factors. Consistent with this, incubation of lupus plasma with control B cells enhanced SYK phosphorylation following IgM crosslinking, which was blocked by pre-incubation of plasma with anti-IFN but not irrelevant Ab. Incubation of healthy control cells with recombinant IFN-␣ enhanced SYK phosphorylation, proliferation, and survival following IgM crosslinking, particularly of the transitional B cell subset. Conclusion: IFN-␣ alters transitional B cell function leading to enhanced survival and proliferation. As purging of transitional B cells plays an important role in preventing autoreactive B cells from entering the mature B cell pool, it is likely that elevated levels of IFN-␣ exacerbate the breach of B cell tolerance in lupus.

35

Sunday, October 27

promoted apoptotic Ag delivery and LT stimulation to FDCs via breaking down the marginal zone barrier and finally exaggerated the production of pathogenic auto-antibodies.

of autophagy) is abnormally enhanced in T lymphocytes from lupus mice and patients (Gros et al., 2012). In this work, we show that a selective form of autophagy, chaperone-mediated autophagy (CMA), is a key target of P140 and demonstrate that the P140 inhibitory effect on CMA results from its ability to alter the integrity of the HSC70/hsp90 heterocomplex of lysosomal chaperones. Expression of HSC70 and LAMP-2A, the two main CMA components, which is increased in MRL/lpr B cells, is down-regulated after P140 treatment. P140 enters MRL/lpr B lymphocytes via a clathrin-dependent endo-lysosomal pathway and accumulates at the lysosomal lumen. There, it may act both by directly hampering HSC70 chaperoning functions and, as a result of loss of hsp90 function, by destabilizing LAMP-2A in lysosomes. This dual effect may interfere with the endogenous (auto)antigen processing and loading to MHCII molecules and as a consequence, lower activation of autoreactive T cells. Conclusion: Our findings provide the first demonstration showing that P140 peptide acts directly on CMA. These results shed light on mechanisms by which P140 can modulate lupus disease and by which it may operate in humans affected by this disorder.

Disclosure: H. Li, None; H. C. Hsu, None, 2; Q. Wu, None; J. Li, None, 2; P. Yang, None; Y. X. Fu, None; J. D. Mountz, None, 2.

38 Use Of An In Vitro Whole Blood Depletion ASSAY To Compare The Efficacy Of B CELL Depleting Agents In Patients With Systemic LUPUS Erythematosus. Venkat Reddy1, Geraldine Cambridge1, D.A. Isenberg1, Mark Cragg2 and Maria Leandro1. 1University College London, London, United Kingdom, 2Southampton University, Southampton, United Kingdom. Background/Purpose: Variability in clinical response to B-cell depletion therapy (BCDT) with the anti-CD20mAb rituximab (RTX) has been well described in Systemic Lupus Erythematosus (SLE). Poor clinical response is associated with incomplete depletion which suggests that improving the efficiency of depletion might result in improved therapeutic outcome. GA101 is a recombinant, afucosylated fully human type II anti-CD20 mAb that has shown more effective depletion and clinical response in phase II trials in lymphoma. We have therefore compared the in vitro B-cell cytotoxicity (cytotoxicity index, CTI) of BHH2 (glycosylated GA101) with RTX, in lymphocytes from patients with SLE. Methods: We included 23 patients with SLE, who met the American College of Rheumatology revised classification criteria. An in vitro autologous whole blood depletion assay (WBD) was used to assess the CTI. Briefly, 100␮l of heparinised blood was incubated with either RTX, BHH2 or without antibody, at a concentration of 1␮g/ml at 37°C, 5% CO2for 24hours. Samples were then analysed by flow cytometry for CD45 (all lymphocytes), CD3 (T cells) and CD19 (B cells). The CTI was calculated using the formula:CTI of mAb⫽100-[(number of B:Tcells in sample without antibody-number of B:Tcells with mAb)/number of B:T cells in sample without antibody) ⫻ 100] and the mean from triplicate well calculated. The relationship between the relative expression (mean fluorescence intensity;MFI) of CD20 and CD32B (FcgRIIB) on B cells and CTI was determined using spearman rank correlation. Concurrent clinical and laboratory parameters including anti-dsDNA and C3 were collected and assessed. Results: The mean CTI of BHH2 was higher than RTX in all but one patient. Median CTI in 23 SLE patients was 20% (range9–70) and 15% (range 1–42), for BHH2 and RTX, respectively (p⫽0.0002). CTI was ⬍25% in 5 (21%) and 16 (73%) patients, for BHH2 and RTX, respectively. The mean⫾SD MFI of CD20 and CD32B on SLE-B cells was 9079⫾4025 and 4223⫾1587, respectively. The CTI of neither mAb correlated with the expression of CD20 (r2⫽⫺0.332,0.204, for BHH2 and RTX, respectively). Also, there was no correlation between the CTI of mAbs and lymphocytecount, serum creatinine, total IgG, C3, positivity for ENAs or anti-dsDNA. Conclusion: These results indicate that BHH2 is superior to RTX at inducing cytotoxicity in vitro in B cells from patients with SLE. This study provides the preliminary data to consider type II mAbs (GA101-like) as an alternative BCD agent for SLE in a clinical trial setting.

Disclosure: S. Muller, None.

40 Preclinical Characterization Of a Humanized Antagonistic Anti-CD40 Mab. Kerry Ralph, Amy Nicoletti, Eunice Musvasva, Susan Cannan, Susan VanTongeren, Diann Blanset, Scott Brodeur, Jennifer Ahlberg, Hua Li, Steve Fogal, Sudha Desai, Kathy O’Shea, Rachel Kroe-Barrett, Gerald Nabozny, Helen Wu, Gale Hansen, Keith Canada, Sanjaya Singh, Meera Ramanujam and Christine Grimaldi. Boehringer Ingelheim Pharmaceuticals Inc., Ridgefield, CT. Background/Purpose: CD40-CD40L interactions play a central role in T-B cell costimulation. Targeting this pathway has generated great interest; however early attempts to target CD40L failed mainly due to thrombotic complications observed in the clinic. In addition, developing antagonistic CD40 antibodies with suitable potency, favorable PK and no agonistic activity has proven to be challenging. BI 655064 is an anti-human CD40 mAb being developed for the treatment of autoimmune disorders. BI 655064 is engineered as a human IgG1 molecule with a mutated Fc region to abrogate effector function. Here we describe the preclinical characterization of BI 655064. Methods: Binding of BI 655064 to CD40 expressed on primary B cells was measured by flow cytometry. The ability of BI 655064 to block CD40L-induced B cell proliferation in vitro was measured by tritium uptake. A subcutaneous PK/PD study in the cynomolgus monkey was performed to establish correlations between BI 655064 exposure, target coverage using a receptor occupancy assay and pharmacodynamic effects using an ex vivo CD54-induction assay. In vivo blockade of B cell function was also tested in a human-peripheral blood lymphocyte (huPBL) induced SCID mouse model of graft-versus-host disease (GvHD) and in cynomolgus monkeys immunized with keyhole limpet hemocyanin (KLH). Results: BI 655064 binds human CD40 on B cells present in whole blood with an EC90 value of 6.85 nM ⫾ 0.74. Blockade of CD40L-induced B cell proliferation was observed at an average IC50 of 0.4 nM. In the PK/PD study, cynomolgus monkeys dosed with greater than 1 mg/kg of BI 655064 exhibited complete blockade of ex vivo CD54 upregulation corresponding to full CD40 target coverage on B cells. In vivo, BI 655064 demonstrated clear effects on B cell function in the GvHD model where both human IgM and IgG responses were abrogated. As part of a repeat dose tolerability study, BI 655064 given to cynomolgus monkeys prior to immunization with KLH resulted in inhibition of KLH-specific IgM and IgG antibody responses. At doses of 5 mg/kg and above, germinal center size was decreased microscopically in Peyer’s patches, lymph nodes, spleens and tonsils in treated monkeys. Conclusion: BI 655064 is a humanized antagonistic anti-CD40 mAb which is to be tested in human clinical trials for autoimmune disorders to establish safety and efficacy. BI 655064 demonstrated relevant pharmacologic in vitro and in vivo activity, blocking CD40-related functions at clinical dosing levels of ⬎ 1 mg/kg in animal models.

Disclosure: V. Reddy, None; G. Cambridge, None; D. A. Isenberg, None; M. Cragg, None; M. Leandro, None.

39 Chaperone-Mediated Autophagy As a Target Of Therapeutic P140 Peptide Used In Lupus. Sylviane Muller. CNRS, Strasbourg, France. Background/Purpose: In the pipeline of molecules with a potential for treating lupus patients, the P140 peptide/Lupuzor holds a lot of promise. P140 is a 21-mer linear peptide (sequence 131–151) that is derived from the small nuclear ribonucleoprotein U1–70K and that is phosphorylated at the Ser140position. In a multicenter, randomized, placebo-controlled phase IIb study, Lupuzor was safe and met its primary efficacy end points in lupus patients (Zimmer et al., 2012). These results confirm data generated in MRL/lpr lupus-prone mice in which the preclinical studies were performed. The mechanism of action of P140 was further studied in this mouse model. Methods: Immunocytochemical analyses were used to identify the way of P140 entry into B cells, and CMA activity was assessed in fibroblasts stably expressing a photoactivable CMA reporter. FACS analyses after LysoTracker and LysoSensor staining of isolated B cells were used to study the number and acidity of lysosomes in B cells from MRL/lpr mice. Results: We found previously that P140 reduces autophagic flux in MRL/lpr B cells (Page et al., 2011) and that macroautophagy (the best characterized type

Disclosure: K. Ralph, Boehringer Ingelheim, 3; A. Nicoletti, Boehringer Ingelheim, 3; E. Musvasva, Boehringer Ingelheim, 3; S. Cannan, Boehringer Ingelheim, 3; S. VanTongeren, Boehringer Ingelheim, 3; D. Blanset, Boehringer Ingelheim, 3; S. Brodeur, Boehringer Ingelheim, 3; J. Ahlberg, Boehringer Ingelheim, 3; H. Li, Boehringer Ingelheim, 3; S. Fogal, Boehringer Ingelheim, 3; S. Desai, Boehringer Ingelheim, 3; K. O’Shea, Boehringer Ingelheim, 3; R. Kroe-Barrett, Boehringer Ingelheim, 3; G. Nabozny, Boehringer Ingelheim, 3; H. Wu, Boehringer Ingelheim, 3; G. Hansen, Boehringer Ingelheim, 3; K. Canada, Boehringer Ingelheim, 3; S. Singh, Boehringer Ingelheim, 3; M. Ramanujam, Boehringer Ingelheim, 3; C. Grimaldi, Boehringer Ingelheim, 3.

S16

Targeting Cereblon With The High Affinity Immunomodulatory Compound CC-220: A Novel Therapeutic Agent For Autoimmunity. Peter Schafer1, Emily Rychak2, Derek Mendy2, Stacey Parton1, Lori Capone1, Antonia Lopez-Girona2, Dorota Cedzik1, Jolanta Kosek1, Ling-Hua Zhang1 and Rajesh Chopra1. 1Celgene Corporation, Summit, NJ, 2Celgene Signal Research, San Diego, CA. Background/Purpose: Cereblon (CRBN) is a component of the E3 ubiquitin ligase complex including CUL4A, DDB1, and ROC-1. CC-220 is a novel immunomodulatory compound currently in development for the treatment of immune conditions. The effects of CC-220 on CRBN binding and ubiquitination, and on immune cell responses were profiled. Methods: Binding studies to CRBN were conducted using endogenous CRBN from human U266 plasmacytoma cells and affinity beads in a competition assay by quantitative immunoblot determination. CRBN ubiquitination was measured in HEK293T cells transfected with a His-biotintagged CRBN construct, treated with the MG132 proteasome inhibitor (to arrest degradation of ubiquitinated proteins). Cells were lysed and processed to measure CRBN ubiquitination by SDS-PAGE and immunoblot analysis. T-cell costimulation was measured in purified primary human T cells stimulated using immobilized anti-CD3 antibody. Cytokine secretion was measured by ELISA. Immunoglobulin M and G (IgG and IgM) production was measured from normal donor peripheral blood mononuclear cells by culturing in the presence of the B cell differentiation factors IL-2, IL-6, IL-10, IL-15, CD40L, and TLR9 ligand. IgM and IgG were measured by ELISA. Cell proliferation studies were conducted in H929 plasmacytoma cells assessed by 7-aminoactinomycin D (7-AAD) staining. Toll-like Receptor 4 (TLR4) stimulation was performed in human PBMC from subjects with SLE and SSc using lipopolysaccharide (LPS). Results: In the competitive CRBN binding studies, CC-220 had a 50% inhibitory concentration (IC50) of approximately 0.1 ␮M, compared with 3 ␮M for pomalidomide. CRBN ubiquitination studies in the transfected HEK293T cells resulted in an IC50 of 0.19 ␮M. CC-220 costimulated IL-2 production by T cells with an EC50 of approximately 0.29 nM, compared with 10 nM for pomalidomide. CC-220 inhibited IgM and IgG production with an IC50 of 0.35 and 2.1 nM, respectively, compared to 17 nM and 63 nM for pomalidomide. The IC50 value for inhibition of proliferation by CC-220 was 0.01 ␮M in the H929 plasmacytoma cell line. A 50% decrease in cell cycle (S-phase) was evident after 24 hours of treatment of H929 cells with CC-220. At 48 hours, CC-220 decreased expression of survivin and retinoblastoma protein (pRB) and increased expression of the cyclin-dependent kinase inhibitor p27. In LPS-stimulated PBMC from systemic lupus erythematosus (SLE) and systemic sclerosis (SSc) patients, CC-220 significantly reduced production of pro-inflammatory cytokines including TNF-␣, IL-1␣, IL-1␤, and IL-6, and increased production of IL-10. Conclusion: The results indicate that CC-220 binds to CRBN and inhibits CRBN ubiquitination CC-220 enhances T cell IL-2 production, but inhibits B cell production of immunoglobulin and reduces plasma cell line proliferation. This compound also reduces the production of several key inflammatory mediators by SLE and SSc patients cells stimulated via the TLR4 pathway. In summary, CC-220 is a novel high affinity CRBN ligand with immunomodulatory effects and is currently in development for the treatment of immune and inflammatory diseases.

Disclosure: P. Schafer, Celgene, 3; L. Capone, Celgene, 3; L. Wu, Celgene, 3; R. Chopra, Celgene, 3.

ACR Poster Session A Biology and Pathology of Bone and Joint (Cartilage and Osteoarthritis) Sunday, October 27, 2013, 8:30 AM–4:00 PM

43 Overexpression Of Lamin A In Mesenchymal Stem Cells Compromises Chondrogenesis and Enhances Adipogenesis. Jesu´s Mateos1, Arancha Landeira1, Alexandre De la Fuente1, Ivań Lesende-Rodrı´guez2, Pablo Fernańdez-Pernas2, Maria Carmen Arufe1 and Francisco J. Blanco3. 1 CIBER-BBN, Zaragoza, Spain, 2Department of Medicine, Area of Anatomy and Human Embryology, University of A Corunã-INIBIC, A Corunã, Spain, 3 Department of Medicine. University of Santiago de Compostela, Santiago de Compostela, Spain.

Disclosure: P. Schafer, Celgene, 3; E. Rychak, Celgene, 3; D. Mendy, Celgene, 3, Celgene, 1; S. Parton, Celgene, 3; L. Capone, Celgene, 3; A. Lopez-Girona, Celgene, 3; D. Cedzik, Celgene, 3; J. Kosek, Celgene, 3; L. H. Zhang, Celgene, 3; R. Chopra, Celgene, 3.

Background/Purpose: Previous work by our group and others indicated that an accumulation of lamin A (LMNA) is associated with the osteoarthritis (OA) chondrocyte phenotype. Mutations of this protein are linked to laminopathies and specifically to Hutchinson-Guilford Progeria Syndrome (HGPS), an accelerated aging disease. Some authors have proposed that a deregulation of LMNA affects the differentiation potential of stem cells. In the present study, we examined the effect of the over-expression of LMNA, on the mesoderm and, specifically, chondrocyte differentiation potential of Mesenchymal Stem Cells (MSCs). Methods: MSCs from human umbilical cord (UC) stroma have previously been isolated, expanded and differentiated towards mesoderm cell lineages. For efficient gene delivery of wt LMNA and GFP (Green Fluorescence Protein) as control, we used a lentiviral expression system. Osteogenic potential was studied by with alizarin red staining and Real-Time PCR of ALP, OC and Runx2 to assess early and late osteogenic differentiation.

42 Inhibition Of B Cell Differentiation To The Plasmablast and Plasma Cell Lineage By CC-220, a Potent Modulator Of The Cereblon E3 Ubiquitin Ligase Complex. Peter Schafer, Lori Capone, Lei Wu and Rajesh Chopra. Celgene Corporation, Summit, NJ. Background/Purpose: CC-220 is an immunomodulatory compound which binds to the CUL4 family E3 ubiquitin ligase complex protein cereblon (CRBN) with high affinity and modulates the ubiquitination of substrate proteins. To explore the effects of CRBN targeting on the differentiation of B cells to the plasmablast and plasma cell lineages, CC-220 was tested in an in vitro model of primary human B cell differentiation. Methods: CD19⫹ peripheral blood human B cells from normal donors, or total PBMC for patients with systemic lupus erythematosus (SLE), were

S17

Sunday, October 27

cultured in the presence of cytokines and stimulatory ligands for one week. Cells were counted, viability assessed, and expression of CD20, CD38, CD44, and CD83 were measured by flow cytometry. Plasmablast lineage factors IRF-4, BLIMP-1, XBP-1, and IgJ, and germinal center markers PAX-5 and BCL-6 were measured by qRT-PCR. Intracellular protein expression was measured by laser scanning cytometry. Secreted immunoglobulins IgG and IgM were measured by ELISA. Results: In B cell cultures, CC-220 decreased the percentage of viable CD20-CD38⫹ plasmablasts on day 7 from 30.4% in control cultures in a dose-dependent manner to 27.3%, 2.1%, and 0.4% at 2 nM, 20 nM, and 200 nM CC-220, respectively. On Day 7, qRT-PCR analysis showed that CC-220 (20 nM) reduced expression of the plasmablast lineage factors IRF-4, BLIMP-1, XBP-1, and IgJ gene expression to 20.5%, 14.3%, 15.1%, and 31.5% of control, respectively (P ⱕ0.001). By intracellular flow cytometry, CC-220 (20 nM) significantly decreased IRF-4 (P⬍0.5), BLIMP-1 (P⬍ 0.05), and XBP-1 (P⬍0.05) protein expression at Day 4, but significantly increased BCL-6 (P⬍0.05) protein expression on Day 7. By laser scanning cytometry on Day 7, CC-220 (20 nM) reduced CD38⫹ cell intracellular protein expression of IRF-4 (Pⱕ0.001), and BLIMP-1 (Pⱕ0.001), and increased BCL-6 expression (Pⱕ0.05) (n⫽3). CC-220 inhibited secreted IgG production with an IC50⫽1.8 nM (n⫽3). In PBMC from SLE patients, CC-220 (20 nM) had similar effects as in normal B cells, reducing BLIMP-1, XBP-1, and IgJ gene expression to 52.8%, 49.2%, and 13.6% of control, respectively (Pⱕ0.001, n⫽3). CC-220 (20 nM) significantly reduced CD38⫹ plasmablast intracellular protein expression of BLIMP-1 (Pⱕ0.01) and IRF-4 (P ⱕ0.001), and increased BCL-6 (Pⱕ0.05) (n⫽3). CC-220 inhibited secreted IgM and IgG production by SLE patient PBMC with IC50s of 0.9 nM and 3.2 nM, respectively (n⫽3). Conclusion: These results demonstrate that targeting of the E3 ubiquitin ligase complex substrate co-receptor CRBN with the small molecule immunomodulator compound CC-220 results in potent inhibition of B cell differentiation to the plasmablast lineage, as shown by a reduction in the percentage of viable CD38⫹ cells, a decrease in BLIMP-1, XBP-1, IRF4, and IgJ gene and protein expression, and inhibition of secreted immunoglobulin production. These data implicate the CUL4-CRBN complex in the differentiation of B cells to the plasma cell lineage furthermore define a novel pathway for treatment of immune inflammatory conditions. CC-220 is currently entering clinical development for the treatment of immune diseases associated with autoantibody production.

41

Sunday, October 27

Adipogenic potential was studied with Oil Red staining and Real-Time PCR of LPL, FABP and ADIPOQ, for early and late adipogenic differentiation. Chondrogenesis and hypertrophy were studied using immunohistochemistry and Real-Time PCR of Aggrecan, MMP-13, Type II Collagen, Type I Collagen and Type I Collagen. Reactive oxygen species (ROS) generation was measured by flow cytometry and oriented migration capacity was tested in a Modified Boyden⬘s chamber assay. Results: We found that over-expression of LMNA by lentiviral gene delivery leads to alterations in differentiation potential. Adipogenic capacity is increased in LMNA-MSCs, accompanied by an accumulation of Fatty Acid Synthase, as revealed by Western-blotting. The chondrogenic potential is defective in LMNA-MSCs, showing a decreased COL2/COL1 ratio and an increase in hypertrophy markers. These cells present lower levels of manganese superoxide dismutase (MnSODM) and an increase of mitochondrial MnSODM-dependent reactive oxygen species (ROS). This stable extra ROS generation is accompanied by alterations in the oriented migratory capacity of these cells under the effect of pro-inflammatory cytokines. ROS synthesis was partially and totally reverted by incubation with the ROS scavenger N-Acetyl Cystein (NAC) for 1 hour in culture. In addition, defects in chondrogenesis detected by immunohistochemistry and Real Time-PCR are partially reversed by periodic incubations with NAC for 1 hour. Conclusion: Overall, our results show that chondrogenic differentiation of UC-MSCs is compromised by Lamin A deregulation whereas adipogenic differenciation is enhanced. We also demonstrated that this deregulation alters the oxidative stress balance in MSCs, modifies their migratory properties and induces defects in their capacity to differentiate into chondrocytes in our in vitro model. Further experiments are necessary to determine the in vivo significance of those alterations and to explore putative key targets for stem cell therapies in OA and other aging syndromes.

cartilage defects). Leptin responses were significantly higher in the cartilage samples with low SOCS-3 expression than in the samples with high SOCS-3 expression. In the ANOVA model, SOCS-3 was confirmed to negatively explain leptin-induced expression of NO, IL-6, MMP-1, MMP-3, iNOS and COX-2 independently of BMI and age (p ⫽ 0.008; 0.009; 0.064; 0.008; 0.001; 0.006, respectively). Accordingly, downregulation of SOCS-3 by siRNA in H4 chondrocytes enhanced leptin-induced expression of NO, iNOS, IL-6, MMP-3 and MMP-13. Conclusion: Previous studies have shown that leptin levels are increased in obesity and leptin has detrimental effects in cartilage. The present results show that SOCS-3 negatively regulates catabolic and proinflammatory effects of leptin in cartilage, suggesting that SOCS-3 could be a future drug target in the treatment or prevention of obesityinduced OA. Disclosure: A. Koskinen, None; K. Vuolteenaho, None; R. Korhonen, None; T. Moilanen, None; E. Moilanen, None.

45 WITHDRAWN

46 Alarmins S100A8/A9 Regulate Osteophyte Formation In Experimental Osteoarthritis With High Synovial Activation. Rik Schelbergen1, Arjen B. Blom1, Wouter de Munter1, Annet W. Sloetjes1, Thomas Vogl2, Johannes Roth2, Wim B. van den Berg3 and Peter L.E.M. van Lent1. 1Radboud University Nijmegen Medical Centre, Nijmegen, Netherlands, 2University of Muenster, Muenster, Germany, 3Rheumatology Research and Advanced Therapeutics, Department of Rheumatology, Radboud University Nijmegen Medical Centre, Nijmegen, Netherlands.

Disclosure: J. Mateos, None; A. Landeira, None; A. De la Fuente, None; I. Lesende-Rodrı´guez, None; P. Fernańdez-Pernas, None; M. C. Arufe, None; F. J. Blanco, None.

Background/Purpose: Osteophytes are cartilage capped, bony outgrowths that limit joint movement and originate from the periosteum or from ligaments during osteoarthritis (OA). There is increasing belief that the synovium contributes to OA joint pathology since a large subset of OA patients shows synovial activation. As shown recently in our lab, alarmins S100A8 and S100A9 (major products of activated synovial macrophages) are involved in cartilage degradation and synovial activation during human and murine OA. In the current study, we explored the involvement of S100A8/A9 in osteophyte formation in experimental OA Methods: Experimental OA was elicited in C57Bl/6 (WT) mice and S100A9⫺/⫺ mice (in which peripheral myeloid cells lack functional S100A8), either by intra-articular collagenase injection (CIOA), or by transsection of the medial anterior meniscotibial ligament (DMM). Osteophyte size was assessed by a blinded observer using image analysis software. Chondrogenesis was induced by bringing human fetal mesenchymal stem cells (hMSCs) in pellet or murine C3H10T1/2 in micromass culture and stimulating for 5 (hMSCs) or 21 days (C3H10T1/2) with BMP-2 and TGF␤1, with 1 or 5 ␮g/ml human or mouse recombinant S100A8. Proteoglycan content was quantified on SafO stained sections, expression of mRNA with RT-qPCR Results: Synovial activation, which is high in CIOA, was significantly reduced in S100A9⫺/⫺ mice. Osteophyte size at day 42 of CIOA was dramatically reduced in the S100A9⫺/⫺ compared to WT in the medial collateral ligament (92,5% reduction, Figure 1), but also significantly at the medial side of both tibia and femur (68,2% and 64,6% reduction) (n⫽10). One explanation for the reduced osteophyte size in S100A9⫺/⫺ mice may be a direct effect of S100-proteins on chondrogenesis. To investigate this, we first stimulated murine C3H10T1/2 MSCs in micromass culture with 5 ␮g/ml S100A8 (in the presence of BMP-2 and TGF␤1) and found a marked increase in MMP3 and aggrecan mRNA as well as a strongly altered morphology, indicating increased remodeling. In line with that, stimulation of human MSCs in pellet culture with 1 or 5 ␮g/ml S100A8 (again together with BMP-2 and TGF␤1) strongly increased proteoglycan deposition as measured by redness in SafO staining (27% and 71% increase respectively). Finally, we determined osteophyte size in the DMM model, in which synovial involvement is very low. At day 56, we observed no significant differences in osteophyte size between the S100A9⫺/⫺ and WT at the medial femur and tibia (105% and 136% of WT, n⫽8).

44 Suppressor Of Cytokine Signaling 3 Negatively Modulates LeptinMediated Catabolic and Proinflammatory Effects In Cartilage - New Potential Mechanism To Target Obesity-Induced Osteoarthritis. Anna Koskinen, Katriina Vuolteenaho, Riku Korhonen, Teemu Moilanen and Eeva Moilanen. The Immunopharmacology Research Group, University of Tampere School of Medicine and Tampere University Hospital, Tampere, Finland. Background/Purpose: Adipocytokine leptin has been suggested to link obesity and osteoarthritis (OA). Initially leptin was found to regulate energy metabolism through central nervous system. More recent studies have shown that leptin is also a proinflammatory factor in arthritis and has detrimental effects on cartilage including upregulation of proinflammatory and catabolic factors. However, we and others have found a significant variation in the leptin responses in cartilage samples between different donor patients. One of the factors that regulate the metabolic effects of leptin in hypothalamus is suppressor of cytokine signaling 3 (SOCS-3). SOCS-3 is also known as an important negative feedback mechanism of inflammatory signals in leukocytes. The aim of the present study was to investigate SOCS-3 as a possible modulator of leptin’s detrimental effects in cartilage. Methods: Cartilage samples from 97 OA patients undergoing knee replacement surgery were collected. Cartilage explants were cultured with leptin (10 ␮g/ml) for 42 hours. The expression of SOCS-3, inducible nitric oxide synthase (iNOS) and cyclooxygenase (COX)-2 were measured by Western blotting. Matrix metalloproteinase (MMP)-1, MMP-3 and IL-6 were measured in the culture media by ELISA, and nitric oxide (NO) by the Griess reaction. The results were analyzed in an ANOVA model allowing for intergel variation of SOCS-3, and adjusting for BMI and age. The role of SOCS-3 in leptin signaling was further studied in H4 chondrocytes by downregulating SOCS-3 with siRNA. Results: Leptin significantly enhanced the expression of NO, IL-6, MMP-1, MMP-3, iNOS and COX-2 in the cultured cartilage samples (fold of increase: 2.7 (6.2); 8.3 (30.5); 2.8 (3.4); 1.8 (1.2); 11.7 (160); 6.9 (18); median (IQR), respectively). There was a considerable variation in these responses between the cartilage samples from different donor patients, which was not explained by any clinical factor measured (BMI, age, sex, diabetic status, radiographic scaling of OA or macroscopic scaling of

S18

Sunday, October 27

Conclusion: Exogenous H2S donors NaHS and GYY4137 show antiinflammatory, anti-catabolic and pro-anabolic properties in an in vitro model with human OA chondrocytes. However, in the conditions of the present study, they do not reduce pathological ROS levels found in OA chondrocytes and do not increase the anti-oxidant capacity of the cells.

Conclusion: S100A8/S100A9 play a crucial role in osteophyte formation in an OA model that shows high synovial involvement, probably by stimulating chondrogenesis. Considering also the deleterious effect of S100A8/A9 on joint destruction in OA, targeting these alarmins during OA may be very promising.

Disclosure: E. F. Burguera, None; A. Vela-Anero, None; R. Meijide-Failde, None; F. J. Blanco, None.

48

Disclosure: R. Schelbergen, None; A. B. Blom, None; W. de Munter, None; A. W. Sloetjes, None; T. Vogl, None; J. Roth, None; W. B. van den Berg, None; P. L. E. M. van Lent, None.

The In Vivo Role Of Bone Specific EphB4 Receptor Overexpression In Osteoarthritic Synovial Membrane. Gladys Valverde-Franco1, David Hum1, Bertrand Lussier2, Koichi Matsuo3, Jean-Pierre Pelletier1, Mohit Kapoor1 and Johanne Martel-Pelletier1. 1Osteoarthritis Research Unit, University of Montreal Hospital Research Centre (CRCHUM), Montreal, QC, 2 Department of Clinical Sciences, Faculty of Veterinary Medicine, University of Montreal, Saint-Hyacinthe, QC, 3Laboratory of Cell and Tissue Biology, School of Medicine, Keio University, Tokyo, Japan.

47 Exogenous Hydrogen Sulfide Donors Show Anti-Catabolic and AntiInflammatory Properties But Limited Anti-Oxidant Capability On Human Articular Osteoarthritic Chondrocytes. Elena F. Burguera1, Angela Vela-Anero1, Rosa Meijide-Failde2 and Francisco J. Blanco1. 1Tissue Engineering and Cellular Therapy Group (CBTTC-CHUAC), CIBER BBN/ ISCIII, A Corunã, Spain, 2Department of Medicine, University of A Corunã, A Corunã, Spain.

Background/Purpose: Osteoarthritis (OA) is characterized by progressive destruction of all joint tissues including inflammation of the synovial membrane. The histological changes that occur in OA present a range of abnormalities in the synovial membrane including fibrosis. Members of the ephrin family, the EphB4 receptor and its specific ligand ephrin-B2, were found to positively impact the abnormal structure and metabolism of OA subchondral bone and cartilage. In the context of evaluating the in vivo effect of the EphB4 receptor, we further investigated in vivo in mice the effect of bone-specific EphB4 overexpression (TgEphB4) on synovial membrane during OA. Methods: Knee OA was surgically induced by the destabilization of the medial meniscus (DMM) in 10-week-old male EphB4 homozygous (TgEphB4) and wild-type (WT) mice. Synovial membrane evaluations were performed at 12 weeks post-surgery using histology, immunohistochemistry, and real-time PCR. Results: Data demonstrated a significant decrease in the synovial membrane thickness (pⱕ0.02), pro-collagen type I (pⱕ0.01), and fibrin (pⱕ0.04) in DMM-TgEphB4 compared to DMM-WT. The expression of the fibrotic markers connective tissue growth factor (CTGF, pⱕ0.02), smooth muscle actine ␣ (SMA␣, pⱕ0.03) and serum cartilage oligomeric matrix protein (COMP, pⱕ0.03) were all significantly reduced in DMM-TgEphB4 compared to DMM-WT. Although the TGF-␤ was decreased in the DMMTgEphB4 mice, the difference did not reach statistical significance. However, the synthesis of a member of the heat shock protein family (HSP), HSP90␤, known to have a crucial role in enhancing TGF-␤ signaling, was significantly decreased (pⱕ0.03) in DMM-TgEphB4. Conclusion: This is the first in vivo evidence showing that protecting the subchondral bone prophylactically reduces the severity of pathological changes in the synovial membrane during the OA process. This study thus stresses the in vivo importance of subchondral bone biology in OA tissues. It also provides evidence that changes in the synovial membrane are an integral part of the OA disease process. In addition, these data define the EphB4 receptor as a potential novel therapeutic avenue for the treatment of the disease.

Background/Purpose: Osteoarthritis (OA) is characterized by an imbalance between catabolism and anabolism, higher than normal levels of matrix metalloproteinases (MMPs), an increase in inflammatory markers (i.e. interleukins (ILs) or eicosanoids), and an increase in reactive oxygen species (ROS). Hydrogen sulfide (H2S) is a novel endogenous gas and signaling molecule. Here we looked into the effects of H2S donors as anti-catabolic, anti-oxidant and anti-inflammatory agents in human articular chondrocytes (hCs) from OA tissue. Methods: We used an in vitro model in which hCs were stimulated with one of two H2S donors: NaHS or GYY4137; 50mM to 1000mM; 48h; H2S-only-treated (S)-group. In some tests, IL1␤ (5ng/mL) or Lipopolysaccharide (LPS, 1␮g/mL) were also added. We chose MMP1, 3 and 13, collagen-II (COLII) and aggrecan (ACAN) as matrix degradation and synthesis markers. The prostaglandin E2 (PGE2) synthesis pathway and IL6, for inflammation. Superoxide dismutase(SOD)-2 and catalase (CAT) as scavenger enzymes. ROS synthesis (mitochondrial(M)-ROS and cytoplasmatic(C)-ROS) was measured with cell cytometry. Effects on gene expression were quantified with qRT-PCR and those on proteins with immunocytochemistry (ICC) or enzymatic immunoassays (EIA). Results: 1. Catabolism: The two H2S donors reduced MMP3 and MMP13 expression in the S-group, although p⬎0.05. On IL1␤-group, MMP3 and 13 (Fig.1A) expressions in the IL1␤ condition (182 and 52 fold vs. the basal (B)), were reduced to 124 and 10 (1000mM GYY4137, respect.) and 91 and 13 (1000mM NaHS, respect.). There was a marked reduction in MMP3 protein levels (ICC). 2. Anabolism: 200mM NaHS and GYY4137 counteracted COLII and ACAN repressed expression on the IL1␤-group. 3. Inflammation: H2S donors reduced IL-6 (Fig.1B), cyclooxygenase (COX)-2 mRNA expression in the IL1␤-group without significantly affecting COX-1. 4.ROS: Only 200mM NaHS reduced C-ROS in the S-Group (from 27.7⫾5.7 to 21.5⫾1.9UA). On the IL1␤-group, H2S donors did not reduce ROS. SOD2 (Fig.1C) and CAT expression levels were unaffected (p⬎0.05) in either group. Only 200mM GYY4137 reduced M-ROS in the LPS-Group to B level.

Disclosure: G. Valverde-Franco, None; D. Hum, None; B. Lussier, None; K. Matsuo, None; J. P. Pelletier, None; M. Kapoor, None; J. Martel-Pelletier, None.

S19

this study we aim to investigate the matrix changes and calcification of cartilage during the development of OA- like changes in mice. Methods: The tip-toe walking (ttw/ttw) mouse that carries a mutation in the enpp1 gene encoding for NPP1 was used as a natural model of OA. Using von Kossa staining in combination with Safranin-orange staining of knee sections we assessed the calcification of articular cartilage and the severity of OA using the Mankin-Score over a time course from 4 to 22 weeks. We analysed the compostion of cartilage matrix using toluidine blue, alcian blue/ PAS staining. To investigate cartilage matrix remodelling we performed immunohistological stainings for collagen II, collagen X and the aggrecan cleavage marker BC-3. We performed the same stainings in knee joint sections of wild type mice with induced OA (DMM model). The influence of basic calcium phospahe (BCP) crystals on neonatal chondrocytes was investigated in micro mass cultures with alcian blue and alizarin red staining. Using Western Blot for ␤- catenin and pCamKII we investigated the activation of WNT signalling. Results: We found a loss of proteoglycans in the joint cartilage of ttw/ttw mice already at an age of 8 weeks, whereas no changes of cartilage matrix were detectable in wild type mice. An increased Mankin Score in ttw/ttw compared to wild type mice also reflected these changes. Early matrix remodelling in ttw/ttw cartilage was followed by an increased calcification starting at the border to the subchondral bone. The straight border of the tide mark gets broken down by multiple calcified chondrocytes that give the border a wave-like appearance. Looking at the matrix changes in more detail we found an increase in sulphated proteoglycans in the ttw/ttw cartilage as well as in the induced OA cartilage. The same pattern was observed in the growth plate in the area of hypertrophic chondrocytes. We found an increased expression of collagen X in the cartilage and collagen II was down-regulated with increasing age and OA severity. Using neonatal chondrocytes of wild type and ttw/ttw mice we found that BCP crystals induce production of highly sulphated matrix and reduce proliferation of chondrocytes. Chondrocytes of ttw/ttw mice show an increase in ␤- catenin, which was also observed in wild type chondrocytes stimulated with BCP crystals. Conclusion: We conclude from our data that matrix remodelling precedes cartilage calcification. The calcification seems to be associated with activation of canonical WNT signalling. Taken together, the data support the notion that OA is characterized by the re-initiation of developmental programms associated with endochondral ossification.

Sunday, October 27

49 Chondrocyte-Specific Bone Morphogenetic Protein-2 Overexpression Results In Severe Aggravation Of Osteophyte Formation In Experimental Osteoarthritis Without Altering Cartilage Damage In Young Mice. Esmeralda N. Blaney Davidson1, Elly L. Vitters1, Miranda B. Bennink2, Fons AJ Loo1, Wim B. van den Berg2 and Peter M. van der Kraan1. 1Radboud University Nijmegen Medical Centre, Nijmegen, Netherlands, 2Rheumatology Research and Advanced Therapeutics, Department of Rheumatology, Radboud University Nijmegen Medical Centre, Nijmegen, Netherlands. Background/Purpose: In murine osteoarthritis (OA) models chondrocytes surrounding lesions express elevated levels of Bone Morphogenetic Protein-2 (BMP2). This growth factor is well known for its ECM inducing qualities and capability to induce new cartilage and bone. The functional consequence of these elevated BMP2 levels near OA lesions is unclear. We made a collagen type II dependent, doxycycline-inducible BMP2 transgenic mouse to investigate the consequence of elevated chondrocyte-specific BMP2 on experimental OA (DMM-model). Methods: We cloned a lentivirus with BMP2 controlled by minCMV promoter coupled to a tet responsive element (Lv-TRE-BMP2).We crossed a Col2-cre with a floxed rtTA mouse and transfected homozygote embryo’s with Lv-TRE-BMP2 to gain a mouse expressing collagen type 2 dependent BMP2 solely in chondrocytes only upon doxycycline (dox) exposure (Col2rtTA-TRE-BMP2). Experimental OA was induced (DMM model) in Col2rtTA-TRE-BMP2 mice with or without dox exposure in food starting one week before DMM, lasting until the end of the experiment (8 weeks after DMM). We isolated knee joints for histology. Left knee joints served as non-OA controls. Results: Mice with DMM alone showed osteophyte formation predominantly on the medial side of the joint originating from the periost near the femoral-tibial joint and extruding from the medial meniscus. Upon chondrocyte-specific BMP2 exposure, DMM-induced osteophyte formation aggravated severely resulting in outsized osteophytes, enthesophytes on the enthesis of the medial collateral ligament, new bone formation within the collateral ligament near the extruded meniscus and osteophyte formation originating from the femoral growth plate, predominantly but not exclusively on the medial side of the joint. There were no significant differences in non-OA knee joints comparing dox versus non-dox treated transgenics. In contrast, dox treatment resulted in large osteophytes in thoracic or cervical area of the spine. Strikingly, despite apparent changes in knee joint morphology due to large osteophytes there was no detectible difference, with regard to structural damage and Safranin O staining intensity, in cartilage damage when comparing DMM with or without dox exposure. Conclusion: Elevated BMP2 levels in chondrocytes did not induce structural changes in articular cartilage of young mice. Moreover, unchallenged dox treated joints had no structural alterations in cartilage. In DMM, BMP2 overexpression greatly enhanced de novo cartilage formation, eventually turning into bone (osteophytes), which was not seen in non-OA knee joints with BMP2. We postulate that newly-induced chondrocytes produce high BMP2 levels in dox-treated transgenics, that boost outgrowth of these structures. In spine however, osteophytes developed in dox-treated transgenics without additional triggers. This could rely on chondrocyte precursor subpopulation that is stimulated to undergo chondrogenesis under control of BMP2 alone. Our data show that chondrocyte-specific elevation of BMP2 levels does not alter the course of cartilage damage in an OA model in young mice but results in severe aggravation of osteophyte formation.

Disclosure: J. Bertrand, None; T. Kra¨ft, None; Y. Nitschke, None; T. Hawellek, None; J. Hubert, None; L. Godmann, None; T. Pap, UCB, Servier, Abbott, Bioiberica, MSD, Pfizer, Eli Lilly, 5, UCB, Servier, Abbott, Bioiberica, MSD, Pfizer, Eli Lilly, 8.

51 Collagen-Based Microspheres Delivering TGF-␤3 For Mesenchymal Stem Cell Differentiation: An Innovative Strategy For Cartilage Engineering. Marc Mathieu1, E Belamie2, M-N Labour2, Sylvain Vigier2, Christian Jorgensen3 and Daniele Noel4. 1INSERM U844, Montpellier, France, 2ICGM UMR 5253, MONTPELLIER, France, 3Inserm U844, CHU Saint-Eloi, Universite´ Montpellier 1, CHU Lapeyronie, Montpellier, France, 4 UM1, Montpellier, France. Background/Purpose: Because of a poor self-healing ability, joint cartilage undergoes progressive degradation in the course of aging or following traumatic injuries. One promising therapeutic approach is the use of mesenchymal stem cells (MSC), which have the potential to differentiate into chondrocytes. However, anchorage of MSC to the sites of injury and their differentiation in situ require combining cells with a biomaterial releasing a differentiation factor such as TGF␤3. Our objective was to design and test a new scaffold to support chondrogenic differentiation. Methods: Microspheres were formed using an acidic type I collagen solution and perfluorated oil, stabilized by a surfactant. Spherical microparticles of fibrillar collagen were obtained through a sol-gel transition induced in aqueous droplets in contact with ammonia vapors. Microspheres were impregnated with TGF␤3 and release of this factor was measured using a reporter gene assay. Impregnated microspheres were combined with human MSC and cultivated in vitro or subcutaneously injected into immunodeficient mice. Expression of chondrocyte markers was monitored by RT-qPCR and immunohistochemistry. Results: Microspheres are constituted by a gel of striated collagen fibrils and are 300⫾67 ␮m in diameter. Fibrils occupy ca. 5% of the total volume, forming an entangled network with pores of 1–10 ␮m. After MSC adhesion onto the microspheres and in vitro culture for 21 days, they differentiate into chondrocytes expressing the specific markers, such as collagen type II variant

Disclosure: E. N. Blaney Davidson, None; E. L. Vitters, None; M. B. Bennink, None; F. A. Loo, None; W. B. van den Berg, None; P. M. van der Kraan, None.

50 Cartilage Matrix Remodeling and Activation Of Canonical Wnt- Signaling Precedes Calcification and OA-Like Changes In Mice. Jessica Bertrand1, Tabea Kra¨ft2, Yvonne Nitschke2, Thelonius Hawellek3, Jan Hubert3, Lars Godmann1 and Thomas Pap1. 1University Hospital Mu¨nster, Mu¨nster, Germany, 2University Hospital Mu¨nster, Muenster, Germany, 3 University Hospital Hamburg, Hamburg, Germany. Background/Purpose: Calcification of cartilage is a common finding during osteoarthritis (OA) and is directly linked to the severity of cartilage degradation. We have found in a previous study that NPP1 is an important player in OA associated cartilage calcification in mouse and humans. The observed cartilage changes resemble aspects of endochondral ossification. In

S20

B and aggrecan. The type I collagen matrix is progressively degraded and replaced by the cartilaginous matrix. In vivo, MSC form a tissue histologically resembling cartilage which stain positive for collagen II and aggrecan. Conclusion: The biomaterial described here is promising for cartilage engineering. Future improvements aim at obtaining microspheres of smaller and more homogenous sizes, which will facilitate their injection; covalently linking the TGF␤3 to the microspheres to prevent its dispersal in the organism.

53

Background/Purpose: Osteoarthritis, the most common type of arthritis in the United States, results from trauma and other mechanical factors as well as metabolic changes in bone and cartilage. We have previously demonstrated that adenosine, acting via the A2A receptor, inhibits inflammation and plays a critical role in regulating bone metabolism. Thus, adenosine A2A receptor knockout mice are osteopenic as a result of increased osteoclast number and activity. We have observed that aging adenosine A2A receptor knockout mice experience difficulty in movement, taking food and walking and so we determined whether changes in their bone or joint structure or function could explain these changes. Methods: 4 month old C57Bl/6 wild type (WT) and A2AKO mice (n⫽4) were sacrificed and knee joints were prepared for microCT analysis and histology. PAS (Periodic Acid Staining) and Trichrome staining of decalcified sections of leg were carried out. MicroCT analysis of knees of WT and KO mice were carried out and analysis of bone was performed on the distal femur below the growth plate. Results: microCT analysis of the knees of A2AKO mice showed osteophyte formation together with mild remodeling and subchondral sclerosis when compared to WT mice. As we have previously reported A2A KO mice suffered from osteopenia; bone volume/total volume (BV/TV) was significantly decreased in A2AKO mice when compared to control (33.324⫾0.56 vs 35.782⫾0.78, respectively, p⬍0.01). The same decrease was observed for trabecular thickness (0.0685⫾0.0035 vs 0.081⫾0.002, p⬍0.05) and bone mineral density (BMD) (0.3795⫾0.003 vs 0.4265⫾0.01, respectively, p⬍0.5), with no change in trabecular number (4.8795⫾0.17 vs 4.646⫾0.23, p⫽ns). H&E and trichrome staining showed a loss in collagen in the cartilage together with diminished subchondral bone and chondrocyte hypertrophy and proliferation. PAS staining correlated with these results showing a loss in proteoglycans both in the articular cartilage and in the growth plate in A2AKO mice together with an increase in chondrocyte proliferation. Conclusion: Deficiency in adenosine A2A receptors leads to spontaneous osteoarthritis and suggests that adenosine receptors may be novel targets for development of therapies to ameliorate or prevent osteoarthitis.

Disclosure: M. Mathieu, None; E. Belamie, None; M. N. Labour, None; S. Vigier, None; C. Jorgensen, None; D. Noel, None.

52 Altered Lipid Metabolism In Osteoarthritis With Subsequent Proinflammatory Properties Of Apolipoprotein A-I. Dominique de Seny1, Gae¨l Cobraiville1, Edith Charlier1, Sophie Neuville1, Laurence Lutteri2, Denis Malaise3, Olivier Malaise1, Jean-Paul Chapelle2, Biserka Relic1 and Michel G. Malaise1. 1GIGA Research - University of Lie`ge - CHU Lie`ge, Lie`ge, Belgium, 2Medical Chemistry - CHU Lie`ge, Lie`ge, Belgium, 3University of Lie`ge, Lie`ge, Belgium. Background/Purpose: Osteoarthritis (OA) is associated with a local inflammatory process. It is now considered as a metabolic syndrome rather than due to aging or mechanical stress. Several evidences point to the direction of an altered lipid metabolism as an underlying cause for the development of OA. Recently, we have studied the role played by an apolipoprotein, the acute phase serum amyloid A (A-SAA), as a proinflammatory marker in OA joints (de Seny et al., Plos One, 2013). Apolipoprotein A-I (ApoA1) is another major protein component of highdensity lipoprotein (HDL) cargo molecules in plasma, and both are playing a major role in back cholesterol transport from peripheral tissues to the liver. Methods: Lipoproteins, anti-oxidized LDL antibodies, cholesterol, ApoA1 levels and inflammatory parameters (IL-6, MMP-1 and MMP-3) in blood and synovial fluids of OA (n⫽29) and rheumatoid arthritis (RA) (n⫽27) patients were quantified and compared to those in matched healthy volunteers (HV) (n⫽35). Primary chondrocytes and fibroblast-like synoviocytes (FLS) were isolated respectively from cartilage and synovial membrane obtained from OA patients during joint replacement. Cells were stimulated with purified human ApoA1 in the presence or not of recombinant human A-SAA (rhSAA) protein, and with lipoproteins at physiological concentration encountered in OA. IL-6, MMP-1 and MMP-3 expression levels were quantified by ELISA after stimulation. Results: 1) In the serum of OA patients, LDL/HDL ratio was significantly higher compared to HV and RA but remained similar in both OA and RA synovial fluid. 2) Although LDL and cholesterol serum levels were higher in OA than in RA, both were lower in the OA synovial fluid than in the RA. 3) In the OA and RA synovial fluid, cholesterol levels were positively correlated to LDL, HDL and ApoA1 levels, and HDL levels were positively correlated to ApoA1 levels. But of interest, OA synovial fluid had the unique characteristic of LDL levels being positively correlated to HDL and ApoA1 levels, which was not observed neither in the synovial fluid of RA patients nor in the serum of OA, RA and HV. 4) LDL and ApoA1 levels were also significantly correlated to IL-6 levels, another uncommon characteristic of OA synovial fluid suggesting local dysregulated processes within lipidic and inflammatory parameters. We also observed using in vitro experiments that the purified human ApoA1 likewise rhSAA induced IL-6, MMP-1 and MMP-3 expression in primary chondrocytes and FLS obtained from OA patients. ApoA1-induced IL-6, MMP-1 and MMP-3 expression was downregulated by TAK242, a specific TLR4 inhibitor. Presence of HDL at OA physiological concentration did not abolished ApoA1-induced IL-6, MMP-1 and MMP-3 expression. Conclusion: Lipid diffusion into the joint cavity is dependent on the degree of inflammation. If inflammation is largely superior in RA compared to OA, we can nonetheless hypothesize that local inflammatory process and lipid diffusion inside OA joint cavity can also occur. In this study, several arguments have been raised in favour of an abnormal lipid profile in OA synovial fluid. This abnormal lipid profile was linked to the local proinflammatory process.

Disclosure: A. Mediero, Filed a patent on use of adenosine A2AR agonists to prevent prosthesis loosening (pending)., 9; T. Wilder, None; B. N. Cronstein, Canfite Pharma, 1, NIH, Gilead, Takeda, AstraZeneca, 2, NYU School of Medicine, 3, MerckSeronoBristol-Myers Squibb, Novartis, CanFite Biopharmaceuticals, Cypress Laboratories, Regeneron (Westat, DSMB), Endocyte, Protalex, Allos, Inc., Savient, Gismo Therapeutics, Antares Pharmaceutical, Medivector, 5, Multiple patents on adenosine receptors and bone metabolism, pharmacology, 9.

54 Anti-Fibrotic and ANTI-Hypertrophic Effect Of Adipose Mesenchymal Stromal Cells On Osteoarthritic Chondrocytes. Marie Maumus1, C. Manferdini2, Karine Toupet1, A. Piacentini2, A. Gabusi2, A. Facchini2, G. Lisignoli2, Christian Jorgensen3 and Daniele Noel4. 1INSERM U844, Montpellier, France, 2IOR BOLOGNA, BOLOGNA, Italy, 3Inserm U844, CHU Saint-Eloi, Universite´ Montpellier 1, CHU Lapeyronie, Montpellier, France, 4 UM1, Montpellier, France. Background/Purpose: Mesenchymal stem cells (MSC) isolated from bone marrow (MSC) or adipose (ASC) tissues secrete a large amount of trophic factors. The possibility that these cells, through their paracrine potential, may influence the course of chronic degenerative disorders and prevent cartilage degradation is promising for the treatment of osteoarthritis (OA). Indeed, the aim of our work was to evaluate the effects of these cells on OA chondrocyte phenotype. Methods: OA ASC were isolated from intra-articular (Hoffa-ASC) or hip (hip ASC) subcutaneous adipose tissue and healthy ASC from subcutaneous abdominal depot (abdo-ASC). BM-MSC and chondrocytes were obtained from OA donors. ASC or MSC were co-incubated with chondrocytes cultured in monolayer during 7 days using cell culture inserts. Expression of markers specific for mature and hypertrophic chondrocytes or fibroblasts was quantified by RT-qPCR or ELISA. Isotypic or anti-HGF antibodies (100ng/ml) were added during the coculture. Results: After 7 days of ASC or MSC co-cultures with chondrocytes, a stable expression of the markers specific for mature chondrocytes (Col IIB, Agg, link, Sox9) was observed, while expression of hypertrophic (MMP13,

Disclosure: D. de Seny, None; G. Cobraiville, None; E. Charlier, None; S. Neuville, None; L. Lutteri, None; D. Malaise, None; O. Malaise, None; J. P. Chapelle, None; B. Relic, None; M. G. Malaise, None.

S21

Sunday, October 27

Adenosine A2A Receptor As a Potential New Therapeutic Target For The Prevention/Treatment Of Osteoarthritis. Aranzazu Mediero1, Tuere Wilder1 and Bruce N. Cronstein2. 1NYU School of Medicine, New York, NY, 2NYU School of Medicine, Division of Rheumatology, New York, NY.

Sunday, October 27

degradation of articular cartilage. TGF␤-superfamily signaling plays an important role in cartilage homeostasis via induction of Smad phosphorylation (pSmad). pSmad2/3 is a potent inhibitor of chondrocyte terminal differentiation and cell death, whereas pSmad1/5/8 induces these processes. In cartilage, TGF␤ receptors that induce Smad phosphorylation are ALK1 for Smad1/5/8 and ALK5 for Smad2/3. We showed an age-related shift in receptor balance towards dominant Alk1 expression compared to Alk5 expression, and linked elevated Alk1 levels to increased Mmp13 and ColXexpression and cartilage damage in mice. However, OA development and ageing are concurrent in mice, and possibly both processes interfere. To uncouple ageing and OA-related processes, we investigated if age-related changes in TGF␤-superfamily signaling occur without interference of the OA process, and if these changes result in an altered cellular response to growth factors. We chose healthy bovine cartilage explants as a model system because bovine samples can be obtained in a wide age range without concurrent OA development. Methods: From the metacarpophalangeal joint of cows (Bos Taurus) (⬍ 3 h post mortem), full thickness cartilage explants were collected. Macroscopically none of the animals showed any signs of OA. Samples were either immediately frozen in liquid nitrogen or, after 24 h equilibration in serum free medium, stimulated with growth factors for 24 h. Subsequently, mRNA was isolated and gene expression was measured by qPCR, using cDNA specific primers and SYBR-green. Results: Gene expression analysis of bovine explants ranging from 0.5 up to 12 years old showed a very significant 100-fold (R2 ⫽ 0.7, p ⬍ 0.0001) decrease in Col2a1 expression and a 3-fold decrease in Tgfb1 (R2 ⫽ 0.3, p ⫽ 0.002) expression with increasing age compared to 4 reference genes; Gapdh, Rpl22, Rps14 and Gusb. Furthermore, Alk5 expression decreased 4-fold (R2 ⫽ 0.3, p ⫽ 0.0004) whereas Alk1 expression remained unchanged. In 8 year old explants, expression of the Smad2/3p response gene Serpine1 (Pai1) was 11-fold less (p ⫽ 0.01) responsive to 1 ng/ml of the ALK5 ligand TGF␤1 compared to 1 year old explants. In contrast, the response of 8 year old cartilage to the ALK1 ligand BMP9 (5 ng/ml) was not significantly different compared to 1 year old cartilage as measured by expression of the Smad1/5/8 response gene Id1. Conclusion: The loss of Col2a1 and Tgfb1 expression are two characteristics of ageing cartilage, and confirm our system as valid to study age-related changes in cartilage. Because no signs of OA could be detected on macroscopic level, changes in TGF␤-superfamily receptor expression appear to occur independently of the OA process. The observed decrease in Alk5 expression results in loss of Smad2/3p signaling in response to TGF␤1, indicating that TGF␤1’s cartilage protective function is lost with increasing age. We propose that this loss of response predisposes chondrocytes for the OA process.

AP) and fibrotic (Col I and III) markers was significantly decreased. Compared to abdo-ASCs, Hoffa- and Hip-ASC reduced less efficiently the expression of hypertrophic/fibrotic markers and some markers of mature chondrocytes were decreased with Hip-ASC. Finally, factors known to be involved in fibrosis and matrix remodeling (TIMP-1 and -2, MMP-1 and -9, IL1-RA) were not changed. On the contrary, HGF secretion was induced and addition of neutralizing anti-HGF antibody reversed the anti-fibrotic effect of ASC whereas the hypertrophic markers were not modulated. Conclusion: ASC from abdominal subcutaneous fat and MSC were the most efficient to reduce hypertrophy and dedifferentiation of articular chondrocytes. This effect was at least partly due to the induction of HGF secretion confirming the interest of using ASC in therapies of osteo-articular diseases. Disclosure: M. Maumus, None; C. Manferdini, None; K. Toupet, None; A. Piacentini, None; A. Gabusi, None; A. Facchini, None; G. Lisignoli, None; C. Jorgensen, None; D. Noel, None.

55 Glucocorticoid Receptor Modulator Compound A Does Not Induce Leptin In Human Osteoarthritic Synovial Fibroblasts and In Dedifferentiated Chondrocytes. Olivier Malaise, Biserka Relic, Edith Charlier, Mustapha Zeddou, Florence Quesada-Calvo, Sophie Neuville, Dominique de Seny and Michel G. Malaise. GIGA Research - University of Lie`ge - CHU Lie`ge, Lie`ge, Belgium. Background/Purpose: Leptin is generally considered detrimental in osteoarthritis (OA) as it can induce IL-8 in synovial fibroblasts (SF) and matrix metalloproteinases in chondrocytes. We recently showed that SF themselves were able to produce leptin and express leptin receptor (Ob-R) in vitro, strongly induced by glucocorticoid (GC) prednisolone, through TGF-␤ pathway and Smad1/5 phosphorylation (Stem Cell Dev 2012; 21: 1948–55). It is generally believed that transrepression of transcription factors by monomeric GC receptor (GR) is associated with GC anti-inflammatory properties, whereas side-effects (mellitus diabetes, lipid abnormalities, osteoporosis) would be mostly resulting in GR dimer-dependent transactivation. Compound A (CpdA), a plant-derived phenyl aziridine selective GR agonist, is efficient to reduce inflammation in a murine arthritis model and does not induce mellitus diabetes nor osteoporosis, offering a better benefit/risk ratio than GC. The aim of this study is to determine, in human OA SF and dedifferentiated chondrocytes (DDC), whether CpdA, as prednisolone, is leptin and Ob-R inducer. Methods: Human SF and chondrocytes were isolated from OA patients during knee or hip surgery. Cells were treated with prednisolone (1 ␮M), TNF-␣ (10 ng/ml) and/or CpdA (1, 5 or 10 ␮M). ELISA measured leptin and IL-6 production in culture supernatant. Ob-R, GR, TGF-␤RII, phosphoSmad1/5, phospho-Smad2, Smad1, Smad2, GILZ and ␣-tubulin were analyzed in total cell extracts by Western Blot. Results: 1. Like prednisolone, CpdA down-regulated endogenous and TNF-␣-induced IL-6 secretion by SF and DDC in a dose-dependent manner, supporting its anti-inflammatory action. 2. By contrast to prednisolone, CpdA did not induce leptin and further decreased endogenous Ob-R expression in a dose-dependent manner in the same cells. Moreover, unlike prednisolone, CpdA did not induce TGF-␤RII expression nor Smad1/5 phosphorylation and did not reduce Smad2 phosphorylation. Lastly, also opposite to prednisolone, CpdA poorly induced GR degradation and did not induce GILZ expression. Conclusion: Unlike prednisolone, CpdA, a dissociative glucocorticoid receptor agonist, did not induce leptin nor Ob-R in human OA SF and in DDC and could even decrease Ob-R expression. As leptin favors osteoblastic differentiation of mesenchymal stem cells, our findings might explain why CpdA also fails to induce such a differentiation. Leptin is a new target where dissociative properties of CpdA can apply.

Disclosure: A. van Caam, None; E. N. Blaney Davidson, None; E. Thijssen, None; W. B. van den Berg, None; P. M. van der Kraan, None.

57 Role Of Procatabolic Cytokines In Dysregulation Of O-Linked N-Acetylglucosamine Modified Proteins In Human Osteoarthritic Chondrocytes. Raquel Largo1, Ane Larranaga-Vera2, Sandra Pe´rez-Baos2, Gabriel Herrero-Beaumont2 and Jessica Andre´s-Bergo´s3. 1Jimeńez Dıáz Foundation University Hospital, Madrid, Spain, 2Joint and Bone Research Unit. IIS-Fundacioń Jimeńez Dıáz. UAM, Madrid, Spain, 3Tissue Engineering Repair and Regeneration Program, Hospital for Special Surgery and Weill Cornell Medical College, New York, NY. Background/Purpose: Nuclear and cytoplasmic protein glycosylation by the addition of O-linked N-Acetylglucosamine (O-GlcNAc) to serine and threonine residues is a widespread post-translational modification that has been described in degenerative and age-related diseases, such as Alzheimer⬘s and diabetes. This modification is catalyzed by the O-GlcNAc transferase (OGT) which uses UDP-GlcNAc as a donor, and O-GlcNAc can be removed by the O-GlcNAc-glucosaminidase (OGA). We have seen that O-GlcNAc glycosylation levels in the articular cartilage of patients with OA were increased as well as the expression of the different isoforms of OGT and OGA was dysregulated. The goal of this study was to assess whether a procatabolic cytokine, IL-1␣, can be responsible of the changes in O-GlcNAc system found in OA human cartilage. Methods: OA knee articular cartilage was obtained during replacement surgery. Twelve OA patients were included in this study (6 woman/6 men; mean age 71⫾3 years), with a Mankin score for knee specimens between 11

Disclosure: O. Malaise, None; B. Relic, None; E. Charlier, None; M. Zeddou, None; F. Quesada-Calvo, None; S. Neuville, None; D. de Seny, None; M. G. Malaise, None.

56 Age-Related Loss Of The Transforming Growth Factor Beta Receptor ALK5 Precedes Osteoarthritis Development In Cartilage. Arjan van Caam, Esmeralda N. Blaney Davidson, Eva Thijssen, Wim B. van den Berg and Peter M. van der Kraan. Radboud University Nijmegen Medical Centre, Nijmegen, Netherlands. Background/Purpose: Osteoarthritis (OA) is the most common joint disease and (old) age is its main risk factor. One of OA’s main hallmarks is

S22

experiments showed that palmitate accelerated cartilage destruction driven by IL-1␤ in bovine explants, as evidenced by a significant increase in cell death and glycosaminoglycan release. Conclusion: Saturated fatty acids, specifically palmitate, have catabolic effects on articular cartilage and can accelerate its breakdown in the context of low-grade inflammation by synergizing with IL-1␤ to increase proinflammatory mediators, ER stress and caspase-associated cell death. Collectively, our data suggest that elevated levels of saturated FFA may contribute to OA pathogenesis in obese patients. Disclosure: O. Alvarez-Garcia, None; N. H. Rogers, None; R. G. Smith, None; M. K. Lotz, None.

59 Downregulated Forkhead-Box Class O Transcription Factors Attenuate Oxidative Stress Resistance In Human Chondrocytes. Yukio Akasaki and Martin K. Lotz. The Scripps Research Institute, La Jolla, CA. Background/Purpose: Aging-associated changes in articular chondrocytes represent a major risk factor for osteoarthritis (OA). The major signaling pathway that regulates cellular aging is the Insulin/IGF-1/ Pl3k/Akt/forkhead-box class O (FoxO) transcription factor axis. The FoxO factors play a central role in oxidative stress resistance regulating antioxidants, autophagy and protein quality control. Previously, we observed that FoxO factors are dysregulated in aged and OA cartilage. FoxO expression is reduced during the aging process and phosphorylation of FoxOs (inactivation form) is increased in OA cartilage. However, the impact of downregulated FoxOs on chondrocytes is still unknown. The objective of this study was to investigate the role of FoxOs in antioxidative stress resistance in human chondrocytes Methods: Normal human cartilage was obtained at autopsy from 4 adult donors. OA human cartilage was obtained from 5 patients undergoing knee replacement surgery. Small interference RNA for FoxO1 and FoxO3 were transfected into chondrocytes. The effects of down-regulated FoxO proteins on cell viability against the oxidant tert-Butyl hydroperoxide (t-BHP) were analyzed by MTT assay. Antioxidants and autophagy related proteins in FoxO-siRNA transfected cells were analyzed by Western blot. The changes in inflammatory mediators and extracellular matrix components following treatment with IL-1␤ or TGF-␤ were also examined Results: FoxO1 and FoxO3 but not FoxO4 mRNA expression was significantly lower in OA cartilage compared to normal cartilage. MTT assay showed that the combination of siFoxO1 and siFoxO3 significantly reduced cell viability compared to control siRNA under treatment with t-BHP (ⱖ 50 ␮M) (Figure 1). Knock-down of FoxO1 and FoxO1⫹3 resulted in significant reductions of GPX-1, catalase, LC3, Beclin1, p62, and Sirt1 proteins following treatment with t-BHP. Expression of iNOS was significantly increased in FoxO-siRNA transfected chondrocytes.

Disclosure: R. Largo, None; A. Larranaga-Vera, None; S. Pe´rez-Baos, None; G. Herrero-Beaumont, None; J. Andre´s-Bergo´s, None.

58 Saturated Fatty Acids Act Synergistically With Interleukin 1 Beta To Increase Inflammation, Endoplasmic Reticulum Stress and Cell Death In Human Articular Chondrocytes. Oscar Alvarez-Garcia1, Nicole H Rogers2, Roy G Smith3 and Martin K. Lotz4. 1Department of Molecular and Experimental Medicine, The Scripps Research Institute, La Jolla, CA, 2 California Institute for Biomedical Research (Calibr), La Jolla, CA, 3Department of Metabolism and Aging, The Scripps Research Institute, Jupiter, FL, 4 The Scripps Research Institute, La Jolla, CA. Background/Purpose: Obesity is a major risk factor for the development of osteoarthritis (OA). Altered biomechanics have been postulated as the main causative mechanism, but recent data showing a higher incidence of hand OA in obese patients suggest the existence of a systemic component. Obesity is commonly associated with a state of low-grade inflammation, increased circulating adipokines and free fatty acids (FFA) released by adipose tissue. Whereas low-grade inflammation has received some attention in the recent years, little is known about the role of FFAs in OA pathogenesis. The aim of this study was to analyze effects of saturated (palmitate) and monounsaturated (oleate) FFAs on articular cartilage and chondrocytes. Methods: Human Articular Chondrocytes (hAC) obtained from young healthy donors (21.3⫾2.9 years) were grown in monolayer cultures until confluence. For cell viability and caspase 3/7 activation assays, hAC were treated with 0.5 mM palmitate, 0.5 mM oleate or vehicle, in the presence or absence of 1 ng/ml IL-1␤ for 72 hours. For gene expression studies, shorter treatment time (24-hours) and lower IL-1␤ concentration (10 pg/ml) were used. Levels of proinflammatory factors, extracellular matrix proteins, extracellular proteases and endoplasmic reticulum (ER) stress markers were measured. In addition, bovine articular cartilage explants were cultured with FFA for 72 with or without IL-1␤ and cartilage integrity was studied. Results: Palmitate induced caspase 3/7 activation and cell death in IL-1␤ -stimulated hAC, and significantly upregulated IL-1␤ -induced IL-6 and Cox-2 gene expression as well as IL-6 protein secretion, whereas oleate had no additive effect. Exposure to FFAs did not modify type II collagen, aggrecan, ADAMTS-4, -5, and MMP13 gene expression after 24 hours. ER stress markers were potently increased by palmitate alone, but not oleate, and this effect was further enhanced by IL-1␤. Preliminary

Conclusion: Down-regulated FoxO transcription factors in chondrocytes increased susceptibility to cell death induced by oxidative stress, and this was associated with reduced antioxidants and autophagy related proteins. Our data provide evidence for a key role of FoxO transcription factors as regulators of oxidative stress resistance in human chondrocytes. Disclosure: Y. Akasaki, None; M. K. Lotz, None.

S23

Sunday, October 27

and 14. Human osteoarthritic chondrocytes (HOC) were isolated by protease digestion for primary culture. At confluence, cells were stimulated with different concentrations of IL-1␣ for different doses and periods of time after they were deprived of serum for 48 h. The level of O-GlcNAcylation, OGT and O-GlcNAcase expression was assessed by western-blot employing specific antibodies. Results: The amount of O-GlcNAcylated proteins was increased in HOC 24 h after stimulation with different doses of IL-1␣ (1, 10 and 100 ng/ml), showing the highest accumulation with 10 ng/ml. This concentration was chosen for stimulation of HOC during different periods of time (3, 6, and 24 h). The increase in O-GlcNAc proteins amount showed a peak of induction at 6 h. This accumulation occurs especially in proteins between 50 to 150 KDa therefore showing a similar O-GlcNAcylated proteins pattern to that observed in OA cartilage. We also observed an increased expression of the three OGT isoforms, nucleo-cytoplasmic (ncOGT), mitochondrial (mOGT) and small (sOGT) in HOC as early as 3 h, although the peak of induction was at 6 h. Three isoforms were increased in a similar grade, without prevalence of sOGT as found in OA cartilage. IL-1␣ also increased the presence of both OGA isoforms, long (OGA-L) and short (OGA-S) after 3 and 6 h of stimuli. A clear prevalence of long isoform was found along IL-1␣ stimulation whereas OGA-S had the highest prevalence in OA cartilage. Conclusion: To our knowledge, this is first evidence of procatabolic cytokines involvement in O-GlcNAc modifications. IL-1␣ can induce accumulation of O-GlcNAc proteins in human OA chondrocytes. Although both regulatory enzymes of this modification, OGT and OGA, are also increased with IL-1␣ stimulation, characteristic distribution of the different isoforms in OA cartilage was not reached. IL-1␣ is in part responsible of a dysregulation in the O-GlcNAc proteins modification during OA although other processes must be involved.

epigenome-wide DNA methylation changes in osteoarthritic compared to healthy cartilage from the same joints. Methods: Twelve femoral heads were obtained at the time of hip arthroplasty for primary OA. Articular cartilage tissue was dissected from grossly affected and grossly normal areas, flash frozen in liquid nitrogen, and DNA was extracted. Following sodium bisulfite-treatment, DNA methylation was quantified at ⬎485,000 CpG sites across the genome using Illumina HumanMethylation450 arrays. CpG sites with an absolute methylation difference between OA and normal cartilage (Db) ⱖ 15%, and P ⬍0.01 after correction for multiple testing using a false discovery rate (FDR) of 5%, were considered statistically significant and used for subsequent analysis. Results: We found 442 differentially methylated CpG sites in OA compared to normal cartilage from the same joints: 260 hypo- and 182 hypermethylated. Overrepresented gene sets included “Connective tissue disorder” (n⫽53, p⫽4.73E-6 to 7.36E-3), “Developmental disorder” (n⫽44, p⫽4.73E-6 to 7.36E-3), and “Skeletal & muscular disorder” (n⫽59, p⫽4.73E-6 to 7.36E-3). Particularly interesting methylation changes in OA include hypermethylation of COL11A2, which functions to maintain spacing and diameter of type II collagen and is mutated in patients with Stickler syndrome and OSMED. Additional DNA methylation changes detected include hypermethylation of COL6A2, hypomethylation of the fibrillar collagen gene COL1A1, and hypermethylation of COL18A1. Hypermethylation was also noted in multiple CpG sites within the WNT pathway co-receptor LRP5, which is associated with OA in mice and is a genetic susceptibility gene for OA in humans. Hypomethylation was found at multiple CpG sites in the transcription factor RUNX1. Upstream regulator analysis identified significant association of TGF␤1 with 44 differentially methylated genes. Finally, canonical pathway analysis identified enrichment of several pathways, most prominently the ERK signaling pathway among differentially methylated genes (p⫽1.51E-4). Conclusion: We detected significant methylation changes in multiple collagen genes in OA. Our data suggest an epigenetic basis for defective collagen production in OA. Furthermore, we found evidence for epigenetic dysregulation of WNT signaling, and enrichment of genes associated with TGF␤- and ERK-pathway signaling, both of which are enticing targets for OA therapeutics. This work reinforces a role for genetic/epigenetic interaction in the pathogenesis of OA.

Sunday, October 27

60 O-Linked N-Acetylglucosamine Modified Proteome Of Human Osteoarthritic Cartilage: Biological Significance. Jessica Andre´s-Bergo´s1, Marıá Luisa Hernaéz2, Miguel Otero1, Ane Larranaga-Vera3, Sandra Pe´rez-Baos3, Mary B. Goldring1, Gabriel Herrero-Beaumont3 and Raquel Largo4. 1Tissue Engineering Repair and Regeneration Program, Hospital for Special Surgery and Weill Cornell Medical College, New York, NY, 2Proteomics Unit, Universidad Complutense de Madrid-Parque Cientı´fico de Madrid, ProteoRed ISCIII, Madrid, Spain, 3Joint and Bone Research Unit. IIS-Fundacioń Jimeńez Dıáz. UAM, Madrid, Spain, 4IIS-Fundacioń Jimeńez Dıáz, Madrid, Spain. Background/Purpose: O-linked N-acetylglucosamine (O-GlcNAc) posttranslational modifications have been implicated in the control of different signaling cascades and in the development and progression of degenerative and age-related diseases. We previously showed that accumulation of O-GlcNAc-modified proteins alters chondrocyte gene expression, leading to increased expression of chondrogenic markers in vitro and increased chondrogenic differentiation in vivo (Andre´s-Bergo´s J, JBC 2012). We also showed that the total amount of O-GlcNAcylated proteins was significantly increased in human osteoarthritic (OA) cartilage compared to that of healthy cartilage (Tardıó L, Arthritis Rheum, submitted). Here, we aimed to better profile and identify the O-GlcNAc-modified proteins in human OA cartilage. Methods: Human OA cartilage was isolated from patients undergoing total knee replacement surgery (n⫽6), while healthy cartilage was obtained from age- and gender-matched donors (n⫽6). Wheat germ agglutinin (WGA) chromatography was used to selectively pull-down O-GlcNAc-modified proteins. To this end, equal amounts of proteins isolated from OA and healthy cartilage were loaded on WGA-affinity columns. The eluted fractions containing O-GlcNAc-modified proteins were processed by nano-LC-M/MS, using a 5500QTRAP mass spectrometer (AB Sciex), and subsequently analyzed using GeneCodis software, which grouped proteins based on predefined biological processes. Results: Healthy and OA human cartilage shared a number of O-GlcNAc-modified proteins; however, our WGA-based analysis identified subsets of proteins O-GlcNAc-modified differentially in health and disease. Around a 26% of healthy cartilage specific proteins were involved in cytoskeleton organization, and more than 16% play an important role in cell adhesion. None of these biological processes were found in the analysis of OA differential proteins suggesting that these protein modifications may have an important role in preserving cell structural integrity in a healthy state. Interestingly, O-GlcNAc modifications in proteins involved in growth plate cartilage development, cartilage condensation, or inhibition of angiogenesis were absent in OA samples and prominent in healthy cartilage. More importantly, O-GlcNAc modifications of proteins implicated in extracellular matrix degradation, cell growth, or complement activation were higher in OA cartilage than in non-OA/healthy cartilage, clearly separating healthy and damaged tissue based upon biological processes and subsets of O-GlcNAcmodified proteins. Conclusion: We report here for the first time subsets of differentially O-GlcNAc-modified proteins in healthy and OA human articular cartilage. Biological processes involving O-GlcNAc proteins are different in OA cartilage in comparison with the healthy one. Our results further highlight the contribution of this means of protein control to the development and progression of OA and, more importantly, may contribute to the identification of potential biomarkers and/or therapeutic targets in OA.

Disclosure: M. A. Jeffries, None; A. H. Sawalha, None; J. A. James, Pfizer Inc, 2, GSK, 5.

62 Tenogenic Effect Of Homeobox Mohawk gene for Bone Marrow Mesenchymal Stem Cells. Koji Otabe, Hiroyuki Nakahara, Akihiko Hasegawa, Martin K. Lotz and Hiroshi Asahara. The Scripps Research Institute, La Jolla, CA. Background/Purpose: Mohawk homeobox (MKX) has been demonstrated as a tendon and ligament specific transcriptional factor. MKX knock-out mice showed hypoplastic tendons throughout the body, due to reduced type I collagenproduction in tendon cells. Our group also demonstrated expression of MKX decreased in mature human knee ligament tissues, during joint aging and further decreased in osteoarthritis (OA). The aims of this study were to characterize factors that may promote tenogenic differentiation of bone marrow derived mesenchymal stem cells (BMMSCs) and to investigate the role of MKX in ligament/tenogenic differentiation of BMMSCs. Methods: Human BMMSCs (n⫽9 from 3 donors) were purchased from Lonza, and treated with putative tenogenic factors (50ng/ml BMP12, 5ng/ml TGFbeta or 50ng/ml PDGF). Tendon/ligament related gene expression was analyzed by qRT-PCR at 1–5 days after treatment. BMMSCs were infected with adenoviruses expressing MKX or Scleraxis (Scx), also a tendon specific transcription factor and tendon related gene expression was analyzed levels at days 1, 2 and 7. Results: Basal MKX and SCX expression levels were low in BMMSCs. BMP12 increased expression of MKX and SCX genes in BMMSCs at day 5 (p⬍0.01, n⫽6). COL1a1 levels were higher in BMP12 and TGFbeta treated groups (p⬍0.01, n⫽6). Decorin (DCN) and tenascin B (TNXB), important extracellular matrix (ECM) components of ACL, were also higher in BMP12 treated group (p⬍0.05, n⫽6). The expression of COL1a1 and TNXB were increased by overexpression of MKX in BMMSCs (p⬍0.05, n⫽9). Tenomodulin (TNMD), a marker of late stage tenogenic differentiation, was expressed at higher levels in Ad-Mkx and Ad-Scx groups than in Ad-Mock group (p⬍0.05, n⫽9).

Disclosure: J. Andre´s-Bergo´s, None; M. L. Hernaéz, None; M. Otero, None; A. Larranaga-Vera, None; S. Pe´rez-Baos, None; M. B. Goldring, None; G. HerreroBeaumont, None; R. Largo, None.

61 Epigenome-Wide DNA Methylation Study Reveals Hypermethylated Collagen Genes and Suggests a Role for TGF␤ in Osteoarthritis. Matlock A. Jeffries1, Amr H. Sawalha2 and Judith A. James3. 1University of Oklahoma Health Sciences Center, Oklahoma City, OK, 2University of Michigan, Ann Arbor, MI, 3Oklahoma Medical Research Foundation, Oklahoma City, OK. Background/Purpose: Osteoarthritis (OA) is the leading cause of chronic disability in the U.S., affecting 40% of individuals over the age of 70 and costing $128 billion annually in the US alone. Late-stage OA chondrocytes exhibit a host of gene transcription changes leading to upregulation of enzymes that contribute to cartilage breakdown. Herein, we characterize

S24

Background/Purpose: Osteoarthritis (OA) is the most common joint disease. Among the earliest lesions during OA development is disruption of the superficial zone (SZ) of articular cartilage. The SZ of articular cartilage contains adult mesenchymal progenitor cells and is unique in extracellular matrix composition. Recently, we found that the EFEMP1 gene encoding fibulin-3 is specifically expressed in SZ. However, the expression pattern of fibulin-3 in articular cartilage and its role is unknown. The objectives of this study were to examine fibulin-3 expression patterns in joint aging and OA and to investigate the role of fibulin-3 in OA pathogenesis. Methods: Immunohistochemical analysis was performed on human and mouse knee cartilage to determine changes in fibulin-3 expression during joint aging and in OA. To examine the role of fibulin-3 in cartilage homeostasis, experimental OA was induced in wild type and fibulin-3⫺/⫺mice. The mice were euthanized 10 weeks after the knee surgery for histological analysis. To address whether fibulin-3 is involved in regulating chondrocyte differentiation, human articular chondrocytes were transfected with fibulin-3-specific siRNA. Cells were harvested at 48 hours and 72 hours for quantitative PCR and Western blot analyses. To further address the role of fibulin-3 during chondrogenesis, bone marrow mesenchymal stem cells (MSC) were transduced with lentivirus (LV) encoding EFEMP1 or control LV expressing LacZ, then MSC pellets were prepared and analyzed for chondrogenesis. Results: Fibulin-3 was specifically expressed in the SZ of normal cartilage in human and mouse knee joints. Fibulin-3 expression was intracellular and in the extracellular matrix (ECM) and declined with aging. Both aging-related OA and experimental OA were significantly more severe in fibulin-3⫺/⫺mice compared with wild type mice. Fibulin-3 expression was high in MSC and decreased during chondrogenesis. Suppression of fibulin-3 by siRNA in MSC significantly increased SOX9, collagen II and aggrecan in articular chondrocytes, while the overexpression of fibulin-3 inhibited chondrogenesis in MSC. Conclusion: We found that fibulin-3 is specifically expressed in the SZ of articular cartilage and its expression is reduced in aging and OA. Fibulin-3 regulates survival and differentiation of adult progenitor cells and its agingrelated decline is an early event in OA pathogenesis. Preventing or restoring aging-associated loss of fibulin-3 in SZ chondrocytes has potential to delay or prevent onset of OA.

Disclosure: K. Otabe, None; H. Nakahara, None; A. Hasegawa, None; M. K. Lotz, Tanabe Research Laboratories, 4, Cargill, 2; H. Asahara, None.

63 Investigations On The Role Of Delta/Notch Like EGF-Related Receptor In The Pathogenesis Of Osteoarthritis. Lucija Berninger1, Anna Balkenhol1, Clemens Baier2, Stefan Rehart3, Markus Rickert4, Ulf Mu¨ller-Ladner1, Elena Neumann1 and Matthias Geyer1. 1Justus-Liebig-University of Gie␤en, Kerckhoff-Klinik, Bad Nauheim, Germany, 2University of Regensburg, Bad Abbach, Germany, 3Markus-Hospital, Frankfurt, Germany, 4University Hospital Giessen and Marburg, Giessen, Germany. Background/Purpose: Delta/Notch like EGF-related receptor (DNER) is a single-pass transmembrane protein with characteristic EGF-like repeats in the extracellular domain, similar to those of the Notch receptor and its ligand Delta. In our previous study, an overexpression of DNER mRNA was observed in lesional areas of human osteoarthritis (OA) articular cartilage compared to unaffected zones of the same tissue sample. Based on these findings and on the fact that Notch signaling pathway plays a major role in cell fate regulation and differentiation of human articular chondrocytes, we analyzed the role of DNER in the pathogenesis of OA. Methods: Human articular cartilage was obtained from knee joint explants of OA patients undergoing total joint replacement surgery. The expression of DNER in human cartilage was analyzed by immunohistochemistry. Isolated primary chondrocytes were nucleofected with siRNA targeting human DNER or with a plasmid for DNER overexpression (pDNER), respectively. Knockdown and overexpression were confirmed by immunocytochemical staining for DNER in the nucleofected chondrocytes at different time points of cultivation. RNA was isolated and the effects of the artificially modified DNER expression on specific cartilage metabolic parameters (collagen type 2 and 1, Sox9, MMP9, ADAMTS4 and 5) analyzed using quantitative real-time PCR. Results: Histologically, DNER was significantly increased in the affected cartilage of all patients (n⫽3) on the protein level compared to the intact zones of the same tissue. An overexpression of DNER in primary human OA chondrocytes peaked 24 h after nucleofection (increase: 50000 fold). 72 h after nucleofection, a down-regulation of collagen type 2 (⫺7,89 fold) and an up-regulation of collagen type 1 (2,07 fold) as well as Sox 9 (7.39 fold) mRNA was measured.The overexpression of DNER was detectable at different time points on the protein level. In contrast, following knockdown of DNER, its expression was significantly decreased (pⱕ0.001) but there was no significant change in the expression profile of collagen type 2 (0.94 fold; pⱕ0.621), collagen type 1 (0.74 fold; pⱕ0.097), Sox9 (0.83 fold), MMP9 (0.97 fold), ADAMTS4 (0.64 fold) and 5 (1.2 fold). The confirmation of the DNER knockdown on the protein level at different time points identified that DNER is still to be constitutively found in the membrane. Conclusion: Since DNER overexpression led to a down-regulation of collagen type 2 in human articular chondrocytes whilst the expression of collagen type 1 and Sox9 were up-regulated, our preliminary results could lead to the assumption that DNER might contribute to the dedifferentiation of human OA chondrocytes within a “regenerative” attempt in order to deorganize and re-assemble adapted cartilage extracellular matrix to reconstitute physiological homeostasis and the ability to re-sustain to physical load during the course of human OA.

Disclosure: A. Hasegawa, None; T. Yonezawa, None; N. Taniguchi, None; K. Otabe, None; Y. Akasaki, None; T. Matsukawa, None; M. Saito, None; M. Neo, None; L. Y. Marmorstein, None; M. K. Lotz, None.

65 The Frequency Of Bone Marrow Oedema Adjacent To The Cruciate Ligament Peri-Entheseal Vascular Channels In Inflammatory and Degenerative Arthritis. Daniel Binks1, Melissa Matzelle2, Diane Bergin3, Richard J. Hodgson1, Ai Lyn Tan1, Ellen M. Gravallese4, Dennis McGonagle5 and Aleksandra Radjenovic1. 1University of Leeds, Leeds, United Kingdom, 2University of Massachusetts Medical School, Worcester, MA, 3Galway University Hospitals, Galway, Ireland, 4UMass Memorial Medical Center, Worcester, MA, 5Leeds Institute of Rheumatic and Musculoskeletal Medicine, University of Leeds and NIHR Leeds Musculoskeletal Biomedical Research Unit, Leeds Teaching Hospitals NHS Trust, Leeds, United Kingdom. Background/Purpose: We noted in mouse models of inflammatory arthritis (IA) that an early point of entry of inflammation into the marrow space occurred at sites where penetrating vessels entered the bone. The role of penetrating vessels in erosion formation in man has not been explored in IA, nor has the role of vascular channels in bone marrow oedema (BME) in osteoarthritis (OA) been explored. The purpose of this work was to investigate the frequency of BME in inflammatory and degenerative arthropathies in close proximity to the peri-entheseal ACL and PCL vascular channels Methods: Normal microanatomy was defined in 21 cadaveric knees using 3T MRI and histology. MRI of 89 patients from the Osteoarthritis Initiative study and 27 patients with inflammatory arthritis were evaluated for BME at the same locations. An animal model of inflammatory arthritis was evaluated to ascertain whether the putative peri-entheseal vascular regions influenced the propensity of osteitis and bone erosion Results: Data from the animal model of IA showed inflammation entering the marrow space along the adventitia of blood vessels that penetrate the cortical bone in close proximity to the cruciate ligament insertions. On 3T MRI, a vascular channel adjacent to the ACL tibial insertion was observed in 64% of cadaveric specimens examined. Similarly, a vascular channel adjacent to the PCL was seen in 71% of cases. BME was observed in the regions corresponding to the location of the vascular channel in 51% of knees for both

Disclosure: L. Berninger, None; A. Balkenhol, None; C. Baier, None; S. Rehart, None; M. Rickert, None; U. Mu¨ller-Ladner, None; E. Neumann, None; M. Geyer, None.

64 Fibulin-3 In Joint Aging and Osteoarthritis Pathogenesis. Akihiko Hasegawa1, Tomo Yonezawa2, Noboru Taniguchi2, Koji Otabe2, Yukio Akasaki2, Tetsuya Matsukawa2, Masahiko Saito3, Masashi Neo4, Lihua Y. Marmorstein5 and Martin K. Lotz2. 1Osaka Medical College, Takatsuki, Osaka, Japan, 2The Scripps Research Institute, La Jolla, CA, 3Toho University Sakura Medical Center, Sakura, China, 4Osaka Medical College, Osaka, Japan, 5University of Arizona, Tucson, AZ.

S25

Sunday, October 27

Conclusion: This is the first report to compare several potential tenogenic factors in BMMSCs and to investigate the MKX expression and function in human BMMSCs. The present findings demonstrate that BMP12 most effectively enhanced tenogenesis related genes including Mkx, and overexpression of MKX was associated with increased tendon ECM production. Thus, MKX is represents a key factor for tenogenic differentiation of BMMSCs.

Sunday, October 27

the anterior and posterior channel. Histological evaluation of 10 cadaveric specimens confirmed the location of the vascular channels along with the presence of associated subclinical microdamage including subchondral bone damage (80% of cases) and micro-cyst formation (50%). Evaluation of patient MRIs showed the prevalence of oedematous features in the same topographic locations in patients with early OA (41% ACL, 59% PCL) and IA (44% ACL, 33% PCL), (Figure 1).

COL1A2, COL2A1, COL10A1, SOX5, SOX9, TGFBR2, CTSK, GREM1, IGFBP4, WISP1. We also demonstrated that FSTL1-deficient MSC displayed defects in TGFb-induced differentiation into chondrocytes, including decreased production of ECM proteoglycans and reduced expression of type II collagen. TGFb-induced phosphorylation of SMAD3, p38 MAPK, and AKT were significantly decreased in FSTL1 KO cells. Conclusion: These results demonstrate that FSTL1 is a crucial component of the regulatory mechanism controlling chondrocyte proliferation, differentiation, and expression of ECM molecules. Our findings may lead to the development of novel strategies for cartilage repair and provide new disease modifying treatments for OA. Disclosure: Y. Chaly, None; S. Smith, None; D. Bushnell, None; B. Campfield, None; H. Blair, None; R. Hirsch, University of Pittsburgh, 9.

Figure 1. Sagittal fluid-sensitive MRI showing examples of A, tibial vascular channel (arrow) posterior to the anterior cruciate ligament insertion (arrowheads) in a patient with inflammatory arthritis and B, bone marrow lesions (arrows) seen in the region anterior to the PCL insertion (arrowheads) in a patient with osteoarthritis. Bone marrow lesions/oedema were seen adjacent to both cruciate ligament insertions in both inflammatory and degenerative types of arthritis.

67 Heparin-Binding Epidermal Growth Factor-Like Growth Factor (HBEGF) As a Potential Mediator In Osteoarthritis. Richard F. Loeser1 and David A. Long2. 1Wake Forest School of Medicine, Winston-Salem, NC, 2 Wake Forest University, Winston-Salem, NC. Background/Purpose: We recently reported changes in gene expression in knee joint tissue from 12 week-old and 12 month-old mice after OA was induced by surgical destabilization of the medial meniscus (DMM). Heparin-binding EGF-like growth factor (HB-EGF) was found by gene array to be significantly up-regulated at 8 weeks after surgery in both age groups. HB-EGF is a known ligand for the EGF receptor (EGFR) which signals through MAP kinases and the small GTPase Rac. EGFR signaling has been implicated in OA pathogenesis but a potential role for HB-EGF has not been examined. The purpose of this study was to determine if HB-EGF was present in human OA cartilage and to test the ability of HB-EGF to stimulate production of chondrocyte MMP-13 which can contribute to matrix destruction in OA. Methods: Real-time PCR was used to confirm the HB-EGF gene array results using RNA isolated from knee joint tissues (n⫽9 mice per surgical group) from 12 week-old and 12 month-old male C57/BL6 mice that had undergone DMM or sham control surgery. Immunohistochemistry for HB-EGF was performed using knee joint tissue collected from normal tissue donors or age-matched patients who had undergone knee replacement for OA (n⫽4 each). In vitro stimulation studies used cultured human articular chondrocytes isolated from normal donor tissue (n’3 independent donors for each experiment). Fibronectin fragments (FN-f) were used as a catabolic stimulus relevant to OA. Effects of recombinant HB-EGF, ⫾ FN-F, on MMP-13 production were measured by immunoblotting and real time (RT)-PCR and on Rac activity using an activity assay. Results: HB-EGF expression was about two-thirds lower (p⬍0.05) in the sham control joints of the older mice compared to young controls but increased by ⬎6-fold (p⬍0.001) in the older DMM joints vs 1.4-fold in the young DMM when compared to the age-matched sham. Immunohistochemical staining revealed increased HB-EGF in human OA cartilage relative to age-matched normal where little to no staining was observed. Treatment of cultured human chondrocytes with FN-f increased HB-EGF expression which peaked at almost 60-fold (by RT-PCR) at 3 hours. Overnight FN-f stimulation increased HB-EGF secretion into conditioned media. We found that HB-EGF at 10ng/ml stimulated chondrocyte Rac activity and phosphorylation of the EGFR, ERK and p38 but not JNK. High dose (100ng/ml) but not low dose (10ng/ml) HB-EGF stimulated MMP-13 production in conditioned media, detected by immunoblotting. However, low dose HB-EGF enhanced the MMP-13 production induced by FN-F by 1.4-fold over FN-f alone and this was completely inhibited by blocking Rac1 with 100␮M NSC23766 or EGFR with 250 nM AG1478. Conclusion: HB-EGF was detected in human OA cartilage but not normal cartilage. FN-fs, known to be present in OA cartilage and to stimulate catabolic pathways, were found to stimulate HB-EGF production by normal chondrocytes. Given the finding that HB-EGF signals through the chondrocyte EGFR and promotes MAP kinase and MMP-13 production and that it is present in human OA cartilage, further evaluation of its role in OA is warranted.

Conclusion: Our findings show that the vascular channels adjacent to the anterior and posterior cruciate ligament entheses are common locations for erosion formation and damage in both inflammatory and degenerative arthritis. Furthermore, we have found a significant clustering of subclinical microdamage in these regions in normal cadaveric samples. Therefore, we conclude that peri-entheseal vascular channels are likely to present a common pathogenesis focus for both OA and IA. Disclosure: D. Binks, None; M. Matzelle, None; D. Bergin, None; R. J. Hodgson, None; A. L. Tan, None; E. M. Gravallese, None; D. McGonagle, None; A. Radjenovic, None.

66 Follistatin-Like Protein 1 Is An Important Regulator Of Chondrogenesis. Yury Chaly1, Sonja Smith1, Daniel Bushnell2, Brian Campfield2, Harry Blair3 and Raphael Hirsch1. 1University of Iowa Carver College of Medicine, Iowa City, IA, 2Children’s Hospital of Pittsburgh of UPMC, Pittsburgh, PA, 3University of Pittsburgh School of Medicine, Pittsburgh, PA. Background/Purpose: Chondrocytes play a pivotal role in osteoarthritis (OA) because they are the only cells in the articular cartilage that are responsible for maintenance of the extracellular matrix (ECM). Articular cartilage intrinsic regenerative capacity is strongly downregulated in humans with age. The use of mesenchymal stromal cells (MSC) is an attractive approach for cartilage repair because MSC are capable of proliferating and differentiating into chondrocytes in vitro. However, without detailed knowledge of how chondrogenic pathways are induced it has not been possible to produce true hyaline cartilage in vitro or in vivo, or to maintain cartilage growth. The present study was undertaken to determine whether follistatin-like protein 1 (FSTL1) regulates chondrogenic differentiation of MSC. We hypothesized that FSTL1 may enhance cell proliferation and anabolic activity in articular chondrocytes and, therefore, play an important role in maintaining cartilage homeostasis. Methods: To study the role of FSTL1 in chondrogenesis, we made use of FSTL1 knockout (KO) mice generated in our laboratory. MSC were obtained from skulls of E18.5 embryos. Proliferative capacity of MSC was analyzed by flow cytometry using CFSE labeling. Differentiation of MSC into chondrocytes was carried out in a pellet culture system. Gene expression microarray analysis was performed on MSC isolated from FSTL1 KO and wild-type embryos. For signaling study, protein extracts from TGFb-stimulated MSC were analyzed for phospho (p)-SMAD3, p-p38 MAPK, p-AKT by Western blotting. Results: Homozygous FSTL1 KO mice showed extensive skeletal defects, supporting a role for FSTL1 in chondrogenesis. FSTL1 KO embryos had decreased cellularity in the embryonic vertebral cartilage and FSTL1 KO MSC had reduced proliferative capacity. Microarray analysis of gene expression in FSTL1 KO MSC revealed dysregulation of multiple genes known to be involved in chondrogenesis, including

Disclosure: R. F. Loeser, None; D. A. Long, None.

S26

Autophagy Activation Protects From Mitochondrial Dysfunction In Human Chondrocytes. Beatriz Carame´s1, Paloma Lo´pez de Figueroa1, Martin Lotz2 and Francisco J. Blanco3. 1Osteoarticular and Aging Research Lab, INIBIC–Complejo Hospitalario Universitario A Corunã, SPAIN, A Corunã, Spain, 2The Scripps Research Institute, La Jolla, USA, La Jolla, CA, 3 Osteoarticular and Aging Res. Lab. CIBER-BBN. INIBIC- University of A Corunã, A Corunã, Spain. Background/Purpose: A common feature of aging-related diseases, such as osteoarthritis (OA), is the progressive accumulation of damaged macromolecules leading to cell dysfunction and death. Autophagy, a key pathway for cellular homeostasis by removing such damaged molecules and organelles, including mitochondria, has a protective and survival-promoting function. Recent studies indicated that autophagy decreases with aging and OA contributing to the accumulation of damaged macromolecules. In addition, there is increasing evidence that mitochondrial dysfunction plays a critical role in accelerating the aging process and several lines of evidence demonstrated mitochondrial dysfunction in OA cartilage. The objective of this study is to determine whether activation of autophagy protects from mitochondrial dysfunction in human chondrocytes. Methods: Human chondrocytes were treated with Rotenone (10 ␮g/ml), Antimycin (40 ␮g/ml) and Oligomycin (10 ␮g/ml), a mitochondrial respiratory chain (MRC) inhibitors of complex I, III and IV, respectively. Mitochondrial function and cell death were evaluated by Flow Cytometry (FC) and Fluorescence Microscopy (FM). Autophagy activation was analyzed by determination of LC3, a main marker of autophagy activation by Immunofluorescence (IF). To investigate whether autophagy protects from mitochondrial dysfunction, autophagy was induced by mTOR inhibition, using mTORC1 selective inhibitor Rapamycin (Rapa, 10 ␮M) and mTORC1 and mTORC2 inhibitor Torin 1 (50 nM). The effects on autophagy, mitochondrial function and chondrocyte viability were analyzed by IF, FC and FM. Results: Mitochondrial dysfunction was induced by 6 h treatment with MRC inhibitors, which significantly decreased mitochondrial membrane potential (⌬⌿m) (ROT: 26.17 ⫾ 5.9; AA: 18.21 ⫾ 3.28; Oligo: 41.74 ⫾ 7.59, expressed as % vs control; *p ⬍ 0.01). These results are consistent with increased ROS production (26.8 %, 44.6 % and 25.7 % for ROT, AA and Oligo, respectively; *p ⬍ 0.001) and cell death by apoptosis at 12 h (Control: 13.56 ⫾ 1.83; ROT: 33.66 ⫾ 5.55; AA: 29.05 ⫾ 4.262; *p⬍0.05). Autophagy activity determined by LC3 expression significantly reduced in response to MRC inhibitors at 12 h. To evaluate whether autophagy regulates mitochondrial dysfunction, chondrocytes were pretreated with Rapa and Torin 1 for 4 h and then treated with the MRC inhibitors for 12 h. The results show an increase in LC3 expression compared to MRC inhibitors alone. Furthermore, autophagy inducers Rapa and Torin1 increased ⌬⌿m (Rapa: 125.8 ⫾ 20.74; Rapa⫹ROT: 44.67 ⫾ 10.37; Rapa⫹AA: 30.71 ⫾ 5.949; Rapa ⫾ Oligo: 108.5 ⫾ 55.03 and Torin 1: 90.34 ⫾ 9.17; Torin 1⫹ROT: 35.24 ⫾ 2.7; Torin 1⫹AA: 41.98 ⫾ 6.49; Torin 1⫹Oligo: 98.19 ⫾ 16.81; *p⬍0.05), decreased ROS production (*p⬍0.05) and reduced cell death, suggesting a protective effect of autophagy activation on pharmacologically induced mitochondrial dysfunction. Conclusion: These data identify autophagy activation as a protective mechanism from mitochondrial dysfunction in human chondrocytes. Pharmacological interventions that enhance autophagy may have chondroprotective activity in articular cartilage with defects on mitochondrial function.

Disclosure: M. D. Mayan, None; P. Carpintero-Fernańdez, None; R. Gago-Fuentes, None; P. Fernańdez-Puente, None; P. Filgueira-Fernandez, None; N. Goyanes, None; V. Valiunas, None; P. Brink, None; G. Goldberg, None; F. J. Blanco, None.

70 Chondrocyte Hypertrophy, Measured By The Secretion Of Collagen Type X, Is Associated With Cartilage Degradation and Systemic Inflammation In Osteoarthritis. Yi He1, Natasja Stæhr Gudman2, Niels Ulrik Hansen1, Jianxia Wang3, Di Su3, Qinlong Zheng3, Ole Simonsen4, Kristian Kjaer Petersen5, Moustapha Kassem6, Morten Asser Karsdal2 and Anne C. Bay-Jensen2. 1Nordic Bioscience, Herlev, Denmark, 2Nordic Bioscience, Biomarkers and Research, Herlev, Denmark, 3Nordic Bioscience China, Beijing, China, 4Frederikshavn Hospital, Frederikshavn, Denmark, 5Center for Sensory-Motor Interaction, Aalborg, Denmark, 6Odense University Hospital and University of Southern, Odense, Denmark.

Disclosure: B. Carame´s, None; P. Lo´pez de Figueroa, None; M. Lotz, None; F. J. Blanco, None.

69 Interconnected Cellular Projections and Gap Junctions Mediate Metabolic Coupling Between Chondrocytes Located In Different Layers Of The Tissue: Cell-To-Cell Communication In Normal and Osteoarthritic Cartilage. Maria Dolores Mayan1, Paula Carpintero-Fernańdez2, Raquel GagoFuentes2, Patricia Fernańdez-Puente3, Purificacion Filgueira-Fernandez3, Noa Goyanes3, Virgin Valiunas4, Peter Brink4, Gary Goldberg5 and Francisco J. Blanco6. 1Correspondence to: [email protected] and [email protected], A Corunã, Spain, 2These authors contribute equally to this work, A Corunã, Spain, 3Osteoarticular and Aging Research Group. Rheumatology Division, Biomedical Research Center (INIBIC). Hospital Universitario A Corunã, Xubias de Arriba 84, 15006, A Corunã, Spain, 4Department of Physiology and Biophysics. State University of New York, Stony Brook, New York, SC, 5 Department of Molecular Biology. Medical Center Drive, University of Medicine and Dentistry of New Jersey, Stratford, NJ, 6Osteoarticular and Aging Res. Lab. CIBER-BBN. INIBIC- University of A Corunã, A Corunã, Spain.

Background/Purpose: Osteoarthritis (OA) is the most common degenerative joint disease, of which the pathogenesis is inadequately understood. The hypertrophy-like changes, such as expression of hypertrophy markers and matrix calcification have been observed in the initiation and progression of OA. The aim of this study was to investigate the relationships of chondrocyte hypertrophy, cartilage degradation and systemic inflammation by measuring 3 biomarkers in serum of 283 OA patients. Methods: i) A competitive ELISA, C-Col10, was developed as a marker of chondrocyte hypertrophy, through measurement of type X collagen (ColX). ii) C-Col10, C2M (matrix metalloproteinase-derived fragments of type II collagen) and hsCRP were quantified by ELISA in serum of the patients, stratified by Kellgren-Lawrence (KL) score 0–4. iii) Association between serum levels of the 3 biomarkers were analyzed (Pearson correla-

S27

Sunday, October 27

Background/Purpose: We have recently found that articular chondrocytes from human adults contain long cytoplasmic arms that cross the extracellular matrix and physically connect two chondrocytes located in distant lacuna. Besides, we have also demonstrated that primary chondrocytes from adults are able to communicate with each other via gap junctions (GJ) as evidenced by the intercellular transfer of tracer dyes. Methods: For immunohistochemistry (IHC) assays, in situ cartilage was frozen immediately using Tissue-Tek O.C.T. and isopentanol in liquid nitrogen. For Scanning Electron Microscope (SEM), samples were embedded in cacodylate buffer before dehydratation. Dual Voltage-clamp methods along with the study of glucose and oligonucleotides transference and in situ electroporation were used for the study of gap junctional communication (InSitu Porator™ Cell Projects Ltd). Transwell co-culture system and mass spectrometry were used for the identification and quantification of transjuctional amino acids. Primary chondrocytes were previously cultured in SILAC® Dulbecco’s Modiffied Eagle’s Medium containing L-lisina [13C6] and L-arginina [13C6, 15N4]. Prior to LC analysis amino acids were derivatized using the tEZ: faastTM kit. Detection of analytes was performed with a Thermo Scientific LTQ-Orbitrap. Results: IHC, electrophysiology experiments and “metabolic capture” assays revealed that GJs play a metabolic function by exchanging nutrients including glucose and essential amino acids between cells. Microinjection of the fluorescent glucose derivative 2-NBDG showed a progressive fluorescence intensity increase in the contacted recipient cells during 22 and 35 minutes after dye injection. In situ electroporation of donor cells confirmed that adult chondrocytes efficiently exchange glucose between adjacent cells (18–20 cells deep). When cells were pre-incubated for 1 hour with 250␮g/ml of the GJ inhibitor GAP27, the transference of glucose was severely reduced (1–2 cells deep). For Transwell co-culture system, donors previously cultured in “heavy medium” were allowed to settle for 4 hours, time enough to form cell projections on the membrane right side up containing receivers. Mass spectrometry analysis revealed the transference of 3,5 pmol/ml of [13C6]-Llysina and 3 pmol/ml of [13C6, 15N4]-L-arginina. The results were confirmed using in situ electroporation and fluorescent labelled amino acids. IHC experiments in cartilage from patients with osteoarthritis (OA) revealed significantly elevated levels of the GJ protein Cx43 in the superficial zone and down through the next 1000 ␮m of tissue. Besides OA condrocytes contain longer cytoplasmic projections. Conclusion: The results here presented demonstrated that chondrocytes from human adult cartilage are physically connected to each other forming a three-dimensional cellular network, and are able to transfer nutrients such as glucose and essential amino acids through GJ channels. The levels of Cx43 protein and the length of cells projections suggest alterations in the normal homeostasis control and cell-cell communication between chondrocytes in cartilage of patients with OA.

68

Sunday, October 27

tions were done on log transformed data). The data is shown as mean [95%-CI]. iv) Full-depth cartilage biopsies from OA patients with different disease stages were immunostained with the C-Col10 and C2M antibodies. Results: The C-Col10 assay was specific for the C-terminal of type X collagen. The measurement range of the C-Col10 assay was 22–500pg/ml, with intra- and inter-assay CVs of 4.2% and 13%. There was a trend towards increasing C-Col10 levels with increasing KL score: KL0 52[24–80] pg/ml, (n⫽10); KL1 67 [56–78] pg/ml, (n⫽59); KL2 87[74–99] pg/ml, (n⫽144); KL3 80pg/ml [60–101], (n⫽36), and KL87 [47–127] pg/ml, (n⫽22). There was significant correlation of levels between C-Col10 and hsCRP (r⫽0.23, P⬍0.0001), and C2M (r⫽0.55, P ⬍0.0001). There was no correlation between C2M and hsCRP. Age and BMI adjustment didn’t change the significant correlations. OA patients with above normal hsCRP (⬎5) levels showed increased C-Col10, however this was independent from cartilage degradation. These data was supported by immunoloclization of C-Col10 and C2M in the OA cartilage biopsies, which showed distinct staining patterns: C-Col10 was observed in the deep zone around the pre-hypertrophic chondrocytes in mild OA, and around chondrocyte clusters in severe OA. Whereas C2M was consistently observed in all layers of the OA cartilage. Conclusion: Elevated C-Col10 levels were measured in OA patients and significantly higher in patients with above normal hsCRP levels, suggesting that inflammation is associated with chondrocyte hypertrophy. Correlation between C-Col10 and cartilage degradation indicated that chondrocyte hypertrophy may be involved in the cartilage degradation. All data show that chondrocyte hypertrophy is an essential step in the pathogenesis of OA and C-Col10 measurement can provide the critical information of OA disease status.

subchondral region suggests that this marker may be a non-invasive surrogate for active bone turnover in knee OA. Such markers may assist in the identification patients whom may benefit from strong anti-resorptives which inhibit bone and subchondral bone turnover. Disclosure: M. A. Karsdal, Nordic Biosciece, 1, Nordic Bioscience Diagnostic, 3; J. L. Huebner, None; V. B. Kraus, None; D. J. Leeming, Nordic Bioscience Diagnostic, 3; E. Coleman, None; G. E. McDaniel, None; K. M. Huffman, None; K. Henriksen, Nordic Bioscience Diagnostic, 3; A. C. Bay-jensen, None.

72 Autophagy Changes During Aging: A Study In GFP-LC3 Mice. Beatriz Carames1, William B. Kiosses2, Merissa Olmer3 and Martin Lotz4. 1Osteoarticular and Aging Research Lab, INIBIC–Complejo Hospitalario Universitario A Corunã, SPAIN, A Corunã, Spain, 2The Scripps Research Institute, La Jolla, CA, 3The Sripps Research Institute, La Jolla, CA, 4The Scripps Research Institute, La Jolla, USA, La Jolla, CA. Background/Purpose: Aging is a main risk factor for osteoarthritis (OA), the most prevalent musculoskeletal disorder. Aging-associated changes in autophagy, an essential cellular homeostasis mechanism, have recently been observed in several tissues, including articular cartilage. The objective of this study is to establish the patterns of autophagy in young and aging cartilage using GFP-LC3 mice. Methods: In GFP-LC3 transgenic mice, GFP-LC3 is ubiquitously expressed under the control of CAG promoter, and the accumulation of GFP punctate, which represents autophagosome, is observed in almost all tissues after 24–48 h fasting period. Homozygous GFP-LC3 mice (tg/tg) were employed to monitore autophagy in liver and joint tissues in response to aging (6 months old and 28 months old mice) by confocal microscopy analysis of vibratome sections. Two sets of control mice were included: Wild-type C57BL/6J mice were used to determine the background level of green fluorescence and young homozygous GFP-LC3 mice (6 months old) to provide a baseline for autophagy activity in liver and joints. Morphological changes in the articular cartilage were examined by histology using a semiquantitative scoring system. Furthermore, cellularity was determined by Hematoxylin-Eosine staining (H&E) and expression of the autophagy proteins Atg5 and LC3 was analyzed by immunohistochemistry (IHC) in 6, 18, 24 and 28 months old mice. Results: GFP-LC3 mice from 6 and 28 months old were employed to analyze the effect of aging on autophagy activation in liver and cartilage in the knee joints. In response to aging (28 months old mice), we observed a reduction in the GFP-LC3 signal in liver and cartilage. This reduction was statistically significant based on the total number of vesicles per cell (P ⬍ 0.01) and on the total area of vesicles per cell (um2) (P ⬍ 0.01) compared to the control group (6 months old mice). In addition, the histological evaluation showed a significant decrease in cartilage score (P ⬍0.01) in the old mice compared to the young control group. These results were accompanied with a reduction on cartilage cellularity and with a decrease in the expression of Atg5 and LC3 in an age-dependent manner, indicating a correlation between autophagy defects and aging-related OA in mice. Conclusion: These results support the hypothesis that autophagy is decreased with aging and that compromised autophagy represents a novel mechanism in the development of OA.

Disclosure: Y. He, Nordic Bioscience, 3; N. S. Gudman, Nordic Bioscience, 3; N. U. Hansen, Nordic Bioscience, 3; J. Wang, Nordic Bioscience China, 3; D. Su, Nordic Bioscience China, 3; Q. Zheng, Nordic Bioscience China, 3; O. Simonsen, None; K. K. Petersen, None; M. Kassem, None; M. A. Karsdal, Nordic Bioscience, 3; A. C. Bay-Jensen, Nordic Bioscnce, 3.

71 Subchondral Bone Turnover and Osteophyte Formation Are Key Aspects In The Progression Of Osteoarthritis and May Be Assessed and Predicted By a-CTX. Morten Asser Karsdal1, Janet L. Huebner2, Virginia Byers Kraus2, Diana J. Leeming1, Edward Coleman2, Gary E. McDaniel2, Kim M. Huffman2, Kim Henriksen1 and Anne C. Bay-jensen1. 1Nordic Bioscience, Biomarkers and Research, Herlev, Denmark, 2Duke University Medical Center, Durham, NC. Background/Purpose: Osteoarthritis (OA) is the most common form of arthritic disease. It is characterized by pathological changes in both bone and cartilage turnover as well as structure in which subchondral bone remodeling is speculated to be both an initiator and driver of OA. The aim of the current study was to evaluate a biomarker of high localized bone resorption, a-CTX, previously used for bone metastasis. OA patients were characterized by; radiographic knee OA, severity, progression and localized knee bone turnover as assessed by bone scintigraphy, a dynamic and sensitive indicator of symptomatic knee OA. Methods: A total of 149 participants (111 women, 38 men) were included who met ACR criteria for symptomatic OA and had the presence of Kellgren Lawrence (K/L) grade 1–4 radiographic OA in at least one knee. Late-phase bone scan images of both knees were obtained 2 hours after administration of 99m Tc-MDP. The intensity of uptake was scored semi-quantitatively and summed for each joint site. Radiographic knee OA severity, based on the summed scoring of osteophytes (OST) and joint space narrowing (JSN) was assessed using the standardized OARSI radiographic atlas. Progression status was determined by calculating the sum of the change score in either JSN or OST between baseline and the 3-year follow-up assessment. The concentrations of urinary a-CTX and CTX-II were determined by ELISA and normalized to creatinine concentration. Results: a-CTX was related to OST formation independent of the effects of age, gender, BMI, and HRT (p⫽0.009). a-CTX did not correlate with severity of knee OA based on OST and JSN, but did correlate with bone turnover assessed by intensity of 99mTc-MDP uptake in the medial knee compartment. Concentrations of urinary CTX-II were strongly associated with knee OA severity based on osteophyte, intensity of total knee 99mTcMDP uptake, and joint space narrowing. Conclusion: a-CTX was associated with subchondral bone turnover, and osteophyte formation, both central features of the pathogenesis of OA. CTX-II was correlated to JSN, and burden of disease as previously reported. The significant and strong association of a-CTX to bone turnover in the

Disclosure: B. Carames, None; W. B. Kiosses, None; M. Olmer, None; M. Lotz, None.

73 Fibrillin-1 Expression and Function Is Needed For Normal Joint Function and Mutations Leads To Osteoarthritis. Wasabha Ramanayake1, Helen Jones1, Isabel Orriss2, Tim Arnett2, Andrew Pitsillides3, Christopher P. Denton1, David J. Abraham1 and Blandine Poulet1. 1UCL Medical School, London, United Kingdom, 2University College London, London, United Kingdom, 3Royal Veterniary College, London, United Kingdom. Background/Purpose: Osteoarthritis (OA) is a common degenerative disease leading to pain and disability in ⬎6 million british people. The TGFb signalling pathway has been shown to play a major role in joint homeostasis and in OA development. Fibrillin-1 is an extracellular matrix protein that may play a structural role in the matrix of articular cartilage, but also regulates the bioavailability of TGFb to the cell. Thus Fibrillin-1 mutations, such as those seen in the Tight Skin mouse (or TSK), have been shown to increase TGFb signalling. The aim of this study is therefore to determine changes in

S28

ZCCHC11 gene indicating that ZCCHC11 is required for IL-6 mRNA stability. Conclusion: The data presented here for the first time show the expression of ZCCHC11 in human cartilage. Differential expression of ZCCHC11 in damaged and undamaged cartilage and its stimulation by IL-1b, a cytokine up-regulated during OA, point to a role of this enzyme in osteoarthritis. Furthermore, our gene silencing studies show the effect of ZCCHC11 on secretion of IL-6, an important cytokine involved in the pathogenesis of OA, via modifying its mRNA at the 3’ end. Taken together, the present study reveals a novel RNA modifying enzyme ZCCHC11 as an important player in OA pathophysiology. Disclosure: A. Haseeb, None; M. Shahidul Makki, None; A. Arida, None; T. M. Haqqi, None.

75 Decorin-Deficiency Alters Cartilage Stiffness and Attenuates The Development Of Osteoarthritis In Mice. Tobias Gronau1, Uwe Hansen1, Daniela Seidler1, Renato Iozzo2, Attila Aszodi3, Carina Prein4, Hauke ClausenSchaumann4, Karsten Kru¨ger5, Frank Mooren5, Jessica Bertrand1, Thomas Pap1, Peter Bruckner1 and Rita Dreier1. 1University Hospital Mu¨nster, Mu¨nster, Germany, 2Thomas Jefferson University, Philadelphia, PA, 3 Ludwig-Maximilians-University Munich, Munich, Germany, 4Munich University of Applied Sciences, Munich, Germany, 5Justus-Liebig University Giessen, Giessen, Germany.

Disclosure: W. Ramanayake, None; H. Jones, None; I. Orriss, None; T. Arnett, None; A. Pitsillides, None; C. P. Denton, None; D. J. Abraham, None; B. Poulet, None.

Background/Purpose: In articular (AC) and growth plate (GP) cartilage, the small leucine-rich proteoglycan decorin is mainly present within the interterritorial regions (ITs) of the extracellular matrix. Within the extracellular matrix it is associated with thick, well-banded collagen fibrils. Here we analyzed the impact of decorin-deficiency (Dcn⫺/⫺) on the biomechanical properties of cartilage and the development of osteoarthritis via forced exercise. Methods: Diverse cartilaginous tissues of Dcn⫺/⫺ mice and wild-type (WT) mice were examined at different developmental time points by histology/immunohistochemistry and by atomic force microscopy (AFM). The levels of active TGF-␤ were analyzed by ELISA. Expression of glycosaminoglycan (GAG) modifing enzymes was assesed via semiquantitave RT-PCR. Osteoarthritic changes (OA) were induced in 3 monthold Dcn⫺/⫺ and WT mice via forced running on a treadmill. The severity of the disease was evaluated via a modified Mankin score. Results: In P14.5 and adult Dcn⫺/⫺ mice, Alcian blue (pH 1.0) stained highly sulfated GAGs intensely throughout the entire GP and AC. In WT mice, staining was weaker and mainly restricted to the pericellular/territorial zones. Likewise, antibodies to stubs of chondrotin-4-sulfate (⌬C4S) chains intensely labeled all zones of GP and AC in Dcn⫺/⫺ but not in WT mice. Futhermore, mutant mice showed an increased expression of enzymes involved in GAG modification (PAPSS1, C4ST2). By contrast, the distribution and staining intensities of core proteins like aggrecan or biglycan were similar in both genotypes. AFM analysis of AC sections at 3 months of age revealed slightly altered fibril architecture and increased compressive stiffness of the ITs from Dcn⫺/⫺ mice. Those abnormalities were accompanied by enhanced levels of TGF-␤ in mutant mice. Both genotypes exhibited osteoarthritic changes in knee joint cartilage after six weeks of forced running. However, the changes were less severe in Dcn⫺/⫺ than in WT mice (Mankin score 4.5 versus 6.5). Conclusion: We propose the following mechanism attenuating OA in Dcn⫺/⫺ mice: Reportedly, decorin sequesters the cytokine TGF-␤ in the extracellular matrix. Higher levels of active TGF-␤ in cartilage of Dcn⫺/⫺ mice are likely to prevent inappropriate chondrocyte hypertrophy and, hence, induction of osteoarthritis. TGF-␤ is also known to affect sulfation of GAGs. Probably an increased sulfation leads to an augmented immobilization of water, increased osmotic swelling pressure and accordingly a stiffer cartilage matrix. Therefore, Dcn⫺/⫺ mice are less prone to develop osteoarthritis due to a combination of elevated active TGF-␤ levels and enhanced compressive stiffness of the articular cartilage.

74 Terminal Uridyltransferase Enzyme ZCCHC11 Regulates Interleukin-6 Expression In Primary Human Osteoarthritis Chondrocytes. Abdul Haseeb1, Mohammad Shahidul Makki1, Ahmad Arida2 and Tariq M. Haqqi2. 1 Northeast Ohio Medical University (NEOMED), Rootstown, OH, 2North East Ohio Medical University, Rootstown, OH. Background/Purpose: Non-template addition of nucleotides on the 3’ termini of mRNAs acts as a mechanism that controls mRNA stability. ZCCHC11 is a member of the non-canonical poly A polymerase (ncPAP) family. It has been implicated in post-transcriptional regulation of various genes, including some cytokines, by modification of their poly A tails or by modification of microRNAs that target these mRNAs. Role of ZCCHC11 in chondrocyte physiology and in the pathogenesis of osteoarthritis (OA) has not been studied. In the present study we determined the expression of ZCCHC11 in damaged and undamaged cartilage obtained from OA patients. We also studied the effect of IL-1b on the expression of ZCCHC11 in primary human chondrocytes. We further investigated the effect of gene silencing of ZCCHC11 on the secreted levels of several cytokines. Methods: Primary human chondrocytes were isolated from the deep zone of the cartilage obtained from OA patients who underwent total knee joint replacement surgery. Silencing of ZCCHC11 gene was carried out by using ON-TARGETplus siRNA pool (Thermo Fisher Scientific, Waltham, MA). Protein levels of 40 cytokines in culture supernatants after siRNA transfections were analyzed simultaneously using a glass-based human cytokine microarray (RayBio Human Cytokine Array). Poly A tail length analysis was performed by reverse transcription followed by PCR using an adaptor reverse primer and a forward primer specific for IL-6 mRNA. The PCR products were run on 2% agarose gel and visualized by EtBr staining. mRNA levels of ZCCHC11 and IL-6 genes were quantitated by using the TaqMan assays (Applied Biosystems, Carlsbad, CA). Data were derived using Origin 6.1 software and P⬍0.05 was considered significant Results: There was higher expression of ZCCHC11 mRNA in damaged cartilage as compared to unaffected smooth cartilage obtained from OA patients. mRNA expression of ZCCHC11 was significantly up-regulated by treatment of human primary chondrocytes with IL-1b. Cytokine protein array analysis revealed a significant decrease in secreted levels of IL-6 and a subset of cytokines upon siRNA mediated silencing of ZCCHC11 gene. Quantitative PCR analysis showed that mRNA expression of IL-6 gene was also suppressed in human chondrocytes with depleted ZCCHC11 mRNA. Poly A tail length analysis showed a significant shortening of IL-6 mRNA poly A tail upon silencing of

Disclosure: T. Gronau, None; U. Hansen, None; D. Seidler, None; R. Iozzo, None; A. Aszodi, None; C. Prein, None; H. Clausen-Schaumann, None; K. Kru¨ger, None; F. Mooren, None; J. Bertrand, None; T. Pap, None; P. Bruckner, None; R. Dreier, None.

S29

Sunday, October 27

fibrillin-1 expression during osteoarthritis in mice and to characterise the effect of Fibrillin-1 mutations in ageing-induced OA development in TSK mice. Methods: Immunohistochemistry for Fibrillin-1 expression was performed in Str/ort mice, a known model of spontaneous OA, and in control aged-matched non-OA CBA mice, in knee joints with different degrees of OA severity. The knees of TSK and littermate control male mice of 60–80wks of age were fixed and microCT scanned. Abnormal ectopic calcified regions were analysed. After scanning, joints were decalcified and processed for paraffin embedding; serial coronal sections were cut at 6um and sections at regular intervals stained with Toluidine Blue. Results: Fibrillin-1 was localised in the pericellular matrix of articular chondrocytes in normal joints. During the development of OA in Str/ort mice, however, Fibrillin-1 immunolabelling was decreased, in particular around articular cartilage lesions. TSK mice microCT images showed important ectopic calcification in various ligaments, as well as osteophyte formation on the margins of the joints. Preliminary data suggest that articulate cartilage degradation was increased in TSK mice compared to aged-matched control mice. Conclusion: Our preliminary study shows that Fibrillin-1 protein expression is decreased during OA development in a well known model of spontaneous OA, suggesting TGFb bioavailability may be modified during this period. In addition, mutations in Fibrillin-1 in the TSK mouse lead to abnormal ossification in the knee joint as well as severe OA development. These data suggest that Fibrillin-1 plays an important role in joint homeostasis and that abnormal expression or mutations in its gene can lead to OA development.

Results: Haematoxilin-Eosin staining showed that hBM-MSCs have grown through the scaffolds. The cellular percentage respect the whole scaffold area was higher than 50% in all the biomaterials and it was observed a big amount of extracellular matrix (ECM) in all of them, except in Col⫹OLH3. The ECM showed proteoglycans, by means of safranin O (SO) staining, and immunopositivity for Col II in all the constructs, except in Col⫹OLH3 (Figure). By measurement of relative expression levels (REL) we observed that COLII were expressed by cells in Col I (0.63), Col I⫹Col II (1.00), Col I⫹3%HS (0.47), Col I⫹Col II⫹HS (0.93), Col I⫹Col II⫹CHS (3.57) and Col I⫹OLH3 (5.20). Moreover, REL of AGG were higher in Col I⫹Col II⫹HS (6.47) constructs than in Col I (0.00), Col I⫹Col II (0.95),Col I⫹3%HS (0.17), Col I⫹Col II⫹CHS (0.00) and Col I⫹OLH3 (0.00). Electron microscopy showed cells with a big amount of mitochondria and with spherical and oval shapes (Figure). Col released was detected in all supernatant cultures, obtaining the highest concentration (␮g/total volume) at all times, in Col I⫹3%HS cultures except at day 21: Col I (137.33), Col I⫹Col II⫹HS (3.33), Col I⫹Col II (0.00),Col I⫹3%HS (0.00), Col I⫹Col II⫹CHS (13.90) and Col I⫹OLH3 (0.00).

Sunday, October 27

76 Protective Effect Of Intra-Articular Injection Of Alginate-Chitosan Beads In An Hydrogel Against The Development Of Experimental Osteoarthritis Lesions In Rabbit. Fre´de´ric Oprenyeszk1, Mickael Chausson2, Ve´ronique Maquet2, Jean-Emile Dubuc3 and Yves Henrotin1. 1 Bone and Cartilage Research Unit, Lie`ge, Belgium, 2KitoZyme SA, Herstal, Belgium, 3Orthopaedic Department, Bruxelles, Belgium. Background/Purpose: Today, there is no treatment to cure osteoarthritis (OA) or to delay effectively its progression. Current treatments are mainly based on the alleviation of painful symptoms but are unable to restore the cartilage. The development of biomaterial is a promising approach. Herein, we report the evaluation of a new biomaterial composed of alginate-chitosan (AC) beads dispersed in an (H) hydrogel derived from chitosan on OA rabbit model. Methods: OA was surgically induced by the transection of the anterior cruciate ligament (ACLT) in HYLA albino rabbits. One week after surgery, animals were randomly divided into 3 groups: group I (n⫽7): mix of AC beads and H hydrogel; group II (n⫽7): H hydrogel alone; group III (n⫽7): saline solution (control). The treatments (900 ␮l) were injected intraarticularly. X-rays from the right knee were performed before surgery, at the time of injection and at sacrifice. The standard radiographs were acquired in extension and scored by the Kellgren and Lawrence (K&L) scale. After 6 weeks, animals were euthanized and the right joint was dissected. The macroscopic evaluation of cartilage from femoral condyles and tibial plateaus stained with India ink was done. Histological sections from bearing areas of each compartment and synovial membrane were stained with Safranine-O/ fast green or hematoxylin/eosin. Histological evaluation was done according to the OARSI histological score where 0 represented a normal situation and 24 points the maximum severity score. Blood samples were collected the day of injection and prior the sacrifice. Prostaglandin E2 (PGE2) and C-reactive protein (CRP) were measured in serum using immunoassays. Results: The X-rays analysis showed a significant decrease (p ⬍0.05) of the K&L score in group I (AC beads and H hydrogel; 1.5 ⫾ 0.2) compared with group II (H hydrogel; 2.2 ⫾ 0.5) and group III (saline solution; 3.0 ⫾ 0.4). The size and the severity of the macroscopic OA cartilage lesion tended to decrease in group I compared to the other groups. The histological global score that refers to all compartments of the knee joint was significantly decreased in group I (11.0 ⫾ 0.7) compared to group II (14.4 ⫾ 0.6, p ⬍0.01) and group III (14.8 ⫾ 0.6, p ⬍0.001). The injection of AC beads and H hydrogel also tended to reduce the synovial membrane inflammation. No significant variation of PGE2 and CRP serum levels were observed in each after 6 weeks follow-up whatever the treatment injected. Conclusion: Alginate-chitosan beads dispersed in H hydrogel prevented OA in rabbit after ACL transection. This effect was not observed with the hydrogel alone, suggesting that AC beads play a role in joint protection. The preventive effect was observed in all joint compartments indicating a global protective effect of this new injectable biomaterial.

Figure. Images collected from the different constructs by electron microscopy analysis, histochemistry and immunohistochemistry. Vertically in columns are shown the constructs from the different types of collagen scaffolds. Ordered in rows are shown the different analysis: Scanning Electron Microscopy (SEM), Transmission Electron Microscopy (TEM), Safranin O staining (SO) and Collagen type II immunostaining (Col 2).

Conclusion: Data showed that hBM-MSCs are capable of growing on collagen scaffolds. On HS enriched collagen scaffolds was kept the differentiated phenotype. The ECM characteristics and the gene expression showed the synthesis of a cartilage-like neotissue which could be useful for cartilage tissue engineering. Acknowledgements: Opocrin S.P.A.; CAM (S2009/MAT1472); CIBER-BBN CB06-01-0040; SAI-UDC; REDICENT; Diputacioń de A Corunã; Jimeńez Dıáz Fundation. Disclosure: C. Sanjurjo-Rodrı´guez, None; A. H. Martıńez-Sańchez, None; T. Hermida-Go´mez, None; I. M. Fuentes-Boquete, None; F. J. De Toro, None; J. Bujań, None; S. Dıáz-Prado, None; F. J. Blanco, None.

Disclosure: F. Oprenyeszk, None; M. Chausson, None; V. Maquet, None; J. E. Dubuc, None; Y. Henrotin, None.

78

77

Norepinephrine (NE) Inhibits Mesenchymal Stem Cell (MSCs) Chondrogenesis By Accelerating Hypertophy – Relevance For Cartilage Regeneration. Zsuzsa Jenei-Lanzl1, Peter Angele2, Frieder Kees3, Georg Pongratz4 and Rainer H. Straub5. 1Laboratory of Experimental Rheumatology and Neuroendocrine Immunology, Regensburg, Germany, 2Department of Trauma Surgery, University Hospital Regensburg, Germany, Regensburg, Germany, 3University of Regensburg, Regensburg, Germany, 4University Hospital Regensburg, Regensburg, Germany, 5Laboratory of Exp. Rheumatology and Neuroendocrino-Immunology, University Hospital of Regensburg, Regensburg, Germany.

Human Mesenchymal Stem Cells Cultured On Collagen Scaffolds For Cartilage Tissue Engineering. Clara Sanjurjo-Rodrı´guez1, Adela Helvia Martıńez-Sańchez1, Tamara Hermida-Go´mez1, Isaac M. Fuentes-Boquete2, Francisco J. De Toro2, Julia Bujań3, Silvia Dıáz-Prado2 and Francisco J. Blanco1. 1Osteoarticular and Aging Res. Lab. CIBER-BBN. INIBIC- University of A Corunã, A Corunã, Spain, 2INIBIC-Universidade da Corunã, A Corunã, Spain, 3Department of Medical Specialties, Madrid, Spain. Background/Purpose: Osteoarthritis is a degenerative disease without a treatment. Tissue Engineering could provide an alternative tool for cartilage repair. The aim of this study was to evaluate the neotissue formed using human bone marrow mesenchymal stem cells (hBM-MSCs) and collagen scaffolds. Methods: hBM-MSCs were cultured on collagen (Col), Col⫹heparan sulfate (HS), Col⫹chondroitin sulfate (CHS) and Col⫹heparin (OLH3) scaffolds, in chondrogenic medium supplemented by TGF␤-3. Chondrogenic differentiation and the constructs were evaluated by histochemistry, immunohistochemistry, and molecular biology. Cellular morphology and ultrastructure were studied by electron microscopy. Culture supernatants were collected and the amount of secreted collagen was measured by ELISA.

Background/Purpose: The high potential of mesenchymal stem cells (MSCs) in cartilage regeneration is undoubtful. The presence of MSCs in healthy or arthritic cartilage has also been confirmed. In addition, MSCs migrating from synovium into the cartilage and differentiating into chondrocytes has been described, which might be the reason for increased MSC number in OA. Furthermore, it is known that sympathetic nerve fibers are present in healthy and osteoarthritic (OA) synovium and that the sympathetic nervous system mediates numerous effects on adult skeletal system. Therefore, the aim of this study was to investigate the effects of the most important sympathetic neurotransmitter norepinephrine (NE) on MSC chondrogenesis.

S30

(inhibition 92% and 43% respectively). Next, we explored the effect in a condition with less synovial inflammation. Synovial thickness at day 42 was 62% lower when compared to the former study (from 2.6 ⫹/⫺1 to 1 ⫹/⫺0.2) which was reflected by lower S100A8/A9 serum levels (at day 14 around 50 ng/ml compared to 800 ng/ml). Injection of the same dose of ADSCs at day 7 after induction of OA, did not inhibit synovial thickening nor chondrogenesis in the collateral (1.3⫹/⫺0.2 versus WT controls 1.1⫹/⫺0.3) and cruciate ligaments (1.8⫹/⫺0.3 versus WT controls 1.9⫹/⫺0.2) when measured at day 42. Conclusion: Our study indicates that synovial activation rapidly drives anti-inflammatory effects of ADSCs after local administration in murine knee joints with experimental OA and protects against development of ligament damage. Disclosure: P. L. E. M. van Lent, None; R. Schelbergen, None; M. C. ter Huurne, None; A. B. Blom, None; J. Roth, None; T. Vogl, None; C. Jorgensen, None; W. B. van den Berg, None.

80 Tidemark Duplication In Osteoarthritis: Evidence Of Incremental Progression? Martine P. Roudier, Paul A. Manner and Peter A. Simkin. University of Washington, Seattle, WA. Background/Purpose: Osteoarthritis (OA) is commonly envisioned as a gradual, relentlessly progressive process of “wear and tear” leading to eventual joint failure. As we see the tidemark (TM) between hyaline and calcified cartilage, we read a different message. Methods: We studied the femoral condyles and tibial plateaux of 18 elderly body donors for evidence of tidemark duplication. We used 62 specimens representing at least 2 specimens per donor. Specimens were photographed, sampled, radiographed, decalcified using 10% formic acid, processed and embedded in paraffin, sectioned then stained with H&E, Safranine O and Mallory Aniline Blue. They were graded using the OARSI 2005 6 point-grading system (G0-G6). When two lesions existed in the same specimen, each lesion was individually graded. A TM number was recorded for each graded lesion. A Spearman correlation between the lesion grade and the number of tidemarks was determined using Graphpad Prizm software. Results: TM number correlated significantly with OA histological grade (r⫽0.81, p⬍0.0001). Thirty three of the 62 specimens had 2 lesions generating 95 grades. Twelve grades were without visible TM (1 G5, 7 G6 and 4 technical issues) and were excluded from the study. Of the 83 remaining grades, 17 were G0 and had 1 TM, 27 were G1 and had 1 or 2 TMs, 22 were G2 and had 2 or 3 TMs, 9 were G3 and had 1, 2 or 3 TMs, and 8 were G4 and had 3 or 5 TMs.

Disclosure: Z. Jenei-Lanzl, None; P. Angele, None; F. Kees, None; G. Pongratz, None; R. H. Straub, None.

79 Synovial Activation In Experimental OA Drives Immuno suppressive Effects Of Adipose-derived Stem Cells After Local Administration and Protects Against Chondrogenesis In Ligaments. Peter L.E.M. van Lent1, Rik Schelbergen1, Menno C. ter Huurne1, Arjen B. Blom1, Johannes Roth2, Thomas Vogl2, Christian Jorgensen3 and Wim B. van den Berg4. 1Radboud University Nijmegen Medical Centre, Nijmegen, Netherlands, 2University of Muenster, Muenster, Germany, 3Inserm U844, CHU saint-Eloi, Universite´ Montpellier 1, CHU Lapeyronie, Montpellier, France, 4Rheumatology Research and Advanced Therapeutics, Department of Rheumatology, Radboud University Nijmegen Medical Centre, Nijmegen, Netherlands. Background/Purpose: A substantial subpopulation of patients with early osteoarthritis (OA) show a thickened synovial lining layer comprising macrophages expressing an activated phenotype (reflected by production of IL-1␤ and S100A8/A9). Recently we have shown that adipose derived stem cells (ADSC) inhibit joint destruction after local application to knee joints with experimental OA. The goal of our study was to explore the effect of synovial activation on the immunosuppressive capacity of ADSCs to modulate joint destruction after local administration to experimental OA knee joints. Methods: ADSCs were isolated from fat surrounding the popliteal lymph nodes and cultured for two weeks. Collagenase induced OA (CIOA) was induced by injection of collagenase into murine knee joints. ADSCs were injected into the same joints at day 7 after induction of OA. Joint destruction was measured within 6 weeks after induction. Total knee joints were isolated and processed for histology. Synovial thickening was measured using an arbitrary scale of 0–3. Synovium was isolated at various time-points after injection of ASC and washouts were measured for S100A8/A9 and IL-1 using Luminex. Chondrogenesis/bone formation inside ligaments was measured using image analysis. Results: A single dose of ADSCs (20⫻103) was injected into the knee joint of mice, 7 days after induction of CIOA. Thickening of the synovial lining layer, which is characteristic for this model, was high (2,6⫹/⫺1) and was significantly inhibited by ADSC treatment at day 14 (9%) and day 42 (35%) when compared to control (serum) treated OA joints. Washouts of synovium taken at 6 hrs, 48 hrs, day 14 and day 42 after injection of ADSCs showed that protein levels of IL-1␤ and S100A8/A9 were significantly decreased already 48 hrs after injection of ADSC (IL-1␤ 57% and S100A8/A9 by 22%) and rapidly declined thereafter. Serum levels of S100A8/A9 were inhibited with 85% (from 794⫹/⫺ 294 to 117⫹/⫺ 29 ng/ml) at day 14 after ADSC treatment suggesting that the effect on synovial activation is very rapid. Strikingly, ADSCs had a protective effect on chondrogenesis in medial and cruciate ligaments at day 42 after treatment

Conclusion: The number of TMs correlated highly with the OARSI histological grade, suggesting that the TM number is a reflection of the OA disease stage. Since the TMs were distinct, we believe that duplication represents the legacy of abrupt, sequential changes leading to regional calcification of deep hyaline cartilage. The calcification buries the previous tidemark and causes the tidemark formation process to begin anew at a higher level. The most probable precipitating event would be a change in joint loading such as that resulting from a focal microfracture of subchondral bone. Disclosure: M. P. Roudier, None; P. A. Manner, None; P. A. Simkin, None.

S31

Sunday, October 27

Methods: To find possible natural sources of local catecholamines in human joint material, we studied sympathetic nerve fibers (tyrosine hydroxylase (TH) expression), catecholamine biosynthesis and synovial fluid levels. Human bone marrow derived MSCs were expanded and chondrogenesis was initiated in a 3D aggregate culture. Endogenous catecholamine production of hMSCs was analyzed during proliferation and chondrogenesis via HPLC. Parallel to control conditions, aggregates were incubated with different NE concentrations and specific ␤-adrenoceptor agonist (isoproterenol) or antagonist (nadolol). After 21 days, chondrogenesis quality was evaluated macroscopically, histologically and biochemically. Furthermore, specific adrenoceptors (AR) were detected. Results: In the human joint, TH-positive fibers and/or single cells were present in synovial tissue, meniscus, and subchondral bone marrow. In addition, knee-traumatized patients demonstrated high NE concentrations in synovial fluid. At various stages of human MSC chondrogenesis, ␤AR were expressed. No endogenous catecholamine synthesis was detected during chondrogenesis. Chondrogenic MSC aggregates treated with NE or isoproterenol synthesized lower amounts of type II collagen and glycosaminoglycans. NE and isoproterenol dose-dependently increased markers of cartilage hypertrophy (collagen type X and MMP-13). Nadolol reversed inhibition of chondrogenesis and up-regulation of cartilage hypertrophy. Conclusion: The suppression of MSC-dependent chondrogenesis by high NE or isoproterenol suggests that the ␤-adrenoceptors mediate this effect. By inhibition of cartilage repair, these sympathetic influences can be important after multiple minor and major joint traumata. These findings can be a basis for novel neuro-chondrogenic therapeutic options.

minimising accessory fibrocartilages are frequently observed in the immediately adjacent joint cavities. Collectively, these structures are termed synovioentheseal complexes (SECs) and are subject to the full gamut of degenerative changes associated with the osteoarthritis (OA) disease process.1 Here we have investigated the SEC formed at the posterior cruciate ligament (PCL) tibial insertion using high resolution magnetic resonance imaging (MRI) and matched histology to show the involvement of the PCL-SEC in the phenotypic expression of knee OA. Methods: High-resolution 3T MRI and comparative histological evaluation were performed on 14 normal cadaveric knee joints. SEC microanatomy and OA features of the PCL tibial attachment were evaluated on serial histological sections stained with H&E and Masson’s Trichrome. Bilateral 3T MRI images were evaluated in 49 patients with early OA from the Osteoarthritis Initiative cohort for SEC morphology and for lesions associated with the early OA disease process at the PCL tibial enthesis. Differences among groups were determined by calculation of chi-square values. Results: Histological evaluation confirmed the SEC microanatomical structure comprising the PCL, adjacent tibial bone plateau and intervening synovium along with sesamoid, periosteal and enthesis accessory cartilages and the posterior horn of the medial meniscus. Evidence of microanatomical damage was observed throughout the PCL-SEC, including; neovascularisation, chondrocyte cell clustering, tide mark duplication and mild inflammatory changes. In the early OA cohort, 3T MRI showed a high prevalence of SEC related lesions including bone marrow oedema (BME) immediately anterior to the PCL insertion (seen in 61% of knees) and osteophytosis lateral to the PCL insertion (85%). Signal hyperintensity on fluid sensitive sequences compatible with posterior recess synovitis immediately adjacent to the PCL was seen in 48% of knees and correlated with SEC cartilage abnormality (␹2 ⫽7.25, p⬍0.01). Furthermore, the frequency of BME at the PCL insertion was associated with a medial meniscus posterior horn deficient SEC anatomical variant (␹2 ⫽10.02, p⬍0.05). Conclusion: The PCL has a prominent SEC configuration that is associated with microscopic evidence of OA in normal tissue. In early knee OA, pathological features were commonly observed at the PCL-SEC and immediately adjacent locations on 3T MRI. Although the PCL rarely fails in OA, the PCL-SEC could play a hitherto unappreciated but important role in the phenotypic expression of early knee OA.

Sunday, October 27

81 Synovial Expression Of CCL19, MCP-1, and Their Receptors In Patients With Meniscal Injury: Variability and Relationship With Knee Symptoms. Justin B. Gan, Anjali Nair, Kanta Saha, Nikhil Verma, Charles Bush-Joseph, Matthew Tetreault, Anne-Marie Malfait, Kumar B. Rajan and Carla R. Scanzello. Rush University Medical Center, Chicago, IL. Background/Purpose: Patients with meniscal injury are at increased risk for symptomatic knee osteoarthritis (OA), and often already have intraoperative signs of early-stage cartilage degradation. We previously reported that patients with meniscal tears often exhibit synovial inflammation, which is associated with more severe knee symptoms. We further demonstrated that synovitis was associated with expression of mRNA for CCL19, a lymphocyte chemoattractant, and its receptor CCR7. Others have reported a relationship between MCP-1, a macrophage chemoattractant, and knee pain in patients with knee injuries including meniscal tears. We sought to determine whether CCL19 and MCP-1 are measurable in synovial fluid (SF) of patients undergoing meniscal arthroscopies, and if protein levels are associated with knee symptoms or dysfunction. We also investigated expression patterns of CCR7 (CCL19 receptor) and CCR2 (MCP-1 receptor) within synovium to gain insight into the cell types that might be responsive to these factors. Methods: We recruited subjects undergoing arthroscopy for meniscal tears. SF CCL19 and MCP-1 were measured by ELISA (R&D systems). Knee symptoms and dysfunction were preoperatively assessed using the Knee Injury and Osteoarthritis Outcome Score (KOOS), consisting of 5 subscales measuring pain, other knee symptoms, function in daily living (ADL), recreational activities, and quality of life. Associations with KOOS scores were tested using Pearson correlations. Multivariate linear regression models adjusted for age, gender, BMI, and degree of radiographic OA (KellgrenLawrence scores). Synovial CCR7 and CCR2 expression was characterized using immunohistochemical (IHC) staining, and compared to staining patterns in synovium from asymptomatic organ donors. Results: 32 patients with meniscal tears were included. Median age (IQR) was 55 (47–61), and gender ratio M:F was 17:15. CCL19 was detectable in 25/32 (78%) while MCP-1 was detectable in all patients. Mean concentration⫹/⫺SD of CCL19 was 374.28⫹/⫺367.58 pg/mL, and MCP-1 was 450.13⫹/⫺273.85 pg/mL. CCL19 concentration correlated with worse KOOS-ADL subscores (Pearson r⫽ ⫺0.362, p⫽0.049), but was not independent of other factors in adjusted regression analyses. In the model, BMI was an independent predictor of KOOS-ADL subscores (p⫽0.01), and there was a correlation between BMI and CCL19 levels (Pearson r⫽0.430, p⫽0.02). MCP-1 levels did not correlate with KOOS subscores. IHC staining identified CCR7 expression in the synovial lining layer, endothelium, and perivascular inflammatory infiltrates, while CCR2 expression was identified primarily in the lining layer. CCR7 and CCR2 staining was more pronounced in patients with meniscal tears than in asymptomatic organ donors. Conclusion: In patients with meniscal tears, CCL19 and MCP-1 are detectable in SF. Although SF CCL19 was not an independent predictor of knee symptoms, CCL19 may be related to increased BMI. CCR7 did appear increased in meniscal tear patients, and its expression by multiple cell types in synovium suggests that CCL19/CCR7 activity may be involved in development of synovial changes seen in these patients.

References: 1. Benjamin M, McGonagle D. Arthritis & Rheumatism 2007 56:3601–09. Disclosure: D. Binks, None; D. Bergin, None; T. Freemont, None; A. Radjenovic, None; D. McGonagle, None.

83 Oleuropein Or Rutin Consumption Decreases The Spontaneous Development Of OA In Hartley Guinea Pig. Christelle Sanchez1, Marie-Noe¨lle Horcajada2, Fanny Membrez Scalfo2, Pierre Drion3, Fanny Comblain3, Elizabeth Offord2 and Yves Henrotin3. 1University of Lie`ge, Liege, Belgium, 2 Nestle Research Center, Lausanne, Switzerland, 3University of Lie`ge, Lie`ge, Belgium. Background/Purpose: To assess the potential protective effect of oleuropein and rutin, two polyphenols found in olive oil, fruits or vegetables, on osteoarthritis development Methods: Sixty 4 weeks old Hartley guinea pig were randomized in four groups and received daily during 31 weeks either standard guinea pig food (control group) or a standard guinea pig food enriched with oleuropein, rutin or rutin/curcumin association. Animals were weighted each week and blood sampled every 6 weeks and at the time of euthanasia (week 35). Biomarkers COLL2-1, COLL2-1NO2, Fib3-1, Fib3-2, as well as PGE2, were quantified in the serum. Histological assessments of knee cartilage and synovial membrane were performed at week 35. Results: At week 35, guinea pigs in the control group spontaneously developed important cartilage lesions with mild synovial inflammation. The histological scores of cartilage lesion and synovitis were well correlated with the increase of serum biomarkers level.. Histologically, all treated groups significantly reduced the cartilage degradation (p⬍0.01), and oleuropein group showed a significant decrease of the synovial modification (p⬍0.05) compared to the control group. Oleuropein decreased the PGE2 levels found in serum at week 35 (p⬍0.01). Serum COLL2-1 and Fib3-1 were decreased by rutin and rutin/curcumin mixture, Fib3-2 was only decreased by rutin/ curcumin mixture, while COLL2-1NO2 was significantly decreased by all treatments (p⬍0.05).

Disclosure: J. B. Gan, None; A. Nair, None; K. Saha, None; N. Verma, None; C. Bush-Joseph, None; M. Tetreault, None; A. M. Malfait, None; K. B. Rajan, None; C. R. Scanzello, U.S. patent pending for synovitis markers in osteoarthritis and joint injury, 9.

82 Anatomical Variation In The Morphology Of The Posterior Cruciate Ligament Synovio-Entheseal Complex and Correlation With Degenerative Change. Daniel Binks1, Diane Bergin2, Tony Freemont3, Aleksandra Radjenovic1 and Dennis McGonagle4. 1University of Leeds, Leeds, United Kingdom, 2Galway University Hospitals, Galway, Ireland, 3University of Manchester, Manchester, United Kingdom, 4Leeds Institute of Rheumatic and Musculoskeletal Medicine, University of Leeds and NIHR Leeds Musculoskeletal Biomedical Research Unit, Leeds Teaching Hospitals NHS Trust, Leeds, United Kingdom. Background/Purpose: Ligament attachments are often found to form complex anatomical functional units involving the ligaments themselves as well as associated synovium and bony tuberosities. Furthermore, stress

S32

deposition does indeed occur in patients with osteoarthritis of the knee who have never experienced an episode of gout. Methods: We recruited patients with advanced osteoarthritis of the knee and no history of gout who were scheduled to undergo knee replacement surgery. All subjects underwent dual-energy scanning of the target joint. CT images were screened for areas of uric acid deposition.During knee replacement surgery, samples of tibial and femoral condyle cartilage, menisci and synovium were obtained. Areas of possible uric acid deposition as identified through DECT were particularly targeted for sample collection.A musculoskeletal pathologist performed light (DeGalantha and H&E stains) and polarizing microscopy and recorded the presence of uric acid crystals in a descriptive manner. Results: 5 patients with advanced osteoarthritis of the knee did undergo DECT. In all subjects, DECT indicated areas of uric acid deposition within the joint cartilage and the menisci. In 4 of these 5 patients, histopathology exam confirmed presence of uric acid crystal clusters in the cartilage and/or menisci. In 2 patients, uric acid deposition did also involve the synovium. Of note, synovial fluid in all patients was negative for MSU crystals as determined through polarizing microscopy. The mean serum uric acid level was 5.7mg/dl, range 5.0–6.4 mg/dl. Conclusion: This pilot study did reveal evidence of cartilaginous uric acid deposition in the majority of patients with advanced osteoarthritis.Future studies will have to confirm this finding in a larger patient cohort and clarify if intracartilaginous urate crystals accelerate cartilage damage or are simply a result of matrix exposure and subsequent MSU crystallization without significant effects on disease progression.

Disclosure: C. Sanchez, Nestle´, 2; M. N. Horcajada, None; F. Membrez Scalfo, None; P. Drion, None; F. Comblain, None; E. Offord, None; Y. Henrotin, Artialis, 1, BioIberica, Artialis, Tilman, Expanscience, 5, Nestle´, 2.

84 New Formulation With Potential To Prevent and Treat Osteoporosis and Osteoarthritis. A. Torrent1, E. Montell1, J. Verge´s1, P. Dalmau2, R. Ruhı´2, M.C. Carceller3, A. Blanco3, M.C. Terencio3, M.L. Ferrańdiz3 and M.J. Alcaraz3. 1Pre-Clinical R&D Area, Pharmascience Division, BIOIBERICA S.A., Barcelona, Spain, 2Technological Extraction Dept, BIOIBERICA S.A., Palafolls, Spain, 3Department of Pharmacology and IDM, University of Valencia, Burjassot, Spain. Background/Purpose: Osteoarthritis (OA) is a multidimensional disease that affects all anatomical joint structures, particularly cartilage, synovium and subchondral bone. In turn, osteoporosis (OP) is a skeletal disorder characterized by a compromised bone strength which substantially increases the risk of fracture. In the past, attention was focused on a supposed inverse relationship between OA and OP, since both disorders usually affect the elderly, but were regarded to rarely coexist in a single person. However, recent studies have revealed several factors which contribute to the pathogenesis of both disorders (Bultink et al, 2013). Despite this, there is not any drug at the moment approved for the simultaneous prevention and treatment of osteoporosis and osteoarthritis. The aim of this study was to investigate the effect of a new formulation in a combined rat model of OP and OA. The formulation (BIS076) contains Vitamin D3, Hydroxyapatite as a source of calcium and a natural extract from porcine cartilage. The latter is rich in bioactive substances due to the mild conditions used in the manufacturing process. Methods: OP was induced by ovariectomy (OVX) in female Wistar rats and, two weeks after, OA was induced by Anterior Cruciate Ligament Transection (ACLT). Sixty female rats were assigned into the following groups: Sham Group, OVX ⫹ ACLT Group (Vehicle) and BIS076 Groups. BIS076 was administered daily during 12 weeks at two doses, 163.5 mg/kg/day and 245 mg/kg/day which correspond approximately to 1400 mg/day and 2100 mg/day in humans. For the assessment of OA, histology was performed and cartilage degeneration was evaluated by means of the OARSI score (Pritzker et al, 2006). Synovitis degree was estimated according to the score proposed by Krenn et al (2006). Bone microarchitecture and density were assessed by Micro-Computed Tomography (microCT). Results: The preparation BIS076 has been shown to induce, at the 2 doses tested, a significant reduction (approximately 50%) of the cartilage degradation according to the OARSI score. Synovial inflammation was strongly reduced as well. In addition, microCT revealed that BIS076 treatment exerted a positive effect in bone structure, especially at the high dose: Significant increase in bone volume (p⬍0.05), bone surface density (p⬍0.01), trabecular number (p⬍0.01) and significant reduction in the trabecular bone pattern factor (p⬍0.01) compared to the Vehicle Group. Conclusion: Our data demonstrate that treatment with BIS076 could be an effective strategy to control the progression of experimental Osteoarthritis and Osteoporosis. This approach holds promise for the development of improved therapies targeting these chronic and disabling diseases.

Disclosure: T. Bongartz, None; A. M. Oliveira, None; R. J. Sierra, None; A. D. Hanssen, None; M. J. Taunton, None.

86 The Role Of Meniscal Cells In Osteoarthritis Calcification. Andrea Roberts1, David Mauerhan2 and Yubo Sun2. 1Carolinas Healthcare System, Charlotte, NC, 2Carolinas Medical Center, Charlotte, NC. Background/Purpose: Osteoarthritis (OA) is a disease that is characterized by the breakdown or loss of articular cartilage due to biomechanical and biochemical changes in the joints of the knees, hips and hands. Chondrocytes, the cells present in articular cartilage, have traditionally been thought to be exclusively responsible for inducing the calcification observed in OA, and therefore are believed to play an essential role in the development of OA pathogenesis. However, recent reports suggest that meniscal degeneration and calcification are correlated with articular cartilage degeneration in the knee of OA patients. Joint calcification correlates greatly with OA, as 70% of OA patients show an increase in calcium crystal deposition compared to non-OA patients. There are two main types of crystals present in advanced OA patients: calcium pyrophosphate dehydrate (CPPD) and basic calcium phosphate (BCP). However, it is still unclear which tissue (cartilage or meniscus) produce these calcium crystals. The role of meniscal cells in calcification is poorly understood at present, thus the purpose of this study is to determine the role of meniscal cells in the pathogenesis of osteoarthritis (OA) in the knee. Our central hypothesis states that meniscal cells play an unrecognized role in the development of OA in part by inducing calcification in the meniscus. Methods: Tissue was collected from menisci and articular cartilage obtained from OA patients undergoing joint replacement surgery. Calcium Crystals were detected using alizarin red and Eosin Y staining. Cell were isolated from OA menisci and articular cartilage tissue. The cells were treated with and without extracellular ATP (a calcification inducer) and calcification medium for seven days in a monolayer culture. Alizarin Red staining was used to detect calcification present in a monolayer. Images were obtained using Sony Imaging software with an inverted scope and alizarin red was quantified measuring HCL absorbance at 405 wavelength using a microplate reader. 45 Calcium Assay was used to quantify calcification induced by chondrocytes and meniscal cells in the presence and absence of extracellular ATP. The cells were cultured in a monolayer environment for 5 days. 45Calcium was measured as counts per minute using a liquid scintillation counter. Results: Our data, using a 45Calcium assay and alizarin red staining, suggest that meniscal cells, similar to chondrocytes, induce the formation of calcium crystals in the presence of extracellular ATP and calcification medium, with meniscal cells inducing calcification at a slightly higher rate than chondrocytes. Additionally, OA patients have been shown to contain many types of calcium crystals in the knee cartilage, with BCP and CPPD crystals being the most abundant. Both of which have been implicated in the development and progression of knee OA. Our preliminary data shows the

Disclosure: A. Torrent, BIOIBERICA, 3; E. Montell, BIOIBERICA S.A., 3; J. Verge´s, BIOIBERICA S.A., 3; P. Dalmau, BIOIBERICA S.A., 3; R. Ruhı´, BIOIBERICA S.A., 3; M. C. Carceller, None; A. Blanco, None; M. C. Terencio, None; M. L. Ferrańdiz, None; M. J. Alcaraz, None.

85 Cartilaginous Uric Acid Deposition In Advanced Osteoarthritis: Innocent Bystander Or Promotor Of Cartilage Destruction? Tim Bongartz, Andre M Oliveira, Rafael J Sierra, Arlen D Hanssen and Michael J Taunton. Mayo Clinic, Rochester, MN. Background/Purpose: Recent studies have suggested that synovial fluid uric acid could contribute to tissue inflammation, disease severity and progression of osteoarthritis (OA). Intraarticular uric acid has been associated with OA disease severity and may promote cartilage destruction through inflammasome activation.We aimed to explore if intraarticular uric acid

S33

Sunday, October 27

Conclusion: Oleuropein and rutin significantly slow down the progression of OA lesions in guinea pig developing spontaneously OA. Furthermore, oleuropein significantly decreased PGE2 and COLL2-1NO2 serum levels, and reduced synovitis, indicating the potent anti-inflammatory properties of these compounds.

Sunday, October 27

meniscus of OA patients contains both BCP and CPPD crystals of which about 44% are CPPD crystals. Conclusion: Our results suggest meniscal cells like chondrocytes induce calcium crystal formation and both BCP and CPPD crystals are present in the meniscus of OA patients.

Disclosure: L. March, None; D. Hoy, None; E. Smith, None; R. Buchbinder, None; M. Cross, None; P. Brooks, None; T. Vos, None; A. D. Woolf, None.

Disclosure: A. Roberts, None; D. Mauerhan, None; Y. Sun, None.

88

account for the greatest proportion of MSK burden globally. In many countries MSK disorders are the leading cause of disability.

Musculoskeletal Diseases Have The Worst Impact On Physical Health Compared With Other Diseases - Results Of The Dutch cross-Sectional Study. Antje van der Zee-Neuen1, Polina Putrik1, Sofia Ramiro2, Andras P. Keszei3, Rob de Bie1, Astrid M. Chorus4 and Annelies Boonen5. 1Maastricht University, Maastricht, Netherlands, 2Academic Medical Center, University of Amsterdam, Amsterdam, Netherlands, 3School for Oncology and Developmental Biology, Maastricht, Netherlands, 4Netherlands Organization for Applied Scientific Research, Leiden, Netherlands, 5Maastricht University Medical Center, Maastricht, Netherlands.

ACR Poster Session A Epidemiology and Health Services I Sunday, October 27, 2013, 8:30 AM–4:00 PM

87 Global and Country Specific Burden Of Musculoskeletal Disorders:a Report From The Global Burden Of Diseases Musculoskeletal Expert Group. Lyn March1, Damian Hoy2, Emma Smith3, Rachelle Buchbinder4, Marita Cross3, Peter Brooks5, Theo Vos2 and Anthony D. Woolf6. 1University of Sydney Institute of Bone and Joint Research, Royal North Shore Hospital, St Leonards, Australia, 2School of Population Health, University of Queensland, Brisbane, Australia, 3University of Sydney, Insitute of Bone and Joint Research, Royal North Shore Hospital, St Leonards NSW, Australia, 4 Monash Department of Clinical Epidemiology at Cabrini Hospital, Department of Epidemiology and Preventive Medicine, Monash University, Malvern, Victoria, Australia, 5Australian Health Workforce Institute, University of Melbourne, Melbourne, Australia, 6Royal Cornwall Hospital, Truro Cornwall, United Kingdom.

Background/Purpose: Musculoskeletal conditions (MSKC) are among the most common chronic conditions. Increasingly, patients suffer from more than one disease. Moreover, the presence of a co-morbid disease adds to the burden of single diseases worldwide. The aim of this study was to understand 1) what is the impact of the number of morbidities on health and 2) whether MSKC has a higher impact on health compared with other diseases or has an important impact when being a co-morbidity. Methods: In a Dutch cross-sectional study, 8904 subjects (⬎18 years old, random sample) completed a questionnaire on socio-demographic factors (age, gender, education, work status, origin and place of residence (postal code code)), BMI, self-reported physician-diagnosed diseases and the 12-Item Short-Form Health Survey (SF-12). Complete cases (n⫽7600) were analyzed. Multivariable linear regression was computed to identify 1) whether multimorbidity in terms of number of diseases was significantly associated with the SF-12 physical (PCS) and mental (MCS) subscales and 2)which diseases contributed the most to changes in health related quality of life (HrQol). Interactions of all diseases with MSKC were checked. Models were adjusted for age, gender, education (5 groups, from no education to university) origin (western vs. non-western) and BMI. Results: Multimorbidity was present in 1432 subjects (19%). MSKC confirmed by a physician was reported by 1438 (19%) participants. 408 (5%) reported diabetes, 1460 (19%) CVD, 199 (3%) cancer, 547 (7%) a respiratory condition, 526 (7%) a skin condition, 462 (6%) a mental disorder, 334 (4%) migraine and 262 (3%) bowel disease. A linear relation between the number of diseases and health (both univariably and in fully adjusted model) was observed (Table 1). MSKC had the highest negative impact on PCS compared to other diseases (Table 2).

Background/Purpose: The first Global Burden of Diseases (GBD) Study was conducted by the World Bank and WHO during the 1990’s and ranked diseases by their contribution to burden as measured by disability adjusted life years (DALYs). All MSK disorders, with 2.1% of DALYs, ranked 12th. These were considered to be underestimates. The GBD 2010 Study aimed to repeat these DALYs to look for trends in diseases from 1990 to 2010 with updated disease definitions and literature review. The MSK Expert Group (EG) aimed to identify data for OA, RA, Back Pain, Neck Pain, Gout and Other MSK. Methods: Systematic reviews of Medline, EMBASE, CINAHL, SIGLE and WHOLIS were screened from 1980 to 2010. Data were extracted from population-based studies using a Quality Assessment tool developed by the MSK EG. These prevalence and incidence data, together with estimates of possible covariates and relative risk of mortality, were entered into the DISMOD3 software developed by the IHME. Disability weights for health states related to MSK conditions were adjusted for distribution of severity and duration and frequency of each condition to calculate years lived with disability (YLDs) for each of the MSK conditions – OA of hip or knee, RA, Low Back Pain (LBP), Neck Pain (NP), Gout and Other MSK. Results: Globally non-communicable chronic disorders were identified as the growing concern and of these MSK was highlighted. In the overall global burden (DALYs) MSK disorders had increased from 4.7% in 1990 to 6.8% in 2010, and ranked 5th behind Cardiovascular (11.8%), Injuries (11.2%), All neoplasms (7.6%) and Mental and behavioural disorders (7.4%). For burden (DALYs) of specific disorders, LBP ranked 6th; NP 21st; and Other MSK 24th. In the global disability (YLD) estimates, mental and behavioural disorders combined at 22.7% ranked 1st, while combined MSK disorders at 21.3% was a close 2nd. Collectively MSK conditions affect 2,933,367,000 people, account for 166 million YLDs and represent a relative increase of 44.7% since 1990. For disease specific causes of disability (YLDs), LBP ranked 1st globally with 83.1 million; NP ranked 4th with 33.6 million; Other MSK 6th with 28.2 million; OA was 11th with 17.1 million; and RA was 6.7 million. OA was felt to be underestimated in this analysis. OA was identified as one of the most rapidly rising (64% increase) conditions with a rank of 15th in 1990 to 11thin 2010. When developed country level data are examined MSK conditions play an even greater role, for example in Australia where combined MSK conditions (15% of DALYs) were a close second only to combined Cancer (16% of DALYs) as the leading causes of disease burden, and LBP was estimated as the leading disease specific cause of total burden (DALYs). For disability in Australia MSK conditions were the leading cause with 27% of YLDs. Country level data are derived from ihme.org. Conclusion: Data are lacking from the African nations, South America, Eastern Europe and Australasia. OA Knee is the most common MSK condition and is showing the greatest increase. Low back and neck pain

Table 1. Association of number of morbidities and health (physical and mental component SF-12) SF-12 Physical Component B1 [95% CI] Number of morbidities 1 2 3 4 5

univariable ⫺4.83 [⫺5.01; ⫺4.65] ⫺9.66 [⫺10.03; ⫺9.30] ⫺14.49 [⫺15.04; ⫺13.94] ⫺19.32 [⫺20.05; ⫺18.59] ⫺24.15 [⫺25.06; ⫺23.24]

Mental Component B1 [95% CI]

multivariable* ⫺4.06 [⫺4.24; ⫺3.87] ⫺8.11 [⫺8.49; ⫺7.73] ⫺12.17 [⫺12.73; ⫺11.61] ⫺16.22 [⫺16.97; ⫺15.47] ⫺20.28 [⫺21.22; ⫺19.34]

univariable ⫺1.62 [⫺1.82; ⫺1.43] ⫺3.25 [⫺3.63; ⫺2.87] ⫺4.87 [⫺5.45; ⫺4.29] ⫺6.50 [⫺7.27; ⫺5.73] ⫺8.12 [⫺9.08; ⫺7.16]

multivariable* ⫺2.07 [⫺2.27; ⫺1.87] ⫺4.14 [⫺4.54; ⫺3.74] ⫺6.20 [⫺6.81; ⫺5.59] ⫺8.27 [⫺9.08; ⫺7.46] ⫺10.34 [⫺11.35; ⫺9.33]

1 understandardized coefficient

*adjusted for age, gender, education (5 groups, from no education to university), BMI **adjusted for age, gender, education (5 groups, from no education to university), origin (western vs. non-western), BMI

Table 2. Association of type of morbidity and health (physical and mental component SF-12) SF-12 Physical Component B1 [95% CI] Type of morbidity univariable CVD ⫺6.55 [⫺7.04; ⫺6.05] Diabetes ⫺6.70 [⫺7.59; ⫺5.81] Cancer ⫺9.80 [⫺9.06; ⫺6.54] Respiratory ⫺7.41 [⫺8.18; ⫺6.64] condition Skin condition ⫺2.99 [⫺3.79; ⫺2.19] Mental disorder ⫺4.13 [⫺4.98; ⫺3.28] MSKC ⫺11.38 [⫺11.84; ⫺10.92] Migraine ⫺4.56 [⫺5.54; ⫺3.57] Bowel disease ⫺9.26 [⫺10.35; ⫺8.17]

multivariable* ⫺2.60 [⫺3.07; ⫺2.13] ⫺2.05[⫺2.83;⫺1.281] ⫺5.24 [⫺6.29; ⫺4.19] ⫺3.92 [⫺4.57; ⫺3.27]

Mental Component B1 [95% CI] univariable ⫺0.17 [⫺0.65; 0.30] ⫺0.20 [⫺0.63; 1.04] ⫺0.80 [⫺1.98; 0.38] ⫺1.73 [⫺2.46; ⫺1.00]

multivariable** ⫺0,91 [⫺1.40; ⫺0.42] ⫺0.46 [⫺1.27; 0.34] ⫺1.05 [⫺2.146; 0.05] ⫺0.62 [⫺1.30; 0.07]

⫺0.49 [⫺1.15; 0.17] ⫺2.13 [⫺2.87; ⫺1.39] ⫺1.17 [⫺1.86; ⫺0.48] ⫺1.82 [⫺2.53; ⫺1.11] ⫺14.82 [⫺15.54; ⫺14.10] ⫺13.93 [⫺14.66; ⫺13.19] ⫺8.93 [⫺9.37; ⫺8.48] ⫺1.25 [⫺1.73; ⫺0.77] ⫺0.30 [⫺0.76; 0.16] ⫺2.37 [⫺3.20; ⫺1.55] ⫺3.73 [⫺4.65; ⫺2.82] ⫺1.44 [⫺2.30; ⫺0.58] ⫺4.68 [⫺5.61; ⫺3.75] ⫺3.99 [⫺5.02; ⫺2.97] ⫺1.42 [⫺2.39; ⫺0.46]

1 understandardized coefficient

*adjusted for age, gender, education (5 groups, from no education to university), BMI **adjusted for age, gender, education (5 groups, from no education to university),origin (western vs. nonwestern),BMI

S34

Disclosure: A. van der Zee-Neuen, None; P. Putrik, None; S. Ramiro, None; A. P. Keszei, None; R. de Bie, None; A. M. Chorus, None; A. Boonen, None.

89 Meta-Analysis of Randomized Controlled Trials Assessing the Effects of Vitamin D Supplementation On Skeletal Muscle Strength. Charlotte Beaudart1, Fanny Buckinx2, Ve´ronique Rabenda1, Sophie Gillain3, Etienne Cavalier3, Jean Petermans3, Jean-Yves Reginster1 and Olivier Bruyere1. 1 University of Liege, Liege, Belgium, 2University of Lie`ge, Lie`ge, Belgium, 3 CHU de Liege, Liege, Belgium. Background/Purpose: Currently, there is growing evidence that vitamin D plays a role on several tissues including skeletal muscle. Previous studies suggested that vitamin D deficiency is associated with low muscular strength. The objective of this meta-analysis is to summarize the effects of vitamin D supplementation on muscle strength Methods: A systematic research of randomized controlled trials (RCTs), performed between 1966 and February 2013, assessing the effect of vitamin D supplementation on muscle strength has been conducted by two independent reviewers (data sources: Medline, Cochrane Database of Systematics Reviews, Cochrane Central Register of Controlled Trials, manual review of the literature and congressional abstracts). All forms and doses of vitamin D supplementation, with or without calcium supplementation, compared with placebo or control were included. Muscle strength was assessed either by grip strength and/or lower limb muscle strength. The quality of the RCTs was evaluated using the Jadad criteria. Results: Out of the 214 potentially relevant articles, 19 RCTs involving 4824 individuals (mean age: 66.0 years) met the inclusion criteria. Studies show a mean quality score of 3.8/5 points. Results reveal a significant positive effect of vitamin D supplementation on global muscle strength (Figure) with a standardized mean difference (SMD) of 0.096 (95% CI⫽0.007–0.184; p⫽0.034). No significant between-study heterogeneity is found (Q-value⫽ 23.6; p⫽ 0.21; I2⫽ 19.6%). No publication bias was observed as shown with the Egger’s regression analysis (p⫽0.13). Regarding the individual type of strength, 13 studies assessed the effect of vitamin D supplementation on grip strength and 15 on lower limb muscle strength. Results show no significant effect of vitamin D supplementation on grip strength (SMD⫽0.062, p⫽0.264), but a significant positive effect on lower limb muscle strength (SMD⫽0.169, p⫽0.03).

Disclosure: J. M. A. Wijnands, None; W. Viechtbauer, None; K. Thevissen, None; I. C. W. Arts, None; P. C. Dagnelie, None; C. D. A. Stehouwer, None; S. van der Linden, None; A. Boonen, None.

91 Risk Factors For Incident Hyperuricemia During Mid-Adulthood In African American and White Men and Women Enrolled In The Atherosclerosis Risk In Communities Study. Mara McAdams DeMarco1, Andrew Law2, Janet W. Maynard3, Josef Coresh1 and Alan N. Baer4. 1Johns Hopkins Bloomberg School of Public Health, Baltimore, MD, 2Johns Hopkins, Baltimore, MD, 3Johns Hopkins University School of Medicine, Division of Rheumatology, Baltimore, MD, 4Johns Hopkins University, Baltimore, MD. Background/Purpose: Increased serum urate levels are associated with poor outcomes including but not limited to gout. Better understanding of which patients are at risk of developing hyperuricemia may aid in clinical decision-making about treatment of asymptomatic hyperuricemia. However, hyperuricemia risk prediction has been limited because previously published studies were not prospective in nature. The objective of this study is to identify risk factors for the development of hyperuricemia over 9 years in a population-based study, ARIC Methods: This cohort recruited 15,792 individuals in 1987–1989 from 4 US communities and contained 9-years of follow-up; 8,212 participants who had serum urate levels ⬍7.0 mg/dL were included in this analysis. Risk factors for 9-year incident hyperuricemia (plasma urate level of > 7.0 g/dL at visit 4) were identified using an AIC-based selection approach in a modified Poisson regression model. We considered baseline, 3-year, and change in serum urate level over 3 years. Results: The 9-year cumulative incidence of hyperuricemia was 4% and subgroup cumulative incidences were: 5% for men; 3% for women; 6% for African Americans and; 3% for whites. The final adjusted model included 8 predictors for incident hyperuricemia over 9 years: male sex (RR⫽1.78, 95% CI: 1.39–2.27), African-American race (RR⫽1.54, 95% CI: 1.19–2.00), current smoking (RR⫽1.36, 95% CI: 1.04–1.79), as well as basic education (RR⫽1.28, 95% CI: 0.99–1.64), hypertension (RR⫽1.73, 95% CI: 1.37–2.19), coronary heart disease (RR⫽1.57, 95% CI: 0.97–2.54), obesity (class I obesity: RR⫽2.82, 95% CI: 1.97–4.04 and ⱖclass II, RR⫽ 3.93, 95% CI: 2.62–5.91) and eGFR⬍60 (RR⫽3.04, 95% CI: 1.73–5.34). In separate models, serum urate levels at baseline, 3 years after baseline, and change in serum urate level over 3 years in addition to demographic and clinical risk factors were all associated with the development of hyperuricemia (Table).

Figure: Forest plot for summary standardized mean difference for global muscle strength

Conclusion: Based on the studies included in this meta-analysis, vitamin D supplementation has a positive impact on global muscle strength, and more especially, on lower limb muscle strength. Disclosure: C. Beaudart, None; F. Buckinx, None; V. Rabenda, None; S. Gillain, None; E. Cavalier, None; J. Petermans, None; J. Y. Reginster, None; O. Bruyere, None.

90 Insufficient Evidence For An Increase In Prevalence and Incidence Of Gout: A Systematic Review and Meta-Regression Analysis. Jose´ M.A. Wijnands1, Wolfgang Viechtbauer1, Kristof Thevissen2, Ilja C.W. Arts1, Pieter C. Dagnelie1, Coen D.A. Stehouwer2, Sjef van der Linden2 and Annelies Boonen2. 1Maastricht University, Maastricht, Netherlands, 2Maastricht University Medical Center, Maastricht, Netherlands.

S35

Sunday, October 27

Background/Purpose: Estimates on the prevalence and incidence of gout in the general population vary widely and an increase is commonly reported. However, studies on the occurrence of gout have not been reviewed and appraised systematically. The aims of this study were: 1) to estimate the pooled prevalence and the incidence of gout in the general population; 2) to explore which factors contribute to the variation in estimated prevalence and incidence. Methods: Pubmed, Embase and Web of Science were systematically searched for primary studies on the prevalence and incidence of gout in the general population. Two reviewers independently extracted data on prevalence and incidence as well as sources of clinical heterogeneity, methodological heterogeneity, and variation in outcome reporting. Prevalence rates were pooled using a random-effects model and were adjusted for a series of clinical and methodological study characteristics in a meta-regression analysis using the mixed-effects model. Results: Of 1466 articles screened, 77 articles were included, of which 71 reported the prevalence and 12 the incidence of gout. The pooled prevalence based on a random effects model was 0.8% (95% CI 0.6; 1.0) with a high level of heterogeneity (I2 ⫽99.9%). Results from a mixed-effects metaregression model indicated that gender (p⬍0.001), continent (p⬍0.001), consistency in data collection (p⫽0.003), and case definition (p⫽0.001) were significantly associated with gout prevalence and jointly accounted for 77.2% of the heterogeneity. Start of data collection was not associated with the prevalence. The incidence in the total population ranged from 0.06 to 2.68 per 1000 person-years and an increase in incidence was only seen in two studies. Conclusion: The large variation in the prevalence data on gout in the general population is explained not only by well-known factors such as gender and continent on which the study was performed, but also by the case definition of gout. There was insufficient evidence for an increase in prevalence or incidence of gout in recent years.

Conclusion: An increasing number of morbidities is negatively associated with physical and mental HrQol. MSKC are responsible for the largest decrease of physical health.

Results: A total of 10,858 ARIC participants were included in the analysis. The study population was 43% male, 21% African American and the mean age at cohort entry was 54 years (SD⫽5.7). The mean eGFR was 92 (SD⫽14.9) ml/min/1.73 m2. At baseline, 2% participants were classified as having eGFR⬍60 ml/min/1.73 m2; 41% with an eGFR between 60 and 90 ml/min/1.73 m2; and 57% with an eGFR ⬎90 ml/min/1.73 m2. The mean alcohol intake was 40.2 grams/week. There were 274 incident gout cases. There was slight evidence that the risk of gout by kidney function was higher in those who drank ⱖ50 g/week of alcohol (Table; P for interaction⫽0.08). For those with alcohol intakeⱖ50 g/week and an eGFR⬍60, there was 1.65-times (95% CI: 1.10–2.48, P⫽0.02) the risk of gout for every 10 ml/min/1.73 m2 decrease in eGFR. The risk of gout by eGFR was similar for those with alcohol intake ⬍50 g/week. Conclusion: There is slight evidence, potentially due to limited power, that the impact of reduced kidney function is greater for adults with alcohol intake ⱖ50 g/week than for those with lower alcohol intake.

Table. Predictors of incident hyperuricemia over 9 years.

Sunday, October 27

Risk Factor

Male sex Black race High school education or higher Hypertension Coronary heart disease Smoking status Current smoker Former smoker Never smoker Body mass index, kg/m2 BMI ⬎⫽ 35 30 ⬍⫽ BMI ⬍ 35 25 ⬍⫽ BMI ⬍ 30 BMI ⬍ 25 eGFR, ml/min/1.73m2 ⬍60 60–90 ⬎⫽90 Serum urate level at baseline, 1 mg/dL Serum urate level at 3 year follow-up, 1 mg/dL 3-year change in serum urate level, 1 mg/dL

Model 1:

Model 2:

Model 3:

Baseline Serum Urate Adjusted 1.09 (0.84, 1.42) 1.54 (1.20, 1.99) 1.28 (0.99, 1.64)

3 Year Serum Urate Adjusted 1.02 (0.80, 1.31) 1.38 (1.08, 1.77) 1.16 (0.90, 1.50)

Change in Serum Urate Adjusted 1.57 (1.24, 2.01) 1.43 (1.11, 1.84) 1.26 (0.98, 1.61)

1.49 (1.17, 1.90) 1.56 (0.96, 2.54)

1.29 (1.01, 1.65) 1.53 (0.94, 2.50)

1.51 (1.18, 1.93) 1.52 (0.93, 2.48)

1.34 (1.03, 1.75) 0.82 (0.62, 1.08) Reference

1.31 (1.00, 1.71) 0.78 (0.59, 1.03) Reference

1.34 (1.02, 1.75) 0.83 (0.63, 1.10) Reference

2.14 (1.40, 3.27) 1.92 (1.34, 2.77) 1.72 (1.25, 2.36) Reference

2.33 (1.54, 3.52) 2.04 (1.42, 2.94) 1.85 (1.35, 2.54) Reference

3.75 (2.50, 5.63) 2.72 (1.90, 3.89) 2.19 (1.59, 3.00) Reference

2.02 (1.15, 3.54) 0.94 (0.75, 1.19) Reference 2.43 (2.02, 2.93)

2.10 (1.27, 3.45) 1.04 (0.83, 1.31) Reference –

2.97 (1.75, 5.04) 1.19 (0.95, 1.50) Reference –

– –

1.94 (1.80, 2.09) –

Risk of Gout by eGFR stratified by alcohol intake Binary alcohol intake (>ⴝ50 g per week) HR (95% CI) Nⴝ2,542 p-value

–

eGFR* eGFRⱖ90* 602eGFR⬍90* eGFR ⬍60*

1.62 (1.49, 1.76)

Source of support: This study was funded by Takeda Pharmaceuticals U.S.A., Inc. The Atherosclerosis Risk in Communities Study is carried out as a collaborative study supported by National Heart, Lung, and Blood Institute contracts (HHSN268201100005C, HHSN268201100006C, HHSN268201100007C, HHSN268201100008C, HHSN268201100009C, HHSN268201100010C, HHSN268201100011C, and HHSN268201 100012C). Alan Baer was supported by the Donald B. and Dorothy Stabler Foundation. Conclusion: Our results suggest that demographic and clinical risk factors that re routinely collected as part of regular medical care predict the development of hyperuricemia in middle-aged adults.

1.18 (1.03, 1.34) 0.92 (0.68, 1.25) 1.25 (0.96, 1.61) 1.65 (1.10, 2.48)

0.01 0.61 0.09 0.02

Binary alcohol intake ( 25) 57.7% 50% 72.92% Smoker 51.4% 63.27% 27.08% Anti-CCPⴙ 84.1% 80% 92.57% RF ⴙ 71.1% 65.35% 82.35% TJC 3.42 ⴞ 5.02 2.36 ⴞ 3.38 5.53 ⴞ 6.84 SJC 2.72 ⴞ 3.91 2.29 ⴞ 3.33 3.57 ⴞ 4.8 PGH 37.71 ⴞ 19.31 34.41 ⴞ 17.89 44.31 ⴞ 20.52 DAS28 3.5 ⴞ 1.46 3.16 ⴞ 1.33 4.17 ⴞ 1.49 Remission DAS28 < 2.6 % 32% 40.2% 15.7% Low disease activity DAS28 < 3.2 % 48.4% 56.86% 31.37% Pain VAS 34.75 ⴞ 26.97 30.16 ⴞ 21.64 44.74 ⴞ 34.24 mHAQ 0.374 ⫾ 0.474 0.373 ⫾ 0.45 0.371 ⫾ 0.52 CRP 0.82 ⫾ 1.22 0.73 ⫾ 1.2 1.01 ⫾ 1.26 ESR 21.52 ⴞ 19.16 19.44 ⴞ 19.89 25.71 ⴞ 17.01 Hemoglobin 129.60 ⴞ 12.93 132.04 ⴞ 11.81 124.68 ⴞ 13.65 Glucocorticosteroids % 52.9% 48.04% 62.75% DMARDs % 82.2% 81.2% 86.75% Biologic therapy % 38.2% 36.63% 41.18% Erosions % 51.9% 51.09% 53.49% Vitamin D 19.78 ⴞ 10.75 21.69 ⴞ 10.64 15.88 ⴞ 9.98

p 0.523 0.982 0.892 0.554 0.595 0.009 0.000 0.053 0.029 0.003 0.057 0.003 0.000 0.002 0.003 0.027 0.986 0.173 0.056 0.001 0.086 0.355 0.586 0.795 0.003

Table 1. Joint effect of family history (FH) and GRS-HLA for RA phenotypes in the Nurses’ Health Study (NHS) and the Epidemiological Investigation of RA (EIRA) study. FH NHS All RA Seropositive RA Seronegative RA EIRA All RA ACPAⴙ RA

BMI: body mas index, RF: rheumatoid factor, TJC: tender joint count, SJC: swollen joint count, PGH: patient global health asessment, DAS: disease activity index, VAS: visual analogue scale, mHAQ: modified health asnessment quiestionnarie, CRP: C reactive protein, ESR: erythrocyte sedimentation rate, DMARDs: disease-modifying anti-rheumatid drugs.

ACPA-RA

Low GRS-HLA Cases/ controls OR (95% CI)

High GRS-HLA Cases/ controls OR (95% CI)

None Any None Any None Any

252/367 78/30 140/367 41/30 112/367 37/30

1.0 (Ref) 4.04 (2.54–6.56) 1.0 (Ref) 3.93 (2.28–6.86) 1.0 (Ref) 4.41 (2.55–7.73)

120/96 42/8 72/96 23/8 48/96 19/8

1.84 (1.33–2.56) 8.56 (3.87–18.93) 1.93 (1.31–2.84) 7.59 (3.17–18.16) 1.72 (1.12–2.61) 10.17 (4.16–24.88)


580/783 39/27 214/783 22/27 366/783 17/27

1.0 (Ref) 1.98 (1.18–3.32) 1.0 (Ref) 3.04 (1.66–5.55) 1.0 (Ref) 1.34 (0.71–2.54)

1013/528 120/23 699/528 93/23 314/528 27/23

2.28 (1.95–2.76) 6.17 (3.86–9.85) 4.32 (3.54–5.27) 13.20 (8.03–21.71) 1.16 (0.96–1.41) 9.50 (3.95–22.83)

All models are adjusted for age, smoking pack-years, body mass index, alcohol intake, education, and parity. EIRA models are also adjusted for sex and occupational exposures.

Conclusion: Latin-American mestizo patients had differing prognostic factors and more active RA than Spanish Caucasian patients, despite having the same access to health care and receiving similar antirheumatic therapy.

Table 2. Joint effect of family history (FH) and GRS-39 for RA phenotypes in the Nurses’ Health Study (NHS) and the Epidemiological Investigation of RA (EIRA) study.

Disclosure: V. Ruiz-Esquide, None; S. Cabrera, None; J. Ramirez, None; M. V. Hernańdez, None; J. Inciarte, None; J. D. Can˜ete, None; R. Sanmarti, None.

111

Low GRS-HLA Cases/ controls OR (95% CI)

High GRS-HLA Cases/ controls OR (95% CI)


240/349 72/27 129/349 35/27 111/349 37/27

1.0 (Ref) 4.26 (2.63–7.07) 1.0 (Ref) 4.09 (2.29–7.39) 1.0 (Ref) 4.77 (2.72–8.48)

132/114 48/11 83/114 29/11 49/114 19/11

1.65 (1.20–2.27) 6.63 (3.30–13.31) 1.94 (1.33–2.82) 6.73 (3.12–14.52) 1.33 (0.87–2.02) 6.39 (2.86–14.30)


826/984 56/36 370/984 35/36 456/984 21/36

1.0 (Ref) 1.70 (1.09–2.66) 1.0 (Ref) 2.43 (1.47–4.01) 1.0 (Ref) 1.17 (0.672.07)

767/327 103/14 543/327 80/14 224/327 23/14

2.43 (2.06–2.87) 8.24 (4.64–14.64) 3.86 (3.18–4.68) 14.20 (7.82–25.75) 1.32 (1.07–1.62) 3.65 (1.83–7.26)

FH NHS All RA

Identifying Groups At Increased Risk Of Developing Rheumatoid Arthritis Using Family History and Genetic Risk Scores. Jeffrey A. Sparks1, Chia-Yen Chen2, Xia Jiang3, Johan Askling4, Linda T. Hiraki5, Lars Klareskog6, Lars Alfredsson3, Karen H. Costenbader1 and Elizabeth W. Karlson1. 1 Brigham and Women’s Hospital, Harvard Medical School, Boston, MA, 2 Harvard School of Public Health, Boston, MA, 3Karolinska Institutet, Stockholm, Sweden, 4Karolinska University Hospital, Stockholm, Sweden, 5 Brigham and Women’s Hospital, Harvard School of Public Health, Boston, MA, 6Karolinska University Hospital, Karolinska Institutet, Stockholm, Sweden.

Seropositive RA Seronegative RA EIRA All RA ACPAⴙ RA ACPA-RA

Background/Purpose: The identification of high risk groups is crucial for RA prevention strategies. Individuals with family history (FH) of autoimmunity are at increased risk for RA. Genetic markers have been associated with RA and genetic risk scores (GRS) have incorporated these markers for RA risk. We aimed to identify groups at increased risk of RA using FH and GRS. Methods: We investigated the association between FH and GRS and risk of RA in a nested case-control study in the Nurses’ Health Study (NHS) and replicated in the Epidemiological Investigation of RA (EIRA). RA cases in NHS were validated by chart review and matched to controls. In EIRA, RA cases at diagnosis were matched to controls. All cases satisfied the 1987 ACR criteria for RA classification and were Caucasian. FH data were obtained from questionnaires (FH of RA or lupus for NHS)

All models are adjusted for age, smoking pack-years, body mass index, alcohol intake, education, and parity. EIRA models are also adjusted for sex and occupational exposures.

Conclusion: Using genetic risk scores and family history, we have identified groups at increased risk of developing RA. Genetic risk scores utilizing only HLA shared epitope alleles performed similarly to genetic risk scores incorporating a larger set of genetic markers for RA. Genetic testing, particularly HLA shared epitope alleles, among those with family history may identify groups at increased risk for RA. Disclosure: J. A. Sparks, None; C. Y. Chen, None; X. Jiang, None; J. Askling, Pfizer Inc, 2; L. T. Hiraki, None; L. Klareskog, No own commercial interests, 2; L. Alfredsson, None; K. H. Costenbader, None; E. W. Karlson, None.

S44

113

Sugar-Sweetened Soft Drink Consumption and Risk Of Developing Rheumatoid Arthritis In Women. Yang Hu1, Karen H. Costenbader2, Frank Hu3, Daniel H. Solomon4, Elizabeth W. Karlson2 and Bing LU5. 1 Brigham and Women’s Hospital, Boston, MA, 2Brigham and Women’s Hospital, Harvard Medical School, Boston, MA, 3Harvard School of Public Health, Boston, MA, 4Harvard Medical School, Brigham and Women’s Hospital, Division of Rheumatology, Division of Pharmacoepidemiology, Boston, MA, 5Brigham & Women’s Hospital and Harvard Medical School, Boston, MA.

Posttraumatic Stress Disorder and Risk For Incident Rheumatoid Arthritis. Yvonne C. Lee1, Susan Malspeis1, Jessica Agnew-Blais2, Katherine Keyes3, Laura Kubzansky2, Andrea Roberts2, Karestan Koenen3 and Elizabeth Karlson4. 1Brigham and Women’s Hospital, Boston, MA, 2Harvard School of Public Health, Boston, MA, 3Columbia University Mailman School of Public Health, New York, NY, 4Brigham and Woman’s Hospital, Boston, MA. Background/Purpose: Posttraumatic stress disorder (PTSD) is associated with autoimmune dysfunction, but the relationship between PTSD and the incidence of autoimmune disorders has not been studied prospectively. We examined the prospective association between PTSD and rheumatoid arthritis (RA) risk and tested markers of adverse health behaviors, cigarette smoking and body mass index (BMI), as possible mediators Methods: The Nurses’ Health Study II (NHSII) is a longitudinal cohort of 116,430 female nurses enrolled in 1989 at ages 25–42 years. A subset (N ⫽ 50,347) completed the Brief Trauma Questionnaire and the Lifetime PTSD screen, validated questionnaires that include date of worst trauma (proxy for date of PTSD onset). Participants were categorized into 5 groups based on trauma exposure and PTSD symptoms: 1) no trauma (referent), 2) trauma but no PTSD symptoms, 3) trauma and 1–3 PTSD symptoms, 4) trauma and 4–5 PTSD symptoms and 5) trauma and 6–7 PTSD symptoms. Incident RA (N ⫽ 185) after 1989 was confirmed by medical record review. Cox proportional hazards models were used to calculate hazard ratios (HRs) and 95% confidence intervals (CIs) for the association between PTSD and RA risk, adjusted for age, questionnaire year, race and parental education. We also examined the association between PTSD and seropositive and seronegative RA risk. To assess whether BMI and smoking mediated these associations, BMI and smoking measured after PTSD onset were added to the models. To assess whether BMI and smoking were confounders, we performed a subgroup analysis, including variables assessed before PTSD onset (BMI at age 18 and smoking before age 19) among women who experienced PTSD onset after age 20 Results: Compared to women unexposed to trauma, women exposed to trauma and reporting more than 4 PTSD symptoms were at higher risk for incident RA. RA risk increased with increasing number of PTSD symptoms (4–5 PTSD symptoms: HR 1.78, 95% CI 1.03–3.09; 6–7 PTSD symptoms: HR 1.91, 95% CI 1.03–3.56) (Table). Results were similar for the association between PTSD and seropositive RA risk (4–7 PTSD symptoms: HR 1.90, 95% CI 1.09–3.33). This trend was also noted in a subanalysis predicting seronegative RA, but the HR was lower and did not reach statistical significance (1.68, 95% CI 0.69–4.08). Both BMI and smoking status attenuated the association between PTSD and incident RA. In secondary analyses among patients who developed PTSD at age ⬎ 20, the HR for association with PTSD was lower (HR 1.57, 95% CI 0.56–4.43), but neither BMI at age 18 nor smoking before age 19 changed the HR (HR 1.58, 95% CI 0.56–4.45).

Background/Purpose: Sugar-sweetened soft drink (SSSD) consumption is associated with weight gain, obesity, and increased risk of type 2 diabetes (T2DM) and cardiovascular disease (CVD). Previous studies have found that obesity may increase risk of rheumatoid arthritis (RA), while T2DM and CVD may share common pathways with RA development. We aimed to examine the relationship between SSSD intake and incident seropositive RA or incident seronegative RA in the Nurses’ Health Study (NHS) and Nurses’ Health Study II (NHS II), two large prospective cohort studies. Methods: The NHS was established in 1976 and enrolled 121,701 US female registered nurses, ages 30–55 years. The NHS II was initiated in 1989 and comprised 116,430 female registered nurses, aged 25–42 years. RA cases were initially self-reported then confirmed by medical record review for 1987 ACR criteria. Seropositive RA was defined as positive RF or ACPA. SSSD consumption (colas and carbonated non-cola, but not low-calorie soft drinks) was assessed with a validated food frequency questionnaire (FFQ) completed every 4 years from 1980–2006 in NHS and 1991 – 2009 in NHS II. To reduce the measurement error and the potential reverse causality bias, we calculated average values from the preceding SSSD measures excluding the most recent dietary record. Time-varying Cox models were used to study the association between SSSD consumption and risk of seropositive and seronegative RA after adjusting for potential confounders. Pooled hazard ratio estimates from the two cohorts were calculated using a fixed effects model Results: During the 1,906,661 person-years of follow-up from 1980 to 2008 in NHS, 366 seropositive and 215 seronegative RA cases were confirmed, and during 1,524,433 person-years from 1991 to 2009 in NHS II, 197 seropositive and 105 seronegative RA cases were confirmed. In the pooled analysis after adjusting for age, smoking, alcohol, BMI, reproductive factors, physical activity, multivitamin use,, diet quality, and diabetes history, we found a significant positive association between SSSD and increased risk of seropositive RA (P for trend 0.005); women consuming ⱖ1 serving of SSSD per day had 1.71 (95% confidence interval 1.23– 2.38) fold increased risk of developing seropositive RA compared to those consuming none or less than 1 serving per month (Table). However, we did not find any significant association between SSSD and seronegative RA in both cohorts. Table. Hazard ratios for incident seropositive RA according to sugar-sweetened soft drink consumption in Nurses’ Health Study (NHS, 1980–2008) and Nurses’ Health Study II (NHS II, 1991–2009) 1 serving/ day

p for trend

116/539,330 1.17 (0.91,1.50) 1.25 (0.96,1.62)

31/97,461 1.82 (1.23,2.71) 1.96 (1.29,2.98)

0.015 0.006

48/365,248 1.04 (0.74,1.46) 1.10 (0.76,1.60)

19/103,555 1.55 (0.95,2.53) 1.34 (0.78,2.32)

0.258 0.351

1.12 (0.92,1.37) 1.20 (0.97,1.49)

1.71 (1.26,2.33) 1.71 (1.23,2.38)

0.009 0.005

Table. Reported PTSD in the NHSII cohort analysis through 2009 (N ⫽ 49,276, including 185 women who developed RA)

b

Variable

RA cases

Person Years

Age-adjusted PTSD* HR (95% CI)

Multivariate* HR (95% CI)

No trauma, no PTSD symptoms Trauma, no PTSD symptoms Trauma, 1–3 PTSD symptoms Trauma, 4–5 PTSD symptoms Trauma, 6–7 PTSD symptoms

36 92 23 20 14

239023 462310 107007 65928 42360

1.00 (REF) 1.26 (0.85, 1.85) 1.28 (0.75, 2.16) 1.78 (1.03, 3.08) 1.89 (1.02, 3.53)

1.00 (REF) 1.26 (0.86, 1.86) 1.27 (0.75, 2.15) 1.78 (1.03, 3.09) 1.91 (1.03, 3.56)

*P-value for trend: age-adjusted model: 0.0142, multivariate model: 0.0134 ** Cox proportional hazards models, adjusted for age, questionnaire-year, race (Caucasian, non-Caucasian) and parental education (2 high school, ⬎ high school)

a. Adjusted for age, median income (quartiles), cigarette smoking pack-years (never, ⬍20 pack years, ⱖ20 pack years), alcohol consumption (⬍5.0, 5.0–15.0, ⱖ15 grams/day), age at menarche (⬍12, 12, ⬎12 years), parity and breast feeding (nulliparous, parous/no breastfeeding, parous/1–12 months breastfeeding, parous/ ⬎12 months breastfeeding), hormone use (pre-menopausal, post-menopausal with never use, current use and past use), physical activity (0–3, 3–9, 9–18, 18–27, ⱖ27 METs/week), multi-vitamin use, healthy eating index (quartiles), body mass index (⬍20, 20–22.9, 23–24.9, 25–29.9, ⱖ30kg/m2), diabetes history and total energy (Kcal, quintiles). b. P-values for heterogeneity were non-significant (p⬎0.05) in all pooled estimates

Conclusion: These results suggest that PTSD may increase RA risk. Additional studies are needed to elucidate the pathways by which PTSD is associated with RA risk, specifically to assess whether this association is due to adverse health behaviors, increased reporting and healthcare utilization or alterations in immune pathways due to dysregulated stress response.

Conclusion: Data from two large prospective cohort studies suggest that regular consumption of SSSD is associated with increased risk of developing seropositive RA in women.

Disclosure: Y. C. Lee, Forest Research Insitutute, 2, Merck Pharmaceuticals, 1, Novartis Pharmaceutical Corporation, 1; S. Malspeis, None; J. Agnew-Blais, None; K. Keyes, None; L. Kubzansky, None; A. Roberts, None; K. Koenen, None; E. Karlson, None.

Disclosure: Y. Hu, None; K. H. Costenbader, None; F. Hu, None; D. H. Solomon, Lilly, Amgen, CORRONA, 2, Lilly, Novartis, BMS, Pfizer, 6, Lilly, BMS, Novartis, 9; E. W. Karlson, None; B. LU, None.

S45

Sunday, October 27

112

Sunday, October 27

114

115

Patients With Regular Physical Activity Before Onset Of Rheumatoid Arthritis Present With Milder Disease. Maria E.C. Sandberg1, Sara Wedren2, Christina H. Opava3, Lars Klareskog4, Lars Alfredsson1 and Saedis Saevarsdottir4. 1Karolinska Institutet, Stockholm, Sweden, 2Rheumatology Unit, Department of Medicine, Karolinska University Hospital, Karolinska Institute, Stockholm, Sweden, 3Karolinska Institutet, Huddinge, Sweden, 4Karolinska University Hospital, Karolinska Institutet, Stockholm, Sweden.

Association Between History Of Periodontitis and Risk Of Rheumatoid Arthritis In Individuals With and Without Diabetes Mellitus: A Population-Based Cohort Study. Hsin-Hua Chen1, D.Y. Chen2 and Nicole Huang1. 1National Yang-Ming University, Taipei, Taiwan, 2Taichung Veterans General Hospital, Taichung, Taiwan. Background/Purpose: To examine whether the association between periodontitis (PD) history and the rheumatoid arthritis (RA) risk differs between individuals with and without diabetes mellitus (DM) Methods: We conducted a population-based retrospective cohort study using the 1997–2009 National Health Insurance (NHI) claims data of one million representative individuals from all NHI enrollees. Adults with DM (aged ⱖ20 years) newly treated during 2001–2009 were classified as DM subjects and those without a DM diagnosis during 1997–2009 were classified as non-DM subjects. We identified 7097 DM subjects with PD history within one year before initiating anti-diabetes treatment (index date). By matching these 7097 subjects for age on the index date, sex, and year of the index date, we randomly extracted 14,194 DM subjects without PD history, 14,194 non-DM subjects with PD history, and 28,388 non-DM subjects without PD history. Adjusted hazard ratios (aHRs) with a 95% confidence interval (CI) were calculated by applying Cox proportional hazards models to quantify the association between PD history and RA risk. Results: Compared with subjects without PD history, the adjusted HR of RA among those with PD history was 2.26 (95%CI, 1.26–4.04). In subgroup analysis, this association was evident among DM subjects (aHR, 3.77; 95%CI, 1.48–9.60) but not among non-DM subjects (aHR, 1.56; 95%CI, 0.73–3.37; P for interaction 0.212). Conclusion: The effect of PD exposure on RA risk was evident in newly treated DM subjects but not in non-DM subjects; no significant modifying effect of DM was observed.

Background/Purpose: Physical activity is a crucial factor in human health; lack thereof is the third most important cause of preventable deaths today. Several biological mechanisms are affected by physical activity and decreasing markers of chronic inflammation have been shown in numerous studies. The aim of this study was to investigate whether the clinical presentation of rheumatoid arthritis (RA) might be affected by physical activity before disease onset. Methods: Among the cases from the large, population-based, casecontrol study EIRA, with clinical data from the Swedish Rheumatology Quality Register, we compared patients with regular physical activity 5 years before diagnosis (N⫽288), to less active patients (N⫽329), we also investigated a possible dose-response relationship with physical activity in four levels. Logistic regression was used to calculate the odds of having above median level of 28-joint disease activity score (DAS28), physician assessment (5 categories), pain (visual-analog scale, VAS-pain) and activity limitation (health assessment questionnaire, HAQ). The analyses were adjusted for potential confounders (sex, age at diagnosis, period of diagnosis, smoking, body mass index, alcohol consumption, socioeconomic status, vegetable intake and physically demanding work (5 years before diagnosis). Results: RA patients with regular physical activity before diagnosis had statistically significant decreased odds of having DAS28, physician assessment, and/or VAS-pain above median, while HAQ was not affected (see Table). We further found indications of a dose-response relationship for physical activity and DAS28, physician assessment and VAS-pain, for HAQ, however, only in the highest category of physical activity gave an effect (OR⫽ 0.49 [95% CI: 0.24 – 0.99]). Further; statistically significant effects were found both for the combined objective components (swollen joint count, erythrocyte sedimentation rate, C-reactive protein) and combined subjective components (tender joint count, patient global assessment) of DAS28. Neither anti-CCP positivity, body-mass index, sex, physically demanding work nor socioeconomic status modified the observed associations. Physical activity at leisure time

Events (N)

OR

95% CI

DAS28 ⱖ5.3

No regular physical activity Regular physical activity

189 122

1.00 0.56

Ref. 0.39–0–82

Physician assessment ⬎2


130 82

1.00 0.62

Ref. 0.42–0.92

VAS-pain ⬎50


181 127

1.00 0.67

Ref. 0.47–0.98

HAQ ⱖ1


194 142

1.00 0.84

Ref. 0.58–1.23

Disclosure: H. H. Chen, None; D. Y. Chen, None; N. Huang, None.

116 Lack Of Association Between Preclinical Markers For Cardiovascular Disease and Rheumatoid Arthritis-Related Autoimmunity In FirstDegree Relatives Without Rheumatoid Arthritis. Jill M. Norris1, Ryan W. Gan1, Jan M. Hughes-Austin2, Kevin D. Deane3, M. Kristen Demoruelle3, Elaine M. Urbina4, Kerrie Moreau3, Peter K. Gregersen5, Michael H. Weisman6 and V. Michael Holers3. 1Colorado School of Public Health, Aurora, CO, 2University of California, San Diego, La Jolla, CA, 3University of Colorado School of Medicine, Aurora, CO, 4Cincinnati Children’s Hospital Medical Center, Cincinnati, OH, 5Feinstein Institute for Medical Research, Manhasset, NY, 6Cedars-Sinai Medical Center, Los Angeles, CA. Background/Purpose: Rheumatoid arthritis (RA) is characterized by systemic inflammation and immune dysregulation, including the presence of autoantibodies and elevated inflammatory biomarkers in subjects with classifiable RA as well as prior to the development of clinically-apparent RA. In addition, the risk for cardiovascular disease (CVD) is greatly increased in patients with RA, with autoantibodies and systemic inflammation believed to be major contributors to the pathogenesis of CVD. Furthermore, the increased risk for CVD may precede the development of classifiable RA, leading to the hypothesis that autoantibodies and systemic inflammation are influencing the development of CVD in subjects even prior to the onset of joint symptoms in RA. The Studies of the Etiology of RA (SERA) demonstrated previously an association between autoantibody positivity and increased levels of circulating cytokines in subjects without RA but at elevated risk for future RA, as they are first-degree relatives (FDRs) of probands with RA. We utilized SERA FDRs to test the hypothesis that CVD may be apparent in subjects at-risk for future RA, and related to the presence of RA-related autoantibodies. Methods: Eighty-six autoantibody positive and negative FDRs from the larger SERA cohort were evaluated after a 10-hour fast for the following measures related to CVD: carotid intima media thickness (cIMT), carotid stiffness, flow-mediated dilation (FMD) of the brachial artery, abdominal adipose tissue using computed tomography (CT), blood pressure, lipids, lipoproteins and adipokines. Levels of these pre-clinical CVD phenotypes were log-transformed and compared by current autoantibody positivity status using mixed linear models to account for varying group sizes, adjusting for age, sex and ever smoked status.

Conclusion: The findings of this study implicate that RA patients who were physically active before disease onset present with a milder disease according to both patient-reported, physician-reported and laboratory measures. No previous studies have evaluated the influence of physical activity on the clinical presentation of RA. However, DAS28 has been reported to improve after physical activity interventions among established RA-patients, and some trials also reported improvements in VAS-pain and/or HAQ. The results of the present study adds to the growing evidence of the general health benefits of physical activity, and might be an important, helpful message for individuals at increased risk of RA. Disclosure: M. E. C. Sandberg, None; S. Wedren, None; C. H. Opava, None; L. Klareskog, No own commercial interests, 2; L. Alfredsson, None; S. Saevarsdottir, None.

S46

but the literature is somewhat contradictory. The use of postmenopausal hormone (PMH) has previously been observed to provide increased [1], decreased [2], as well as no association [3] with the risk of RA. The aim of the present work was to study the association between the use of PMH and the risk of developing RA in postmenopausal women (aged 50–70), by stratifying the cases according to presence/absence of antibodies to citrullinated peptides (ACPA). Methods: Data from the Swedish population-based EIRA (Epidemiological Investigation of RA) case-control study was analyzed. In total, 380 incident postmenopausal female cases, aged 50–70 years, and 780 randomly selected controls (matched by age and residential area) were included between 2006–2009. Of the cases, 239 (62.9%) were ACPA-positive. The use of PMH was assessed by means of an identical questionnaire answered by the participating cases and controls. Current and past users of PMH were compared with never users to obtain odds ratios (ORs) with 95% confidence intervals (CI) by means of unconditional logistic regression models. Results: A decreased risk of developing ACPA-positive RA was observed among current users compared with never users of PMH (OR 0.5, 95% CI 0.3–0.9). Among past users of PMH no association was found when comparing with never users (OR 1.2, 95% CI 0.8–1.7). For ACPA-negative RA no association was found among current (OR 1.0, 95% CI 0.6–1.8) or past (OR 1.2, 95% CI 0.8–2.0) users of PMH. Conclusion: Our results indicate that current use of PMH is associated with a decreased risk of ACPA-positive RA in postmenopausal women aged 50–70, but has no association with the risk of ACPA-negative RA. The striking difference in the results for ACPA-positive and ACPA-negative RA adds further evidence to the notion that RA consists of two different entities with partly different etiology. Further research is needed in order to explore the biological mechanisms behind our findings.

Table. CVD-related measures between CCP positive, RF positive and Ab Negative FDRs. Adjusted for age, gender, race, and smoking status; values are back-transformed. CCP Positive nⴝ19† Mean (SE) Age (years)* Sex (% Female) Ethnicity (% non-Hispanic white) Ever Smoker (% Yes) Blood Pressure Systolic (mm/Hg) Diastolic (mm/Hg) Lipids, lipoproteins and adipokines LDL (mg/dL) HDL (mg/dL) Cholesterol (mg/dL) Triglycerides (mg/dL) ApoB (mg/dL) ApoA (mg/dL) ApoB/ApoA Adiponectin (ug/mL) Leptin (ug/mL) ICAM (ug/mL) VCAM (ug/mL) E-Selectin (ug/mL) Adiposity BMI (kg/m2) Subcutaneous Fat Area (cm2) at L4/5* Visceral Fat (cm2) at L4/5* Carotid Intima Medial Thickness (averaged over left and right cIMT) cIMT, carotid bulb (max mm) cIMT, carotid bulb (avg mm) cIMT, common carotid artery (max mm) cIMT, common carotid artery (avg mm) cIMT, internal carotid artery (max mm) cIMT, internal carotid artery (avg mm) Measures of Carotid Stiffness Peterson’s Elastic Model (mmHg) Circumferential Arterial Strain (no units) Beta Stiffness Index (no units) Elastic Modulus, Incremental (mmHg) Young’s Elastic Pressure Modulus (mmHg/mm) Arterial Compliance (mm2/mmHg) Flow Mediated Dilatation (FMD) (⌬%)**

RF Positive nⴝ22 Mean (SE)

Ab Negative nⴝ45 Mean (SE)

P

51.42 (3.37) 84.2 84.2 21.1

49.86 (3.72) 77.3 95.5 27.3

53.44 (2.00) 60.0 88.9 42.2

0.67 0.11 0.52 0.20

118.6 (4.1) 72.0 (2.5)

115.9 (4.9) 69.9 (2.9)

118.1 (2.8) 73.6 (1.8)

0.85 0.38

93.2 (9.0) 51.4 (3.8) 167.0 (11.3) 113.1 (15.5) 78.7 (5.5) 151.0 (6.5) 0.53 (0.04) 8.9 (1.6) 15.8 (4.4) 91.8 (11.9) 605.6 (64.3) 30.2 (4.3)

82.2 (8.0) 50.9 (4.6) 161.5 (8.5) 105.4 (18.5) 71.6 (4.8) 150.8 (7.1) 0.48 (0.04) 9.1 (1.9) 13.6 (4.7) 80.5 (10.6) 599.4 (71.6) 36.9 (6.1)

91.2 (7.0) 47.0 (2.6) 161.5 (8.5) 107.2 (11.3) 79.4 (4.3) 141.5 (4.5) 0.58 (0.03) 8.9 (1.2) 16.9 (3.5) 106.2 (11.3) 591.8 (59.0) 35.7 (3.3)

0.33 0.34 0.63 0.87 0.17 0.17 0.02 0.99 0.72 0.06 0.98 0.35

28.8 (1.8) 362.3 (110.0) 155.2 (90.0)

28.4 (2.3) 324.5 (74.3) 98.9 (36.7)

29.9 (1.4) 379.2 (55.0) 136.9 (27.0)

0.68 0.70 0.50

1.0 (0.09) 0.8 (0.06) 0.7 (0.04) 0.6 (0.03) 0.6 (0.05) 0.5 (0.03)

1.1 (0.07) 0.8 (0.05) 0.7 (0.03) 0.6 (0.02) 0.7 (0.05) 0.5 (0.03)

0.08 0.15 0.85 0.69 0.16 0.19

0.9 (0.07) 0.7 (0.05) 0.7 (0.04) 0.6 (0.04) 0.6 (0.06) 0.5 (0.04) 413.4 (48.9) 0.1 (0.01) 4.3 (0.5) 1873.8 (286.3) 355.2 (49.3) 0.1 (0.01) 6.6 (1.7)

379.3 (53.5) 0.1 (0.02) 4.1 (0.5) 1688.2 (292.1) 327.0 (51.9) 0.1 (0.01) 6.7 (1.4)

423.4 (44.9) 0.1 (0.01) 4.4 (0.4) 1909.5 (270.5) 361.4 (48.2) 0.1 (0.01) 5.7 (0.9)

References: 1. Merlino LA, Cerhan JR, Criswell LA et al. (2003) Estrogen and other female reproductive risk factors are not strongly associated with the development of rheumatoid arthritis in elderly women. Semin Arthritis Rheum 33, 72–82. 2. Vandenbroucke JP, Witteman JC, Valkenburg HA et al. (1986) Noncontraceptive hormones and rheumatoid arthritis in perimenopausal and postmenopausal women. JAMA 255, 1299–1303. 3. Walitt B, Pettinger M, Weinstein A et al. (2008) Effects of postmenopausal hormone therapy on rheumatoid arthritis: the women’s health initiative randomized controlled trials. Arthritis Rheum 59, 302–310. Disclosure: C. Orellana, None; S. Saevarsdottir, None; L. Klareskog, None; L. Alfredsson, None; C. Bengtsson, None.

0.66 0.84 0.75 0.69 0.75

ACR Poster Session A Fibromyalgia, Soft Tissue Disorders and Pain I Sunday, October 27, 2013, 8:30 AM–4:00 PM

0.34 0.58

*Sample size reduced as not everyone participated in CT scan (CCP n⫽5, RF n⫽16, Ab- n⫽28). **Sample size reduced as not everyone was measured for FMD (CCP n⫽11, RF n⫽13, Ab- n⫽20). † 11 CCP positive cases were also positive for RF.

118

Conclusion: This analysis suggests that CCP and RF positive FDRs did not have worse biochemical, structural or functional indicators of vascular health compared to autoantibody negative FDRs, which is contrary to our a priori hypothesis. These results tentatively indicate that the increased risk for CVD that is seen in RA is not detectable in the early preclinical autoantibody phase of the disease, using the biochemical, structural or functional vascular changes measured herein. It is possible that the increased CVD risk seen in RA is manifested outside of these pathways and other mechanisms/measures should be explored.

The Controversies and Points Of Debate That Remain a Challenge For Fibromyalgia Care. Mary-Ann Fitzcharles1, Peter A. Ste-Marie1, Don L. Goldenberg2, John X. Pereira3, Susan Abbey4, Manon Choinie`re5, Gordon Ko6, Dwight E. Moulin7, Pantelis Panopalis1, Johanne Proulx8 and Yoram Shir1. 1McGill University Health Centre, Montreal, QC, 2Newton-Wellesley Hosp, Newton, MA, 3University of Calgary, Calgary, AB, 4University of Toronto, Toronto, ON, Canada, Toronto, ON, 5University of Montreal, Montreal, QC, 6University of Toronto, Toronto, ON, 7University of Western Ontario, London, ON, 8Patient Representative, Montreal, QC.

Disclosure: J. M. Norris, None; R. W. Gan, None; J. M. Hughes-Austin, None; K. D. Deane, None; M. K. Demoruelle, None; E. M. Urbina, None; K. Moreau, None; P. K. Gregersen, Janssen Pharmaceutica Product, L.P., 2; M. H. Weisman, None; V. M. Holers, None.

Background/Purpose: Fibromyalgia (FM) continues to present challenges for the health care community, with perceptions of disease and attitudes of physicians being highly polarized. Neurophysiologic studies provide reassurance of validity of FM, but clinical care remains dependant on the traditional “art of medicine”. In the process of formulating Canadian guidelines for care of FM patients, debate was generated on a number of highlighted clinical issues (1). This report examines these contentious issues in order to provide insight regarding challenges surrounding FM. Methods: In developing evidence based guidelines for the clinical care of FM, 11 of 60 recommendations did not achieve 80% approval at the first round of voting by a nationwide multidisciplinary health care panel. Modifications according to panel input were made and resubmitted for a second voting, at which time all passed, and form the basis of this report.

117 Postmenopausal Hormone Therapy and The Risk Of Rheumatoid Arthritis: Results From The Swedish EIRA Study. Cecilia Orellana1, Saedis Saevarsdottir1, Lars Klareskog2, Lars Alfredsson1 and Camilla Bengtsson1. 1Karolinska Institutet, Stockholm, Sweden, 2Rheumatology unit, Karolinska University Hospital, Karolinska Institute, Stockholm, Sweden. Background/Purpose: The importance of hormonal/reproductive factors has been hypothesized to contribute to the risk of Rheumatoid arthritis (RA),

S47

Sunday, October 27

Results: Nineteen FDRs were positive for anti-cyclic citrullinated peptide (either CCP2 or CCP3.1), 22 FDRs were positive for rheumatoid factor (RF) (either RF by nephelometry, or RF isotype-IgM, IgA, IgG), and 45 were negative for CCP and RF. The ApoB/ApoA ratio differed across autoantibody groups, with the Ab negative FDRs having the highest ratio, indicating increased CVD risk. Otherwise, there were no significant differences in CVD-related measures and current autoantibody status in FDRs (see Table).

Sunday, October 27

Results: Contention was seen in the following areas. 1. Criteria for diagnosis: The healthcare community remains fixated on application of diagnostic criteria, the absence of which causes uncertainty. The tender point count (TPC) remains entrenched in clinical practice. Agreement was obtained by recommending that the 2010 ACR criteria may be used to validate a clinical diagnosis, and TPC may be done by choice, but neither is required for diagnosis. 2. Treatments: Initial recommendation to discourage use of complementary and alternative medicine (CAM) treatments in absence of evidence for efficacy was modified to emphasize lack of evidence to support CAM use, thereby allowing for individual choice. Although antidepressants in all classes have shown some efficacy, individual bias favoured the newer serotonin norepinephrine reuptake inhibitors. Resolution was achieved by recommending that the specific choice be tailored to the patient according to physicians’ knowledge and evidence for efficacy. 3. Terminology: Recommendation to change the terminology of antidepressants and anticonvulsants to pain modulators was rejected in favour of promoting their pain modulating effects rather than a nomenclature change. 4. Past/triggering events: Causation in FM is contentious with experts recommending acknowledgement of previous negative lifetime events, but without excessive emphasis on its importance. 5. Work and disability: Recommendation to remain in the workforce was softened to state that continued work is ideal, with application of a rehabilitation program to improve function and possible return to work. Conclusion: The healthcare community must adopt a rational and unified approach to the management of FM to dispel false notions that hinder management. The abundance of anecdotal literature, strong advocacy from various groups and the subjective nature of FM symptoms may all have played a part in diagnosis and treatment uncertainty. These guidelines reflect the available evidence with clinically applicable input from health care workers from various disciplines and adhere to strict standards of development. Continued dialogue will help dispel misperceptions and facilitate optimal patient care.

Though it is not a spesific questionnaire for FM, it successfully puts forward the severity of the disease in a quicker and easier way. These preliminary results justify further investigation of the usage of this questionnaire in FM. Disclosure: E. Kaptanoglu, None; O. Sahin, None; Y. Durmaz, None; A. K. Cengiz, None; S. Hizmetli, None.

120 The Symptom Severity In Korean Patients With Fibromyalgia Is Associated With Socioeconomic Status, But Not With Obesity. DongJin Park1, Shin-Seok Lee2, Seong-Ho Kim3, Seong-Su Nah4, Ji Hyun Lee5, Seong-Kyu Kim6, Yeon-Ah Lee7, Seung-Jae Hong7, Hyun-Sook Kim8, Hye-Soon Lee9, Hyoun Ah Kim10, Chung-Il Joung11 and SangHyon Kim12. 1Chonnam National University Medical School, Gwangju, South Korea, 2Chonnam National University Medical School and Hospital, Gwangju, South Korea, 3Division of Rheumatology, Inje University College of Medicine, Haeundae Paik Hospital, Busan, South Korea, 4 Soonchunhyang University Cheonan Hospital, Cheonan, South Korea, 5 Maryknoll Medical Center, Busan, South Korea, 6Catholic University of Daegu School of Medicine, Daegu, South Korea, 7Kyung Hee University, Seoul, South Korea, 8Division of Rheumatology, Department of Internal Medicine, Soonchunhyang University of Korea, Seoul, Korea, Seoul, South Korea, 9Hanyang University Guri Hospital, Guri, South Korea, 10 Ajou University Hospital, Suwon, South Korea, 11Konyang university hospital, Daejeon, South Korea, 12Division of Rheumatology, Department of Internal Medicine, Keimyung University School of Medicine, Daegu, South Korea. Background/Purpose: Fibromyalgia (FM) is a chronic pain disorder characterized by widespread pain, tenderness, and various associated symptoms including sleep disturbances, chronic fatigue, depression, and other somatic symptoms. Several longitudinal and cross-sectional studies in western countries have shown that a high body mass index (BMI) is a strong and independent risk factor for future development of FM and is associated with higher levels of FM symptoms. In addition, obese patients have more physical and emotional impairments compared with nonobese patients. The purpose of this study was to determine whether obesity and socioeconomic factors influence symptom severity in Korean patients with FM. Methods: A total of 343 patients with FM were recruited from outpatient clinics at 11 medical centers across the Republic of Korea. All patients met the ACR 1990 classification criteria for FM. We interviewed these patients using a structured questionnaire that included sociodemographic data, current or past FM symptoms, and current use of relevant medications at the time of enrollment. Tender point counts and scores were assessed by thumb palpation. Patients were asked to complete a Korean version of the Fibromyalgia Impact Questionnaire (FIQ), the Brief Fatigue Inventory (BFI), the SF-36, the Beck Depression Inventory (BDI), the State-Trait Anxiety Inventory (STAI), the Self-Efficacy Scale, and the Social Support Scale. Results: Based on an obesity definition of a BMI of ⱖ25, 76 (22.1%) of the 343 patients were obese. Obese patients were not different from nonobese patients in terms of tender points and scores, FIQ, BFI, SF-36, BDI, STAI, Self-Efficacy, and Social Support scores. After age-, gender-, and symptom duration adjustment by propensity score matching, no significant differences were also found between obese and nonobese patients. However, socioeconomic status such as employment, insurance, and education were significantly associated with symptom severity of FM. The unemployed patients had higher FIQ scores (p ⫽ 0.011), higher BFI scores (p ⫽ 0.013), lower physical and mental component SF-36 scores (p ⫽ 0.012, p ⫽ 0.005), higher BDI scores (p ⫽ 0.005), and higher STAI II scores (p ⫽ 0.041). Lower-income patients had higher FIQ score (p⫽0.040), lower physical and mental SF-36 scores (p ⫽ 0.047, p ⫽ 0.006), higher BDI scores (p ⬍ 0.000), and lower Self-Efficacy scores (p ⫽ 0.016). Finally, patients with an education of ⬍12 years had higher tender points (p ⫽ 0.034), higher BDI scores (p ⫽ 0.007), and higher STAI II scores (p ⫽ 0.045). Conclusion: Our findings show that, contrary to Western patients, symptom severity in Korean patients with FM is associated with socioeconomic status, but not with obesity.

Disclosure: M. A. Fitzcharles, Purdue Pharma L.P., 5, Eli Lilly and Company, 5, Pfizer Inc, 5, Valeant, 5; P. A. Ste-Marie, None; D. L. Goldenberg, Pfizer Inc, 5; J. X. Pereira, None; S. Abbey, Lundbeck, 5, Pfizer Inc, 5, Eli Lilly and Company, 5, Forest Laboratories, 5, Lundbeck, 8, Pfizer Inc, 8; M. Choinie`re, Pfizer Inc, 5; G. Ko, Eli Lilly and Company, 5, Eli Lilly and Company, 8, Sanofi-Aventis Pharmaceutical, 8, Valeant, 8; D. E. Moulin, Pfizer Inc, 2, Pfizer Inc, 5, Eli Lilly and Company, 5, Johnson & Johnson, 5, Purdue Pharma L.P., 5, Merck Pharmaceuticals, 5; P. Panopalis, Amgen, 5, Bristol-Myers Squibb, 5, Abvie, 5; J. Proulx, None; Y. Shir, Purdue Pharma L.P., 8, Paladin Labs, 8, Paladin Labs, 5.

119 Routine Assessment Of Patient Index Data 3 In Fibromyalgia: A Rapid and Reliable Instrument For Evaluating Disease Severity? Ece Kaptanoglu, Ozlem Sahin, Yunus Durmaz, Ahmet Kivanc Cengiz and Sami Hizmetli. Cumhuriyet University, Sivas, Turkey. Background/Purpose: Routine Assessment of Patient Index Data 3 (RAPID 3) is a patient based questionnaire which was designed for evaluation of disease activity without formal joint counts for rheumatoid arthritis (RA) for busy clinical settings. Besides RA, it’s usage is being serached for some other diseases, too. We aimed to search the effectiveness of RAPID 3 in evaluating severity of fibromyalgia (FM) which is a generalized pain syndrome sharing many symptoms of RA like stiffness and fatigue besides many somatic symptom. Methods: 38 female volunteered patients with FM were involved in the study. All patients were requested to complete the RAPID 3, Fibromyalgia Impact Questionnaire (FIQ), Revised Fibromyalgia Impact Questionnaire (FIQR) and FM symptom severity scale. Results: Mean age of the patients were 46⫾10,9 years and the mean disease duration was 4,8⫾3.8 years. There were significant correlations between RAPID 3 and FIQR (r⫽0.598 p⫽0.000), symptom severity (r⫽0.573 p⫽0.000), FIQ total (r⫽0.742 p⫽0.000) and subgroups of FIQ (FIQ physical impairment r⫽0.434 p⫽0.007, feel good r⫽0.524 p⫽0.001, work missed r⫽0.577 p⫽0.000, do work r⫽0.634 p⫽0.000, pain r⫽0.733 p⫽0.000, fatigue r⫽0.045 p⫽0.005, stiffness r⫽0.492 p⫽0.002, anxiety r⫽0.378 p⫽0.019, depression r⫽0.398 p⫽0.013 except rested r⫽0.197 p⫽0.236. The correlation of FIQR and FIQ total was also good (r⫽0.763 p⫽0.000). The correlations of FIQ total and FIQR with symptom severity were r⫽0.626 p⫽0.000 and r⫽0.481 p⫽0.005 respectively. Conclusion: RAPID 3 has a good correlation with FIQ, FIQR and symptom severity scale which are gold standarts in follow up of FM.

Disclosure: D. J. Park, None; S. S. Lee, None; S. H. Kim, None; S. S. Nah, None; J. H. Lee, None; S. K. Kim, None; Y. A. Lee, None; S. J. Hong, None; H. S. Kim, None; H. S. Lee, None; H. A. Kim, None; C. I. Joung, None; S. H. Kim, None.

S48

Comparative Burden Of Chronic Widespread Pain and Fibromyalgia In The US Population. Caroline Schaefer1, Rachael Mann2, Elizabeth T. Masters3, Joseph C. Cappelleri4, Shoshana Daniel5, Gergana Zlateva3, Heather McElroy6, Arthi B. Chandran3, Edgar H. Adams1, Annlouise R. Assaf4, Michael McNett7, Philip Mease8, Stuart L. Silverman9 and Roland Staud10. 1Covance Market Access Services Inc., Gaithersburg, MD, 2Covance Market Access Services Inc., San Diego, CA, 3Pfizer Inc., New York, NY, 4Pfizer Inc., Groton, CT, 5Covance Market Access Services Inc., Conshohocken, PA, 6Covance Market Access Services, Sydney, Australia, 7 Aurora Pain Program, Milwaukee, WI, 8Swedish Medical Center, Seattle, WA, 9Cedars-Sinai Medical Center, UCLA Center of Excellence, Beverly Hills, CA, 10University of Florida, Gainesville, FL.

122 Rheumatologists Lack Confidence in Their Knowledge of Cannabinoid Molecules and Use in the Management of Rheumatic Disease Patients: Analysis of a Needs Assessment. Mary-Ann Fitzcharles1, Peter A. SteMarie1, Daniel J. Clauw2, Shahin Jamal3, Jacob Karsh4, Sharon Leclercq5, Jason J. McDougall6, Yoram Shir1, Kamran Shojania3 and Zach Walsh7. 1 McGill University Health Centre, Montreal, QC, 2University of Michigan, Ann Arbor, MI, 3University of British Columbia, Vancouver, BC, 4Ottawa Hosp Riverside, Ottawa, ON, 5University of Calgary, Calgary, AB, 6Dalhousie University, Halifax, NS, 7University of British Columbia, Kelowna, BC.

Background/Purpose: Few data exist on the comparative burden of chronic widespread pain (CWP) and fibromyalgia (FM) in the general population. Methods: 8,382 nationally representative participants (ⱖ18 years old) completed an online screener including sociodemographic questions and the London Fibromyalgia Epidemiology Study Screening Questionnaire (LFESSQ). Subjects who screened positive (bilateral pain, above/below waist, lasting ⱖ1 week in the past 3 months; CWP⫹), and a control group without CWP (CWP-) were invited for a site visit for physician evaluation of FM including a tender point exam. Of 1,331 CWP⫹ and 502 CWP- subjects who consented to be scheduled for a visit, mutually exclusive groups of CWP(n⫽125), CWP⫹ (n⫽176) and confirmed FM subjects based on physician evaluation (n⫽171) completed a visit and an online questionnaire to assess clinical characteristics and patient-reported outcomes (including Brief Pain Inventory-short form [BPI-SF], Medical Outcomes Study Sleep Scale [MOSSS], 12-Item Short Form Health Survey v2 [SF-12], EQ-5D-3L). Statistical significance was tested at the 0.05 level across the 3 groups using ANOVA for continuous variables and chi-square or Fisher’s exact test for categorical variables. Results: Age and race were similar among the 3 groups, but mean body mass index (kg/m2) increased from CWP- (28.8) to CWP⫹ (30.7) to FM (32.1) (P⫽.0044). Among those reporting comorbidities, the mean (SD) number increased from 2.4 (1.6) to 3.2 (2.0) and 4.9 (3.0) for CWP-, CWP⫹, and FM, respectively (P⬍.0001), with a corresponding greater prevalence of specific physical conditions (e.g. arthritis, headache/migraine, irritable bowel syndrome; all P⬍.003) and affective disorders (anxiety, depression; both P⬍.0001). Pain severity and interference with function progressively increased from CWP- to CWP⫹ to FM (P⬍.0001) (Table), including all BPI-SF interference subscales. Sleep quantity and quality significantly differed across the groups as indicated by the MOS-SS total score (Table) and subscales. The EQ-5D-3L and SF-12 showed progressive reduction in health status, and physical and mental health, respectively (Table); all SF-12 domains scores were lowest with FM followed by CWP⫹ and CWP(P⬍.0001). Relative to CWP-, higher proportions of CWP⫹ and FM groups were taking pain-related prescription medications, 32.8%, 52.8% and 62.6%, respectively (P⬍.0001); opioids were the single most frequently reported pain medication class.

Background/Purpose: The pharmacologic treatment of chronic rheumatic pain is often sub-optimal, leading patients to seek alternate treatments. Although herbal cannabis (marihuana) has had medicinal use for pain management for centuries, scientific study of cannabinoid effects is recent. With patient advocacy for access to cannabinoids, regulatory bodies worldwide are considering the merits of legalizing medical cannabis. As arthritis is cited as a common reason for medical cannabis use, rheumatologists should be better informed to advise patients. We have assessed rheumatologists’ self-reported confidence in their knowledge of cannabinoids and their perceived competence in providing prescriptions. Methods: Using a 19-question needs assessment survey, sent via email to the entire Canadian Rheumatology Association membership, we have examined rheumatologists’ confidence in 1) knowledge of cannabinoids in general, including phyto-, syntheto- and endocannabinoids, and 2) perceived competence and ability to advise patients regarding indications, use and precautions for cannabinoids in general, and herbal cannabis specifically. Results: 128 (25%) of all 510 members responded. Over three quarters were not confident in their knowledge of cannabinoid molecules, with two thirds reporting poor knowledge of the physiology of the endocannabinoid system. While 45 % of respondents stated no current role for any cannabinoid preparations for rheumatology patients, 70% believe this applies specifically to medical cannabis. Only 16 (13%) had ever previously recommended a trial of medical marihuana. Over 90% were not confident in writing a prescription for medical cannabis when required to indicate dosing, frequency and method of administration. When respondents were grouped as “Confident” in knowledge of cannabinoid molecules (n⫽33) vs. “Not-Confident” (n⫽95), the following were reported respectively: Current role for medical cannabis 48% vs. 23%; Previous prescription of pharmacological cannabinoid 33% v.12%; Previous recommendation for medical cannabis 27% vs.7%; No previous recommendation for either 39% vs. 81%; Would not recommend any cannabinoid in future 33% vs. 67%. Only 33% of knowledge confident respondents reported competence in prescribing medical cannabis. Concerns about risks of marihuana use were in line with current literature. Conclusion: The overwhelming majority of rheumatologists reported lack of confidence in their knowledge of cannabinoids, and uncertainty about their competence to prescribe cannabinoid treatments and herbal cannabis in particular. This survey highlights a major disconnect between patients’ advocacy, policy makers and physician need to provide competent patient care within the bounds of medical ethics and deontology. Guidance is required to inform rheumatologists on the prevailing evidence for the safe and effective use of cannabinoids in rheumatic conditions.

Table. Comparative Burden in Subjects by Chronic Widespread Pain (CWP) and Fibromyalgia (FM) Status Measure (nⴝ125) BPI-SFa Pain Severity Index Pain Interference Index MOS-SS total scorea SF-12b Physical Component Summary score Mental Component Summary score EQ-5D-3Lb Health state valuation Overall health status rating

CWPⴚ (nⴝ176)

Mean (standard deviation) CWPⴙ (nⴝ171)

FM

P-value

3.0 (2.1) 2.5 (2.5) 32.8 (18.7)

4.4 (2.1) 4.3 (2.6) 48.2 (18.5)

5.5 (2.0) 5.8 (2.4) 59.6 (17.5)

⬍.0001 ⬍.0001 ⬍.0001

48.1 (10.8) 49.1 (10.4)

38.9 (11.0) 44.9 (10.8)

33.7 (10.5) 40.4 (10.8)

⬍.0001 ⬍.0001

0.85 (0.13) 79.8 (16.0)

0.73 (0.18) 68.7 (18.8)

0.61 (0.21) 58.7 (20.0)

⬍.0001 ⬍.0001

Disclosure: M. A. Fitzcharles, Purdue Pharma L.P., 5, Eli Lilly and Company, 5, Pfizer Inc, 5, Valeant, 5; P. A. Ste-Marie, None; D. J. Clauw, Nuvo, Merck, Cerephex, Pfizer, 2, Lilly, Pfizer, Forest, Nuvo, Cerephex, Purdue, 5; S. Jamal, Abbott Immunology Pharmaceuticals, 5, Amgen, 5, Bristol-Myers Squibb, 5, Roche Pharmaceuticals, 5, Janssen Pharmaceutica Product, L.P., 5, Ucb, 5, Pfizer Inc, 5; J. Karsh, Roche Pharmaceuticals, 2, Bristol-Myers Squibb, 5, Amgen, 5, Roche Pharmaceuticals, 5, Abvie, 5, AstraZeneca, 5; S. Leclercq, Roche Pharmaceuticals, 5; J. J. McDougall, Eli Lilly Co. USA, 2; Y. Shir, Purdue Pharma L.P., 8, Paladin Labs, 8, Paladin Labs, 5; K. Shojania, None; Z. Walsh, None.

aHigher score⫽worse outcomes; blower score⫽worse outcomes

Conclusion: In a nationally representative sample, CWP⫹ and FM were characterized by a high disease burden relative to CWP- including more comorbidities and pain-related medications, poorer sleep, and reduced function and health status; the burden was highest with FM. Health status observed in FM was lower than has been reported for chronic conditions, such as cancer and diabetes (Luo et al. Health Outcomes Res Med.2011;e203–14).

S49

Sunday, October 27

Disclosure: C. Schaefer, Covance Market Access Services, 3; R. Mann, Covance Market Access Services, 3; E. T. Masters, Pfizer Inc., 3, Pfizer Inc., 1; J. C. Cappelleri, Pfizer Inc., 1, Pfizer Inc., 3; S. Daniel, Covance Market Access Services, 3; G. Zlateva, Pfizer Inc., 1, Pfizer Inc., 3; H. McElroy, Covance Market Access Services, 3; A. B. Chandran, Pfizer Inc., 1, Pfizer Inc., 3; E. H. Adams, Covance Market Access Services, 3; A. R. Assaf, Pfizer Inc., 1, Pfizer Inc., 3; M. McNett, Pfizer Inc., 5, Lilly, 8, Pfizer Inc., 8; P. Mease, Forest Laboratories, 2, Lilly, 2, Pfizer Inc, 2, Forest Laboratories, 5, Lilly, 5, Pfizer Inc., 5; S. L. Silverman, Amgen, Lilly, Medtronics and Pfizer/Wyeth, 2, Amgen, Lilly, Pfizer/Wyeth, 8, Amgen, Genetech, Lilly, Novartis, Pfizer/Wyeth, 5, Cedars-Sinai Medical Center, 3; R. Staud, Pfizer Inc., 2, Forest Laboratories, 2.

121

depression by Arthritis Impact Measurement Scale (AIMS). Between-group differences in discrete and continuous variables were assessed for statistical significance with the Chi-Square test and the independent-samples t-test, respectively. Linear regression was used to assess between-group differences in disease activity while adjusting for potential confounders. Results: The prevalence of disablement was 30.8%. Disabled patients were significantly older (49.1 vs. 45.9; P⫽0.020), more likely to smoke cigarettes (33.8% vs. 15.6%; P⫽0.006) or use marijuana (13.0% vs. 3.3%; P⫽0.020). No significant differences were observed in pain duration (10.7 years) and gender (female: 91.0%). Prior/current occupation type differed significantly between groups: disabled patients were more likely previously employed in manual professions or service industry, with employed patients occupied in education/clerical/health fields (P⫽0.001). Significant between-group differences were observed for management strategies: disabled patients used a greater count of medications (P⫽0.001), more opioids (P⫽0.001), antidepressants (P⫽0.032), tranquilizers (P⬍0.001), and cannabinoids (P⫽0.053), and participated less in exercise activity (P⫽0.009). Those disabled demonstrated more allodynia (P⫽0.027) and pain related behaviour (P⫽0.002). Except for depression and anxiety, all other parameters were significantly higher in the disabled group: pain VAS (P⬍0.001), PGA (P⬍0.001), FIQ (P⬍0.001), HAQ (P⬍0.001), MPQ (P⬍.0001), PCS (P⫽0.005) and PDI (P⬍0.001). All associations remained significant except for HAQ, MPQ, and PCS when adjusted for age and education. Conclusion: The results of this analysis suggest that a significant proportion of FM patients are unemployed due to disability. The subjective report of symptom severity for those disabled may be explained by true disease severity, negative impact of medications, or patient perception of illness and suffering. Alternately, justification for ongoing disablement may be the driver for augmentation of subjective illness report. As all measurements in FM are subjective, disabled patients may be an important confounder for understanding outcome in FM.

Sunday, October 27

123 Association Of Rheumatic Diseases With Symptom Severity, Quality Of Life, and Treatment Outcome In Patients With Fibromyalgia. Juan Jiao, John M. Davis III, Ann Vincent, Connie A. Luedtke, Stephen Cha and Terry H. Oh. Mayo Clinic, Rochester, MN. Background/Purpose: Fibromyalgia is often associated with various comorbid conditions. A high prevalence of fibromyalgia in patients with rheumatic diseases has been reported. Previous studies showed that patients with rheumatoid arthritis (RA) and fibromyalgia had worse RA symptoms, greater disease activity, and worse physical and mental quality of life (QOL) compared with patients who have RA without fibromyalgia. How a comorbidity of rheumatic disease affects symptom severity, QOL, and treatment outcome in patients with fibromyalgia is unknown. This study aims to evaluate the association of rheumatic diseases with symptom severity, QOL, and treatment outcome in patients with fibromyalgia. Methods: The initial study population consisted of 978 patients who were confirmed to have fibromyalgia and completed the brief multidisciplinary fibromyalgia treatment program (FTP), with emphasis on cognitive behavioral therapy (CBT). We based the present analysis on the 536 patients who returned survey questionnaires during the 6- to 12-month follow-up period (mean follow-up duration was 11.7 months; overall survey response rate, 54.8%). The Fibromyalgia Impact Questionnaire (FIQ) and the Short Form-36 Health Status Questionnaire (SF-36) were completed at initial evaluation and at 6- to 12-months follow-up period. Information about the presence or absence of inflammatory rheumatic disease (IRD) of each participant at initial evaluation in the FTP was collected retrospectively from the electronic medical record by an investigator. Presence of IRD was determined by physician diagnosis. Two-sample t test and multivariate regression analyses were performed to compare the rheumatic and nonrheumatic groups. Results: Thirty-six patients (6.7%) had documented IRDs, with undifferentiated inflammatory arthritis and rheumatoid arthritis being the common IRDs. At baseline, the rheumatic group had poorer scores in SF-36 physical functioning (P⫽.02), pain index (P⫽.01), and physical component summary (P⫽.009) than the nonrheumatic group. After the FTP, both groups tended to improve; however, the rheumatic group had significantly less improvement in the FIQ subscales in pain (P⫽.02) and work missed days (P⫽.02), and in the SF-36 physical functioning (P⫽.02) compared to the nonrheumatic group. Conclusion: Our findings suggest that IRD is a relatively common comorbidity among patients with fibromyalgia, with a prevalence of about 7% in the study population. Fibromyalgia patients with IRDs have worse SF-36–assessed physical health and pain but not for mental health at baseline compared to fibromyalgia patients without rheumatic diseases. In addition, the rheumatic group demonstrated lower treatment response to the FTP compared to the nonrheumatic group. Our study suggests that our brief FTP based on CBT is less efficacious in patients with fibromyalgia who have IRDs than in those who do not have IRDs. Further studies that seek to identify the relations and the pathophysiologic mechanisms linking fibromyalgia and IRDs will likely advance the understanding of these overlapping chronic conditions.

Disclosure: E. Rampakakis, None; M. A. Fitzcharles, Purdue Pharma L.P., 5, Eli Lilly and Company, 5, Pfizer Inc, 5, Valeant, 5; P. A. Ste-Marie, None; J. S. Sampalis, None; Y. Shir, Purdue Pharma L.P., 8, Paladin Labs, 8, Paladin Labs, 5.

125 Correlation Of Findings In Clinical and High Resolution Ultrasonography Examinations Of The Painful Shoulder. Raphael Micheroli1, Diego Kyburz2, Adrian Ciurea2, Beat Dubs3, Martin Toniolo2, Samuel Bisig4 and Giorgio Tamborrini5. 1University of Zurich, Zurich, Switzerland, 2University Hospital of Zurich, Zurich, Switzerland, 3Sonography Institute Bethanien Zurich, Zurich, Switzerland, 4Swiss Federal Institute of Technology Zurich, Zurich, Switzerland, 5Bethesda Hospital Basel, Basel, Switzerland. Background/Purpose: High resolution ultrasonography (HRUS) is a non-painful and non-invasive imaging technique which is increasingly used for evaluating patients with musculoskeletal disorders. In particular, HRUS is useful for an assessment of shoulder pain causes, as clinical examination often does not allow an exact diagnosis. The aim of this study was to establish the role of HRUS in the diagnosis of shoulder problems as well as to form an evidence base for clinical interpretation of most common clinical tests of the painful shoulder. Methods: Non-interventional observational study of 100 adult patients suffering from unilateral shoulder pain. Exclusion criteria were shoulder fractures, prior shoulder joint surgery and prior shoulder injections (local anesthetics or steroids) in the past month. The HRUS examination was performed according to the guidelines of the Swiss Society of Ultrasound in Medicine (SGUM, musculoskeletal section). The clinical tests were: bursitis sign, Jobe’s test, painful arc, drop arm sign, Hawkins and Kennedy impingement tests, lift off test, belly press test, m. infraspinatus test, acromioclavicular (AC) joint tenderness, Abbott-Saunders test, palm up test, Yergason test and Hueter sign. The physicians performing clinical and HRUS examination were blinded to each other. In the statistical analysis HRUS findings were used as the gold standard. Results: In the HRUS examination pathologies of the bursa subacromialis were found in 87 %, of the m. supraspinatus tendon in 67 %, of the AC joint in 24 %, of the long biceps tendon in 20 %, of the m. subscapularis tendon in 11 %, of the m. infraspinatus tendon in 10 % and of the m. pectoralis major tendon in 1 % of all cases. In order to detect pathology of the m. supraspinatus tendon, the Hawkins Kennedy impingement tests showed the highest sensitivity (0.86) whereas the Jobe’s test showed the highest specificity (0.55). To identify m. subscapularis tendon pathology the lift off test showed a

Disclosure: J. Jiao, None; J. M. Davis III, None; A. Vincent, None; C. A. Luedtke, None; S. Cha, None; T. H. Oh, None.

124 Disability In Fibromyalgia Is Associated With Greater Self-Reported Symptoms and Functional Impairment. Emmanouil Rampakakis1, MaryAnn Fitzcharles2, Peter A. Ste-Marie2, John S. Sampalis1 and Yoram Shir2. 1 JSS Medical Research, St-Laurent, QC, 2McGill University Health Centre, Montreal, QC. Background/Purpose: It is intuitive that disablement due to illness should be reflected in illness severity. When illness measurement is based on subjective report only, without objective validation, the reliability of symptom report is crucial and requires critical evaluation. Societal costs for fibromyalgia (FM) are high with disability rates up to 30% reported in the developed world. We have examined clinical characteristics of FM patients currently employed or receiving disability payments. Methods: Of 246 participants in a tertiary care cohort study of FM patients, 90 were employed, 77 receiving disability payments. Demographic and disease severity measures included: pain visual analog scale (VAS), patient global assessment (PGA), Fibromyalgia Impact Questionnaire (FIQ), Health Assessment Questionnaire (HAQ), McGill Pain Questionnaire (MPQ), Pain Disability Index (PDI), Pain Catastrophizing Scale (PCS), anxiety and

S50

sensitivity of 1, whereas the belly press test showed the higher specificity (0.72) as the lift off test (0.55). The m. infraspinatus test showed a high sensitivity (0.90) and specificity (0.74). All three AC tests (painful arc IIa, AC joint tendernessb, cross body actionc) showed low sensitivities (a0.25, b 0.38, c0.38) but high specificities (a0.96, b0.99, c0.96). To evaluate the long biceps tendon the palm up test showed the highest sensitivity (0.47), the Yergason test showed the highest specificity (0.88). Conclusion: Knowledge of sensitivity and specificity of various clinical tests is important for the interpretation of clinical examination test results and the identification of cases in which further imaging procedures are necessary to make a distinct diagnosis. Thus, clinical test results should be confirmed by HRUS examination, which allows a reliable differentiation of the various pathologies leading to a painful shoulder.

Table. Clinical characteristics used to define the clusters. Values are expressed as medians and interquartile ranges.a

BPI pain intensity Fatigue Sleep problems HADS Depression Illness burdenf Catastrophizing

Cluster 1 (N ⴝ 89) 0.0 (0.0–1.0) 3.0 (2.0–4.0) 20.0 (10.0–30.0) 27.2 (16.1–41.1) 3.0 (1.0–5.0) 1.0 (0.0–3.0) 6.0 (1.0–12.0)

Cluster 2 (N ⴝ 57)

Cluster 3 (N ⴝ 23) c

2.0 (0.0–4.0) 3.0 (2.0–5.0)c 70.0 (50.0–80.0)c 38.3 (27.2–46.7)c 4.0 (1.0–7.0)c 2.0 (1.0–3.0)c 12.0 (5.0–21.0)c

P-valueb d,e

12.0 (10.0–14.0) 3.0 (2.0–5.0) 60.0 (25.0–80.0)d 35.6 (17.5–47.8) 5.0 (2.0–8.0)d 1.0 (0.0–2.0)e 9.0 (3.0–18.0)

a BPI ⫽ Brief Pain Inventory, HADS ⫽ Hospital Anxiety and Depression b P-value that any one cluster is different from the others. c Cluster 2 is significantly different from cluster 1 at P ⱕ 0.05. d Cluster 3 is significantly different from cluster 1 at P ⱕ 0.05. e Clusters 3 is significantly different from cluster 2 at P ⱕ 0.05. f

⬍0.0001 0.03 ⬍0.0001 0.009 0.004 0.06 ⬍0.0001

Scale.

Illness burden was quantified using a count of patient-reported symptoms of headaches, migraines, poor concentration, poor memory and poor word-finding.

Disclosure: R. Micheroli, None; D. Kyburz, None; A. Ciurea, None; B. Dubs, None; M. Toniolo, None; S. Bisig, None; G. Tamborrini, None.

Conclusion: Although most patients had low levels of inflammation, pain and fatigue, 47.3% continued to report moderate to high pain and fatigue. These results indicate that: 1) chronic widespread pain syndromes are common among patients with established RA; 2) active inflammatory disease does not explain all problems with pain, fatigue and psychosocial problems; and 3) chronic widespread pain syndromes are associated with significantly diminished quality of life (assessed by depression, fatigue and sleep problems), even compared to patients with active inflammatory disease.

126 Subgrouping Of Rheumatoid Arthritis Patients Based On Pain, Fatigue, Inflammation and Psychosocial Factors. Yvonne C. Lee1, Michelle A. Frits1, Christine K. Iannaccone1, Michael E. Weinblatt1, Nancy A. Shadick1, David A. Williams2 and Jing Cui1. 1Brigham and Women’s Hospital, Boston, MA, 2Univ of MI Hlth System-Lobby M, Ann Arbor, MI.

Disclosure: Y. C. Lee, Forest Research Insitutute, 2, Merck Pharmaceuticals, 1, Novartis Pharmaceutical Corporation, 1, Cubist, 1, Elan Corporation, 1; M. A. Frits, None; C. K. Iannaccone, None; M. E. Weinblatt, MedImmune, 2, Crescendo Biosciences, 2, Bristol-Myers Squibb, 2, MedImmune, 5, Crescendo Biosciences, 5, Bristol-Myers Squibb, 5; N. A. Shadick, MedImmune, 2, Crescendo Biosciences, 2, Amgen, 2, Abbott Immunology Pharmaceuticals, 2, Genentech and Biogen IDEC Inc., 2; D. A. Williams, Pfizer Inc, 5, Forest Laboratories, 5, Eli Lilly and Company, 5, Healthy Focus, Inc., 5; J. Cui, None.

Background/Purpose: Among rheumatoid arthritis (RA) patients, pain may be due to peripheral inflammation or other causes, such as central pain mechanisms. The objective was to use self-report measures and physical examination to identify clusters of RA patients who may have diverse causes of pain and different prognoses and treatment options. Methods: Data were analyzed from 169 RA patients in a prospective observational study who had any pain ⬎ 0/10 and completed questionnaires about pain, fatigue, depression, anxiety, catastrophizing and sleep problems. Systemic inflammation was assessed using serum C-reactive protein (CRP) levels. A hierarchical agglomerative clustering procedure with Ward’s method was used to obtain subgroups. Multivariate analysis of variance was used to determine the contribution of each variable in a cluster. General linear regression models were used to examine differences in clinical characteristics across subgroups. Discriminant analyses were performed to determine coefficients for linear combinations of variables that assigned cluster membership to individual cases. Results: Cluster analyses identified three groups of RA patients (Figure). Cluster 1 consisted of the largest number of patients (N ⫽ 89, 52.7%) and was characterized by the lowest swollen joint counts and the lowest levels of fatigue and depression (Table). Cluster 2 consisted of 57 participants (33.7%), characterized by low swollen joint counts and high levels of fatigue, catastrophizing and sleep problems, indicative of a chronic widespread pain syndrome. Cluster 3 consisted of 23 participants (13.6%), characterized by high swollen joint counts and moderate levels of fatigue, catastrophizing and sleep problems, consistent with active inflammatory disease.

127 Clinical Application Of Pain Diagrams In Fibromyalgia. Amanda Steele1, Dana Dailey2 and Kathleen Sluka2. 1University of Iowa Hospitals and Clinics, Iowa City, IA, 2University of Iowa, Iowa City, IA. Background/Purpose: Fibromyalgia is a condition characterized by chronic widespread pain. Accurately assessing the subjective experience of pain using objective measures remains a challenging clinical problem, particularly among patients with Fibromyalgia. Commonly utilized selfreport pain measures are often time consuming (completing and scoring), potentially decreasing their clinical utility. The current study sought to establish a reliable scoring method of pain diagrams completed by patients with primary fibromyalgia. Methods: Prior to data collection, IRB approval was received and written consent was obtained. 43 people with fibromyalgia (42 female, 1 male), aged 25–76 years, completed this study (duration of fibromyalgia 7.55 years; range 0.25 to 20 years). Subjects filled out the Fibromyalgia Impact Questionnaire (FIQ), McGill Pain Questionnaire (MPQ) and a visual analog scale (0–10) for pain at rest and movement (walking). Patients were instructed to mark a body diagram showing areas of pain. To assess inter-rater reliability, a scoring protocol, which included applying a transparent overlay to each patient’s diagram, was designed and tested. A score of 0–46 was assigned to each diagram, based on the number of areas marked. Ten randomly selected diagrams were scored independently by two raters three times and inter-rater reliability assessed (intraclass correlation, ICC). The number of body areas was correlated with pain, FIQ, and MPQ scores (Pearson’s, *p⫽0.05, **p⫽0.01). Results: Means ⫾ SEM: Average pain at rest was 4.39 ⫾ 0.39, pain with movement was 5.23 ⫾ 0.39, FIQ was 58.17 ⫾ 2.24 and MPQ was 44.93 ⫾ 2.5. Scores on the body diagram (n⫽43) averaged 16.55 ⫾ 1.3 (95% CI 13.92 to 19.18) with the most common areas as neck (91%), shoulders (81%), and low back (63%). Inter-rater reliability for the body diagram scoring was excellent (ICC⫽0.952; 95% CI 0.912 to 0.974). Higher scores on the body diagram significantly correlated with higher scores on MPQ (R2⫽ 0.44**), pain at rest (R2⫽0.324*), pain with movement (R2⫽0.380*) but not FIQ (R2⫽0.298). Conclusion: Preliminary results indicate this protocol may be a reliable method of scoring pain diagrams among patients with Fibromyalgia and that higher scores significantly correlated with worse pain but not the impact of pain on function and quality of life (movement-pain,

S51

Sunday, October 27

Characteristic

Sunday, October 27

FIQ). These diagrams provide a quick, simple and clinically applicable self-report measure that can be completed and scored in a very brief period of time. Thus, quantitative analysis of the body diagram may be useful to gain an understanding of the severity of pain but not the global impact of pain as measured by the FIQ.

129 Identifying Predictors Of A Fibromyalgia Diagnosis: A Retrospective Electronic Medical Record Analysis. Elizabeth T. Masters1, Jack Mardekian1, Andrew Clair1 and Stuart L. Silverman2. 1Pfizer Inc, New York, NY, 2Cedars-Sinai Medical Center, UCLA Center of Excellence, Beverly Hills, CA.

Disclosure: A. Steele, None; D. Dailey, None; K. Sluka, None.

Background/Purpose: Fibromyalgia (FM) is a chronic pain condition characterized by a constellation of symptoms and comorbidities. Consequently, FM diagnosis can be challenging and it often goes undetected. This study used electronic medical record (EMR) data to identify factors associated with FM that may facilitate earlier identification and diagnosis. Methods: Subjects ⱖ18 years old who had ⱖ1 diagnosis code for common pain conditions were extracted from the Humedica de-identified EMR database, which has broad geographic representation and includes information on demographics, diagnoses, inpatient/outpatient encounters, medications, procedures, lab results, and select data from physicians’ notes. Records are linked using a unique patient identifier. Subjects with continuous enrollment in an integrated healthcare delivery system in 2010 and a first FM diagnosis in 2011 (cases) were compared with subjects without an FM diagnosis during 2011 (controls). Patients with an FM diagnosis prior to 2011 were excluded. FM diagnosis was based on ICD-9 code 729.1 (myalgia and myositis, unspecified). Sequential stepwise logistic regression was performed with FM diagnosis as the response variable, and demographic, clinical, and healthcare resource use as predictor variables. Variables with significant associations (Pⱕ.05) were retained in the model and expressed as odds ratios (OR) with their 95% confidence intervals (95% CI). Results: Subjects were 2,823 individuals with an FM diagnosis and 210,495 without an FM diagnosis in 2011. Although mean (SD) age was similar between groups, 51.4 (15.3) years for cases and 51.4 (16.4) years for controls, the FM population had more females (72.3% vs 60.4%; P⬍.0001), and significant differences between groups were observed for other baseline characteristics including race and healthcare resource use (higher in cases; P⬍.001), and the presence of specific comorbidities (higher in cases; Pⱕ.05). The model identified 17 variables significantly associated with an FM diagnosis. The first variable (i.e., the variable with the smallest p-value when included as a predictor by itself) that made it into the model was the number of pain medication prescriptions (OR 1.03 (95% CI 1.02, 1.04), followed by the number of musculoskeletal pain conditions (OR 1.19 (95% CI 1.16, 1.23). Several clinical variables, including the presence of gastrointestinal and sleep disorders were also predictive: OR 1.38 (95% CI 1.26, 1.51) and 1.33 (95% CI 1.15, 1.55), respectively. Other healthcare resource utilization variables that entered into the model included number of outpatient visits and hospitalizations. Conclusion: The model identified several demographic and clinical variables as significant predictors of an FM diagnosis. These results suggest analysis of EMR data can help identify variables associated with FM in a real world setting, and may inform earlier identification of FM patients.

128 Frequency Of Axial Spondyloarthropathy Among Patients Suffering From Fibromyalgia A Magnetic Resonance Immaging Study With Application Of The Assessment Of Spondylo-Arthritis International Society Classification Criteria. Jacob N. Ablin1, Iris Eshed2, Valerie Aloush1, Irena Wigler3, Dan Caspi1, Maria Likhter1, Mark Berman1, Yonatan Wolman1, Daphna Paran4, Marina Anouk5 and Ori Elkayam3. 1Tel Aviv Sourasky Medical Center, Tel Aviv, Israel, 2Sheba Medical Center, Tel Hashomer, Israel, 3Tel Aviv Sourasky Medical Center and the Sackler Faculty of Medicine, Tel Aviv University, Tel Aviv, Israel, 4Tel Aviv Sourasky Medical Ctr, Tel-Aviv university, Tel Aviv, Israel, 5Tel Aviv Sourasky Medical Ctr, Tel-Aviv University, Tel Aviv, Israel. Background/Purpose: Fibromyalgia Syndrome (FMS), considered the result of increased processing of pain by the central nervous system, is a non-inflammatory condition characterized by chronic, widespread musculoskeletal pain and tenderness. Axial spondyloarthritis (SpA) is the hallmark of Ankylosing Spondylitis (AS), an inflammatory joint disease primarily involving the sacroiliac joints and axial spine. Although FMS and axial SpA differ vastly in their pathogenesis, a considerable clinical overlap may exist between these conditions. Chronic nocturnal back pain and disturbed sleep may accompany either condition. We have previously described an increased prevalence of (secondary) FMS among female AS patients. The Assessment of Spondylo-Arthritis international Society (ASAS) has published updated classification criteria for axial SpA. These criteria are based on the evaluation of patients suffering from chronic back pain with an age of onset of less than 45. Objective: To evaluate the prevalence of axial SpA among FMS patients, utilizing the 2010 ASAS criteria (1). Methods: Patients suffering from FMS (ACR 1990 classification criteria) were recruited consecutively from a specialized FMS clinic. Patients were interviewed regarding the presence of SpA features, as defined by the ASAS group (Inflammatory back pain, arthritis, enthesitis, uveitis, dactylitis, psoriasis, Crohn’s / colitis, good response to NSAIDSs, family history of SpA) and underwent HLA-B27 testing and CRP measurement. MRI examinations of the sacro-iliac joints were performed on a 1.5 T MRI unit using semicoronal T1-weighted, STIR and FSPGR pre- and post-contrast injection sequences. FMS severity was assessed by the FIQ and SF-36 questionnaires and physical examination of the tender points using a dolorimeter. SpA symptom severity was assessed by the BASDAI questionnaire Results: 61unselected patients were recruited and MRI results were available for 56.

Disclosure: E. T. Masters, Pfizer Inc, 1, Pfizer Inc, 3; J. Mardekian, Pfizer Inc, 1, Pfizer Inc, 3; A. Clair, Pfizer Inc, 1, Pfizer Inc, 3; S. L. Silverman, Amgen, Lilly, Medtronics and Pfizer/Wyeth, 2, Amgen, Lilly, Pfizer/Wyeth, 8, Amgen, Genetech, Lilly, Novartis, Pfizer/Wyeth, 5, Cedars-Sinai Medical Center, 3.

Table 1. Summarizes MRI findings, HLA-B27 results and ASAS criteria positivity among FMS patients (N⫽56): Sacroiliitis (2) 4 (7.1%)

Erosions

Sub-chondral sclerosis

Fatty replacement

HLA-B27

ASAS criteria positive

11 (19.6%)

12 (21.4%)

4 (7.1%)

3 (5.4%)

4 (7.1%)

130 Mindfulness Is Associated With Psychological Symptoms, Self-Efficacy and Quality Of Life Among Patients With Fibromyalgia. Nani Morgan1, Lucas Morgan2, Lori Lyn Price3 and Chenchen Wang4. 1University of Hawaii, Honolulu, HI, 2University of Massachusetts, Honolulu, HI, 3Tufts University, Boston, MA, 4Tufts Medical Center, Boston, MA.

Sacroiliitis, based on ASAS definition (2), was found among 4 patients, 3 of which fulfilled ASAS SpA criteria. One additional patient fulfilled SpA criteria based on HLA-B27 positivity and additional SpA features. Conclusion: Imaging changes suggestive of inflammatory SpA were common among patients presenting with a clinical diagnosis of FMS. Definite sacroiliitis and ASAS criteria SpA positivity were diagnosed among 7.1% of patients and additional changes typical of SpA were frequent. These findings suggest that FMS may mask an underlying SpA, a diagnosis with important therapeutic implications. Thus, physicians involved in the management of FMS should remain vigilant to the possibility of underlying inflammatory disorders and actively search for such co-morbidities.

Background/Purpose: Mindfulness is the ability to observe, describe, or be aware of present moment experiences without judgment or reactivity. Preliminary evidence suggests that interventions aimed at increasing mindfulness may be effective in reducing chronic pain as well as symptoms of anxiety and depression commonly experienced among patients with fibromyalgia. Our objective was to evaluate whether mindfulness is associated with the overall impact of fibromyalgia on physical and psychological impairment, sleep quality, self-efficacy, and quality of life. Methods: We conducted a secondary analysis of baseline data from our randomized trial comparing Tai Chi and aerobic exercise among patients with fibromyalgia as defined by the American College of Rheumatology criteria. Patients enrolled in the trial completed the Five Facet Mindfulness Questionnaire (FFMQ), a 39-item, self-report questionnaire, in which higher total

References: 1. Sieper J et al. Ann Rheum Dis 2009;68:8(Suppl II):ii1–ii44 2. Rudwaleit M et al. Ann Rheum Dis 2009; 68(10):1520 Disclosure: J. N. Ablin, Pfizer Inc, 8; I. Eshed, None; V. Aloush, None; I. Wigler, None; D. Caspi, None; M. Likhter, None; M. Berman, None; Y. Wolman, None; D. Paran, None; M. Anouk, None; O. Elkayam, None.

S52

scores indicating higher levels of mindfulness in daily life. Patients also completed well-validated measures commonly used to assess patients with fibromyalgia (Revised Fibromyalgia Impact Questionnaire [FIQR], Perceived Stress Scale [PSS], Beck Depression Inventory Second Edition [BDI-II], Pittsburgh Sleep Quality Index [PSQI], Arthritis Self-Efficacy Scale Short Version [ASES-8] and Medical Outcomes Short Form-36 [SF-36]). We calculated Pearson’s correlation coefficients to evaluate hypothesized associations between mindfulness and measures of fibromyalgia impact on physical and psychological impairment, sleep quality, self-efficacy, and quality of life. Results: Our analysis included data from 81 patients with an average age of 52.04 (SD⫽11.92); 92% were female. All correlations were in the hypothesized direction. Patients reporting higher levels of mindfulness tended to report fewer symptoms of anxiety and depression as measured by the PSS (r⫽⫺0.55, p⬍0.0001) and BDI-II (r⫽⫺0.49, p⬍0.0001), respectively. They also tended to report higher self-efficacy as well as higher quality of life as measured by the ASES-8 (r⫽0.26, p⫽0.02) and SF-36 Mental Component Summary (r⫽0.51, p⬍0.001), respectively. There were no significant correlations between mindfulness and the FIQR, PSQI, or SF-36 Physical Component Summary. Conclusion: Our results suggest that mindfulness is associated with psychological symptoms, self-efficacy, and quality of life among patients with fibromyalgia. Mindfulness may help to change patients’ relationship to their symptoms and functional limitations through promotion of awareness and acceptance. Longitudinal studies are underway to evaluate whether changes in mindfulness are associated with changes in psychological symptoms, self-efficacy, and quality of life in patients with fibromyalgia and other chronic pain conditions.

Table 1. Multivariable baseline predictors of fibromyalgia Variable

Odds Ratio

P-Value

95% CI

2.92 3.57 3.09 0.97 0.98 1.05

0.005 0.002 0.088 0.052 0.019 0.042

1.37, 6.25 1.59, 8.04 0.85, 11.30 0.95, 1.00 0.96 1.00 1.00, 1.00

3.44 13.22

0.120 0.004

0.72, 16.39 2.24, 77.99

Disclosure: B. T. Walitt, None; F. Wolfe, None.

132 Relation Of Age With Symptom Severity and Quality Of Life In Patients With Fibromyalgia. Juan Jiao, Ann Vincent, Connie A. Luedtke, Stephen Cha and Terry H. Oh. Mayo Clinic, Rochester, MN. Background/Purpose: The relationship of age and fibromyalgia symptom severity and quality of life (QOL) is still debated. Reports range from no differences between patients younger and older than 60 years, to less symptom burden in older patients, and worse symptoms as well as QOL in older patients with fibromyalgia. The goals of this present study were to examine the relationship of age with symptom severity and QOL in patients with fibromyalgia and to compare QOL in physical and mental health of our female patients with the female general population. Methods: Nine hundred seventy eight patients with fibromyalgia who presented to a tertiary care fibromyalgia clinic were divided into 3 age groups: young (ⱕ 39 years), middle-aged (40–59 years) and older (ⱖ60 years). They completed the Fibromyalgia Impact Questionnaire (FIQ) and the Short Form-36 Health Status Questionnaire (SF-36) at the time of their evaluation. A standardization of the SF-36 physical and mental component summaries of our female patients was made in accordance with the normative data from the US female general population. One-way ANOVA and Post Hoc pairwise t-tests analyses were performed to detect differences across age-groups. Results: The mean age of 978 patients was 48.6 years (range, 19 to 87 years). Our patients age distribution in young, middle-aged and older were 233 (23.8%), 560 (57.3%), and 185 (18.9%), respectively. The young and middle-aged patients were more likely to be unmarried, employed, and current smokers, and to have a higher education level and more abuse history, a lower BMI, and a shorter duration of symptoms compared with the older patients. Pairwise comparison within 3 age-groups showed the young and middle-aged patients having worse fibromyalgia symptoms in the FIQ total score and all subscales except for the anxiety subscale when compared to the older patients (Psⱕ0.014). Similarly, those young and middle-aged patients had worse QOL in the SF-36 mental component summary, as well as SF-36 general health perceptions, vitality, social functioning, and mental health index compared to the older patients (Ps⬍0.001). When QOL of our female patients was compared to the U.S. female general population of similar age, all age groups had lower QOL in physical as well as mental health with more prominent reduction on physical health, particularly in the young patients. Conclusion: Our study illustrates that symptom severity and QOL differs across age-groups in patients with fibromyalgia. Young and middle-aged patients had poorer QOL and worse fibromyalgia symptoms compared to older patients. When the QOL of our female patients was compared with the US female general population of similar age, all age-groups had lower QOL in physical as well as mental health, particularly on physical health in the young patients. Further studies are needed to demonstrate the physiologic and pathological mechanisms underneath this phenomenon in patients with fibromyalgia according to age.

Disclosure: N. Morgan, None; L. Morgan, None; L. L. Price, None; C. Wang, None.

131 Prevalence and Antecedents Of Fibromyalgia In Elderly Women. Brian T. Walitt1 and Frederick Wolfe2. 1Washington Hospital Center, Washington, DC, 2National Data Bank for Rheumatic Diseases, Wichita, KS. Background/Purpose: Although epidemiological studies demonstrate that fibromyalgia (FM) prevalence increases with age, there have been no large-scale studies of FM in the elderly. In the clinic, elderly subjects usually receive treatment for common problems like diabetes, hypertension, back pain syndromes and other chronic illnesses; and the presence of FM is rarely recognized. In 2013 we performed a pilot study to estimate FM prevalence in the elderly by studying 882 women participants of the Women’s Health Initiative (WHI). Because the WHI study is longitudinal, we used prior data to investigate factors that predicted subsequent FM development. Methods: In 2013, we administered the 2010 modified American College of Rheumatology (ACR) survey criteria questionnaire to 882 randomly selected WHI participants. We utilized a wide range of questionnaire data that had been collected at WHI enrollment 16 years previously, including a range of questions from the SF-36, CESD depression scale, symptoms scales, and other questionnaires. We used logistic regression and baseline variables to predict current FM. Results: In 2013, the age of participants was 78.4 (SD 6.0) and they had been followed by the WHI for 16.4 years. The (2013) prevalence of fibromyalgia was 6.6% (95% CI 4.9–8.2). The average questionnaire scores in the population were: widespread pain index (WPI, 2.8 (2.8)), the symptom severity scale (SS, 2.9 (2.3), and the polysymptomatic distress score (PSD - sum of WPI and SS) 5.7 (4.3). To identify predictors of FM at initiation, we first excluded possible FM cases (15%) at study onset from further analysis if they reported moderate or severe body pain during the past 4 weeks. Of the remaining 746 subjects, the 2013 FM prevalence was 4.7% (3.1–6.2). In univariate analyses the following variables predicted FM 16 years later: social function, physical function, body pain, pain interference, general health, energy/fatigue, sadness in last two weeks, upset stomach, depression, headaches, low back pain, neck pain, diarrhea, aches and pain, constipation, and tiredness. In multivariate analyses, the following variables were included after backward stepwise regression filtering: constipation, pain severity, general health, energy/ fatigue, neck pain, pain interference, and aches and pains. The area under the ROC for the final logistic model was 0.85. Conclusion: Fibromyalgia (by criteria) was common in WHI participants as they approached 80 years of age. FM appeared to develop primarily in those baseline pain, aching, fatigue, impaired general health, and GI symptoms. Psychological factors did not seem to be important predictors.

Disclosure: J. Jiao, None; A. Vincent, None; C. A. Luedtke, None; S. Cha, None; T. H. Oh, None.

133 Cross-Sectional Neurocognitive Data Do Not Support a Transition From Fibrofog To Alzheimer’s Disease In Fibromyalgia Patients. Robert S. Katz1 and Frank Leavitt2. 1Rush Medical College, Chicago, IL, 2Rush University Medical Center, Chicago, IL. Background/Purpose: The issue of progressive decline in cognitive abilities is a source of concern to many patients with fibromyalgia. The fear

S53

Sunday, October 27

Constipation (Y/N) Neck pain (Y/N) Pain interference (Y/N) SF-36 General Health SF-36 Energy/fatigue SF-36 pain Aches and pains Aches and Pains-Mild Aches and Pains-Moderate or ⬎

Sunday, October 27

of progressing from fibrofog in middle age to Alzheimer’s or other forms of dementia is central to the worries shared by many. To determine if cognitive impairment progresses over time in fibromyalgia, a cross sectional study of cognitive functioning was conducted with a battery of neuro-cognitive tests known to be sensitive to cognitive change. Methods: In a cross-sectional study, cognitive disparities on 14 neurocognitive measures were evaluated in two cohorts of fibromyalgia subjects who met the new criteria for fibromyalgia. The first cohort contained 69 subjects with a short duration of cognitive problems (12 months or less). The second cohort contained 39 subjects with a long duration of cognitive problems (ⱖ 84 months or greater). Results: The two groups were similar in terms of mean (⫾SE) education (15.0⫾2.2 vs. 14.9⫾2.5), Vocabulary Scale score (11.2 ⫾ 2.3 vs. 11.9 ⫾ 2.8), and depression (17.5⫾9.4 vs. 17.8⫾9.3). The Long Duration Cohort was significantly older (45.6 ⫾ 10.8 vs. 52.3⫾ 9.1 years; p⬍0.001). Cognitive data of the two cohorts are displayed in Table 1. No significant differences were found between the two duration derived cohorts on 13 of the 14 cognitive measures. Performance declined with the duration of cognitive problems on Trails A, a measure of spatial scanning and cognitive sequencing. However, the skills of Trails A are incorporated inTrails B, where performance of the two groups was essentially equivalent. Measures of episodic memory and processing speed, markers of preclinical Alzheimer’s disease, are in the normal range for both cohorts Conclusion: Fibromyalgia patients’ fear of developing Alzheimer’s or other forms of dementia was not borne out by the data. Duration of cognitive problems was not a determiner of cognitive severity. People with a duration of 84 months or greater were not significantly more cognitively disabled than those with a duration of 12 months or less. The data do not support later transitions to dementia. The fact that individuals suffering cognitive dysfunction from 7 to 26 years did not exhibit a broad based deficit relative to those suffering cognitive dysfunction for a duration of a year of less should allay the worries of many with fibromyalgia who fear that fibrofog in the middle years is the start of a dementing process.

abuse (P⫽0.016), current unemployment (P⬍0.001) and higher score in the following measures of symptom severity: PGA (P⫽0.019), FIQ (P⫽0.002), HAQ (P⫽0.001), MPQ (P⫽0.026), PDI (P⫽0.031), and PCS (P⫽0.015). These associations remained significant even upon adjusting for age and gender differences. No significant differences in pain severity, anxiety, and depression were observed between groups. Conclusion: Similar to other health conditions, FM patients with lower SES reported greater symptom severity, functional impairment and unemployment, but not mood disorder. Although FM spans all socioeconomic groups, societal factors, rather than specific disease characteristics or mental status, appear to play an important role in patients’ perception of illness. Disclosure: M. A. Fitzcharles, Purdue Pharma L.P., 5, Eli Lilly and Company, 5, Pfizer Inc, 5, Valeant, 5; E. Rampakakis, None; P. A. Ste-Marie, None; J. S. Sampalis, None; Y. Shir, Purdue Pharma L.P., 8, Paladin Labs, 8, Paladin Labs, 5.

135 Evaluation Of The Quality Of Life and Sleep In Patients With Benign Joint Hypermobility Syndrome. Ilknur Albayrak1, Halim Yilmaz1, Ekrem Akkurt1, Ali Salli2 and Gulten Karaca3. 1MD, Konya, Turkey, 2MD, KONYA, Turkey, 3Associate Professor, Konya, Turkey. Background/Purpose: Traction injuries at tendon or ligament attachment sites, joint instability, low back, back and neck pain can be seen in patients with benign joint hypermobility syndrome (BJHS). The most important complaint of such patients is pain. Due to pain, impairment in the quality of sleep and life and fatigue resulting in effects on psychological condition. In this study, we aimed to evaluate pain, fatigue, quality of sleep, depression level and quality of life in patients with BJHS and compare these parameters with those of healthy controls. Methods: 26 patients who had diagnosed with BJHS according to Brighton diagnostic criteria and 40 healthy controls were included in the study. In all patients and healthy controls, pain severity evaluation was performed using visual analog scale (VAS), fatigue severity was performed using Checklist Individual Strength Scale (CIS), quality of sleep was performed using Pittsburg Sleep Quality Index (PSQI), depression level was evaluated using Beck Depression Scale (BDS), and quality of life was evaluated using Short Form-36 (SF-36). Results: Mean age was 31.4⫾7.6 years in the patient group, and 31.3⫾6.9 years in the control group. There were no statistically significant differences between the two groups with respect to age, gender, body mass index and marital status (p⬎0.05). In the patient group, VAS score during movement was 6.6⫾0.9. Total CIS scores were 4.3⫾1.09 in the patient group and 3.2⫾1.2 in the control group (p⫽0.001), global PSQI scores were 7.8⫾3.1 in the patient group and 4.9⫾2.9 in the control group (p⬍0.001), BDS scores were 17.6⫾8.8 in the patient group and 6.8⫾5.8 in the control group (p⬍0.001), SF-36 physical component scale scores were 52.7⫾14.6 in the patient group and 56.09⫾9.07 in the control group (p⬎0.05), and SF-36 mental component scale scores were 42.01⫾15.04 in the patient group and 54.3⫾10 in the control group (p⬍0.001). Conclusion: This study shows that the levels of pain, fatigue and depression are increased and the quality of sleep and quality of life are adversely affected in patients with BJHS. Therefore, patients with BJHS should be evaluated in terms of pain, fatigue, depression, quality of sleep and quality of life during their diagnostic procedures and follow-up, and multifaceted treatment programs should be planned.

Disclosure: R. S. Katz, None; F. Leavitt, None.

134 Lower Socioeconomic Status Associates With Increased Symptom Severity and Functional Impairment In Fibromyalgia. Mary-Ann Fitzcharles1, Emmanouil Rampakakis2, Peter A. Ste-Marie1, John S. Sampalis2 and Yoram Shir1. 1McGill University Health Centre, Montreal, QC, 2JSS Medical Research, St-Laurent, QC. Background/Purpose: Persons with lower socioeconomic status (SES) have poorer health status for many medical conditions. Reasons for this finding are multiple, but access to care, health related behaviours and adherence likely play a role. As fibromyalgia (FM) is a clinical construct with psychosocial implications, it is possible that SES may influence symptom expression and severity. We have examined the effects of SES (measured as proxy by level of education) for disease severity in a cohort of FM patients. Methods: In a prospective cohort study of patients with FM followed at a tertiary care multidisciplinary clinic, patients were stratified according to education level: high school or less (Group 1; N⫽99), college (Group 2; N⫽84), and university (Group 3; N⫽63). Demographic and disease severity measures included pain visual analog scale (VAS), patient global assessment disease activity (PGA), Fibromyalgia Impact Questionnaire (FIQ), Health Assessment Questionnaire (HAQ), McGill Pain Questionnaire (MPQ), Pain Disability Index (PDI), Pain Catastrophizing Scale (PCS), and anxiety and depression by Arthritis Impact Measurement Scale (AIMS). Between-group differences in discrete and continuous variables were assessed for statistical significance with the Chi-Square test and one-way analysis of variance, respectively. Linear regression was used to assess between-group differences in disease activity while adjusting for potential confounders. Results: The cohort comprised 246 patients with a mean ⫾ SD age of 47.8 ⫾ 10.4 years, disease duration of 10.8 ⫾ 9.8 years, and 91.1% female. Baseline values were: pain VAS 6.5 ⫾ 2.3, PGA 6.7 ⫾ 2.1, FIQ 67 ⫾ 17, HAQ 1.19 ⫾ .59, MPQ 41 ⫾ 15, PDI 38 ⫾ 14, PCS 29 ⫾ 12, AIMS anxiety 6.3 ⫾ 1.8, AIMS depression 4.9 ⫾ 1.8, with a mean medication count of 2.6 ⫾ 1.3 per patient. There were no significant differences between groups for the following parameters: disease duration, marital status, cigarette smoking, previous eating disorder or alcohol abuse, current medication categories, and total number of medications used per patient. Higher education was associated with greater use of alternative medicines (P⬍0.001) and chiropractic, massage or osteopathic treatments (P⫽0.021). Lower education level was significantly associated with older age (P⫽0.039), previous drug

Disclosure: I. Albayrak, None; H. Yilmaz, None; E. Akkurt, None; A. Salli, None; G. Karaca, None.

136 The Detection Of Primary Hyperparathyroidism In Patients Diagnosed With Fibromyalgia. Michael Tsoukas, Peter A. Ste-Marie, Yoram Shir, Mary-Ann Fitzcharles and Elliot Mitmaker. McGill University Health Centre, Montreal, QC. Background/Purpose: Some illnesses, such as hypothyroidism or small fiber neuropathy, may present similarly to fibromyalgia (FM), or may in turn lead to an augmented polysymptomatic distress, mimicking FM, by concomitant illness. Primary hyperparathyroidism (HPT), diagnosed biochemically by the measurement of parathyroid hormone in the context of serum calcium and phosphate levels, has a reported general population prevalence in the order of 0.1%. Its symptoms of musculoskeletal pain, fatigue, mood disorders, and sleep disturbances closely mimic those of FM. Given the commonality of

S54

estimates. Estimated prevalence (95% CI) of FM was 9.5% (7.2, 11.7) after adjusting for CWP status, gender, age, and race in the screened sample and weighted by 2010 census results for gender to the US population. Genderspecific prevalence estimates were 12.4% (11.0, 13.8) and 6.8% (4.1, 9.5) among adult women and men, respectively. The majority (55.0%) of subjects with physician-confirmed FM were obese per WHO criteria (BMI ⱖ30 kg/m2). Subjects with physician-confirmed FM had a mean (SD) of 14.4 (3.3) tender points, and 90.1% had ⱖ11 tender points. Conclusion: Based on the results of this nationally representative sample, the FM prevalence among US adults was higher than previously estimated by regional prevalence studies. The higher prevalence estimate may be a function of changing demographic and clinical characteristics of the US population, increased awareness of FM among patients and providers, study design and factors that influenced study participation. Clinical characteristics of subjects with FM in this study are consistent with the FM clinical profile in the literature. Disclosure: E. H. Adams, Covance Market Access Services, 3; S. Daniel, Covance Market Access Services, 3; A. B. Chandran, Pfizer Inc., 1, Pfizer Inc., 3; H. McElroy, Covance Market Access Services, 3; C. Schaefer, Covance Market Access Services, 3; R. Mann, Covance Market Access Services, 3; G. Zlateva, Pfizer Inc., 1, Pfizer Inc., 3; J. C. Cappelleri, Pfizer Inc., 1, Pfizer Inc., 3; E. T. Masters, Pfizer Inc, 1, Pfizer Inc, 3; A. R. Assaf, Pfizer Inc., 1, Pfizer Inc., 3; M. McNett, Pfizer Inc., 5, Lilly, 8, Pfizer Inc., 8; S. L. Silverman, Amgen, Lilly, Medtronics and Pfizer/Wyeth, 2, Amgen, Lilly, Pfizer/Wyeth, 8, Amgen, Genetech, Lilly, Novartis, Pfizer/Wyeth, 5, Cedars-Sinai Medical Center, 3; R. Staud, Pfizer Inc., 2, Forest Laboratories, 2; P. Mease, Forest Laboratories, 2, Lilly, 2, Pfizer Inc, 2, Forest Laboratories, 5, Lilly, 5, Pfizer Inc., 5.

Disclosure: M. Tsoukas, None; P. A. Ste-Marie, None; Y. Shir, Purdue Pharma L.P., 8, Paladin Labs, 8, Paladin Labs, 5; M. A. Fitzcharles, Purdue Pharma L.P., 5, Eli Lilly and Company, 5, Pfizer Inc, 5, Valeant, 5; E. Mitmaker, None.

138

137

Long Term Outcomes In Fibromyalgia. Lesley M. Arnold and Thomas Blom. University of Cincinnati College of Medicine, Cincinnati, OH.

A Population-Based Survey and Physician Assessment Of The Characteristics and Prevalence Of Fibromyalgia. Edgar H. Adams1, Shoshana Daniel2, Arthi B. Chandran3, Heather McElroy4, Caroline Schaefer1, Rachael Mann5, Gergana Zlateva3, Joseph C. Cappelleri6, Elizabeth T. Masters7, Annlouise R. Assaf6, Michael McNett8, Stuart L. Silverman9, Roland Staud10 and Philip Mease11. 1Covance Market Access Services Inc., Gaithersburg, MD, 2Covance Market Access Services Inc., Conshohocken, PA, 3Pfizer Inc., New York, NY, 4Covance Market Access Services, Sydney, Australia, 5 Covance Market Access Services Inc., San Diego, CA, 6Pfizer Inc., Groton, CT, 7Pfizer Inc, New York, NY, 8Aurora Pain Program, Milwaukee, WI, 9 Cedars-Sinai Medical Center, UCLA Center of Excellence, Beverly Hills, CA, 10University of Florida, Gainesville, FL, 11Swedish Medical Center, Seattle, WA.

Background/Purpose: There are few studies on the long-term prognosis of fibromyalgia. This study assessed the natural history of fibromyalgia in patients receiving usual medical care and identified predictors of outcome. Methods: Over about 3 years, patients from the Fibromyalgia Treatment Program completed questionnaires for demographics, pain, mood, fatigue, sleep, health-related quality of life, coping mechanisms, stressful life events, and global assessments. Clinicians evaluated pain thresholds, psychiatric comorbidity, medical history, and global assessments. Patient Global Impression of Improvement (PGI) at endpoint was used to define two groups: 1) definitely improved (much or very much better); and 2) not markedly improved (minimally improved, no change, or worse). Spearman correlation coefficients were calculated for baseline variables versus PGI endpoint scores. A general linear model was used to examine the relationship between baseline variables and PGI score at endpoint. Results: Sixty-nine patients completed the baseline assessment and had at least 1 follow-up visit. About half (n⫽34) of the patients definitely improved. There were no significant differences in any demographic variable between the two groups. Among those who reported global improvement, there was significant improvement in fatigue. Having baseline maladaptive coping (catastrophizing), lower pain thresholds, more negative fibromyalgia impact, and impairment in quality of life were correlated with poorer global assessment at endpoint. In the regression analysis, only baseline catastrophizing remained a significant predictor of endpoint PGI. Conclusion: Fibromyalgia patients who improved were less likely to use catastrophizing at baseline. Interventions directed at improving pain coping skills should be a major component of fibromyalgia management.

Background/Purpose: No large-scale, national study estimating fibromyalgia (FM) prevalence in the United States (US) has been conducted involving physician evaluation. The study objective was to provide contemporary data on characteristics and prevalence of FM among a subset of a nationally representative sample of US adults who completed an online subject screener and physician evaluation. Methods: This multi-stage study included an online subject screener, a site visit for physician evaluation of FM, and an online subject questionnaire. Subjects were required to be at least 18 years old. Nationally representative participants were invited to complete the online screener, which included the London Fibromyalgia Epidemiology Study Screening Questionnaire (LFESSQ) to identify those with chronic widespread pain (bilateral, above/ below waist for ⱖ1 week in the past 3 months) (CWP) and sociodemographic questions. Subjects who screened positive (CWP⫹) and a control group who screened negative (CWP-) were invited to schedule a site visit at one of 20 sites for physician evaluation of FM including a manual tender point exam until the established cap was met. CWP⫹ subjects who were diagnosed with FM based on the physician’s evaluation were designated as having FM and included in prevalence estimates. Point prevalence estimates of FM were calculated using an adjusted approach for a two-phase study design, and sampling weights reflecting the underlying screened population were used to refine estimates. Following the site visit subjects completed the subject questionnaire including validated patient-reported outcome instruments to measure FM impact, symptoms, and pain. Results: 8,382 (1,939 CWP⫹, 6,443 CWP-) subjects completed the screener. A total of 475 subjects attended a site visit: 350 of 1,331 CWP⫹ and 125 of 502 CWP- who consented to be scheduled for a site visit. The mean age (years) of the screened sample was 44.1, and there was representation across racial groups with 75.1% white and 12.3% black/African American. Females were intentionally over-sampled (79.3%) in order to generate precise

Disclosure: L. M. Arnold, None; T. Blom, None.

139 Pain and Other Symptoms During The Childhood Of Fibromyalgia Patients. Robert S. Katz1, Alexandra Small2, Ben J. Small3 and Susan Shott4. 1 Rush Medical College, Chicago, IL, 2University of Illinois Medical School, Chicago, IL, 3Rush University Medical School, Chicago, IL, 4Rush University Medical Center, Chicago, IL. Background/Purpose: We investigated the severity of pain and other symptoms during childhood, before age 12 and before age 18, that might be related to fibromyalgia (FMS). Methods: 115 FMS patients and 63 control patients with other rheumatic diseases answered a rheumatology office questionnaire that included questions about symptoms experienced before the ages of 12 and 18. Patients were asked to rate each symptom as 1 ⫽ no problem, 2 ⫽ seldom, 3 ⫽ occasional,

S55

Sunday, October 27

symptoms and possible underdiagnosis of HPT, we have examined the prevalence of HPT in a pilot study of FM patients. Methods: A retrospective chart review of a convenience sample of FM patients attending a tertiary care multidisciplinary pain center was undertaken to assess parathyroid and calcium status. Consecutive patients attending the clinic between December 2012 and March 2013, had a primary diagnosis of FM as defined by clinical criteria, and had routine biochemistry screening during the preceding year. Routine blood tests included measurements for bone health, vitamin D, parathyroid hormone, calcium and phosphate levels. Hyperparathyroidism was diagnosed by the following biochemical parameters: parathyroid hormone (PTH) level ⬎ 9.3 pmol/L (93 pg/mL) with inappropriately non-suppressed calcium (true HPT or normocalcemic HPT), and/or an elevated serum ionized calcium ⬎ 1.32 mmol/L (5.3 mg/dL) with inappropriately high PTH levels (normohormonal HPT). Results: Thirty-eight patients, 35 female, with a mean ⫾ standard deviation of age 51.6 ⫾10.6 years, and similar disease duration comprised the cohort. Based on their constellation of biochemical indices, 4 patients of the cohort, representing 10.5% of subjects studied, were diagnosed with hyperparathyroidism. Conclusion: In this small exploratory study, we have identified that unsuspected hyperparathyroidism in diagnosed FM patients has a prevalence of greater than 10%, as compared to 0.1% in the general population, representing a 100-fold difference. As no previous study has examined HPT in FM patients, these preliminary findings call for a more thorough study of parathyroid hormonal status in FM.

Sunday, October 27

4 ⫽ moderate, and 5 ⫽ severe problem. The chi-square test of association and Fisher’s exact test were used to compare percentages, and the Mann-Whitney test was done to compare FMS and control patients with respect to age and symptom severity ratings. A 0.05 significance level was used and all tests were two-sided. Results: The mean age was 48.1 ⫾12.3 years for FMS patients and 50.7 ⫾13.6 for rheumatic disease control patients (p ⫽ 0.092). 81.7% of the FMS patients and 61.9% of the control patients were women (p ⫽ 0.004). FMS patients had significantly worse severity ratings than did control patients for the following symptoms before age 18 (all with p ⬍ 0.001): “growing pains” (FMS 2.1⫾ 1.5 vs. controls 1.4⫾ 0.9), sleep problems (FMS 2.3⫾ 1.4 vs. controls 1.3⫾ 0.9), fatigue (FMS 2.2 ⫾ 1.4 vs. controls 1.4 ⫾0.9), concentration problems (FMS 2.3 ⫾1.4 vs. controls 1.5⫾ 0.8), memory problems (FMS 2.2⫾1.5 vs. controls 1.4 0.8), anxiety (FMS 2.1 ⫾1.4 vs. controls 1.4⫾ 1.0), headaches (FMS 2.7 ⫾1.4 vs. controls 1.6 1.0), stomach aches (FMS 2.3 ⫾1.4 vs. controls 1.3 ⫾0.8), and depressed mood (FMS 2.1⫾ 1.3 vs. controls 1.3⫾ 0.7). Similar results were obtained for the same symptoms before age 12 (all with p⬍ 0.022). Conclusion: Pain, sleep problems, fatigue, concentration, and memory complaints were reported to be more common before the ages of 18 and before 12 in FMS subjects. Though recall bias could be a limitation of this study, the data suggest that fibromyalgia patients frequently have symptoms in childhood. Other studies have suggested that fibromyalgia may be related to a trait (fibromyalgianess). This study points out that the vulnerability to this illess might be manifested early in life.

awareness of the Registry among pregnant women who have taken milnacipran and healthcare providers remains a major focus. Disclosure: D. Hirsch, Forest Research Institute, 3; D. Leclair, Forest Research Institute, 3; R. H. Palmer, Forest Research Institute, 3; V. Brown, Forest Laboratories, Inc., 5; S. Sinclair Roberts, Forest Laboratories, Inc., 5; J. McLean, Forest Laboratories, Inc., 5.

141 Skin Disease and Fibromyalgia: Dermatologic Diagnoses Encountered In a Series Of 845 Patients With Fibromyalgia At Mayo Clinic In 2008. Valerie Laniosz, David Wetter and Desiree Godar. Mayo Clinic, Rochester, MN. Background/Purpose: To determine the common dermatological diagnoses and skin-related complaints in a cohort of patients with fibromyalgia seen in a tertiary referral center. Methods: A retrospective chart review was performed of all patients with a diagnosis of fibromyalgia from January 1st, 2008-December 31st2008, whose diagnosis of fibromyalgia was confirmed in the Fibromyalgia Clinic at Mayo Clinic in Rochester, Minnesota. Charts were reviewed for dermatologic conditions and cutaneous symptoms. Demographic and clinical data were collected to assess for the frequency of skin-related issues in patients with fibromyalgia. Results: Of 2233 patients screened, 845 patients met the inclusion criteria of having a confirmed diagnosis of fibromyalgia. Amongst these 845 patients with fibromyalgia, various dermatologic conditions and cutaneous complaints were identified including hyperhidrosis in 270 (31.8%); burning of the skin or mucous membranes in 29 (3.4%); and various unusual cutaneous sensations in an additional 14 (1.7%). Pruritus without identified cause was noted by 28 patients (3.3%); with another 16 (1.9%) with neurotic excoriations, prurigo nodules, or lichen simplex chronicus. 77 patients (9.1%) were found to have some form of dermatitis other than neurodermatitis. Conclusion: Patients with fibromyalgia may have skin-related symptoms associated with their fibromyalgia. There does not appear to be any single dermatologic diagnosis overrepresented in this population with the exception of a subjective increase in sweating.

Disclosure: R. S. Katz, None; A. Small, None; B. J. Small, None; S. Shott, None.

140 Fibromyalgia and Pregnancy: Challenges Of The Savella® (Milnacipran) Pregnancy Registry. Damien Hirsch1, Denise Leclair1, Robert H. Palmer1, Vikki Brown2, Susan Sinclair Roberts2 and Janice McLean2. 1Forest Research Institute, Inc., Jersey City, NJ, 2INC Research, LLC, Wilmington, NC. Background/Purpose: During pregnancy, women with fibromyalgia (FM) often experience worsening of symptoms, which include widespread pain, fatigue, sleep disturbances, cognitive difficulties, and reduced daily functioning. Although the safety and efficacy of milnacipran in FM patients has been demonstrated in large clinical trials, these studies excluded pregnant women and milnacipran safety in this population is largely unknown. Milnacipran is a serotonin and norepinephrine reuptake inhibitor approved in the US for the management of FM. As with other drugs in this class, milnacipran is currently designated by the FDA as a Pregnancy Category C drug. As FM disproportionally affects women and, as a result, there is potential for exposure to milnacipran in women of child-bearing age, a pregnancy registry was established in October 2009. Methods: This ongoing, voluntary, prospective, observational, US-based registry employs an exposure-registration cohort design that allows for the evaluation of birth defects and other pregnancy outcomes in women ⱖ18 years exposed to milnacipran at any time during pregnancy. Women are followed throughout pregnancy, with their infants followed until 1 year of age. Enrollment in the Registry is initiated by the woman or by her healthcare provider. Pregnancies with evidence of birth defects prior to enrollment are considered retrospective and excluded from the primary analysis to minimize selection bias. All data are reviewed at least yearly by the Registry’s Scientific Advisory Committee. Results: To date, 2 prospective subjects have enrolled in the Registry. Both gave birth to premature infants (at 32 and 35 weeks of gestation, respectively). No birth defects have been reported at this time. Enrollment is considerably lower than projected, and is far from the goal of 350 exposed pregnancies and 196 live births. Postmarketing surveillance data from 01Nov11 to 31Oct12 indicate that milnacipran was primarily prescribed to women (88.7%); however, only 21.5% of these women were 18–40 years of age (ie, women with the highest childbearing potential). Healthcare utilization data suggest that the pregnancy rate in women prescribed milnacipran may be ⬍0.5% per year, well below the estimated general population rate of 7%. Other reasons for low enrollment may include pregnancy avoidance due to FM symptoms, the Category C classification, or lack of interest in participating in a voluntary registry. Conclusion: Possibly due to the demographics of the target population, enrollment in the Registry has been challenging. The sample size is insufficient to reach conclusions about milnacipran safety in pregnancy. However, the Registry will continue to monitor milnacipran-exposed pregnancies for teratogenicity signals or other negative pregnancy outcomes. Increasing

Disclosure: V. Laniosz, None; D. Wetter, None; D. Godar, None.

142 Home Tender Point Measurement In Fibromyalgia Patients. Robert S. Katz1, Hannah Bond2 and Jessica L. Polyak2. 1Rush Medical College, Chicago, IL, 2Rheumatology Associates, Chicago, IL. Background/Purpose: To determine whether: 1) patients with the fibromyalgia syndrome (FMS) can accurately evaluate their own tender points, similar to self-administered testing for other purposes (blood pressure and glucose), and 2) whether this might be helpful for monitoring the illness. Methods: We enrolled 24 patients with a known diagnosis of FMS (meeting the 2010 ACR criteria) and 27 healthy controls. Dolorimetry was performed on each patient by a rheumatology nurse, who also instructed the patient in self-dolorimetry using a pressure gauge smaller than a standard dolorimeter, (Wagner Force Dial FDK/FDN Series Push Pull Force Gage). Tender points were assessed using self-administered dolorimetry. Tender points were checked at 8 paired fibromyalgia tender point areas: occiput, lower cervical spine, chest wall, lateral epicondyles, knees, trapezius muscles, gluteus muscles, and greater femoral trochanters. Results: FM patients’ ages were 45 y.o. ⫹/⫺20, 18F and 10M. Normal controls ages were 48 y.o. ⫹/⫺15, 15F and 10M. Self-measured dolorimetry readings for fibromyalgia patients ranged from 3–20 pounds of pressure and for normal controls from 10–30. When we compared readings performed by a nurse with those done by patients, there were no significant differences. Dolorimetry exams of the trapezius muscle yielded the following results: 32 FMS patients recorded a mean reading of 5.86 pounds per pressure, while 32 controls had a mean reading of 12.05 pounds per pressure. Conclusion: Using self-administered dolorimetry, our results show that tender points can be measured by patients easily and with reasonable accuracy. The sites providing the best discrimination between patients and controls were the occiput, trapezius muscles, and gluteal muscles. Currently, clinician rated VAS scales are used to monitor the efficacy of

S56

Disclosure: R. S. Katz, None; H. Bond, None; J. L. Polyak, None.

143 Rapid Infrared Microscopy (IRMS) For Differential Diagnosis Of Fibromyalgia and Other Central Sensitivity Syndromes. Kevin V. Hackshaw, Luis Rodriguez-Saona, Irving L. Rosenberg, Marcal Plans and C.A. Tony Buffington. The Ohio State University, Columbus, OH.

Comparison of oxidative stress markers, BDNF and cytokines in FMS patients and healthy controls

Background/Purpose: The aim of this study was to investigate the ability of a rapid blood-spot-based method for diagnosis of Fibromyalgia (FM) using mid-infrared microspectroscopy (IRMS) to differentiate patients with FM from those with osteoarthritis (OA) and rheumatoid arthritis (RA) as well as another central sensitization disorder – Interstitial Cystitis (IC) - and to identify molecular species associated with the spectral patterns. Methods: Under IRB approval, blood samples were collected from patients diagnosed with FM (n ⫽ 14), RA (n ⫽ 15), or OA (n ⫽ 12) and IC (n ⫽ 20). Samples were prepared, placed onto a highly reflective slide, and spectra were collected using IRMS and compared using multivariate statistical modeling to differentiate groups. Aliquots of samples also were subjected to metabolomic analysis. Results: IRMS separated subjects into classes based on spectral information with no misclassifications among FM, RA, OA and IC patients. Discriminating power was greatest based on spectral bands centered at 1560cm⫺1, which are associated with indole ring vibrations of tryptophan or its derivatives. Metabolomic analysis revealed that RA and OA groups were metabolically similar, whereas biochemical differences were identified in the FM that were quite distinctive from those found in the other groups. Conclusion: These results support the use of IRMS to differentiate patients with FM from those with other chronic pain or central sensitivity syndromes. The accuracy and simplicity of the test may make it useful for earlier diagnosis of these syndromes.

Neuropsychiatric biomarkers TBARS (␮M of MDA) BDNF (ng/mL) Carbonil (nmol/mg) Cytokines (fg/mL) IL-6 IL-10 TNF-␣

FMS patients (nⴝ69) median (IQR)

Healthy controls (nⴝ61) median (IQR)

p value*

13.2 (9.8, 20.5) 30.8 (21.8, 37.1) 0.019 (0.015, 0.022)

14.4 (8.7, 29.2) 30.7 (21.1, 35.4) 0.021 (0.015, 0.027)

0.808 0.595 0.132

1498.7 (743.5, 2595.6) 593.5 (348.4, 890.2) 187.3 (145.7, 670.3)

1077.1 (720.0, 2119.1) 441.4 (266.6, 569.8) 187.3 (76.1, 352.9)

0.128 0.006 0.255

*Mann-Whitney test; the p value is not adjusted for multiple comparisons. IQR: interquartile range

Conclusion: The significant increase of serum IL-10 in patients with FMS found in this study may indicate a disregulation of antiinflammatory cytokines in this disease. This finding was also seen in patients with chronic fatigue syndrome, disease that is also characterized for chronic pain and tiredness. Disclosure: A. Ranzolin, None; C. A. D. C. Neto, None; B. M. Ascoli, None; B. Wollenhaupt-Aguiar, None; A. L. B. P. Duarte, None; M. Bredemeier, None; F. Kapczinski, None; R. M. Xavier, Pfizer, Roche and Merck, 5.

145 Autonomic Nervous System “Decomplexification” In Fibromyalgia. A Proof Of Concept Study Looking At The Fractality Of Heart Rhythms. Manuel Martinez-Lavin, Claudia Lerma, Laura Aline Martinez-Martinez, Oscar Infante and Angelica Vargas. National Institute of Cardiology, Mexico City, Mexico.

Disclosure: K. V. Hackshaw, None; L. Rodriguez-Saona, None; I. L. Rosenberg, None; M. Plans, None; C. A. T. Buffington, None.

Background/Purpose: Prevailing linear-reductionist medical model seems unable to explain complex diseases like fibromyalgia (FM) and similar maladies. As consequence of this divorce, some physicians disparage the fibromyalgia concept. Paradigms derived from the new complexity theory may provide a coherent framework for these elusive illnesses. Along these lines is the proposal that FM represents a degradation of our main complex adaptive system (the autonomic nervous system), in a failed effort to adjust to a hostile environment (Semin Arthritis Rheum 2008;37:260). Resilient complex systems have fractal structures (fractal is defined as an object or quantity that displays self-similarity at different scales). Fractality loss would signal degradation of a given system. Heart rate variability (HRV) fractal scaling index (alpha) is a novel non-linear method to assess autonomic nervous system resilience: high vagal influence produces less auto-correlated behavior (alpha values below 1.0) while increased sympathetic activity produces more auto-correlated and stiffer behavior (alpha greater than 1.0). Alpha values close to 1.0 indicate resilient fractal-like behavior (Open Cardiovasc Med J 2009;3:110). The objective of this study was to estimate the HRV fractal scaling index in FM patients, and to correlate this scaling index with clinical symptoms. Methods: We studied 30 women with FM (1990 ACR criteria) that were free from any medication that may affect the autonomic nervous system. Thirty age - matched healthy women served as controls. HRV was measured during 24 hrs with a Meigaoyi DSM-3 Holter monitor, while subjects were doing their routine activities. The fractal scaling index was estimated with the detrended fluctuation analysis method using custom computer programs which were previously validated. The scaling index was estimated in short-term scales or alpha 1 (4 to 11 heartbeats) and long-term scales or alpha 2 (greater than 11 heartbeats). Mean values between groups were compared with Student t test. Correlations between alpha indexes and symptoms scales were calculated with Pearson⬘s or Spearman⬘s correlations. Results: Demographic features and main results are shown in the table. The short-term fractal scaling index (alpha1) was higher in FM patients when compared to controls (1.22 ⫹/⫺0.10 vs. 1.16 ⫹/⫺ 0.10 p ⫽ 0.035). There was a positive correlation of fractal scaling index alpha 1 with total FIQ score (Rho⫽0.322, p⫽0.012).

144 Cytokines, Oxidative Stress Markers and Brain-Derived Neurotrophic Factor In Fibromyalgia Syndrome. Aline Ranzolin1, Claudio Antoˆnio da C. Neto2, Bruna Maria Ascoli3, Bianca Wollenhaupt-Aguiar4, Angela Luzia B. Pinto Duarte2, Markus Bredemeier5, Fla´vio Kapczinski3 and Ricardo M. Xavier6. 1Programa de Po´s-Graduaçaõ em Cieˆncias Me´dicas da Universidade Federal do Rio Grande do Sul, Porto Alegre, Brazil, 2Faculdade de Medicina, Universidade Federal de Pernambuco - Hospital das Clıńicas de Pernambuco, Recife, Brazil, 3Programa de Po´s-Graduaçaõ em Cieˆncias Me´dicas da Universidade Federal do Rio Grande do Sul: Psiquiatria, Porto Alegre, Brazil, 4 Programa de Po´s-Graduaçaõ em Cieˆncias Biolo´gicas: Bioquı´mica, da Universidade Federal do Rio Grande do Sul, Porto Alegre, Brazil, 5Hospital Nossa Senhora da Conceiçaõ - Grupo Hospitalar Conceiçaõ, Porto Alegre, Brazil, 6Faculdade de Medicina, Universidade Federal do Rio Grande do Sul - Hospital de Clıńicas de Porto Alegre, Porto Alegre, Brazil. Background/Purpose: The pathophysiology of Fibromyalgia Syndrome (FMS) is incompletely understood. An imbalance of pro- and antiinflammatory cytokines is assumed to play a role in the induction and maintenance of pain, but the many studies that have investigated cytokine levels in patients with FMS presented diverse results. The objective of our study was to analyze different cytokines, oxidative stress markers and brain-derived neurotrophic factor (BDNF) in patients with FMS. Considering that depression is also a possible cause of alteration in some of these markers, its presence was analyzed. Methods: A total of 130 female subjects (69 FMS and 61 healthy controls) were included in the present study. The cytokines (IL-6, IL-10 and TNF␣), oxidative stress markers (Thiobarbituric Acid Reactive Substances – TBARS – and carbonyl) and BDNF expression patterns were evaluated in all subjects. The Hamilton and Beck‘s depression scales, as well as the Fibromyalgia Impact Questionnaire (FIQ), were applied in FMS patients. BDNF serum levels were measured by sandwich-ELISA; the concentration of serum cytokines was determined by flow cytometry; levels of lipid peroxidation were measured by the method of TBARS and oxidative damage to proteins were analyzed by the determination of carbonyl groups content in proteins.

S57

Sunday, October 27

Results: Age was similar between groups (44.5 ⫾ 6.4 years in FMS and 44.0 ⫾ 6.7 in controls; p⫽ 0.613). In the FMS group, the mean of FIQ was 70.3 ⫾ 18.1 and almost all patients had depression (95.6 %) considering an abnormal score on specific scales. The majority of patients had mild/moderate depression (52.2% on Beck scale and 66.6% on Hamilton scale). IL-10 levels where significantly higher in FMS subjects, but other cytokines and biomarkers did not differ between the groups. (table). There was no correlation of any biomarker or cytokine with the FIQ and the Hamilton and Beck‘s depression scales.

treatment in FMS. Self-administered dolorimetry may also allow patients to assess treatment efficacy. This method of home monitoring would be a convient and inexpensive option for patients to monitor flares of disease.

Sunday, October 27

Group FM n ⴝ30 Controls n ⴝ 30 Standard Standard Mean Deviation Mean Deviation Age BMI FIQ score Fractal scaling index alpha 1 Fractal scaling index alpha 2

31 23.8 63.24 1.22 0.93

8 4.4 16.38 0.10 0.04

31 24.4 10.30 1.16 0.95

8 3.2 10.05 0.10 0.05

Background/Purpose: The use of whole blood gene expression to predict and follow the response to TNF inhibition therapy in RA has been challenging due to the complex nature of the data. Methods: Here we employ an approach to gene expression analysis that is based on gene expression “modules” previously reported by Chassubel (Immunity 2:150–64, 2008). Whole blood RNA (PAXgene) was obtained at baseline and 12 or 14 weeks on two cohorts of rheumatoid arthritis patients beginning anti-TNF therapy. Results: The initial cohort was enrolled by the ABCoN and contains 50 subjects stratified by EULAR Good Responders (N⫽14), Moderate Responders (N⫽21) and Non Responders (N⫽15) at 12 weeks after starting therapy. Good and Moderate Responders exhibited highly significant changes in multiple modules using a hypergeometric analysis. These included dramatic decreases in modules related to the myeloid lineage and inflammation, along with increases in B cell and plasma cell modules as well as in a number of modules related to the MHC and ribosomal proteins and other “undefined” module groups. Strikingly, Non Responders exhibited very little change in any modules. We have replicated these data in patients enrolled in a clinical trial of Simponi®, with full expression data available on 29 Good Responders, and 37 Moderate Responders. We observed nearly identical modular changes to those identified in the ABCoN Responders after 14 weeks of treatment. Only 6 Non Responders were available for study in this dataset, making convincing replication difficult for this subgroup. Several other replication datasets are currently being analyzed. Conclusion: These data suggest that using gene expression modules related to inflammatory disease may provide a valuable method of characterizing the responder status of RA patients treated with TNF inhibitors or other biologic therapies.

P 0.856 NS 0.550 NS ⬍0.0001 0.035 0.143

Conclusion: The short-term HRV fractal scaling index is altered in FM patients with values indicating stiffer autonomic behavior. This fractal index correlated with total FIQ score. This tangible non-linear finding supports the notion that FM represents a degradation of our main complex adaptive system, namely the autonomic nervous system. Disclosure: M. Martinez-Lavin, None; C. Lerma, None; L. A. Martinez-Martinez, None; O. Infante, None; A. Vargas, None.

ACR Poster Session A Genetics and Genomics of Rheumatic Disease I Sunday, October 27, 2013, 8:30 AM–4:00 PM

146 The Arthritis Severity Locus Cia4 Is An Early Regulator of IL-6, IL-1␤, and NF␬B activators’ Expression in Pristane-Induced Arthritis. Max Brenner, Teresina Laragione and Percio Gulko. Hosftra North Shore-LIJ School of Medicine, Manhasset, NY.

Disclosure: M. Curran, Janssen Research & Development, LLC., 3; M. Oswald, None; S. Lamberth, Janssen Research & Development, LLC., 3; C. Brodmerkel, Janssen Research & Development, LLC., 3; Z. Cherkas, Janssen Pharmaceutica Product, L.P., 3; P. K. Gregersen, Janssen Pharmaceutica Product, L.P., 2.

Background/Purpose: Cia4 is a quantitative trait locus on rat chromosome 7 that regulates disease severity and joint damage in three models of rheumatoid arthritis including pristane-induced arthritis (PIA). To identify cellular and molecular processes regulated by Cia4, we studied the expression over 23,000 genes in synovial tissues from MHC identical DA (severe erosive disease) and DA.F344(Cia4) congenics (mild and non-erosive disease) rats. Methods: Synovial tissues from DA and DA.F344(Cia4) congenics were collected at pre-clinical (day 10 and day 14 post-pristane administration; qPCR) and recent onset (day 18; microarray and qPCR) stages following the induction of PIA and analyzed for gene expression levels. Results: Il6 levels were 135-fold higher in DA compared with congenics at very early pre-clinical stages (day 10), and remained significantly increased. The Il6 increase preceded the modest increase in Il1b (4.2-fold) suggesting that Il6 could be driving cytokine expression and the early histologic inflammatory infiltration. 187 genes had significantly different expression and included inflammatory mediators expressed in increased levels in DA such Slpi (10.94-fold) and its receptor Plscr1 (2.31-fold), Cd163 (5.85-fold), Ccl7 (5.17-fold) and Litaf (2.09-fold). Syk or NF␬B pathway activating and interacting genes were increased in DA synovial tissues. Fifty-nine genes implicated in cancer-related phenotypes were increased in DA, while genes involved in cell metabolism, transport across membranes and tissue protection such as Acat1, Dgat1, Dhcr7, Slc25a29, and Slc1a1 were increased in DA.F344(Cia4) congenics. 21 genes differentially expressed, or expressed in only one of the two strains were located within the Cia4 interval, and could be the gene accounting for the arthritis effect. Conclusion: The Cia4 interval contains a new arthritis gene that regulates early Il6, Il1b expression, and other inflammatory mediators central to arthritis, and processes involved in cancer that could be mediating the development of synovial hyperplasia and invasion, and cartilage and bone destruction.

148 Microarray Analysis Of Synovial Specimen of Early Human (CHECK) and Experimental Osteoarthritis to Identify Pathways and Processes Associated with Pathology. Arjen B. Blom1, Peter L. van Lent1, Martijn H. van den Bosch1, Hans Cats2, Frank H.J. van den Hoogen3, Peter M. van der Kraan1 and Wim B. van den Berg4. 1Radboud University Nijmegen Medical Centre, Nijmegen, Netherlands, 2Rheumatology Centre Sint Maartenskliniek, Nijmegen, Netherlands, 3Rheumatology Centre Sint Maartenskliniek and University Medical Centre St Radboud, Ubbergen (Nijmegen), Netherlands, 4Rheumatology Research and Advanced Therapeutics, Department of Rheumatology, Radboud University Nijmegen Medical Centre, Nijmegen, Netherlands. Background/Purpose: The majority of osteoarthritis (OA) patients show synovial inflammation, even relatively early during the disease. We used microarray analysis of synovial tissue of early OA patients and of experimental OA, to identify common pathways that determine joint damage in this disease. Methods: Expression analysis was performed on murine synovial tissue at day 7, day 21 and day 42 in collagenase induced OA (CIOA) and the surgically induced DMM model (destabilization of the medial meniscus). CIOA was induced by intra-articular injection of collagenase, which causes joint instability. From a subpopulation of patients (n⫽ 25) that entered the CHECK Cohort study (Cohort Hip and Cohort Knee) and 7 controls, synovial biopsies were collected at year 0, 2 and 5. CHECK is a prospective 10-year follow-up study on participants with early osteoarthritis-related complaints (less than 6 months before inclusion) initiated by the Dutch Arthritis Association. Kellgren&Lawrence-score (KL) at inclusion was determined (n⫽18) and follow up measurements were performed at 2 and 5 years. Affymetrix was used for microarray, and pathway analysis was done using DAVID. Results: Among the genes that were strongly upregulated on all 3 time points after induction of CIOA were MMP-3 (6-fold), MMP-13 (16-fold), MMP-14 (6-fold). Wound healing, phagocytosis, chemotaxis and metalloproteases were significantly enriched, as were the complement pathway, the TLR-, TGF␤, BMP and wnt-signaling pathways. Highly similar results were obtained in the DMM model for OA. However, at day 42 in this model very view genes were still regulated in the synovium compared

Disclosure: M. Brenner, None; T. Laragione, None; P. Gulko, None.

147 Modular Analysis Of In Peripheral Blood Gene Expression In Rheumatoid Arthritis Captures Reproducible Gene Expression Changes In TNF Responders. Mark Curran1, Michaela Oswald2, Sarah Lamberth1, Carrie Brodmerkel1, Zhenya Cherkas1 and Peter K. Gregersen3. 1Janssen Research & Development, LLC., Spring House, PA, 2The Feinstein Institute for Medical Research, Manhasset, NY, 3Feinstein Institute for Medical Research, Manhasset, NY.

S58

to other time points or CIOA, indicating that synovial activation differs late between the models. This was underlined by histological examination. All in all, the expression patterns in experimental OA showed compelling similarities with human OA synovium. Gene expression profiles of control synovia were compared to CHECK synovia. Analysis using DAVID indicated enrichment of several biological processes and signaling pathways, including macrophage presence, cell migration, TGF␤-, BMP- and wnt-signaling. This indicates activation of the synovium in the early OA versus controls. Next we compared synovial tissue of CHECK-patients with radiological damage (KLⱖ1) with CHECK-patients without joint damage (KL⫽0). In the top 30 genes that were associated with cartilage damage were MMP-1 (18-fold), MMP-3 (10-fold) and S100A8 (6-fold), all of which have been associated with cartilage damage. FAC analysis further underlined response to wounding, chemotaxis, innate immune response and metalloproteases to be strongly enriched. In particular, complement-activation pathway, TGF␤- and BMP-signaling and TLRactivation were striking. Conclusion: Activation pathways and processes in the two models for OA were highly similar. A major difference lies in the presence of late synovial activation. The data suggest an active role for the synovium in OA pathology, and identifies pathways that may be involved. Activation of the complement-pathway was strongly associated with damage. In addition, synovial MMP expression was associated with joint damage, underlining an active role of synovium in OA pathology.

and GPA. In Particular, GPA was characterized by high expression level of the proteins associated with the activity of the neutrophil.

150 A Subset Of Osteoarthritis Individuals Has Elevated IL-6 Pathway Activation Associated With Worse Symptoms. Michael Kuziora1, Zheng Liu1, Brandon W. Higgs1, Philip Brohawn1, Kim Lehmann1, Fernanda Pilataxi1, Katie Streicher1, Lydia Greenlees1, Meina Liang2, Rozanne Lee2, Amy Schneider2, Raffaella Faggioni2, Yihong Yao1 and Koustubh Ranade1. 1MedImmune, LLC, Gaithersburg, MD, 2MedImmune, LLC, Hayward, CA. Background/Purpose: Elevated levels of the pro-inflammatory cytokine IL-6 have been found in the affected joint of patients with osteoarthritis (OA), although the prevalence of OA patients demonstrating IL-6 pathway activation is not well-characterized. Our goal was to estimate the proportion of OA patients with increased IL-6 signaling in the diseased joint, and to correlate IL-6 pathway activation with Western Ontario and McMaster Universities Arthritis Index (WOMAC) scores. Methods: Synovial membranes were obtained from OA patients (N⫽ 101) undergoing knee replacement surgery. The WOMAC pain questionnaire was used to assess levels of pain on a visual analog scale of 100mm. Scores within each WOMAC subfunction (pain, stiffness and physical function) were averaged to derive a single composite subfunction score, and the WOMAC total score was calculated as the average of all subfunction scores. Synovial fluid and serum collected at the time of surgery was assayed for IL-6 and CCL8 proteins. Gene expression levels were measured using Affymetrix HG-U133Plus2 microarrays. Transcripts with ⬎2-fold over-expression following treatment with IL-6 and IL6R in synovial fibroblasts were identified. Hierarchical clustering using expression levels of these inducible transcripts in synovial tissue was used to identify OA patients showing high levels of IL-6 activity. A signature score was calculated as the median expression intensities of IL-6 –inducible genes for each subject. Results: CCL8 and SOCS3 were identified as IL-6 inducible genes in synovial fibroblasts and clustering of synovial membrane specimens from OA subjects revealed a cluster (19/101 subjects, 19%) with a 3.5-fold increase in the median expression of these two genes compared to the remaining subjects (388.4⫾270.1 vs 111.4 ⫾ 38.7, p⫽3.0⫻10⫺4) suggesting significantly elevated IL-6 pathway activation in this subgroup. These “IL-6 signature high subjects” also had significantly higher mean levels of IL-6 (155.2⫾ 4.6 pg/mL vs 58.6 ⫾ 4.1 pg/mL p⫽7.4⫻1-⫺3) and CCL8 (30.5 ⫾ 2.5 pg/mL vs 14.4 ⫾ 2.1 pg/mL p⫽7.2⫻10⫺5) proteins in synovial fluid. The levels of CCL8 and IL-6 proteins in synovial fluid were significantly correlated (rho⫽0.52, p⫽1.6⫻10⫺7). Compared to subjects with low levels of IL-6 inducible genes in the synovium, with those in the IL-6 signature high cluster were more likely to be women (Odds ratio 4.2 95% C.I. [1.1, 15.4] p⫽0.03), and had higher mean WOMAC scores for pain (65.6 ⫾ 16.4 vs 55.6 ⫾18.3, p⫽ 4⫻10⫺2), stiffness (76.9 ⫾ 17.0 vs 60.5 ⫾ 24.8, p⫽3.4⫻10⫺3), physical activity (67.2 ⫾15.8 vs 56.3 ⫾ 18.2, p⫽4.4⫻10⫺2) and total score (69.9 ⫾ 14.6 vs 57.5 ⫾ 18.1, p⫽8.9⫻10⫺3). Conclusion: An IL-6 inducible gene signature consisting of CCL8 and SOCS3 was used to identify a subgroup of OA subjects with elevated IL-6 pathway activation in the synovial membrane. OA subjects in this IL-6-signature high group were more likely to be women and had higher mean WOMAC pain scores. CCL8 is a potential biomarker of IL-6 pathway activation in the affected joint.

Disclosure: A. B. Blom, None; P. L. van Lent, None; M. H. van den Bosch, None; H. Cats, None; F. H. J. van den Hoogen, None; P. M. van der Kraan, None; W. B. van den Berg, None.

149 Comparative Proteomic Analysis Of Neutrophils From Patients With Microscopic Polyangiitis and Granulomatosis With Polyangiitis. Teisuke Uchida1, Kohei Nagai2, Toshiyuki Sato3, Nobuko Iizuka3, Mitsumi Arito3, Yukiko Takakuwa4, Hiromasa Nakano4, Seido Ooka4, Manae Kurokawa3, Naoya Suematsu3, Kazuki Okamoto3, Shoichi Ozaki4 and Tomohiro Kato3. 1St. Marianna University School of Medicine, Kawasaki, Japan, 2Kinki University, Wakayama, Japan, 3St. Marianna University Graduate School of Medicine, Kawasaki, Japan, 4St. Marianna University School of Medicine, Kawasaki, Japan. Background/Purpose: Both microscopic polyangiitis (MPA) and granulomatosis with polyangiitis (GPA) belong to ANCA-associated vasculitis (AAV), in which neutrophils are thought to be involved in their pathology. Clinically, it is often difficult to distinguish MPA from GPA. To discriminate between MPA and GPA, protein profiles of peripheral blood polymorphonuclear cells (PMNs) of MPA and GPA patients and healthy controls (HC) were analyzed. Methods: Proteins extracted from peripheral blood PMNs of 11 MPA patients, 9 GPA patients, and 10 HC were separated by two-dimensional difference gel electrophoresis (2D-DIGE). Differentially expressed protein spots were identified by mass spectrometry analysis. Then, to find biomarker candidates which discriminate between MPA and GPA, the obtained protein profiles were subjected to the multivariate data analysis using SIMCA-P⫹ containing principal component analysis (PCA) and orthogonal partial-least-squares-discriminate analysis (OPLS-DA), and subjected to the receiver operating characteristic (ROC) analysis. Results: In all the 864 protein spots detected, intensity of 55 spots was found to be significantly different (p ⬍ 0.05) among the three groups by an analysis of variance (ANOVA). 31 out of the 55 spots were identified by mass spectrometry. The OPLS-DA analysis revealed that the expression profile of the protein spots discriminated the AAV group from the HC group completely and also discriminated the MPA group from the GPA group completely. 13 protein spots were considered as biomarker candidates to distinguish between MPA and GPA. In those, spots whose intensity was higher in MPA than in GPA included actin with various pI values, while a considerable part of spots whose intensity was higher in GPA than MPA were proteins related with the activity of neutrophils. Among the candidate proteins, ROC analysis showed that a combination of neutrophil gelatinase-associated lipocalin and a kinase anchor protein 7 isoforms beta had a high diagnostic potential. Conclusion: In the study, we determined that the protein profile of the neutrophil was clearly different between AAV and HC, and between MPA

Disclosure: M. Kuziora, AstraZeneca, 1, MedImmune, 3; Z. Liu, MedImmune LLC, 3, AstraZeneca, 1; B. W. Higgs, MedImmune LLC, 3, AstraZeneca, 1; P. Brohawn, MedImmune LLC, 3, Astra Zeneca, 1; K. Lehmann, MedImmune, 3, AstraZeneca, 1; F. Pilataxi, AstraZeneca, 1, MedImmune, 3; K. Streicher, AstraZeneca, 1, MedImmune, 3; L. Greenlees, AstraZeneca, 1, MedImmune, 3; M. Liang, AstraZeneca, 1, MedImmune, 3; R. Lee, AstraZeneca, 1, MedImmune, 3; A. Schneider, AstraZeneca, 1, MedImmune, 3; R. Faggioni, AstraZeneca, 1, MedImmune, 3; Y. Yao, MedImmune, 3, AstraZeneca, 1; K. Ranade, AstraZeneca, 1, MedImmune, 3.

S59

Sunday, October 27

Disclosure: T. Uchida, None; K. Nagai, None; T. Sato, None; N. Iizuka, None; M. Arito, None; Y. Takakuwa, None; H. Nakano, None; S. Ooka, None; M. Kurokawa, None; N. Suematsu, None; K. Okamoto, None; S. Ozaki, None; T. Kato, None.

Background/Purpose: The Osteoarthritis Initiative (OAI) is a multicentre study focused on identifying and evaluating risk factors of knee osteoarthritis (OA). Previous studies carried out by our group showed the role of the mitochondrial DNA (mtDNA) haplogroups on the prevalence and severity of OA. The aim of this study is to replicate our previous findings about the influence of the mtDNA haplogroups on OA progression in the well characterized cohort of the OAI. Methods: We assigned the mtDNA haplogroups in 891 Caucasian samples of the progression subcohort of the OAI to analyse their influence on radiographic OA progression attending to: KL grade, joint space narrow (JSN), presence of osteophytes and subchondral sclerosis in media tibial compartment between baseline and visit 6 (48 months). The progression criteria for KL grade consisted in an increase of at least one (KL or OARSI) grade at any visit and any knee; for JSN if an increase in score by a grade3 0.5 of the OARSI scale at any visit and any knee occurred. Additionally, we also analysed the four-year change in radiographic medial joint space width (mJSW) at x⫽0.225 in (N⫽265) patients with baseline unilateral JSN in both JSN knees (OARSI atlas grade ⱖ 1) and non-JSN knees (OARSI atlas grade ⬍ 1). Appropriate statistical analyses adjusting by gender, age and body mass index (BMI) at baseline were carried out using SPSS software (v.19) and R software v2.10.0 (The R Foundation for Statistical Computing). Results: OA patients in the progression subcohort that carry the mtDNA haplogroup T showed the lowest significant cumulative probability of progression in terms of KL grade (Relative Risk (RR) ⫽ 0.490; 95% Confidence Interval (CI): 0.208–0.762; p ⬍ 0.05), osteophytes (RR ⫽ 0.490; 95% CI: 0.208–0.762; p ⬍ 0.05), subchondral esclerosis (RR ⫽ 0.490; 95% CI: 0.208–0.762; p ⬍ 0.05) and JSN (RR ⫽ 0.490; 95% CI: 0.208–0.762; p ⬍ 0.05) when compared with OA patients that carry the most common mtDNA haplogroup H. Regarding to mJSW at x⫽0.225 in both knees of patients with unilateral JSN during the follow-up period of 48 months, the results obtained showed that carriers of the haplogroup T had a significant smaller decline in the mJSW at x⫽0.225 in non-JSN knees (p ⫽ 0.033). For JSN knees, though carriers of the haplogroup T showed the smallest decline too, the differences did not reach the statistical significance. Conclusion: This work strength the hypothesis that mitochondrial genome is a key factor in the progression of the OA disease. The early identification and classification of patients with haplogroup T and the most common haplogroup H would permit to carry out a more personalized follow-up of the disease.

Sunday, October 27

151 The mtDNA Haplogroups Influence The Cartilage Integrity In Osteoarthritis. Data From The Osteoarthritis Initiative (OAI). Ignacio RegoPe´rez1, Angel Soto-Hermida1, Mercedes Fernańdez-Moreno1, Sonia Pe´rtegaDıáz2, Juan Fernańdez-Tajes1, Marıá Eugenia Va´zquez-Mosquera1, Estefanıá Corte´s-Pereira1, Sara Relanõ-Fernańdez1, Natividad Oreiro-Villar1, Carlos Fernańdez-Lo´pez1 and Francisco J. Blanco3. 1INIBIC-Hospital Universitario A Corunã. Rheumatology Division. Genomic Group, A Corunã, Spain, 2 INIBIC-Hospital Universitario A Corunã. Epidemiology and Public Health Unit, A Corunã, Spain, 3Osteoarticular and Aging Res. Lab. CIBER-BBN. INIBIC- University of A Corunã, A Corunã, Spain. Background/Purpose: Magnetic resonance imaging (MRI) is ideally suited for detect structural changes and degradation in cartilage against the limitations of radiographic common methods. Changes in volume and thickness are the most widely used variables to measure progression in OA with MRI. The aim of this study is to elucidate the influence of the mtDNA haplogroups on cartilage integrity based on MRI data obtained from different OAI projects. Methods: We analysed the influence of the mtDNA haplogroups on quantitative measurements of cartilage morphology from MRI scans, mean cartilage thickness and volume of cartilage, in the whole knee joint by means of both cross-sectional and longitudinal studies during a follow-up period of 24 months. We selected for analysis 326 patients from project 18 (crosssectional) and 358 from project 9 (longitudinal), all of them belonging to the progression subcohort of the OAI. Appropriate statistical analyses adjusting by gender, age and body mass index (BMI) at baseline were carried out using SPSS software (v.19) and R 2.10.0 (The R Foundation for Statistical Computing). Results: The cross-sectional analyses showed that OA patients that carry the mtDNA haplogroup T had increased median values of cartilage volume in medial tibia (2.35⫻103 mm3 vs 1.95⫻103 mm3; p⫽0.009) and medial tibia femoral (3.56⫻103 mm3 vs 3.02⫻103 mm3; p⫽0.015) compartments, when compared with the most common mtDNA haplogroup H. In relation to mean cartilage thickness, OA patients that carry the mtDNA haplogroup T also showed higher mean cartilage thickness in medial tibia, however this difference borderline the statistical significance (1.85 mm vs 1.73 mm; p⫽0.07). During the follow-up period of 24 months, OA patients that carry the haplogroup T suffered a significant smaller decline of volume in medial tibia femoral (p⫽0.015) and central medial femur (p⫽0.016) compartments when compared with the most common mtDNA haplogroup H. Even when the normalized cartilage volume of these two subregions was analysed, the results obtained were similar (p⫽0.023 and p⫽0.031 respectively). In relation to mean cartilage thickness, patients with the mtDNA haplogroup T suffered a significant smaller decline in central medial tibia femoral (weight bearing) (p⫽0.011), medial tibia femoral (p⫽0.019), medial tibia (anterior) (p⫽0.007) and central medial femur (center) (p⫽0.013) compartments when compared with the mtDNA haplogroup H. The statistical models applied in the longitudinal approach showed both age and gender (female) as risk factors for volume loss (p⬍0.02 and p⬍0.0001 respectively) and thickness loss (p⬍0.05 and p⬍0.0001 respectively) in articular cartilage over time. Conclusion: The mtDNA haplogroups manifest an intrinsic relation with OA progression in terms of cartilage integrity; carriers of this mitochondrial variant show a slower disease progression than haplogroup H carriers (most common in Caucasian population). The early haplogroup assignment could be crucial for an effective follow-up of the disease and a fit treatment.

Disclosure: A. Soto-Hermida, None; I. Rego-Pe´rez, None; M. Fernańdez-Moreno, None; S. Pe´rtega-Dıáz, None; J. Fernańdez-Tajes, None; M. E. Va´zquez-Mosquera, None; E. Corte´s-Pereira, None; S. Relanõ-Fernańdez, None; N. Oreiro-Villar, None; C. Fernańdez-Lo´pez, None; F. J. Blanco, None.

153 Identification Of Autoimmune Disease Risk Alleles That Are Under Recent Selection In The Sea Island Gullah African Americans. Carl D. Langefeld1, Satria Sajuthi1, Jasmin Divers1, Yiqi Huang2, Uma Nayak2, Wei-Min Chen2, Kelly J. Hunt3, Diane L. Kamen3, Gary S. Gilkeson3, Jyotika K. Fernandes3, Ida J. Spruill3, W. Timothy Garvey4, Miche`le M. Sale2 and Paula S. Ramos3. 1Wake Forest School of Medicine, Winston-Salem, NC, 2University of Virginia, Charlottesville, VA, 3Medical University of South Carolina, Charleston, SC, 4University of Alabama at Birmingham, Birmingham, AL.

Disclosure: I. Rego-Pe´rez, None; A. Soto-Hermida, None; M. Fernańdez-Moreno, None; S. Pe´rtega-Dıáz, None; J. Fernańdez-Tajes, None; M. E. Va´zquez-Mosquera, None; E. Corte´s-Pereira, None; S. Relanõ-Fernańdez, None; N. Oreiro-Villar, None; C. Fernańdez-Lo´pez, None; F. J. Blanco, None.

Background/Purpose: The reasons for the ethnic disparities in rheumatic and autoimmune diseases (AIDs) are largely unknown. Given the increasing evidence of selection at loci associated with human diseases, identification of alleles under selection may provide further insight into disease susceptibility. The Gullahs form a unique population of African ancestry in the U.S. In addition to their relative genetic and environmental homogeneity and low European admixture, a shorter genetic distance between the Gullahs and Sierra Leoneans has also been reported, suggesting that population genetic signals, such as regions under recent selection, may be more easily detected in the Gullahs than in other African American (AA) populations. We sought to capitalize upon the relative closeness between the Gullah and Sierra Leoneans to identify regions that differentiate both populations and may hence be under recent population-specific selective pressures. The goal of this study was to identify regions that might be under recent positive selection in the Gullah and that harbor risk loci for AIDs, which may help explain the higher prevalence of autoimmunity in AA.

152 The Uncoupling mtDNA Haplogroup T Associates With Better Radiographic Progression Of Osteoarthritis. Data From The Osteoarthritis Initiative (OAI). Angel Soto-Hermida1, Ignacio Rego-Pe´rez1, Mercedes Fernańdez-Moreno1, Sonia Pe´rtega-Dıáz2, Juan Fernańdez-Tajes1, Marıá Eugenia Va´zquez-Mosquera1, Estefanıá Corte´s-Pereira1, Sara RelanõFernańdez1, Natividad Oreiro-Villar1, Carlos Fernańdez-Lo´pez1 and Francisco J. Blanco3. 1INIBIC-Hospital Universitario A Corunã. Rheumatology Division. Genomic Group, A Corunã, Spain, 2INIBIC-Hospital Universitario A Corunã. Epidemiology and Public Health Unit, A Corunã, Spain, 3CIBERBBN-ISCIII, Madrid, Spain.

S60

between the Gullah and the YRI, revealing an enrichment of genes involved in cytokine/chemokine mediated immunity (P-Bonferroni⫽ 0.0015), but only two regions were being selected for in the Gullahs, none of which has previously shown evidence of association with an AD. Regions associated with ADs showing evidence of selection in the YRI population include those of the the HLA region (all ADs), the CCR2 (celiac disease), and SMOC2 genes (vitiligo), with CCR2 also showing evidence of selection in the Sierra Leone population. This scan also identified the region around the APOL1 gene, which has been shown to be under selection and associated with multiple kidney diseases in AAs. Conclusion: We have identified several risk alleles for ADs that are targets of recent positive selection in an AA population. Given the increased prevalence of several ADs in AAs and the homogeneity of the Gullahs, identification of these regions in the Gullahs has the potential to elucidate AD risks in AAs and help explain the ethnic disparity. Disclosure: P. S. Ramos, None; N. Titus, None; S. Sajuthi, None; J. Divers, None; Y. Huang, None; U. Nayak, None; W. M. Chen, None; K. J. Hunt, None; D. L. Kamen, None; G. S. Gilkeson, None; J. K. Fernandes, None; I. J. Spruill, None; W. T. Garvey, None; M. M. Sale, None; C. D. Langefeld, None.

155 Exome Sequencing For Identification Of Potential Causal Variants For Diffuse Cutaneous Systemic Sclerosis. Angel CY Mak1, M Kari Connolly2, Tamiko Katsumoto3, Paul Wolters4, Clare Cleveland3, Blanca M Herrera1, Paul LF Tang1, Simi Mathauda1, Richard Lao1, Pui-Yan Kwok1 and Lindsey A. Criswell3. 1University of California, San Francisco, Institute for Human Genetics, San Francisco, CA, 2University of California, San Francisco, Department of Dermatology, San Francisco, CA, 3University of California, San Francisco, Rosalind Russell Medical Research Center for Arthritis, San Francisco, CA, 4University of California, San Francisco, Department of Medicine, Pulmonary Division, San Francisco, CA.

Disclosure: C. D. Langefeld, None; S. Sajuthi, None; J. Divers, None; Y. Huang, None; U. Nayak, None; W. M. Chen, None; K. J. Hunt, None; D. L. Kamen, None; G. S. Gilkeson, None; J. K. Fernandes, None; I. J. Spruill, None; W. T. Garvey, None; M. M. Sale, None; P. S. Ramos, None.

154 Identification Of Autoimmune Disease Genes In Regions Under Selection In The Gullah African American Population Of South Carolina. Paula S. Ramos1, Nathan Titus2, Satria Sajuthi2, Jasmin Divers2, Yiqi Huang3, Uma Nayak3, Wei-Min Chen3, Kelly J. Hunt1, Diane L. Kamen1, Gary S. Gilkeson1, Jyotika K. Fernandes1, Ida J. Spruill1, W. Timothy Garvey4, Miche`le M. Sale3 and Carl D. Langefeld2. 1Medical University of South Carolina, Charleston, SC, 2Wake Forest School of Medicine, Winston-Salem, NC, 3University of Virginia, Charlottesville, VA, 4University of Alabama at Birmingham, Birmingham, AL.

Background/Purpose: Scleroderma is a genetically complex autoimmune disease with substantial phenotypic heterogeneity. Previous genome-wide association studies (GWAS) have identified a large number of gene regions associated with disease risk. However, GWAS directly capture only common genetic variation that are presumably in linkage disequilibrium with causal variants, which may be rare in the general disease population. Our goal was to identify rare and potentially causal variants through performance of whole exome sequencing. We focused on patients with severe disease, and specifically those with diffuse cutaneous systemic sclerosis (dcSSc), to limit disease heterogeneity. Methods: Our initial exome-sequencing studies were performed on 24 dcSSc patients using the Illumina HiSeq2500 platform and the Nimblegen SeqCap EZ v3.0 exome enrichment protocol. We also studied 60 healthy control subjects who underwent exome sequencing in the same laboratory using comparable methods. Exome sequencing reads were processed using the automated pipeline for next-generation sequencing data (bcbionextgen, https://github.com/chapmanb/bcbio-nextgen) and variants were annotated with ANNOVAR. We applied a gene burden test to identify genes that were enriched or depleted with rare (MAFⱕ0.01) and functionally deleterious (SIFTⱕ0.05) variants in dcSSc patients compared to control individuals. Results: Paired 100bp end reads were generated with a mean coverage of 56X on the targeted exome regions (64Mb). We identified 31 genes that were enriched or depleted with rare and functionally deleterious variants in dcSSc patients. Among the 31 genes, 6 genes (ADAMTS8, ALMS1, CHST15, DACH1, FMNL2, RELN) were in previously identified scleroderma susceptibility loci or pathways implicated in scleroderma pathogenesis. Conclusion: Using exome sequencing and gene burden analysis, we identified 31 genes that contain rare and potentially functionally deleterious variants that may contribute to the development of dcSSc. This pilot study demonstrates the potential value of whole exome sequencing for the identification of causal variants that contribute to scleroderma risk and/or severity.

Background/Purpose: Most autoimmune diseases (ADs) are more prevalent in specific ethnic groups. We hypothesize that one reason for the ethnic disparity may be an effect of population-specific selection influencing the allele frequencies at some loci. Selection for specific alleles that improves fitness and survival in a specific environment can lead to increased disease risk in a different environment. Given the growing number of disease-associated loci in regions that show evidence of selection, identification of alleles under selection may provide insight into disease susceptibility. Relative to other African-Americans (AA), the AA Gullah population has lower European admixture and higher ancestral homogeneity from the Sierra Leone area in Far-West Africa. The shorter genetic distance between the Gullahs and Sierra Leoneans suggests that population genetic signals, such as regions under recent selection, may be more easily detected in the Gullahs than in other African American (AA) populations. The goal of this study was to identify regions under recent positive selection in the Gullahs that may harbor risk loci for ADs. Methods: We computed the cross population extended haplotype homozygosity test (XP-EHH) to identify alleles with higher than expected frequency relative to their haplotype length in one population relative to another. We compared 277 control Gullahs to 400 Sierra Leonean controls and to 203 HapMap Yorubans (YRI) from Nigeria. A total of 679,513 SNPs with MAF⬎5% met all statistical quality control criteria. Regions that met genome-wide significance ((兩XP-EHH兩⬎4, P⬍E-07) were followed to assess for overlap with regions known to be associated with ADs. Results: We identified multiple regions with evidence of recent selection. Over 20 regions showed evidence of selection between the Gullah and the Sierra Leoneans (兩XP-EHH兩⬎4, P⬍E-07). They revealed an enrichment of genes involved in resistance to HIV infection (P-Bonferroni⫽0.017). Half of these regions showed evidence of selection in the Gullahs, including regions associated with ADs: DAB1 (Kawasaki disease), KCNH8 (Crohn’s disease, psoriasis), and LPP genes (vitiligo, celiac disease). Twenty four regions showed evidence of selection

Disclosure: A. C. Mak, None; M. K. Connolly, None; T. Katsumoto, None; P. Wolters, None; C. Cleveland, None; B. M. Herrera, None; P. L. Tang, None; S. Mathauda, None; R. Lao, None; P. Y. Kwok, None; L. A. Criswell, None.

S61

Sunday, October 27

Methods: We computed a linear regression model using the principal component of the HapMap Yoruba (YRI) population from Nigeria (PC2) as a quantitative outcome, using 277 Gullah and 400 Sierra Leonean samples genotyped on the Affymetrix Genome-Wide Human SNP Array 6.0. We adjusted for European admixture via inclusion of the HapMap Caucasian (CEU) population component (PC1) as a covariate. In total, 679,513 SNPs with MAF⬎5% were used in this analysis. In order to exclude spurious loci, only regions where at least one SNP met genome-wide significance (P⬍5E10) and a second significant SNP (P⬍E-07) in LD with it were considered. We then checked for the presence of immune/autoimmune-related genes in these regions. Results: Nine regions met our criteria as those that best differentiate the Gullah from the Sierra Leonean. The most significant was a ⬃2 Mb region at Xq22.2-q22.3 around the IL1RAPL2 gene, where 4 SNPs had P⬍5E-10. Of the other 8 regions, half encompass immune-function genes: TLR3 at 4q35.1, CD83 at 6p23, the extended HLA at 6p22.1-21.32, and RAPGEF5-IL6 at 7p15.3. The CD83 region is associated with rheumatoid arthritis, the HLA region with all AIDs, and the RAPGEF5-IL6 region with dialysis-related mortality in AA and CRP levels. Conclusion: We have identified several regions that harbor immune function genes and differentiate the Gullah from the Sierra Leoneans, suggesting that recent selection may be operating at these loci. Given the relative homogeneity of the Gullah and their genetic proximity to Africans from Sierra Leone, identification of regions that might be under selection and harbor immune-related genes in the Gullah has the potential to elucidate AID risks in AA.

Sunday, October 27

156

157

Cartilage Peptide Profiling: Identifying Novel OA Biomarkers For Early Disease Detection. Valentina Calamia, Lucıá Lourido, Diego Lo´pez, Jesus Mateos, Patricia Fernańdez-Puente, Carolina Fernańdez-Costa, Beatriz Rocha, Cristina Ruiz-Romero and Francisco J. Blanco. Osteoarticular and Aging Research Laboratory, Proteomics Unit-Associated Node to ProteoRedISCIII, INIBIC-CHUAC, A Corunã, Spain.

Application Of a Multiplex Gene Polymorphism Assay For Variants Associated With Rheumatoid Arthritis Susceptibility: Results Of 168 Single Nucleotide Polymorphisms In The Optima Study. Jeffrey F. Waring1, Viswanath Devanarayan2, Kenneth Idler1, Feng Hong2, Josef S. Smolen3, Arthur Kavanaugh4, Hartmut Kupper5, Hendrik Schulze-Koops6 and Alla Skapenko6. 1AbbVie Inc., North Chicago, IL, 2AbbVie Bioresearch Center, Worchester, MA, 3Medical University of Vienna and Hietzing Hospital, Vienna, Austria, 4University of California San Diego, La Jolla, CA, 5 AbbVie Deutschland GmbH & Co. KG, Ludwigshafen, Germany, 6University of Munich, Munich, Germany.

Background/Purpose: Cartilage secretome contains a complex matrix of proteolytically derived peptides that may provide useful information of the biological events occurring in the articular joint in normal and pathological condition. Moreover, the low-molecular-weight subset of the cartilage proteome may have a diagnostic potential in osteoarthritis (OA) research, not fully exploited yet. The aim of this study is to investigate cartilage secretome by means of peptidome analysis and to provide a novel source for OA biomarker discovery. Methods: Tissue explants were obtained from the dissection of human OA cartilages, both from the wounded zone (WZ) and the unwounded zone (UZ). The study was approved by the local ethical committee. Cartilage shavings from each zone were cut into 6 mm discs and five discs/zone were placed into 96 wells plates and incubated (37 °C/5% CO2) in serum-free DMEM up to 6 days. Conditioned media from each condition were collected at day 1, 3 and 6. Selective extraction of endogenous peptides was realized by combining ultrafiltration and solid phase extraction. Then, the peptides were directly analyzed by nano reversed-phase liquid chromatography (RP-LC) in an Easy nLC II system coupled to an ion trap LTQ Orbitrap Velos Pro mass spectrometer (Thermo Scientific) without trypsin digestion. Protein/peptide identifications were searched against human database using the SEQUEST algorithm (Proteome Discoverer 1.3, Thermo Scientific). Results: Firstly we compared the profiles of peptides released to the media at different times (0, 3, and 6 days in culture). At day 3 we detected the highest number of unique peptides and proteins, and the lowest serum contamination. Thus, we selected day 3 as the best point for the next proteomic analysis in which we compared peptide profiles from WZ and UZ. This study led to the detection of a panel of 262 peptides corresponding to 36 proteins that were differentially released from the WZ and UZ in OA cartilage. Most of them were cartilage ECM proteins or proteins with well-established matrix functions, such as collagens and proteoglycans, thus indicating the high cartilage-specific biomarker concentration present in our samples. 18 peptides were detected only in the WZ (corresponding to 5 proteins: K1C9, K1C10, PRPC, FIBA, and DCD) while 60 peptides only in the UZ (corresponding to 20 proteins: ACAN, CILP1, CILP2, COL11A2, COL1A2, COL25A1, LTBP2, MGP, and PRELP among others) (Table 1). 11 proteins (BGN, CLU, COL1A1, COL2A1, COL3A1, COL5A2, COMP, and FN1 among others) were detected in both zones but in all the cases the number of peptides was higher in the UZ than in the WZ.

Background/Purpose: Genetic factors have been identified that may be associated with the development and severity of rheumatoid arthritis (RA), disease progression, or response to treatment. In-depth knowledge about RA susceptibility genes may thus prove useful for the development of future diagnostic tests or personalized therapeutics. Frequencies from previously investigated RA susceptibility gene polymorphisms from patients enrolled in the OPTIMA trial were compared to a control database to identify genetic risk factors for RA. Methods: OPTIMA was a 78-week, multicenter, randomized study designed to assess adalimumab plus methotrexate therapy in early RA patients.(1) Of 1032 patients, genetic analysis was conducted on 921 patients who signed an additional informed consent for genetic analyses. Frequency rates for 168 identified RA susceptibility alleles from 88 genes were compared to the European cohort of the 1000 Genomes Project as a reference control.(2) Genotypic variants were assayed using the Illumina BeadXpress GoldenGate Assay. The largest deviations in the OPTIMA minor allele frequencies (MAFs), those ⬎ 0.05 compared to the control, were further investigated. Logistic regression was used to calculate odds ratios (OR) for RA susceptibility. Association of alleles with 28-joint count disease activity score (DAS28) and age were conducted using ANOVA comparing the OPTIMA cohort with the control cohort. Results: The majority of single nucleotide polymorphism (SNP) MAFs from the OPTIMA study did not deviate ⬎ 0.05 from MAFs reported in the European cohort from the 1000 Genomes Project. Within known RA susceptibility genes or those with treatment response, 7 of 63 SNPs previously investigated as being associated with increased RA risk deviated ⬎ 0.05 from reference controls in OPTIMA patients and were investigated further; 11 SNPs that had not been previously associated with RA susceptibility had MAF differences ⬎ 0.05 comparing OPTIMA with the reference control. The BTLA SNP rs9288952 G allele displayed the highest OR for RA risk, 3.60 (95% CI, 2.49 – 5.20), was weakly associated with higher baseline DAS28, and also associated with patient age (mean ages were 50.9, 49.0, and 44.3 for genotypes A/A, A/G, and G/G, respectively), suggesting a potential association with onset at earlier or later age. STAT4 rs7574865 minor allele homozygosity was associated with higher baseline DAS28 scores. Of the previously unidentified RA susceptibility gene SNPs, only PDZD2 rs1532269 was significantly associated with higher baseline DAS28 scores.

Table 1. Examples of peptides detected at high levels (High) in UZ samples and not detected (ND) in WZ samples, potentially useful as early OA biomarkers. Protein name Cartilage intermediate layer protein 1 (CILP1)

Matrix Gla protein (MGP)

Accession no.

Sequence

MHⴙ [Da]

UZ

WZ

O75339

ETNIIPLGEV

1084,58911

High

ND

P08493

DWTPAGSTGQV WTPAGSTGQV SLPGGAPASGAA NANTFISPQQR NANTFISPQ NPFINR VYGYNA

1118,51025 1003,48511 955,48291 1275,64385 991,48334 760,40991 686,31451

High High High High High High High

ND ND ND ND ND ND ND

Gene Previously Investigated RA BTLA Susceptibility Gene SNPs FNDC1/TAGAP HLA-DPB1/HLA-DPB2 HLA-DQB1/HLA-DQA2 IL23R/IL12RB2 PTPN22 STAT4 Previously Unidentified FAM167A/BLK RA Susceptibility Gene HLA-DPB1 SNPs NOS3 PDZD2 PTPN22 PTPRC PTPRC STAT4 STAT4 STAT4 STAT4

Conclusion: To our knowledge this is the first proteomic study specifically focused on cartilage peptidome analysis. In this work we analysed the cartilage endogenous peptides profile and highlighted its potential application as novel OA biomarker source for early disease detection.

SNP rs9288952 rs394581 rs3117213 rs6457617 rs1495965 rs2476601 rs7574865 rs13277113 rs3135021 rs2070744 rs1532269 rs2488458 rs1326279 rs12144388 rs12463658 rs3024877 rs1031509 rs11693480

1000 Difference Genomes OPTIMA Between MAF MAF MAFs 0.04 0.32 0.26 0.45 0.49 0.01 0.23 0.33 0.36 0.33 0.44 0.34 0.38 0.37 0.47 0.37 0.20 0.46

0.15 0.26 0.18 0.33 0.42 0.13 0.29 0.25 0.28 0.41 0.37 0.27 0.31 0.31 0.40 0.30 0.27 0.40

0.11 0.06 0.08 0.12 0.07 0.12 0.06 0.08 0.08 0.07 0.07 0.07 0.07 0.06 0.07 0.07 0.07 0.06

Odds Ratio (95% CI) 3.60 (2.49–5.20)*** 0.75 (0.62–0.90)** 0.62 (0.51–0.76)*** 0.62 (0.52–0.73)** 0.76 (0.64–0.90)*** 1.31 (0.99–1.72)# 1.37 (1.13–1.67)** 1.49 (1.23–1.81)*** 1.48 (1.23–1.78)*** 0.73 (0.61–0.87)*** 1.35 (1.14–1.61)*** 1.41 (1.17–1.70)*** 1.36 (1.13–1.63)*** 1.31 (1.09–1.57)** 1.35 (1.14–1.60)*** 1.36 (1.14–1.64)*** 0.70 (0.57–0.85)*** 1.29 (1.08–1.53)**

MAF, minor allele frequency; SNP, single nucleotide polymorphism. ** and *** denote statistical significance at the p ⬍ 0.01 and 0.001 levels, respectively. #significant when homozygous for the minor allele.

Disclosure: V. Calamia, None; L. Lourido, None; D. Lo´pez, None; J. Mateos, None; P. Fernańdez-Puente, None; C. Fernańdez-Costa, None; B. Rocha, None; C. Ruiz-Romero, None; F. J. Blanco, None.

Conclusion: Genes identified with increased RA risk have known roles in immune responses including mediation of lymphocyte signaling,

S62

Methods: We analyzed whole blood RNA (PAXgene RNA tubes) from 38 patients with sarcoidosis and 20 healthy controls using Affymetrix U133 Plus 2.0 microarrays (Affymetrix Inc., Santa Clara, CA). Correcting with the Benjamini-Hochberg method, we found differentially expressed (DE) genes that were upregulated in sarcoidosis compared to controls. We performed Ingenuity Pathway Analysis (IPA) and then hierarchical clustering (MultiExperiment Viewer) using selected genes representative of canonical pathways highly up-regulated in sarcoid. Standardized expression levels of DE genes were correlated with measures of pulmonary function testing (PFT). Whole blood DE genes expression levels were compared to levels found in publically available sarcoidosis tissue gene expression datasets (Gene Expression Omnibus GSE32887, GSE19976). Results: 193 DE genes associated with a B-statistic ⬎ 0 (adjusted p value ⬍0.006) were upregulated in sarcoidosis. These genes were associated with interferon (IFN), TNF␣, and TGF-␤1 signaling pathways. Hierarchial clustering of subjects using genes representative of IFN and TNF␣ pathways revealed different patient subsets, including subjects with IFN-high/TNF␣-low or IFN-low/TNF␣-high signatures, and subjects with dual IFN/TNF␣-high or dual IFN/TNF␣ low signatures. Expression levels of TGF-␤1-associated genes (SKIL, USP15, LPCAT2) correlated with PFT measures of restrictive lung volumes and decreased oxygenation, including FVC% predicted (R2 ⫽ 0.39, p⬍0.0001), TLC% predicted (R2 ⫽ 0.26, p ⫽ 0.001), and DLCO% predicted (R2 ⫽ 0.2, p ⫽ 0.004). Expression levels of TNF␣-inducible genes (TNFAIP2, PLAUR, ICAM1) correlated with degree of pulmonary obstruction as measured by FEV1/ FVC (R2⫽ 0.26, p ⫽ 0.001). Expression of these pathway-specific genes was elevated in lesional sarcoid skin and progressive fibrotic sarcoid lung biopsy gene expression datasets. Conclusion: Based on these data we hypothesize that IFN-, TNF␣-, and TGF-␤1-specific gene expression biomarkers reflect distinct subgroups of patients with sarcoidosis and signaling pathways that may play a pathogenic role in sarcoidosis. TGF-␤1-associated DE gene expression levels may reflect severity of pulmonary fibrosis. We plan to test this hypothesis in a larger longitudinal cohort to 1) validate our groupings, 2) determine whether these signatures are stable over time or vary with disease activity, 3) determine whether these groupings have prognostic value.

Disclosure: J. F. Waring, AbbVie Inc., 1, AbbVie Inc., 3; V. Devanarayan, AbbVie Inc., 1, AbbVie Inc., 3; K. Idler, AbbVie Inc., 1, AbbVie Inc., 3; F. Hong, AbbVie Inc., 1, AbbVie Inc., 3; J. S. Smolen, AbbVie Inc., Amgen, AstraZeneca, BMS, Celgene, Centocor-Janssen, GSK, Eli Lilly, Pfizer, MSD, Novo-Nordisk, Roche, Sandoz, UCB, 2, AbbVie Inc., Amgen, AstraZeneca, BMS, Celgene, Centocor-Janssen, GSK, Eli Lilly, Pfizer, MSD, Novo-Nordisk, Roche, Sandoz, UCB, 5; A. Kavanaugh, AbbVie Inc., Amgen, Astra-Zeneca, BMS, Celgene, Centocor-Janssen, Pfizer, Roche, and UCB, 2, AbbVie Inc., Amgen, Astra-Zeneca, BMS, Celgene, Centocor-Janssen, Pfizer, Roche, and UCB, 5; H. Kupper, AbbVie Inc., 1, AbbVie Inc., 3; H. Schulze-Koops, AbbVie Inc., 5; A. Skapenko, None.

158 The Use Of “Autoantigenomics” To Rapidly Identify Targets Of Human Autoantibodies. Wei-Hong Yang, Emily G. Bloch, Daniel F. Berenson, Rita L. Galdos, Pankaj Arora, Connie Wu and Donald B. Bloch. Massachusetts General Hospital and Harvard Medical School, Boston, MA. Background/Purpose: The messenger RNA (mRNA) processing body (P-body) is a cellular structure that regulates the stability of cytoplasmic mRNA. Approximately 5% of patients with the autoimmune disease primary biliary cirrhosis (PBC) have antibodies directed against this structure and some patients have antibodies that react with several different known P-body components. In this study, we used a proteomic array and serum from a PBC patient to rapidly and efficiently identify a new P-body component. Methods: Serum from patient 0081 was used to probe a high density protein macroarray which contains approximately 17,000 proteins. The proteins were produced in situ on a polyvinylidene difluoride (PVDF) membrane by Escherichia coli transformed with a prokaryotic expression cDNA library. This library was derived from mRNA prepared from phytohemagglutinin treated human T-lymphocytes. Human auto-antibodies that bound to the protein macroarray were detected using horse-radish peroxidaseconjugated rabbit anti-human IgG antiserum and chemiluminescence. Immunoreactive protein targets of these autoantibodies were determined by their locations on the PVDF membrane. Immunoblot was then used to confirm that serum 0081 contained antibodies that reacted with newly indentified autoantigens. Human epidermoid cancer cells (HEp-2 cells) were transfected with plasmids encoding green fluorescent protein (GFP) fused to the new P-body component candidates, to confirm that the proteins localized to P-bodies. Results: Serum 0081 reacted with fifty-six proteins on the protein macroarray membrane. One of these new proteins, Limkain B (LMKB), was chosen for further study because it was previously reported to localize to a “subset of peroxisomes”, a staining pattern that might appear similar to the P-body pattern. Antibodies in serum 0081 reacted with glutathione S-transferase (GST)-LMKB, but not GST alone, by immunoblot. A plasmid encoding full-length LMKB fused to GFP was used to express the protein in HEp-2 cells. LMKB localized to cytoplasmic dots and co-localized with Ge-1, a known P-body component. Conclusion: Our findings demonstrate that the combination of proteomic array technology and human autoantibodies provides a useful tool for identifying new autoantigens. “Autoantigenomics” can be used to rapidly and efficiently discover new proteins in cellular structures. Using this method, we identified LMKB as a novel component of mRNA processing bodies.

Disclosure: R. Su, None; M. Li, None; N. Bhakta, None; M. Agarwal, None; P. Woodruff, None; L. Koth, None.

160 Association Analysis Of The Organic Anion Transporter 4 and Urate Transporter 1 Locus With Gout In New Zealand Case-Control Sample Sets Reveals Multiple Ancestral-Specific Effects. Tony R. Merriman1, Amanda Phipps-Green1, Jade E. Hollis-Moffatt1, Marilyn E. Merriman1, Ruth Topless1, Grant Montgomery2, Brett Chapman2, Lisa K. Stamp3, Nicola Dalbeth4 and Tanya Flynn1. 1University of Otago, Dunedin, New Zealand, 2Queensland Institute of Medical Research, Brisbane, Australia, 3University of Otago, Christchurch, Christchurch, New Zealand, 4University of Auckland, Auckland, New Zealand. Background/Purpose: There are genetic variants in urate transporters SLC22A11 (OAT4) and SLC22A12 (URAT1) that influence serum urate levels in European Caucasian. However, there is no consistent evidence for association with risk of gout. The aim of this study was to test genetic variation across the SLC22A11 and SLC22A12 locus for association with gout risk in New Zealand sample sets. The sample sets include M␤ori and Pacific Island (Polynesian) participants, who exhibit higher prevalence and more severe gout than European Caucasian. Methods: Twelve single nucleotide polymorphism (SNP) variants in four haplotype blocks were genotyped using TaqMan¨ and Sequenom MassArray in 1003 gout cases and 1156 controls. Gout was classified according to the 1977 American Rheumatism Association criteria. Association analysis of single markers and haplotypes was done using PLINK and STATA. Results: SNP rs17299124 (upstream of SLC22A11) in haplotype block 1 was associated with gout in the Polynesian sample sets with higher Polynesian ancestry, but not in Maori and Pacific with lower Polynesian ancestry and European Caucasian (OR⫽3.38, P⫽0.001; OR⫽0.94, P⫽0.54, respectively) (Table). A protective block 1 haplotype, rarer in Polynesian, was driving the association (OR⫽0.28, P⫽0.001). Within haplotype block 2 (SLC22A11) we could not replicate previous reports of association of rs2078267 with gout in European Caucasian and people

Disclosure: W. H. Yang, None; E. G. Bloch, None; D. F. Berenson, None; R. L. Galdos, None; P. Arora, None; C. Wu, None; D. B. Bloch, None.

159 Transcriptomic Blood Markers and Molecular Stratification In Sarcoidosis. Robert Su, Michael Li, Nirav Bhakta, Misha Agarwal, Prescott Woodruff and Laura Koth. University of California San Francisco, San Francisco, CA. Background/Purpose: Sarcoidosis is a heterogeneous disease with diverse manifestations and varied clinical courses. Currently there are no reliable non-invasive biomarkers of disease activity or severity. The aim of this study was to identify “pathway-specific” blood gene expression signatures in subjects with sarcoidosis and to ascertain whether these signatures identify distinct subsets of patients and correlate with measures of disease severity.

S63

Sunday, October 27

proliferation, and differentiation. The MAFs of 18 gene SNPs in the OPTIMA study differed from a healthy control population; further investigation into these SNPs will clarify their role in RA susceptibility and disease progression.

Sunday, October 27

with lower Polynesian ancestry (OR⫽1.00, P⫽1.00), however this SNP was associated with gout in people with higher Polynesian ancestry (OR⫽1.84, P⫽0.012). Within haplotype block 3 (including SLC22A12) there was evidence for association for SNP rs3825018 in all ancestral groups (OR⫽1.27, P⫽0.002). Analysis of haplotypes revealed a haplotype with trans-ancestral protective effects (OR⫽0.80, P⫽0.004), and a second haplotype conferring protection in the less admixed Polynesian sample sets (OR⫽0.63, P⫽0.028) but risk in European Caucasian samples (OR⫽1.33, P⫽0.039).

162 Genotypic and Haplotypic Effects Of 7 Single-Nucleotide Polymorphisms In the CRP Gene On Levels Of C-Reactive Protein and DAS28 In a Cohort Of 180 Untreated Newly Diagnosed Rheumatoid Arthritis Patients (OPERA Study). Christian G. Ammitzbøll1, Rudi Steffensen2, Peter Junker3, Mikkel Østergaard4, Julia Johansen5, Jan Pødenphant6, Merete Lund Hetland7, Hanne M. Lindegaard8, Torkell Ellingsen9 and Kristian Stengaard-Pedersen1. 1Arhus University Hospital, Aarhus, Denmark, 2Aalborg University Hospital, Aalborg, Denmark, 3University of Southern Denmark, Odense, Denmark, 4Copenhagen University Hospital Glostrup, Copenhagen, Denmark, 5Copenhagen University and Glostrup Hospital, Copenhagen, Denmark, 6Copenhagen University at Gentofte, Hellerup, Denmark, 7The Danish Rheumatologic Database (DANBIO), Glostrup Hospital, Copenhagen, Denmark, 8Odense University Hospital, Odense, Denmark, 9Diagnostic Centre Region Hospital Silkeborg Denmark, 8600 Silkeborg, Denmark.

Table. Allelic associations SNP (haplotype block) rs475414 (1) rs17299124 (1) rs693591 (2) rs17300741 (2) rs2078267 (2) rs3825018 (3) rs475688 (3) rs7932775 (3) rs476037 (3) rs478607 (3) rs12289836 (4) rs642803 (4)

European Caucasian and low ancestry Polynesian OR [95% CI] P Het P1 1.02 0.94 0.99 1.08 1.00 1.24 1.31 1.32 1.05 1.33 0.87 0.97

[0.85–1.22] [0.76–1.16] [0.78–1.26] [0.90–1.29] [0.84–1.20] [1.02–1.50] [1.06–1.60] [1.05–1.66] [0.78–1.40] [1.05–1.70] [0.72–1.05] [0.81–1.16]

0.86 0.54 0.94 0.40 1.00 0.028 0.011 0.017 0.77 0.020 0.16 0.73

0.99 0.43 0.56 0.45 0.50 0.66 0.36 0.92 0.87 0.37 0.44 0.27

High ancestry Polynesian OR [95% CI] P Het P1 1.15 3.38 1.25 1.04 1.84 1.33 0.95 1.21 1.25 1.04 0.89 0.95

[0.92–1.43] [1.69–6.79] [1.00–1.57] [0.78–1.38] [1.14–2.95] [1.05–1.69] [0.76–1.20] [0.97–1.51] [0.67–2.34] [0.79–1.36] [0.70–1.13] [0.76–1.19]

0.23 0.001 0.051 0.79 0.012 0.020 0.68 0.095 0.48 0.79 0.33 0.66

0.73 0.67 0.053 0.17 0.80 0.57 0.22 0.26 0.014 0.15 0.72 0.86

All ancestries OR [95% CI] P 1.07 1.71 1.12 1.07 1.08 1.27 1.13 1.26 1.08 1.19 0.88 0.96

[0.93–1.23] [0.89–3.28] [0.95–1.33] [0.92–1.24] [0.91–1.28] [1.10–1.48] [0.97–1.32] [1.08–1.48] [0.89–1.31] [1.00–1.43] [0.76–1.02] [0.84–1.11]

0.38 0.11 0.17 0.40 0.37 0.002 0.11 0.004 0.42 0.056 0.085 0.59

Het P1 0.86 0.010 0.087 0.39 0.17 0.82 0.093 0.55 0.070 0.17 0.86 0.82

Background/Purpose: Single nucleotide polymorphisms (SNPs) in the CRP gene are implicated in the regulation of the basal C-reactive protein (CRP) expression and its response to pro-inflammatory stimuli. Previous reports suggest these effects may have an impact on clinical decision-making based on CRP, e.g. DAS28 (1). We aimed to investigate for the first time the possible association between 7 SNPs in the CRP and the serum level of CRP/ DAS28 in a cohort of 180 untreated inflammatory active early RA patients Methods: 180 DMARD naı¨ve RA patients with disease duration ⬍6 months were included in a randomized double blind placebo-controlled trial (OPERA-study,NCT00660647) of methotrexate, intraarticular glucucorticoids ⫹ either adalimumab or placebo. SNPs were analyzed by the TaqMan OpenArray system. The 7 SNPs (Table 2) were selected based on previously reported effects on CRP levels(1). CRP was measured using CRP QUICK-READ (range 8–160 mg/l). The associations between SNPs (and haplotypes of SNPs) and CRP and DAS28 levels were evaluated using linear regression analysis adjusted for age, sex and treatment. For the analysis of genotypic and haplotypic effects, the common allele homozygous genotype/haplotype was selected as reference, and the effects are presented as percentage. ‘Haplo.stats’ package for R was used Results: Baseline characteristics were similar in the two groups, Table 1. There were no significant genotypic or haplotypic effects of the 7 SNPs on CRP levels at baseline or one year (Pⱖ0.080). Homozygosity for the minor allele of rs2808632 reduced borderline significant the baseline DAS28 levels to 54% P⫽0.055, and heterozygosity for rs1800947 increased DAS28 levels at one year to 158% P⫽0.03. Six haplotypes were constructed encompassing 94% of the cohort, Table 3. The H4 haplotype reduced baseline DAS28 score to 51% P⫽0.009, and the H6 haplotype increased the DAS28 score at one year to 168% P⫽0.02. No further haplotypic effect on DAS28 were observed at baseline or one year.

1Heterogeneity P-value, P ⬍ 0.05 indicates significant heterogeneity.

Conclusion: Our analysis demonstrates several ancestral-specific effects across the SLC22A11/SLC22A12 locus indicating that multiple common variants influence the activity of both OAT4 and URAT1 in relation to risk of gout. These results clarify some of the previous inconsistent gout associations seen, and indicate that further fine mapping of the association signal is needed. Disclosure: T. R. Merriman, None; A. Phipps-Green, None; J. E. HollisMoffatt, None; M. E. Merriman, None; R. Topless, None; G. Montgomery, None; B. Chapman, None; L. K. Stamp, None; N. Dalbeth, None; T. Flynn, None.

161 H3K9 Demethylation By LSD1 Contributes To IL-1-Induced mPGES-1 Expression In OA Chondrocytes. Fatima Ezzahra El Mansouri and Hassan Fahmi. Osteoarthritis Research Unit, University of Montreal Hospital Research Centre (CRCHUM), Notre-Dame Hospital, Montreal, QC. Background/Purpose: Microsomal prostaglandin E synthase-1 catalyzes the terminal step in the biosynthesis of PGE2, which plays a critical role in the pathophysiology of osteoarthritis. To investigate the role of histone H3 (H3K9) methylation in interleukin-1b (IL-1)-induced microsomal prostaglandin E synthase-1 (mPGES-1) expression in human osteoarthritic (OA) chondrocytes. Methods: Chondrocytes were stimulated with IL-1 and the expression of mPGES-1mRNA was analyzed using real-time reverse transcriptasepolymerase chain reaction. H3K9 methylation and the recruitment of the histone demethylase LSD1 to the mPGES-1 promoter were evaluated using chromatin immunoprecipitation assays. The role of LSD1 was further evaluated using the the amino oxidase inhibitor tranylcypromine (a potent inhibitor of LSD1 activity). Results: Treatment with IL-1 induced mPGES-1 expression in a time dependent manner. The induction of mPGES-1 expression by IL-1 was associated with H3K9 demethylation at the mPGES-1 promoter. These changes were concomitant with the recruitment of the histone demethylase LSD1. Treatment with tranylcypromine inhibited IL-1-induced H3K9 demethylation as well as IL-1-induced mPGES-1 expression. Conclusion: These results indicate that H3K9 demethylation by LSD1 contributes to IL-1-induced mPGES-1 expression and suggest that this pathway could be a potential target for pharmacological intervention in the treatment of OA and possibly other arthritic diseases.

Table 1. Patient characteristics

Patient characteristics Female sex Age, years Disease duration, days Anti-CCP positive IgM-RF positive DAS28 C-reactive protein, mg/l Tender joint count(28) Swollen joint count(28) VAS-patient global, mm Baseline x-ray erosions (ESⱖ1) Disease activity, 1 year DAS28 C-reactive protein, mg/l

OPERA Placebo treated group (nⴝ91)

OPERA Adalimumab treated group (pⴝ89)

P value

69% 54 (28–77) 83 (42–150) 70% 74% 5.6 (3.8–7.3) 15 (7–109) 11 (3–24) 8 (2–22) 65 (17–96) 52%

63% 56 (26–78) 88 (42–162) 60% 70% 5.5 (3.8–7.8) 15 (7–133) 10 (3–27) 8 (2–26) 70 (12–100) 54%

0.46 0.71 0.74 0.17 0.67 0.53 0.54 0.78 0.66 0.27 0.94

2.6 (1.7–4.7) 7 (7–44)

2.0 (1.7–5.2) 7 (7–21)

0.009 0.21

Values are medians with 5–95% percentile values in parentheses, unless otherwise stated. Anti-CCP ⫽ anti-cyclic citrullinated peptide, RF ⫽ rheumatoid factor, DAS28 ⫽ disease activity score 28 joints, VAS ⫽ visual analogue scale, ES ⫽ Sharp/van der Heijde Erosion Score.

Disclosure: F. E. El Mansouri, None; H. Fahmi, None.

S64

CRP levels relative to major genotype in % Rs-number Genotype % rs11265257

rs1130864

rs1205

CC

rs2808632

rs876538

⫺

Year one 100 (ref.)

P value ⫺

DAS28 levels relative to major genotype in % Baseline 100 (ref.)

P value ⫺

Year one

P value ⫺

100 (ref.)

46 103 (75–141) P ⫽ 0.87

TT

16

GG

49 100 (ref.)

AG

41

92 (67–125) P ⫽ 0.58

97 (84–113) P ⫽ 0.73 119 (84–167) P ⫽ 0.32 109 (84–142)

P ⫽ 0.51

AA

10 119 (72–199) P ⫽ 0.50

88 (69–112) P ⫽ 0.30 117 (67–207) P ⫽ 0.58 109 (71–167)

P ⫽ 0.71

99 (85–115) P ⫽ 0.88

95 (67–136) P ⫽ 0.78

99 (75–130)

P ⫽ 0.92

79 (51–122) P ⫽ 0.29 105 (85–129) P ⫽ 0.68

85 (52–139) P ⫽ 0.52

98 (68–141)

P ⫽ 0.90

⫺

⫺

CC

46 100 (ref.)

CT

44 105 (77–143) P ⫽ 0.75 10

100 (ref.)

100 (ref.)

⫺

89 100 (ref.) 11

TT

46 100 (ref.)

GT

46 115 (85–155) P ⫽ 0.37

⫺

100 (ref.)

⫺

⫺

⫺

100 (ref.)

⫺

100 (ref.)

P ⫽ 0.28

99 (81–123) P ⫽ 0.96 103 (59–182) P ⫽ 0.91 102 (66–157)

P ⫽ 0.93

⫺

100 (ref.)

⫺

⫺

100 (ref.)

96 (75–122) P ⫽ 0.73 127 (76–213) P ⫽ 0.37 158 (103–241) P ⫽ 0.03 100 (ref.)

⫺

100 (ref.)

⫺

⫺

100 (ref.)

97 (84–112) P ⫽ 0.65 113 (81–158) P ⫽ 0.46 104 (81–135)

8 113 (64–200) P ⫽ 0.67 117 (89–154) P ⫽ 0.26 ⫺

100 (ref.)

86 (66–113)

100 (ref.)

88 (55–140) P ⫽ 0.59 ⫺

⫺

89 (78–101) P ⫽ 0.08 101 (71–142) P ⫽ 0.97

69 (42–114) P ⫽ 0.15

CC CG

GG rs3093077

38 100 (ref.)

P value

CT

TT rs1800947

Baseline

⫺

⫺

11

CC

60 100 (ref.)

CT

37 128 (94–174) P ⫽ 0.12 101 (87–117) P ⫽ 0.93 131 (93–184) P ⫽ 0.12 111 (85–144)

⫺

100 (ref.)

3 124 (55–281) P ⫽ 0.60

⫺

85 (56–129) P ⫽ 0.44

100 (ref.)

80 (47–136) P ⫽ 0.41 100 (ref.)

⫺

P ⫽ 0.75 P ⫽ 0.98 ⫺

89 100 (ref.)

AC

94 (58–151) P ⫽ 0.79 110 (88–137) P ⫽ 0.41

100 (ref.)

99 (61–161)

AA

TT

100 (ref.)

54 (29–101) P ⫽ 0.06

80 (54–118)

P ⫽ 0.26 ⫺

100 (ref.)

57 (23–140) P ⫽ 0.22

79 (38–165)

P ⫽ 0.45 P ⫽ 0.53

Numbers in parantheses are 95% confidence intervals. Calculated by linear regression, corrected for treatment (placebo/ adalimumab), age and gender. ref.⫽reference

Table 3. Linear regression analyses of the haplotype effect on the mean CRP and DAS28 levels relative to the major haplotype. CRP levels relative to major genotype in %

DAS28 levels relative to major genotype in %

Haplotype

Frequency

SNPs#

Baseline

Year one

Baseline

H1

30 %

CACCTAC

100 (ref.)

100 (ref.)

100 (ref.)

100 (ref.)

Year one

H2

26 %

TGTCTAC

93 (72–121) P ⫽ 0.58

93 (72–121) P ⫽ 0.58

91 (68–121) P ⫽ 0.51

83 (67–103) P ⫽ 0.10

H3

21 %

CGCCGAT

115 (86–154) P ⫽ 0.35

115 (86–154) P ⫽ 0.35

95 (69–131) P ⫽ 0.77

99 (77–126) P ⫽ 0.93

H4

7%

TGCCGAC

80 (51–126) P ⫽ 0.33

80 (51–126) P ⫽ 0.33

H5

5%

CGCCTCC

96 (59–156) P ⫽ 0.87

96 (59–156) P ⫽ 0.87

78 (46–133) P ⫽ 0.36

75 (51–111) P ⫽ 0.15

H6

5%

TGTGTAC

93 (56–153) P ⫽ 0.76

93 (56–153) P ⫽ 0.76

115 (66–200) P ⫽ 0.62

168 (108–261) P ⫽ 0.02

51 (31–84) P ⫽ 0.009

115 (80–165) P ⫽ 0.45

Disclosure: C. N. Son, None; S. Y. Bang, None; S. C. Bae, None; J. B. Jun, None.

(#) The 7 SNPs defining the 6 haplotypes are listed as follows (rs11265257, rs1130864, rs1205, rs1800947, rs2808632, rs3093077, rs876538). ref.⫽reference

164 Conclusion: Seven selected CRP gene SNPs had no impact on preand one year post-treatment levels of CRP. Minor genotypic and haplotypic effects on DAS28 scores were observed, but these were not consistent between baseline and one year. This study shows that DAS28 can be used for clinical decision-making without adjustment for CRP gene variants.

Haptoglobin Chains As Potential Biomarkers In Serum Of Osteoarthritis Disease. Carolina Fernandez-Costa1, Valentina Calamia1, Patricia Fernandez-Puente1, Jesus Mateos1, Beatriz Rocha1, Lucia Lourido1, Jose Luis Capelo2, Cristina Ruiz-Romero3 and Francisco J. Blanco1. 1Osteoarticular and Aging Research Laboratory, Proteomics Unit-Associated Node to ProteoRed-ISCIII, INIBIC-CHUAC, A Corunã, Spain, 2BIOSCOPE Group, REQUIMTE, Departamento de Quı´mica, Faculdade de Cieˆncias e Tecnologia, FCT, Universidade Nova de Lisboa,, Caparica, Portugal, 3CIBER-BBN, INIBIC-CHUAC, A Corunã, Spain.

(1) Nat Rev Rheum 2011;7(5):282–9. Disclosure: C. G. Ammitzbøll, None; R. Steffensen, None; P. Junker, None; M. Østergaard, None; J. Johansen, None; J. Pødenphant, None; M. L. Hetland, None; H. M. Lindegaard, None; T. Ellingsen, None; K. Stengaard-Pedersen, None.

Background/Purpose: The aim of this study is to search for osteoarthritis (OA) biomarkers in human serum using an easy and fast approach and the validation of the potential biomarkers found with two independent methods. Methods: The serum samples for the biomarker discovery experiment were obtained from 20 OA patients and 20 non-symptomatic controls. Samples were grouped into four pools of 10 samples each. The pools were subjected to a chemical sequential depletion protocol to reduce the high dynamic range of the proteins. Then, the proteomics comparison between OA and control sera was performed across two dimensional difference in-gel electrophoresis (2D-DIGE) experiment. The quantitative image analysis was performed using Same Spots software. For protein identification, gel spots were digested and analyzed by mass spectrometry (MALDI-TOF/TOF) and identified using Mascot with SwissProt knowledgebase. Haptoglobin chains validation was performed with two different methods, immunoblotting and multiple reaction monitoring (MRM) technology with the QTRAP mass spectrometer. Results: We studied the combination of a chemical sequential depletion method combined with 2D-DIGE for the search of OA biomarkers in 40 serum samples. The analysis resulted in 46 spots significantly and reproducibly altered between OA and control samples (29 increased and 17 decreased). These 46 spots correspond to 14 different proteins, Table 1. The most interesting result was the modulation of the protein haptoglobin (HPT), three different spots were opposite modulated and were identify as the three chains of haptoglobin. In human, HPT exists in two allelic forms that produce three known phenotypes. The presence of the different alpha chains in patients depends on which phenotype their express. Interestingly, HPT beta and HPT alpha-1 chains were increased in OA sera whereas HPT alpha-2 chain was decreased in OA sera versus control. Using western blot analysis and MRM mass spectrometry technique on 30 new individual samples from OA and control donors (15

163 The Frequency Of Single Nucleotide Polymorphisms In Urate Transporter Genes and Their Association With Uric Acid Concentration Based On Data From Genome-Wide Association Studies In The Korean Population. Chan-Nam Son1, So-Young Bang2, Sang-Cheol Bae1 and Jae-Bum Jun1. 1Hanyang University Hospital for Rheumatic Diseases, Seoul, South Korea, 2Hanyang University Guri Hospital, Guri, South Korea. Background/Purpose: Gouty arthritis is characterized by hyperuricemia, which results from overproduction of, or impaired renal excretion of, uric acid. Recently, interest has increased in renal urate transporters, which control the excretion of uric acid. Genome-wide association studies (GWASs) are now identifying risk alleles among renal urate transporter genes. At present, few studies of factors affecting uric acid regulation have been conducted on Koreans. We therefore aimed, first, to investigate the minor allele frequencies (MAFs) of single nucleotide polymorphisms (SNPs) associated with serum uric acid (SUA) level in the Korean population, and compare these with data from other ethnic groups (Study 1). Second, to investigate whether the SNPs are associated with altered SUA levels (Study 2). Methods: Study 1: We used datasets derived from two available GWASs. The first study was the Korean RA GWAS, including 800 rheumatoid arthritis (RA) cases and 757 controls, and the second was Korean systemic lupus erythematosus (SLE) GWAS, comprising 400 SLE cases (a total of 1957 subjects). We explored the GWAS results already obtained from subjects without gout to examine the frequencies of risk alleles, and investigated the MAFs of 40 previously described SNPs associated with SUA level in the Korean population, and compared results

S65

Sunday, October 27

with data for other ethnic groups. Study 2: A total of 402 RA subjects satisfying the inclusion and exclusion criteria were selected from Study 1. The representative value of SUA level in this study was determined by the highest value among SUA levels, obtained by searching the medical records of subjects in the GWAS from December 2000 to October 2012. Subjects with renal insufficiency as well as SLE patients were excluded. Also, we used the highest SUA values obtained before use of drugs that could increase SUA concentrations, such as antituberculosis medication. We analyzed associations with serum uric acid concentrations based on data from GWASs in the Korean population, and also tested whether polymorphism of any of the 40 SNPs associated with SUA identified previously were associated with SUA levels. Results: Study 1: We compared the MAFs of 13 SNPs in Koreans with those in other ethnic groups. Overall, the MAFs of SNPs associated with SUA level in the Korean population were quite similar to those among Japanese, but not in populations of European descent. Study 2: SNP rs12734001 (PPP1R12B) proved to have the most probable association with SUA concentrations (P_trend ⫽ 2.29 ⫻ 10⫺9). We also analyzed 13 SNPs shown previously by meta-analysis to be associated with SUA, and SNP rs3741414 (INHBC) was found to have ther strongest association with SUA level observed in the present study (P_trend ⫽ 0.01). Conclusion: The pattern of variants controlling SUA levels in the Korean population is quite similar to that in the Japanese population, but not in populations of European descent. SNP rs12734001 (PPP1R12B) is significantly associated with SUA level, and SNP rs3741414 (INHBC), previously identified SNP, is strongly associated with SUA levels among Koreans.

Table 2. Linear regression analyses of the genotype effect on the mean CRP and DAS28 levels relative to the major genotype.

Sunday, October 27

from each condition), we confirmed the different modulations of the alpha and beta HPT chains.

expression of UCP2 (3.86⫾0.12) as well as a decreased expression of PDK4 (0.20 ⫾0.01) was detected in OA cartilage samples (p⬍0.05). Gene expression of OLR-1 was only observed in OA cartilage samples. The experiments of glucose overload carried out in the cell line TC28a2 showed a direct correlation between glucose concentration and the expression levels of UCP2 (correlation coefficient (c.c)⫽0.965; p⫽0.102), MMP-3 (c.c⫽0.992; p⫽0.08) and IL-6 (c.c⫽0.996; p⫽0.05) as well as an inverse correlation with PDK4 (c.c⫽ ⫺0.914; p⫽0.265). The analysis of the interaction between the mtDNA haplogroups and the SNP rs659366 showed that carriers of both haplogroups J or T (mtDNA cluster TJ) that also carry the T allele of rs659366 have a significant decreased risk of OA (OR⫽0.581; CI: 0.357 – 0.948; p⫽0.029). Conclusion: This work provides strong evidence on metabolic alterations that occur in OA. These alterations involve the lipid and glucose metabolism, through the mitochondrial pathway, as well as the switch that modulates the balance between both metabolisms which, in turn, would lead to the increased inflammatory process that takes place in OA.

Table 1. Proteins identified in this work as altered in Osteoarthritis (OA) vs Control (N) sera. Protein ID A1AT FETUA ANGT APOA1 HPT HBB A2GL TTHY HPT APOA2 APOA4 APOC3 HPT IGKC LAC2 SAA

Proteins increased in OA

Ratio OA:N

Alpha-1-antitrypsin Alpha-2-HS-glycoprotein Angiotensinogen Apolipoprotein A-I Haptoglobin beta chain Hemoglobin subunit beta Leucine-rich alpha-2-glycoprotein Transthyretin Haptoglobin alpha-1 chain Proteins decreased in OA Apolipoprotein A-II Apolipoprotein A-IV Apolipoprotein C-III Haptoglobin alpha-2 chain Ig kappa chain C region Ig lambda-2 chain C regions Serum amyloid A protein

1,4 2 1,1 2,2 2 1,6 1,5 2,2 1,4 0,7 0,6 0,8 0,5 0,6 0,6 0,8

Disclosure: I. Rego-Pe´rez, None; M. E. Va´zquez-Mosquera, None; A. SotoHermida, None; M. Fernańdez-Moreno, None; J. Fernańdez-Tajes, None; E. Corte´s-Pereira, None; S. Relanõ-Fernańdez, None; N. Oreiro-Villar, None; C. Fernańdez-Lo´pez, None; F. J. Blanco, None.

166 Conclusion: We were able to identified 16 protein forms altered in the disease (9 increased and 7 decreased), and we verified for the first time the OA- dependent alteration of the haptoglobin chains. The haptoglobin protein provides two different OA biomarkers easily to measure in serum samples at the same time.

The Potential Role Of Protein Tyrosine Phosphatase Receptor D (PTPRD) Gene Copy Number Variation In Susceptibility To Rheumatoid Arthritis. Su Jin Yoo1, Mi Kyoung Lim2, Donghyuk Sheen2, In Seol Yoo1, Jinhyun Kim1, Seong Wook Kang1 and Seung-Cheol Shim3. 1Chungnam National University School of Medicine, Daejeon, South Korea, 2Eulji University Hospital, Daejeon, South Korea, 3Chungnam National University Hospital, Daejeon, South Korea.

Disclosure: C. Fernandez-Costa, None; V. Calamia, None; P. Fernandez-Puente, None; J. Mateos, None; B. Rocha, None; L. Lourido, None; J. L. Capelo, None; C. Ruiz-Romero, None; F. J. Blanco, None.

Background/Purpose: Since it is important to explore genetic variations associated with rheumatoid arthritis (RA), genome-wide association studies (GWAS) have led to the identification of RA genetic variants putatively associated with susceptibility. Recently, copy number variation (CNV) may also affect susceptibility to diseases, which have been already observed in diverse autoimmune diseases. Protein tyrosine phosphatase receptor D (PTPRD) is a member of the receptor-like PTP which expresses in the B cell lines and thymus and could be involved in the pathogenesis of autoimmune diseases. In this study, we investigated whether the variation of the PTPRD gene copy number related with susceptibility to RA. Methods: To investigate whether the variation of the PTPRD gene copy number influence the pathogenesis to RA, blood samples and clinical records were obtained from 217 RA patients (184 females, 33 males) and 205 healthy controls. The genomic DNA of RA patients and healthy controls was extracted from leukocytes in peripheral blood using the Genomic DNA Extraction kit (iNtRON Biotechnology, Korea). To measuring the copy number of PTPRD gene, the quantitative real-time PCR (QPCR) was carried out using Mx3000P QPCR system (Stratagene, La Jolla, CA) and each sample for each gene was assayed in triplicate. Western blot was conducted to detect expression levels of PTPRD. Results: The copy number of PTPRD gene in RA patients was compared with that in healthy controls. The proportion of the individuals with ⬍2 copy of VPREB1 was significantly higher in patients than in controls, while that of the individuals with ⬎2 copy was lower in patients than in controls. The average relative copy number of the PTPRD gene in RA patients (1.14, 95 % CI (1.12–1.16)) was significantly lower than that in healthy controls (1.65, 95 % CI (1.12–1.16), p ⬍ 0.0001). Furthermore, we also investigated association between copy number of PTPRD and RA phenotype such as RF factor and anti-CCP levels, which showed no association between copy number of PTPRD and both RA phenotypes. Western blot showed the lower expression of PTPRD in patients with RA compared to control subjects. Conclusion: This is the first evidence showing the association between low copy number of the PTPRD gene and susceptibility to RA, which may help understanding the pathogenesis of RA and other autoimmune disorders like affecting maturation and differentiation of T cell and B cells.

165 New Insights Into The Metabolic Origin Of Osteoarthritis. Ignacio Rego-Pe´rez1, Marıá Eugenia Va´zquez-Mosquera1, Angel Soto-Hermida1, Mercedes Fernańdez-Moreno1, Juan Fernańdez-Tajes1, Estefanıá Corte´sPereira1, Sara Relanõ-Fernańdez1, Natividad Oreiro-Villar1, Carlos Fernańdez-Lo´pez1 and Francisco J. Blanco2. 1INIBIC-Hospital Universitario A Corunã. Rheumatology Division. Genomic Group, A Corunã, Spain, 2 INIBIC-Hospital Universitario A Corunã, A Corunã, Spain. Background/Purpose: Metabolic alterations take place in osteoarthritis (OA) and the mtDNA haplogorups influence the prevalence and severity of the disease. The aim of this work is to analyze the metabolic behavior in OA from different points of view, including the analysis of gene polymorphisms, mitochondrial variants and gene expression. Methods: The first approach consisted in the analysis of the influence of the mtDNA haplogroups in the body mass index (BMI) in a cohort of 638 OA patients and 474 healthy controls from Spain. This approach was followed by a second one that included the analysis of the expression levels of genes related to both lipid and glucose metabolism in RNA isolated from 24 OA and 22 healthy normal (N) cartilage samples; for this purpose we also used the chondrocyte cell line TC28a2 in order to analyse the glucose overload (at 1, 4.5 and 18 g/l) during 48 hours in the expression levels of these genes as well as of both MMP-3 and IL-6; finally, we search for possible interactions between the mtDNA haplogroups and the SNP -866C/T (rs659366) in the promoter region of UCP2 to assess their incidence in the prevalence of OA in a cohort of 137 OA patients and 133 healthy controls from Spain. Nucleic acid isolation was carried out using commercial kits following the manufacturer recommendations with some modifications. Appropriate data analysis was performed with SPSS software (v19) and qBase plus software (Biogazelle). Results: The mtDNA haplogroup J and OA are independently associated with BMI, so that carriers of this mitochondrial variant show significant lower BMI values (p⫽0.006) meanwhile OA patients show significant higher values (p⬍0.001), being the BMI a risk factor for OA. The obtained OA/N expression ratio of GLUT1 (0.8⫾0.26), GLUT3 (0.73⫾0.01), GLUT5 (0.68⫾0.19), HK1 (1.09⫾0.02), HK2 (1.59⫾0.21), INSR (0.99⫾0.08), CPT1B (1.30⫾0.1), ACADM (2.65⫾0.36), HADHB (1.68⫾0.09), PPARg (0.61⫾0.11), PPARa (1.41⫾0.33) did not show significant differences (p ⬎ 0.05). However, a significant increased

Disclosure: S. J. Yoo, None; M. K. Lim, None; D. Sheen, None; I. S. Yoo, None; J. Kim, None; S. W. Kang, None; S. C. Shim, None.

S66

Association Of Functional Polymorphisms In IRF2 With Systemic Lupus Erythematosus In a Japanese Population. Aya Kawasaki1, Hiroshi Furukawa2, Nao Nishida3, Eiji Warabi1, Yuya Kondo1, Satoshi Ito4, Isao Matsumoto1, Makio Kusaoi5, Akiko Suda6, Shouhei Nagaoka7, Keigo Setoguchi8, Tatsuo Nagai9, Shunsei Hirohata9, Katsushi Tokunaga10, Yoshinari Takasaki5, Hiroshi Hashimoto11, Takayuki Sumida1, Shigeto Tohma2 and Naoyuki Tsuchiya1. 1University of Tsukuba, Tsukuba, Japan, 2Sagamihara Hospital, National Hospital Organization, Sagamihara, Japan, 3National Center for Global Health and Medicine, Ichikawa, Japan, 4Niigata Rheumatic Center, Shibata, Japan, 5Juntendo University, Tokyo, Japan, 6Yokohama City University, Mecical Center, Yokohama, Japan, 7Yokohama Minami Kyousai Hospital, Yokohama, Japan, 8Tokyo Metropolitan Cancer and Infectious Diseases Center Komagome Hospital, Tokyo, Japan, 9Kitasato University School of Medicine, Sagamihara, Japan, 10The University of Tokyo, Tokyo, Japan, 11 Juntendo University School of Medicine, Tokyo, Japan. Background/Purpose: Interferon regulatory factor (IRF) families are transcription factors involved in type I interferon (IFN) pathway. Recent genetic studies identified association of IRF family genes, IRF5, IRF7, and IRF8, with systemic lupus erythematosus (SLE). IRF2 is thought to negatively regulate the type I IFN signals. In addition, IRF2 has a role in induction of Th1 differentiation and NK cell development. So far, association of IRF2 with SLE has not been published. In this study, we conducted an association study to examine whether IRF2 contributes to susceptibility to SLE in a Japanese population. Methods: A case-control association study was performed in 501 Japanese patients with SLE and 551 healthy controls on 46 tag SNPs in the IRF2 gene, selected based on linkage disequilibrium (LD) and minor allele frequencies (r2ⱖ0.8, minor allele frequencyⱖ0.05), using the DigiTag2 and the TaqMan allele discrimination assays. To identify functional SNPs, resequencing of the IRF2 region was performed using the next-generation sequencing. Effect of IRF2 SNPs on transcriptional activity was analyzed using a luciferase assay. Results: Among the IRF2 tag SNPs, rs13146124T in intron 1 was most significantly associated with SLE (dominant model, P⫽7.4⫻10⫺4, odds ratio [OR] 1.60, 95% confidence interval [CI] 1.22–2.11). Resequencing identified SNPs in LD with rs13146124. Eight SNPs including rs62339994 were in almost absolute LD with rs13146124 (r2: 0.98–1), whereas rs66801661 showed moderate LD (r2: 0.52). We next examined association of rs66801661 with SLE. rs66801661A allele was significantly increased in SLE (allelic model, P⫽3.7⫻10⫺4, OR 1.75, 95%CI 1.29–2.39). Haplotype analysis identified three haplotypes constituted by rs66801661 and rs13146124, and the haplotype carrying both of the risk alleles, rs66801661A - rs13146124T, was significantly increased in SLE (SLE: 10.8%, control: 6.4%, permutation P⫽2.0⫻10⫺4). On the other hand, the haplotype formed by the non-risk alleles, rs66801661G - rs13146124C, was decreased in SLE (SLE: 83.1%, control: 87.9%, permutation P⫽0.0014). To examine the functional significance of the risk haplotype, IRF2 region containing the promoter region, exon 1 and the 5’ part of intron 1 encompassing rs66801661 and rs62339994, which was in tight LD with rs13146124, were inserted upstream of the luciferase gene. Three constructs corresponding to the naturally-occurring haplotypes were transfected, and the luciferase activities were compared. The construct carrying both of the risk alleles, rs66801661A and rs62339994A (in LD with rs13146124T), was associated with the highest transcriptional activity in Jurkat cells under IFN␥ stimulation (ANOVA, P⫽1.3⫻10⫺4). Conclusion: IRF2 polymorphisms are associated with susceptibility to SLE in a Japanese population. The SLE risk haplotype appeared to be associated with transcriptional activation of IRF2.

Disclosure: H. W. van Steenbergen, None; R. Tsonaka, None; T. W. J. Huizinga, None; S. le Cessie, None; A. H. M. van der Helm-van Mil, None.

169 Effects Of Chondroitin Sulfate On The Gene Expression Profile In Interleukin-1␤ Stimulated Synovial Fibroblast Cells Cultures. Cećile Lambert1, Jean-Emile Dubuc2, Eulalia Montell3, Josep Verges3 and Yves Henrotin1. 1Bone and Cartilage Research Unit, Lie`ge, Belgium, 2Orthopaedic Department, Bruxelles, Belgium, 3Pre-Clinical R&D Area, Pharmascience Division, BIOIBERICA S.A., Barcelona, Spain. Background/Purpose: Chondroitin sulfate (CS) is one the most used molecules in the management of OA. Its mechanism of action remains to be detailed. In this study, we perform a microarray analysis to identify a differential expression profile between control and IL-1␤ stimulated synovial fibroblast cells cultures and to investigate the effects of CS on this gene expression profile Methods: OA synovial specimens were obtained from 12 patients undergoing knee replacement. Synovial fibroblast cells (SFC) were enzymatically isolated and used after four passages (P4). SFC were pre-treated 1 hour with highly purified bovine CS (200 ␮g/ml, Bioibe´rica S.A., Barcelona, Spain) before treatment with IL-1␤ (1 ng/ml) for 24 hours. Gene expression profiling was performed using Illumina’s multi-sample format Human HT-12 BeadChip (Illumina Inc.). Differential analysis was performed with the BRB array tools software. Class comparison test between control (Ctl) and interleukin (IL)-1␤ conditions, Ctl and Ctl/CS and IL-1␤ and IL-1␤/CS conditions was based on paired t-test where Ctl and IL-1␤, Ctl and Ctl/CS and IL-1␤ and IL-1␤/CS were paired for each patient. The biological relevance of regulated genes was analyzed with Ingenuity Pathways Analysis (Ingenuity® Systems). Probes with a p-value below 0.001 were chosen and classified as up- or down-regulated. Results: 3308 genes were identified as differentially expressed genes between Ctl and IL-1␤ conditions. A differential profile of expression of major pathways involved in OA pathogenesis was observed. In the inflammatory network, the most upregulated cytokines were IL-8 and IL-6 (fold change: 156.25 and 58.8 respectively). We identified several chemokines, enzymes and metallothioneins (MTs). Complement factor B (CFB) and complement component 3 (C3) are two factors upregulated in the inflamma-

Disclosure: A. Kawasaki, None; H. Furukawa, None; N. Nishida, None; E. Warabi, None; Y. Kondo, None; S. Ito, None; I. Matsumoto, None; M. Kusaoi, None; A. Suda, None; S. Nagaoka, None; K. Setoguchi, None; T. Nagai, None; S. Hirohata, None; K. Tokunaga, None; Y. Takasaki, None; H. Hashimoto, None; T. Sumida, None; S. Tohma, None; N. Tsuchiya, None.

168 Predicting The Severity Of Joint Damage In Rheumatoid Arthritis; The Contribution Of Genetic Factors. Hanna W. van Steenbergen, Roula Tsonaka, Tom W.J. Huizinga, Saskia le Cessie and Annette H.M. van der Helm-van Mil. Leiden University Medical Center, Leiden, Netherlands.

S67

Sunday, October 27

Background/Purpose: The severity of radiologic progression is variable between rheumatoid arthritis (RA) patients. Recently several genetic severity variants have been identified and replicated. This study determined the contribution of the established genetic severity factors to the explained variance in radiologic progression and evaluated whether genetic factors, in addition to traditional clinical and serological risk factors, improved the accuracy of predicting the severity of radiologic progression. Methods: 417 early RA patients with yearly radiologic follow up were studied. The main outcome measure was the progression in Sharp-van der Heijde scores (SHSs) over six years. Genetic variants in the following genes were studied: HLA-DRB1, CD40, IL-15, DKK-1, IL2RA, GRZB, IL-4R, SPAG16, C5orf30, MMP-9 and OPG. Linear regression analyses were used to determine the explained variance and the net improvement in reclassifications of prediction models without and with genetic risk factors. For studying reclassification, the continuous outcome was categorized based on the severity of radiologic progression in progression in SHSs over six years ⱕ6, 7–30 and ⬎30 units, indicating mild, moderate and severe radiologic progression. Results: Treatment effects and a combination of traditional risk factors explained 7.1% and 31.2% of the variance in progression in SHSs over six years. The genetic factors together explained 18.1%. When added to treatment effects and traditional factors, the genetic risk factors additionally explained 7.4% of the variance. Compared to a prediction model without genetic factors, a prediction model also including genetic factors yielded a net correct reclassification of 5.9% of the patients; now 62% of those patients were correctly classified. Thus with a model including known traditional and genetic factors 38.1% of the patients were still not correctly classified. Evaluating the reclassifications per severity group, the net percentages of patients that were additionally correctly classified were 0%, 5.1% and 13.2% for the groups with mild, moderate and severe progression respectively. Hence, including genetic factors led to improved identification of patients with severe radiological progression in particular. Sensitivity analyses using imputation of missing radiographs yielded comparable results. Conclusion: When added to a model consisting of traditional factors, genetic risk factors improved the predictive accuracy. Nonetheless, the predictive performance was still insufficient for use in clinical practice.

167

Sunday, October 27

tory complement cascade. We identified some genes implicated in the angiogenesis pathway. The most upregulated was Stanniocalcin 1 (STC1) (fold change: 9.09). The differential expression of intermediates in cartilage anabolism and catabolism revealed an imbalance in favour of catabolism. MMP-3 was largely upregulated (fold change: 62.5). Wnt 5A and low density lipoprotein receptor-related protein (LRP8) were significantly upregulated while frizzled homolog 2 (FZD2) and dickkopf homolog 3 (DKK3) were downregulated in the Wnt signaling. The class comparison test highlighted 660 differentially expressed genes between Ctl and Ctl/CS conditions and 241 genes between IL-1␤ and IL-1␤/CS. Among them, our attention was focused on two genes upregulated in the presence of CS: lysyl oxidase-like 4 (LOXL4) and claudin 11 (CDLN11), two genes that negatively regulate cell invasion. Conclusion: We here evidenced in synovial fibroblast cells the modulation of gene expression following IL-1␤ stimulation. We also demonstrated the modulatory effects of CS on gene expression and isolated several CS-modulated genes of interest such as LOXL4 and CDLN11, which could constitute new mechanisms of action of the molecule and contribute to explain the symptomatic efficacy of CS in the treatment of OA.

detailed. The aim of the present work is to identify the differentially expressed genes between the inflammatory (I) and normal/reactive (N/R) synovial areas using a unique ex vivo culture model and to investigate the genetic modulatory effects of CS in this model. Methods: Synovial cells (SC) were isolated from OA synovial specimens from 12 patients undergoing knee replacement. The synovial membrane was dissected and SC from N/R and I areas cultured separately for a period of 7 days with or without highly purified bovine CS (200 ␮g/ml, Bioibe´rica S.A., Barcelona, Spain). Gene expression profiling was performed using Illumina’s multi-sample format Human HT-12 BeadChip (Illumina Inc.). Differential analysis was performed with the BRB array tools software. Class Comparison test between N/R and I conditions, N/R and N/R-CS conditions and I and I-CS conditions was based on paired t-test where N/R and I, N/R and N/R-CS and I and I-CS were paired for each patient. The biological relevance of upand down-regulated genes was analyses with Ingenuity Pathways Analysis (Ingenuity® Systems). Results: From among 47000 probes, 18253 were filtered out. Probes with a p-value below than 0.005 were chosen and classified as up- or downregulated ones. By this way, 465 differentially expressed genes between N/R and I areas were identified. Many inflammatory mediators appear differentially expressed. The interferon alpha-inductible protein 6 (IFI6) was the most up-regulated. We also identified the hydroxysteroid (11-beta) dehydrogenase 1 (HSD11B1), the cathepsin K (CTSK), the chemokine (C-X-C motif) ligand 1 (CXCL1) and the EBV-induced G-protein coupled receptor 2 (EBI2). The differential expression of intermediates involved in angiogenesis pathway was also revealed between N/R and I areas. Among them, R-spondin-3 (RSPO3), the secreted phopshoprotein 1 (SPP1) and aquaporin 9 (AQP9) were upregulated whereas ADAMTS1 was down-regulated. Finally, in the Wnt signaling, RSPO3 was up-regulated unlike dickkopf homolog 3 (DKK3) which was in turn down-regulated. We next performed a class comparison test between N/R and N/R-CS in one hand and between I and I-CS the other hand. 489 genes were identified as differentially expressed genes between N/R and N/R-CS conditions while 219 genes were identified between I and I-CS conditions. In this latter, our attention was focused on the downregulated genes. Among them, we identified a number implicated in angiogenesis and cell migration pathways. Thus, the endothelial cell-specific molecule-1 (ESM1), the Transmembrane-4-L-six-family-1 (TM4SF1), the 5’-Ectonucleotidase (NT5E) and the growth arrest-specific gene 6 (GAS6) were down-regulated by CS. Conclusion: Our work demonstrates the differential gene expression profile between paired inflammatory and normal/reactive areas of synovial membrane as well as the modulatory effects of CS on gene expression in the inflammatory areas, especially regarding genes involved in both angiogenesis and cell migration.

Disclosure: C. Lambert, BioIberica, 2; J. E. Dubuc, None; E. Montell, BioIberica, 3; J. Verges, BioIberica, 3; Y. Henrotin, Artialis SA, 1, BioIberica, Artialis, Tilman, Expanscience, 5, BioIberica, 2.

170 Association Of BMI, 8 SNPs Reported To Be Related To Gout Phenotype and Their Interaction In Gout Incidence In Framingham Heart Study. Jasvinder A. Singh1, Ana Vazquez2, Richard Reynolds2, Vinodh Srinivasasainagendra3, S. Louis Bridges Jr.2 and David Allison2. 1University of Alabama, Tuscaloosa, AL, 2University of Alabama at Birmingham, Birmingham, AL, 3University of alabama at birmingham, birmingham, AL. Background/Purpose: We aim to assess the association of 8 serum urate SNPs and BMI and their interactions with incident gout in a population-based cohort study. Methods: We used the Framingham Study including subject from the Original and the Offspring cohort (N⫽4,967). We assessed the effect of 8 SNPs in the recently described genetic urate score on incident gout. Subjects were genotyped with Affy500K platform, and two of the SNPs associated to gout rs1165196 and rs1106766 were present in that platform. The SNPs rs1967017, rs780093, rs13129697, rs2199936, rs675209 and rs2078267 were not present in the Affy SNP array, thus we imputed them with IMPUTE2. BMI was taken at the time of gout or, at the time of censoring if the subject is healthy or died without gout. We fitted logistic regression models to assess the associations. Estimated effects in the liability scale, odd ratios, and p-values are reported for all SNPs, covariates, and interactions. 408 patients had incident gout (74% males), with 169 from the FHS original cohort and 293 from the Offspring cohort. Results: In a model with main effects of BMI and SNPs, three of eight SNPs and BMI were significantly associated with incident gout (p⬍⫽0.008 for all). Zero SNPs showed significant main effects on gout in the model that adjusted for all BMI*SNP interaction terms. However, BMI remained significant (p⫽0.003) and the rs2199936*BMI interaction was nearly significant (p⫽0.06). Gender and duration of follow-up were also significant predictors (p⬍⫽0.009) in all models. Conclusion: SNPs known to predict urate levels moderated the association of BMI with gout, suggesting the possibility that the extent to which BMI increases the risk for got may depend on a person’s a priori genetic risk for high urate levels.

Disclosure: C. Lambert, BioIberica, 2; J. E. Dubuc, None; E. Montell, BioIberica, 3; J. Verges, BioIberica, 3; Y. Henrotin, Artialis SA, 1, BioIberica, Artialis, Tilman, Expanscience, 5, BioIberica, 2.

172 Transcript Profiling Of Kidney Biopsies From Chinese Patients With Lupus Nephritis Suggests a Prevalence Of Infiltrating CD8ⴙ T Cells, Monocytes/Macrophages, and B Cells In ISN/RPS Class IV Disease. Zheng Liu1, Chris Morehouse1, Xinfang Huang2, Philip Brohawn1, Nan Shen3, Gabor G. Illei1, Yihong Yao1 and Brandon W. Higgs1. 1MedImmune, LLC, Gaithersburg, MD, 2Renji Hospital, Shanghai Jiaotong University School of Medicine, Shanghai, China, 3Cincinnati Children’s Hospital Medical Center, cincinnati, OH. Background/Purpose: Lupus nephritis (LN) affects ⬃60% of patients with systemic lupus erythematosus (SLE). The International Society of Nephrology and the Renal Pathology Services (ISN/RPS) classification system categorizes LN into 6 main classes based on the underlying renal pathology. Among these, Classes I and II rarely require immunosuppressive treatment and Class-VI is associated with irreversible glomerulosclerosis. Class-IV (diffuse proliferative nephritis) is the most severe form with the largest proportion of all diagnosed cases (⬃40%) followed by Class-III (focal proliferative nephritis), representing ⬃25% of all diagnosed cases. These classes have similar histopathological and clinical characteristics while Class-V (membranous nephritis) representing ⬃10% of LN cases is different both in its clinical course and histological appearance. We examined kidney biopsies from a group of ISN/RPS Class III, IV, and V LN subjects to evaluate molecular differences between these classes. Methods: Kidney biopsies were procured from 6 healthy donor (HD) and 25 LN (8 Class-III, 6 Class-IV, and 11 Class-V) Chinese subjects. The

Disclosure: J. A. Singh, Takeda, Savient, 2, Savient, Takeda, Ardea, Regeneron, Allergan, 5, URL pharmaceuicals Novartis, 5; A. Vazquez, None; R. Reynolds, None; V. Srinivasasainagendra, None; S. L. Bridges Jr., None; D. Allison, None.

171 Effects Of Chondroitin Sulfate On The Gene Expression Profile In The Inflamed Synovial Membrane. Cećile Lambert1, Jean-Emile Dubuc2, Eulalia Montell3, Josep Verges3 and Yves Henrotin1. 1Bone and Cartilage Research Unit, Lie`ge, Belgium, 2Orthopaedic Department, Bruxelles, Belgium, 3Pre-Clinical R&D Area, Pharmascience Division, BIOIBERICA S.A., Barcelona, Spain. Background/Purpose: Chondroitin sulfate (CS) is one the most used molecules in the management of OA. Its mechanism of action remains to be

S68

Miyashita, None; K. Migita, None; A. Suda, None; S. Nagaoka, None; N. Tsuchiya, None; S. Tohma, Research grants from Pfizer Japan Inc., 2, Research grants from Astellas Pharma Inc., 2.

174 Genetic Association On Disease Severity In Rheumatoid Arthritis: A Validation Study In Japanese Patients. Shinji Yoshida1, Katsunori Ikari1, Koichiro Yano1, Yoshiaki Toyama2, Atsuo Taniguchi1, Hisashi Yamanaka3 and Shigeki Momohara1. 1Institute of Rheumatology, Tokyo Women’s Medical University, Tokyo, Japan, 2Department of Orthopaedic Surgery, Keio University School of Medicine, Shinjuku, Japan, 3Tokyo Women’s Medical University, Tokyo, Japan. Background/Purpose: The disease severity of rheumatoid arthritis (RA) is objectively measured by radiographic joint destruction, which is reflective of the cumulative burden of inflammation. Recently, the heritability has been proven to be partly involved in progression of joint destruction in RA, and several genetic variants and gene-gene interactions have been reported to be associated with progression of joint destruction in European multi-cohort studies. However, the presence of genetic heterogeneity in RA susceptibility genes has been demonstrated in many population-based studies, and the genetic predisposition factors for progression of joint destruction remain to be poorly understood in Asian patients with RA. The purpose of this study was to identify genetic variants associated with progression of joint destruction in Japanese patients with RA. Methods: This study included 865 Japanese patients with RA for whom Sharp/van der Heijde scores (SHS) of hands were available at a disease duration of 5 years. DNA samples of the subjects were obtained from the Institute of Rheumatology Rheumatoid Arthritis cohort study (IORRA) DNA collection. All of the patients who donated DNA samples consented to participate in this study in accordance with the process approved by the Tokyo Women’s Medical University Genome Ethics Committee, and satisfied the American College of Rheumatology 1987 revised criteria for the classification of RA. Nine of the single nucleotide polymorphisms (SNPs) reported in European multi-cohort studies were selected and genotyped in the DNA samples: rs6821171, and rs1521761 which is in absolute linkage disequilibrium (r2⫽1) with rs7667746, in interleukin15 (IL15); rs8192916, in granzyme B (GZMB); rs1896368, rs1896367, rs1528873, and rs10762715, in Dickkopf-1 (Dkk1); rs4792909, and rs6503475, in sclerostin (Sost). SNP genotyping was performed by using the TaqMan fluorogenic 5’ nuclease assay (Applied Biosystems, Tokyo, Japan). Univariate linear regression analyses were performed to examine the association of each SNP and several combinations of SNPs with progression of joint destruction in the first 5 years after onset of RA, calculated as SHS of hands at the 5-year disease duration. All SHS were log-transformed to obtain a normal distribution. Results: Univariate linear regression analyses revealed that all of the SNPs and combinations of SNPs tested in this study were not associated with progression of joint destruction (Table 1).

Figure 1. Distribution of A) macrophage, B) CD8⫹ T cell, and C) Ig gene signatures for LN patients with Class-III, -IV, or -V disease. Disclosure: Z. Liu, MedImmune LLC, 3, AstraZeneca, 1; C. Morehouse, Astra Zeneca, 1; X. Huang, None; P. Brohawn, MedImmune LLC, 3, Astra Zeneca, 1; N. Shen, MedImmune, 5; G. G. Illei, AstraZeneca, 1, MedImmune, 3; Y. Yao, None; B. W. Higgs, MedImmune LLC, 3, AstraZeneca, 1.

173 HLA-DRB1*08:02 Is Associated With Bucillamine-Induced Proteinuria In Japanese Rheumatoid Arthritis Patients: A Case-Control Study. Hiroshi Furukawa1, Shomi Oka1, Kota Shimada2, Shoji Sugii2, Atsushi Hashimoto1, Akiko Komiya1, Naoshi Fukui1, Taiichiro Miyashita3, Kiyoshi Migita3, Akiko Suda4, Shouhei Nagaoka5, Naoyuki Tsuchiya6 and Shigeto Tohma1. 1Sagamihara Hospital, National Hospital Organization, Sagamihara, Japan, 2Tokyo Metropolitan Tama Medical Center, Tokyo, Japan, 3Nagasaki Medical Center, National Hospital Organization, Omura, Japan, 4Yokohama City University, Mecical Center, Yokohama, Japan, 5Yokohama Minami Kyousai Hospital, Yokohama, Japan, 6University of Tsukuba, Tsukuba, Japan.

Table 1

Background/Purpose: Bucillamine (Buc) is one of the commonly used disease-modifying anti-rheumatic drugs (DMARDs) in Japan. Drug-induced proteinuria can occur in rheumatoid arthritis (RA) patients treated with Buc, and represents a drug hypersensitivity reaction. Striking associations of human leukocyte antigen (HLA) alleles with adverse reactions have recently been reported for many drugs. Methods: We investigated the association of HLA class II with Bucinduced proteinuria (BI-Pro) in 485 Japanese RA patients treated with Buc, of which 25 had developed BI-Pro. Results: This study showed a highly significant association of DRB1*08:02 with BI-Pro (P⫽1.09⫻10⫺6, corrected P [Pc]⫽1.96⫻10⫺5, odds ratio [OR] 25.17, 95% confidence interval [CI] 7.98–79.38). DQB1*04:02 was also significantly associated with increased risk of BI-Pro (P⫽2.44⫻10⫺5, Pc⫽2.69⫻10⫺4, OR 10.35, 95%CI 3.99–26.83). Conclusion: We detected striking HLA class II associations with proteinuria induced by Buc in Japanese RA patients. This association merits confirmation in future large-scale studies for its potential clinical usefulness as a biomarker to prevent adverse reactions.

Gene

P value 0.48 0.36 0.21 0.36 0.91 0.65 0.21 0.16 0.09 0.31 0.28 0.90 0.59

additive

0.01

0.78

additive

0.02

0.59

Allele (minor/major)

MAF

Tested model

C/A

0.30

rs1521761

T/A

0.15

GZMB rs8192916

A/G

0.45

Dkk-1

rs1896368

A/G

0.42

rs1896367

A/G

0.30

rs1528873

C/A

0.22

rs1528873 rs6503475 rs10762715 rs4792909 rs10762715 rs6503475

C/A G/A T/C T/G T/C G/A

0.22 0.45 0.22 0.40 0.22 0.45

additive recessive additive recessive additive recessive additive recessive additive recessive additive recessive additive

Dkk-1 Sost

Disclosure: H. Furukawa, Research Grants from Takeda Science Foundation, 2, Research Grants from Mitsui Sumitomo Insurance Welfare Foundation, 2; S. Oka, None; K. Shimada, None; S. Sugii, None; A. Hashimoto, Research grants from Mitsuibishi Tanabe Pharma Corporation, 2; A. Komiya, None; N. Fukui, None; T.

␤ ⫺0.05 ⫺0.13 ⫺0.10 ⫺0.13 ⫺0.01 0.05 0.07 0.12 ⫺0.11 ⫺0.13 ⫺0.07 ⫺0.02 ⫺0.02

SNP rs6821171

L15

Conclusion: Genetic variants in IL15, GZMB, Dkk-1, and Sost were not associated with progression of joint destruction in Japanese patients with RA.

S69

Sunday, October 27

Affymetrix U133⫹ array was used to transcript profile specimens. Fold changes were calculated for each LN subject relative to the average HD expression. Previously defined cell type-specific gene signatures were identified to be differential between ISN/RPS classes. Results: Several gene signatures were significantly over-expressed (p⬍0.05) in Class-IV subjects relative to both Class-III and -V subjects. These included: a chemokine, MHC class I, monocyte, macrophage, and CD8⫹ T cell gene signatures. An immunoglobulin (Ig) signature was also significantly over-expressed in Class-IV subjects compared to Class-V subjects, but not Class-III subjects. The following gene signatures did not significantly differ between the three classes: neutrophil, CD4⫹ T cell, and type I IFN. Conclusion: Transcript profiling was used to correlate the molecular differences between LN disease classifications and cell infiltrate prevalence. In addition to pathways upregulated in all three forms of LN, we identified cell type-specific signatures which were over-expressed in Class-IV. The increased CD-8⫹ signature suggests that CD8⫹ effector T cell infiltration may correlate with more active disease status. Monocytes/macrophages are also more prevalent in this more active disease class, suggesting an increase in infiltration from damage caused by these CD8⫹ T effector cells, which may also lead to an increase in Ig products.

Sunday, October 27

These results indicate the presence of genetic heterogeneity in risk loci of progression of joint destruction in RA between Caucasian and Asian patients. The heritability of the rate of joint destruction might be different among ethnic populations.

176 Identification Of Major Histocompatibility Complex Class II Alleles Associated With Systemic Sclerosis Through Imputation Strategy. Jose Ezequiel Martin1, Carmen P. Simeoń2, Norberto Ortego-Centeno3, Patricia Carreira4, Miguel A. Gonzalez-Gay5, Nicolas Hunzelmann6, Shervin Assassi7, Filemon K. Tan7, Frank C. Arnett7, Xiaodong Zhou7, T.R.D.J. Radstake8, Maureen D. Mayes7, Paul de Bakker9, Javier Martin10 and B. P. C. Koeleman8. 1Instituto de Parasitologıá y Biomedicina Lo´pez-Neyra, IPBLNCSIC, Granada, Spain, 2Hospital Valle de Hebron, Barcelona, Spain, 3Hospital Clıńico San Cecilio, Granada, Spain, 4Hospital Universitario 12 de Octubre, Madrid, Spain, 5Epidemiology, Genetics and Atherosclerosis Research Group on Systemic Inflammatory Diseases, Rheumatology Division, IFIMAV, Santander, Spain, 6University of Cologne, Cologne, Germany, 7 University of Texas Health Science Center at Houston, Houston, TX, 8 University Medical Center Utrecht, Utrecht, Netherlands, 9Brigham and Women’s Hospital, Harvard Medical School, Boston, MA, 10Instituto de Parasitologia y Biomedicina Lo´pez-Neyra, IPBLN-CSIC, Granada, Spain.

Disclosure: S. Yoshida, None; K. Ikari, Pfizer, Takeda, 2, Asahikasei, AbbVie, Astellas, Eisai, Kaken, Mitsubishi Tanabe, Taisho Toyama, 8; K. Yano, None; Y. Toyama, None; A. Taniguchi, None; H. Yamanaka, Abbott, AbbVie, Asahikasei, Astellas, AstraZeneca, Bristol-Myers Squib, Chugai, Daiichi Sankyo, Eisai, GlaxoSmithKline, Janssen, Mitsubishi Tanabe, MSD, Nippon Kayaku, Pfizer, Santen, Taishotoyama, Takeda, Teijin, 2, Abbott, AbbVie, Astellas, AstraZeneca, Bristol-Myers Squib, Chugai, Daiichi Sankyo, Eisai, Mitsubishi Tanabe, Nippon Kayaku, Pfizer, Takeda, Teijin, 5, Abbott, AbbVie, Astellas, Bristol-Myers Squib, Chugai, Eisai, Mitsubishi Tanabe, Pfizer, Takeda, Teijin, 8; S. Momohara, None.

175 Dynamics Of The Inflammatory Response In Rheumatoid Arthritis Patients Following Traumatic Injury: Insights From In Vivo, In Silico, and Single Nucleotide Polymorphism Studies. Rajaie Namas1, Rami Namas2, Harpreet Sagar3, Felix Fernandez-Madrid3, Timothy Billiar2 and Yoram Vodovotz2. 1Wayne State University/ Henry Ford Health System, Detroit, MI, 2University of Pittsburgh, Pittsburgh, PA, 3Wayne State University, Detroit, MI.

Background/Purpose: Different alleles of MHC class II molecules have been associated either with risk to systemic sclerosis (SSc) or its subphenotypes. Due to the high cost of HLA typing, studies have been limited to small sample sizes, preventing definitive statements as to which HLA alleles are likely causal Methods: We have imputed the MHC class I and II alleles of 2,296 cases and 5,356 controls from four countries with a method previously developed. Besides classical HLA alleles, we also imputed amino acid changes encoded by genetic variants within the different MHC molecules. We compared the frequencies of the different alleles between cases and controls for SNPs, amino acids and classical HLA alleles Results: The accuracy of the imputations ranged from 84% to 94% depending on the alleles being imputed with an average of 92% for all alleles in both populations. We confirmed previous associations of HLA alleles with SSc or its auto-antibody positive subgroups (e.g. HLA-DRB1*0701, HLADPB1*1301, HLA-DRB1*1104, HLA-DQB1*0501). We define in deeper detail some of these associations down to the level of aminoacidic positions which affect epitope binding and relocate some previous SSc associations within auto-antibody positive subgroups. Furthermore, we describe new association of the HLA allele HLA-DRB1*0801 with the presence of anti-centromere auto-antibodies. Conclusion: Our data indicate that most HLA associations are specific to the presence of auto-antibodies. Also, only MHC class II alleles are associated and not MHC class I.

Background/Purpose: Both traumatic injury/hemorrhagic shock (T/ HS) and rheumatoid arthritis (RA) are inflammatory diseases that are major sources of morbidity worldwide. We hypothesize that RA modifies T/HS-induced dynamic inflammatory networks, and that antiinflammatory therapy for RA may modify T/HS-induced inflammation and outcomes. Accordingly, we sought to compare a prolonged time course of post-injury inflammatory mediators and outcomes in T/HS patients with or without RA. Methods: From a cohort of 472 blunt trauma survivors studied following IRB approval, 12 patients diagnosed with seropositive RA according to 2010 ACR criteria (8 males and 4 females; age: 65⫾4; ISS: 17⫾1, receiving Methotrexate and tumor necrosis factor-a [TNF-a]inhibitors) were matched with 12 non-RA control patients (8 males and 4 females; age: 68⫾4, ISS: 17⫾2). Serial blood samples were obtained from all patients (3 samples within the first 24 h and then daily from days 1 to 7 post-injury). Twenty-three plasma inflammatory mediators were assayed using Luminex™, along with NO2⫺/NO3⫺ assayed using nitrate reductase, respectively. Statistically significant differences (P ⬍ 0.05) between both groups were determined using the two-way analysis of variance (ANOVA) followed by the Holm-Sidak post hoc test. In silico Dynamic Bayesian Network (DBN) inference was utilized to infer causal relationships among inflammatory mediators based on probabilistic measures. Human genomic DNA was isolated and amplified for known TNF-␣, IL-6, IL-10, and IFN-␥ single nucleotide polymorphisms (SNPs) using PCR, along with TLR4 (Asp299Gly and Thr300I1e) SNPs using gel-DNA sequencing. Results: There was no statistical difference in the ICU length of stay, total length of stay, days on mechanical ventilation, and Marshall score (a measure of organ dysfunction) in RA vs. non-RA patients. Plasma levels of TNF-␣, IL-6, IL-10, IL-17, MIG, MIP-1␤, and NO2⫺/NO3⫺were significantly elevated within the 24 h in the RA group and up to day 7 when compared to the non-RA group. In contrast, plasma IP-10 and MCP-1 levels were significantly lower in the RA group within the first 24 h when compared to controls. DBN suggested that the inflammatory response in RA trauma patients is mainly driven by MIG, whereas MCP-1 drives the inflammatory response via forward-positive loops with IP-10 and MIG in non-RA patients. The incidence of all of the measured SNPs was not significantly different between trauma patients with or without RA. Conclusion: To our knowledge, this is the first pilot study that characterizes the inflammatory response in RA patients following blunt trauma. Our analysis suggested that different cytokine patterns emerge, particularly in the early events post-injury, in RA patients independent of the injury severity score on admission. Network analysis points to MIG as a key driver of this process. Larger cohorts are needed to clarify the role that T/HS plays in RA disease activity and therapy.

Disclosure: J. E. Martin, None; C. P. Simeoń, None; N. Ortego-Centeno, None; P. Carreira, None; M. A. Gonzalez-Gay, None; N. Hunzelmann, None; S. Assassi, None; F. K. Tan, None; F. C. Arnett, None; X. Zhou, None; T. R. D. J. Radstake, None; M. D. Mayes, None; P. de Bakker, None; J. Martin, None; B. P. C. Koeleman, None.

177 Differences and Overlap Of Immunological Pathways Implicated In The Aetiology Of Anti-Citrullinated Peptide Antibody Positive and Negative Rheumatoid Arthritis. Sebastien Viatte1, Paul Martin2, Andrew Brass3, Mark Lunt4, Anne Barton4 and Stephen Eyre4. 1Arthritis Research UK Epidemiology Unit, Manchester Academic Health Science Centre, The University of Manchester, Manchester, United Kingdom, 2Arthritis Research UK Epidemiology Unit, Manchester Academic Health Sciences Centre, Institute of Inflammation and Repair, The University of Manchester, Manchester, United Kingdom, 3School of Computer Science, University of Manchester, Manchester, United Kingdom, Manchester, United Kingdom, 4 Arthritis Research UK Epidemiology Unit, University of Manchester, Manchester, United Kingdom. Background/Purpose: Genome-wide association studies have been successful in identifying single nucleotide polymorphisms (SNPs) associated with disease. However, previous studies have been under-powered to detect differences between anti-citrullinated peptide antibody (ACPA) positive and negative patients. Using a custom Illumina® Infinium® array, the RA Consortium for Immunochip (RACI) allows comprehensive analysis of disease associations in both ACPA positive and negative patients. Pathway analyses can identify whether a molecular pathway is associated with disease by testing for enrichment of genes associated with disease. This involves mapping SNP associations to genes, which can often be subjective.

Disclosure: R. Namas, None; R. Namas, None; H. Sagar, None; F. FernandezMadrid, None; T. Billiar, None; Y. Vodovotz, None.

S70

observed a significant overrepresentation of FCGR2A-27Q⬎W (OR 1.5 [95%CI 1.1–1.9, p⫽0.006) and the -386G⬎C promoter polymorphism of FCGR2B and FCGR2C (OR 1.5 [95%CI 1.1–1.9], p⫽0.004) in KD. No significant difference was found in the allele frequency of FCGR2A-131H⬎R in KD patients and controls (43.8% versus 45.4%, p⫽0.467). The frequency of CNV in the FCGR2C, FCGR3A and FCGR3B genes was not significantly different between the groups. A positive correlation was observed between gene expression and dosage of FCGR2C, FCGR3A and FCGR3B, which may affect the activating Fc␥Rs expression level on leucocytes. Results of the patient-parent trios data using transmission disequilibrium tests (TDT) will be used for further validation of our fine-mapping studies on the FCGR2/3 genes in KD. Conclusion: A significant overrepresentation of FCGR2C-ORF, as well as FCGR2A-27Q⬎W and the FCGR2B/C-386G⬎C promoter polymorphisms, was observed in Caucasian KD patients compared to the controls. Genetic variation at the FCGR2/3 gene cluster may lead to a shift in the balance towards the activating Fc␥Rs and may therefore contribute to the development of the inflammatory response in KD. Disclosure: C. E. Tacke, None; W. B. Breunis, None; L. T. Hoang, None; E. Png, None; J. Geissler, None; S. Nagelkerke, None; J. Ellis, None; S. Davila, None; C. C. Khor, None; M. Levin, None; D. Burgner, None; C. Shimizu, None; J. C. Burns, None; M. L. Hibberd, None; T. W. Kuijpers, None.

179 The Association Between Single Nucleotide Polymorphism ABCG2 rs2231142 and Hyperuricemia: A Case-Control Study. Bingqing Zhang1, Weigang Fang2, Yun Zhang2, Shufen Liu2, Yuanjie Li2 and Xuejun Zeng2. 1 Peking Union Medical College, Beijing, China, 2Peking Union Medical College Hospital, Beijing, China.

Disclosure: S. Viatte, None; P. Martin, None; A. Brass, None; M. Lunt, None; A. Barton, None; S. Eyre, None.

Background/Purpose: The prevalence of hyperuricemia and gout in China increased markedly in the past decades. Genome-wide association studies in Caucasian population have identified multiple gene loci associated with hyperuricemia and gout, including single nucleotide polymorphism (SNP) of rs2231142 in the ABCG2 gene, which leads to an amino acid change from glutamine to lysine at position 141 of the membrane transporter. However, this association was not consistently observed in Asian population. Methods: A case-control study was carried out to investigate the association between locus ABCG2 rs2231142 polymorphism and hyperuricemia in a Chinese cohort of faculty and staff members at an academic health center in Beijing. Blood samples were drawn and stored during their annual health examination in 2008. Physical examination and biochemical assay were performed for all participants at that time. Hyperuricemia was defined as serum urate ⱖ7 mg/dl in men or ⱖ6 mg/dl in women. Each case of hyperuricemia with glomerular filtration rate ’30ml/min/1.73m2 was matched with one or two controls that were randomly sampled from individuals with the same sex, age group, chronic kidney disease stage and body mass index classification as the case in the cohort. SNP rs2231142 was assayed for both cases and controls by amplification refractory mutation system polymerase chain reaction (ARMS-PCR) in 2013. Conditional logistic regression analysis was performed to study the association between SNP rs2231142 and hyperuricemia. Results: A total of 198 subjects and 370 controls were identified in the cohort. The overall A allele frequency of the ABCG2 gene was 30.81% in the cohort, but it was significantly higher in the cases than the controls (38.38% vs. 26.76%, p⬍0.01). Compared to individuals with genotype CC, the Odds Ratios (ORs) of hyperuricemia in individuals with A allele (genotype AA ⫹ CA), genotype CA and AA were 2.89 (95% CI 1.91–4.37, p⬍0.01), 2.84 (95% CI 1.86–4.32, p⬍0.01) and 3.30 (95% CI 1.60–6.81, p⬍0.01), respectively. After adjustment of hypertension, hyperglycemia and dyslipidemia, the ORs of A allele, genotype CA and AA were 2.99 (95% CI 1.94–4.62, p⬍0.01), 2.99 (95% CI 1.93–4.66, p⬍0.01) and 2.99 (95% CI 1.39–6.48, p⬍0.01), respectively. Subgroup analysis showed the ORs of A allele, genotype CA and AA were higher in males (3.83, 3.62, and 6.34, respectively) than those in females (2.30, 2.34, and 2.04, respectively), but the interaction of allele and gender was not significant with OR 1.66 (95% CI 0.71–3.87, p⫽0.23). Conclusion: ABCG2 SNP rs2231142 A allele is an independent risk factor of hyperuricemia in Chinese. The risk of hyperuricemia in individuals with A allele appears to be higher in males than that in females, but the interaction of allele and gender is not significant.

178 Fc-Gamma Receptor Genetic Variation In Kawasaki Disease. Carline E. Tacke1, Willemijn B. Breunis2, Long T. Hoang3, Eileen Png3, Judy Geissler2, Sietse Nagelkerke2, Justine Ellis4, Sonia Davila3, Chiea Chuen Khor3, Michael Levin5, David Burgner4, Chisato Shimizu6, Jane C. Burns6, Martin L. Hibberd3 and Taco W. Kuijpers2. 1Emma Children’s Hospital/Academic Medical Center (AMC), Amsterdam, Netherlands, 2Sanquin Resarch and Landsteiner Laboratory (AMC), Amsterdam, Netherlands, 3Genome Institute of Singapore, Singapore, Singapore, 4Murdoch Childrens Research Institute, Parkville, Australia, 5Imperial College London, London, United Kingdom, 6 UC San Diego, School of Medicine, La Jolla, CA. Abstract on behalf of the International Kawasaki Disease Genetics Consortium. Background/Purpose: Kawasaki disease (KD) is a pediatric vasculitis of unknown etiology with a substantial genetic contribution to susceptibility. KD is complicated by coronary artery aneurysms in 25% of untreated patients. Intravenous immunoglobulin (IVIG), a polyclonal IgG preparation, decreases the aneurysm rate to ⬍10%. In our recent GWAS, a functional variant in FCGR2A (131H⬎R; rs1801274) was associated with KD at genome-wide significance. FCGR2A encodes the Fc-gamma receptor (Fc␥R) IIa, one member of a family of glycoproteins on leukocytes that bind the Fc-domain of IgG. The balance between the various activating and a single inhibiting Fc␥R (Fc␥RIIb) determines the level of cell activation. Hence, altered Fc␥R expression may result in unbalanced immunity or autoinflammation. FCGR2A is located within the FCGR2/3 gene cluster. This cluster contains extensive gene copy number variations (CNVs) and several important single nucleotide polymorphisms (SNPs). We hypothesized that CNV and additional SNPs within the FCGR locus, regulating gene transcription and expression of the Fc␥Rs, may be important in KD susceptibility. Methods: The FCGR-specific Multiplex-Ligation Probe-dependent Amplification (MLPA) assay was performed in 426 patients and 710 controls of Caucasian origin, and in 510 patient-parent trios of mixed ethnic origin. First, the association between FCGR genotype status (including CNV, FCGR2A131H⬎R, FCGR2A-27Q⬎W, FCGR2C-ORF and FCGR2B -386G⬎C) and KD susceptibility was analyzed. Second, the genetic variation at the FCGR2/3 locus was correlated to the expression of these genes. Transcriptional expression data (pre-IVIG acute and post-IVIG convalescent samples) were available in a subset of 169 patients. Results: An increased allele frequency of the open-reading frame (ORF) of FCGR2C encoding an additional activating Fc␥RIIc was found in patients as compared to controls (OR 1.4 [95%CI 1.1–1.7], p⫽0.016). We also

Disclosure: B. Zhang, None; W. Fang, None; Y. Zhang, None; S. Liu, None; Y. Li, None; X. Zeng, None.

S71

Sunday, October 27

Objectives: The objectives of this study, therefore, were to develop a robust workflow-based method to assign SNPs to genes and to compare the pathways significantly enriched in ACPA positive and negative patients. Methods: Excluding HLA, the top 100 most associated SNPs from the Immunochip analysis were selected for ACPA positive patients along with the top 100 most associated SNPs from the ACPA negative analysis. Genes were assigned to these SNPs using a Taverna workflow which defines an associated region using SNPs in linkage disequilibrium (LD) with the associated SNP. This prevents the bias introduced by researchers assigning SNPs to ‘biologically plausible’ genes and provides a robust, reproducible method for assigning genetic associations to genes. These genes were subsequently tested for enrichment in PANTHER pathways and Gene Ontologies (GO) using the Expression Analysis tool to calculate an enrichment binomial probability. To correct for the enrichment of genes with immunological functions present on the Immunochip, the statistical significance of enrichment binomial probabilities was calculated by permutation. Results: Overall we show that the proportion of pathways shared between the two RA serological subtypes is significantly larger than expected by chance, indicating a high degree of overlap. On the other hand, we identified some pathways specifically and significantly enriched in the ACPA positive gene list (Interleukin Signalling Pathway: permutation p-value for enrichment ⫽ 7.0E-03, while in ACPA negative p ⫽ 0.46) and in the ACPA negative gene list (Interferon-Gamma Signalling Pathway: p ⫽ 1.2E-03, while in ACPA positive p ⫽ 0.33). Conclusion: This study provides the first comparative pathway analysis of RA serological subtypes based on large numbers of RA cases. The methods developed here present a quantitative framework to compare different diseases at the pathway level. The findings not only suggest shared processes between the two disease sub-groups, but also identify serotypespecific pathways.

Sunday, October 27

180

ACR Poster Session A Health Services Research, Quality Measures and Quality of Care – Rheumatoid Arthritis Sunday, October 27, 2013, 8:30 AM–4:00 PM

Familial Aggregation Of Female Premenopausal Gout –Monogenic, Polygenic Or Clinical Coincidence? Bingqing Zhang1, Shufen Liu2, Nuo Si3, Yun Zhang2 and Xuejun Zeng2. 1Peking Union Medical College, Beijing, China, 2Peking Union Medical College Hospital, Beijing, China, 3 Chinese Academy of Medical Science & Peking Union Medical College, Beijing, China.

181

Background/Purpose: Primary gout is a multifactorial disease, in which genetic background and environmental factors interact with each other. Genetic predisposition is particularly prominent in those seemingly low risk patients, such as the young and females. Women usually do not have gout until menopause, with menses as one of the most well-known protective factors of gout and hyperuricemia. We have found a rare family with three female gout patients, all of whom developed gout before menopause. The concentration of female patient with gout before menopause in this family highly suggests the involvement of genetic factors. The aim of this study is to explore the genetic factors related to gout in this female premenopausal gout family. Methods: Medical history, physical examination and laboratory work-up of this family were collected. Allele sharing examination and exon sequencing were carried out to study the causative genes of familial juvenile hyperuricemia nephropathy, including UMOD, RENIN, HNF1B and FJHN3, as well as HPRT1 of Lesch-Nyhan syndrome. Serum uric acid concentration quantitative trait locus ABCG2, SLC2A9, and SLC22A12 were screened by exon sequencing Results: The family was composed of 4 generations and 21 members (figure 1). All three patients were females; none of the male members had gout or hyperuricemia. The proband (III-5) was a female patient who had gout at the age 28; she started allopurinol since the second attack, but her renal function deteriorated. Her mother (II-5) had recurrent acute monoarthritis since the age of 16; she was diagnosed with gout at the age of 38, with obvious decreased renal function. The aunt of the proband (II-1) had gout at the age of 46, when she still had menses. None of the three patients had other known disease prior to gout attack.

Rheumatoid Arthritis Quality Measure Results In a Large Managed Care Population. Roxanne Meyer1, Lorie A. Ellis2, Susan C. Bolge2, Andrew Howe3 and Yunping Zhou3. 1Janssen Scientific Affairs, LLC, Horsham, NY, 2Janssen Scientific Affairs, LLC, Horsham, PA, 3Comprehensive Health Insights, Louisville, KY. Background/Purpose: Improvement in quality of care is of growing interest to both physicians and payers. Both the American College of Rheumatology and National Quality Forum (NQF) recommend monitoring of quality measures in rheumatoid arthritis (RA) patients. The purpose of this study is to describe the proportion of RA patients in a large managed care population that achieved NQF quality measures specific to RA. Methods: This is a retrospective analysis of Humana’s Commercial and Medicare database. Claims were reviewed to evaluate compliance with four NQF quality measures in calendar year 2011. NQF definitions were used for the numerator (N) and the denominator (D) for each measure. Descriptive statistics were calculated on the specific measures. NQF Measure 0054 Arthritis: disease modifying antirheumatic drug (DMARD) therapy in rheumatoid arthritis 0590 Rheumatoid Arthritis New DMARD Baseline Liver Function Test

0592 Rheumatoid Arthritis Annual ESR or CRP

0601 New Rheumatoid Arthritis Baseline ESR or CRP within Three Months

Definition Percentage of patients 18 years and older, diagnosed with rheumatoid arthritis who have had at least one ambulatory prescription dispensed for a DMARD This measure identifies adult patients with a diagnosis of rheumatoid arthritis who received appropriate baseline liver function testing (AST or ALT) within 90 days before to 14 days after the new start of sulfasalazine, methotrexate, leflunomide, azathioprine, cyclosporine or cyclophosphamide during the measurement year. This measure identifies adult patients with a history of rheumatoid arthritis who have received erythrocyte sedimentation rate (ESR) or C-reactive protein (CRP) lab tests during the measurement year This measure identifies adult patients newly diagnosed with rheumatoid arthritis during the first 8 months of the measurement year who received erythrocyte sedimentation rate (ESR) or C-reactive protein (CRP) lab tests either 4 months (3 months ⫹ 1-month grace period) before or after the initial diagnosis.

Results: The proportion of adult RA patients who received at least one prescription for a DMARD (measure 0054) was 72.8% (N⫽17,066, D⫽23,440). Liver function testing in patients with a new DMARD as defined by measure 0590 was achieved by 84.7% (N⫽3,130, D⫽3,694) of adult RA patients. The proportion of newly diagnosed adult patients receiving a baseline ESR or CRP (measure 0601) was 54.7% (N⫽2,595, D⫽4,742). For measure 0592 the proportion of adult patients receiving an annual ESR or CRP was 58.2% (N⫽13,125, D⫽22,538). Conclusion: This analysis of an RA population in a large managed care population suggests that 15% to 45% of patients do not attain NQF quality measure standards. These findings suggest that improvements in patient care could be made with the goal of improving the proportion of patients achieving RA specific NQF measures. Further studies are warranted to understand whether attainment of these measures translates into better overall health or RA-related outcomes and may enhance physician and patient awareness about the benefits of quality of care.

Figure 1. Pedigree of the family.

RENIN, the causative gene of Hyperuricemic Nephropathy, Familial Juvenile, type2 (HNFJ, MIM: #613092) was found to be shared in this family but not UMOD, HNF1B or FJHN3. Exon sequencing of RENIN and UMOD of patient II-5 was proved to be normal. HPRT1 of the rare Lesch-Nyhan syndrome was excluded by exon sequencing. Multiple single nucleotide polymorphisms of gene ABCG2, SLC2A9 and SLC22A12 were found in patient II-5, including one locus known to be highly associated with gout, SNP rs2231142 (C⬎A) of gene ABCG2. Nearly all of the family members carry the risk A allele. Conclusion: In this premenopausal gout family, all known causative genes of gout were found to be negative. ABCG2 rs2231142 (C⬎A) polymorphism was found in both female patients and male healthy members of this family.

Disclosure: R. Meyer, Janssen Scientific Affairs, LLC, 3; L. A. Ellis, Janssen Scientific Affairs, LLC, 3; S. C. Bolge, Janssen Scientific Affairs, LLC, 3; A. Howe, Janssen Scientific Affairs, LLC, 5; Y. Zhou, Janssen Scientific Affairs, LLC, 5.

Disclosure: B. Zhang, None; S. Liu, None; N. Si, None; Y. Zhang, None; X. Zeng, None.

S72

182 National Quality Forum Measures Among Rheumatoid Arthritis Patients In a Large Managed Care Population. Roxanne Meyer1, Lorie A. Ellis2, Susan C. Bolge2, Joseph Tkacz3, Peter Kardel3 and Charles Ruetsch3. 1Janssen Scientific Affairs, LLC, Horsham, NY, 2Janssen Scientific Affairs, LLC, Horsham, PA, 3Health Analytics, LLC, Columbia, MD.

NQF Measure Results 2007–2011

2007

2008

2009

2010

2011

21,156 70.0%

26,514 71.2%

27,034 71.1%

24,852 72.1%

26,420 70.5%

2,438 32.3%

3,868 31.3%

3,388 35.8%

3,181 38.3%

3,160 27.7%

2,681 29.7%

4,183 30.3%

3,716 36.5%

3,524 39.0%

3,488 26.7%

14,792 58.4%

21,702 57.7%

25,474 55.8%

25,416 54.9%

26,776 55.5%

7,126 30.9%

8,743 33.2%

8,589 37.4%

7,977 37.3%

8,285 33.8%

7,126 72.4%

8,743 72.7%

8,589 72.4%

7,977 72.8%

8,285 71.8%

7,126 38.6%

8,743 41.0%

8,589 44.7%

7,977 45.0%

8,285 41.6%

4,144 67.8%

4,283 69.2%

4,540 71.6%

4,709 71.5%

Conclusion: This analysis of a large national health plan suggests that between 30% and 70% of RA patients do not meet NQF quality measure criteria. Further studies are needed to understand: 1) the relationship between NQF measures and health outcomes and cost; 2) drivers of meeting NQF quality standards; and 3) interventions that improve NQF scores within health plans.

Background/Purpose: The American College of Rheumatology and National Quality Forum (NQF) recommend monitoring quality measures among rheumatoid arthritis (RA) patients. This study describes the proportion of RA patients within a large managed care population meeting the criteria of RA specific NQF quality measures. Methods: Definitions of NQF RA quality measures 0054, 0589, 0590, 0592, 0597–0599, and 0601 were applied to claims data of a commercially-insured population from Optum Insight’s Clinformatics database for calendar years 2007–2011. NQF definitions may be found at www.qualityforum.org/Home.aspx. Results: Measure 0054 defines the proportion of RA patients treated with a DMARD in a defined measurement year. At each year studied, approximately 70% achieved this measure. Measures 0589, 0590 pertain to laboratory monitoring of adult RA patients with a new DMARD in the measurement year. Measures 0589 (baseline serum creatine) and 0590 (baseline liver function test; LFT) were achieved by only one-third of patients. The proportion of patients achieving these measures appeared to increase annually between years 2007 and 2010 but declined during 2011. Measures 0597, 0598 and 0599 were designed to monitor adult RA patients within 12 weeks of a new methotrexate (MTX) prescription. Approximately one-third of patients met the criteria for Measure 0597 (LFT within 12 weeks), and 40% met the criteria for 0599 (serum creatinine within 12 weeks). In contrast, approximately 70% of new MTX-treated patients met the criteria for 0598 (complete blood count within 12 weeks). Slight variations in the proportions of patients achieving these measures were observed from year to year. Measures 0592 and 0601 pertain to monitoring of ESR or CRP in adult RA patients. Measure 0601 (proportion of newly diagnosed RA patients who received an ESR or CRP measure within 3 months of diagnosis) was achieved by approximately 70% of patients. Proportions of patients achieving this measure were improved slightly over time. However, annual ESR or CRP (0592) was achieved in slightly more than half the patients.

0054 Arthritis: disease modifying antirheumatic drug (DMARD) therapy in rheumatoid arthritis Denominator (f) Percent (%) 0589 Rheumatoid Arthritis New DMARD Baseline Serum Creatinine Denominator (f) Percent (%) 0590 Rheumatoid Arthritis New DMARD Baseline Liver Function Test Denominator (f) Percent (%) 0592 Rheumatoid Arthritis Annual ESR or CRP Denominator (f) Percent (%) 0597 Methotrexate: LFT within 12 weeks Denominator (f) Percent (%) 0598 Methotrexate: CBC within 12 weeks Denominator (f) Percent (%) 0599 Methotrexate: Creatinine within 12 weeks Denominator (f) Percent (%)

3,249 66.2%

Disclosure: R. Meyer, Janssen Scientific Affairs, LLC, 3; L. A. Ellis, Janssen Scientific Affairs, LLC, 3; S. C. Bolge, Janssen Scientific Affairs, LLC, 3; J. Tkacz, Janssen Scientific Affairs, LLC, 5; P. Kardel, Janssen Scientific Affairs, LLC, 5; C. Ruetsch, Janssen Scientific Affairs, LLC, 5.

183 The Performance Of a Point Of Care Test For Detection Of AntiMutated Citrullinated Vimentin and Rheumatoid Factor In Early Rheumatoid Arthritis. Preeda Rojanasantikul1, Prapa Pattrapornpisut2, Kulvara Anuruckparadorn3 and Wanruchada Katchamart1. 1Division of Rheumatology, Department of Internal Medicine, Siriraj Hospital, Mahidol University, Bangkok, Thailand, 2Department of Internal Medicine Siriraj Hospital, Mahidol University, Bangkok, Thailand, 3Department of Transfusion Medicine, Siriraj Hospital, Mahidol University, Bangkok, Thailand. Background/Purpose: Besides rheumatoid factor (RF), anti-cyclic citrullinated peptide antibody (ACPA), a newer serologic marker, was added to the new American College of Rheumatology (ACR)/European League Against Rheumatism (EULAR) classification criteria 2010. Most commercially available assay used for detection of ACPA and RF is the ELISA-based method. This assay is a multi-step and time-consuming procedure requiring special equipments and experienced technicians. ACPA therefore may not be available worldwide. Moreover, the results of these tests may take too long to be of value in daily practice, leading to delayed diagnosis. Recently, a serological point-of-care test (POCT) for detection of anti-MCV and RF using a lateral flow immunochromatographic assay (LFIA) has been developed for a rapid result, and it also has wider availability. The diagnostic performance of this POCT was documented in established RA; however, its performance in patients with ’early’ RA is still unknown. Our objective was to determine the diagnostic performance of a point of care test (POCT) for detection of anti-mutated citrullinated vimentin (MCV) and rheumatoid factor (RF) in early rheumatoid arthritis (RA) with 2 years of disease duration or less. Additionally, we evaluated the agreement of these tests when using EDTA whole blood and capillary blood. Methods: Patients with RA and other rheumatic disorders were consecutively recruited from the Rheumatology outpatient clinic. The POCT for detection of anti-MCV and RF using capillary blood and EDTA whole blood was performed in 78 patients with early RA, 55 patients with other rheumatic disorders, and 55 healthy blood donors. Results: The sensitivity and specificity of anti-MCV POCT in patients with early RA were 64% and 97%, respectively, while the sensitivity and specificity of RF POCT were 51% and 95% respectively. The positive likelihood ratio of the POCT for anti-MCV was higher than those for RF (23.5 vs.9.4). The negative likelihood was 0.37 for anti-MCV and 0.52 for RF. There were 3 cases with false positive for anti-MCV including a patient with psoriatic arthritis and the other two with systemic sclerosis. The agreement between capillary blood and EDTA whole blood testing for anti-MCV and RF was low to moderate with Cohen’s kappa of 0.58 and 0.49, respectively. Conclusion: This POCT for detection of anti-MCV and RF yielded high specificity and may be a valuable tool for the diagnosis of early RA. Using this POCT with EDTA whole blood instead of capillary blood is not recommended. Disclosure: P. Rojanasantikul, None; P. Pattrapornpisut, None; K. Anuruckparadorn, None; W. Katchamart, None.

S73

Sunday, October 27

0601 New Rheumatoid Arthritis Baseline ESR or CRP within Three Months Denominator (f) Percent (%)

diagnosis, discordance (i.e. difference between patient’s and doctor’s global scores ⬎20 mm) was independent of physician’s age and gender. This is reassuring for the monitoring of patients in routine care. Having specialized in rheumatology seemed to be associated with increased odds of discrepancy compared to less experienced colleagues.

Sunday, October 27

184 Are Physician Gender, Age and Clinical Experience Associated With Discrepancy In Global Disease Score In Rheumatoid Arthritis, Ankylosing Spondylitis and Psoriatic Arthritis? Data From The Nationwide Danbio Registry. Cecilie Lindstrom Egholm1, Niels Steen Krogh2, Lene Dreyer3, Torkell Ellingsen4, Bente Glintborg5, Marcin Kowalski6, Tove Lorenzen7, Ole Rintek Madsen8, Henrik Nordin9, Claus Rasmussen10 and Merete L. Hetland11. 1Regional Research Unit, Region Zealand, Roskilde, Denmark, 2ZiteLab ApS, Copenhagen, Denmark, 3Copenhagen University Hospital at Gentofte, Copenhagen, Denmark, 4Diagnostic Centre Region Hospital Silkeborg Denmark, 8600 Silkeborg, Denmark, 5Copenhagen Center for Arthritis Research, Center for Rheumatology and Spine Diseases, Glostrup Hospital, Glostrup, Denmark, 6Aalborg University Hospital, Aalborg, Denmark, 7Department of Rheumatology, Region Hospital Silkeborg, Silkeborg, Denmark, 8Copenhagen University Hospital in Gentofte, Copenhagen, Denmark, 9Department of Infectious Diseases and Rheumatology, Rigshospitalet, Copenhagen, Denmark, 10Vendsyssel Teaching Hospital/ Aalborg University, Hjoerring, Denmark, 11Copenhagen University Hospital Glostrup, Copenhagen, Denmark.

Disclosure: C. L. Egholm, None; N. S. Krogh, None; L. Dreyer, None; T. Ellingsen, None; B. Glintborg, None; M. Kowalski, Abbvie, 6; T. Lorenzen, Phizer, Roche, 6; O. R. Madsen, None; H. Nordin, None; C. Rasmussen, None; M. L. Hetland, None.

185 No Secular Trends In Real Live Radiographic Progression Of Rheumatoid Arthritis In Recent Year. Miriam Ga¨rtner1, Farideh Alasti1, Gabriela Supp1, Josef S. Smolen2 and Daniel Aletaha1. 1Medical University of Vienna, Vienna, Austria, 2Medical University of Vienna and Hietzing Hospital, Vienna, Austria. Background/Purpose: Progression of joint damage as measured radiographically is a hallmark of rheumatoid arthritis (RA), and its reduction a key claim for traditional and novel antirheumatic drugs. Data from clinical trials may be interpreted to indicate that the observed progression of RA has decreased over the past two decades.1 Although these observations mainly stem from clinical trials, it has been hypothesized that RA in general may have become a less destructive disease. It was the aim of this study, to evaluate the frequency and degree of radiographic progression in real life routine RA patients over the past decade Methods: RA patients of our outpatient clinic, who had two x-rays performed over a three to five year interval between the years 2000 and 2012 were included. X-rays of the hands and feet were scored according to the Sharp van der Heijde (SvdH; range 0–448) method.2 Clinical and demographic data were collected from the patients’ charts. Patients were separated into three periods of time (2000–2004, 2005–2008, 2009–2012) and the annual radiographic progression was calculated based on the observed progression in these periods. Disease activity was assessed using the time averaged clinical disease activity index (CDAI). The mean number of joints with radiographic progression per patient, as well as the mean grade of progression per joint (erosion/joint space narrowing, JSN) were evaluated. Lack of progression was defined as no progression in total SvdH Results: Of the 444 patients included (mean duration of RA 7.4⫾9.4 years, 64.8% rheumatoid factor positive, 63.7% ACPA positive), 406 (91.4%) showed radiographic progression over a mean of 3.8⫾0.6 years. We found no significant difference neither in clinical nor in radiographic results comparing patients between the three time periods. In all progressors we found no difference in rates of annual radiographic progression, erosion score, or JSN, within the three different time periods (Figure 1). Although the annual rate of progression was stable, baseline SvdH scores decreased across the three periods (43.2⫾57.4 2000–2004 vs 43.0⫾49.7 2005–2008 vs 33.2⫾48.9 2009–2012).

Background/Purpose: A global estimate on a 100 mm Visual Analogue Scale (VAS) assessed by patients (PATGL) and physicians (DOCGL) is commonly used to measure disease activity and to monitor treatment response in patients with rheumatoid arthritis (RA), ankylosing spondylitis (AS) and psoriatic arthritis (PsA). Several studies in RA have shown that PATGL and DOCGL are discordant (⬎20 mm difference) in about 1/3 of the encounters and have mainly focused on whether clinical measures and patient characteristics explain this. The influence of physician characteristics has not yet been examined across different arthritis diagnoses. Studies in other clinical areas suggest that treatment decisions are affected by physician gender and clinical experience. The aim of the study was therefore to assess whether physician characteristics are associated with discordance between DOCGL and PATGL in patients with RA, AS and PSA. Methods: Physician characteristics were collected by a questionnaire sent to DANBIO physicians (n ⫽ 265). Patient characteristics and clinical measures as well as PATGL and DOCGL scores were obtained from DANBIO first encounters. A difference between DOCGL and PATGL of up to 20 mm was considered concordant, yielding three groups: PATGL⬎DOCGL, PATGL⫽DOCGL and DOCGL⬎PATGL. The two latter groups were merged due to few patients in DOCGL⬎PATGL. We used mixed effects logistic regression analyses with discordance (yes/no) as the dependent variable, performing independent analyses for each diagnosis. The model was adjusted for patient and clinical variables (e.g. age, gender, disease activity, treatment, disease duration), which were entered as covariates. Results: 90 physicians (44% females, 61% rheumatologists, median age 52 years) returned the questionnaire (34%) and were pairwise matched with 7,619 RA, 1,291 PsA and 469 AS first encounters. Discordance was independent of physician’s gender and age, both for patients with RA, and for patients with AS or PSA, see table. Less experienced physicians (i.e. not yet specialized) had lower odds of discordance 0.69–0.92, although it only reached statistical significance in patients with RA, see table. Table. Mixed effects logistic regression of predictors of discordance between patient and physician assessments of global disease activity. RA (N: 7,619 encounters) Adjusted OR (95% CI) P Physician gender, male (female⫽1.0) Physician age ⬎52 years (ⱕ52 ⫽1.0) Rheumatologist (specialized), no (yes⫽1.0)

AS (N: 469 encounters) Adjusted OR (95% CI)

p

PsA (N: 1,291 encounters) Adjusted OR (95% CI) p

0.83 (0.69–1.01)

NS

1.25 (0.70–2.26)

NS

1.04 (0.73–1.49)

NS

1.10 (0.89–1.37)

NS

1.16 (0.59–2.30)

NS

1.17 (0.79–1.74)

NS

0.72 (0.55–0.93)

*

0.69 (0.31–1.52)

NS

0.92 (0.56–1.51)

NS

Concordant group is reference group. The model was adjusted for swollen and tender joint counts (0–28), C-reactive Protein, disease duration (years), biological treatment (yes/no), country of medical exam, and doctor’s subjective rating of factors important for DOCGL (inflammation, fibromyalgia, structural joint damage, comorbidity and social factors). To account for clustering by physician and patient, these variables were included in the model as random effects. ***⫽p⬍0.001, **⫽p⬍0.01, *⫽p⬍0.05, NS⫽not significant

Figure 1. Patients with radiographic progression (n⫽406) during three time periods. The figure shows the mean annual radiographic progression rates for total SvdH-Score, erosions as well as JSN. There was no significant difference between the different time periods.

Conclusion: Data from clinical practice covering ⬎9,000 DANBIO first encounters between 90 physicians and their patients with rheumatoid arthritis, ankylosing spondylitis and psoriatic arthritis showed that, regardless of

S74

Conclusion: The delay in starting treatment with DMARDs was about 5 months. However, one-quarter of patients took longer than one year to start the treatment. The implementation of early arthritis clinics, a fast referral system and limitation of health barriers could be a good strategy to optimize the prompt treatment of the patients with RA.

References: (1) Rahman MU et al. Ann Rheum Dis 2011. (2) van der Heijde D. J Rheumatol 1999.

Disclosure: N. Zamora, None; C. A. Waimann, None; M. F. Marengo, None; G. Citera, None; A. Granel, None; A. I. Marcos, None; E. Buschiazzo, None; R. V. Juarez, None; D. Pereira, None; G. Pendoń, None; A. D’Orazio, None; M. Salcedo, None; J. Sarano, None; M. V. Collado, None; G. Go´mez, None; A. Medina, None; M. E. Lara, None; J. C. Barreira, None; M. A. La´zaro, None; A. Cusa, None; L. Gonza´lez Lucero, None; M. S. Yacuzzi, None; R. N. Martıńez, None; P. Arturi, None.

Disclosure: M. Ga¨rtner, None; F. Alasti, None; G. Supp, None; J. S. Smolen, None; D. Aletaha, None.

186

187

Delay In Consultation and Starting Disease Modifying Anti-Rheumatic Drugs In Patients With Rheumatoid Arthritis In Argentina. How Early Arthritis Clinics Impact On Health Barriers? Natalia Zamora1, Christian A. Waimann1, Maria Florencia Marengo1, Gustavo Citera1, Amelia Granel2, Ana Ine´s Marcos2, Emilio Buschiazzo3, Ricardo V. Juarez3, Dora Pereira4, Gisela Pendoń4, Andrea D’Orazio5, Mariana Salcedo6, Judith Sarano7, Marıá Victoria Collado7, Graciela Go´mez7, Alejandra Medina8, Marıá Eugenia Lara9, Juan C. Barreira9, Marıá Alicia La´zaro10, Alejandra Cusa10, Luciana Gonza´lez Lucero11, Marıá Silvia Yacuzzi11, Rau´l Nicolas Martıńez11 and Pablo Arturi12. 1Instituto de Rehabilitacioń Psicofı´sica, Buenos Aires, Argentina, 2Hospital San Roque de Gonnet, La Plata, La Plata, Argentina, 3Hospital Senõr del Milagro, Salta, Argentina, 4Hospital Ricardo Gutierrez, La Plata, La Plata, Argentina, 5Hospital de la Asociacion Medica de Bahıá Blanca, Bahıá Blanca, Argentina, 6Consultorio Privado, San Nicola´s, Argentina, 7Instituto de Investigaciones Medicas Alfredo Lanari, Buenos Aires, Argentina, 8Hospital General de Agudos “Dr. E. Tornu”, Ciudad Autońoma de Buenos Aires, Argentina, 9Buenos Aires British Hospital, Buenos Aires, Argentina, 10Hospital central de San isidro Dr A. Posse, San Isidro, Argentina, 11Hospital Angel C. Padilla de San Miguel de Tucumań, San Miguel de Tucumań, Argentina, 12Hospital Italiano de La Plata, La Plata, Argentina.

The Health and Economic Consequences Of Delay In Starting DiseaseModifying Anti-Rheumatic Drugs In Australian Patients With Early Rheumatoid Arthritis. Danny Liew, Mark Tacey and Sharon Van Doornum. The University of Melbourne, Melbourne, Australia. Background/Purpose: Several international studies suggest that the time between symptom onset and DMARD initiation in RA patients is longer than is considered optimal. We sought to assess the health economic impact of this delay in an Australian context. Methods: The delay in DMARD initiation was estimated from a study of 135 Australian RA patients referred to hospital-based and community-based rheumatologists. RA-associated utilities and costs were sourced from published data. Patients not taking and taking DMARD therapy were assumed to have utilities of 0.59 and 0.64, respectively. The annual direct cost of DMARD therapy was assumed to be AUD$1200, consistent with the most expensive first-line agents in Australia. It was also conservatively assumed that DMARD therapy did not reduce non-DMARD RA costs. (NB: The Australian dollar is near parity with the US dollar.) Results: The median time from time of symptom onset to initiation of DMARD therapy was 0.46 years. Over this time a mean of 0.27 qualityadjusted life years (QALYs). Had DMARDs been commenced at symptom onset, 0.30 QALYs would have been lived per patient, and AUD$555 of additional healthcare costs incurred. Hence early initiation of DMARDs would have saved 0.023 QALYs, equating to an incremental costeffectiveness ratio (ICER) of AUD$24,000 per QALY saved. An additional AUD$600 could be spent per patient to reduce the time to DMARD initiation before the ICER breached the arbitrary cut-off of AUD$50,000 per QALY saved. Our analysis was conservative in it did not consider the long-term health and cost savings associated with avoidance of permanent joint damage. Conclusion: The considerable delay in the initiation of DMARD therapy among patients with RA leads to significant health loss. Reducing the time to initiation of DMARDs represents a cost-effective means of reducing the burden of RA

Background/Purpose: An early and intensive treatment has become the cornerstone in the treatment of Rheumatoid Arthritis (RA). For this reason, an early consult with a rheumatologist is crucial. The aim of our study was to evaluate the delay in consultation and starting disease modifying anti-rheumatic drugs (DMARDs) in patients with RA, and to assess the impact of early arthritis clinics and health barriers on such delay. Methods: We carried out a retrospective multicenter study including patients with a diagnosis of RA of less than 5 years of disease evolution. Data collected included clinical, economic, sociodemographic characteristics, health insurance status and health care center (private vs. public hospital; early arthritis clinics vs. routine care hospitals). In addition, we evaluated the presence of health barriers in those patients with ⬍2 years of disease duration, including geographical location, social support, family and work responsibilities, economics issues, accessibility to health care, self-treatment and patientphysician relationship. Three time-points were evaluated: time to first medical contact, time to rheumatologist and time to initiation of first DMARD. The association between variables and time-points were assessed using univariate and multivariate models. Results: A total of 316 patients were included; 86% were female, mean age was 47 ⫾ 14 years, disease duration 7 ⫾ 5 years, 73% had a low income (less than 1000 dollars/month), and 11 ⫾ 4 years of formal education. Two third were from public hospitals and one third from private sector, 8% lived alone and 23% were unemployed. The median time to first medical contact was 30 (IQR 10–60) days, being the clinician the initial physician (52%) followed by traumatologist and rheumatologist (24% for both). The total time to arrive to the rheumatologist was 90 (IQR 35–210) days, but 32% of the patients delayed more than six months to contact a rheumatologist. Twenty five percent of patients were never referred by their clinicians duplicating the access time to rheumatologists to 165 (IQR 30–365) days. The median time to start DMARDs treatment by the rheumatologist was 24 (IQR 6–365) days, which result on a total median time from onset of symptoms of five months. One quarter of patients took longer than 12 months to start a DMARD. Three health barriers were significantly associated with delay in consultation and treatment: geographical location, family and work responsibilities, and lack of economic issues. After adjusting for multiples confounders (health insurance, health care center, marital status, household members, comorbidities and age), the presence of health barriers was independently associated with increased delay to rheumatologist and initiation of the first DMARD. In addition, being attended in early arthritis clinics was significantly associated to shorter delay in receiving DMARD treatment.

Disclosure: D. Liew, Abbvie Australia, 2, Abbvie Australia, 9; M. Tacey, None; S. Van Doornum, AbbVie Pty Ltd, 2.

188 Can MDHAQ/RAPID3 Be Used In Usual Care Of Patients With Systemic Lupus Erythematosus? Narender Annapureddy1, Theodore Pincus2, Joel A. Block1 and Meenakshi Jolly1. 1Rush University Medical Center, Chicago, IL, 2NYU Hospital for Joint Diseases and New York University School of Medicine, New York, NY. Background/Purpose: Disease-specific patient reported outcome (PRO) tools are uniquely informative for the management of systemic lupus erythematosus (SLE) patients. However, it is difficult to incorporate distribution of appropriate disease specific PRO measures in busy clinical settings. A single, simple PRO measure for all patients, irrespective of their underlying disease diagnosis, can easily be integrated into the clinic check-in processes without requiring additional processes/ personnel. Multi-Dimensional Health Assessment Questionnaire (MDHAQ) is a PRO tool developed for patients with musculoskeletal diseases; RAPID 3 (Routine Assessment of Patient Index Data) scores are easily and rapidly calculated from the MDHAQ. We compared RAPID3 and LupusPRO (a disease specific PRO tool for SLE) and disease activity indices in patients with SLE seen during usual care. Methods: 78 consenting patients meeting 4 ACR criteria for SLE completed both MDHAQ and LupusPRO during a routine clinic visit. Disease activity was assessed using SELENA-SLEDAI (Physician Global Assessment-PGA 0–3), and demographic data were obtained. We compared

S75

Sunday, October 27

Conclusion: Only a minority of patients did not progress structurally over three to five years. The temporal trend analysis indicates that on the background of similar average disease activity and similar mix of treatment, the overall annual progression was constant during the last decade. However, in general progression of joint damage was low.

Sunday, October 27

LupusPRO domain scores for 8 Health Related Quality of Life-HRQOL and 4 Non-Health Related Quality of Life-Non-HRQOL scales (all scored 0–100,higher scores⫽Better Quality of Life) and physician global assessment (DOCGL), SELENA-SLEDAI stratified by disease severity to RAPID3 scores in 4 categories,ⱕ 3 Remission; 3.1–6.0 Low severity; 6.1–12 Medium Severity and ⬎12 High severity. Statistical significance was analyzed using Kruskall-Wallis non-parametric tests; a p value of ⱕ 0.05 was considered significant on two tailed tests. Results: Mean (SD) age, Physician global assessment (or DOCGL) and total SELENA-SLEDAI scores were 42.4 (13.8) yrs, 0.7 (0.7) and 3.9 (4.7) respectively. Mean (SD) RAPID3 score was 9.8 (7.6). Table 1 shows the median (IQR) disease activity and LupusPRO domain scores for the RAPID3 disease severity categories. Disease activity scores did not vary between RAPID3 disease severity categories. This phenomenon has been previously reported with all PRO tools (Generic or Disease Specific) used in SLE. Seven of the 12 LupusPRO domains differed significantly in patients in the 4 RAPID3 disease severity categories - lupus symptoms, cognition, physical health, pain-vitality, emotional health, body image, and desires- goals, while 5/12 domains - Procreation, Lupus Medications, Social Support, Coping and Satisfaction with Treatment, did not differ in the 4 groups. The composite HRQOL LupusPRO scores differed significantly in the RAPID3 disease severity categories, while the composite non-HRQOL did not differ.

questions on methotrexate self-injection (graded as a score out of 6). The nurse recorded the number of minutes spent providing direct education to the patient Results: 29 patients participated in this study: 15 had standard (S) teaching and 14 were in the intervention group (VS). Average age (49), gender (VS⫽6 males, S⫽5), & education level was similar in both groups. Both groups were very satisfied with the quality of teaching (9.9/10). There was no difference in pre-confidence (S⫽5.5/10 vs. VS⫽4.7/10, p⫽0.44) or post-confidence (S⫽8.8, VS⫽8.8, p⫽0.93) between the groups, although there was a trend to greater improvement in confidence in the video group (p⫽0.15). Further, there was a trend towards improved patient knowledge in the video group vs. the standard group, with an average of more correct answers in the video group (S⫽4.7/6, VS⫽5.5/6, p⫽0.15). Nurse teaching time was significantly less in the video group compared to the standard group (S⫽60 minutes, VS⫽44 minutes, p⫽0.012). Conclusion: A web based education video may be a good supplement to standard in person nurse teaching for methotrexate self-injection. It has equally as good benefit on patient confidence and knowledge base while decreasing teaching time by 25%. Disclosure: S. J. Katz, None; S. Leung, None.

190

Table 1. Disease Activity and LupusPRO scores stratified by RAPID3 Disease Severity. Median (IQR)

RAPID 3 Based Disease Severity 0–3 3.1–6

Physician Global Assessment Total SELENA-SLEDAI LupusPRO-HRQOL Lupus Symptoms Cognition Procreation Lupus Medications Physical Health Pain-Vitality Emotional Health Body Image Composite HRQOL LupusPRO-Non-HRQOL Desires-Goals Social Support Coping Satisfaction Treatment Composite Non-HRQOL

6.1–12

>12

0.0 (1.0) 2.0 (4.0)

0.25 (0.50) 2.0 (6.0)

1.0 (1.0) 4.0 (5.8)

0.5 (1.1) 4.0 (4.0)

91.7 (8.3) 100 (37.5) 100.0 (0.0) 100.0 (0.0) 100.0 (0.0) 95.0 (20.0) 83.3 (33.3) 100 (15.0) 91.7 (14.1)

90.6 (14.6) 81.3 (43.8) 100.0 (0.0) 75.0 (46.9) 100 (12.5) 75.0 (20.0) 75.0 (11.5) 97.5 (38.8) 86.7 (4.3)

66.7 (33.3) 56.3 (46.9) 100.0 (0.0) 81.3 (50.0) 80.0 (43.8) 50.0 (30.0) 79.2 (45.8) 77.5 (38.8) 76.3 (19.4)

50.0 (41.7) 50.0 (50.0) 100.0 (0.0) 87.5 (37.5) 55.0 (32.5) 30.0 (17.5) 45.8 (30.2) 50 (57.6) 57.9 (18.6)

⬍0.001 ⬍0.001 0.120 0.030 ⬍0.001 ⬍0.001 ⬍0.001 ⬍0.001 ⬍0.001

93.8 (25.0) 100 (37.5) 83.3 (41.7) 87.5 (62.5) 75.0 (26.0)

78.1 (29.7) 100.0 (0.0) 91.7 (25.0) 100 (0.0) 87.5 (9.9)

68.8 (25.0) 87.5 (50.0) 66.7 (41.7) 100 (34.4) 75.0 (30.7)

46.9 (42.2) 93.8 (40.6) 75.0 (27.1) 90.6 (31.3) 72.7 (12.9)

⬍0.001 0.227 0.314 0.146 0.300

A Meta-Analysis and Systematic Literature Review: Effect Of The Method Used To Measure Adherence In Rheumatoid Arthritis On Its Rate, and Evaluation Of Associated Factors. Anat Scheiman-Elazary1, Cortney Shourt1, M. Cameron Hay2 and Daniel Furst3. 1Rheumatology UCLA, Los Angeles, CA, 2Anthropology Miami University, Oxford, OH, 3 David Geffen School of Medicine, UCLA, Los Angeles, CA.

P value 0.101 0.130

Background/Purpose: Adherence reported in the literature for RA varies widely (49.5–98%)1. This variability may result in part from different methods used to measure adherence. Our aim was to examine whether adherence is influenced by the method used to measure it through a meta-analysis. We also conducted a systematic literature review (SLR) on factors that are associated with adherence. Methods: Systematic literature review was conducted using PubMed and Cochrane central database as well as associated reference lists from reviews from 1970 to Nov. 2012. Papers were included if they reported DMARD adherence data or factors associated with adherence in adult RA patients and if adherence was defined. Aspirin/NSAID or non pharmaceutical therapy were excluded. Papers were divided according to the method used to measure adherence. A Random effects model estimated adherence and a Binomial significance test was used to identify significant factors associated with adherence. Results: Out of an initial 273 articles, 26 papers were included, including 34,700 RA patients. Figure 1 details the results of the meta-analysis for adherence, showing 2–8 articles for each method. Adherence rates were statistically similar when measured by interview, questionnaires, pill count, drug level or medication effect monitoring system (MEMS), although there was an 11% mean numerical range. There was a numerically but not statistically lower adherence when measured by prescription claims (0.59 for prescription claims versus 0.72–0.83 for the other methods).

Conclusion: RAPID3 may be used as a PRO for HRQOL in busy clinical settings to add to management of SLE patients. Disclosure: N. Annapureddy, None; T. Pincus, None; J. A. Block, Ferring, Inc., 5, PL Pharma, Inc., 9; M. Jolly, GSK, 5, Medimmune, 5.

189 Implementation Of a Methotrexate Self-Injection Education Video – Reducing Teaching Time While Maintaining Patient Confidence and Knowledge. Steven J. Katz and Sylvia Leung. University of Alberta, Edmonton, AB. Background/Purpose: To compare the effect of standard nurse led methotrexate self-injection patient education to a methotrexate self-injection web based education video added to standard teaching on 1) Patient satisfaction with teaching 2) Patient self-confidence in performing the injection 3) Patient knowledge around injection technique & precautions & 4) nursing time required Methods: Consecutive rheumatology patients at the University of Alberta Medicine Clinic seen by the Clinic Nurse for methotrexate self-injection education were asked to participate in the study. Each patient’s pre-education confidence for self-injection (Visual Analog Scale 1–10), age, gender, & education level was ascertained. Patients were randomized 1:1 to either standard teaching or the intervention: a 12 minute video reviewing the steps before, during and after methotrexate self-injection, followed by further in-person nurse education with a focus on reviewing patient concerns and questions arising from the video, and practicing self-injection similar to the non-intervention group. At the end of the teaching session, patients recorded their post-education confidence for self-injection, their satisfaction with the teaching process (VAS 1–10), and answers to 4 specific knowledge based

Associated factors- Three factors were associated with increased adherence. 1. Patients belief about the necessity of medications and the ability to control pain, disease activity or negative mood related to arthritis. 2. Use of Infliximab compared to either Etanercept, Methotrexate or Anakinra. 3. Use of DMARDs prior to anti TNF therapy. Two factors were associated with decreased adherence, use of Sulfasalazine compared to Methotrexate and use

S76

Tenderness

Ref. 1. Pasma A et al. Semin Arthritis Rheum, 2013.

Swelling and/or Tenderness

Disclosure: A. Scheiman-Elazary, None; C. Shourt, None; M. C. Hay, None; D. Furst, AbbVie, 2, Actelion Pharmaceuticals US, 2, Amgen, 2, BMS, 2, Gilead, 2, GSK, 2, NIH, 2, Novartis Pharmaceutical Corporation, 2, Pfizer Inc, 2, Roche/Genentech, 2, UCB, 2, Janssen Pharmaceutica Product, L.P., 5.

0.114* 0.075* 0.042* 0.100* 0.014 0.102* ⫺0.045* 0.130* 0.096* 0.015 0.105* 0.008 0.095* ⫺0.031

0.075* 0.060* 0.042* 0.112* 0.002 0.103* ⫺0.029 0.082* 0.069* 0.041* 0.088* ⫺0.016 0.097* ⫺0.034*

0.090* 0.065* 0.049* 0.109* 0.007 0.106* ⫺0.034* 0.103* 0.072* 0.043* 0.086* ⫺0.015 0.101* ⫺0.040*

*P⬍0.05. Standardized coefficients upon adjusting for age, gender, and number of affected joints

191 Does Specific Joint Involvement In Rheumatoid Arthritis Patients Predict Patient Reported Outcomes? Implications For Clinical Practice. Regan Arendse1, William G. Bensen2, Andrew Chow3, Jude F. Rodrigues4, Sanjay Dixit5, Dalton E. Sholter6, Philip Baer7, Milton F. Baker8, Denis Choquette9, Isabelle Fortin10, Algis Jovaisas11, Emmanouil Rampakakis12, John S. Sampalis12, Francois Nantel13, Susan M. Otawa13, May Shawi13 and Allen J. Lehman13. 1University of Saskatchewan, Saskatoon, SK, 2St. Joseph’s Hospital and McMaster University, Hamilton, ON, 3Credit Valley Rheumatology, Mississauga, ON, 4Rheumatology, Windsor, ON, 5McMaster University, Burlington, ON, 6Rheumatology Associates, Edmonton, AB, 7 Rheumatology, Scarborough, ON, 8University of Victoria, Victoria, BC, 9 Institut de Rhumatologie de Montreál, Montreal, QC, 10Centre de Rhumatologie de l’Est du Que´bec, Rimouski, QC, 11University of Ottawa, Ottawa, ON, 12JSS Medical Research, St-Laurent, QC, 13Janssen Canada Inc, Toronto, ON.

Conclusion: Significant variability in PROs exists based on the presence of swelling and/or tenderness in specific joint groups. Swelling/ tenderness of shoulders, wrists and knees were the main drivers of HAQ-DI, PtGA and pain. The results have important implications for the achievement of disease remission and would suggest that the joint type in addition to the number of affected joints has unique impact on PROs. Disclosure: R. Arendse, None; W. G. Bensen, None; A. Chow, None; J. F. Rodrigues, None; S. Dixit, None; D. E. Sholter, None; P. Baer, Janssen Pharmaceutica Product, L.P., 5; M. F. Baker, None; D. Choquette, None; I. Fortin, None; A. Jovaisas, Janssen Pharmaceutica Product, L.P, 5; E. Rampakakis, None; J. S. Sampalis, None; F. Nantel, None; S. M. Otawa, Janssen Canada, 3; M. Shawi, Janssen Canada, 3; A. J. Lehman, Janssen Canada, 3.

Background/Purpose: Assessment of functional (dis)ability in rheumatoid arthritis (RA) is subject to patient judgment when appraising their ability to do daily activities. The aim of this analysis was to describe the impact of specific joint involvement on patient reported outcomes (PROs) - functional activity, pain and patient global assessment of disease activity (PtGA) - and to identify joints most resistant to treatment over time. Methods: Biologic Treatment Registry Across Canada (BioTRAC) is an ongoing, prospective, registry of patients initiating treatment for RA, AS, or PsA with infliximab or golimumab as first biologics or after having been treated with a biologic for less than six months. In this analysis, RA patients treated with infliximab between 2002 and 2012 or with golimumab between 2010 and 2012 were included. Based on joint involvement seven groups were created: shoulder(s), elbow(s), metacarpophalangeal (MCP), wrist(s), proximal interphalangeal (PIP), knee(s), and thumb(s). The impact of specific joints on patient outcomes was assessed with the independent-samples t-test. Linear regression was used in order to produce adjusted estimates. Results: A total of 935 RA patients with 4854 assessments were included. Swelling, tenderness, and swelling and/or tenderness in all joint groups were associated with significantly (P⬍0.001) higher HAQ-DI, PtGA, and pain. Upon adjusting for age, gender and the total number of swollen (SJC28) and tender (TJC28) joints, swelling in all joint groups but the thumb(s) had a significant impact on PtGA, pain and HAQ. Similarly, tenderness in all joints but PIP had a significant impact on these parameters. Overall, swelling and/or tenderness at specific joints, shoulders, wrists and knees, had the greatest impact on HAQ-DI, PtGA, and pain (P⬍0.001). Swollen and/or tender PIP(s) did not have a significant effect on any PRO. At baseline, the MCP joint(s) and the wrist(s) were the most commonly swollen (86.4% and 67.9% of patients, respectively) or tender (82.9% and 73.1%, respectively) joints. Upon 12 months of treatment, the MCP joints were the joints most resistant to treatment, still remaining affected.

192 Can Routine Assessment Of Patient Index Data (RAPID)3 Be Used As a Patient Reported Assessment Tool For Patients With Ankylosing Spondylitis? Muhammet Cinar1, Sedat Yilmaz1, Fatma Ilknur Cinar2, Suleyman Serdar Koca3, Hakan Erdem1, Salih Pay1, Yusuf Yazici4, Ayhan Dinc5 and Ismail Simsek1. 1Gulhane School of Medicine, Ankara, Turkey, 2Gulhane Military Medical Academy, School of Nursing, Ankara, Turkey, 3Department of Rheumatology, Faculty of Medicine, Firat University, Elazig, Turkey, 4NYU Hospital for Joint Diseases and New York University School of Medicine, New York, NY, 5Patio Clinic, Ankara, Turkey. Background/Purpose: The assessment of disease activity in rheumatology is essential in both clinical trials and practice. The patient reported questionnaires have shown to reflect disease activity appropriately and is especially feasible in busy clinical settings. On the other hand, distribution of multiple questionnaires to patients with different diseases in a reception area might be difficult. A single questionnaire regarding to disease activity for all rheumatic diseases may present advantages to introduce quantitative measurement into routine care. Recently, a questionnaire that has been widely used and validated for rheumatoid arthritis, routine assessment of patient index data 3 (RAPID3), consists of patient function, pain, and patient global assessment, and has been suggested as suitable index for all rheumatic diseases. The aim of this study was to evaluate correlation of RAPID3 with Bath Ankylosing Spondylitis Disease Activity Index (BASDAI), AS Disease Activity Score (ASDAS), Bath AS Functional Index (BASFI), patient global assessment (PGA), and physician’s global assessment (PhGA) in AS patients. Methods: This single-centre, cross-sectional study was performed in a rheumatology department at a tertiary-care hospital. We included 341 consecutive AS patients who met the modified New York classification criteria. All patients completed BASDAI, BASFI, and MDHAQ at each visit, and their physicians completed physician global assessment. ASDAS scores were calculated using defined formulas. Spearman’s rho correlation test was used to determine the linear association between RAPID3 and other disease activity measures. Results: The median age of AS patients was 34.0 (21.0–69.0) years and the median disease duration 10.0 (2.0–35.0) years; 273 patients (80.1%) were male. Median scores for RAPID3, BASDAI, BASFI, ASDAS-CRP, ASDAS-ESR, PGA and PhGA were 13.0 (0.0–27.3), 4.7 (0.0–9.7), 3.0 (0.0–9.4), 3.0 (0.4–5.8), 2.5 (0.5–6.3), 5.0 (0.0–10.0) and

Table 1. Impact of Specific Joint Involvement on PROs Joint Group Swelling

Shoulder(s) Elbow(s) MCP Wrist(s) PIP Knee(s) Thumb(s) Shoulder(s) Elbow(s) MCP Wrist(s) PIP Knee(s) Thumb(s)

Shoulder(s) Elbow(s) MCP Wrist(s) PIP Knee(s) Thumb(s)

HAQ-DI

PtGA

Pain

0.055* 0.090* 0.085* 0.106* 0.048* 0.085* 0.032

0.049* 0.059* 0.103* 0.113* 0.032 0.098* 0.015

0.061* 0.053* 0.100* 0.108* 0.044* 0.102* 0.014

S77

Sunday, October 27

of Anakinra compared to Etanercept. Limitations- there were few articles using any one method and quality was variable, resulting in increased variability and decreased ability to separate the methods. Conclusion: There was a numerical tendency for decreased rate of adherence when measured by prescription claims versus the other methods. Patient beliefs and specific DMARDs are associated with adherence.

Sunday, October 27

3.5 (0.0–9.5), respectively. RAPID3 was strongly correlated with PGA, BASDAI, and ASDAS-ESR (r⫽0.843, r⫽0.842, r⫽0.815; p⬍0.001, respectively) (Table 1). Table 1. Correlation analysis between all disease activity measures. RAPID3 BASDAI BASFI ASDAS-CRP ASDAS-ESR PGA

PhGA

PGA

ASDAS-ESR

ASDAS-CRP

BASFI

BASDAI

0.637 0.581 0.545 0.672 0.718 0.606

0.843 0.796 0.737 0.824 0.821

0.815 0.828 0.717 0.888

0.790 0.829 0.673

0.793 0.764

0.842

Conclusion: RAPID3 was found to have good levels of positive correlation with BASDAI, BASFI, ASDAS-CRP, ASDAS-ESR, PGA and PhGA and seems to be as informative as those established disease activity indices. One might conclude that this index can be replaced for them in the follow-up patients with AS and it may offer an additional advantage of being easier to implement in a routine care setting. Figure. Proportion of Rheumatologists’ RA patients consistently on no treatment

Disclosure: M. Cinar, None; S. Yilmaz, None; F. I. Cinar, None; S. S. Koca, None; H. Erdem, None; S. Pay, None; Y. Yazici, None; A. Dinc, None; I. Simsek, None.

Conclusion: A modest proportion of RA patients are treated with no biologics or DMARDs over a sustained period of time. Even after controlling for RA-related and other patient characteristics, meaningful variability in the proportion of patients on no RA treatment was observed between physicians, suggesting the possibility of inappropriate variation in quality of care. Further efforts to explain and/or ameliorate quality gaps in RA appear warranted and have implications on reporting systems that judge rheumatologists’ quality of care. Except in the circumstance where patients are in clinical remission, codifying other evidence-based reasons why offering no RA treatment is appropriate is likely to be useful.

193 Physician Variability In Rheumatoid Patients Not Receiving Biologics Or Non-Biologic Dmards: Implications For Quality Reporting. Dimitrios A. Pappas1, Lang Chen2, Leslie R. Harrold3, George W. Reed4, Joel M. Kremer5, Jeffrey D. Greenberg6 and Jeffrey R. Curtis7. 1Columbia University, College of Physicians and Surgeons, New York, NY, 2 University of Alabama at Birmingham, Birmingham, AL, 3University of Massachusetts Medical School, Worcester, MA, 4CORRONA, Inc., Worcester, MA, 5Center for Rheumatology, Albany Medical College, Albany, NY, 6NYU Hospital for Joint Diseases, New York, NY, 7University of Alabama at Birmingham, Division of Clinical Immunology and Rheumatology, Birmingham, AL.

Disclosure: D. A. Pappas, Novartis, 9; L. Chen, None; L. R. Harrold, Corrona, 5, Takeda, 2; G. W. Reed, Corrona, Inc, 3; J. M. Kremer, Pfizer, Lilly, 2, Pfizer, Lilly, Vertex, 5; J. D. Greenberg, None; J. R. Curtis, Roche/Genentech, UCB, Janssen, CORRONA, Amgen, Pfizer, BMS, Crescendo, Abb Vie, 2, Roche/Genentech, UCB, Janssen, CORRONA, Amgen, Pfizer, BMS, Crescendo, Abb Vie, 5.

194

Background/Purpose: Current quality of care guidelines recommend that all patients with Rheumatoid Arthritis (RA) be treated with biologic and/or non-biologic (nb) DMARDs. However, some RA patients are not treated with these agents for a variety of reasons, which may impact perceptions of the quality of care their physicians are providing. To estimate physician-level variability on the proportion of RA patients who receive no biologic or nbDMARDs over a sustained period of time in a large U.S. Registry Methods: Data from the Consortium of Rheumatology Researchers of North America (CORRONA) was used to identify rheumatologistdiagnosed RA patients. Patients were characterized as being on RA treatment (biologics and/or nbDMARDs, irrespective of glucocorticoids/ NSAID/narcotics) based upon their most 3 consecutive visits in the registry. They were categorized as consistently on no treatment (0 of 3 visits), consistently on treatment (3 of 3 visits), and inconsistently on treatment (1–2 of 3 visits). Alternating logistic regression was used to evaluate physician-level clustering, controlling for patient-level factors including patient age, sex, RA disease duration, disease activity measured by CDAI, disability (by mHAQ), and comorbidities. Results: A total of 361 CORRONA physicians and 22,889 of their RA patients contributed to the analysis (median [IQR] number of RA patients per physician 19 [4, 78]). Using data from the 3 most recent CORRONA visits, 3.5 % of patients were consistently on no treatment, 84.7% consistently on treatment, and 11.8% inconsistently on treatment. The proportion of physicians’ patients on no RA treatment is shown (Figure) and ranged from 0 – 20%. For patients consistently on no RA treatment (n ⫽ 805), 31.4% were in remission, 37.1% were in low disease activity, and 31.4% were in moderate or high disease activity at the most recent CORRONA visit. After controlling for patient factors, physician clustering was independently associated with RA patients being on no RA treatment (adjusted OR ⫽ 1.81, 95% CI 1.17 – 2.80). Ongoing work is evaluating other potential reasons why patients may be untreated.

Team Managed Care From a Patient’s Perspective: A Study Of Biological Patients At a Canadian Centre. Dawn Heap1, Margaret Saldanha1, Melissa Deamude1, Cynthia Mech1, Helena Ross1, Kathy Kislinsky1, Lauri Vanstone2, Arthur N. Lau3 and William G. Bensen4. 1Rheumatology Health Team, St. Joseph’s Hospital Hamilton, Hamilton, ON, 2Rheumatology Health Team, Hamilton, ON, 3McMaster University, Hamilton, ON, 4St. Joseph’s Hospital and McMaster University, Hamilton, ON. Background/Purpose: In Canada, there is a widening care gap in Inflammatory Arthritis due to the increased aging population, higher disease prevalence, dwindling numbers of practicing rheumatologists and decreased access to care. As a result, there is delay in diagnosis, delay in initiating treatment, and delay in starting biological therapy. In 2008, our practice developed a team based program where trained Registered Nurses were assigned as the primary care nurse to patients receiving biological therapy in order expedite and more efficiently manage patient care. Currently more than 1300 biological patients are followed in this clinic. In this unique to Canada practice, patients are first seen and assessed by a registered nurse who do a complete history, review medications, assess joints, and provide health teaching. The nurse then discusses the patient and their needs with the Rheumatologist, and together they devise a plan which the nurse then implements and follows through. Methods: 175 consecutive patients with a confirmed diagnosis of inflammatory arthritis (Rheumatoid Arthritis, Ankylosing Spondylitis or Psoriatic Arthritis) receiving biological therapy between May and June of 2013 were sampled and given a confidential survey with 15 questions to determine their level of satisfaction with a multidisciplinary, team approach to care. They were asked questions regarding their general demographics, current treatment, and their confidence in the nurse that primarily follows them, whether or not the team addresses their questions or concerns, and treatment goals. 22 questionnaires had more than 2 questions left blank and therefore withdrawn from analysis.

S78

Results: More than 150 patients surveyed were included in this analysis studying patient satisfaction using a team approach model of patient care. The majority of patients surveyed were middle aged (40–75) and predominantly female. 85 percent of patients strongly agreed that they felt comfortable discussing their health and issues centring on their arthritis treatment with the nurse and confirmed that these goals were discussed at each visit. 97 percent of patients were confident in the nurse’s assessment skills and ability to manage their arthritis care and 99 percent of patients valued this team based model of care in comparison to seeing the physician alone. Conclusion: Our results from this study demonstrate a high degree of confidence in team- based rheumatology care. Currently this practice manages the biological volume of about 10 clinical rheumatologists who practice independently. Nurses are able to establish a rapport with patients in a unique and different way than a physician, and are not only able to address the patient’s physical needs but their psychosocial issues as well. Patients have the confidence that these highly trained specialty nurses are providing them with optimal care, allowing them to work toward achieving targeted treatment goals and better arthritis disease control. This study further affirms that team managed care is the only way we will be able to meet treat to target guidelines in our exponentially growing disease prevalence.

196

Background/Purpose: The RCR provides ACR members with an infrastructure for quality reporting related to rheumatoid arthritis, gout, osteoarthritis, osteoporosis, and drug safety. Currently in its fourth year of operation, the RCR contains data on over 32,000 patients. Here we report the uptake of the RCR by U.S. rheumatologists and performance on quality measures regarding functional status, disease modifying drug (DMARD) use, tuberculosis screening, prognosis, and disease activity assessment for RA patients in 2011 and 2012. Methods: Data derive from retrospective medical records abstractions performed by providers and/or designated practice staff for a sample of patients seen by rheumatologists. Reporters submit data on quality measures via a secure, web-based registry system. Patients included in the denominator of all quality measures are ⬎18 years of age with a diagnosis of RA who are receiving treatment by the reporting rheumatology provider. Additional details of each measure are listed in Table 1. We report the mean performance on each quality measure, defined as percentage of eligible patients receiving recommended care. Results: Table 1 summarizes performance on RA measures reported through the RCR. From January 1, 2012 to December 31, 2012, 301 rheumatology providers in 138 practices submitted data on 9154 encounters with RA patients. Comparative data are provided for the previous reporting period. These data summarize 8096 encounters with RA patients submitted by 257 rheumatology providers in 143 practices from January 1, 2011– December 31, 2011. Reporting providers practice in sites ranging from solo offices to large academic medical centers.

Disclosure: D. Heap, None; M. Saldanha, None; M. Deamude, None; C. Mech, None; H. Ross, None; K. Kislinsky, None; L. Vanstone, None; A. N. Lau, None; W. G. Bensen, None.

195 Major Cost Savings Associated With Reduced Biologic Dosing Frequency In Inflammatory Arthritis. Claire-Louise Murphy1, Miriam O’Sullivan2, Sohail Awan2, Shawn Chavrimootoo2, Clara Bannon2, Linzi Martin2, Trevor Duffy2, Eithne Murphy2 and Maurice Barry2. 1Connolly hospital, Dublin 15, Dublin, Ireland, 2Connolly hospital, Dublin, Ireland.

Table 1. Performance on RA Measures Assessed through the RCR CY2011

CY2012

Patients (N)

QM Performance Rate

Patients (N)

QM Performance Rate

8075

43.3%

6485

54.4%

8077

70.5%

6485

86.6%

7808

97.9%

6485

86.6%

1650

73.6%

1048

92.9%

7771

49.5%

6441

73.4%

Background/Purpose: Biologic agents are highly effective in Inflammatory Arthritis (IA) but are extremely expensive. A sustained reduction in biologic dosing frequency would lead to major cost savings. The purpose of this study was to explore whether patients with Inflammatory Arthritis would remain in remission following a reduction in biologic dosing frequency and therefore lead to cost savings. Methods: This prospective non-blinded non-randomised study commenced in 2010. Patients with Inflammatory Arthritis (Rheumatoid Arthritis, Psoriatic Arthritis or Ankylosing Spondylitis) being treated with a biologic agent were screened for disease activity. A cohort of those in remission according to standardized disease activity indices (DAS28⬍2.6, BASDAI⬍4) was offered a reduction in the dosing frequency of two commonly used biologic therapies (etanercept 50mg once per fortnight instead of weekly, adalimumab 40mg once per month instead of fortnightly). Patients were assessed for disease activity at 3, 6 and 12 months following reduction in dosing frequency. Cost saving was calculated as the difference between the cost of the actual amount of biologic agent used over one year compared with the cost if the dosing interval had not changed. Results: Seventy nine patients with inflammatory arthritis in remission were recruited. Fifty seven per cent had rheumatoid arthritis (n⫽45), 13% psoriatic arthritis (n⫽10) and 30% ankylosing spondylitis (n⫽24). Fifty seven per cent (n⫽45) were taking etanercept and 43% (n⫽34) adalimumab. The percentage of patients in remission at 12 months was 61% (n⫽48). Using paired sample t-tests in SPSSv20, no significant difference in measures of disease activity or functional status (DAS28, HAQ or BASDAI scores) was identified from baseline to 12 months in those who remained in remission. This resulted in an actual saving to the state of approximately 300,000 euro over the year. Conclusion: This small study suggests reduction in biologic dosing frequency is feasible in inflammatory arthritis. The study resulted in considerable cost savings at 1 year. The potential for major cost savings in biologic usage should be pursued further.

Discordance Of Patient-Physician Assessments Of General Health In a US Hispanic Population With Rheumatoid Arthritis. Leyda DıázCorrea1, Mariely Nieves-Plaza2, Yesenia C. Santiago-Casas1, Tania Gonza´lez-Rivera3, Grissel Rıós1 and Luis M. Vila´1. 1University of Puerto Rico Medical Sciences Campus, San Juan, PR, 2University of Pittsburgh, Pittsburgh, PA, 3University of Michigan School of Medicine, Ann Arbor, MI.

Disclosure: C. L. Murphy, None; M. O’Sullivan, None; S. Awan, None; S. Chavrimootoo, None; C. Bannon, None; L. Martin, None; T. Duffy, None; E. Murphy, None; M. Barry, None.

Background/Purpose: Instruments to measure rheumatoid arthritis (RA) activity such as the 28-Joint Disease Activity Score (DAS28) are used to assess response to treatment. One component of the DAS28 is the patient

Disease activity assessed at least once within 12 months, using a standardized descriptive or numeric scale or composite index, and classified as low, moderate or high Functional status assessment performed at least once within 12 months, and documented using a standardized descriptive or numeric scale, standardized questionnaire, or notation of assessment of the impact of RA on patient activities of daily living Patient prescribed, dispensed, or administered at least one ambulatory prescription for a DMARD within 12 months Documentation of TB screening performed and results interpreted within 6 months prior to receiving first course DMARD Assessment and classification of disease prognosis at least once within 12 months

Conclusion: Performance rates increased on four out of five measures from CY2011 to CY2012. Based on preliminary results, the ACR plans to complete in-depth analyses of both measure and population performance. Disclosure: J. Yazdany, None; S. Kazi, None; R. Myslinski, None; M. Francisco, None.

197

S79

Sunday, October 27

Uptake Of The American College Of Rheumatology’s Rheumatology Clinical Registry: Quality Measure Summary Data. Jinoos Yazdany1, Salahuddin Kazi2, Rachel Myslinski3 and Melissa Francisco3. 1University of California, San Francisco, San Francisco, CA, 2UT Southwestern Medical Center, Dallas, TX, 3American College of Rheumatology, Atlanta, GA.

Sunday, October 27

global assessment of general health. This is a subjective measurement that has been shown to be influenced by multiple factors such as depression. Moreover, some studies have reported discordance between patient and physician ratings of general health; however, the factors associated with these discrepancies have not been well established. The aims of this study were to determine the level of discordance of patient-physician assessments of general health in a group of US Hispanics with RA, and to evaluate the factors associated with such discordance. Methods: A cross-sectional study was conducted in 213 US Hispanics with RA (per 1987 ACR criteria) participating in an observational study. Demographic parameters, clinical manifestations, disease activity (per DAS28), comorbid conditions, functional status (per Health Assessment Questionnaire [HAQ]), pharmacologic profile, and patient and physician global assessments (measured on a visual analogue scale [VAS]) were determined. Positive discordance of general health was defined as a patient rating minus physician rating ⬎25 mm on a 100-mm VAS. Differences between study groups were evaluated using multivariable logistic regression analysis. All variables with a pⱕ0.10 in the bivariable analysis plus age, gender and disease duration were entered into the regression model. Results: The mean (standard deviation [SD]) age of RA patients was 56.6 (13.6) years; 88.0% were females. The mean (SD) disease duration was 11.2 (10.0) years. Positive patient-physician discordance was found in 78 (36.6%) patients. Only 2 patients exhibited negative discordance. In the multivariable analysis, patients with positive discordance were more likely to have a higher global assessment of pain (OR⫽1.05, 95% CI 1.03–1.07, p⬍0.001) and HAQ scores (OR⫽3.19, 95% CI 1.70–6.00, p⬍0.001), and were more likely to receive therapy with biologic agents (OR⫽3.06, 95% CI 1.40–6.71, p⫽0.005) than those without discordance. Conversely, patients with positive discordance had lower physician’s assessment of function (OR⫽0.96, 95% CI 0.93–0.98, p⫽0.002). No association was found for comorbidities including depression and fibromyalgia syndrome. Conclusion: In this group of patients with RA, a large number differed from their physicians in the assessment of general health. Positive discordance was associated with patients’ self-report of pain and disability, and the use of biologics. The awareness of these factors may help to better assess disease activity and treatment response in this population.

reporting of the disease by the patient and its usage to rule out the disease. In 48 cases, a patient reported to their primary care physician that they suffered from rheumatoid arthritis when in fact they had another diagnosis. These patients with inaccurate ICD-9 codes will be inappropriately categorized as receiving poor levels of care by not meeting RA quality measures. In patients with true disease, a lower percentage is noted to not meet the quality measure of DMARD treatment. Interestingly, it was observed that injectible medications were not documented in the primary care medication list. This could be due to the fact that patients often bring physical bottles for medication reconciliation and not the injectables. Taking this into consideration, the percentage of RA patients not meeting quality measures is likely considerably lower. In conclusion, since quality measures do not address the issue of inaccurate diagnosis, its emphasis on management and its potentially associated penalties should be reevaluated. Further studies are needed to determine if patients with inaccurate coding are subjected to unnecessary treatment or testing. Disclosure: S. Ogando, None; K. M. Weiss, None; H. D. Fischer, None.

199 Overcoming The Barriers To Adequate Hydroxychloroquine Retinal Toxicity Screening. Cristina Arriens1 and Elizabeth Blair Solow2. 1University of Texas Southwestern Medical Center at Dallas, Dallas, TX, 2UT Southwestern Medical Center, Dallas, TX. Background/Purpose: Hydroxychloroquine (HCQ) is considered a minimal risk drug in the treatment arsenal for rheumatoid arthritis (RA), systemic lupus erythematosus (SLE), and other rheumatic diseases; however it has a well-documented risk of retinal toxicity in approximately 1% of patients following 5 years of use. Adequate ophthalmologic examination monitoring can limit the impact of toxicity on vision by early detection. The American College of Rheumatology’s position statement regarding screening for HCQ retinopathy recommends baseline ophthalmologic examination within the first year of treatment and annual examinations after 5 years of therapy as the minimum for healthy patients. It was recognized that multiple patients in a busy county hospital system receiving treatment with HCQ lacked retinopathy screening exams. A quality improvement project to ascertain the factors that lead to failure within this system was devised with the goal of improving compliance with the accepted guidelines. Methods: In this busy county hospital outpatient rheumatology clinic we performed a chart review to evaluate provider, patient, and system factors to determine where the level of breakdown in achieving baseline eye examinations occurred. A referral to ophthalmology clinic was required for the provider to adequately fulfill their role. A scheduled appointment with ophthalmology was required for the system to meet its responsibility. The patients’ role was assessed by their appointment attendance. A physicianinitiated, patient driven intervention was instituted in which rheumatology physicians provided the ophthalmology clinic phone number and asked the patient to call to schedule their appointment. Compliance with screening guidelines was re-assessed 1 year later. Results: The average age of the 60 patients was 48 years and included 30 RA, 20 SLE, and 10 other connective tissue diseases. Prior to the intervention, 32 patients (53%) had completed a baseline retinal toxicity exam, 2 (3%) were not referred, 6 (10%) had no-showed an appointment, and 20 (33%) had a referral without an appointment scheduled. Failure occurred at the system level. Following the intervention, of the 53 patients still taking HCQ, now 36 (68%) had completed their screening visit, 0 (0%) were not referred, 1 (0.2%) had no-showed, and 16 (30%) were referred with no appointment. Pre- intervention 25 of the remaining 53 were compliant with screening examination and eleven more patients became compliant postintervention, a statistically significant improvement determined by McNemar’s test (p⫽0.0026). Conclusion: In this quality improvement project the major barrier to patients receiving baseline eye exams for HCQ toxicity was found to be a system issue. The physician-initiated, patient driven intervention was successful in improving compliance in the 1 year time frame of study. In addition to this intervention, future efforts will be aimed at inter-departmental communication and education between ophthalmology and rheumatology regarding the necessity of timely retinal evaluations.

Disclosure: L. Dıáz-Correa, None; M. Nieves-Plaza, None; Y. C. Santiago-Casas, None; T. Gonza´lez-Rivera, None; G. Rıós, None; L. M. Vila´, None.

198 Accuracy Of International Classification Of Diseases (Ninth Revision) Coding For Rheumatoid Arthritis In The Primary Care Setting. Sheena Ogando, Karolina M. Weiss and Harry D. Fischer. Albert Einstein College of Medicine at Beth Israel Medical Center, New York, NY. Background/Purpose: Current quality measures are diagnosis driven and focus on management. For rheumatoid arthritis (RA), the Physician Quality Reporting System requires that a disease modifying anti-rheumatic drug (DMARD) therapy be prescribed. Quality measures do not account for incorrect diagnoses. Diagnostic error can stem from a physician or from a patient report of a medical history. Prior studies have shown a discrepancy between patient report and physician diagnosis. One study reported diabetes - a well defined disease - had no variance, while arthritis - a less clearly defined disease - had a 15% variance, with patients over reporting the disease. Our objective is to determine the accuracy of RA diagnosis in our institution’s primary care offices and to evaluate if quality measures are met Methods: Adult patients with the International Classification of Diseases, Ninth Revision (ICD-9) code for Rheumatoid Arthritis (714.0) were searched from April 2011 to April 2013 using our institution’s primary care electronic medical record. The medication list was reviewed for the presence of a DMARD. The paper chart of patients seen in our institution’s rheumatology clinic was reviewed for a more accurate diagnostic code. Results: 246 patients were identified by ICD-9 code using the primary care electronic medical record; 172 were used for evaluation. The average age of the patient population was 60.4 with a standard deviation of 12.7. 88.4% (152) were female. The rate of incorrect ICD-9 coding by primary care physicians was 24–38% [CI of 95%]. The top 5 rheumatologist established diagnoses in these patients were osteoarthritis (20), other inflammatory arthritis (10), undifferentiated connective tissue disease (8), osteoporosis/ osteopenia (5), and systemic lupus erythematosus (5). In patients with rheumatologist confirmed RA, 12–26% [CI of 95%] did not have a DMARD in the primary care medication list. Conclusion: The high incidence of primary care physicians erroneously utilizing the 714.0 ICD-9 code in our institution can be attributed to inaccurate

Disclosure: C. Arriens, None; E. B. Solow, None.

S80

Hydroxychloroquine Use Among Systemic Lupus Erythematosus Patients In An Academic Rheumatology Practice. James G. Miceli1 and Harry D. Fischer2. 1Beth Israel Medical Center, New York, NY, 2Albert Einstein College of Medicine at Beth Israel Medical Center, New York, NY. Background/Purpose: Hydroxychloroquine (HCQ) is a mainstay in the current treatment of systemic lupus erythematosus (SLE). HCQ has been shown to reduce the incidence of SLE flares, reduce the risk of irreversible end-organ damage, and improve survival in SLE patients. Yet, few studies have been devoted to examining its prevalence of use. Prior papers have either examined HCQ use as a secondary endpoint (56–64%) or as a primary endpoint in a community practice (55%). Our study aims to identify the prevalence of HCQ use in an urban academic rheumatology practice and to identify the reasons for nonuse Methods: A search of medical records with a diagnostic code for systemic lupus erythematosus (710.0) was performed for patients seen within a three-month period (September 1–December 1, 2011) at Beth Israel Medical Center’s rheumatology clinic. Charts for 211 patients were reviewed for active use of HCQ; if the patient was not on HCQ, the chart was further reviewed and an explanation for nonuse was documented if present. Results: A total of 161 out of 211 SLE subjects (76.3%) were found to be actively using hydroxychloroquine. Among nonusers, 21 (42%) did not have a definitive diagnosis of SLE; 9 (18%) were either non-compliant or refused treatment; 6 (12%) had an allergy or intolerance to HCQ; 6 (12%) were not taking HCQ for unclear reasons but were asymptomatic; 4 (8%) reported a history of HCQ-induced oculotoxicity; and 2 (4%) developed thrombocytopenia presumed secondary to HCQ (Figure 1).

Disclosure: G. Ciancio, None; I. Farina, None; F. Pignatti, None; M. Govoni, None.

202 Analysis Of Non-Steroidal Anti-Inflammatory Drug Burden Among Rheumatoid Arthritis Patients Using The Dougados Algorithm. E Alemao1, L Xie2, R Wong1, G Lltalien1 and O Baser3. 1Bristol-Myers Squibb, Princeton, NJ, 2SATinMED Research, Ann Arbor, MI, 3STATinMED Research and University of Michigan, Ann Arbor, MI Background/Purpose: Non-steroidal anti-inflammatory drugs (NSAIDs) are used to alleviate RA pain symptoms, but they may be associated with adverse events. Previous studies examining clinical and economic outcomes associated with NSAID use in RA have not implemented a systematic algorithm for NSAID use, which would more precisely define the NSAID burden. This study evaluated the correlation between the NSAID intake score developed by Dougados et al.1 and other traditional measures of NSAID burden. In addition, clinical and economic outcomes in patients with high versus low NSAID scores were compared. Methods: A retrospective analysis was performed on RA incidence populations using National Medicare claims data (01/01/2006–12/31/2010) with a 12-month baseline and 24-month follow-up period. Patients had to have at least two RA diagnoses at least 60 days apart between January 1, 2007 and December 31, 2009. The date of the first RA diagnosis was designated as the index date. High, middle and low cohorts were created based on the distribution of the NSAID intake score.1 The correlation between NSAID score and traditional measures was determined by univariate analysis. Clinical and economic outcomes during follow-up were compared with a generalized linear model (GLM). Results: Approximately 19% (n⫽10,313) of 52,973 incident patients with RA used NSAIDs during the baseline period. Dougados NSAID score correlated with percentage of days with at least one NSAID intake (0.69 [p⬍0.05]) and number of NSAID tablets (0.59 [p⬍0.05]). At baseline, 36.34% of patients had a low NSAID score (ⱕ10% follow-up time on diclofenac 150 mg or equivalent), 58.33% had a medium NSAID score (11–90% time on diclofenac 150 mg or equivalent) and 5.32% had a high NSAID score (⬎90% time on diclofenac 150 mg or equivalent). Compared with the high NSAID cohort, the low NSAID cohort were older (mean age 75.9 vs 74.4 years, p⬍0.05), had higher Charlson Comorbidity Index scores (3.2 vs 2.8, p⫽0.05), and higher rates of chronic kidney disease (CKD) (11.39 vs 6.38%, p⬍0.05), chronic obstructive pulmonary disease (22.9 vs 18.94%, p⬍0.05), and respiratory disorders (57.54 vs 52.28%, p⬍0.05). Prior to

Conclusion: The use of HCQ in SLE patients at our institution exceeded rates of use reported elsewhere. Nonetheless, almost 1⁄4 of potential candidates for treatment were not prescribed HCQ. Almost half of the non-prescribed patients lacked a definitive diagnosis of SLE. The majority of the other non-takers had a documented explanation. Our results help to better define reasons for nonuse of HCQ in patients with SLE. It is hoped that recognition of these factors could help to further increase the use of HCQ in SLE patients. Disclosure: J. G. Miceli, None; H. D. Fischer, None.

201 Comparison To Organisational Models For EARLY Rheumatoid Arthritis Management: Routine Care Versus EARLY Arthritis Clinic. Giovanni Ciancio1, Ilaria Farina1, Federica Pignatti2 and Marcello Govoni1. 1 Rheumatology Unit-Azienda Ospedaliera-Universitaria Sant’Anna, Ferrara, Italy, 2Rheumatology Unit-Azienda Ospedaliera-Universitaria Sant’Anna, Ferrara, Italy, ferrara, Italy Background/Purpose: although the Early Arthritis Clinic (EAC) institution is considered the best approach for the management of rheumatoid arthritis (RA) at an early stage, to the best of our knowledge there is no formal demostration that this organisational model is better than a conventional model in terms of public health care and of improved clinical, therapeutic and radiographic outcomes. Our purpose was to evaluate whether significant differences in clinical and therapeutic outcomes exist between two groups of patients with Rheumatoid Arthritis followed in a EAC and in a routin care (RC) model, respectively.

S81

Sunday, October 27

Methods: Two groups of RA patients fulfilling 1987/ACR criteria were retrospectively analised. In the first group, RA diagnosis was made in our centre between 2002 and 2008 and patients were followed with a RC model. In the second one, RA diagnosis was made in our centre between 2009 and 2013 and patients were followed in our EAC. Patients with a follow-up of at least 2 years were included. For each patient of the two groups, lag time from symptom onset to diagnosis and from symptom onset to the beginning of DMARDs therapy were calculated. Disease activity were compared between the two groups using DAS28 and EULAR response criteria at baseline (T0) and after 6 (T6), 12 (T12) and 24 (T24) months. The ratios (%) of patients who started a biological therapy within 24 years after diagnosis were compared between the two groups Results: A total of 273 RA patients, divided into two groups, were evaluated: 209 (mean age 59 ⫾ 24.04 years,165 F, 44 M) followed in RC and 64 (mean age 59 ⫾ 13.44 years, 50 F,14 M) in EAC. Lag time from symptom onset to diagnosis resulted significally lower (p⬍0,0001) in patients assessed in EAC (5.73 ⫾ 8.09 mesi) compared to that of the other group (20.90 ⫾ 30.3 mesi) and treatment beginning occured before in EAC population (7.17⫾ 8.26 mesi) respect to that followed in RC (21.98⫾32.39 mesi) with a significant difference (p⬍0.0004). At baseline DAS28 evaluation was similar between groups (4.76⫾1.24 in RC vs 4.94 ⫾ 1.41 in EAC) and was significally reduced to 2.82 ⫾ 1.24 (RC) and 2.46 ⫾1.12 (EAC) after 24 months (p⬍0.036). A significant statistical difference emerged from the comparison to the mean VES value (19.9 ⫾15.9 RC vs 13.6 ⫾ 11.3 EAC, p⬍0.003) and SJ (swollwn joints) value (1 ⫾ 1.8 RC vs 0.43 ⫾ 1.2 EAC, p⬍0.02) at T-24. Within 24 months, biologic therapy was initiated in 29% of patients followed in RC and in 9.3% of patients followed in EAC population (p⬍0,0013). Conclusion: In comparison with the RC model, the EAC institution has allowed in our experience a significant reduction of the time for diagnosis and for the therapeutic intervention, with an improvement in clinical outcomes, less use of biological drugs and a significant long-term savings on pharmaceutical spending

200

Sunday, October 27

baseline adjustments, patients in the low NSAID cohort had fewer gastrointestinal perforation cases (0.53 vs 1.28%, p⬍0.05), and lower per patient per year (PPPY) total costs ($26,235 vs $30,078, p⬍0.05) and outpatient pharmacy costs ($4,347 vs $6,547, p⬍0.05) than patients in the high NSAID cohort. After adjusting for baseline differences in demographic and clinical characteristics with the GLM, patients in the high NSAID cohort had a higher risk for CKD (20.23 vs 17.31%, p⬍0.05), higher PPPY inpatient costs ($13,187 vs $10,652, p⬍0.05) and higher PPPY outpatient pharmacy costs ($4,605 vs $3,515, p⬍0.05) during the follow-up period. Conclusion: The Dougados NSAID algorithm, developed to evaluate NSAID burden in spondyloarthritis, is also applicable to the RA patient population. Medicare patients with RA with high NSAID scores had a higher probability of CKD and higher average in/outpatient costs. Therapies that reduce NSAID burden in elderly patients may protect those with RA from CKD and help reduce cost.

and rapid way to identify patients who are more likely to be non-adherent with DMARD therapy. Disclosure: W. Sun, None; D. C. T. Bautista, None; X. Xin, None; Y. T. Saw, None; W. B. Tan, None; K. Balasubramaniam, None; W. P. Lee, None; S. T. H. Oo, None; Y. B. Cheung, None; J. Thumboo, None.

204 Evaluation Of Hospitalizations and Costs In Patients With Rheumatoid Arthritis In United States Medicare Population. E Alemao1, L Wang2, G Lltalien1, O Baser3, H Yuce4 and M Hochberg5. 1Bristol-Myers Squibb, Princeton, NJ, 2STATinMED Research, Dallas, TX, 3STATinMED Research and University of Michigan, Ann Arbor, MI, 4New York City College of Technology (CUNY), Brooklyn, NJ, 5University of Maryland, Baltimore, MD

Reference: 1. Dougados M, et al. Ann Rheum Dis 2011;70:249–51.

Background/Purpose: RA-related hospitalization and surgery (e.g. total joint arthroplasty [TJA]) are important long-term outcomes in RA. While advances in treatment for RA have resulted in decreased admissions for active disease, the national rates of RA-related procedures, hospitalizations and their costs in elderly patients with RA have not been reported. We evaluated the rates of first and subsequent RA-related surgery, hospitalization and mortality in a US Medicare population, analyzed predictors of RA-related surgery and compared total costs between RA patients with and without surgery. Methods: Data from patients enrolled in the US Medicare Claims Database from January 1999 to December 2009, who had ⱖ2 RA diagnoses (ICD 714.0) ⱖ2 months apart during the identification period were analyzed. The date of the first RA diagnosis was designated as the index date. Those with a baseline period ⬍12 months (i.e. those diagnosed during 1999 and 2000) were considered prevalent cases. Kaplan–Meier analysis was used to estimate the cumulative incidence of orthopedic surgery following RA diagnosis. Cox proportional hazards (CPH) modeling identified factors associated with surgery (TJA, TJA-associated procedures, non-TJA) and overall mortality. Estimated healthcare costs (mean and standard deviation) for all RA patients with surgery were compared with those for RA patients without surgery. Results: The study population comprised 360,912 patients with RA enrolled in Medicare who met the study inclusion criteria. Cumulative 4- and 10-year TJA incidence rates were 7.5 and 13.2%, respectively; mortality rates were 13.2 and 27.9%. Of all RA patients with surgical experience, 86.5% had a TJA procedure. Patient characteristics are shown (Table). Based on CPH models, predictors of surgery varied by type of procedure; positive factors often included regional demographic and co-morbid osteoarthritis (OA) at baseline, and negative predictors often included follow-up therapy and minority race. For example, for TJA, patients with OA and patients living in the Midwest had a higher hazard (hazard ratio [HR]⫽2.11, p⬍0.01; HR⫽1.38, p⬍0.01, respectively). Patients receiving combination (MTX ⫹ biologic DMARD) RA therapies during follow-up had a lower hazard (HR⫽0.41, p⬍0.01). RA patients with surgery had almost double the average inpatient cost at $19,382 vs $10,282 for patients without surgery (p⬍0.01). RA patients with surgery also had significantly higher outpatient, outpatient emergency room, and office costs (p⬍0.01).

Disclosure: E. Alemao, Bristol-Myers Squibb, 1, Bristol-Myers Squibb, 3; L. Xie, Baistol-Myers Squibb, 5; R. Wong, Bristol-Myers Squibb, 1, Bristol-Myers Squibb, 3; G. Lltalien, Bristol-Myers Squibb, 1, Bristol-Myers Squibb, 3, SimplySmiles (www.simplysmiles.org), 6; O. Baser, Bristol-Myers Squibb, 5.

203 Belief In Self-Adjustability Of Medication Dosing Is Negatively Correlated With Medication Adherence In Patients With Rheumatoid Arthritis. Wenxin Sun1, Dianne Carrol Tan Bautista1, Xiaohui Xin2, Yu Ting Saw2, Wee Boon Tan2, Kanchanadevi Balasubramaniam2, Wan Pin Lee2, Steve Thein Htay Oo2, Yin Bun Cheung1 and Julian Thumboo2. 1Duke-NUS Graduate Medical School, Singapore, Singapore, 2Singapore General Hospital, Singapore, Singapore. Background/Purpose: Medication adherence in Rheumatoid Arthritis (RA) is reported to be generally low. This major problem needs to be addressed because it leads to reduced treatment benefits and greater healthcare costs. We hypothesized that specific medication beliefs would influence adherence to disease-modifying anti-rheumatic drugs (DMARDs) in RA patients, and assessed this in a cross-sectional study. Methods: Based on focus groups and a quantitative survey, we identified 5 belief factors in patients with RA: that DMARDs harm health; that DMARDs benefit health; that DMARD dosing/regimens can be changed by patients; that DMARD side effects can be effectively managed; that DMARD side effects cannot be effectively managed. In this IRB approved study, we administered a Medication Behavior Survey (n⫽21, developed from focus group data) assessing these 5 belief factors and the 8-item Morisky Medication Adherence Scale (MMAS-8, score range 0–8.0) to measure non-adherence in a consecutive sample of English-speaking RA patients during routine outpatient visits from November 2012 to April 2013 at our tertiary referral center. Multiple linear regression (MLR) was used to examine the relationship between the 5 belief factors and MMAS-8, as well as MMAS-8 intentional and unintentional non-adherence subsets, adjusting for demographic factors (gender, race, age and education). Results: Among 279 English-speaking RA patients (82% females; 56% Chinese, 15% Malays, 22% Indians; mean age (SD) 53.8 (12.7) years). MLR with MMAS-8 scores showed that a stronger belief that DMARD dosing/regimens can be changed by patients was significantly negatively correlated with MMAS-8 scores (beta⫽⫺0.53, 95% CI ⫺0.75 to ⫺0.37, p⬍0.001) in a model that also included the other 4 factors (none of which was significantly associated with MMAS-8 scores). Similarly, MLR with MMAS-8 subsets showed that a stronger belief that DMARD dosing/regimens can be changed by patients was negatively correlated with both intentional and unintentional non-adherence MMAS-8 subsets (beta⫽⫺0.25, 95% CI ⫺0.39 to ⫺0.11, p⬍0.001; beta⫽⫺0.28, 95% CI ⫺0.42 to ⫺0.14, p⬍0.001, respectively). Additionally, belief that DMARDs side effects cannot be effectively managed was negatively associated with intentional non-adherence MMAS-8 subset (beta ⫽ ⫺0.18, 95% CI ⫺0.32 to ⫺0.03, p⫽0.019). Conclusion: We found that stronger belief that DMARD dosing/ regimens can be changed by patients was an important factor associated with non-adherence in an urban Asian population. Other beliefs do not apparently provide additional value in accounting for non-adherence. If confirmed in prospective studies, this observation may provide a simple

Mean age, years White Mean Charlson’s Comorbidity Index scores OA High RA severity score

Patients with surgery

Patients without surgery

Standard deviation

74 88% 2.27

76 82% 2.60

22.45 17.76 17.04

63% 35%

46% 27%

33.85 17.91

Conclusion: Medicare patients with RA continue to experience high rates of RA-related surgery, hospitalization and overall mortality. However, biologic DMARD therapies appear to have a protective effect on all outcomes. Greater availability of efficacious RA therapies has the potential to reduce RA-related surgery, hospitalization and costs. Disclosure: E. Alemao, Bristol-Myers Squibb, 1, Bristol-Myers Squibb, 3; L. Wang, Bristol-Myers Squibb, 5; G. Lltalien, Bristol-Myers Squibb, 1, Bristol-Myers Squibb, 3, SimplySmiles (www.simplysmiles.org), 6; O. Baser, Bristol-Myers Squibb, 5; H. Yuce, None; M. Hochberg, Abbott Laboratories, Amgen Inc., BMS, Eli Lilly and Company, EMD Serono Inc., Genentech/Roche, Merck & Co., Inc., Novartis Pharma AG, Pfizer Inc, 5, Bioberica SA, IBSA, 8, NIH, 2.

S82

Q-Med, 5, NiCox, S.A., 5, Flexion, 5, Bioberica, 5; D. Prieto-Alhambra, Bioiberica, 2, Amgen, 2; A. Judge, Servier, 2, Anthera, 5, Roche Pharmaceuticals, 2.

205

206 American College Of Rheumatology’s Rheumatology Informatics System For Effectiveness Registry Pilot. Peter J. Embi1, William Stephens2 and Rachel Myslinski3. 1The Ohio State University, Columbus, OH, 2Ohio State University, Columbus, OH, 3American College of Rheumatology, Atlanta, GA Background/Purpose: There is a significant need in the rheumatology community to access and integrate data across diverse patient populations in order to aid quality improvement efforts, help rheumatologists meet quality reporting requirements, as well engage in clinical research. We report on the pilot results of the American College of Rheumatology’s registry expansion initiative that links disparate clinical data resources across multiple clinical sites and systems in support of rheumatology practice and research. This effort is aimed at providing a reliable, cost-effective means of connecting data from multiple EHR systems, using these data for quality improvement, quality reporting, and research querying. Methods: The design and execution of effective quality improvement projects and clinical studies requires access to high quality, longitudinal data. In most instances, such data are collected, formalized, stored and retrieved using project- or organization-specific disease registries or data warehouses. It is increasingly desirable to access data across multiple clinical sites for quality improvement and clinical research purposes, but disparate EHR systems remain difficult to connect for data interchange. Furthermore, in these types of settings, organizational and policy barriers often preclude the use of centralized repositories. To address this need, the ACR is piloting this system – called the Rheumatology Informatics System for Effectiveness (RISE) – to enhance registry efforts to benefit rheumatic disease research and quality reporting efforts. RISE is a federated system of interconnected clinical data repositories. The RISE architecture allows data to remain at the sites thereby allowing sites full control of their data. Data is accessed by a Federated Query Processor service that maintains no source practice data internally, but instead acts as a secure router and aggregator for information as it passes through the grid. The RISE system enables access to clinical data across multiple sites while maintaining high levels of data security. Results: The model employed by RISE uses an approach to enable the federated query of geographically distributed data sources. This platform is being implemented at 6 sites, with several more planned. The initial sites include a variety of academic and community centers, which demonstrate feasibility for implementation in a variety of practice settings. The following table shows aggregate data from two of the initial sites:

Background/Purpose: Disease modifying anti-rheumatic drugs (DMARDs) are standard initial treatments for rheumatoid arthritis (RA). Many RA treatment guidelines have been published including from the American College of Rheumatology (ACR), European League Against Rheumatism (EULAR) and British Society for Rheumatology (BSR). However, there is little evidence to demonstrate that guidelines alter prescribing practice. BSR guidelines for treatment of RA were published in 2006 and recommended initiation of DMARDs in the first 90 days after RA diagnosis. We aimed to test whether publication of these guidelines changed clinical practice. Methods: We used the Clinical Practice Research Datalink, consisting of primary care records & prescriptions, secondary care referral details, procedures and diagnoses, for ⬎10 million patients from a representative sample of primary care practices in the UK. DMARDs are usually initiated in the UK by a rheumatologist, with ongoing DMARD prescribing usually performed by primary care physicians with supervision by a rheumatologist. Incident diagnoses of RA in persons aged ⱖ18 years between 1995–2010 were identified. For each calendar year and quarter, the proportion of patients receiving any DMARD prescription within 90 days and 1 year of diagnosis was ascertained. An interrupted time-series regression model was used to assess whether trends in the proportion of patients receiving DMARDs within 90 days and 1 year changed following publication of the 2006 BSR Guidelines. Results: Between 1995 and 2010, 25,963 incident cases of RA were identified, 69% female, average age of diagnosis 59 years. There was a progressive increase in the proportion of patients prescribed a DMARD by a primary care physician within 90 days from 16.6% in 1995 to 23.5% in 2005. The proportion receiving DMARD/s within 90 days increased more rapidly following publication of the BSR guidelines (Figure 1) to 27.9% by end 2006 and 36.0% in 2010 (p⫽0.01). A similar pattern was observed for the proportion of patients receiving DMARD/s within 365 days of diagnosis (p⫽0.002).

Query

Aggregate Result

Number of patients with an RA diagnosis Number of patients with a RAPID score ⱖ 3 Number of patients taking methotrexate All patients in the registry

6,131 260* 6,724 44,390

Conclusion: The design, deployment and initial use of the ACR RISE network addresses the need for data access across disparate sites using otherwise non-interoperable information systems and creates the opportunity for a robust source of rheumatology clinical data that can be used for multiple purposes. We believe that such an approach to distributed data sharing in rheumatology will help advance science and improve clinical practice. Disclosure: P. J. Embi, None; W. Stephens, None; R. Myslinski, None.

207 Frequency and Predictors Of Hypertension Communication In Rheumatoid Arthritis Visits. Christie M. Bartels, Heather Johnson, Elizabeth A. Jacobs, Patrick McBride and Maureen Smith. University of Wisconsin School of Medicine and Public Health, Madison, WI

Conclusion: The proportion of patients prescribed a DMARD in primary care in this study is lower than generally accepted estimates. This may reflect reality for patients with RA outside of specialist centres although secondary care prescribing is not captured within this database. Nevertheless, prescriptions that are captured should be representative of overall trends in prescriptions within the UK and suggests that guidelines published by a national body can improve the proportion of patients receiving DMARD treatment in the first year after diagnosis of RA.

Background/Purpose: Patients with rheumatoid arthritis (RA) have a 60% higher cardiovascular disease (CVD) event risk, yet previously we demonstrated they were ⬃30% less likely to have hypertension diagnosed. We sought to examine the frequency and predictors of documented hypertension/blood pressure (BP) communication during rheumatology visits for RA patients with undiagnosed and/or uncontrolled hypertension. Methods: Electronic health record searches from 2004–11 identified RA patients (defined by ICD-9 algorithm) who had uncontrolled or undiagnosed

Disclosure: G. L. Wallace, None; C. J. Edwards, Roche Pharmaceuticals, 2, UCB, 2, Abbott Immunology Pharmaceuticals, 2, Pfizer Inc, 2, GlaxoSmithKline, 2; N. K. Arden, Merck Pharmaceuticals, 5, Roche Pharmaceuticals, 5, Smith & Nephew, Inc., 5,

S83

Sunday, October 27

Temporal Trends In The Prescribing Of Disease Modifying AntiRheumatic Drugs For Rheumatoid Arthritis and The Impact Of Guidelines. Gemma L Wallace1, C. J. Edwards2, Nigel K. Arden3, Daniel PrietoAlhambra4 and Andrew Judge5. 1University of Oxford, Oxford, United Kingdom, 2University Hospital Southampton, Southampton, United Kingdom, 3Oxford NIHR Musculoskeletal Biomedical Research Unit, University of Oxford, Oxford, United Kingdom, 4URFOA-IMIM, Parc de Salut Mar; Idiap Jordi Gol; University of Oxford; University of Southampton, Barcelona, Spain, 5Oxford University, Oxford, United Kingdom

Sunday, October 27

hypertension and who had encounters at 1 of 3 rheumatology clinics. Using Joint National Committee-7 criteria, undiagnosed was defined as ⱖ3 BPs ⱖ140/90 mmHg or 2 readings ⱖ160/100 mmHg without prior diagnosis or antihypertensive medication; BP control was ⱖ3 consecutive BP’s ⬍140/90 mmHg. We abstracted rheumatology clinic notes from first eligible date for hypertension diagnosis/control through date of hypertension control or end of data. We examined the frequency and predictors of any hypertension communication beyond vital sign documentation. We used multivariate logistic regression to examine the impact of systolic BP stage at a rheumatology visit on the predicted probability (PP) of hypertension communication. Covariates included patient sociodemographics, comorbidity, utilization, visit date, and rheumatologist. Results: Among 1267 RA patients, 501 were studied who had uncontrolled hypertension, and 232 of the 501 patients met criteria and lacked a hypertension diagnosis. Mean age was 62 years, 76% were female, 11% were current and 35% former smokers, and prior CVD was noted in ⬎20%. In 2677 abstracted visits, BP’s varied with 20% Normotensive (⬍120 mmHg) at a given visit, 45% Prehypertensive (ⱖ120 - ⬍140 mmHg), 32% with Stage I (ⱖ140 - ⬍160 mmHg) and 11% with Stage II (ⱖ160 mmHg) elevations. Overall 23% of RA visits contained any hypertension communication. After adjustment, even in visits with systolic BPs ⱖ160 mmHg, only 25% contained hypertension communication (PP 25%, CI 20,31). Compared to Stage I systolic elevations, Stage II elevations did not significantly increase the probability of communication (Figure 1), although both were higher than normotension and prehypertension. When examining other predictors, active tobacco users were least likely to have hypertension communication (PP 12%, CI 8,16).

savings. To assess whether managed care policies that limit access to infusion biologics in order to shift new patients to injection biologics (SQ) have had the desired effect. Methods: The share of utilization of biologic products for plans identified with ST policies, were evaluated versus a comparison cohort using three separate methodologies. ST plans were first identified using data from Symphony Health Solutions Corp.’s ProMetis database. This database provides a unique opportunity for this type of analysis, as it spans multiple identifiable plans which can be cross-referenced to known ST policies. Data included payer, prescription and procedure claims with unique anonymised patient identifiers. Analyses #1 & #2 included data from plans with ST policies throughout 2010. Analysis #1 compared the percentage of patients with biologic claims for plans with ST policies to all other plans in the database during the same time period (ALL OTHER). Analysis #2 compared the percentage of patients with biologic claims for plans with ST policies to a set of plans matched roughly to the ST plans based on region and relative size (MATCHED). Analysis #3 included data from 1/1/2006 through 4/30/2011 for plans with a known ST implementation date (index date) and with available biologic claims within 365 days before and after this index date. Analysis #3 evaluated the period after ST policy implementation to the period before implementation (PRE/POST). Data for each analysis were descriptive of the number of patients with access and using any biologic in the ST and comparison cohorts. The net change in percentage of patients with infusion claims is reported. Results: Sixteen different plans were identified with a ST policy. The only analysis to demonstrate a shift in patient utilization away from IV biologics was the ALL OTHER comparison (⫺5.1%). When comparison plans were MATCHED to ST plans, or when evaluating existing plans using PRE/POST methods, patient access and use of infusion biologics increased (⫹7.0% and ⫹2.8% respectively). Conclusion: These data suggest that policies designed to reduce overall patient proportions using infusion biologics may not have the intended effect, or may at most affect 5.1% of biologic users in any given year. The limited potential benefit needs to be weighed against the cost of implementing and monitoring this type of policy, as well as the provider resource implications. Disclosure: M. P. Ingham, Janssen Scientific Affairs, LLC, 3; A. Paris, Janssen Scientific Affairs, LLC, 5; L. A. Ellis, Janssen Scientific Affairs, LLC, 3; C. Kozma, Janssen Scientific Affairs, LLC, 5.

209 Focus on cardiovascular Risk Factor Recording In a Rheumatology Outpatient Clinic. Eirik Ikdahl1, Silvia Rollefstad1, Inge C. Olsen1, Tore K. Kvien2, Inger Johanne Widding Hansen3, Dag Magnar Soldal4, Glenn Haugeberg4 and Anne Grete Semb1 1Diakonhjemmet Hospital, Oslo, Norway, 2University of Oslo, Oslo, Norway, 3Hospital of Southern Norway Trust, Kristiansand, Norway, Kristiansand, Norway, 4Hospital of Southern Norway Trust, Kristiansand, Norway

Conclusion: Despite higher CVD risk, many RA patients had undiagnosed or uncontrolled hypertension. Even with systolic BPsⱖ160 mmHg only 1 in 4 rheumatology encounters documented communication about hypertension, demonstrating lost opportunities for identifying and managing modifiable CVD risks. Future research should investigate patient and provider role perceptions and other barriers and facilitators of hypertension management to systematically improve CVD prevention in RA patients.

Background/Purpose: There is an unmet need of implementing the knowledge of cardiovascular (CV) risk in patients with rheumatoid arthritis (RA) into clinical practice. Our aim was to evaluate CV risk factor (CVRF) recording in a rheumatology outpatient clinic (ROC), where the standard was annual CVRF recording for all patients. Moreover, we evaluated how various clinical models influenced the extent of CVRF recording, comparing a regular ROC (RegROC) to an arthritis clinic (AC), a structured, team-based model. Methods: Of the 1142 RA patients visiting ROC at the Hospital of Southern Norway during 2012, 612 attended RegROC and 530 attended the AC. Allocation to either RegROC or the AC was based on clinical judgement by the rheumatologist and did not include any inclusion criteria. For all patients, CVRFs were to be recorded in the medical journal as well as in a computerized journal program, GoTreatIT-rheuma (GTI-r). We searched both journal systems to ascertain how many patients had CVRFs recorded. Results: Overall, 38.2% of patients had CVRFs recorded and only 26.9% had recorded all the CVRFs included in the CV risk calculator, SCORE. When comparing the AC to the RegROC, odds ratios (OR) for CVRFs being recorded in the medical journal was for: lipids: 5.0–6.0, blood pressure (BP): 12.4, glucose: 9.1, and HbA1c: 6.1 (p for all ⬍ 0.001) (Table). In the GTI-r

Disclosure: C. M. Bartels, None; H. Johnson, None; E. A. Jacobs, None; P. McBride, None; M. Smith, None.

208 Three Approaches To Evaluating Step Therapy Policies For Immune Disorder Specialty Pharmaceuticals. Michael P. Ingham1, Andrew Paris2, Lorie A. Ellis1 and Chris Kozma3. 1Janssen Scientific Affairs, LLC, Horsham, PA, 2Vigilytics, Victor, NY, 3CK Consulting, Saint Helena Island, SC Background/Purpose: Rheumatology practices are increasingly exposed to patient health care benefit plans that include “step therapy” (ST) policies which affect medication choice. Managed care companies implement ST policies for specialty pharmaceuticals to reduce utilization of infused biologics (IV) in new patients, based on an assumption that there will be cost

S84

adherence while other factors including gender and gout-related factors (e.g., flares) are less consistent across studies. Conclusion: This is the first systematic review of medication adherence, with particular focus on gout patients. Adherence rates may vary according to methods used to measure adherence. Overall, synthesis of current evidence suggest that medication non-adherence is substantial in gout. Findings highlight the importance of discussing adherence with gout medications during health care professional encounters with gout patients.

Table. Cardiovascular risk factors recorded in patients attending rheumatology consultations in a rheumatology outpatient clinic Cardiovascular Risk factors Medical journal: n (%) Brachial BP Total cholesterol LDL-cholesterol HDL-cholesterol Triglycerides Fasting blood glucose HbA1c GTI-r journal n (%) Brachial BP Lipid values Smoking Complete risk profile CV medication CV co-morbidities

ROC (nⴝ1142)

AC (nⴝ530)

RegROC (nⴝ612)

OR* (95% CI)

AC vs. RegROC P-value*

576 (50.4) 537 (47.0) 503 (44.1) 515 (45.1) 472 (41.3) 351 (30.7) 385 (33.7)

421 (79.4) 354 (66.8) 347 (65.5) 350 (66.0) 333 (62.8) 281 (53.0) 284 (53.6)

155 (25.3) 183 (29.9) 156 (25.5) 165 (27.0) 139 (22.7) 70 (11.4) 101 (16.5)

12.36 (9.27, 16.48) 5.02 (3.89, 6.48) 5.87 (4.53, 7.62) 5.59 (4.31, 7.23) 6.02 (4.63, 7.82) 9.11 (6.71, 12.35) 6.10 (4.62, 8.04)

⬍ 0.001 ⬍ 0.001 ⬍ 0.001 ⬍ 0.001 ⬍ 0.001 ⬍ 0.001 ⬍ 0.001

422 (37.0) 378 (33.1) 756 (66.2) 307 (26.9) 251 (22.0) 231 (20.2)

371 (70.0) 321 (60.6) 378 (69.3) 276 (52.1) 198 (37.4) 184 (34.7)

51 (8.3) 57 (9.3) 378 (61.8) 31 (5.1) 53 (8.7) 47 (7.7)

27.46 (19.41, 38.85) 15.90 (11.45, 22.08) 1.44 (1.12, 1.85) 20.97 (14.04, 31.33) 6.31 (4.52, 8.82) 6.43 (4.53, 9.14)

⬍ 0.001 ⬍ 0.001 0.05 ⬍ 0.001 ⬍ 0.001 ⬍ 0.001

Study Prescription Records Riedel 2004 Sarawate 2006 Briesacher 2008 Solomon 2008 Halpern 2009* Harrold 2009 Harrold 2010* Park 2012 Zandman 2013 Electronic Monitoring de Klerk 2003

*Adjusted for age and gender ROC: rheumatology outpatient clinic, RegROC: Regular rheumatology outpatient clinic, AC: Arthritis clinic, CV: Cardiovascular, LDL: Low-density lipoprotein, HDL: High-density lipoprotein, BP: Blood pressure, HbA1c: Glycated haemoglobin, GTI-r: GoTreatIT-rheuma, Complete risk profile: Complete lipid values, smoking and blood pressure, CV medication: Anti-hypertensive and statins, CV co-morbidities: Hypertension, angina pectoris, acute myocardial infarction, percutaneous coronary intervention, coronary artery bypass graft surgery, cerebrovascular accident, premature familiar cardiovascular disease

Self-Report Silva 2010 Dalbeth 2011* Dalbeth 2012* Martini 2012*

N

Follow-up

5,597 2,405 9,715 9,823 10,070 4,166 4,166

2 2 1 1 1 6 6

yr yr yr yr yr yr yr

Adherence Measure weighted average compliance rate medication possession ratio ⬎0.80 medication possession ratio ⬎0.80 proportion days covered ⬎0.80 medication possession ratio ⬎0.80 medication possession ratio ⬎0.80 therapy gap (⬎60 days no refill)

Main Result 18% adherent 26% adherent 36.8% adherent 36% adherent 44% adherent 44% adherent 30% adherent (no gap) 27% adherent 17% adherent

352 7,644

1 yr 6 yr

proportion days covered ⬎0.80 proportion days covered ⬎0.80

29

1 yr

taking compliance dosing compliance

colchicine: 65% allopurinol: 84% colchicine: 44% allopurinol: 74%

34 142 273 60

1 yr n/a n/a n/a

taking medication regularly questionnaire score (45⫽high) questionnaire score (45⫽high) semi-structured interview

16.7% adherent 39.8 40.2 79% adherent

*medication adherence evaluated as exposure and not outcome

Conclusion: Despite high focus on CV disease, recording of CVRFs in the ROC was low; however it was augmented, but still not satisfactory in a structured, team-based model. There is necessity for improving systems for CVRF recording in patients with RA.

Disclosure: M. De Vera, None; S. Rai, None; V. Bhole, None.

Disclosure: E. Ikdahl, None; S. Rollefstad, None; I. C. Olsen, None; T. K. Kvien, None; I. J. W. Hansen, None; D. M. Soldal, None; G. Haugeberg, None; A. G. Semb, None.

The Economic Burden Of Gout: A Systematic Review Of Direct and Indirect Costs. Sharan Rai1, Aliya Haji1, Lindsay C Burns2 and Hyon K. Choi3. 1Arthritis Research Centre of Canada, Richmond, BC, 2University of British Columbia, Vancouver, BC, 3Boston University School of Medicine, Boston, MA

211

210 Medication Adherence In Patients With Gout: A Systematic Review. Mary De Vera1, Sharan Rai1 and Vidula Bhole2. 1Arthritis Research Centre of Canada, Richmond, BC, 2EpiSolutions Consultancy Services, Thane, India

Background/Purpose: The prevalence of gout, an excruciating and disabling joint disease, has been increasing in recent decades such that it now constitutes the most common inflammatory arthritis in the US. A high rate of uncontrolled disease and high comorbidity burden suggest that the economic impact of gout could potentially be substantial. To clarify this impact, we systematically reviewed the literature on the productivity loss and healthcare costs associated with gout. Methods: We conducted a mapped search of MEDLINE, EMBASE, IPA, and CINAHL databases for articles published between Jan 1993 to Jun 2013 that reported either direct or indirect costs of gout, encompassing healthcare utilization and productivity loss, respectively. Our search strategy employed mapped subject headings terms together with keywords for unindexed terms relating to the themes of gout and cost. Titles and abstracts were reviewed for preliminary inclusion criteria of: 1) full-text, original, published article; 2) gout patient population; 3) direct or indirect costs reported; 4) English language. Non-human studies and case-reports/series were excluded. Where possible, data were abstracted and tabulated on annual all-cause and gout-related direct and indirect costs per patient. Results: Our search strategy yielded a total of 2,954 unique articles. Of these, 11 studies reporting direct and indirect costs for gout met our inclusion criteria and were included for review. Seven studies reported gout-specific data on direct costs (medical services ⫹ prescriptions), 5 on medical services, and 4 on indirect costs (work productivity loss). US estimates of the annual direct costs per patient ranged from $3,985–$22,562 for all-cause costs and $192–$5,924 for gout-specific costs. Annual per-patient medical services costs ranged from $3,122– $14,866 and corresponding indirect costs ranged from $915–$3,900 US. Patient characteristics associated with increased costs included: 1) ⱖ 3 flares per year; 2) serum uric acid levels ⱖ 6 mg/dL; 3) presence of tophi (Table).

Background/Purpose: Recent data suggesting the growing problem of medication non-adherence in gout has called for the need to understand the magnitude of the problem as well as its determinants. To date, only one systematic review has summarized medication adherence across a variety rheumatic diseases, including three studies in gout. To update this data as well as better understand patterns and determinants of adherence in this population, our objective was to a conduct a systematic review of the literature examining medication adherence among patients with gout. Methods: We conducted a systematic search of MEDLINE (1946-), EMBASE (1974-), and INTERNATIONAL PHARMACEUTICAL ABSTRACTS (1970-) databases and selected original studies that included patients with gout and measured and/or reported medication adherence in real-world settings. We extracted information on: 1) study design, 2) sample size, 3) length of follow-up, 4) data source (e.g. prescription records vs. electronic monitoring vs. self-report), and 5) adherence measures and reported estimates. We assessed quality of studies by adapting and applying published recommendations for the reporting, assessment, and evaluation of medication adherence studies. Results: After screening 963 potential articles, 14 met inclusion criteria. We divided studies according to methods used to measure adherence including prescription records (9), electronic monitoring devices (1), and self-report (4). Studies using prescription records have the largest sample sizes and lowest reported adherence, ranging from 17% to 44%. Higher adherence rates (or scores) were reported in studies based on self-report, which may be due to social desirability bias. With respect to determinants, younger age has been shown as a predictor of poor

S85

Sunday, October 27

journal the discrepancies between the AC and RegROC were even more pronounced. OR for CVRFs being recorded was for lipids: 15.9, BP: 27.5 and for having recorded all the CVRFs included in SCORE: 21.0 (p for all ⬍ 0.001).

Sunday, October 27

Gout Population

DIRECT (MEDICAL SERVICES ⴙ PRESCRIPTION) COSTS All-Cause Gout-Related Healthcare Cost Healthcare Cost Country Cost Year

⬍3 attacks

$9,009a-$10,547b

$192b

USA

ⱖ3 attacks/year

$9,748a-$17,603b

$870b-$5,924b

USA

Employees with gout SUA ⬍ 6.0 mg/dL

$3,985c $11,365d-$15,237e

⫺ $332d-$505b

USA USA

SUA ⱖ6.0 and ⬍ 9.0 mg/dL SUA ⱖ 9.0

$11,551d-$14,935e

$353d-$696b

USA

$14,474d-$18,340e

$663d-$723e

USA

Gout patients with tophi Gout patients without tophi

$22,562b

⫺

USA

$14,574b

⫺

USA

2005

Wu et al. 2008

Physician diagnosed gout ⱖ3 attacks/year

$14,734b

$876b

USA

2005

Wu et al. 2008

€ 1259f

⫺

Spain

2007

Sicras-Mainar et al. 2013

Gout Population ⬍3 attacks/year ⱖ3 attacks/year Employee with gout ⱖ3 attacks/year Gout Population ⬍3 attacks/year ⱖ3 attacks/year Employee with gout SUA ⱖ 6.0 mg/dL ⱖ3 attacks/year 1–2 attacks/year

2011 Lynch et al. 2013; Saseen et al. 2012 2008–2011 Lynch et al. 2013; Saseen et al. 2012; Wu et al. 2012 2001–2004 Brook et al. 2006 2002–2010 Halpern et al. 2009; Park et al. 2012; Wu et al. 2008; 2002–2010 Halpern et al. 2009; Park et al. 2012; Wu et al. 2008; 2005–2010, Halpern et al. 2009; NS* Park et al. 2012; Wu et al. 2008; 2005 Wu et al. 2008

MEDICAL SERVICES COSTS All-Cause Gout-Related Healthcare Cost Healthcare Cost Country Cost Year $7,332a-$8,209b $8,505b-$14,866b

$176b $834b-$5,477b

USA USA

$3,122c € 2517f

⫺ ⫺

USA Spain

USA USA USA USA Spain Spain

References

2011 Lynch et al. 2013; Saseen et al. 2012 2008–2011 Lynch et al. 2013; Saseen et al. 2012; Wu et al. 2012 2001–2004 Brook et al. 2006 2007 Sicras-Mainar et al. 2013

INDIRECT (WORK PRODUCTIVITY LOSS) COSTS Indirect Cost Country Cost Year $915a $2,021a $2,885c $3,900b € 88f € 29f

feet at both visits. Two readers analysed the DECT scans for the total urate volume in both feet. Paired scans were read in chronological order. The readers were blinded to each other’s measurements and all clinical measures including serum urate concentrations. Analysis included inter-reader intraclass correlation coefficients (ICCs) and limits of agreement, and calculation of the smallest detectable change (SDC). Results: Allopurinol was prescribed in 85% patients during the study period. Mean (SD) SU concentration for both timepoints was 6.3 (1.5) mg/dL. The median (IQR) baseline DECT urate volume was 0.49 (0.16, 2.18) cm3, and change in DECT urate volume was ⫺0.01 (⫺0.40, 0.28) cm3. Interreader ICCs were 1.00 for baseline DECT volumes and 0.93 for change values. Inter-reader bias (SD) for baseline volumes was ⫺0.18 (0.63) cm3 and for change was ⫺0.10 (0.93) cm3. The SDC was 0.91cm3. There were 47 (64%) patients with baseline DECT urate volumes ⬍0.91cm3. Higher mean SU concentrations were observed in patients with increased DECT urate volumes above the SDC (one way ANOVA p⫽0.006). However, a relationship between changes in DECT urate volumes and mean SU concentrations was not observed in the entire group (Figure).

References

2011 2011 2001–2004 2006 2007 2007

References Lynch et al. 2013 Lynch et al. 2013 Brook et al. 2006 Edwards et al. 2011 Sicras-Mainar et al. 2013 Sicras-Mainar et al. 2013

aAdjusted for age, sex, marital status, race, exempt status, full-time status, salary, tenure, region, and history of flares bUnadjusted cost cAdjusted for age, sex, annual salary, tenure, exempt status, race, marital status, location, and Charlson comorbidity index dAdjusted for age, sex, insurance, Charlson comorbidity index, presence of hypertension, and number of all-cause prescriptions eAdjusted for age, sex, index year, Charlson comorbidity index, and medication use fAdjust for age, sex, resource utilization, and Charlson comorbidity index

Conclusion: Overall, the economic burden of gout was found to be substantial, with direct healthcare costs comparable to chronic rheumatic conditions such as rheumatoid arthritis. There was a paucity of data on indirect costs associated with gout, the population costs of gout (rather than patient subgroups), a lack of standardized cost reporting, and a lack of validated instruments for cost assessment in gout. Characteristics associated with increased costs generally reflected poorly controlled disease and were largely modifiable. As gout represents a metabolically-driven arthropathy that can be fully controlled with proper therapeutic approaches, substantial resources could be spared through by closing the gap between guideline recommendations and practice.

Figure. Scatter plot showing the relationship between change in DECT urate volumes and mean serum urate concentrations (baseline and Year 1).

Conclusion: DECT urate volume measurement has high inter-reader agreement but low sensitivity to change over one year in patients with tophaceous gout receiving conventional urate-lowering therapy. These data raise questions about the role of DECT as an outcome measure for clinical studies of gout. Disclosure: A. Rajan, None; O. Aati, None; R. Kalluru, None; G. Gamble, None; A. Horne, None; A. Doyle, None; F. M. McQueen, None; N. Dalbeth, None.

Disclosure: S. Rai, None; A. Haji, None; L. C. Burns, None; H. K. Choi, None.

213 ACR Poster Session A

Digital Tomosynthesis for Measurement of Bone Erosion in Gout: Comparison With Computed Tomography. Nicola Dalbeth1, Anthony Doyle1, Mark Roger2, Angela Gao2 and Fiona M. McQueen1. 1University of Auckland, Auckland, New Zealand, 2Auckland District Health Board, Auckland, New Zealand

Imaging of Rheumatic Diseases I: Imaging in Gout, Pediatric, Soft and Connective Tissue Diseases Sunday, October 27, 2013, 8:30 AM–4:00 PM

212

Background/Purpose: Digital tomosynthesis is a recently developed imaging method in which multiple projected images obtained at different angles are collected with a digital detector. These projected images are used to reconstruct multiple thin planar images of the scanned object in a plane parallel to the detector. Recent reports indicate that this technology allows low-cost, low-radiation detection of bone erosion in patients with inflammatory arthritis. The aim of this study was to determine the sensitivity and reproducibility of tomosynthesis, compared with conventional computed tomography (CT), for measurement of bone erosion in gout. Methods: Tomosynthesis and CT scans of the dominant wrist were prospectively obtained in 37 patients with gout (mean disease duration 18 years, 86% with subcutaneous tophi) (Figure). Each scan was scored separately by two radiologists for bone erosion using semi-quantitative volume assessment (sites and 0–10 scoring as for RAMRIS erosion method).

Dual-Energy Computed Tomography for Monitoring of Urate Deposition in Tophaceous Gout: A Prospective Longitudinal Study Examining Sensitivity to Change. Ashwin Rajan1, Opetaia Aati1, Ramanamma Kalluru2, Gregory Gamble1, Anne Horne1, Anthony Doyle1, Fiona M. McQueen1 and Nicola Dalbeth1 1University of Auckland, Auckland, New Zealand, 2Department of Rheumatology, Auckland District Health Board, Auckland, New Zealand Background/Purpose: Dual-energy computed tomography (DECT) is an advanced imaging method with potential for monitoring urate deposition in patients with gout. The aim of this prospective longitudinal study was to analyse the sensitivity to change of DECT urate volume measurement. Methods: Seventy-three patients with tophaceous gout attended study visits at baseline and 12 months. All patients had a comprehensive clinical assessment including serum urate (SU) testing and DECT scanning of both

S86

Figure. A. Tomosynthesis and B. corresponding coronal CT images of the right wrist in a participant with tophaceous gout.

Results: The mean total erosion score in the tomosynthesis scans was lower than CT scans (10.8 vs 13.2, p⫽0.004). For many individual sites (radius, scaphoid, triquetrum, trapezium, capitate, MC1 base), tomosynthesis erosion scores were significantly lower than CT scores. There were no sites where tomosynthesis scores were higher than CT. The inter-reader intraclass correlation coefficients (ICC) for the total erosion score was 0.64 (0.59–0.89) for tomosynthesis and 0.80 (0.64–0.94) for CT. At all individual sites except MC2 base, ICCs were lower for tomosynthesis than CT. We were unable to identify a combination of sites for which mean scores or ICCs for tomosynthesis approximated those of CT. There was a high correlation between tomosynthesis and CT total scores (r⫽0.95, p⬍0.0001) and at most individual sites. Total erosion scores for both tomosynthesis and CT correlated highly with hand tophus count, grip strength, and Health Assessment Questionnaire scores (p⬍0.001 for all). Conclusion: Tomosynthesis erosion scores correlate highly with CT erosion scores and clinical measures of disease severity in patients with gout. However, tomosynthesis has lower reproducibility and detection of erosion compared with CT. These findings indicate that CT remains the gold standard for measurement of bone erosion in gout. Disclosure: N. Dalbeth, None; A. Doyle, None; M. Roger, None; A. Gao, None; F. M. McQueen, None.

214 Use Of Dual-Energy Computed Tomography In Evaluation Of Axial Gout. Kristin Logee1, Ranadeep Mandhadi1, William Traverse2 and Santhanam Lakshminarayanan3. 1University of Connecticut, Farmington, CT, 2 Saint Francis Hospital and Medical Center, Hartford, CT, 3University of Connecticut School of Medicine, Farmington, CT Background/Purpose: Axial (spinal) involvement has become increasingly recognized as a potential manifestation of gouty arthritis. The presentation can vary widely from mild, asymptomatic disease to severe back pain, radiculopathy or frank spinal cord compression. Axial gout is frequently a missed diagnosis for a variety of reasons: it can mimic other spine diseases, is under-recognized as a possible manifestation of gout, and current imaging techniques (X-ray, MRI and CT) often produce non-specific findings which do not distinguish gouty erosions and tophi from other pathologies. Dual-energy CT has recently been recognized as very sensitive and specific for identifying monosodium urate (MSU) crystal deposition. By using a specific display algorithm that assigns different colors to materials of different chemical composition, MSU can be color-coded and thereby distinguished from surrounding structures. Up until now, it has been used mainly to image peripheral sites but has not been evaluated as a diagnostic tool for axial gout. Methods: We present two cases that demonstrate the efficacy of dualenergy CT in detecting MSU deposition within the axial skeleton. Patient X was a 91 year-old male with severe neck pain, fevers and leukocytosis. MRI showed abnormal T2 signal around the C4-C5 facet joint with fluid consistent with synovitis. However, septic arthritis could not be ruled out. The patient also complained of right knee pain and swelling. Aspirate of the knee revealed intracellular MSU crystals with negative cultures, consistent with acute gout flare. This raised suspicion that the patient’s presentation could be secondary to spinal involvement of gout. Patient Y was a 29 year-old male with a long-standing history of chronic tophaceous gout who presented with incapacitating mid-low back pain. Traditional CT showed extensive subchondral erosions throughout the lumbar spine and right sacroiliac joint.

S87

Sunday, October 27

Results: Dual-energy CT of the cervical spine of Patient X (Image 1) showed MSU deposition on the cervical facet joints (color-coded in green) which correlated with the MRI findings. The patient’s symptoms quickly resolved with colchicine and IV methylprednisolone. Similarly, lumbar spine dual-energy CT of Patient Y (Image 2) showed extensive MSU deposition along the transverse processes and pelvic bony structures. Symptoms remitted with colchicine and oral prednisone.

Sunday, October 27

Conclusion: Dual-energy CT can be used to visualize the presence of axial MSU deposition. This may lead to appropriate diagnosis and management of axial gout while avoiding invasive procedures and erroneous treatment.

Conclusion: MSU crystals are frequently present in joints affected by radiographic damage in patients with gout. These findings support the concept that MSU crystals directly interact with articular tissues to influence the development of structural joint damage in this disease.

Disclosure: K. Logee, None; R. Mandhadi, None; W. Traverse, None; S. Lakshminarayanan, None.

Disclosure: N. Dalbeth, None; O. Aati, None; R. Kalluru, None; A. Horne, None; A. Doyle, None; F. M. McQueen, None.

215 216

Relationship Between Structural Joint Damage and Urate Deposition In Gout: A Site-By-Site Analysis Using Plain Radiography and Dual Energy Computed Tomography. Nicola Dalbeth1, Opetaia Aati1, Ramanamma Kalluru2, Anne Horne1, Anthony Doyle1 and Fiona M. McQueen1. 1University of Auckland, Auckland, New Zealand, 2Department of Rheumatology, Auckland District Health Board, Auckland, New Zealand

Systematic Staging For Uric Acid Deposits With Dual-Energy Computed Tomography and Ultrasound In Suspected Gout. Wolfgang A. Schmidt1 and Alexander Huppertz2. 1Med Ctr Rheumatology Berlin Buch, Immanuel Krankenhaus Berlin, Berlin, Germany, 2Imaging Science Institute Charite´, Berlin, Germany

Background/Purpose: Structural joint damage, including erosion, joint space narrowing (JSN) and new bone formation (NBF), is frequently observed in patients with tophaceous gout. Although imaging studies have implicated tophi in the development of these changes, the relationship between monosodium urate (MSU) crystals and structural joint damage has not been examined. The aim of this work was to examine the relationship between joint damage and MSU crystal deposition in patients with tophaceous gout. Methods: Plain radiographs and dual energy computed tomography (DECT) scans of the feet were prospectively obtained from 92 patients with tophaceous gout. Two readers analysed the 10 metatarsophalangeal joints (MTPJs) for the presence of MSU crystals using DECT (920 total joints analysed, inter-reader agreement 94.2% and kappa 0.82). For the purposes of analysis, DECT MSU crystal deposition was considered present if recorded by both readers. A further reader, who was blinded to the DECT results, scored the same joints on plain radiography for Sharp-van der Heijde erosion score (0–10), JSN score (0–4) and presence of NBF features (spur, sclerosis and osteophyte). Results: DECT MSU crystal deposition was more frequently observed in joints with erosion (odds ratio (OR) 11.2), JSN (OR 5.1), spur (OR 13.4), osteophyte (OR 5.3) and sclerosis (OR 9.1), p ⬍0.0001 for all. For those joints with any erosion, erosion scores were higher in joints with DECT urate deposition compared with those without MSU crystals (mean (SD) erosion scores 5.2 (2.7) and 3.3 (2.11) respectively, p⬍0.0001). A strong linear relationship was observed in the frequency of joints affected by MSU crystals with increasing radiographic erosion score (p⬍0.0001). In contrast, for those joints with any narrowing, JSN scores were no different in those with and without DECT urate deposition (p⫽0.75). DECT MSU crystal deposition and all features of joint damage were most frequently observed at the 1st MTPJ (Figure). There was a very high correlation between the number of joints at each site affected by MSU crystal deposition and all features of radiographic joint damage (r⬎0.88 for all, p⬍0.05 for all).

Background/Purpose: To evaluate the diagnostic accuracy of DualEnergy Computed Tomography (DECT) and ultrasound for detecting monosodium urate crystal deposits in patients with clinically suspected gout. Methods: In this case control study DECT and comprehensive ultrasound of feet, knees, hands, wrists, and elbows were performed bilaterally in 60 consecutive patients (49 males; 11 females; mean age, 62 years; age range, 36–82 years) of a tertiary rheumatology referral center with suspected gout and compared with pooled clinical information including polarization microscopy, maximum documented uric acid levels, presence of podagra and the final rheumatological diagnosis as standard-of-reference. Results: Finally, 39 patients were classified as gout positive, 31 of which had been newly diagnosed. Sixteen of these patients had gout and a concomitant rheumatic disease. Although an experienced rheumatologist aimed at receiving material from tissue or joint aspirates of every patient, if necessary with ultrasound guidance, the diagnosis could be confirmed by polarization microscopy only in 46% of the gout patients. DECT had a sensitivity of 84.6%. Ultrasound had a sensitivity of 100%. The specificity for the diagnosis of gout was 85.7% for DECT and 76.2% for ultrasound. The positive predictive value was 0.92 for DECT and 0.89 for ultrasound. The negative predictive value was 0.75 for DECT and 1 for ultrasound. Uric acid deposits occurred most commonly in the medial and lateral aspects of the knee joint, in the quadriceps tendon and in the first MTP 1 joints. The evaluation on a joint-basis compared to ultrasound as reference revealed a sensitivity of 46.2% and a specificity of 97.6% for DECT. Ultrasound detected smaller crystal deposits than DECT. DECT failed to show crystal deposits on the cartilage, representing the “ultrasound double contour sign”, but delineated larger intra-articular and extra-articular tophi. Ultrasound was false positive in one patient with calcium pyrophosphate dihydrate disease (CPPD), hydroxyapatite deposition disease and rheumatoid arthritis, respectively, and in two patients with severe peripheral arterial occlusive disease. DECT was false positive in 3 patients showing minimal deposits suggestive of gout tophi in the lateral menisci. One patient was finally diagnosed with CPPD, one with psoriatic arthritis and one with undifferentiated oligoarthritis. Small signals that are not representing urate deposits also appear in nails. The DECT volumetry computed a mean uric acid deposit load of 2.0 cm3 (SD 9.6 cm3). A mean effective dose between 0.4 and 0.5 mSv was estimated. Conclusion: DECT is specific for the diagnosis of gout particularly when excluding small signals in the lateral menisci and in the nails. However, it fails to detect small uric acid deposits. It is particularly useful for patients with ambivalent findings, concomitant rheumatic diseases, and for those whose joint aspiration and microscopy did not successfully detect gout crystals. Disclosure: W. A. Schmidt, Esaote, 2, GE Healthcare, 2; A. Huppertz, Siemens, 3.

217 Clinical Value Of 18f-Fluoro-Dexoxyglucose Positron Emission Tomography In Patients With Adult-Onset Still’s Disease. Hiroyuki Yamashita, Kazuo Kubota, Yuko Takahashi, Hiroshi Kaneko, Toshikazu Kano and Akio Mimori. National Center for Global Health and Medicine, Tokyo, Japan Figure. Examples of plain radiographs (upper panel) and corresponding axial DECT images (lower panel) of eroded 1st MTPJs from three separate participants. MSU crystals are shown as red on DECT images.

Background/Purpose: The aim of this study was to assess the usefulness of 18F-fluoro-dexoxyglucose positron emission tomography/Computed tomo-

S88

Mason score evaluation) features were analyzed. Patients were classified according to ACR/EULAR criteria for SS (AEC).[4] Results: all patients with HCV infection were counted out as well as all the patients with one of the exclusion criteria present in the Primary Sjogren Syndrome classification. Male/female ratio was 1/9; 91.2 % of patients presented xerostomia, 75% xerophthalmia. Schirmer test was positive in 72% of the patients; break up time positive in 65%. The biopsy was positive (Chisholm and Mason score ⬎1) in 61% and ultrasound was positive (score ⱖ2 out of 4) in 40.2%. ENA test (SSA and⶿or SSB) was positive in 37.4%. The diagnosis of Sjogren Syndrome has been assessed, according to AECG classification criteria, in 36 patients out of 82 (43.9%). 80.6% of patients with positive ultrasound were classified as affected by Sjogren Syndrome, 89.1% of patients with negative ultrasound didn’t fill the classification criteria. The ultrasound score presented a positive likelihood ratio of 4.58 (95% confidence interval 2.34⬍x⬎8.98) and a negative one of 0.135 (0.058⬍x⬍0.313). The test showed a sensibility of 80.6% and a specificity of 89.1%. Concordance (Cohen K) between classification criteria and US score was 0.701 (medium level), higher than that between classification criteria and biopsy ChisholmMason score (0.572). Surprisingly ENA screening didn’t add diagnostic power to the ultrasound scan due to a wide overlapping of the two tests. Conclusion: in our experience salivary gland ultrasound seems to be an useful tool in the diagnostic process of Sjogren Syndrome due to its good sensibility and specificity. Moreover the exam is simple to perform, fast, incruent and well accepted by patients. References: 1. Cornec D et al. Contribution of salivary gland ultrasonography to the diagnosis of Sjo¨gren’s syndrome: Toward new diagnostic criteria? Arthritis Rheum. 2013 Jan;65(1):216–25. 2. Milic V. et al. Ultrasonography of major salivary glands could be an alternative tool to sialoscintigraphy in the American-European classification criteria for primary Sjogren’s syndrome. Rheumatology (Oxford). 2012 Jun;51(6):1081–5 3. Ariji Y, Ohki M, Eguchi K, et al. Texture analysis of sonographic features of the parotid gland in Sjo¨gren’s syndrome. AJR Am J Roentgenol 1996;166:935–41. 4. Vitali C, Bombardieri S, Jonsson R, et al. Classifi cation criteria for Sjo¨gren’s syndrome: a revised version of the European criteria proposed by the AmericanEuropean Consensus Group. Ann Rheum Dis 2002;61:554–8. Disclosure: G. Germano`, None; N. Possemato, None; O. Addimanda, None; P. Macchioni, None; C. Salvarani, None.

219 Fig. 1. FDG-PET/CT images at diagnosis (A1,B1,C1) and after steroid and Tocilizmab treatment (A2,B2,C2) in a 32-year-old woman (Patient 4) with AOSD.

Contrast-Enhanced Ultrasound Of The Sovraortic Arteries: The Potential Role In Monitoring Disease Activity and Response To Treatment In Large Vessel Vasculitis. Giuseppe Germano`1, Pierluigi Macchioni1, Niccolo` Possemato1 and Carlo Salvarani2. 1Arcispedale S Maria Nuova, IRCCS, Reggio Emilia, Italy, 2Arcispedale S Maria Nuova-IRCCS, Reggio Emilia, Italy

Marked FDG accumulation was observed in the bone marrow, spleen and multiple lymph nodes at the time of diagnosis before treatment. After treatment, bone marrow FDG uptake improved from SUVmax 004.02 (A1) to 2.50 (A2) and spleen FDG uptake decreased from SUVmax 006.05 (A1,C1) to 4.38 (A2,C2). In addition, multiple lymph node FDG uptake observed in the axilla, mediastinum, hilar region of the lung, hilar region of the liver and para-aortic region declined or disappeared (B1/C13 B2/C2).

Background/Purpose: promising data has been recently published about the role of contrast-enhanced ultrasound (CEU) in the diagnosis and follow up of Takayasu arteritis (TA) ⫺1,2Objective: To assess the role of CEU examination of the carotid artery in patients with large vessel vasculitis (LVV) compared to Positron Emission Tomography (PET). Methods: 10 patients (8 TA,2 LVV, mean age 46⫾19y, mean disease duration 3.9⫾3.1y) were contemporary evaluated with total body PET and carotid arteries US gray scale and CEU (Esaote MyLab70, 13-5MH linear probe, contrast agent Sonovue). All the patients underwent complete clinical examination and laboratory determination of acute phase reactants. Imaging results were reported using a semiquantitave score ranging from 0 (no activity) to 3 (high activity). Comparison between the two tests were made using Cohen K test. Results: At US gray scale examination 8/10 patients have high intima/ media thickness (mean 1,5 mm ⫾0.46). In patients with positive contrast enhanced ultrasound the mean carotid intima-media thickness was 2,1mm ⫾0.48, while patients with negative contrast enhanced ultrasound presented a mean wall thickness of 0,9mm ⫾0.47; P ⫽ 0.009. There was no significant differences in clinical, demographic and laboratory data between patients with active and inactive disease. All the 4 patients with high/medium carotid activity at PET examination presented positive (activity score ⬎1) CEU. Five of the 6 patients with low/absent PET signal have negative (activity score ⬍2) CEU (K test ⫽ 0.800, good concordance).

Disclosure: H. Yamashita, None; K. Kubota, None; Y. Takahashi, None; H. Kaneko, None; T. Kano, None; A. Mimori, None.

218 The Role Of Salivary GLAND Ultrasound In Sjogren Syndrome: A Single Center Experience. Giuseppe Germano`1, Niccolo` Possemato1, Olga Addimanda1, Pierluigi Macchioni1 and Carlo Salvarani2. 1Arcispedale S Maria Nuova, IRCCS, Reggio Emilia, Italy, 2Arcispedale S Maria NuovaIRCCS, Reggio Emilia, Italy Background/Purpose: There is increasingly scientific evidence of salivary gland ultrasonography accuracy and usefulness in diagnosing Sjogren Syndrome. [1–2] Methods: 82 patients with suspected Sjogren Syndrome from a single center were evaluated. Demographic (age and sex), clinical (sicca syndrome, artralgia), ultrasonographic (salivary gland ultrasound following the score proposed by Ariji et al [3], serologic (anti nuclear antibodies – ANA, anti ENA, anti HCV) and histological (salivary glands biopsy with Chisholm-

S89

Sunday, October 27

graphy (18F-FDG PET/CT) for the diagnosis or the evaluation of Adult-onset Still’s disease (AOSD) visually. Methods: 7 consecutive AOSD patients who had undergone PET at our department between 2007 and 2012 were included in the study. In addition, the literature review was performed about 6 previously reported AOSD patients who had undergone PET. We evaluate FDG uptake for characteristic findings in patients with AOSD. Results: FDG accumulation was positive mainly in the bone marrow (70%), spleen (80%), lymph nodes (77.8%), and joints (50%). In addition, FDG uptake was positive in the pericardium, pleura, salivary glands, eyelids, muscle, and major blood vessels. 8 patients underwent FDG PET/CT for evaluating the efficacy of treatment. The follow-up PET showed diminished FDG accumulation in the bone marrow, spleen, lymph nodes, with SUVmax being significantly reduced from 4.03⫾0.95 to 2.20⫾0.75 (P⫽0.04), from 4.15⫾1.10 to 2.55⫾1.13(P⫽0.04), and from 5.47⫾5.19 to 2.10⫾1.91 (P⫽0.11), respectively. No significant correlation was found between max SUVs in each site and the laboratory data; the only significant correlation was between LDH and the spleen SUV. Conclusion: Our study suggests that 18F-FDG PET may play a potential role in diagnosing or monitoring of disease activity marker in patients with AOSD. Additionally, we suspect that future PET/CT imaging will reveal that AOSD is a disease involving lesions with a little clinical description.

Sunday, October 27

Conclusion: CEU have a good concordance with PET. These results outline the potential role of CEU to be used in monitoring disease activity and response to treatment in LVV

Results: Examination by the 7Tesla MRI revealed cerebral microvascular lesions(under 100 micrometer) in 6 of 7 NPSLE patient. These findings were not detected in any of the patients with non-NP SLE. These findings were not detected by 1.5 Tesla MRI. There were not findings at all in 1.5Tesla MRI, and the case that included findings was found in only 7TeslaMRI. The cerebral microvascular lesions were found around the central sulcus especially in T2-weighted image. These MRI findings disappeared after central nervous symptom improvement in two of the three NPSLE patient that reexamined three months later. However, we did not find a direct association between the MRI findings and the neuropsychology dysfunction.

References: 1. Magnoni M. et al.; Assessment of Takayasu arteritis activity by carotid contrast-enhanced ultrasound. Circ Cardiovasc Imaging. 2011 Mar;4(2):e1–2 2. Giordana P et al.; Contrast-enhanced ultrasound of carotid artery wall in Takayasu disease: first evidence of application in diagnosis and monitoring of response to treatment. Circulation. 2011;124:245–7 Disclosure: G. Germano`, None; P. Macchioni, None; N. Possemato, None; C. Salvarani, None.

220 Asymptomatic Myocardial Ischemic Disease In Takayasu’s Arteritis: Detection By Magnetic Resonance Imaging. Cloe´ Comarmond1, Philippe Cluzel2, Dan Toledano3, Nathalie Costedoat-Chalumeau4, Richard Isnard5, Fabien Koskas6, Patrice Cacoub Sr.7 and David Saadoun8. 1Hoˆpital Pitie´ Salpe´trie`re, Paris, France, 2Pitie´-Salpe´trie`re, Paris, France, 3Hoˆpital Pitie´-Salpeˆtrie`re, Paris, France, 4Hopital Cochin, Paris, France, 5CHU Pitie´-Salpeˆtrie`re, 47-83 Boulevard de l’hoˆpital, 75651 Paris Cedex 13, Paris, France, Paris, France, 6 Assistance Publique-Hoˆpitaux de Paris, Hopital Pitie´-Salpe´trie`re, Paris, France, 7 CHU Pitie´-Salpeˆtrie`re, Paris, France, 8Pitie´-Salpeˆtrie`re Hospital, APHP, Paris, France

Image: 41 years old female She had both legs motor nerve disorder.

Background/Purpose: Takayasu’s arteritis (TA) may affected myocardium and caused coronary stenosis. The aim of this study was to assess the prevalence and pattern of myocardial disease in patients with TA, using late gadolinium enhancement (LGE) of cardiac magnetic resonance imaging (CMRI). Methods: Twenty-seven consecutive patients with TA and 80 age and sex matched controls without known cardiovascular disease underwent CMRI. The prevalence of myocardial ischemic disease, as revealed by LGE, was compared between patients with TA and controls, and factors associated with myocardial disease were identified in patients with TA. Results: Myocardial ischemic disease, as characterized by LGE on CMRI, was present in 7 (25.9%) of 27 patients with TA, and imaging with LGE showed a typical pattern of myocardial infarction (MI) in 6 patients (22.2%). Although both patients with TA and control subjects shared a similar risk of cardiovascular events, the prevalence of myocardial ischemia was more than 5 times higher in patients with TA (P ⫽ 0.002 versus controls). No association was found between myocardial disease in patients with TA and cardiovascular atherosclerotic risk factors. The presence of myocardial scarring tended to be more closely associated with specific features of TA, such as renovascular hypertension, older age at the onset of TA symptoms, male gender, aneurysmal dilatation, and numano type V. Conclusion: The finding of a significant and unexpectedly high prevalence of occult myocardial scarring in patients with TA indicates the usefulness of CMRI with LGE for the identification of occult myocardial disease in such patients.

The cerebral microvascular lesions were found around the central sulcus in T2-weighted image. It could be find only by using 7TeslaMRI. Conclusion: Cerebral microvascular lesions could be detected in patients with NPSLE by 7 Tesla MRI, suggesting that 7 Tesla MRI is a useful tool for diagnosing the cerebral lesions of NPSLE. It will be necessary to elucidate the association between cerebral microvascular lesions and neuropsychological dysfunction. Disclosure: N. Sasaki, JSPS KAKENHI, 2; O. Murata, None; Y. Ninomiya, None; Y. Oikawa, None; H. Kobayashi, None; K. Yamauchi, None; M. Sasaki, None; T. Sawai, None; Y. Nakamura, None.

222 Comparison Of Whole Body Versus Targeted Magnetic Resonance Imaging For Assessing Disease Activity and Damage In Idiopathic Inflammatory Myopathies. Adam Schiffenbauer1, Evrim Turkbey2, Lisa G. Rider1, Suvimol Hill2, Irene Z. Whitt3, Songtao Liu2, David A. Bluemke2 and Frederick W. Miller1. 1NIEHS, NIH, Bethesda, MD, 2NIH, Bethesda, MD, 3 Duke, Durham, NC Background/Purpose: In the idiopathic inflammatory myopathies (IIM) MRI imaging is traditionally focused on the proximal upper or lower extremities. Whole body MRI (WBMRI) is a technique where the entire body is scanned into one complete study. The purpose of this study was to assess what additional information is gained from WBMRI and to determine its relationship to clinical parameters. Methods: 24 patients with a variety of IIM, including juvenile dermatomyositis (JDM, n⫽7), dermatomyositis (DM, n⫽6), polymyositis (PM, n⫽4), sporadic inclusion body myositis(s-IBM, n⫽3), and familial inclusion body myositis (f-IBM, n⫽4) underwent WBMRI. Using a Siemens 3 Tesla MRI scanner the patients were scanned in a head to toe manner and had T1, T2, and STIR images obtained. These images were then scored in a blinded manner across 34 muscle groups and each group was scored on a scale of 0 to 3 for T2/STIR intensity, T2/STIR area of involvement, and fatty infiltration on T1. WBMRI and thigh MRI (THMRI) scores were evaluated for correlation with a variety of measures of disease activity and damage. Results: 3 out of 24 patients (12.5%) who were scored to have activity (at least a 1 for STIR intensity) on WBMRI had no activity present on THMRI. Similarly, 4 of 24 patients (16.67%) had fatty replacement on WBMRI, but no fatty replacement on THMRI. Patients had MRI STIR abnormalities by WBMRI in many areas that would not be scanned in a MRI of the thigh or proximal arms, including the paraspinal muscles (20% of patients) and the chest (16% of patients) as well asforearms and distal extremities (Table).

Disclosure: C. Comarmond, None; P. Cluzel, None; D. Toledano, None; N. Costedoat-Chalumeau, None; R. Isnard, None; F. Koskas, None; P. Cacoub Sr., None; D. Saadoun, None.

221 Detection Of Cerebral Microvascular Lesions In Patients With Acute Phase Neuropsychiatric Systemic Lupus Erythematosus By 7Tesla MRI. Nobuhito Sasaki, Okinori Murata, Yukari Ninomiya, Yuka Oikawa, Hitoshi Kobayashi, Kohei Yamauchi, Makoto Sasaki, Takashi Sawai and Yutaka Nakamura. Iwate Medical University School of Medicine, Morioka, Japan Background/Purpose: The cerebral microvascular lesions of patients with neuropsychiatric systemic lupus erythematosus (NPSLE) have not been fully elucidated. The 7Tesla MR scanner has high image resolution and can be expected to detect cerebral microvascular lesions that have never before been visible. We examined the cerebral microvascular lesions of NPSLE using a 7Tesla MR scanner. Methods: We studied 7 acute phase-NPSLE patient and 11 non-NPSLE. MRI was performed using a 1.5Tesla MR scanner and 7Tesla MR scanner (T1-weighted imaging, T2-weighted imaging, diffusion-weighted imaging, magnetic resonance angiography). Two radiologists performed the interpretation without clinical information. Diagnosis of was determined by a neurologist, psychiatrist and rheumatologist.

S90

Percentage positive on Right 7.69 30.77 20.83 37.5

Percentage positive on Left 7.14 21.43 45.83 37.5

Sunday, October 27

Region Anterior forearm Posterior forearm Distal lower extremity anterior Distal lower extremity posterior

Among all IIM patients, physician global disease damage on a visual analogue scale (VAS) and a Likert scale correlated with imaged fatty infiltration (p⬍0.0001). R squared for comparison of WBMRI and THMRI fatty replacement with physician global disease damage were 0.796 and 0.695 respectively. Also amongst all patients serum levels of creatine kinase (CK), aspartate aminotransferase (AST), alanine aminotransferase (ALT), and lactate dehydrogenase (LDH) were found to correlate significantly with WBMRI STIR intensity and involvement scores, as well as with THMRI STIR intensity scores (r2⫽0.23–0.49, p⬍0.05 for all). In adult and juvenile DM and PM patients, the physician global activity by VAS correlated with WBMRI and THMRI intensity and involvement scores (r2⫽0.28–0.33, p⬍0.05). WBMRI and THMRI intensity and involvement scores also correlated significantly with CK, AST, ALT, and LDH (r2⫽0.42–0.81, p⬍ 0.05) in this population. Conclusion: WBMRI was able to detect areas of muscle edema and fatty infiltration in IIM patients that would not be detected on traditional focal MRI of the thighs or arms. WBMRI also correlated well with physician assessment of disease activity and damage as well as with multiple laboratory markers of disease activity. These findings suggest utility of WBMRI in assessing patients across the IIM spectrum. Disclosure: A. Schiffenbauer, None; E. Turkbey, None; L. G. Rider, None; S. Hill, None; I. Z. Whitt, None; S. Liu, None; D. A. Bluemke, None; F. W. Miller, None.

Conclusion: Twenty two percent of patients with scleroderma had positive subcutaneous power doppler signal above the dorsal aspect of the MCPs. Patients with this phenomenon had longer disease duration and less lung involvement with moderately good predictive value (ROC AUC- 77). There was not increased synovitis or elevated ESR in this group. We therefore assume that this is not an inflammatory phenomenon. We speculate that SCPD may be related to increased blood flow to the skin during the resolution phase of the skin. Further studies are needed to understand the underling mechanism.

223 Clinical Predictors For Subcutaneous Power Doppler Signals Detected By Ultrasound in Hands Of Scleroderma Patients. Anat Scheiman-Elazary1, Ami Ben-Artzi2, V.K. Ranganath2, Nabeel Borazan1, Philip J. Clements2, Suzanne Kafaja2 and Daniel Furst2. 1Rheumatology UCLA, Los Angeles, CA, 2David Geffen School of Medicine, UCLA, Los Angeles, CA

Disclosure: A. Scheiman-Elazary, None; A. Ben-Artzi, None; V. K. Ranganath, None; N. Borazan, None; P. J. Clements, None; S. Kafaja, None; D. Furst, AbbVie, 2, Actelion Pharmaceuticals US, 2, Amgen, 2, BMS, 2, Gilead, 2, GSK, 2, NIH, 2, Novartis Pharmaceutical Corporation, 2, Pfizer Inc, 2, Roche/Genentech, 2, UCB, 2, Janssen Pharmaceutica Product, L.P., 5.

Background/Purpose: Skin thickening in scleroderma was demonstrated previously in US studies, although subcutaneous tissue was not yet evaluated. We evaluated scleroderma patients for subcutaneous Power Doppler signal (SCPD) on the dorsal aspect of the MCPs of the dominant hand. Out of 74 patients, 17 (22.9%) had a positive signal. Our aim was to compare SCPD positive (N⫽ 17, 22.9%) versus SCPD negative (N⫽57, 77.1%) patients and to look for clinical and laboratory predictors for SCPD. Methods: Seventeen SCPD positive patients were compared with 57 SCPD negative patients. US exam was performed using GE logic E9 scanner with a linear array (5–16 MHz) for PD in subcutaneous tissue. SCPD was defined as a positive PD signal in the subcutaneous tissue on the dorsal aspect of either MCPs 2–5 in both longitudinal and short views that was not involving the tendon. At least one MCP had to be involved. A binary grading was used to score SCPD (0–1). Patients were divided into 2 groups, with or without SCPD. GS and PD examinations of 13 joints (wrist, MCP 2–5, PIPs 2–5) and tendons were done. A step wise linear regression model was used to identify predictors for SCPD. The following parameters were included: age, gender, disease duration, disease type (limited or diffuse), modified Rodnan skin score (MRSS), lung involvement (symptoms, pulmonary function test and high resolution CT), pulmonary hypertension (echo), ESR, CRP, ulcers, calcinosis, sclerodactaly and contractures. Results: After univariate analysis with P⬍0.2 cut-off, MRSS, disease duration and lung involvement were included. MRSS (OR⫽0.93, p⫽0.14) and lung disease(OR⫽0.28, p⫽0.047) decreased the likelihood for SCPD while longer disease duration increased its likelihood (OR⫽1.117, p⫽0.036). The final model included only lung and disease duration with a ROC curve, AUC of 0.77 (figure 1). There was no statistical difference in mean joint GS and PD between both groups (mean joint GS 0.34 and 0.31 in the negative and positive SCPD patients respectively, mean joint PD 0.08 and 0.04 in the negative and positive SCPD patients respectively). Mean ESR was 21 and 18 mm/h in the positive and negative SCPD patients respectively (P value 0.94).

224 Ultrasound Evaluation Of Joints, Tendons and Subcutaneous Tissue Of Hands In 74 Scleroderma Patients. Anat Scheiman-Elazary1, Ami BenArtzi2, Suzanne Kafaja2, V Ranganath2, Nabeel Borazan1, Philip J. Clements2 and Daniel Furst2. 1Rheumatology UCLA, Los Angeles, CA, 2David Geffen School of Medicine, UCLA, Los Angeles, CA Background/Purpose: A previous ultrasound study of the hands in scleroderma (SSC) and rheumatoid arthritis (RA), found mild inflammatory changes in the tendons and joints of the SSC patients. Our aim was to evaluate joints, tendons and subcutaneous tissue in SSC patients compared to RA patients and healthy controls (HC). Methods: A total of 74 SSC patients and 14 HC were examined in this cross sectional study and compared to a database of 32 RA patients. The patients underwent clinical and laboratory assessment. Blinded US exam of 13 joints (Wrist, MCPs 2–5, PIPs 2–5,DIPs 2–5) was performed, using GE logic E9 scanner with a linear array (5–16 MHz). Gray scale (GS) (0–1) and Power Doppler (PD) (0–3) scoring were used for tendons. GS and PD (0–3) scoring were used to evaluate joints. In 13 patients, we were unable to examine all 13 joints due to contractures. We therefore calculated separate GS and PD scores for joints and tendons by summing up scores for individual joints and dividing it by the number of joints examined. Subcutaneous tissue over the dorsal aspect of the MCPs was evaluated by PD (SCPD). Tuft resorption was examined by evaluating shortening of the distal second to fifth phalanges on the palmar and dorsal aspects. RA patients only had data on 5 joints (wrist, MCPs 2–3 and PIPs 2–3) and this subset was compared across the groups. Results: Mean age was similar between all 3 groups. Percentage of females was 82, 90, and 71 for SSC, RA and HC respectively. Mean disease duration was 6.9 and 5.7 in the SSC and RA patients (Table 1). Mean DAS28

S91

Sunday, October 27

and CDAI in the RA patients was 5.07 and 22 respectively. For SSC patients 13.5% had tuft resorptions compared to none of the RA patients and HC. Mean GS and PD score for joints and tendons of the 13 joints exam were significantly higher in the SSC patients then HC (mean joint GS 0.33 and 0.08, mean join PD 0.07 and 0.01, mean tendon GS 0.08 and 0, mean tendon PD 0.09 and 0 for the SSC and HC respectively, p⬍0.05 for all). For the 5 joint exam, mean GS and PD for tendons and joints was significantly higher in the SSC versus HC as well as RA patients except PD for tendons which was not significant in the SSC compared to HC. RA patients had a higher frequency of ⫹2 or ⫹3 PD in joints and tendons while HC had none (Table 1). Twenty one percent of SSC patients with swollen joint count of zero, had ⫹2 or ⫹3 PD signals in joints or tendons. Subcutaneous PD signals were found nearly exclusively in SSC patients.

SSC (N⫽74) RA (N⫽32) HC (N⫽14) Total P value SSC-RA P value SSC-HV P value

Age

Disease duration %F (y)

Mean ESR

53.46 53.75 52.2 0.97

82 90 71 0.26

18.6 0.42 0.09 39.6 1 0.52 NA 0.08 0.004 0.0013 ⬍0.01 ⬍0.01

6.9 5.7 NA 0.012

5 Joint GS

5 Joint PD

5 Tendon GS

5 Tendon PD

% patients with ⴙ2 or ⴙ3 join PD

% Patients with ⴙ2 or ⴙ3 Tendon PD

% SCPD (N)

0.11 0.33 0.01 ⬍0.01

0.1 0.41 0 ⬍0.01

16 93 0 ⬍0.01

24 62 0 ⬍0.01

22 (17) 0.03 (1) 0 ⬍0.01

NA

NA

NA

NA

⬍0.01 ⬍0.01

⬍0.01

⬍0.01

⬍0.01

0.0003

0.0113

NA

NA

NA

NA

⬍0.01 0.0328

0.0133

0.0662

0.208

0.0642

0.0459

The angiography assessment suggested the presence of periaortic fibrosis with “coated aspect” in two cases. In these, epiaortic vessels were surrounded by irregular, non stenosing fibrosis. Conclusion: Our study confirms the frequent cardiovascular involvement in ECD. A systematic cardiac and aortic evaluation by MRI would be indicated in patients with ECD. Disclosure: E. Rossi, None; L. Buttarelli, None; A. Vaglio, None; D. Gianfreda, None; C. Martini, None; M. De Filippo, None.

226 Omeract Definitions For The Sonographic Appearance Of The Normal Pediatric Joint. Johannes Roth1, Sandrine Jousse-Joulin2, Silvia MagniManzoni3, Ana Narrodi4, Nikolay Tzaribachev5, Annamaria Iagnocco6, Esperanza Naredo7, Maria-Antonietta D’Agostino8 and Paz Collado9. 1University of Ottawa, Ottawa, ON, 2C.H.U. La Cavale Blanche, Brest, France, 3 Ospedale Pediatrico Bambino Gesu`, Rome, Italy, 4Rheumatology, Madrid, Spain, 5Center for Rheumatic Diseases, Bad Bramstedt, Germany, 6University La Sapienza, Rome, Italy, 7Hospital General Universitario Gregorio Maranõń, Madrid, Spain, 8Ambroise Pare´ Hospital, and Versailles Saint Quentin en Yvelines University, Boulogne-Billancourt, France, 9Severo Ochoa University Hospital, Madrid, Spain Background/Purpose: Musculoskeletal ultrasound (MSUS) has significant potential in the assessment of disease activity and early structural damage in childhood arthritides. In order to assess pathology, the ultrasonographic characteristics of the normal pediatric joint need to be defined first. The aims of this study were to develop definitions for the various components of the pediatric joint that can be evaluated by MSUS through a consensus process and to validate these definitions in their practical application. Methods: In a first step, members of the pediatric group of the OMERACT Ultrasound Task Force developed definitions of the ultrasonographic appearance for various components of the pediatric joint through a Delphi process. In a second step, standardized scans of the knee and ankle joint were acquired on 4 healthy volunteers aged 12, 10 and 4 years old (2 volunteers) by seven members of the group on the same day. Subsequently, for each of the stored images, the applicability (validity) of each of the definitions was scored by all 7 participants on a 1–5 Likert scale. A minimum of 80 % grade 4 or 5 agreement across all images was required in order to validate the definition. Results: Five definitions were developed through the Delphi process addressing articular bone, cartilage, joint capsule, the epiphyseal ossification centre and the synovial membrane. A total of 224 images were acquired by the seven participants. 172 of these images were selected for analysis and on each of them the five definitions were rated by all 7 participants. In the first practical application on stored images the minimum of 80 % agreement was not met for either definition. After further modification of the definitions 95%, 81%, 86%, 97% and 91% agreement were reached respectively. Conclusion: In this study, definitions of the ultrasonographic appearance for the various components of the normal pediatric joint were successfully developed through a Delphi process and validated in a practical exercise. These results represent the essential first step in order to develop definitions for pathology and support the standardization of the ultrasonographic assessment of the pediatric joint. This carries relevant implications in defining the role of MSUS both in routine clinical assessment and as an outcome measure in research.

HC-Healthy controls, SSC-scleroderma, SCPD- subcutaneous power Doppler, PD-Power Doppler, GS-gray scale, F-females, NA-not applicable

Conclusion: Joint and tendon inflammation in scleroderma is increased compared to HC but milder then in RA. US detected synovitis more frequently than physical examination in the SSC patients. Since it is sometimes difficult to estimate synovitis on physical examination in SSC patients due to tight skin and contractures, US would be useful in the assessment of synovitis in these patients. Ref. 1. Elhai M et al. Arthritis Care Res 2012. Disclosure: A. Scheiman-Elazary, None; A. Ben-Artzi, None; S. Kafaja, None; V. Ranganath, None; N. Borazan, None; P. J. Clements, None; D. Furst, AbbVie, 2, Actelion Pharmaceuticals US, 2, Amgen, 2, BMS, 2, Gilead, 2, GSK, 2, NIH, 2, Novartis Pharmaceutical Corporation, 2, Pfizer Inc, 2, Roche/Genentech, 2, UCB, 2, Janssen Pharmaceutica Product, L.P., 5.

225 Cardiovascular Involvement In Erdheim-Chester Disease: A Magnetic Resonance Imaging Study On Seven Patients. Enrica Rossi, Lorenzo Buttarelli, Augusto Vaglio, Davide Gianfreda, Chiara Martini and Massimo De Filippo. University of Parma, Parma, Italy Background/Purpose: Erdheim-Chester Disease (ECD) is a rare nonLangherans form of histiocytosis, characterized by xanthomatous or xanthogranulomatous infiltration of tissues by foamy histiocytes, surrounded by fibrosis. ECD is a multisystemic disease and its clinical course depends mostly on cardiovascular manifestations, that are responsible for poor prognosis and death. In order to assess the cardiovascular extent of the disease, we studied 7 consecutive cases with biopsy-proven ECD by magnetic resonance imaging (MRI). Methods: The patients underwent cardiac and thoracic aorta MRI. Images were acquired with a 1,5T RM, using T2W with and without fat suppression, long and short axis cine B-TFE and late gadolinium enhancement sequences evaluated by dedicated software. Results: Four patients (57%) showed abnormal heart imaging. Four patients had myocardium involvement with typical pseudo-mass aspects in the anterior and lateral walls of right atrium. In three cases it extended to the right atrioventricular sulcus, sheathing the right coronary artery; only in one case both coronary arteries were surrounded by periarterial infiltration. None of the cases showed signs of coronary stenosis. One case also presented atrial septal thickening. In three cases imaging studies recognized pericardial effusion; just in one case it was massive (50mm), even if without signs of cardiac tamponade. The pericardium facing the right ventriculum was thickened in two cases. Imaging showed a mild cardiomegaly in one patient, but no atrial enlargement was detected. In no case we recognized cardiac insufficiency or cardiac signs of systemic hypertension. No patients had valvular defects.

Disclosure: J. Roth, None; S. Jousse-Joulin, None; S. Magni-Manzoni, None; A. Narrodi, None; N. Tzaribachev, None; A. Iagnocco, None; E. Naredo, None; M. A. D’Agostino, None; P. Collado, None.

227 Quantitative Image Analysis Of Articular Involvement In Blau Syndrome By Radiographic Calpal Length and Ultrasound Assessment. Tsuyoshi Yamatou1, Tomohiro Kubota1, Harumi Akaike1, Yuichi Yamasaki1, Yukiko Nonaka1, Yasuhito Nerome1, Tomoko Takezaki1, Hiroyuki Imanaka1, Kei Ikeda2, Naotomo Kambe3, Syuji Takei1 and Tomokazu Nagakura4. 1Kagoshima University, Kagoshima, Japan, 2Chiba University Hospital, Chiba, Japan, 3Chiba University Graduate School of Medicine, Chiba, Japan, 4House of Meguminoseibo, Usuki, Japan Background/Purpose: Blau syndrome (Blau) is a rare auto-inflammatory disease, and it has now been shown to be caused by NOD2/CARD15 gene

S92

mutations. Clinical features of Blau consist of granulomatous inflammation in the joints, eyes, and skin. However, most of the patients were initially miss-diagnosed as JIA because the symmetrical polyarthritis occurred in early phase of the disease, and “boutonniere” deformities-like fingers appeared during the disease course. However, the articular involvement in Blau has been considered as non-erosive. Therefore, we evaluate the joint damage in Blau patients by measuring calpal length (CL), an objective tool for evaluating cartilage damage in polyariticular JIA (poJIA)1–2. Methods: A total of 6 Blau patients were followed up their CL for mean 9.2 years. Their disease duration at the first CL measurement was 5.2 years. Since biologic agents were used in 5 of 6 of the Blau patients, changes in CL of 46 poJIA patients treated with biologics were also followed up. CL was analyzed by standard deviation (SD) calculated by Poznanski’s formula established from healthy children3). In addition, ultrasound examination was carried out in 4 Blau patients at the last observation. A total of 3 joints and 4 tenosynovial sites in the wrist were scanned and the ultrasound images of gray-scale (GS) or Power Doppler (PD) findings were graded from 0 (normal) to 3 (severe). Results: Though all Blau patients had comptodactyly with “boutonniere” deformities like fingers, CL(SD) maintained normal range (within ⫺2SD to ⫹2SD) during the observation period regardless of biologic therapy (Figure 1). On the contrary, CL(SD) in poJIA decreased with disease course especially before initiating biologic therapy, and there were negative correlation between the CL(SD) and the disease duration in both RF positive (p⫽0.0018) and RF negative (p⫽0.008) poJIA patients.

228

Background/Purpose: Juvenile Dermatomyositis (JDM) is a systemic vasculopathy, primarily involving the micro vasculature. Soft tissue calcification occurs in 15–30% of cases, and is associated with increased morbidity and mortality. The calcifications develop in conjunction with chronic inflammation, and are often located at pressure points of daily trauma (elbows, buttocks, hands). In contrast, calcifications of aortic heart valves in non-JDM patients are more common and are associated with interrelated factors such as hypertension, hyperlipidemia, obesity, smoking and diabetes. The purpose of this investigation was to compare the composition and structure of the calcific deposits obtained from children with JDM with those in removed aortic valves. Methods: Five children (2 boys, 3 girls) previously reported (Arthritis Rheum 54:10 2006, 3345–3350) with definite JDM, had a mean age of JDM onset, 3.3 years ⫾ 1.9 years. They had sustained persistent cutaneous inflammation for 81.3 ⫾ 58.7 months, and donated calcium deposits, after obtaining informed consent. The calcifications, removed by surgery, were a focus of constant pain and contributed to decreased range of motion. Fresh, calcified aortic heart valves were obtained at cardiovascular surgery, after obtaining informed consent, from 5 White adult donors without JDM (age range 43–85, mean age⫽ 69.2 years), two were women, with severe aortic stenosis. Four of the donors were hypertensive, one had diabetes and one smoked. Fourier transform infrared radiography images (FTIRI) were obtained of the calcifications, which had been fixed in 90% ethanol, embedded in polymethyl methacrylate (PMMA) and sectioned. Transmission Electron Microscopy was performed. Results: In JDM, electron microscopy documented extensive macrophage and giant cell infiltration; the mineral definitely formed crystals, even within the cytoplasm of the affected cells, whereas in plaque the calcium appeared to be associated with large lipid deposits were not crystalline (verified by EDAX microanalysis). Average FTIRI Parameters for the 5 Calcified Aortic Valves Min/Mat CO3/min XLR XST Mean 12 0.0104 4.20 1.10 SD 1.7 0.0013 0.90 0.05 Average FTIRI Parameters for the 5 JDM Deposits Mean 25 0.014 5.0 1.12 SD 14 0.013 1.8 0.06 Average FTIRI Parameters for 2 Pediatric Iliac Crest Biopsies Cortical Bone Mean 4.57 0.0065 3.17 1.25 SD 0.64 0.0004 0.02 0.02 Average FTIRI Parameters for 2 Pediatric Iliac Crest Biopsies Cancellous Bone Mean 4.53 0.0065 3.18 1.20 SD 0.20 0.0003 0.08 0.001

Ultrasound examination revealed that tenosynovitis was more frequent and severe than intra-articular synovitis in the 8 wrists of 4 Blau patients; Grade 1–3 findings were found in 62.5% by PD and 31.2% by GS in scanned tenosynovium sites, while it was 29.2% by PD and 4.2% by GS in intra-articular synovium sites (tendon vs intra-articular sites: p⫽0.0169 by PD, p⫽0.0161 by GS). Conclusion: Comptodactyly with “boutonniere” deformities-like findings, a characteristic feature of Blau, may be caused by tenosynovitis but not by cartilage destruction in finger joints. The finding indicates the usefulness of CL and ultrasound examination in differentiating Blau from JIA, and is also suggestive that tenosynovitis may be more predominant than intra-articular synovitis in Blau syndrome.

On FTIRI, the mineral/matrix ratio of the calcified aortic valves was greater than in human cortical or cancellous bone, but less than those in JDM deposits; JDM calcifications were much more punctate. The ratios of carbonate to phosphate and the carbonate/matrix ratios were not different from those of bone, nor were the values for collagen maturity and crystallinity. In JDM calcifications, but not calcified aortic valves, intracellular crystals were present. Conclusion: The composition and structure of calcifications occurring in aortic valves differ from those that develop in the soft tissue of children with JDM, suggesting that the deposition mechanism differs as well.

1) Ravelli A, et al. J Pediatr 1998; 133:262–5. 2) Kubota T et al. #1163, 2012 ACR Annual Meeting. 3) Poznanski AK et al. Radiology 1978;129:661–8. Disclosure: T. Yamatou, None; T. Kubota, None; H. Akaike, None; Y. Yamasaki, None; Y. Nonaka, None; Y. Nerome, None; T. Takezaki, None; H. Imanaka, None; K. Ikeda, None; N. Kambe, None; S. Takei, Chugai, Eisai, Takeda, and Bristol-Myers, 2, Chugai, Eisai, Takeda, Abbvie, Astellas, Teijin, Novartis, Pfizer, and Asahi Kasei, 8; T. Nagakura, None.

Disclosure: L. M. Pachman, None; G. A. Morgan, None; P. M. McCarthy, None; A. Huskin, None; S. C. Malaisrie, None; L. Spevak, None; S. Doty, None; A. Boskey, None.

S93

Sunday, October 27

The Composition and Structure Of Calcifications In Juvenile Dermatomyositis Differs From Calcified Aortic Valves Removed From Adults Without JDM. Lauren M. Pachman1, Gabrielle A. Morgan1, Patrick M. McCarthy2, Anna Huskin2, S. Chris Malaisrie2, Lyudmila Spevak3, Stephen Doty3 and Adele Boskey3. 1Ann & Robert H. Lurie Children’s Hospital of Chicago Research Center, Chicago, IL, 2Northwestern University Feinberg School of Medicine Bluhm Cardiovascular Institute, Chicago, IL, 3Hospital for Special Surgery, New York, NY

Stage 3. Range of movement was assessed and the Shoulder Pain and Disability Index (SPADI) completed. An MRI and X-ray were undertaken of the affected or previously affected shoulder, or matched side (in those with no current or previous shoulder pain). The X-rays and MRIs were read independently by two radiologists blinded to the participant group and each other. All participants provided written informed consent. Results: Overall, 30 participants took part in the study. The mean range of flexion, abduction and external rotation were all lower for the pain group compared to the other two groups. The mean SPADI percentage score for the pain group was 41.80 (SD 20.10, range 18–74) and the mean percentage function score was 31.88 (SD 20.41, range 8.75–60). On X-ray there were few differences between each group. The MRI findings are presented in Table 1. Subacromial bursitis was common (90%) in each group, 90% of participants also demonstrated a degree of acromioclavicular degeneration. There was a slightly higher number of supraspinatus tendinosis/tears and involvement of the LHB in the current pain group.

Sunday, October 27

229 Positron Emission Tomography Assessment Of Children With SystemicOnset Juvenile Idiopathic Arthritis. Taichi Kanetaka, Tomo Nozawa, Kenichi Nishimura, Masako Kikuchi, Tomomi Sato, Nodoka Sakurai, Ryoki Hara, Kazuko Yamazaki and Shumpei Yokota. Yokohama City University School of Medicine, Yokohama, Japan Background/Purpose: Systemic-onset juvenile idiopathic arthritis (sJIA) is a systemic inflammatory disorder manifesting spiking fever, rheumatoid rash, and arthritis. The onset is nonspecific, and may be triggered by infectious agents. A body of evidence indicates involvement of cells and inflammatory cytokines produced in innate immune response in the induction phase of the disease, and then augmented in an auto-inflammatory fashion, suggesting that s-JIA may be one of the auto-inflammatory diseases. However, the causative agents and the responsible organs for exaggerated innate immune responses are still to be elucidated. The 18F-FDG-Positron Emission Tomography (PET) is a hybrid imaging technique displaying the sites of high metabolic turnover of both physiologic and pathologic origin and visualizes infection focus and inflammatory lesions as well as malignancy. Application of 18F-FDG-PET to patients with s-JIA will possibly localize inflammatory lesions of this disease. To examine and delineate inflammatory focus in patients with s-JIA, 18 F-FDG-PET was applied to patients with s-JIA and polyarticular (p)-JIA, and the images of these patients were compared. Methods: Sixty eight children (57 with s-JIA and 11 with p-JIA) were included in this study who needed the diagnosis work-up of fever-ofunknown origin. The diagnosis of s-JIA and p-JIA was done to meet the ILAR criteria (2004). After fasting for at least 6 hours, whole body PET-scans were acquired 60 minutes after intravenous injection of 3–5 MBq/kg 18 F-FDG. The interpretation of 18F-FDG uptake was based on visual characteristics. Results: 18F-FDG uptake in shoulder joint areas was found in 42 out of 57 patients with s-JIA (74%) and in 5 out of 11 patients with p-JIA (45%). 18 F-FDG uptake in knee joint areas was demonstrated in 19 patients with s-JIA (33%) and 7 patients with p-JIA (64%). Although in patients with s-JIA 18 F-FDG uptake was found in shoulder areas, physical examination failed to reveal inflammatory symptoms/signs around shoulders. Additionally, two types of PET images were outstanding in s-JIA; one was 18F-FDG uptake in red bone marrow such as the spine, femur heads, and pelvis as well as spleen (11 cases), and other type was the uptake in the major joints such as hips, elbows, wrists, knees, and ankles (7 cases). The former findings were correlated with elevated levels of inflammatory markers (WBC, ESR, and ferritin) while the latter were with significantly increased levels of MMP-3 (P⬍0.05). Conclusion: There was a noticeable accumulation of 18F-FDG uptake in bone marrow of s-JIA patients which may indicate the inflammatory focus of this disease and play an important role in the pathogenic basis of arthritides and systemic inflammation of s-JIA.

Table 1. MRI findings

Supraspinatus Normal/Equivocal Tendinosis Partial thickness tear with/without tendinosis Full thickness tear with/without tendinosis Infraspinatus Normal/Equivocal Tendinosis Partial thickness tear with/without tendinosis

No shoulder pain n %

Previous shoulder pain n %

Current shoulder pain n %

3 1 4 2

30.0 10.0 40.0 20.0

3 2 3 2

30.0 20.0 30.0 20.0

2 6 2

20.0 60.0 20.0

7 3 -

70.0 30.0 -

4 5 1

40.0 50.0 10.0

7 1 2

70.0 10.0 20.0

4 2 4

40.0 20.0 40.0

3 5 2

30.0 50.0 20.0

10

100.0

10

100.0

8 1 1 -

80.0 10.0 10.0 -

6 3 1 -

60.0 30.0 10.0 -

2 3 4 1

20.0 30.0 40.0 10.0

2 8 -

20.0 80.0 -

3 7 -

30.0 70.0 -

4 5 1

40.0 50.0 10.0

Subscapularis Normal/Equivocal Tendinosis Partial thickness tear with/without tendinosis Teres Minor Normal/Equivocal Long head of biceps Normal/Equivocal Tendinosis Partial thickness tear with/without tendinosis Full thickness tear GHJ cartilage degeneration Normal/Equivocal Mild Moderate

5 1 4

50.0 10.0 40.0

10 100.0

Conclusion: Shoulder pathology is present on imaging in people with shoulder pain, those who have a history of shoulder pain, and those who have never had shoulder pain. Clinical symptoms do not necessarily match the radiological findings. The value of MRI as a clinically useful diagnostic investigation for shoulder pain is questionable.

Disclosure: T. Kanetaka, None; T. Nozawa, None; K. Nishimura, None; M. Kikuchi, None; T. Sato, None; N. Sakurai, None; R. Hara, None; K. Yamazaki, None; S. Yokota, None.

230 An Examination Of Shoulder Pain Using Magnetic Resonance Imaging In Older People. Tiffany K. Gill1, E. Michael Shanahan2, Dale Allison3, Daniel Alcorn3 and Catherine L. Hill4. 1University of Adelaide, Adelaide, South Australia, Australia, 2Flinders University, Bedford Park, South Australia, Australia, 3Bensons Radiology, Adelaide, Australia, 4The Queen Elizabeth Hospital, Woodville, Australia

Disclosure: T. K. Gill, None; E. M. Shanahan, None; D. Allison, None; D. Alcorn, None; C. L. Hill, None.

231 Variable Imaging Characteristics Identified By Point-Of-Care Ultrasound For Greater Trochanteric Pain Syndrome. Minna J. Kohler1, Naina Rastalsky1 and Liana Fraenkel2. 1Massachusetts General Hospital/Harvard Medical School, Boston, MA, 2Yale University School of Medicine, Veterans Affairs Connecticut Healthcare System, New Haven, CT

Background/Purpose: Shoulder pain is a common problem in the general population and can cause significant difficulties with activities of daily living. Magnetic resonance imaging (MRI) is often used in the assessment of shoulder pain. This pilot study aimed to determine if the pathology demonstrated on shoulder MRI correlates with the pain reported by study participants. Methods:Participants for this pilot study were obtained from the North West Adelaide Study (NWAHS), a longitudinal cohort study of 4056 randomly selected adults aged 18 years and over at the time of recruitment from the northern and western regions of Adelaide, Australia. Thirty participants aged 55 to 74 years were recruited from the cohort and allocated to one of three groups, those with no shoulder pain in Stage 2 and 3 of the NWAHS, those with pain in Stage 2 but not Stage 3 and those with pain in Stage 2 and

Background/Purpose: Greater trochanteric pain syndrome (GTPS) is a common cause of musculoskeletal pain that has been typically attributed to trochanteric bursitis. It encompasses a spectrum of disorders [gluteal tendinopathy, tears, bursitis, and iliotibial band (ITB) syndrome] that are difficult to distinguish by clinical exam alone. Few modalities for the treatment of GTPS exist and are recommended without consideration of the structural etiology of pain. Better understanding of ultrasound (US) imaging characteristics in relation to clinical symptoms may be helpful in identifying those patients who would most benefit from the various treatment options available.

S94

US Imaging Characteristics

N (%)

Bony changes (Grade 0) Bony changes (Grade 1) Bony changes (Grade 2) Bony changes (Grade 3) Bursal fluid collections Calcifications of gluteus medius Calcifications of gluteus minimus Calcifications in other locations Enthesophytes in gluteus medius Enthesophytes in gluteus minimus Gluteus medius tendinopathy Gluteus medius partial thickness tear Gluteus medius full thickness tear Gluteus minimus tendinopathy Gluteus minimus partial thickness tear Gluteus minimus full thickness tear Tensor fascia latae abnormality (tendinopathy, tear, fluid) Distal ITB tendinopathy Distal ITB peritendinous fluid

42 (56.8) 18 (24.3) 13 (17.6) 1 (1.4) 34 (46.0) 29 (39.2) 7 (9.5) 4 (5.4) 7 (9.5) 1 (1.4) 57 (77.0) 11 (14.9) 1 (1.4) 10 (13.5) 3 (4.1) 0 (0.0) 14 (18.9) 6 (8.1) 3 (4.1)

(SPADI) and clinical assessment by an independent (blinded) rheumatologist. This clinical assessment included determination of the physician global assessment (PGA), the presence of Hawkins sign and the presence of supraspinatus tendon (SST) tenderness. Data are expressed as median (interquartile range) or as number (percentage). Results: At baseline and at 12 weeks there were no significant differences in assessed parameters. 12 (60%) of the cohort had an abnormal initial MSUS: of these 5 (42%) received a steroid injection and 7 (58%) received active physiotherapy. After six weeks those who received a steroid injection had significantly different clinical parameter measurements than those receiving active physiotherapy: - decrease in PGA was 80% (61–88%) vs 38% (30–43%, p⫽0.003); - decrease in SPADI was 85% (37–90%) vs 14% (10–23%, p⫽0.01); - resolution of SST tenderness occurred in 5 (100%) vs 0 (0%, p⫽0.003); and - resolution of Hawkins sign occurred in 4 (100%) vs 1 (14%, p⫽0.006). 10 (50%) of the cohort received a steroid injection: of these 5 (50%) received had an abnormal MSUS. After six weeks those with an abnormal MSUS had significantly different clinical parameters measurements than those with a normal MSUS: - decrease in PGA was 80% (61–88%) vs 20% (20–35%, p⫽0.008); - decrease in SPADI was 85% (37–90%) vs 19% (10–63%, p⫽0.05); - resolution of SST tenderness occurred in 5 (100%) vs 0 (0%, p⫽0.002); and - resolution of Hawkins sign occurred in 4 (100%) vs 0 (0%, p⫽0.002). Resolution of SST tenderness and Hawkins sign remained significantly different at 12 weeks. Conclusion: The presence of a significant structural abnormality at baseline MSUS suggests that outcome, in the short term at least, may be superior when patients receive a guided injection rather than physiotherapy. And conversely a normal baseline scan may indicate that physiotherapy is the preferred option. Adequately powered randomized clinical trials are required to determine whether treatment decision-making based on MSUS findings is superior to standard management without use of MSUS. Disclosure: M. Khan, None; K. McCreesh, None; A. Saeed, None; T. Ahern, None; A. D. Fraser, None.

233 Ultrasonographic Evaluation Of Patellar Cartilage and Triangular Fibrocartilage In Patients With Familial Benign Hypercalcemia. Alberto Batticciotto1, Diana Certan2, Valentina Varisco1, Marco Antivalle1, Fabiola Atzeni1, Maurizio Bevilacqua2 and Piercarlo Sarzi-Puttini1. 1Rheumatology Unit, L. Sacco University Hospital, Milan, Italy, 2Endocrinology Unit, L. Sacco University Hospital, Milan, Italy

Conclusion: GTPS is commonly attributed to trochanteric bursitis, but only 46% of the subjects had US evidence of true bursitis. The addition of US evaluation to the clinical assessment of GTPS increase may diagnostic accuracy and improve medical decision making.

Background/Purpose: Familial benign hypercalcemia (FBH) is a rare autosomal dominant disease that is caused by a mutation in calcium sensing receptor (CaSR) genes and characterised by hypercalcemia and hypocalciuria with normal serum parathormone levels. The typical clinical symptoms are fatigue, weakness and polydipsia, but rare cases may be associated with chondrocalcinosis, pancreatitis and gallstones. One sensitive and specific means of diagnosis is an oral calcium and peptone load test. The aim of this study was to evaluate the presence of calcium deposits in patellar cartilage and triangular fibrocartilage levels in FBH patients using the same ultrasonographic guidelines as those proposed for patients with gout and chondrocalcinosis. Methods: Twenty-three patients with FBH (2 M, 21F; mean age 69 yrs, range 51–85; mean serum ionised calcium level 1.33 mmol/L, range 1.2–1.5) and 22 controls (1 M, 20 F; mean age 68 yrs, range 45–89; mean serum ionised calcium level 1.1 mmol/L, range 1.1–1.3), with normal serum ionised calcium levels who had negative oral calcium and peptone load test results, were evaluated be means of patellar cartilage and triangular fibrocartilage ultrasonography (ESAOTE My Lab 70, linear probe 13–18 MHz). Results: The two groups were statistically similar in terms of age (p⫽0.709), serum parathormone levels (p⫽0.548), serum vitamin D levels (p⫽0.508) and calciuria levels (p⫽0.194), but different in terms of serum ionised calcium levels (p⬍0.001). Ultrasonography revealed microcrystalline deposits in at least one cartilage in 16/23 (69.6%) FBH patients and in 8/22 (36,4%) controls with a statistically significant difference (p⫽0.026).

Disclosure: M. J. Kohler, None; N. Rastalsky, None; L. Fraenkel, None.

232 Can Musculoskeletal Ultrasonography Examination (MSUS) Predict Outcome In Shoulder Impingement Syndrome (SIS)? A Prospective Blinded Study. Mumtaz Khan1, Karen McCreesh2, Aamir Saeed1, Tomas Ahern3 and Alexander D. Fraser4. 1Limerick University Hospital, Limerick, Ireland, 2University of Limerick, Limerick, Ireland, 3St. Vincent’s University Hospital, Dublin, Ireland, 4Limerick Univeristy Hospital, Limerick, Ireland Background/Purpose: Steroid injection and active physiotherapy are the two standard conservative therapies used to treat SIS and the outcome from these two treatments varies depending on numerous prognostic factors. Outcome predictors may identify patients suitable for specific therapies.The role of MSUS in the diagnosis of rotator cuff disease is well documented, however its utility in determining prognosis and selecting treatment pathways has not been yet assessed. This prospective investigation was designed to assess the utility of MSUS in determining which patients may respond to guided steroid injection or active physiotherapy in SIS. Methods: Twenty consecutive patients with a new diagnosis of isolated SIS (symptoms duration less than 6 months) underwent MSUS. Participants chose to receive either ultrasound guided steroid injection or active physiotherapy. Participants were assessed at baseline, 6 weeks and 12 weeks. Assessments included shoulder pain and disability index

S95

Sunday, October 27

Methods: A prospective, observational, descriptive study was performed to describe the soft tissue and bony structural US findings identified in the lateral hip in patients presenting with GTPS to a dedicated outpatient rheumatology musculoskeletal US clinic. US imaging was obtained using a standardized protocol by 2 US-trained rheumatologists (NR, MK). Eligible subjects included ages 18 and above with lateral hip pain thought to be due to GTPS with pain for at least 1 week and a pain score of at least 2 out of 10 on an 11-point numeric rating scale at rest or with activity. For subjects with bilateral GTPS, the most symptomatic hip was considered the study hip. Most important ineligibility criteria included Body Mass Index (BMI) ⬎40, groin pain, and symptomatic osteoarthritis of the hip. All US images were reviewed by both doctors for consensus of image interpretation. Results: 74 eligible subjects were identified ranging in age from 25–85, with a mean BMI of 28 (⫾4.77); 62 (84.0%) were female. Forty-five (60.8%) had concomitant low back pain and 11 (14.9%) had chronic widespread pain. The mean duration of symptoms was 18 (⫾23.0) weeks. The mean level of pain at rest and activity were 4.8 (⫾2.87) and 6.7 (⫾2.36) respectively. Twenty-five (35.0%) had a prior episode of lateral hip pain. The most common location of bursal fluid was the subgluteus maximus bursa. Frequency and percentage of various imaging characteristics are summarized in the Table.

Sunday, October 27

Conclusion: Patients with hypercalcemia due to FBH have a higher prevalence of ultrasonographically detectable microcrystalline deposits in patellar cartilage and triangular fibrocartilage than normocalcemic controls. Age (yrs) Parathormone (pg/ml) Ionised Calcium (mmol/L) Vitamine D (ng/ml) Calciuria /24h (mg/dl)

FBH

Control Group

P

69 (range 51–85) 49,1 ⫾ 11,4 1.33 ⫾ 0.6 37.3 ⫾ 13.6 216 ⫾ 120.8

68 (range 45–89) 46,67 ⫾ 13,67 1.18 ⫾ 0.5 40⫾13.9 170 ⫾ 88.3

0.709 0.548 0.001 0.508 0.194

ACR Poster Session A Osteoarthritis - Clinical Aspects I: Risk Factors for and Sequelae of Osteoarthritis Sunday, October 27, 2013, 8:30 AM–4:00 PM

235 Meniscal Extrusion Is a Better Predictor Of Radiographic Osteoarthritis Change Over Ten Years Compared To Cartilage Volume Loss. Louisa Chou1, Hussain Ijaz Khan1, Andrew McBride1, Dawn Aitken1, Changhai Ding1, Leigh Blizzard1, Jean-Pierre Pelletier2, Johanne Martel-Pelletier2, Flavia Cicuttini3 and Graeme Jones1. 1Menzies Research Institute Tasmania, University of Tasmania, Hobart, 7000, Australia, 2Osteoarthritis Research Unit, University of Montreal Hospital Research Centre (CRCHUM), Montreal, QC, 3Department of Epidemiology and Preventive Medicine, Monash University, Melbourne, 3004, Australia

Disclosure: A. Batticciotto, None; D. Certan, None; V. Varisco, None; M. Antivalle, None; F. Atzeni, None; M. Bevilacqua, None; P. Sarzi-Puttini, None.

234 The Power Doppler Modalities Of 5 Ultrasound Machines Perform Very Differently, Ascertained By a Microvessel Flow Phantom. David F. Ten Cate, Jolanda J. Luime, Myrthe van der Ven, Johanna M.W. Hazes, Klazina Kooiman, Nico De Jong and Johannes G. Bosch. Erasmus University Medical Center, Rotterdam, Netherlands

Background/Purpose: Change in joint space narrowing (JSN) on radiographs is currently the gold standard for monitoring disease progression of knee osteoarthritis in clinical trials and is considered a surrogate marker for cartilage volume. However, this method can be unreliable due to high measurement error and it is insensitive to early change. Furthermore, cross-sectional studies have shown that JSN not only reflects hyaline articular cartilage but also meniscal pathology. Therefore, the aim of this study was to quantitate the correlation between changes in meniscal extrusion and cartilage volume loss over two years and radiographic osteoarthritis (ROA) change over ten years. Methods: A sample of 220 subjects (mean age 45 years at baseline; age range 26–61 years) was evaluated at baseline, two, and ten years. Approximately half were adult offspring of subjects who had a knee replacement performed for knee osteoarthritis and the remaining were randomly selected controls that were initially matched by age and sex. Knee cartilage volume and meniscal extrusion were determined using T1-weighted fat-suppression MRI techniques at baseline and two years. ROA, JSN, and osteophytes were scored using fixed flexion radiographs at baseline and ten year using standard protocols. Spearman-ranked correlation analysis was used to analyse the correlation between changes in medial meniscal extrusion (MME) and medial tibial cartilage (MTC) loss with changes in ROA, JSN, and osteophytes. All were adjusted for age, sex and BMI. Results: The mean MTC volume was 2234mm3 at baseline with a 2.5% loss per annum over two years. At baseline, 8% of participants had any medial meniscal extrusion and 3% had an incident meniscal extrusion over two years. An increase in MME over two years predicted a change in ROA over ten years in adjusted analysis (r⫽0.22, p⫽0.003), an increase in medial JSN (r⫽0.30, p⫽⬍0.001) and an increase in medial osteophytes (r⫽0.15, p⫽0.046) [Fig 1]. However, a change in MTC over two years did not correlate with a change in ROA over ten years (r⫽⫺0.12, p⫽0.096) or a change in medial JSN (r⫽⫺0.10, p⫽0.182) and weakly with change in medial osteophytes (r⫽⫺0.14, p⫽0.045) [Fig 1].

Background/Purpose: In many patients with rheumatoid arthritis (RA) subclinical disease activity can be detected with ultrasound (US), especially using power Doppler US (PDUS).[1–3] However, in our experience, PDUS is highly dependent on type of machine. This is a problem when conducting multicentre trials or in clinical practice. US machines should be able to detect the blood flow velocities in capillaries as low as 0.5 mm/s. [4–5] The objective of the study was to determine the lowest detectable flow by PDUS of 5 different US machines using microvessel flow phantom. Methods: The flow phantom, consisted of an acrylic (PMMA) container filled with tissue mimicking material(TMM)[6], with 3 microvessels (150, 1000, 2000 micron). The blood mimicking fluid was based on the recipe by Ramnarine.[7] A syringe pump generated the flows. We tested the Aloka ␣7, Esaote MyLab60, Philips iU22, Ultrasonix SonixTouch, and VisualSonics Vevo2100. Settings were optimised to detect the lowest flows by adjusting pulse repetition frequency, velocity range, wall filters, frequency and gain. Lowest possible flow was defined as a continuous PDUS signal. Flow velocities were calculated from pump flow setting and microvessel diameter. Results: Table 1. Lowest detected flow velocity (mm/s), still resulting in a continuous positive PDUS signal; N.D. ⫽ none detected. Vessel size (micron)

2000

Flow velocity 1000

150

Machine (probe) Aloka ␣7(UST-5411) Esaote MyLab 60 (LA 435) Philips iU22 (L9-3) Ultrasonix SonixTouch (L14-5/38) Visual Sonics Vevo2100 (MS200)

4 ⬍0.05 1 1 0.5

2.2 0.06 0.56 0.56 0.33

11.1 0.11 1.68 N.D. N.D.

Conclusion: The performance of the PDUS modality of 5 US machines for detecting very low flows in small vessels was very different, when tested on a microvessel flow phantom. The large differences found between the machines are partly explained by different lower limits of PRF and wall filter, but also by fundamental differences in processing of the PD signal or internal settings inaccessible to users. In the 150 micron vessel, minimal detectable velocity is higher because of sensitivity issues. The actual flow velocity in the 1000 micron vessel was probably slightly higher than calculated and similar to that in the 2000 micron vessel. A reason for this could be compression by the TMM and a more parabolic flow velocity profile in the 1000 micron vessel, with a relatively high peak flow, as compared to the 2000 micron vessel. Caution should be exercised when conducting a multi-machine trial or when making treatment decisions based on PDUS. Based on the results of our study it would be advisable to standardise and validate US machines both for rheumatological clinical practice and for clinical trials. Our flow phantom could be used for this purpose.

Figure 1. Correlation between changes in medial meniscal extrusion and medial tibial cartilage loss over two years with radiographic changes over ten years

Conclusion: In a midlife cohort at higher risk of osteoarthritis, a change in MME, despite being quite rare, is more strongly correlated with subsequent change in ROA than cartilage volume loss. These findings suggest ROA is a composite measure of joint pathology and does not primarily or sufficiently reflect cartilage loss.

References: [1] Peluso (2011) Ann Rheum Dis [2] Saleem (2011) Ann Rheum Dis [3] Brown (2008) Arthritis Rheum [4] Stucker (1996) Microvas Res [5] Stucker (2004) Skin Res Tech [6] Teirlinck (1998) Ultrasonics [7]Ramnarine (1998) Ultrasound Med Biol

Disclosure: L. Chou, None; H. I. Khan, None; A. McBride, None; D. Aitken, None; C. Ding, None; L. Blizzard, None; J. P. Pelletier, ArthroLab, 4; J. Martel-Pelletier, ArthroLab, 4; F. Cicuttini, None; G. Jones, None.

Disclosure: D. F. Ten Cate, None; J. J. Luime, None; M. van der Ven, None; J. M. W. Hazes, None; K. Kooiman, None; N. De Jong, None; J. G. Bosch, None.

S96

Hip Bone Marrow Lesions In Asymptomatic and Osteoarthritic Adults: Prevalence, Risk Factors and Significance. L Dawson1, Yuanyuan Wang1, Anita Wluka1, Kim Bennell2, Andrew Teichtahl1 and Flavia Cicuttini3. 1 Monash University, Melbourne, Australia, 2University of Melbourne, Melbourne, Australia, 3Department of Epidemiology and Preventive Medicine, Monash University, Melbourne, 3004, Australia Background/Purpose: Bone marrow lesions (BMLs) at the knee have an important role in the pathogenesis of osteoarthritis (OA), being associated with increased pain, accelerated cartilage loss, and increased risk of total knee replacement. However, data is limited for the role of BMLs at the hip. The aim of this study was to determine the prevalence and associations of BMLs at the hip in an asymptomatic and an osteoarthritic population. Methods: 142 asymptomatic and 19 participants with hip OA were recruited from existing cohorts. Hip magnetic resonance imaging was performed and used to assess femoral head BMLs, cartilage volume and bone area. Results: The demographic characteristics of the asymptomatic versus the OA population were as follows: age 66.8 ⫹/⫺ 7.4 vs. 59.5 ⫹/⫺ 7.6 years (p⬍0.001), female 55.6% vs. 57.9% (p⫽0.85), body mass index 27.6 ⫹/⫺ 4.8 vs. 27.2.⫹/⫺.4.8 kg/m2. (p⫽0.73). The prevalence of BMLs was 17.6% in the asymptomatic population and 63.2% in the OA population (p⬍0.001). BMLs were strongly associated with OA after adjusting for age, gender and body mass index (odds ratio 5.32, 95% CI 1.78, 15.9, p⫽0.003). BMLs were associated with lower femoral head cartilage volume in the whole population (regression coefficient ⫺245.7 mm3, 95% CI ⫺455.5, ⫺36.0, p⫽0.02). In the OA population, BMLs were also associated with lower femoral head cartilage volume (regression coefficient ⫺426.6 mm3, 95% CI ⫺855.2, 2.14, p⫽0.05) after adjusting for age, gender, body mass index, femoral head bone area and hip OA (for analysis of the total population). Conclusion: Femoral head BMLs are common in those with OA, but are also present in asymptomatic individuals with no clinical hip OA. They are associated with reduced hip cartilage volume. These findings suggest that BMLs at the hip may provide a novel target for the treatment and prevention of hip OA.

Exposure BML⬎0 in any subregion

Level Lesion No Lesion

Cartilage lesion area⬎0 in any subregion

Lesion

Meniscal tear or extrusion

Lesion

No Lesion

No Lesion

Medial TF Compartment MRI damage n/N (%) among among control case knees knees 24/200 (12%) 176/200 (88%) 72/200 (36%) 128/200 (64%) 55/200 (27.5%) 145/200 (72.5%)

63/200 (31.5%) 137/200 (68.5%) 122/200 (61%) 78/200 (39%) 82/200 (41%) 118/200 (59%)

Adjusted Model for Medial Compartment OR (95% CI)

p-value

3.44 (1.99, 5.94) 1

⬍.0001

2.48 (1.63, 3.79) 1

⬍.0001

1.85 (1.19, 2.89) 1

0.0066

Lateral TF Compartment MRI damage n/N (%) among among control case knees knees 21/200 (10.5%) 179/200 (89.5%) 89/200 (44.5%) 111/200 (55.5%) 28/200 (14%) 172/200 (86%)

158/200 (79%) 42/200 (21%) 196/200 (98%) 4/200 (2%) 183/200 (91.5%) 17/200 (8.5%)

Adjusted Model for Lateral Compartment OR (95% CI)

p-value

39.96 (21.11, 75.65) 1

⬍.0001

89.85 (29.69, 271.90) 1

⬍.0001

93.77 (43.46, 202.32) 1

⬍.0001

Conclusion: Knees with isolated radiographic OA in the lateral compartment have more MRI abnormalities of cartilage, meniscus and BMLs in the lateral compartment than knees without radiographic OA. Isolated lateral compartment ROA is also associated with medial compartment MRI lesions, but less strongly. Disclosure: B. L. Wise, Pfizer, Inc., 2; J. Niu, None; F. Liu, None; J. A. Lynch, None; Y. Zhang, None; A. Guermazi, Boston Imaging Core Lab, 1, Merck Serono, 5, Sanofi-Aventis Pharmaceutical, 5, TissueGene, 5; D. T. Felson, None; C. K. Kwoh, None; N. E. Lane, None.

238 Structural Predictors Of Ten Year Knee Cartilage Volume Loss. Dawn Aitken1, Hussain Ijaz Khan1, Changhai Ding1, Leigh Blizzard1, Jean-Pierre Pelletier2, Johanne Martel-Pelletier2, Flavia Cicuttini3 and Graeme Jones1. 1 Menzies Research Institute Tasmania, University of Tasmania, Hobart, 7000, Australia, 2Osteoarthritis Research Unit, University of Montreal Hospital Research Centre (CRCHUM), Montreal, QC, 3Department of Epidemiology and Preventive Medicine, Monash University, Melbourne, 3004, Australia

Disclosure: L. Dawson, None; Y. Wang, None; A. Wluka, None; K. Bennell, None; A. Teichtahl, None; F. Cicuttini, None.

237

Background/Purpose: Cartilage volume loss is considered a key feature of knee osteoarthritis (OA) and increased loss over two years is associated with future knee replacement surgery. Previous studies have identified structural predictors of cartilage loss over 2–3 years including baseline cartilage volume, cartilage defects, bone marrow lesions (BMLs), meniscal pathology and bone area. These are now considered important risk factors for osteoarthritis progression. However no study has examined whether these features predict loss over longer timeframes. The aim of this study was to examine structural predictors of cartilage volume loss over ten years. Methods: 220 participants [mean age 45 (26–61); 57% female] were studied at baseline, two, and ten years. Approximately half were adult offspring of subjects who had a knee replacement performed for knee OA and the remaining were randomly selected controls that were initially matched by age and sex. Cartilage volume (tibial, patella), cartilage defects (tibial, femoral), BMLs (tibial, femoral, patella), meniscal extrusion (medial, lateral) and bone area (tibial) were measured using MRI. Predictors of ten year cartilage loss were examined using linear regression and all models were adjusted for age, sex, BMI, and offspring-control status. Interactions between offspring-control status and knee structures on cartilage loss were explored. Results:Cartilage change over ten years was ⫺2.5% per annum (pa), ⫺1.1% pa, and ⫺2.1% pa for the medial and lateral tibial and patella sites respectively. BMLs predicted tibial (␤ ⫽ ⫺433 mm3 per 1 unit increase in BML size, P ⬍ 0.01) but not patella cartilage loss. Meniscal extrusion predicted lateral tibial cartilage loss (␤ ⫽ ⫺872 mm3 in those with a baseline extrusion versus those without, P ⬍ 0.01) but not medial tibial or patella cartilage loss. Cartilage volume at baseline predicted tibial and patella ten year cartilage loss (␤ ⫽ ⫺0.26 to ⫺0.41 mm3 per unit increase in baseline volume, P⬍ 0.01). There was interaction between offspring-control status for cartilage defects. Baseline defect score predicted lateral tibial cartilage loss in offspring (B⫽ ⫺242 mm3 per unit increase in defect score, P ⬍ 0.01); but, surprisingly defects appeared to be protective against medial and lateral tibial cartilage loss in controls (␤ ⫽ 276 to 327 mm3 per unit increase in defect score, P⬍0.01).

The Association Of Radiographic Disease With Lesions On MRI In Lateral Compartment Knee Osteoarthritis: The Osteoathritis Initiative. Barton L. Wise1, Jingbo Niu2, Felix Liu3, John A. Lynch4, Yuqing Zhang2, Ali Guermazi5, David T. Felson2, C. Kent Kwoh6 and Nancy E. Lane1. 1UC Davis School of Medicine, Sacramento, CA, 2Boston University School of Medicine, Boston, MA, 3University of California at San Francisco, San Francisco, CA, 4University of California-San Francisco, San Francisco, CA, 5 Boston University, Boston, MA, 6University of Pittsburgh, Pittsburgh, PA Background/Purpose: Lateral compartment knee tibio-femoral radiographic osteoarthritis (ROA) is associated with pain and disability, and the epidemiology of this form of knee OA is not well known. We examined MRI abnormalities and their association with lateral compartment ROA. Methods: We analyzed knees from participants in the Osteoarthritis Initiative (OAI), an NIH-funded study of persons aged 45–79 at baseline with, or at high risk of knee ROA. Baseline radiographs with central readings and MRIs were used to assess knee structure lesions. We defined cases as knees with Kellgren/Lawrence (K/L) grade ⬎⫽2 and with joint space narrowing (JSN) ⬎0 in the lateral compartment but JSN ⫽ 0 in the medial compartment; controls were defined as having K/L⫽0 and JSN⫽0 in all compartments. Cases and controls were matched by sex and age in 10-year bands. MRI readings for cases and controls were obtained from two ancillary studies in the OAI and further MRI readings were generated for this analysis. MRIs were scored at each subregion for cartilage lesions, bone marrow lesions, and meniscal tear or extrusion using the MOAKS scoring system. The score for each lesion was dichotomized into “any” or “none”. Lesions in medial and lateral compartments were defined separately for whether any of the subregions in each compartment exhibited that lesion. We compared cases and controls for each structural lesion separately in both lateral and medial compartments using conditional logistic regression, adjusting for age, race, body mass index (BMI), and clinic site. Results: Included in this analyses were 200 cases (mean age: 63.8, 34.5% men, mean BMI 29.5, and 76.5% white) and 200 controls (mean age: 63.1,

S97

Sunday, October 27

34.5% men, mean BMI 27.1, and 87.5% white). Isolated lateral compartment ROA was strongly associated with lateral compartment cartilage lesions (odds ratio (OR)⫽89.9, 95%CI 29.7–271.9), BMLs (OR⫽40.0, 95% CI: 21.1– 75.7) and meniscal tear or extrusion (OR⫽ 93.8, 95% CI 43.5–202.3). Medial compartment cartilage lesions, BMLs and meniscal abnormalities were also associated with isolated lateral compartment ROA; the magnitude of association, however, was much smaller. (See table)

236

Sunday, October 27

Both change in cartilage volume (r ⫽ ⫺0.58 to ⫺0.62, all P ⬍ 0.01) and progression of cartilage defects (r ⫽ ⫺0.18 to ⫺0.58, all P⬍ 0.01) over two years predicted ten year tibial cartilage loss. Bone area did not predict ten year change in cartilage. Conclusion: Structural features which have been identified as risk factors for cartilage loss over 2–3 years appear to predict long-term cartilage changes with the most consistent data for BMLs. This was seen in a largely pre-radiographic cohort and indicates that bone marrow lesions, presence of meniscal extrusion; and to a lesser extent, cartilage volume and cartilage defects, may be used to identify individuals who will lose the most cartilage long-term. Additionally this study demonstrates that cartilage loss and cartilage defect progression over two years are reasonable predictors of long-term cartilage changes.

In unadjusted analysis for cartilage loss, offspring lost more cartilage at the medial tibial site only. This persisted after adjustment for age, sex, baseline cartilage volume and baseline differences. Conclusion: The offspring of subjects having a total knee replacement have greater incidence of radiographic osteoarthritis (both joint space narrowing and osteophytes) and higher medial tibial cartilage loss over ten years. This is mediated, in part, by baseline differences in pain, weight, cartilage defects and tibial bone area and suggests this group could be a target for early prevention strategies. Disclosure: H. I. Khan, None; D. Aitken, None; L. Chou, None; A. McBride, None; C. Ding, None; J. P. Pelletier, None; J. M. Pelletier, ArthroLab, 4; L. Blizzard, None; F. Cicuttini, None; G. Jones, None.

Disclosure: D. Aitken, None; H. I. Khan, None; C. Ding, None; L. Blizzard, None; J. P. Pelletier, ArthroLab, 4; J. Martel-Pelletier, ArthroLab, 4; F. Cicuttini, None; G. Jones, None.

240 Longitudinal Femorotibial Cartilage Thickness Increase In Young Athletes At The End Of Adolescence. Felix Eckstein1, Heide Boeth2, Gerd Diederichs3, Wolfgang Wirth1, Martin Hudelmaier1, Sebastian Cotofana1 and Georg Duda2. 1Paracelsus Medical University, Salzburg, Austria, 2Julius Wolff Institute, Charite´ – Universita¨tsmedizin Berlin, Berlin, Germany, 3 Department of Radiology, Charite´ - Universita¨tsmedizin Berlin, Berlin, Germany

239 A Family History Of Knee Joint Replacement Increases The Progression Of Knee Joint Radiographic Osteoarthritis and Cartilage Volume Loss Over Ten Years. Hussain Ijaz Khan1, Dawn Aitken1, Louisa Chou1, Andrew McBride1, Changhai Ding1, Jean Pierre Pelletier2, Johanne M. Pelletier3, Leigh Blizzard1, Flavia Cicuttini4 and Graeme Jones1. 1Menzies Research Institute Tasmania, University of Tasmania, Hobart, 7000, Australia, 2Institut de Rhumatologie de Montreál, Montreal, QC, 3Osteoarthritis Research Unit, University of Montreal Hospital Research Centre (CRCHUM), Notre-Dame Hospital, Montreal, QC, 4Department of Epidemiology and Preventive Medicine, Monash University, Melbourne, 3004, Australia

Unadjusted Adjusted* Difference in ratios and Radiographic changes 95% confidence interval Increase in medial JSN ⴙ2.0 (ⴙ1.1, ⴙ3.5) ⴙ1.9 (ⴙ1.0, ⴙ3.5) Increase in lateral JSN ⫹0.8 (⫹0.3, ⫹2.8) ⫹0.5 (⫹0.2, ⫹1.9) Increase in total JSN ⫹1.5 (⫹0.9, ⫹2.6) ⫹1.4 (⫹0.82, ⫹2.3) Increase in total tibial osteophytes ⴙ2.1 (ⴙ1.2, ⴙ3.9) ⫹1.7 (⫹0.9, ⫹3.1) Increase in total femoral osteophytes ⴙ2.4 (ⴙ1.1, ⴙ5.0) ⫹1.6 (⫹0.72, ⫹3.4) Increase in total osteophytes ⴙ2.3 (ⴙ1.3, ⴙ4.0) ⫹1.6 (⫹0.9, ⫹2.9) Increase in total ROA score ⴙ1.9 (ⴙ1.3, ⴙ3.0) ⫹1.5 (⫹0.9, ⫹2.3) Cartilage loss Difference in means and 95% confidence interval Medial tibial ⴚ91.5 (ⴚ181.6, ⴙ1.33) ⴚ83.4 (ⴚ166.2,ⴚ0.53) Lateral tibial ⫺3.0 (⫺107.9, ⫹101.8) ⫹32.1 (⫺72.9, ⫹137.1) Total Tibial ⫺89.6 (⫺263.0, ⫹83.8) ⫺56.2 (⫺221.4, ⫹109.0) Patellar ⫺0.9 (⫺171.6, ⫹169.8) ⫹77.2 (⫺70.1, ⫹224.1)

Background/Purpose: Anterior or posterior cruciate ligament (ACL/ PCL) ruptures are thought to lead to early knee osteoarthritis (KOA) and thus considered scientifically valuable “models” of pre-radiographic change. Further, longitudinal studies can be initiated at a precise set-point, which is the time of the trauma. Quantitative measurement of cartilage thickness change with MRI provides a powerful and sensitive tool for longitudinal analysis of structural cartilage change before, during, and after the onset of radiographic KOA. Because ACL/PCL rupture frequently occurs in young athletes, the purpose of this study was to obtain reference data of normal longitudinal subregional femorotibial cartilage thickness change in such subjects, specifically at the end of adolescence. Methods: The knees of the dominant leg (the one used for take off) of 18 young top volleyball athletes (Olympic center Berlin) were examined. One participant was retrospectively excluded because of a history of ACL rupture, and one because of an imaging artifact. The age range (n⫽16; 8 female, 8 male) was 15–17 y at baseline. MR images were acquired using a 3D VIBE with water excitation, at baseline and two years later. Femorotibial cartilage thickness was measured by manual segmentation, using the images and commercially available software (Chondrometrics GmbH, Ainring, Germany). Cartilage thickness data were computed in the medial (MFTC) and lateral femorotibial compartment (LFTC), in tibial and femoral cartilages, and in 16 femorotibial subregions. Results: Of the 16 subjects, 11 displayed open epiphyses at baseline, and all showed closed epiphyses at follow-up. In the MFTC, a cartilage thickness increase of ⫹114⫾126␮m (mean⫾standard deviation) or ⫹3.3⫾3.7% was observed, with a standardized response mean (SRM⫽ mean/standard deviation of change) of ⫹0.90. This increase was less in the LFTC (⫹2.3⫾2.5%; SRM ⫹0.93); and was greater in the weight-bearing medial femur (⫹4.4⫾3.7%; SRM ⫹1.19) than in the medial tibia (⫹2.0⫾4.3%; SRM ⫹0.46). The greatest increase was observed in the internal aspect of the medial femur (⫹5.5⫾5.8%; SRM ⫹0.94) whereas no increase was seen in the internal aspect of the medial tibia (⫺0.5⫾6.6%; SRM ⫺0.07). Over the two years, the subchondral bone area (tAB) increased by 0.8⫾1.5% in the MFTC and by 1.2⫾1.2% in the LFTC. Conclusion: A substantial increase in femorotibial cartilage thickness (and subchondral bone area) was observed in young athletes towards the end of adolescence, i.e. a period during which the epiphyseal line is closing. This increase should be considered when measuring longitudinal cartilage change in young athletes after ACL/PCL rupture. Further studies need to clarify whether the increase is due mainly to normal growth, or to high exercise levels. Longitudinal studies in young adults with cruciate ligament rupture should hence include healthy reference cohorts of similar age (and activity level), to adequately differentiate pathological (post traumatic) change from that occurring physiologically.

Bold denotes statistical significance *Adjusted for age, sex, corresponding baseline measure and baseline differences(BMI, knee pain, cartilage defects and tibial bone area) JSN(Joint Space Narrowing) ROA(Radiographic Osteoarthritis)

Disclosure: F. Eckstein, Chondrometrics, 4, Chondrometrics, 3, MerckSerono, 5, Abbvie, 5, Novartis Pharmaceutical Corporation, 2; H. Boeth, None; G. Diederichs, None; W. Wirth, Chondrometrics, 4, Chondrometrics, 3; M. Hudelmaier, Chondrometrics GmbH, 3; S. Cotofana, Chondrometrics GmbH, 3; G. Duda, None.

Background/Purpose: Osteoarthritis (OA) is a disease of multifactorial origin with a modest but significant heritable effect for disease severity in the knee joint. The aim of this study was to describe radiographic osteoarthritis (ROA) progression and cartilage loss over 10 years in a midlife cohort at higher risk of osteoarthritis and community based controls. Methods: 220 participants [mean age 45 (26–61); 57% female] were studied at baseline and ten years. Approximately half were the adult offspring of subjects who had a knee replacement performed for knee OA and the remaining were randomly selected controls that were initially matched for age and sex. Joint space narrowing (JSN) and osteophytes on radiographs and cartilage volume (tibial, patella) on MRI were assessed using standard protocols. Negative binomial and linear regression were used to describe radiographic changes (expressed as difference in ratios (dr)) and cartilage loss (expressed as difference in means (dm)) respectively. Multivariable analyses were first adjusted for age, sex and the corresponding baseline measures (i.e. baseline cartilage volume for cartilage loss).Then the four baseline measures which were significantly different between offspring and controls (BMI, knee pain, cartilage defects and tibial bone area) were added to the model. Results: In unadjusted analysis for radiographic changes, there was a significant difference between offspring and controls for changes in total ROA, medial JSN, total tibial, total femoral and total osteophyte scores (Table 1). The associations persisted after adjusting for age, sex, and corresponding baseline measures. However, with the exception of medial JSN, most of the other associations were markedly attenuated after adjustment for baseline pain, weight, cartilage defects and tibial bone area. Table 1. Differences between offspring and controls Outcome factor

S98

241 Exploratory Factor Analysis Of a Parsimonious Medial Cartilage Damage Index Reveals One Factor Associated With Radiographic Severity and Symptoms: Data From The Osteoarthritis Initiative. Grace H. Lo1, Lori Lyn Price2, Ming Zhang3, Jeffrey B. Driban3, Daniel Harper3 and Timothy E. McAlindon3. 1Michael E. DeBakey VA Medical Center, Baylor College of Medicine, Houston, TX, 2Tufts University, Boston, MA, 3Tufts Medical Center, Boston, MA

CDI CDI AntWB CDI

Background/Purpose: The cartilage damage index (CDI) is a parsimonious method of measuring articular cartilage approximating regional cartilage volume based on 18 locations within the knee, 9 in the medial tibia and femur (figure). Understanding locations that are simultaneously pathologic may better clarify the pathophysiology of knee osteoarthritis (OA). Therefore, we conducted an exploratory factor analysis of the CDI.

PostTib CDI PostFem CDI Medial JSN KL Score HKA WOMAC

1

AntWB PostTib PostFem CDI CDI CDI

Medial JSN

KL Score

0.97832 0.67338 0.68628 ⫺0.50802 ⫺0.28166 ⬍.0001 ⬍.0001 ⬍.0001 ⬍.0001 ⬍.0001 1 0.61494 0.56546 ⫺0.57167 ⫺0.35091 ⬍.0001 ⬍.0001 ⬍.0001 1 0.45342 ⫺0.1569 ⬍.0001 1

0.0269 ⫺0.1293 0.0687 1

HKA

WOMAC

0.28489 ⫺0.14135 ⬍.0001 0.0459 0.32417 ⫺0.1724

⬍.0001 ⬍.0001 0.0146 ⫺0.10798 ⫺0.04208 ⫺0.04712 0.129 0.12819

0.5581 0.11757

0.5076 0.02053

0.0712 0.1008 0.56874 ⫺0.52189

0.7729 0.09234

⬍.0001 1

⬍.0001 ⫺0.19741 0.0057 1

0.1946 0.29629 ⬍.0001 0.02272 0.7519 1

Conclusion: These findings suggest 3 patterns of cartilage damage occur in the medial tibia and femur. While PostTib and PostFem CDI are correlated with the full CDI, they are at best weakly correlated with medial JSN and not correlated with KL score, static alignment or pain. The AntWB CDI pattern contributes most of the variation in the 18 points and shows construct validity, correlating with medial JSN, KL score, static alignment, and pain, better than the full CDI measurement. Future studies confirming these findings are warranted. Disclosure: G. H. Lo, None; L. L. Price, None; M. Zhang, None; J. B. Driban, None; D. Harper, None; T. E. McAlindon, None.

242 Elevated Systemic Blood Pressure Is Associated With Increased Prevalent Knee Osteoarthritis and Knee Pain: Data From The Osteoarthritis Initiative. Grace H. Lo1, Timothy E. McAlindon2, Jeffrey N. Katz3, Jeffrey B. Driban2, Lori Lyn Price4, Charles Eaton5, Nancy Petersen6, Christie Ballantyne7 and Maria E. Suarez-Almazor8. 1Michael E. DeBakey VA Medical Center, Baylor College of Medicine, Houston, TX, 2Tufts Medical Center, Boston, MA, 3Brigham and Women’s Hospital, Boston, MA, 4Tufts University, Boston, MA, 5Brown University, Providence, RI, 6Michael E. DeBakey VA Medical Center Health Services Research and Development Center of Excellence, Houston, TX, 7Baylor College of Medicine, Houston, TX, 8University of Texas MD Anderson Cancer Center, Houston, TX Background/Purpose: Epidemiologic studies have linked osteoarthritis (OA) to atheromatous vascular disease, both age-related diseases. Systemic hypertension (HTN) is an important risk factor for cardiovascular disease and may also be for OA. The objective of this study was to evaluate whether there is an association between HTN and prevalent knee OA and pain in the Osteoarthritis Initiative (OAI). Methods:This is a longitudinal study of the OAI, a multi-center observational study of knee OA. Systolic (SBP) and extensive medication data were accumulated on participants at their OAI baseline, 12-, 24-, and 36month visits. Bilateral PA knee radiographs were taken at the OAI 48-month visit. Central readers assessed radiographic OA severity, Kellgren and Lawrence (KL) grade (0–4). WOMAC pain scores were assessed for both knees at the 48-month visit. The greatest WOMAC score within one participant represented the maximal WOMAC score. There may be a lag between exposure to HTN and development of OA; therefore, we evaluated a 4 year interval between exposure and outcome. We performed logistic regression of quartiles of baseline SBP and 48-month status of prevalent radiographic OA (KL ⬎ 2) in either knee. We evaluated prevalent instead of incident knee OA to minimize the likelihood of right sided censorship. We performed a least square means of the maximum WOMAC score by quartile of SBP. The adjusted model included age, sex, and body mass index (BMI) as covariates. To eliminate the influence of HTN treatments, we performed subgroup analyses of those not taking anti-HTN medications at the baseline visit. To address the possibility that participants’ SBPs fluctuated over time, we performed a sensitivity analysis using mean SBP over the baseline, 12-, 24-, and 36-month OAI visits as the exposure.

Figure. Spatial diagram of 9 of the femoral CDI points. Each red dot in the grid represents a separate location where CDI of articular cartilage is measured.

Methods: We included a convenience sample of 200 participants of the progression cohort of the Osteoarthritis Initiative (OAI) with 3T knee MRIs, PA radiographs, and WOMAC pain scores (0 – 20) at the OAI 24- month visit. Long limb films obtained at the OAI 12-, 24-, or 36- month visits were used to measure Hip-Knee-Ankle (HKA) angle using a semi-automated program (negative ⫽ varus). One assessor performed CDI measurements on MRI sagittal DESS sequences. Medial joint space narrowing (JSN) (0–3) and Kellgren-Lawrence (KL) score (0–4) were centrally assessed. An exploratory factor analysis was conducted on the 18 CDI points. To assess the characteristics of factors resulting from that analysis, we performed Spearman’s correlations with the full CDI measure, the separate factors, medial JSN (construct we are trying to capture), KL score (OA severity), HKA (risk factor for medial OA), and WOMAC score (OA symptoms). Results: Mean age was 62.5 (9.5) years, mean BMI was 30.0 (4.5) kg/m2, 47% were female. 3 factors resulting from the factor analysis were: (1) anterior and weight bearing portions of the femur and tibia (AntWB) (2) posterior tibia (PostTib) and (3) posterior femur (PostFem), explaining 75%, 11%, and 9% of the data variation.

S99

Sunday, October 27

Table. Spearman’s correlations of CDI, anterior weight-bearing (Ant WB) CDI, posterior tibia (PostTib) CDI, posterior femur (PostFem) CDI, Medial JSN, KL Score, HKA, and WOMAC scores

Sunday, October 27

Results: 3644 people were included with a mean age of 61.1 (9.1), mean BMI of 28.5 (2.7), 42% were male. 61% of participants were not on anti-HTN medications. 20%, 13%, 5%, and 1% were on 1, 2, 3, and ⬎ 4 anti-HTN medications. Table. Systolic blood pressure is associated with prevalent radiographic knee osteoarthritis and knee pain. Logistic regression of SBP quartile and prevalent knee OA. Least Squared Means of maximal WOMAC score by SBP quartiles, adjusted for age, sex, BMI, and radiographic OA status Outcome ⴝ Prevalent Knee OA

N ⴝ 3644 SBP Quartile 1 (76 – 112) SBP Quartile 2 (114 – 122) SBP Quartile 3 (123 – 134) SBP Quartile 4 (135 – 210)

Outcome ⴝ Knee Pain

Prevalence of Knee OA by SBP Quartile

Unadjusted Odds Ratio (95% CI)

Odds Ratio (95% CI) Adjusted for Age, Sex, BMI

531/1031 (52%)

Referent

Referent

LS Means of Maximal WOMAC Score, Adjusted for Age, Sex, BMI, Rad OA status 2.6 (2.4–2.8)

519/846 (61%)

1.5 (1.2–1.8)

1.2 (1.0–1.5)

2.8 (2.6–3.0)

586/865 (68%)

2.0 (1.6–2.4)

1.4 (1.2–1.7)

2.9 (2.7–3.1)

619/902 (69%)

2.1 (1.7–2.5)

1.3 (1.0–1.6)

3.2 (3.0–3.4)

p-for trend ⬍ 0.0001

p-for trend ⫽ 0.006

p trend ⫽ 0.0016

Sensitivity analyses of participants who did not report anti-HTN medications showed similar results, as did using mean SBP over baseline – OAI 36 month as the exposure.

Conclusion:To our knowledge, this is the first epidemiologic study to show elevated systemic blood pressure is associated with increased prevalence of knee OA and greater levels of knee pain. The associations were weakened when adjusting for age, sex, and BMI, but were still significant. These findings suggest a new and promising avenue for research on disease modification as well as symptom control in knee OA. Future epidemiologic studies are needed to confirm these findings. Disclosure: G. H. Lo, None; T. E. McAlindon, None; J. N. Katz, None; J. B. Driban, None; L. L. Price, None; C. Eaton, None; N. Petersen, None; C. Ballantyne, None; M. E. Suarez-Almazor, None.

association between metabolic syndrome and symptomatic hand OA (Table). The strength of association between metabolic syndrome and symptomatic hand OA was attenuated when we adjusted for body mass index (BMI) (cross-sectional association: OR 1.19, 95% CI 0.68–2.10). When evaluating the individual components of metabolic syndrome separately, we found a significant association between central obesity and symptomatic hand OA in the adjusted cross-sectional analyses (Table). In the longitudinal analyses, we found a statistically significantly lower risk of incident symptomatic hand OA associated with elevated triglycerides (Table). The association remained after adjustment for lipid-lowering treatment (data not shown). Similar results were found when we included BMI in the adjusted models (data not shown). Table. Associations between metabolic syndrome and hand OA. Frequency of metabolic syndrome (%) No hand Hand OA OA

Logistic regression OR (95% CI) Adjusted for Crude analyses age and sex

Cross-sectional analyses1 Metabolic syndrome 39.0 48.6 1.48 (0.91–2.40) 1.44 (0.87–2.38) Central obesity 64.2 79.7 2.19 (1.22–3.94) 2.24 (1.22–4.12) Hypertension 40.2 50.0 1.49 (0.92–2.41) 1.27 (0.77–2.11) High blood-glucose 27.5 24.3 0.85 (0.49–1.48) 0.89 (0.49–1.61) High triglycerides 33.4 33.8 1.02 (0.61–1.69) 1.01 (0.60–1.73) Low HDL 64.1 70.3 1.32 (0.78–2.23) 1.30 (0.75–2.23) Longitudinal analyses (only those without hand OA at baseline were included)2 Metabolic syndrome 38.6 31.5 0.73 (0.45–1.18) 0.77 (0.47–1.25) Central obesity 63.5 59.8 0.85 (0.56–1.43) 0.90 (0.56–1.43) Hypertension 37.4 41.3 1.18 (0.74–1.86) 1.24 (0.77–1.20) High blood-glucose 28.0 18.7 0.59 (0.33–1.04) 0.64 (0.35–1.14) High triglycerides 34.0 21.7 0.54 (0.32–0.92) 0.56 (0.33–0.96) Low HDL 64.9 55.4 0.67 (0.43–1.06) 0.67 (0.42–1.05) 1 n⫽674 and n⫽74 without and with symptomatic hand OA at exam 5, respectively, 2 n⫽430 and n⫽92 without and with incident symptomatic hand OA at exam 7, respectively (n⫽152 lost to follow-up),

Conclusion:Metabolic syndrome was not associated with higher probability of presence or development of hand OA. If anything, we observed a lower probability of incident symptomatic hand OA in participants with high triglycerides. Disclosure: I. K. Haugen, None; V. Ramachandran, None; D. Misra, None; T. Neogi, None; J. Niu, None; Y. Zhang, None; D. T. Felson, None.

243 The Association Between Metabolic Syndrome and Hand Osteoarthritis - Data From The Framingham Study. Ida K. Haugen1, Vasan Ramachandran2, Devyani Misra2, Tuhina Neogi2, Jingbo Niu2, Yuqing Zhang2 and David T. Felson2. 1Diakonhjemmet Hospital, Oslo, Norway, 2Boston University School of Medicine, Boston, MA Background/Purpose: Metabolic factors may have a negative effect on cartilage, and may be especially relevant in the pathogenesis of hand OA. Our aim was to investigate whether metabolic syndrome and its components were associated with prevalent and incident hand OA in a large community-based sample. Methods: We included participants from the Framingham Offspring cohort, who were examined for OA at exam 5 (1991–95) and 7 (1998–2001). Inclusion criteria were: available data on symptomatic hand OA, clinical data, no anti-hypertensive, anti-diabetic and lipid-lowering treatment, no rheumatoid arthritis, and age 50–75 years at exam 5. Symptomatic hand OA was defined as Kellgren-Lawrence grade ⱖ2 and pain in the same joint(s). Metabolic syndrome was defined as central obesity (men: ⱖ37.0 inches, women: ⱖ31.5 inches waist circumference) plus two of the following: hypertension (systolic ⱖ130 mmHg and/or diastolic ⱖ85 mmHg), elevated fasting blood glucose (⬎100 mg/dL), elevated triglycerides (⬎150 mg/dL) and low HDL (men: ⬍40 mg/dL, women: ⬍50 mg/dL). In cross-sectional analyses, we examined whether metabolic syndrome and its components were associated with presence of symptomatic hand OA using logistic regression. In longitudinal analyses, we examined the associations between metabolic syndrome and incident symptomatic hand OA at exam 7. We adjusted for age and sex. Results: Of the 748 participants who fulfilled the inclusion criteria, 74 (9.9%) had symptomatic hand OA. Participants with symptomatic hand OA were older than those without (mean (SD) 62.2 (5.9) vs. 57.7 (6.0) years) and more likely to be women (75.7% vs. 51.0%). There was no significant

244 Association Between Serum Adipokine Levels and Extent Of Symptomatic Joint Involvement In Hip and Knee Osteoarthritis. Anthony V. Perruccio1, Vinod Chandran2, Nizar N. Mahomed3 and Rajiv Gandhi4. 1 Toronto Western Hospital, Toronto Western Research Institute, and University of Toronto, Toronto, ON, 2University of Toronto, Toronto Western Hospital, Toronto, ON, 3University Health Network and University of Toronto, Toronto, ON, 4University Health Network, Arthritis Program, Toronto, ON Background/Purpose: Osteoarthritis (OA) has not traditionally been considered a systemic joint disease. However, a growing body of evidence suggests that OA cannot solely be accounted for by a single phenotype, and that systemic factors may play a contributing role in its pathogenesis. Obesity is an important risk factor for OA, but the underlying mechanisms are not fully understood (e.g. strong association between obesity and hand OA cannot be explained by mechanical stresses). Finally, the presence of multiple symptomatic joints among many with OA further suggests potentially underlying systemic mechanisms. In this pilot study, we investigated the association between serum levels of adipokines and the extent of symptomatic joint involvement among patients scheduled for hip and knee replacement surgery for OA. Methods:Serum levels of adipokines (leptin, adiponectin, adipsin, resistin) were determined by ELISA in 78 patients (45 females, 33 males) scheduled for total hip or knee replacement surgery for OA. Individuals reporting inflammatory arthritis, cardiovascular disease or diabetes were excluded. Patients indicated on a homunculus all joints that were symptomatic on most days for at least a month in the past 12 months. A count score of symptomatic regions (neck; spine; shoulders; elbows; wrists; hands; hips; knees; ankles; feet) (not including surgical joint) was developed (range: 0 to

S100

Table 1. Multivariate adjusted associations between regional joint count (outcome) and adipokines. IRR*

Lower 95% CL

Predictors Age BMI Knee vs. hip Leptin (ng/ml) Adiponectin (␮g/ml) Adipsin (mg/ml) Resistin (ng/ml)

1.03 1.01 1.25 1.03 1.03 0.85 1.01

1.005 0.973 0.765 1.010 1.011 0.751 0.987

Age BMI Knee vs. hip Leptin (ng/ml) Adiponectin (ug/ml) Adipsin (mg/ml) Resistin (ng/ml)

1.02 1.03 0.92 0.97 0.97 0.96 0.97

0.979 0.933 0.430 0.923 0.937 0.790 0.942

Upper 95% CL Females 1.058 1.054 2.048 1.047 1.056 0.961 1.038 Males 1.054 1.143 1.949 1.020 1.010 1.176 0.997

Pr > ChiSq 0.0204 0.5407 0.3725 0.0021 0.0031 0.0094 0.3613

regression analyses were used to assess the association of CRP and ESR with changes in muscle strength, adjusting for relevant confounders. Results: Persistently elevated or increased serum CRP values (␤⫽⫺0.23; p⫽0.002), but not ESR values (␤⫽⫺0.05; p⫽0.55), were associated with a decrease in knee muscle strength after two years, even after adjustment for age, sex, comorbidities, NSAIDs use, BMI, pain, physical activity and baseline muscle strength. Conclusion: Our results indicate that persistently elevated or an increase in inflammatory markers are independently related to decreased muscle strength in patients with knee OA. Although the mechanism to explain this relationship is not clear yet, targeting inflammation might have the potential to control the loss of muscle strength in this group of patients. References: (1) Levinger I, Levinger P, Trenerry MK, Feller JA, Bartlett JR, Bergman N et al. Increased inflammatory cytokine expression in the vastus lateralis of patients with knee osteoarthritis. Arthritis Rheum 2011; 63(5):1343–1348. (2) Sanchez-Ramirez DC, van der LM, van der EM, Gerritsen M, Roorda LD, Verschueren S et al. Association of serum C-reactive protein and erythrocyte sedimentation rate with muscle strength in patients with knee osteoarthritis. Rheumatology (Oxford) 2013; 52(4):727–732. Disclosure: D. C. Sanchez-Ramirez, None; M. van der Leeden, None; M. van der Esch, None; L. D. Roorda, None; S. Verschueren, None; J. van Dieen, None; J. Dekker, None; W. F. Lems, None.

0.4017 0.5330 0.8181 0.2485 0.1563 0.7190 0.0307

246

*IRR: Incident rate ratio, per unit of measurement - Poisson analysis; 95% CL: 95% confidence limits.

Conclusion:This work suggests to some degree that there may be a dose-response association between OA (using extent of regional symptomatic joint involvement as an indicator) and serum levels of adipokines. However, these associations differ between women and men. This suggests the possibility of distinct OA phenotypes, and supports potential systemic effects in OA. OA as a systemic disorder entails a need to reevaluate OA treatment and management strategies, with consideration for multimodal approaches. Systemic factors in OA may also help explain high levels of comorbidity in these populations and may be a target for comorbidity prevention. Disclosure: A. V. Perruccio, None; V. Chandran, None; N. N. Mahomed, None; R. Gandhi, None.

245 C-Reactive Protein But Not Erythrocyte Sedimentation Rate Is Associated With Decrease In Muscle Strength In Patients With Knee Osteoarthritis: A 2 Year Follow Up Study. Diana C Sanchez-Ramirez1, Marike van der Leeden2, Martin van der Esch2, Leo D. Roorda3, Sabine Verschueren4, Jaap van Dieen5, Joost Dekker6 and Willem F. Lems6. 1VU university Amsterdam, Amsterdam, Netherlands, 2Reade, Amsterdam, Netherlands, 3 Reade, centre for rehabilitation and rheumatology, Amsterdam, Netherlands, 4 KU University of Leuven, Leuven, Belgium, 5VU University, Amsterdam, Netherlands, 6VU University Medical Center, Amsterdam, Netherlands Background/Purpose: In patients with knee osteoarthritis activity limitations and disease progression have been associated with muscle weakness (OA). Muscle weakness might be determined by factors such as increased pain, immobilization, disuse, avoidance of activities and aging. Additionally, recent cross-sectional studies have found an association between elevated inflammatory markers and a decrease in knee muscle strength in this group of patients (1;2). However, longitudinal studies which assess the association between inflammatory markers and muscle strength in patients with OA are still needed. Therefore, the aim of the study was to examine the associations of serum c-reactive protein (CRP) and erythrocyte sedimentation rate (ESR) with changes in muscle strength in patients with knee osteoarthritis (OA) after two years. Methods: Data from 186 patients with knee OA part of the Amsterdam Osteoarthritis cohort (AMS-OA) were gathered at baseline and at two years follow up. CRP (mg/l) and ESR (mm/l) were measured in serum from patients’ blood. Strength of quadriceps and hamstrings muscles was assessed using an isokinetic dynamometer. Univariable and multivariable linear

Bone Mineral Density and the Risk of Knee Osteoarthritis: The Johnston County Osteoarthritis Project. Kamil E. Barbour1, Jennifer M. Hootman2, Charles G. Helmick1, Louise Murphy3, Jordan B. Renner4 and Joanne M. Jordan5. 1Centers for Disease Control and Prevention, Atlanta, GA, 2Centers for Disease Control and Prevention, Kennesaw, GA, 3CDC, Atlanta, GA, 4 University of North Carolina at Chapel Hill, Chapel Hill, NC, 5University of North Carolina Thurston Arthritis Research Center, Chapel Hill, NC Background/Purpose: There is credible evidence that high bone mineral density (BMD) is associated with an increased risk of incident radiographic osteoarthritis (ROA) of the knee. However, less is known about the relationship of BMD and the outcome with greater clinical and public health relevance, incident symptomatic ROA (sROA). Methods: Using data (N⫽951) from the Johnston County Osteoarthritis Project’s first (1999–2004) and second follow-up (2005–2010), we examined the association between BMD and both incident knee ROA and sROA among participants aged ⱖ45 years. Total hip BMD at baseline was measured using dual-energy X-ray absorptiometry. Participants were grouped into sexspecific BMD quartiles because of large sex-specific differences in BMD. Incident knee ROA was defined as development of a Kellgren-Lawrence grade of ⱖ2 in a knee at second follow-up. Incident knee sROA was defined as onset of both ROA and symptoms in at least one knee at second follow-up. Weibull regression modeling, which accounted for interval censored data, was used to estimate hazard ratios (HR) and 95% confidence intervals (95% CIs). Multivariate models adjusted for age, BMI, sex, race, education, smoking, physical activity, and history of knee injury. Results: Median follow-up time was 6.8 (range⫽4.0–10.2) years. Compared with participants in the lowest BMD quartile, the multivariable adjusted HRs (95% CIs) of sROA for participants in the second, third, and highest quartiles of total hip BMD were 1.4 (0.9 to 2.4), 1.7 (1.1 to 2.7), and 1.6 (1.02 to 2.5), respectively, p trend⫽0.03. Risk of sROA risk did not vary by total hip BMD quartiles, nor was the test of trend significant across BMD quartiles (p trend⫽0.23). Conclusion: Although high levels of BMD may significantly increase one’s risk of knee ROA, we found no evidence of an association between BMD and the more clinically relevant outcome of knee sROA. These findings suggest that adults can achieve and maintain a healthy BMD without the tradeoff of increasing their risk of the painful and potentially debilitating outcome of sROA. Disclosure: K. E. Barbour, None; J. M. Hootman, None; C. G. Helmick, None; L. Murphy, None; J. B. Renner, None; J. M. Jordan, Trinity Partners, Inc., 5, Osteoarthritis Research Society International, 6, Chronic Osteoarthritis Management Initiative of US Bone and Joint Initiative, 6, Samumed, 5, Interleukin Genetics, Inc., 5, Algynomics, Inc., 1.

S101

Sunday, October 27

9). Analysis was stratified by sex, and Poisson analyses used to investigate associations. Results:Bivariate associations suggested opposite effects of leptin, adiponectin, and adipsin between women and men. Adjusted for age, BMI, and hip/knee group membership, higher levels of leptin and adiponectin, and lower levels of adipsin (all p⬍0.01) were associated with a greater regional joint count among women (Table 1). However, among men only resistin was significantly (p⫽0.03) associated in adjusted analyses, with higher levels of resistin associated with lower regional joint counts.

Sunday, October 27

247 Associations Between Levels Of Urinary C-Telopeptide Fragments Of Type II Collagen and Knee Structure In Asymptomatic Middle-Aged Women. Binghui Wang1, Hans Kurniawan Pramono1, Flavia Cicuttini2, Anita Wluka1, Fahad Hanna1, Susan Davis1, Robin Bell1, Andrew Teichtahl1 and Yuanyuan Wang1. 1Monash University, Melbourne, Australia, 2Department of Epidemiology and Preventive Medicine, Monash University, Melbourne, 3004, Australia Background/Purpose: There is evidence for an association between levels of urinary C-telopeptide fragments of type II collagen (uCTX-II) and the risk of knee osteoarthritis. The aim of this study was to examine the association between uCTX-II levels and knee cartilage and bone changes in a cohort of asymptomatic women. Methods: 140 women, aged 40–67 years with no history of knee pain, injury or clinical knee disease, underwent knee MRI at baseline and two years later. Cartilage volume, cartilage defects, tibial plateau bone area and bone marrow lesions (BMLs) were measured using validated methods. Baseline uCTX-II concentrations were determined using enzyme-linked immunosorbent assay (ELISA). Results: Increased uCTX-II level was associated with increased prevalence of medial tibiofemoral cartilage defects (OR 4.5, 95% CI 1.6–12.5, p ⫽ 0.004) and lateral tibiofemoral BMLs (OR 10.6, 95% CI 1.8–61.8, p ⫽ 0.01), greater medial (regression coefficient 80.2 mm2, 95% CI 9.3–151.1, p ⫽ 0.03) and lateral (regression coefficient 86.0 mm2, 95% CI 33.3– 138.7, p ⫽ 0.02) tibial plateau bone area. No significant association was found for tibial cartilage volume. Baseline uCTX-II levels were not associated with the changes in cartilage and bone over two years. Conclusion: In asymptomatic women, increased uCTX-II levels were associated with adverse knee structural changes, as evidenced by increased prevalence of cartilage defects and BMLs, and bigger tibial plateau bone area. These data suggest that uCTX-II may be a sensitive biomarker for early knee cartilage changes. Longitudinal studies with larger sample size and longer follow-up are needed. Disclosure: B. Wang, None; H. Kurniawan Pramono, None; F. Cicuttini, None; A. Wluka, None; F. Hanna, None; S. Davis, None; R. Bell, None; A. Teichtahl, None; Y. Wang, None.

248 Hand Osteoarthritis Is Associated With Increased Type II Collagen Degradation In Women: The Ofely Study. Jean-Charles Rousseau1, Elisabeth Sornay-Rendu1, Cindy Bertholon1, Patrick Garnero2 and Roland Chapurlat3. 1INSERM UMR 1033, Lyon, France, 2INSERM, UMR 1033, Lyon and Cisbio Bioassays, Bagnols/Ce`ze, France, 3INSERM UMR 1033 and Universite´ de Lyon, Hoˆpital Edouard Herriot, Lyon, France Background/Purpose: Hand osteoarthritis (OA) is one of the most common sites of OA and predominantly affects women. Patients with knee, hip or spine OA exhibit increased cartilage type II collagen degradation as detected by the urinary excretion of C-terminal crosslinking telopeptide of type II collagen (CTX-II), but data on hand OA is lacking. The aim of this study was to investigate the relationship between urinary CTX-II and hand OA in women. Methods: We investigated 590 women from the OFELY populationbased study (mean age: 61.8 years ⫾ 10.2), including 475 postmenopausal women. Clinical hand OA was defined according to the ACR criteria (Altman et al, 1990), slightly modified (without functional complains). At the same time of hand OA evaluation, knee and spine OA were assessed by radiographs and self-reported hip OA was recorded. Levels of urinary CTX-II measured by ELISA (Urine CartiLaps®, IDS) in the 186 women with hand OA (mean age: 67.2 years ⫾ 8.4) were compared to those of the 404 other women without hand OA (mean age: 59.4 years ⫾ 10.1). All analyses were adjusted for age and concomitant knee, hip or spine OA. Results: Urinary CTX-II levels were significantly increased in women with hand OA (237 ⫾134 ng/mmolCr) compared to controls (165 ⫾ 97 ng/mmolCr, p ⫽ 0.001) after adjustment for age and for prevalent knee, spine and hip OA. When urinary CTX-II concentrations were considered in quartiles, subjects with levels in the highest quartile had an increased risk of prevalent hand OA, with an odds-ratio of 2.06 (95% CI: 1.3–3.2; p ⫽ 0.002), after adjustment for OA at the other anatomical sites.

Conclusion: Hand OA is characterized by increased type II collagen degradation. Urinary CTX-II could be a useful biomarker for the clinical investigation of hand OA. Disclosure: J. C. Rousseau, None; E. Sornay-Rendu, None; C. Bertholon, None; P. Garnero, None; R. Chapurlat, None.

249 Biomarkers Reflect Differences In Osteoarthritis Phenotypes Of The Lumbar Spine: The Johnston County Osteoarthritis Project. Adam P. Goode1, Amanda E. Nelson2, Virginia B. Kraus3, Jordan B. Renner4 and Joanne M. Jordan2. 1Duke University, Durham, NC, 2University of North Carolina Thurston Arthritis Research Center, Chapel Hill, NC, 3Duke University Medical Center, Durham, NC, 4University of North Carolina at Chapel Hill, Chapel Hill, NC Background/Purpose: We have previously identified differences in associations between joint metabolism biomarkers and radiographic features of the lumbar spine (disc space narrowing (DSN) and osteophytes (OST)). Due to anatomical differences, there may be distinct associations between biomarkers and different phenotypes of osteoarthritis (OA) in the spine. Therefore, our aim was to determine if differences exist in the associations between biomarkers and OA in the spine, defined as 1) facet osteoarthritis (FOA) and 2) the combination of DSN and OST at the same level; we further explored variations by gender, race and low back symptoms. Methods: Of the 1,015 participants enrolled in the Johnston County OA Project from 2003–04, 547 participants had complete data for serum hyaluronan (sHA), serum Cartilage Oligometric Matrix Protein (sCOMP), serum collagen neoepitope (sC2C), serum C-propeptide (sCPII), and urinary N-terminal telopeptide (uNTX-I); 529 participants had complete data for urinary cross-linked C-telopeptide of type II collagen (uCTX-II). The mean age and body mass index (BMI) were 62 (SD 10) years and 30 (SD 6) kg/m2, respectively; 62% were female, 38% were African American (AA), 29% had knee OA, 24% hip OA, 28% hand OA and 49% low back symptoms. Each lumbar spine level was graded for OST and DSN in a semi-quantitative fashion (0–3) and FOA was graded as present or absent according to the Burnett Atlas. Spine OA was defined as the presence of DSN and OST grade 1 or more at the same lumbar level. Biomarkers were natural log (ln) transformed. Linear regression was used to determine adjusted geometric mean values of biomarkers and binary logistic regression models were used to estimate adjusted odds ratios (aOR) and 95% confidence intervals (CI) between biomarkers and spine radiographic variables. All analyses were adjusted for demographic (age, race, gender), clinical (BMI) and radiographic (knee, hip, hand and spine OA or FOA) variables. Interactions were tested with likelihood ratio tests at a p-value⬍0.05. Results: FOA was present at least at one level of the lumbar spine in 57% of participants, while spine OA was present in 49%. Geometric mean biomarker levels of lnHA were significantly (p⫽0.04) greater in the presence of FOA while lnCTX-II levels were significantly (p⬍0.01) greater with in presence of spine OA. Significant associations were found between lnHA and FOA (aOR⫽1.26 ((95% CI 1.02, 1.55)) whereas lnCTX-II was significantly associated with spine OA (aOR⫽1.64 ((95% CI 1.22, 2.21)). The association between lnHA and FOA was significantly (interaction p⬍0.01) greater among Caucasians (aOR⫽1.56 ((95% CI 1.18, 2.05)) than with AA (aOR⫽0.92 ((95% CI 0.68, 1.25)). The association between lnCOMP and spine OA was significantly greater (interaction p⫽0.03) among those with low back symptoms (aOR⫽2.14 ((95% CI 1.08, 4.28)), than those without (aOR⫽0.76 ((95% CI 0.51, 1.68)). Conclusion: Joint metabolism biomarkers reflect the anatomical (HA with FOA and CTX-II with spine OA), demographic (HA with race and FOA) and clinical (COMP with spine OA and symptoms) differences between FOA and a composite definition of spine OA. This may indicate a different pathophysiologic process exists for these two phenotypes of OA in the lumbar spine. Disclosure: A. P. Goode, None; A. E. Nelson, None; V. B. Kraus, None; J. B. Renner, None; J. M. Jordan, Trinity Partners, Inc., 5, Osteoarthritis Research Society International, 6, Chronic Osteoarthritis Management Initiative of US Bone and Joint Initiative, 6, Samumed, 5, Interleukin Genetics, Inc., 5, Algynomics, Inc., 1.

S102

250

Background/Purpose: In patients with knee osteoarthritis (OA), synovitis is associated with severity of knee symptoms. Previously, we demonstrated that in patients undergoing partial meniscectomy without radiographic OA, synovitis is associated with worse preoperative knee pain and function using the Lysholm score, which measures knee-specific pain and disability on a single scale. mRNA expression levels of the chemokine CCL19 and its receptor CCR7 were associated with presence of synovitis, and strongly correlated with worse Lysholm scores. However, unlike the patients in our previous study, most patients with meniscal tears indicated for arthroscopy already have pre-existing cartilage degeneration indicative of early-stage OA. In this study we sought to validate these markers in more typical patients presenting for arthroscopy to determine their utility as markers of more symptomatic early knee disease. To measure knee specific pain and functional deficits we used the Knee Injury and Osteoarthritis Outcome score (KOOS). Methods: Synovial biopsies were collected during surgery from 19 patients undergoing arthroscopic meniscal procedures. Relative mRNA expression levels (RE) of CCL19 and CCR7 were measured using Real Time polymerase chain reaction. The KOOS was administered preoperatively and measures knee function and disability on 5 separate domains: Pain, symptoms, activities of daily living (ADL), sport and recreation function, kneerelated quality of life (QOL). Pearsons correlations were used to determine relationships between mRNA levels and KOOS scores. Multivariable linear regression models were run to test if associations were independent of age, gender, BMI, cartilage degeneration (measured using Outerbridge Classification) and radiographic OA (Kellgren-Lawrence (K/L)) scores. Post-hoc power analysis showed our study was powered at 73% to detect a moderate correlation (␳ ⫽ 0.5) between chemokine levels and KOOS scores. Results: The majority of patients had grade 2–4 Outerbridge scores and median K/L scores of 2, indicative of pre-existing OA. CCL19 and CCR7 transcripts were detected in all patients. Unadjusted analysis revealed CCL19 RE was associated with KOOS ADL (r ⫽ ⫺0.620, p ⫽ 0.005), Pain (r⫽⫺0.547, p⫽0.015), and QOL scores (r⫽⫺0.479, 0.038). Adjusted analyses showed CCL19 RE was independently associated with KOOS ADL scores (␤ ⫽ ⫺4.201, 95% CI [⫺8.071, ⫺0.331], p ⫽ 0.036). Trends were observed for associations with KOOS Pain (CCL19 ␤ ⫽ ⫺3.252, 95% CI [⫺6.748, 0.243], p ⫽ 0.065) and KOOS QOL CCL19 ␤ ⫽ ⫺3.719, 95% CI [⫺7.786, 0.349], p ⫽ 0.070). Conclusion: CCL19 is important in lymphocyte recruitment and was identified in our previous work as a marker of synovitis. Our current work extends these findings by demonstrating a relationship between CCL19 expression and knee-related difficulty with activities of daily living in typical patients presenting for arthroscopic meniscal surgeries. These findings increase the potential for this marker to be useful in identifying patients with synovitis after a meniscal tear and with early-stage OA. Future work is needed to determine the role of this marker in development and pathogenesis of symptomatic OA after meniscal injury. Disclosure: A. Nair, None; C. Bush-Joseph, None; N. Verma, None; M. Tetreault, None; K. Saha, None; A. Margulis, None; L. F. Fogg, None; C. R. Scanzello, N/A, 9.

251 Physical Activity Is Associated With Reduced Incident Disability: Evidence From The Osteoarthritis Initiative. Dorothy D. Dunlop1, Jing Song1, Pamela A. Semanik1, Leena Sharma1, Joan M. Bathon2, Charles Eaton3, Marc C. Hochberg4, Rebecca D. Jackson5, C. Kent Kwoh6, W, Jerry Mysiw5, Michael C. Nevitt7 and Rowland W. Chang8. 1Northwestern University, Chicago, IL, 2Columbia University, New York, NY, 3Brown University, Providence, RI, 4University of Maryland, Baltimore, MD, 5Ohio State University, Columbus, OH, 6University of Pittsburgh, Pittsburgh, PA, 7 University of California, San Francisco, San Francisco, CA, 8Northwestern University Medical School, Chicago, IL Background/Purpose: Over 56 million people in the U.S. are classified as disabled. Physical activity is a low cost, broadly applicable approach to improve cardiovascular fitness, but it is not established if improved fitness translates to reduced disability, a key to containing healthcare costs. This

Disclosure: D. D. Dunlop, None; J. Song, None; P. A. Semanik, None; L. Sharma, None; J. M. Bathon, None; C. Eaton, None; M. C. Hochberg, Abbvie, Inc, 5, Bioiberica SA, 5, Eli Lilly and Company, 5, Genentech, Inc, 5, Iroko Pharmaceuticals, LLC, 5, Merck, Inc, 5, Novartis Pharma AG, 5, Pfizer, Inc, 5, Savient Pharmaceuticals, 5; R. D. Jackson, NIH, 2; C. K. Kwoh, None; W. J. Mysiw, NIH, 2; M. C. Nevitt, None; R. W. Chang, None.

252 The Association Of Knee Buckling With Vibratory Perception and Muscle Strength: The Multicenter Osteoarthritis Study. Najia Shakoor1, David T. Felson2, Jingbo Niu2, Neil A. Segal3, Uyen Sa D.T. Nguyen2, Jasvinder A. Singh4 and Michael C. Nevitt5. 1Rush University Medical Center, Chicago, IL, 2Boston University School of Medicine, Boston, MA, 3 University of Iowa, Iowa City, IA, 4University of Alabama, Tuscaloosa, AL, 5 University of California, San Francisco, San Francisco, CA Background/Purpose: Knee buckling (“giving way”) is a common symptom in knee osteoarthritis (OA). Yet, little is known regarding risk

S103

Sunday, October 27

Synovial Inflammation In Meniscal Tear Patients: CCL19 mRNA Expression Is Independently Associated With Knee Related Disability. Anjali Nair, Charles Bush-Joseph, Nikhil Verma, Matthew Tetreault, Kanta Saha, Arkady Margulis, Louis F. Fogg and Carla R. Scanzello. Rush University Medical Center, Chicago, IL

study evaluated whether accelerometer measured physical activity is related to incident disability. Methods: We prospectively examined the relationship of accelerometer measured physical activity with two-year development of disability measured by instrumental and basic activity of daily living limitations (IADL/ADL). We studied 1666 participants aged 49 years or older from the Osteoarthritis Initiative observational study of adults at high risk for or having knee OA and were free of baseline IADL/ADL limitations. This sample represents a large segment of community-dwelling adults at elevated risk for disability. Hazard ratios were estimated from discrete survival methods to investigate a graded relationship between baseline physical activity quartile groups and the development of disability within two years. Results: Higher physical activity levels had a significant inverse association with the development of disability. For higher active physical activity quartiles, hazard ratios for incident disability were 1.00, 0.67, 0.51, and 0.36, respectively (P for trend, ⬍0.001) controlling for socioeconomic (age, gender, race/ethnicity, education, income) and health factors (comorbidity, depressive symptoms, obesity, smoking, lower extremity pain, function, and knee factors: osteoarthritis severity, pain, symptoms, prior injury). Consistent trends between greater physical activity and less incident disability were demonstrated from analyses stratified by age, gender, and presence/absence of knee osteoarthritis and analyses based on time spent in moderate-intensity or light-intensity physical activities. Sensitivity analyses applying a more stringent definition of disability based on reported IADL/ADL dependency/ assistance also demonstrated a statistically significant inverse graded relationships. Conclusion: These prospective data demonstrated an inverse relationship between higher physical activity levels and the development of disability in community-dwelling adults at elevated risk for disability. Importantly, this study demonstrated greater physical activity, regardless of intensity, was related to less incident disability. Reduced incident disability was associated with greater time spent in light intensity as well as moderate-to-vigorous intensity activities. These findings support policies that encourage adults, including persons with arthritis, to increase their physical activity of any intensity to prevent and slow disability.

Sunday, October 27

factors for buckling. Knee buckling has been found to be associated with muscle weakness, a treatable cause, in one study and this finding needs to be confirmed. In addition, vibratory deficits have been shown to be present in knee OA, and may create alterations in knee position, increasing vulnerability to buckling but this has not been examined. Here, we evaluate the association of muscle strength and vibratory perception with buckling in older adults with, or at high risk for, knee OA. Methods: MOST is a NIH-funded longitudinal study of persons with symptomatic knee OA or at increased risk of OA. At the 60-month visit, participants underwent evaluation of isokinetic muscle strength at the right leg, bilateral evaluation of vibratory perception threshold (VPT) at predetermined anatomic sites, and were asked about buckling and its frequency in the past 3 months. Knee buckling was defined in each subject as (a) any buckling and (b) repeated buckling (ⱖ 2 episodes). VPT was evaluated using a biothesiometer. The applicator tip of the instrument was placed on preselected anatomic bony prominences and the voltage increased by 1 volt/sec until the participant acknowledged sensation. Mean VPT between the limbs was used for analyses. Quadriceps strength was measured as the maximum torque during active isokinetic extension using a dynamometer and scaled to body size by dividing the maximum torque by BMI. VPT and strength were categorized into groups based on ⫾1 SD of the gender-specific mean of the sample. A person-based analyses using logistic regression to estimate adjusted odds ratios for the association of VPT and strength with buckling in the past 3 months was performed. We adjusted in analyses for age, sex, BMI, race, clinic site, and WOMAC knee pain. Results: We evaluated 2,291 subjects (60% women, age⫾SD⫽68⫾8 yrs, BMI⫾SD⫽31⫾6 kg/m2). 16.7% of participants reported buckling and 13.4% reported repeated buckling in the past 3 months. Results are summarized in the Table. A borderline association was found between muscle strength and any buckling episode, however, a significant association was observed with repeated buckling. High quadriceps strength was associated with a significantly decreased odd of repeated buckling (Adj OR: 0.59[0.36,0.99]. There was no significant association found between VPT and buckling. Table Adjusted OR (95% CI)* Any buckling VPT at tibial tuberosity

Low Normal High Quadriceps Low muscle Normal strength High Repeated buckling VPT at tibial Low tuberosity Normal High Quadriceps Low muscle Normal strength High

p value

1.20 (0.88, 1.62) 1 (referent) 1.15 (0.88, 1.49) 1.47 (1.16, 1.87) 1 (referent) 0.51 (0.34, 0.75)

0.244


0.067

0.301 0.002 0.001

0.180 ⬍0.001 ⬍0.001

Adjusted OR (95% CI)**

p value


0.441


0.225

0.312 0.787 0.087

dynamic muscular force and the velocity of muscle contraction, declines earlier and more precipitously than muscle strength across the adult life span. In older populations, peak muscle power is also a more robust predictor of functional outcomes compared to muscle strength, particularly when assessed during closed chain testing. Despite its clinical relevance as a potential determinant of disease burden in knee osteoarthritis (OA), no study to date has comprehensively evaluated lower extremity muscle power output in patients with knee OA. The purpose of this investigation was to examine the relationships between lower extremity muscle power and perceived disease severity in a large population of patients with symptomatic knee OA. We hypothesized that, compared to muscle strength, peak muscle power assessed during bilateral leg press exercise would be a more influential determinant of self-reported pain and health-related quality of life within this patient population. Methods: We used baseline data collected as part of a 4-year randomized controlled clinical trial in 181 patients (mean age: 60.2 ⫾ 10yrs, BMI: 32.7 ⫾ 10 kg/m2, 69% female, 52% Caucasian) who met the American College of Rheumatology criteria for knee OA. Pain was measured using the pain subscale of the Western Ontario and McMaster Osteoarthritis Index (WOMAC) and health-related quality of life was measured using the Medical Outcomes Study Short Form 36 physical component summary and mental component summary. One-repetition maximum (1RM) strength was measured using the bilateral leg press exercise and peak leg press muscle power was measured during 5 repetitions performed as fast as possible with resistance set to 40% of the 1RM (Keiser Pneumatic Leg Press A420, Keiser Corporation, Fresno, CA). Results: In univariate regression analysis, greater peak muscle power output was significantly and inversely associated with lower WOMAC pain score (r ⫽ ⫺0.17, P ⫽ 0.02). Peak muscle power was also significantly and positively associated with SF36 physical component summary score (r ⫽ 0.15, P ⫽ 0.05) but not mental component summary score (r ⫽ ⫺0.02, P ⫽ 0.8). After adjusting for age, BMI, sex, race and depressive symptoms, peak muscle power was a significant and independent predictor of SF 36 physical component summary score (P ⫽ 0.02). Muscle strength was not associated with any measure of disease severity or quality of life (P ⱖ 0.07). Conclusion: Leg press muscle power output is a significant determinant of perceived disease burden in patients with symptomatic knee OA. While the overall magnitude of these relationships are modest, peak muscle power exerts a greater influence on self reported pain and quality of life compared to muscle strength within this patient population. These data suggest that the efficacy of resistance training interventions specifically designed to improve leg extensor muscle power output should be examined in patients with knee OA. Supported by the National Center for Complimentary and Alternative Medicine (R01 AT005521-01A1) Disclosure: K. F. Reid, None; L. L. Price, None; W. F. Harvey, Vindico Medical Education, 5; J. B. Driban, None; R. A. Fielding, None; C. Wang, NIH, 2.

0.204 0.568

254

0.041

*VPT adjusted for muscle strength and muscle strength adjusted for VPT **adjusted for all covariates

Conclusion: In this large cohort of participants with knee OA or at high risk for knee OA, greater quadriceps strength was found to be associated with decreased odds of repeated knee buckling. There was no significant association found between vibratory perception and knee buckling. Future studies should continue to evaluate for other potential risk factors for buckling in knee OA. Disclosure: N. Shakoor, None; D. T. Felson, None; J. Niu, None; N. A. Segal, None; U. S. D. T. Nguyen, None; J. A. Singh, Takeda, Savient, 2, Savient, Takeda, Ardea, Regeneron, Allergan, 5, URL pharmaceuicals Novartis, 5; M. C. Nevitt, None.

253 Higher Leg Extensor Muscle Power Output Is Associated With Reduced Pain and Better Quality Of Life In Patients With Symptomatic Knee Osteoarthritis. Kieran F. Reid1, Lori Lyn Price2, William F. Harvey2, Jeffrey B. Driban2, Roger A. Fielding1 and Chenchen Wang2. 1Tufts University, Boston, MA, 2Tufts Medical Center, Boston, MA Background/Purpose: Muscle strength, the maximal force generating capacity of skeletal muscle, has been widely characterized in patients with knee osteoarthritis. However, skeletal muscle power, defined as the product of

The Relationship Of Foot Pronation During Walking To Risk Of Incident Medial Tibiofemoral and Lateral Patellofemoral Cartilage Damage: The Multicenter Osteoarthritis Study. K. Douglas Gross1, Howard J. Hillstrom2, Yuqing Zhang3, Emily K. Quinn1, Michael C. Nevitt4, Neil A. Segal5, Cora E. Lewis6 and David T. Felson3. 1Boston University, Boston, MA, 2Hospital Special Surgery (HSS), New York, NY, 3Boston University School of Medicine, Boston, MA, 4University of California, San Francisco, San Francisco, CA, 5University of Iowa, Iowa City, IA, 6University of Alabama, Birmingham, Birmingham, AL Background/Purpose: Studies of older adults suggest an association between plantar pressure measures of foot pronation and medial tibiofemoral (med TF) cartilage damage, while studies of younger adults suggest an association with lateral patellofemoral (lat PF) pain. These were crosssectional studies, however. We assessed the longitudinal relationship of foot pronation in walking with 2-year risk of incident med TF and lat PF cartilage damage in at-risk older adults. Methods: The Multicenter Osteoarthritis Study (MOST) includes adults aged 50–79 years that have or are at risk of knee OA. At the 60 month exam, plantar pressure was assessed using an EmedX to quantify Center of Pressure Excursion Index (CPEI) during 5 trials of self-paced walking (14-day retest ICC ⬎ 0.82). Smaller CPEI values indicate increased pronation. From 1.0 T MRIs obtained at 60 and 84 months readers scored one knee per subject for cartilage damage in each sub-region comprising the med TF and lat PF compartments using WORMS grades (weighted kappa ⬎ 0.63). Among

S104

Table. Relative odds (OR) of incident medial tibiofemoral and lateral patellofemoral cartilage damage in quintiles of increasing foot pronation during walking. CPEI (high) 44.2, 24.9 Incident Medial Tibiofemoral Cartilage Damage # Knees # Sub-regions % Damage Adj OR* (95% CI)

Incident Lateral Patellofemoral Cartilage Damage # Knees # Sub-regions % Damage Adj OR* (95% CI)

243 864 3.8% 1.00 (ref) CPEI (high) 44.2, 25.6

160 259 2.3% 1.00 (ref)

Center of Pressure Excursion Index (CPEI) Increasing Pronation 3 24.8, 20.9 20.8, 17.6 17.5, 13.8

220 756 4.1% 0.96 (0.54, 1.70) 25.5, 21.4

174 266 1.9% 0.82 (0.25, 2.70)

214 190 784 654 3.8% 3.2% 0.93 (0.50, 1.73) 0.78 (0.42, 1.44) Increasing Pronation 3 21.3, 17.9 17.8, 13.8

190 290 1.7% 0.78 (0.24, 2.58)

161 237 2.1% 0.97 (0.29, 3.24)

CPEI (low) 13.7, ⫺7.7

189 682 3.7% 0.81 (0.41, 1.58) CPEI (low) 13.7, ⫺7.7

165 243 4.1% 2.04 (0.72, 5.82)

* Adjusted for age, BMI, sex, and non-independent sub-regions of a knee compartment.

Conclusion: In the first longitudinal data on foot pronation and risk of knee cartilage damage, we cannot confirm cross-sectional findings of an association between excessive pronation and med TF disease. Yet, this data does suggest a possible association between extreme foot pronation and elevated risk of lat PF damage. Studies with higher incidence rates are needed to confirm this. Disclosure: K. D. Gross, None; H. J. Hillstrom, JAK Tool, 5; Y. Zhang, None; E. K. Quinn, None; M. C. Nevitt, None; N. A. Segal, None; C. E. Lewis, None; D. T. Felson, None.

255 History Of Knee Injury Is Weakly Associated With Knee Structural Change In Middle Or Older Aged Adults. Hussain Ijaz Khan1, Dawn Aitken1, Changhai Ding1, Leigh Blizzard1, Jean Pierre Pelletier2, Johanne M. Pelletier3, Flavia Cicuttini4 and Graeme Jones1. 1Menzies Research Institute Tasmania, University of Tasmania, Hobart, 7000, Australia, 2Institut de Rhumatologie de Montreál, Montreal, QC, 3Osteoarthritis Research Unit, University of Montreal Hospital Research Centre (CRCHUM), Notre-Dame Hospital, Montreal, QC, 4Department of Epidemiology and Preventive Medicine, Monash University, Melbourne, 3004, Australia Background/Purpose: History of knee joint injury is a strong risk factor for the development of knee osteoarthritis (OA) but it’s not clear how different structures comprising the knee joint are affected by injury. The aim of this study was to describe the association between history of knee injury and knee structural changes on MRI in middle aged and older adults. Methods: This study included two population based samples. (I)the Offspring Study comprised of middle aged adults (n⫽372; mean age 45.0 years, range 26–61 years; 57.5% female). Approximately half were adult offspring of subjects who had a knee replacement performed for knee OA and the remaining were randomly selected controls that were initially matched by age and sex.(II) the Tasmanian Older Adult Cohort (TASOAC) Study comprised of older adults(n⫽430; mean age 63.0 years, range 51 – 79 years; 51% female) which were selected from the electoral roll using sex-stratified simple random sampling without replacement. 1.5 T MRI scans of the right knee at baseline was performed in both studies to measure bone marrow lesions (BMLs), cartilage volume, tibial bone area, cartilage defects, and

meniscal pathology. History of knee injury was assessed using a selfadministered questionnaire. The association between knee injury and knee structures was determined using multiple linear and log binomial regression models. Results: In middle aged adults, only BML prevalence (Prevalence Ratio (PR)⫽1.6 (1.2, 2.0)) was significantly higher in those with knee injury in adjusted analysis (Table 1). In older adults, cartilage defects (PR⫽1.3 (1.0,1.7), total tibial cartilage volume (Difference of Means (df)⫽ ⫺323 (⫺616,⫺32)) and BMLs (PR⫽1.4 (1.0,1.9)) showed significant associations with knee injury in adjusted analysis. Meniscal tears and extrusions showed no significant associations in either of the two cohorts and tears were very common. Table 1. Association between injury and bone marrow lesions in the knee BMLs Absent/ Present(site) OFFSPRING Medial Tibial Lateral Tibial Medial Femoral Lateral Femoral Total TASOAC Medial Tibial Lateral Tibial Medial Femoral Lateral Femoral Total

Injury

No Injury

%

%

Unadjusted PR (95% CI)

Adjusted* PR (95% CI)

33 (13/39) 31 (12/39) 31 (12/39) 21 (8/39) 72 (28/39)

16 (25/159) 25 (40/161) 11 (17/160) 14 (23/159) 49 (82/158)

2.1 (1.2, 3.8) 1.2 (0.7, 2.1) 2.9 (1.5, 5.6) 1.4 (0.7, 2.9) 1.5 (1.1, 1.9)

1.9 (1.0, 3.4) 1.3 (0.8, 2.3) 2.7 (1.3, 5.3) 1.5 (0.7, 3.2) 1.6 (1.2, 2.0)

33 (15/46) 17 (8/46) 17 (8/46) 30 (14/46) 52 (24/46)

18 (63/346) 15 (51/346) 12 (40/346) 17 (58/346) 41 (143/346)

1.9 (1.1, 3.1) 1.5 (0.8, 3.0) 2.0 (1.0, 3.9) 1.7 (0.9, 3.0) 1.4 (1.1, 1.9)

1.8 (1.1, 3.1) 1.6 (0.8, 3.2) 1.9 (1.0, 3.5) 1.7 (0.9, 3.0) 1.4 (1.0, 1.9)

PR-Prevalence Ratio CI-Confidence Interval *Adjusted for age, sex and bmi

Conclusion: In terms of specific structural abnormalities, history of knee injury is weakly and inconsistently associated with knee structural changes in either middle age or later life suggesting most structural change in the knee is atraumatic. Disclosure: H. I. Khan, None; D. Aitken, None; C. Ding, None; L. Blizzard, None; J. P. Pelletier, None; J. M. Pelletier, ArthroLab, 4; F. Cicuttini, None; G. Jones, None.

256 Classification Of Alignment By Self-Report Versus Radiograph Results In Unexpected and Discrepant Pain and Functional Outcomes. E Megan Erickson, Jeffrey B. Driban, Lori Lyn Price, Chenchen Wang, Timothy E. McAlindon and William F. Harvey. Tufts Medical Center, Boston, MA Background/Purpose: Malalignment is a potential risk factor for structural progression of knee osteoarthritis (OA), but data conflict as to its association with knee pain. We hypothesize that subjective alignment measures are less burdensome and yield results similar to radiographic measures. As obesity is also associated with pain in knee OA, there could be interactions between obesity and alignment when assessing pain and function. Thus, we investigated if pain and function differ between obese and non-obese individuals with and without knee malalignment using both radiographic and self-reported classification of alignment. Methods: We used baseline data derived from two clinical trials for symptomatic knee OA. Both trials used similar definitions of knee OA (ACR radiographic criteria). We assessed pain and function at baseline with the WOMAC and Short Form 36 Health Survey (SF-36) subscales and chair stand time. Participants were asked if they perceived themselves as bowlegged, straight-legged, or knock-kneed. One reader measured anatomic axis from posteroanterior fixed-flexion weight-bearing knee radiographs using a validated standard method [ICC (3, 1) ⫽ .94]. We adjusted anatomic axis measures (female, ⫺3.5°; male, ⫺6.4°) to better reflect mechanical axis and categorized them using typical definitions (see table). To analyze the influence of malalignment, obesity, and their interaction, we performed a 2 (obesity) ⫻ 2 (alignment) analysis of variance for each outcome. We conducted analyses using the self-report and radiographic classifications with varus (⬍ 178°) and valgus (⬎ 182°) groups collapsed. Results: Characteristics of our 201 participants, stratified by obesity and radiographic alignment, are shown in the table. Using radiographic classification, malaligned groups had better WOMAC pain and function, SF-36 body pain, and chair stand time, independent of obesity. Using self-report classi-

S105

Sunday, October 27

sub-regions with WORMS ⬍ 2 at 60 months, incident cases had WORMS ⱖ 2 at 84 months. With the highest quintile (least pronated feet) as a reference, separate logistic regression models estimated relative odds of incident med TF and lat PF damage in each case-based CPEI quintile while adjusting for age, BMI, sex, and using GEE to account for non-independent sub-regions of a compartment. Results: 1123 participants (mean age 66.8 ⫾ 7.5 yrs, BMI 29.7 ⫾ 4.8 kg/m2, 62.4% female) contributed one knee each (mean CPEI 19.8 ⫾ 6.6) to the analysis of med TF damage, while 904 participants with similar characteristics contributed to the analysis of lat PF damage. Only 3.6% and 2.6% of sub-regions had incident med TF and lat PF damage, respectively. Although relative odds of lat PF damage were increased (OR ⫽ 2.04; 95% CI: 0.72, 5.82) among the most pronated feet (CPEI 13.7 to ⫺7.7), pronation was not strongly associated with odds of med TF damage (see table). With few incident cases contributing, no findings achieved statistical significance (p ⬎ 0.05).

Sunday, October 27

fication, malaligned groups had worse WOMAC pain and function, SF-36 body pain and physical function, and chair stand time, again independent of obesity, but no outcome reached statistical significance. We found no significant interactions between obesity and either alignment method. Multiple sensitivity analyses using 3 categories of alignment, restricting analyses to women, and examining the trials separately, did not change the results. Table. Study Sample Characteristics

Outcome

Radiographic Alignment Non-obese (< 30 kg/m2) Obese (> 30 kg/m2) p-value Neutrally Neutrally aligned Malaligned aligned Malaligned Obesity Alignment Obesity* (178°–182°) (182°) (178°–182°) (182°) Main Main Alignment (n ⴝ 21) (n ⴝ 65) (n ⴝ 32) (n ⴝ 83) Effect Effect Interaction

Female (n, %) 18 (85.7) Caucasian (n, %) 13 (61.9) K/L Grade ⬎ 2 (n, %) 6 (28.6) Age (years) 59.7 2 BMI (kg/m ) 26.3 WOMAC pain (VAS, 0–500)* 223.7 WOMAC Function (VAS, 0– 818.4 1700)* SF-36 Body Pain (0–100)** 46.4 SF-36 Physical Functioning 52.6 (0–100)** Chair Stand Time (secs) 21.6

site. Beta estimates and 95% confidence intervals report the amount of standard deviation increase or decrease in mode shape among women compared to men. Results: The mean age of subjects in the analysis was 52.7 years (⫾4.3 SD) for both men and women. There were 191 female knees and 149 male knees. 10 modes were derived for tibial shape. The 10 modes described 95.5% of the total variance in proximal tibia shape in the population sampled. Mode 2 had the highest significance for association with sex (p⫽0.009). (See table for beta estimates and description of first 5 modes.) Mode

Adjusted Beta (95%CI)

P-value

1

⫺0.2 (⫺0.42 to 0.02)

0.0797

2

⫺0.29 (⫺0.51 to ⫺0.07)

0.0091

0.4 0.4

3

⫺0.22 (⫺0.43 to ⫺0.01)

0.0376

0.2 0.1

4

⫺0.23 (⫺0.45 to ⫺0.02)

0.0324

5

⫺0.01 (⫺0.23 to 0.2)

0.9097

24 (36.9) 42 (64.6) 35 (53.8) 60.1 26.6 205.3 707.8

27 (84.4) 13 (40.6) 8 (25.0) 57.8 36.1 264.8 968.6

53 (63.9) 46 (55.4) 59 (71.1) 58.3 35.3 223.5 784.8

0.003 0.09 0.1 0.1

⬍ 0.0001 0.1 ⬍ 0.0001 0.7

0.99

0.05 0.02

0.02 0.001

49.9 52.0

37.2 36.1

47.7 47.0

0.1 0.02

0.009 0.09

18.1 27.5 23.2 0.002 0.04 0.8 Self-Reported Alignment 2 2 Outcome Non-obese (< 30 kg/m ) Obese (> 30 kg/m ) p-value Neutrally Malaligned Neutrally Malaligned Obesity Alignment Obesity* aligned (182°) aligned (182°) Main Main Alignment (178°–182°) (n ⴝ 29) (178°–182°) (n ⴝ 35) Effect Effect Interaction (n ⴝ 57) (n ⴝ 80) Female (n, %) 26 (45.6) 16 (55.2) 57 (71.3) 23 (65.7) 0.003 ⬍ 0.0001 Caucasian (n, %) 36 (63.2) 19 (65.5) 45 (56.3) 14 (40.0) 0.09 0.4 K/L Grade ⬎ 2 (n, %) 27 (47.4) 14 (48.3) 42 (52.5) 25 (71.4) 0.1 0.1 Age (years) 59.3 61.3 58.8 56.6 0.1 0.8 0.1 2 BMI (kg/m ) 26.4 26.8 35.1 36.5 WOMAC pain (VAS, 0–500)* 200.6 227.7 235.3 234.2 0.04 0.4 0.3 WOMAC Function (VAS, 0– 714.7 774.3 831.2 846.8 0.02 0.5 0.6 1700)* SF-36 Body Pain (0–100)** 52.0 43.3 44.8 44.8 0.1 0.2 0.1 SF-36 Physical Functioning 55.1 46.4 45.4 40.6 0.01 0.08 0.6 (0–100)** Chair Stand Time (secs) 19.06 18.99 23.90 25.27 0.002 0.7 0.7 *Smaller is better. ** Larger is better.

Conclusion: Conventional wisdom suggests that both obesity and malalignment should cause more pain in people with knee OA. However, our results suggest that assessing the exposure of malalignment using radiographs vs. self-report may yield different, counterintuitive, results. Because more people classified themselves as neutrally aligned, the discrepancy could not have been due to a perceived connection between malalignment and pain. Studying these classifications of exposure with longitudinal outcomes will provide further insights into their relative utilities. Disclosure: E. M. Erickson, None; J. B. Driban, None; L. L. Price, None; C. Wang, NIH, 2; T. E. McAlindon, NIH, 2; W. F. Harvey, Vindico Medical Education, 5.

257 The Association Of Proximal Tibia Shape With Sex: The Osteoarthritis Initiative. Barton L. Wise1, Felix Liu2, Neeta Parimi3, John A. Lynch4, Yuqing Zhang5, Lisa Kritikos1 and Nancy E. Lane1. 1UC Davis School of Medicine, Sacramento, CA, 2University of California at San Francisco, San Francisco, CA, 3California Pacific Medical Center Research Institute, San Francisco, CA, 4University of California-San Francisco, San Francisco, CA, 5 Boston University School of Medicine, Boston, MA Background/Purpose: Knee osteoarthritis (OA) manifests disproportionately in women and the etiology remains unclear. Bone shape has been found to be associated the development of knee OA in prior studies. To further explore this relationship, we examined the association of proximal tibia shape with sex to begin to understand these relationships. Methods: We used information from the NIH-funded Osteoarthritis Initiative (OAI), a cohort of persons aged 45–79 at baseline who either had symptomatic knee OA or were at high risk of it. We randomly sampled knees from women and men from the OAI cohort who had Kellgren/Lawrence grade of 0 in central readings on baseline radiograph who were aged between 45 and 60 years. Using baseline radiographs we characterized proximal tibia shape using Active Shape Modeling to generate the modes of shape difference across our selected population. Given that all modes were normally distributed, we performed linear regression to examine the association of proximal tibia shape with sex, adjusting for age, race, body mass index (BMI) and clinic

Mode Description – primary alteration of shape with increased SD weighting Decreased concavity, elevation of lateral compartment plateau, and decreased shaft width Tibial head shifted laterally in relation to the shaft, and head width increased. The lateral tibial plateau is more concave Slightly increased tibial width, depression of medial plateau and elevation of lateral plateau Increased concavity of medial compartment and elevation of medial lip, with broadening of lateral plateau Narrowed medial plateau width and slight elevation of lateral plateau

Conclusion: Tibial knee shape may be associated with sex. Further work to understand how these shape modes are associated with knee OA, and whether they can explain differences in prevalence of OA by sex, is warranted. Disclosure: B. L. Wise, Pfizer, Inc., 2; F. Liu, None; N. Parimi, None; J. A. Lynch, None; Y. Zhang, None; L. Kritikos, None; N. E. Lane, None.

258 Underdiagnosis and Undertreatment Of Knee Osteoarthritis In The Obese Population: The Need For Physician Education and Advocacy. Janice Lin, Ryan Flanagan, Jay Bhatia, Manish Parikh, Christine RenFielding, Renata La Rocca Vieira, Steven B. Abramson and Jonathan Samuels. NYU Langone Medical Center, New York, NY Background/Purpose: Obesity is a modifiable risk factor of knee osteoarthritis (KOA). While medical treatments for KOA can have a limited effect, an alternative strategy would target weight loss to reduce the arthritis burden. Supportive of other literature, we recently showed in a retrospective analysis that 51% of patients who underwent LapBand surgery reported complete improvement in KOA pain after 18 months. Therefore we began a prospective study to evaluate the prevalence of knee pain and KOA in the obese population, observe how knee pain is treated, and track how bariatric weight loss affects KOA-related pain and physical function. Methods: We screened consecutive patients prior to bariatric surgery with the LapBand, sleeve gastrectomy, or gastric bypass. We enrolled patients (age’21) with knee pain for ⱖ1 month and a visual analog scale (VAS) pain score ⱖ30mm, excluding lupus, inflammatory arthritis, or psoriasis. Treatment history for knee pain was recorded. Baseline assessments included validated questionnaires: Western Ontario McMasters Universities Osteoarthritis Index (WOMAC), Assessment of Obesity-Related Comorbidities, and Knee Injury and Osteoarthritis Outcome Score. Those with radiographic KOA by the Kellgren-Lawrence (KL) grading scale ⬎1 will repeat questionnaires post-surgery. Results: We evaluated 262 patients, with 136 reporting knee pain and 62 consenting for the study (88.7% female, mean BMI 44.5 kg/m2⫾8.3, range: 32.0–60.4, and mean age 44 years⫾10.3, range:22–70). 52% were scheduled for sleeve gastrectomy (mean BMI⫽43.0 kg/m2), 26% for Lapband (BMI⫽43.7), and 22% for gastric bypass (BMI⫽48.8). The mean VAS score was 65.4 (⫾19.1;range:30–100), WOMAC pain 265.1 (⫾101.8;range:13– 466), WOMAC stiffness 99.1 (⫾59.2;range:0–187) and WOMAC physical function 917.1 (⫾427.9;range:0–1589). Baseline radiographs revealed that 96% had evidence of OA (70.6 % KL 2–4 and 25.4% KL1). Despite significant knee pain in this cohort, only 4.8% (3/62) had seen rheumatologists and 17.7% orthopedists – while 55% were treated by primary care and 22.5% had never discussed their knee pain treatment with a physician. Only 37% had taken x-rays previously to evaluate knee pain. Not surprisingly, the ACR OA treatment guidelines were not met in a majority of our cohort: Only 40.3% had been referred for physical therapy, 80.6% tried acetaminophen, 70.9% NSAIDs, 6.4% narcotics, 1.9% SSRI, 12.9% tramadol, 11.2% topical NSAIDs, 16.1% intra-articular steroids, and 3.2% viscosupplementation.

S106

Disclosure: J. Lin, None; R. Flanagan, None; J. Bhatia, None; M. Parikh, None; C. Ren-Fielding, None; R. La Rocca Vieira, None; S. B. Abramson, None; J. Samuels, None.

259 Examining Patient Reported Outcome Measurement Information System Measures In Community Dwelling Adults With Arthritis. AnaMaria Orbai1, Leigh F. Callahan2, Rebecca J. Cleveland3, Sharon R. Ghazarian1, Susan J. Bartlett1 and Clifton O. Bingham III1. 1Johns Hopkins University, Baltimore, MD, 2University of North Carolina, Chapel Hill, NC, 3 University of North Carolina at Chapel Hill, Chapel Hill, NC Background/Purpose: Arthritis is the leading cause of disability in the US and impacts on multiple aspects of health-related quality of life (HRQoL). Many multidimensional HRQoL measures used in arthritis have complex scoring algorithms and are sometimes proprietary limiting their utility in clinical care and research. The Patient Reported Outcome Measurement Information System (PROMIS) was developed by NIH using modern psychometrics to provide publicly available, population normalized measures to assess domains of physical, emotional, and social health. Assessments of PROMIS measures in arthritis have been limited to pain, function and fatigue in small numbers of patients. It was our goal to evaluate PROMIS measures in a large group of adults with self-reported arthritis. Methods: Six PROMIS short forms for pain interference (6 items), physical function (10), fatigue (7), sleep quality (8), depression (8), and satisfaction with social roles (7) were administered to community dwelling adults with self-reported arthritis enrolled in the Walk with Ease (WWE) study. Raw scores were transformed to T scores for analysis, with the population mean 50 and standard deviation 10. Other measures included general health; pain, fatigue and stiffness VAS; and the improved HAQ (score range 0 to 100). Analyses were conducted to examine the effect of age and BMI using Stata 12. Results: Instruments were administered to 439 adults with arthritis: 88% female, 70% White, 26% Black or African-American; general health 34% very good and excellent, 48 % good and 18 % fair; mean (SD) age 64 (12) BMI 30 (7); pain VAS 38.1 (25.3) fatigue VAS 37.3 (28.3), stiffness VAS 42.4 (26.7), IHAQ 14 (14). Mean (SD) PROMIS T scores were: pain interference 54.5 (6.9), fatigue 52.3 (8.3), physical function 42.8 (6.5), depression 49 (7.8), sleep quality 50.1 (9.66), satisfaction with social roles 48.6 (9). After adjustment for age, T scores were significantly worse for obese vs. non-obese for pain interference, fatigue, physical function, satisfaction social roles (Table). Table. PROMIS T scores in obese versus non-obese people with self-reported arthritis T scores Mean (SD) BMI30 (nⴝ254) (nⴝ185) PROMIS Pain Interference PROMIS Fatigue PROMIS Physical Function* PROMIS Depression PROMIS Sleep Quality PROMIS Satisfaction with Social Roles

53.4 (7.4) 50.7 (8.3) 43.8 (6.7) 48.4 (7.3) 49.9 (9.7) 50.1 (8.8)

55.9 (6.3) 54.5 (7.8) 41.4 (5.9) 49.8 (8.4) 50.6 (9.4) 46.4 (9.0)

Age adjusted p-value ⬍0.001 ⬍0.001 ⬍0.001 0.1 0.7 ⬍0.001

*Lower scores represent worse physical function

Conclusion: PROMIS measures identify worse pain interference, fatigue, and physical function in people with self-reported arthritis compared to population-based norms. PROMIS social role satisfaction measures indicate arthritis impacts participation in valued life activities. Significantly worse fatigue (0.4 SD) and decreased satisfaction with social roles (0.4 SD) were seen in obese vs. non-obese, after adjustment for age. Studies are ongoing to evaluate responsiveness of PROMIS measures compared with legacy mea-

sures with an exercise intervention in the WWE study. Further studies are needed to evaluate the impact of obesity on other measures of HRQL in individuals with arthritis. Disclosure: A. M. Orbai, None; L. F. Callahan, None; R. J. Cleveland, None; S. R. Ghazarian, None; S. J. Bartlett, None; C. O. Bingham III, None.

260 Pain, Functional Impairment and Ultrasound Detected Changes In Patients With Erosive and Non-Erosive Hand Osteoarthritis. Olga Sleglova1, Olga Ruzickova2, Karel Pavelka2 and Ladislav Senolt3. 1Institute of Rheumatology, Prague, Prague, Czech Republic, 2Institute of Rheumatology and Department of Rheumatology, 1st Medical Faculty, Charles University Prague, Prague, Czech Republic, 3Institute of Rheumatology, Department of Rheumatology, 1st Faculty of Medicine, Charles University, Prague, Czech Republic Background/Purpose: Hand osteoarthritis (HOA) is a common and frequent cause of pain. HOA is a heterogeneous group of disorders with two main subsets including non-erosive disease and erosive, sometimes referred to as inflammatory, HOA. Few studies demonstrated inflammatory ultrasound changes and more severe clinical symptoms in patients with erosive compared with non-erosive disease, however the results are inconclusive. The aim of the study was to compare pain, stiffness, physical impairment and ultrasound features between patients with erosive and non-erosive HOA in a crosssectional study. Methods: Consecutive patients with symptomatic HOA fulfilling the American College of Rheumatology (ACR) criteria were included in this study. Joint tenderness and swelling were assessed. Patients reported joint pain on 100 mm visual analogue scale (VAS) was completed. Pain, joint stiffness and disability were assessed by the Australian/Canadian OA hand index (AUSCAN). Radiographs of both hands were examined and erosive disease was defined by at least one erosive interphalangeal joint. Effusion, synovial hypertrophy and power Doppler signal (PDS) were scored with ultrasound. Synovitis was graded on a scale of 0–3 and osteophytes were defined as cortical protrusions seen in two planes. Results: Altogether, 81 patients (five male) with symptomatic nodal HOA were included in this study between April 2012 and April 2013. Out of these patients, 46 had erosive disease. Patient’s characteristics are given in table 1. The intensity of pain assessment on VAS (p⬍0.05), duration of morning stiffness (p⬍0.05) and number of clinically swollen joints (p⬍0.05) were significantly higher in patients with erosive compared with non-erosive disease. Accordingly, functional impairments assessed by AUSCAN showed more disability in patients with erosive compared with non-erosive disease (p⬍0.05). US-detected pathologies (gray-scale synovitis, power Doppler signal and osteophytes) were common in both groups of patients. Although, synovial hypertrophy was higher in patients with erosive compared with non-erosive disease (total score: 7.7 vs. 3.8; p⬍0.05), the differences in intensity of power Doppler signal and number of osteophytes did not differ between both groups. Table 1.

Age, years (mean ⫾ SD) Female, no. (%) Disease duration, years (mean ⫾ SD) BMI, kg/m2 (mean ⫾ SD) AUSCAN, total (mean ⫾ SD) AUSCAN A, pain (mean ⫾ SD) AUSCAN B, function (mean ⫾ SD) AUSCAN C, stiffness (mean ⫾ SD) VAS, pain (mm) Tender joints, no. Swollen joints, no. NSAIDs, no. (%) SYSADOA, no. (%)

All patients

Non-erosive HOA

Erosive HOA

66.74 ⫾ 8.87 76 (96.30%) 8.78 ⫾ 8.18

64.49 ⫾ 8.21 33 (94.29%) 7.91 ⫾ 8.57

68.46 ⫾ 8.97 43 (93.48%) 9.43 ⫾ 7.81

28.36 ⫾ 5.42 22.94 ⫾ 11.43

28.98 ⫾ 6.06 20.51 ⫾ 10.51

27.88 ⫾ 4.82 24.78 ⫾ 11.53

8.47 ⫾ 4.41

8.03 ⫾ 4.45

8.80 ⫾ 4.36

2.00 ⫾ 0.89

1.83 ⫾ 1.06

2.13 ⫾ 0.71

12.04 ⫾ 6.52

10.37 ⫾ 6.21

13.30 ⫾ 6.46

43.19 ⫾ 24.20 9.49 ⫾ 6.58 3.49 ⫾ 4.28 37 (45.68%) 55 (67.90%)

39.71 ⫾ 24.80 9.26 ⫾ 7.23 2.83 ⫾ 3.52 16 (45.71%) 23 (65.71%)

45.84 ⫾ 23.40 9.67 ⫾ 6.04 4.00 ⫾ 4.71 21 (46.65%) 32 (69.57%)

Conclusion: This study shows that patients with erosive HOA have more hand pain, joint stiffness and functional limitation associated with US-

S107

Sunday, October 27

Conclusion: In this early phase of our prospective study of bariatric patients, we found that moderate radiographic KOA is common in obese patients with knee pain. In many cases, pain had been attributed to mechanical load from obesity without proper evaluation or treatment. Few patients were referred to rheumatologists, though would benefit from such evaluation and management. These data indicate that knee OA in obese patients is underdiagnosed and undertreated. There is a need to educate primary care and bariatric providers that knee pain from OA and other pathology in obese patients should be diagnosed and treated appropriately to maximize their function and quality of life.

Sunday, October 27

detected synovial hypertrophy, but not with inflammatory signs or osteophyte formation. Acknowledgement: This work was supported by the project (Ministry of Health, Czech Republic) for consensual development of research organization 023728. Disclosure: O. Sleglova, None; O. Ruzickova, None; K. Pavelka, None; L. Senolt, None.

261 Concordance Of Hip Pain With Radiographic Hip Osteoarthritis In An Urban US Community: The Framingham Osteoarthritis Study. Kyu Chan Kim1, David T. Felson2, Katherine D. Linsenmeyer1, Ali Guermazi3, Steven C. Vlad3, Mary M. Clancy3 and Jingbo Niu2. 1Boston Medical Center, Boston, MA, 2Boston University School of Medicine, Boston, MA, 3 Boston University, Boston, MA Background/Purpose: While it is well known that knee pain and radiographic knee osteoarthritis (OA) are often discordant, little is known of the concordance of hip pain with the presence of radiographic hip OA. From a prevalence study of hip osteoarthritis done in Framingham, Massachusetts, we examined the relationship between hip pain and radiographic hip OA. Methods: We performed a community based study of OA among persons age 50–79 living in Framingham in 2002–2005 with recruitment by random digit dialing and subjects studied without respect to joint pain or arthritis. Anteroposterior standing long-limb radiographs of the lower extremities including the pelvis were obtained and were read for radiographic hip OA(ROA) by two trained physicians. Cases of ROA were confirmed by an experienced musculoskeletal radiologist. ROA was defined as KellgrenLawrence score ⱖ 2. Using a homunculus in which the hip joint was depicted as a large circle in the groin, participants were asked whether they had hip pain on most days. Those who said ‘yes’ were defined as having hip pain. If they had hip pain, subjects then answered another question asking location of pain: groin, front of the leg (anterior), outside the leg (lateral), lower back, or buttocks. Also, many participants had a standardized hip exam during which they were asked about pain during passive internal rotation. We examined sensitivity (Sn), specificity (Sp) and positive and negative predictive values (PPV, NPV) of different constellations of hip pain and location specific pain with ROA. Results: Radiographs from 948 participants were evaluated including 419 men and 529 women. The average age was 63.5 years (s.d. 9 yrs). One hundred sixty participants (87 men and 73 women) had ROA (16.9%). One hundred eighty-two participants (19.2%) had hip pain. Only 22% of participants with hip pain had ROA in the same hip whereas 15.7% of participants without hip pain had ROA in that hip. The Sn of hip pain for ROA was 25%, the Sp of hip pain for ROA was 82%, and the PPV for hip pain was 22% (See table). However, when we restricted analyses to those with hip pain who localized this pain to the groin, the PPV rose to 38.1%. But of those with ROA, only 6.3% had hip and groin pain. For hip pain with anterior pain, the PPV was 27.8%, and the Sn was 7%. The diagnostic test performance for other sites of pain was poorer. If a participant had hip pain and pain with passive internal rotation, 18% had ROA. Sensitivity Hip Pain Hip Pain with Groin Pain Hip Pain with Anterior Pain Hip Pain with Lateral Pain Hip Pain with Low Back Pain Hip Pain with Buttocks Pain Hip Pain with Pain on Internal Rotation

25% (40/160) 6.3% (8/128)

Specificity 82% (646/788) 98% (646/659)

PPV

NPV

22.0% (40/182) 38.1% (8/21)

84.3% (646/766) 84.3% (646/766)

7.0% (9/129)

96.4% (646/670)

27.8% (9/33)

84.3% (646/766)

13% (18/138)

89.7% (646/720)

19.6% (18/92)

84.3% (646/766)

10% (12/132)

90% (646/719)

14.1% (12/85)

84.3% (646/766)

5.5% (7/127)

95% (646/680)

17.1% (7/41)

84.3% (646/766)

6.3% (8/128)

94.7% (646/682)

18.9% (8/44)

84.3% (646/766)

Conclusion: We found poor agreement between hip pain on most days and radiographic OA in the ipsilateral hip. Many older persons with frequent hip pain including pain in the groin or front of the thigh did not have positive x-rays in that hip, and many persons with radiographic OA did not have hip pain. Of the constellation of questions for hip pain, those with the highest PPV were hip pain with groin or anterior pain. Disclosure: K. C. Kim, None; D. T. Felson, None; K. D. Linsenmeyer, None; A. Guermazi, Boston Imaging Core Lab, 1, Merck Serono, 5, Sanofi-Aventis Pharmaceutical, 5, TissueGene, 5; S. C. Vlad, None; M. M. Clancy, None; J. Niu, None.

262 Is There a Role For Non-Mendelian Inheritance In Severe Osteoarthritis Of The Knee. Allen D. Sawitzke1, Richard Pimentel2, Jathine Wong2, Helena Martinez3, Marta Herrero4, Josep Verges5 and Daniel O. Clegg6. 1 University of Utah, Salt Lake City, UT, 2University of Utah, Salt Lake, UT, 3 BIOIBERICA, S.A:, Barcelona, Spain, 4Bioiberica, Barcelona, Spain, 5Pharmacological Research Unit, Scientific Medical Department, Bioibe´rica, S.A., Barcelona, Spain, 6George Wahlen VA Medical Center/University of Utah, Salt Lake City, UT Background/Purpose: Although the etiopathogenesis of osteoarthritis of the knee remains largely unknown, it is clear that genetic risks are contributory. Traditional family, twin, case control and GWAS designs have all contributed to the extensive knowledge in this area. Recently, groups have reported that mitochondrial genes are involved (1) and not-involved (2) in OA risk. The Utah Population Database (UPDB), a unique genealogic resource housed at the University of Utah has previously been used to study large founder pedigrees for evidence of many heritable diseases and we have used it to examine traditional Mendelian inheritance in OA of the knee. We hypothesized that these pedigrees could also be used to assess the potential role of non-Mendelian inheritance, specifically mitochondrial inheritance on the development of OA of the knee by comparing the statistical measures of mitochondrial models (MP) to autosomal models (AP) in these same founder families. Methods: All OA cases who underwent TKA in Utah were selected based on billing codes in the International Classification of Diseases and data pulled from statewide hospital discharge data for a ten year period (1996–2007). Cases and controls matched for gender and age and their families were linked to the UPDB for analysis. The software kinship analysis tool (KAT) was used to analyze mitochondrial (MP) with autosomal (AP) by comparing familial standardized incidence ratio (FSIR) under each inheritance models. Results: Over 18,000 OA patients who underwent TKA in Utah hospitals were linked to the UPDB and analysis performed for mitochondrial inheritance. The linkage resulted in 683 founder families with at least 5 affected members for analysis in comparison to randomly selected control families. Pattern

FSIR

Extended Bayes FSIR

PAR

Autosomal Pattern Mitochondrial Pattern

1 1.015

1.75 (1.61–1.91) 1.77 (1.61–1.96)

0.21 (0.18–0.24) 0.14 (0.12–0.15)

MP had a very low population-attributable risk, PAR value, 14, where AP has a higher value of 0.21. It suggested approximately *% of the population who have OA of the knee could be from mitochondrial inheritance, whereas 21 percent of OA of the knee resulting in TKA are familial. Conclusion: The UPDB is a resource that has confirmed a Mendelian heritable risk for OA of the knee with a PAR of 21%. Now, it has also been used to shown non-Mendelian, in particular mitochondrial inheritance is not likely responsible for a large fraction of the heritability of OA of the knee severe enough to result in TKA as its associated PAR is 14. 1. Rego-Perez I, Fernandez-Moreno M, Fernandez-Lopez C, Arenas J, Blanco FJ. Mitochondrial DNA haplogroups: role in the prevalence and severity of knee osteoarthritis. Arthritis Rheum. 2008;58(8):2387–96. 2. Hudson G, Panoutsopoulou K, Wilson I, Southam L, Rayner NW, Arden N, et al. No evidence of an association between mitochondrial DNA variants and osteoarthritis in 7393 cases and 5122 controls. Ann Rheum Dis. 2013;72(1):136–9. Disclosure: A. D. Sawitzke, Bioiberica S.A., 2, Bioiberica S.A., 8; R. Pimentel, None; J. Wong, None; H. Martinez, Bioiberica S.A., 3; M. Herrero, Bioiberica, 3; J. Verges, Bioiberica S.A., 3; D. O. Clegg, None.

263 “Generalized Osteoarthritis”: A Systematic Review. Amanda E. Nelson1, Michael W. Smith2 and Yvonne M. Golightly3. 1University of North Carolina Thurston Arthritis Research Center, Chapel Hill, NC, 2Saint Luke’s Hospitals, Kansas City, MO, 3University of North Carolina at Chapel Hill, Chapel Hill, NC Background/Purpose: To perform a systematic review of definitions, risk factors, and outcomes in “generalized osteoarthritis” (GOA). Methods: We performed a systematic review of Medline literature using the terms osteoarthritis, generalized, polyarticular, multiple joint, multi-joint, and others to obtain articles related to GOA, following PRISMA guidelines.

S108

Author (year) 1 2 3 4 5 6 7 8 9 10 11 12

13 14 15

16

Kellgren (1952) Lawrence (1969) Hordon (1993) Cooper (1996) Multiple Solomon (1976) Loughlin (1994) Multiple Price (1987) Multiple Hopkinson (1992) Dougados (1996)

Huang (2000) Carroll (2006) Hoogeboom (2010)

Summary scores

CMC

Knee

Hip

Spine

Other specified sites

X

X

#

#

#

MTP, TMT

#

#

#

#

#

3⫹

X

#

#

#

#

2⫹ to 5⫹

#

#

#

#

# #

# #

#

# #

No. joints or groups

3⫹ or 5⫹

Nodes

#

2⫹ 2⫹

X

3⫹

X

6⫹

X X

IP

3⫹

5356946 MTP, shoulder, ankle

8252314 8923371

# #

(1) 984909 8000735

X (3⫹) # X X (3⫹) X

X

PMID 14896078

# X #

# X #

# # #

X

X (3⫹)

X

X

X

(2) 3625636 (3) 1525625

#

X

bilateral IP or spine ⫹ bilateral knee

8938865, 19089534

MTP

16755236

X

10662878

⫹ clinical 20594308 signs ⫹ ADL impairment by HAQ (4)

X⫽required, #⫽considered as a potential site. IP⫽interphalangeal; CMC⫽carpometacarpal; MTP⫽metatarsophalangeal; TMT⫽tarsometatarsal; ADL⫽activities of daily living; HAQ⫽health assessment questionnaire. (1) PMID: 15818669, 19575196, 18226556. (2) PMID: 6134929, 9458216, 10402070. (3) PMID: 8447703, 11132207, 2042984, 10070270, 10760642. (4)PMID: 6712295, 3780105, 7840096, 9627016, 16079167, 21572158.

The Proportion Of Knee Osteoarthritis Patients In Southern Sweden That Seek Medical Care. Aleksandra Turkiewicz1, Maria Gerhardsson de Verdier2, Gunnar Engstro¨m3, Stefan Lohmander1 and Martin Englund1. 1 Department of Orthopedics, Clinical Sciences Lund, Lund University, Lund, Sweden, 2Astra Zeneca R&D Molndal, Molndal, Sweden, Molndal, Sweden, 3 Dept of Cardiovascular epidemiology, Clinical Sciences, Lund University, Lund, Sweden Background/Purpose: To provide estimates of the proportion of subjects with knee osteoarthritis (OA) that seek healthcare for knee OA or joint pain within an 8-year period. Methods: In 2007 a random sample of 10 000 56 to 84 year old Region Skåne residents from the Malmo¨ Diet and Cancer Study (Manjer et al 2001) were sent a mailed questionnaire about knee pain in the last 12 months; this being the first part of the Malmo¨ Osteoarthritis Study (MOA). We classified subjects with knee pain with duration of at least 4 weeks as having frequent knee pain. A random sample of 1300 subjects with frequent knee pain and a random sample of 650 subjects without (out of the 7737 questionnaire responders) were invited for a clinical and radiographic examination including assessment of clinical knee OA according to the American College of Rheumatology (ACR) clinical criteria. Participants underwent radiography of both knees in weight-bearing and semi-flexion. An independent radiologist who was blinded to clinical data assessed all frontal and patellofemoral radiographs. Subjects who fulfilled criteria approximating Kellgren and Lawrence (KL) grade 2 or worse were considered as having radiographic knee OA. We considered those having radiographic knee OA and frequent knee pain to have symptomatic knee OA. Using the subject’s personal identification number and individual linkage with the Skåne Health Care Register, covering the entire population in the county, we retrieved information on all doctor visits with a diagnosis of knee OA (ICD-10 code M17) or pain in joint (joint unspecified, ICD-10 code M25.5) for the years 2004 to 2011, i.e. the 8-year period preceding and following the MOA examination. We used weighting to adjust for different sampling probabilities depending on the knee pain status as well as for the nonresponse and volunteer bias in the MOA study. We used multiple imputation to account for missing diagnostic codes in the register. Results: The 10 000 MOA subjects had mean (SD) age of 70 (7.6) years, mean (SD) body mass index was 27.1 (5.0) and 62% were women. The response rate in mailed questionnaire was 77.4% and 1527 invited subjects (78.3%) attended the clinical visit. Out of subjects classified as having symptomatic knee OA, 74.6% (95%CI: 70.0% to 79.3%) had consulted a physician during the 8-year time frame and received a diagnosis of knee OA or pain in joint (Table). Among subjects having clinical knee OA the corresponding percentage was 66.8% (95%CI: 59.1% to 74.6%). The ICD-10 diagnosis of knee OA was set in 62.9% (95%CI: 57.7% to 68.1%) of subjects with symptomatic knee OA and 49.9% (95%CI: 41.8% to 58.1%) of patients with clinical knee OA. Table. The percentage of subjects with knee osteoarthritis (OA) who received a diagnosis of knee OA or pain in joint, respectively, set by a physician during the examined 8-year time frame (2004–2011).

A variety of risk factors and outcomes were considered in the included papers. Increased risk of GOA or its progression was associated with age, female sex, and genetic/familial factors. GOA constructs were generally identified more frequently among individuals of European descent. Associations with increased BMI or other measures of body mass were not consistent across studies. A higher BMD was seen in GOA patients in some series but not others. One study estimated the heritability of GOA at 42%. Biomarker (cartilage oligomeric protein [COMP] and urinary type II collagen C-telopeptide [uCTX-II]) levels increased with greater numbers of involved joints. Increased OA burden was associated with higher mortality, poorer health and function, and increased disability. Conclusion: While there remains no clear or widely agreed upon definition of GOA, this term is commonly used in the literature. There remains substantial debate regarding the existence of GOA as a distinct disease entity, and in what populations, although there appears to be a greater impact on health in the presence of OA in more than one joint. It may be more appropriate for individual studies to clearly define joints evaluated and involved with OA, and use alternate, more descriptive terms, such as multi-joint or polyarticular OA. Disclosure: A. E. Nelson, None; M. W. Smith, None; Y. M. Golightly, None.

264

OA definition Radiographic knee OA† Clinical knee OA† Symptomatic knee OA† Symptomatic or clinical knee OA†

Diagnosis of knee OA or pain in joint % (95% CI)*

Diagnosis of knee OA % (95% CI)*

Diagnosis of pain in joint % (95% CI)*

56.8(49.1–64.5) 66.8(59.1–74.6) 74.6(70.0–79.3) 68.9(63.8–74.0)

43.4(36.5–50.4) 49.9(41.8–58.1) 62.9(57.7–68.1) 53.2(47.6–58.9)

31.3(23.7–39.0) 39.7(31.0–48.4) 40.8(35.2–46.4) 40.1(34.3–45.9)

*- 95%CI: 95% confidence intervals. †- Radiographic knee OA - changes on x-ray approximating Kellgren-Lawrence grade 2 or worse; Clinical knee OA - clinical OA according to the American College of Rheumatology clinical criteria, recursive positioning method; Symptomatic knee OA - knee pain of duration at least 4 weeks in the last. 12 months in combination with radiographic OA as defined above.

Conclusion: Only 2 of 3 subjects with clinical or symptomatic knee OA consult a physician for his/her knee(s) symptoms in an 8-year time frame. Self-management or coping strategies as well as over-the-counter pain treatments are plausible explanations. Disclosure: A. Turkiewicz, None; M. Gerhardsson de Verdier, Astra Zeneca, 1, Astra Zneca, 3; G. Engstro¨m, Astra Zeneca, 3; S. Lohmander, None; M. Englund, None.

S109

Sunday, October 27

The search was completed on 9/25/12 by a professional librarian. The initial searches produced 948 articles for title and abstract review (performed by MWS and AEN), after which 108 underwent full text review (AEN and YMG). Of these, 74, along with 24 identified through bibliographic review of included articles (total⫽98), were included for data extraction (AEN) based on pre-specified criteria including the requirement for a clearly stated definition of OA (clinical, radiographic, or symptomatic) assessed at more than one body site. Results: In the 98 included articles, 24 large cohorts (n⬃30,000) were represented along with numerous clinical series (n⬃9000), across 22 countries and 60 years (1952–2012). The sites assessed, and OA definitions at each site, varied widely but most often included the hands and knees. In 30 studies where a definition of GOA was stated, no less than 15 definitions of GOA were given; at least 6 groups used a summed score of joints or radiographic grades at multiple sites (Table). While interphalangeal joints were almost always included, there was debate about the appropriateness of including other joints such as spine and hip. Estimates of the prevalence of such variably defined GOA ranged from 1–80%, although a majority were in the 5–25% range.

Disclosure: D. Misra, None; D. T. Felson, None; I. K. Haugen, None; M. Englund, None; T. Neogi, None.

Sunday, October 27

265 All-Cause Mortality and Incident Cardiovascular Disease In Knee Osteoarthritis: The Framingham Study. Devyani Misra1, David T. Felson1, Ida K. Haugen2, Martin Englund3 and Tuhina Neogi1. 1Boston University School of Medicine, Boston, MA, 2Diakonhjemmet Hospital, Oslo, Norway, 3Department of Orthopedics, Clinical Sciences Lund, Lund University, Lund, Sweden Background/Purpose: Studies examining the relation of osteoarthritis (OA) to mortality or cardiovascular disease (CVD) have reported conflicting results, raising the question of whether OA itself increases such risks, or rather OA may simply be a marker of these outcomes due to its relation to other factors that heighten such risks, such as obesity. We assessed the relation of knee OA to mortality and incident CVD in a population-based cohort of older adults in which knee OA status, death and CVD were comprehensively assessed in a standardized, validated manner. Methods: We included participants aged 50–75 from the Framingham OA Study comprising the Original and Offspring cohorts with knee x-rays obtained at exam 18 (1983–85) for the Original cohort, and exam 7 (2002–2005) for the Offspring cohort. Radiographic knee OA (ROA) was defined as Kellgren and Lawrence grade ⱖ2, and symptomatic knee OA (SxOA) as ROA plus pain in the same knee. All-cause mortality was defined as death due to any cause. Incident CVD was defined as a new diagnosis of coronary heart disease, intermittent claudication, CHF, stroke or TIA in the absence of any of these diseases prior to knee OA assessment. All deaths and CVD events were adjudicated by a panel of cardiologists and neurologists using published criteria to review clinical data and hospitalization records. Follow-up started from the date of knee OA ascertainment and continued until the outcome (death/incident CVD) or last assessment in 2009. We truncated follow-up at 10 years to avoid universal occurrence of events in these older subjects. In sensitivity analyses, we evaluated outcomes beyond 10 years. For the incident CVD analysis, only those free of CVD at the time of knee OA ascertainment were eligible for inclusion. The relation of knee OA to overall mortality and incident CVD were examined in both cohorts combined as well as separately, using Cox proportional hazards models, adjusting for age and sex in the first model, and then additionally adjusted in a second model for body mass index (BMI), diabetes mellitus, hypertension, renal disease, aspirin use, lipid-lowering medications and smoking status. Results: Among 2037 participants (mean age 66 yrs, 58% women, mean BMI 27 kg/m2, 499 ROA, 163 SxOA), there were 873 deaths and 624 incident CVD cases. No association was found between knee OA (ROA or SxOA) and overall mortality or incident CVD (Table), with effect estimates in the adjusted models ranging from 0.93–1.27. Results were similar when the cohorts were analysed separately and in the sensitivity analyses. While OA was associated with obesity and other risk factors for CVD, additionally adjusting for these factors did not affect the risk of the outcomes. Table. Relation of Knee Osteoarthritis to All-cause mortality and incident cardiovascular disease in Framingham Original and Offspring cohorts combined Knee OA status

N (%)

Mean follow-up (years)

Crude HR

Adjusted HR* (95% CI)

Adjusted HR** (95% CI)

All-cause mortality ROA Yes (N⫽499) No (N⫽1538) (referent grp) SOA Yes (N⫽163) No (N⫽1852) (referent grp)

258 (52) 615 (40)

6.06 5.23

0.99

0.93 (0.78, 1.10)

1.04 (0.85–1.28)

58 (36) 802 (43)

4.53 5.48

0.99

0.98 (0.71, 1.35)

1.00 (0.67, 1.50)

Incident Cardiovascular Events ROA Yes (N⫽499) No (N⫽1538) (referent grp) SOA Yes (N⫽163) No (N⫽1852) (referent grp)

183 (42) 441 (31)

4.22 3.56

1.13

1.06 (0.84, 1.34)

0.97 (0.75, 1.27)

50 (34) 566 (34)

2.28 3.80

1.23

1.23 (0.81, 1.91)

1.27 (0.77, 2.07)

*Adjust for age and sex only. **Additionally adjusted for BMI, diabetes mellitus, hypertension, renal disease, aspirin use, lipid-lowering medication use, smoking.

Conclusion: In this large cohort of community-dwelling older adults, there was no significant relation of knee OA to mortality or incident CVD, suggesting knee OA itself likely does not have systemic effects.

266 Cardiovascular Disease In Osteoarthritis: Hip Versus Knee and The Influence Of Multiple Symptomatic Joint Involvement. Anthony V. Perruccio1, Rita A. Kandel2 and Aileen M. Davis3. 1Toronto Western Hospital, Toronto Western Research Institute, and University of Toronto, Toronto, ON, 2 Mount Sinai Hospital and University of Toronto, Toronto, ON, 3Health Care and Outcomes Research, Toronto Western Research Institute; Departments of Rehabilitation Science and Health Policy, Management and Evaluation, University of Toronto, Toronto, ON Background/Purpose: The strength of association between obesity and osteoarthritis (OA) is reported to vary by joint (e.g. stronger association with knee than hip). As well, the association between obesity and non weightbearing joints, such as the hands, suggests that systemic/metabolic factors are involved. The presence of multiple symptomatic joints among individuals with severe OA in at least one joint further suggests likely systemic factors in OA. Given known associations between systemic/metabolic factors and cardiovascular disease (CVD) risk, this study investigated the likelihood of reporting CVD between those with hip vs. knee OA, and further considered if the number of symptomatic joints influenced this likelihood. Methods: Patients with severe OA (443 hip; 540 knee) reported CVD, diabetes, high blood pressure (HBP), symptomatic joints, height, weight, age and sex. The cross-sectional association between study measures and prevalent CVD was investigated using logistic regression, with the inclusion of an interaction term between symptomatic joint count and knee/hip group membership. Subsequently, the analysis was stratified by knee/hip. Results: Knee: ages 41–88 years, 64% female, 4% CVD, 14% diabetes, 50% HBP. Hip: ages 40–91 years, 55% female, 5% CVD, 9% diabetes, 43% HBP. Mean joint count was 5 (⫾4), ranging from 1–20. Adjusted for age, sex, body mass index, and presence of diabetes or HBP, the knee cohort had odds 4 times greater than hips for reporting CVD (p⬍0.01). There was a significant interaction between joint count and knee/hip group status, such that the effect of increasing joint count was greater for the hip cohort. From stratified analyses, joint count was associated with CVD in the hip cohort only, with a 25% increased odds with each numerical increase in symptomatic joint count (odds ratio (OR): 1.25, p⬍0.001). In the knee cohort, the presence of diabetes or HBP was significantly associated with reporting CVD (OR: 3.74, p⫽0.02), whereas a similar association was not found in the hip cohort (OR: 1.23, p⫽0.68). Conclusion: The variable association observed with CVD between hip and knee OA, alongside the differential influence of the extent of symptomatic joint involvement between joint groups suggests different OA phenotypes, with potentially varying roles for systemic factors. This has potential implications for our understanding of OA, and suggests a potential need for multimodal approaches to treatment and management. Disclosure: A. V. Perruccio, None; R. A. Kandel, None; A. M. Davis, None.

267 Risk Of Falls Increases With Additional Symptomatic Osteoarthritic Joints: The Johnston County Osteoarthritis Project. Adam Dore1, Yvonne M. Golightly1, Vicki Mercer1, Jordan B. Renner1, Xiaoyan A. Shi2, Joanne M. Jordan3 and Amanda E. Nelson3. 1University of North Carolina at Chapel Hill, Chapel Hill, NC, 2SAS Institute, Inc, Cary, NC, 3University of North Carolina Thurston Arthritis Research Center, Chapel Hill, NC Background/Purpose: Falls, a growing problem among older adults, often lead to hospitalizations, surgeries, and death. Knee and hip osteoarthritis (OA) are known risk factors for falls, but whether they together additionally contribute to falls risk is unknown. This study utilizes a cohort of African American and Caucasian men and women with and without OA to examine the influence of body OA burden on risk for future falls. Methods: A longitudinal analysis was performed using data from 2 time points of the Johnston County OA Project. The outcome of interest was falls at follow up (⬃5 years after initial visit), based on the response to the question “In the last 12 months, have you had any falls of any type?” Baseline risk factors included age, sex, race, body mass index, and prior falls, with a focus on symptomatic OA of the hip and knee (Model 1). Symptomatic OA was defined as patient reported symptoms (pain, aching, or stiffness on most days) and radiographic evidence of OA (Kellgren Lawrence grade ⱖ 2 or joint replacement secondary to OA in the tibiofemoral or hip joint) in the same

S110

Table 1. Characteristics of the Sample (n⫽1619) Model 1

Model 2

Characteristic

Frequency Baseline n(%) or mean (range)

OR (95% CI)

OR (95% CI)

Caucasian Male Mean Age (years) Mean BMI (kg/m2) Falls at baseline Symptomatic Knee OA Symptomatic Hip OA Lung problems1 Narcotic use2 CES-D3 Low Back Symptoms4

1245 (71.6) 484 (34.7) 62 (45–89) 30.8 (17.1–65.8) 352 (21.8) 321 (20.2) 196 (12.9) 311 (19.2) 50 (3.1) 6.1 (0–53) 776 (48.0)

1.37 (1.04 ,1.81) 0.72 (0.55 ,0.93) 1.03 (1.01 ,1.04) 1.01 (0.99 ,1.03) 2.42 (1.85 ,3.17) 1.41 (1.05 ,1.91) 1.70 (1.22 ,2.36) – – – –

1.41 (1.05 ,1.88) 0.80 (0.61 ,1.05) 1.03 (1.01 ,1.04) 1.01 (0.99 ,1.03) 2.38 (1.80 ,3.14) 1.28 (0.92 ,1.78) 1.56 (1.08 ,2.25) 1.56 (1.15 ,2.10) 1.70 (0.88 ,3.29) 1.02 (1.00 ,1.03) 1.00 (0.77 ,1.29)

Results: 3070 patients were followed in the BIKER registry with a total observation time of 8389 patient years (py). 13 events of IBD-flares in 12 patients (33% male), 8 with Crohn’s disease (CD, 67%) and 4 with ulcerative colitis (UC, 33%). were documented. 4 patients had extended oligoarthritis (eoJIA, 33%), 3 seronegative polyarthritis (RF-JIA, 25%), 3 enthesitis-related arthritis (ERA, 25%) and 2 psoriatic JIA (PsA, 17%). HLA-B27 was positive in 3 (25%) and anti-nuclear antibodies were positive in 7 IBD patients (58%). Thus patient with IBD as adverse event more commonly had ERA, eoJIA and PsA than expected in the total BIKER cohort, but not statistitically significant. No IBD occurred in systemic JIA, RF positive or pers-oligo JIA. IBD incidence in patients with pers-oligo JIA was significantly decreased. IBD incidence in JIA patients was 143/100,000 py and significantly higher than IBD incidence in the general pediatric population (5.2/100,000 py, p⬍0.001; OR 27.50 [15.54–48.67]). Mean disease duration until IBD was 7.8 ⫾ 4.0 years (1.8–16.8). The total exposure time was 4575PY for NSAIDs, 1981PY for corticosteroids, 195PY for sulfasalazine (SUL), 5455PY for methotrexate (MTX), 212PY for leflunomide (LEF), 3557PY for ETA and 369PY for adalimumab (ADA). Incidence of IBD was significantly higher in patients treated with etanercept (p⬍0.05; OR 4.53 [1.25–16.47]), leflunomide (p⬍0.05; OR 7.03 [1.55–31.91]) and also sulfasalazine (p⬍0.001; OR [12.63 (3.45–46.24]). In patients treated with methotrexate the IBD incidence decreased significantly (p⬍0.05; OR 0.16 [0.04–0.59]). At admission to the registry, JIA patients later developing IBD had been treated with MTX (100%), steroids (67%), SUL (33%), azathioprine (25%), LEF (17%) and hydroxychloroquine (8%). At time of IBD diagnosis 83% were on ETA, 33% on steroids, 25% on MTX, 25% on SUL, 17% on LEF and 1 on ADA. Thus, therapy with non-biologic agents has markedly been reduced. Results Patient cohort Treatment group

JIA registry ref. population* NSAIDs Steroids Sulfasalazine MTX

1. Includes patient report of chronic bronchitis, emphysema or other chronic lung trouble (dichotomous). 2. Defined as self-reported narcotic use for ⬎ 2 weeks (dichotomous). 3. Center for Epidemiologic Studies Depression scale (continuous, 0–60). 4. Answer to: “On most days, do you have pain, aching, or stiffness in your low back?”

Leflunomide Etanercept

Conclusion: This study confirms that symptomatic knee and particularly hip OA are important risk factors for falls and reveals that the risk increases with additional symptomatic knee and hip joints. Future interventions aimed at enhancing fall prevention should not ignore the impact of multiple symptomatic OA joints.

Adalimumab

⫹ ⫺ ⫹ ⫺ ⫹ ⫺ ⫹ ⫺ ⫹ ⫺ ⫹ ⫺ ⫹ ⫺

IBD events no.

IBD-rate/ 1000PY

12 739 6 7 4 9 3 10 3 10 2 11 10 3 1 12

1,430 0,052 1.31 1.84 2.02 1.40 15.41 1.22 0.55 3.41 9.45 1.35 2.81 0.62 2.71 1.50

p-value ⬍0.001

odds ratio (95% CI) 27.50 (15.54–48.67)

0.5442

0.71 (0.24–2.13)

0.5437

1.44 (0.44–4.68)

⬍0.001

12.63 (3.45–46.24)

0.0015

0.16 (0.04–0.59)

0.0032

7.03 (1.55–31.91)

0.0118

4.53 (1.25–16.47)

0.5638

1.81 (0.24–13.95)

* Reference population children aged ⬍16 years (see Sawczenko et al.)

Disclosure: A. Dore, None; Y. M. Golightly, None; V. Mercer, None; J. B. Renner, None; X. A. Shi, None; J. M. Jordan, Trinity Partners, Inc., 5, Osteoarthritis Research Society International, 6, Chronic Osteoarthritis Management Initiative of US Bone and Joint Initiative, 6, Samumed, 5, Interleukin Genetics, Inc., 5, Algynomics, Inc., 1; A. E. Nelson, None.

Conclusion: Incidence of IBD in JIA patients is higher than expected. Especially patients with ERA, PsA and eoJIA are at risk. IBD seems to be associated with treatment of ETA, LEF and SUL, while MTX turned out to be protective. The impact of discontinuation of MTX, AZA and corticosteroids in patients later on developing IBD has to be discussed. Unfortunately case numbers so far are too small for further statistical analyses including linear regression models.

ACR Poster Session A

Disclosure: D. Barthel, None; G. Horneff, AbbVie, Pfizer, Roche, 2, AbbVie, Novartis, Pfizer, Roche, 8.

Pediatric Rheumatology - Clinical and Therapeutic Aspects I: Juvenile Idiopathic Arthritis Sunday, October 27, 2013, 8:30 AM–4:00 PM

269

268 Inflammatory Bowel Disease In Juvenile Idiopathic Arthritis Patients Upon Biologics. Deborah Barthel1 and Gerd Horneff2. 1Asklepios Clinic Sankt Augustin, Sankt Augustin, Germany, 2Centre of Pediatric Rheumatology, Sankt Augustin, Germany Background/Purpose: Inflammatory bowel disease (IBD) is a matter of interest in patients with juvenile idiopathic arthritis (JIA) treated with biologics. Methods: Baseline demografics, clinical characteristics and medication have been documented in the German BIKER JIA registry between 2001 and 2013. Exposure time of treatment was used to calculate IBD incidence. Published data of IBD incidence in the general pediatric population were used as reference.

Marked Improvement In Patient Reported Outcomes Of Children With Active Systemic Juvenile Idiopathic Arthritis With Canakinumab Treatment – Results Of The Phase III Program. Rayfel Schneider1, Hermine I. Brunner1, Nicolino Ruperto2, Nico Wulffraat2, Pierre Quartier3, Riva Brik2, Liza McCann2, Helen E. Foster2, Michael Frosch2, Valeria Gerloni2, Liora Harel2, Claudio Len2, Kristin Houghton1, Rik Joos2, Ken Abrams4, Karine Lheritier5, Sophia Kessabi5, Alberto Martini2 and Daniel J. Lovell1. 1Pediatric Rheumatology Collaborative Study Group (PRCSG), Cincinnati, OH, 2 Pediatric Rheumatology International Trials Organization (PRINTO)-Istituto Gaslini, Genova, Italy, 3Necker-Enfants Malades Hospital, Paris, France, 4 Novartis Pharmaceuticals Corporation, East Hanover, NJ, 5Novartis Pharma AG, Basel, Switzerland Background/Purpose: Efficacy and safety of canakinumab (CAN), a selective, fully human, anti-interleukin-1␤ monoclonal antibody, in systemic

S111

Sunday, October 27

joint. Additional potential contributing factors were 6 comorbid conditions, alcohol and medication use (sleep aids, narcotics, vitamin D, and bisphosphonates), and low back symptoms (Model 2). Logistic regression analyses were used to determine associations (adjusted odds ratios [aOR]) between baseline covariates and falls at follow-up in both Models 1 and 2. Results: Participants (n⫽1,619) had a mean age of 62 years (SD 9.03); 72% were Caucasian, 35% male, 22% had ⱖ 1 fall within 12 months of the baseline time point and 26% had ⱖ1 fall within 12 months of the follow up visit. Caucasians (aOR 1.37), females, older adults, those who reported a history of lung problems (aOR 1.56) and particularly those who had a fall at baseline (aOR 2.42) had greater odds of a fall at follow-up (Model 1; Table 1). When controlling for baseline characteristics (Model 1), patients who had symptomatic knee or hip OA had increased odds for falling (aOR 1.41 and 1.70, respectively). In Model 2, Caucasians (aOR 1.41), older adults, those who had a previous fall (aOR 2.38), reported a history of depression (aOR 1.02), and symptomatic hip OA (aOR 1.56) had increased odds of falls. However, the association between falls and symptomatic knee OA as well as female sex were no longer significant. The odds of falls increased with an increasing number of knee and/or hip joints with symptomatic OA: the aOR for patients with 1 involved joint was 1.56 (95% CI 1.12, 2.17), for 2 joints was 1.85 (CI 1.28, 2.68), and for 3–4 joints was 1.91 (95% CI 1.01, 3.65).

Sunday, October 27

juvenile idiopathic arthritis (SJIA) patients has been demonstrated in two phase 3 trials (1 and 2). 1SJIA is associated with severe functional impairment and chronic pain that markedly affect the health-related quality of life (HRQoL). Here, we present the effects of CAN therapy on patient-reported outcomes (PROs) from the phase 3 studies. Methods: In Trial 1, 84 pts were randomized to CAN (43) and PBO (41) and followed for 1 month. In Trial 2, 177 pts (including 71 from Trial 1) received open label CAN in Part 1, of which 100 CAN-responders entered part 2 and were randomized 1:1 to placebo or continued CAN. PROs considered were functional ability by the Childhood Health Assessment Questionnaire [CHAQ©], pain as measured on a visual analog scale [VAS] of 0–100 mm, and HRQoL by the Child Health Questionnaire–Parent Form (CHQ-PF50 for pts age 5–18 yrs) with rating on physical (PhS) and psychosocial (PsS) health status. Betweentreatment differences were evaluated using analysis covariance models with repeated measures adjusting for treatment group, time, stratification factors and the treatment group-by-time interaction as explanatory variables. Results: At the end of Trial 1, there was significant improvement (p⫽0.0002) in functional ability with CAN, resulting in an estimated difference (ED) in CHAQ score of ⫺0.69 between the CAN and PBO group at Day 29 from baseline [BL]; this ED constitutes about 3.6x the minimal clinically important difference (⫺0.19)2 in the CHAQ score. Pain intensity significantly declined (both p ⬍0.0001) in the CAN group vs. PBO both at Day 15 (ED, ⫺46.42) and Day 29 (ED, ⫺41.86). Similarly, HRQoL significantly improved from BL for CAN vs. PBO, with an ED in CHQ-PF50 PhS scores of 12.07 and 7.28 for the CHQ-PsS (both, p⬍0.005) over 1 month with CAN, respectively. Even more pronounced improvements in functional ability, pain and HRQoL were observed in Trial 2 (Table).

Bristol Myers and Squibb, -Glaxo Smith & Kline, Italfarmaco, MedImmune, Novartis, 8; D. J. Lovell, NIH, 2, Astra-Zeneca, Centocor, Jansen, Wyeth, Amgen, Bristol-Meyers Squibb, Abbott, Pfizer, Regeneron, Hoffman La-Roche, Novartis, Genentech, 5, Genentech, Roche, 8.

270 Canakinumab In The Treatment Of Systemic Juvenile Idiopathic Arthritis: Results From a 12-Week Pooled Post-Hoc Analysis For Efficacy. Hermine Brunner1, Pierre Quartier2, Tamas Constantin3, Shai Padeh3, Inmaculada Calvo3, Muferet Erguven3, Laurence Goffin3, Michael Hofer3, Tilmann Kallinich3, Sheila Oliveira3, Yosef Uziel3, Stefania Viola4, Kiran Nistala3, Carine Wouters3, Karine Lheritier5, Josef Hruska5, Ken Abrams6, Alberto Martini3, Nicolino Ruperto3 and Daniel J. Lovell1. 1Pediatric Rheumatology Collaborative Study Group (PRCSG), Cincinnati, OH, 2Necker-Enfants Malades Hospital, Paris, France, 3Paediatric Rheumatology International Trials Organization (PRINTO), Genova, Italy, 4Pediatria 2, Genova, Italy, 5Novartis Pharma AG, Basel, Switzerland, 6Novartis Pharmaceuticals Corporation, East Hanover, NJ

1.7 (0.8)

0.74 (0.9)

0.5 (0.9)

0.6 (0.8)

66.6 (23.3)

20.2 (25.8)

13.6 (26.9)

17.0 (24.2)

16.1 (14.3) 41.6 (11.1)

37.7 (17.2) 50.7 (11.1)

43.6 (17.4) 53.6 (11.3)

39.0 (18.1) 52.7 (9.8)

Background/Purpose: Interleukin-1␤ (IL-1␤) plays a key role in the pathogenesis of systemic juvenile idiopathic arthritis (SJIA), a severe disabling subtype of JIA characterized by arthritis plus systemic symptoms. Canakinumab (CAN), a selective, fully human, anti-IL-1␤ monoclonal antibody has demonstrated efficacy and safety in 2 phase III trials in SJIA patients (pts). Here we present results of a post-hoc pooled analysis evaluating the 12-week efficacy of CAN 4 mg/kg every 4 weeks. Methods: Pooled data from the 3 phase III trials were used. Pts aged 2–19 yrs with active SJIA were enrolled and received subcutaneous CAN 4 mg/kg or placebo every 4 weeks. The post-hoc analysis presented here focuses on response to CAN therapy in the initial treatment period for a total of 178 CAN-naı¨ve pts. Methodological factors precluded a comparator group, so this analysis is of a descriptive nature. Results: At baseline (BL), 94% of pts had intermittent spiking fever due to SJIA and 73% were on steroids (mean dose 0.38 mg/kg/d). By Week 2, evidence of profound SJIA improvement was observed (Table), with 20% of pts achieving inactive disease status.

Note: Results are based on patients with both baseline and post-baseline values. n⬍N (177/50) for CHQ-PF50 as assessment not done in pts ⬍5 and ⬎18 years old. *PBO pts received CAN during Part 1 of Trial 2.

Table. Percentage of patients with adapted JIA ACR (aACR) response* and inactive disease

Patient Reported Outcomes in Trial 2 Outcome measure, mean (SD) CHAQ disability score Pain intensity by 0– 100 mm VAS CHQ-PF50 PhS score CHQ-PF50 PsS score

Baseline End of Part 1 End of Part 2 End of Part 2 CAN, Nⴝ177 CAN, Nⴝ177 CAN, Nⴝ50 PBO*, Nⴝ50

CAN, Nⴝ178; n (%)

Conclusion: Treatment with CAN demonstrated rapid, marked and continued improvement in patient-reported outcome of SJIA patients, including a significant increase in functional ability, reduction of pain and HRQoL. References: 1. Ruperto N, et al. N Engl J Med 2012;367(25):2396–406. 2. Brunner HI, et al. J Clin Rheumatol 2005;32: 150–61.

aACR30 aACR50 aACR70 aACR90 aACR100 Inactive disease

Week 2

Week 12

142 (80%) 125 (70%) 102 (57%) 65 (37%) 38 (21%) 36 (20%)

125 (70%) 122 (69%) 108 (61%) 87 (49%) 54 (30%) 50 (28%)

*Data from missing patients not shown; aACR response ⫽ ACR response level plus absence of fever

Disclosure: R. Schneider, Hoffmann-La Roche, Inc., 5, Novartis Pharmaceutical Corporation, 5, Innomar Strategies, 5; H. I. Brunner, Novartis, Genentech, Medimmune, EMD Serono, AMS, Pfizer, UCB, Jannsen, 5, Genentech, 8; N. Ruperto, To Gaslini Hospital: Abbott, Astrazeneca, BMS, Centocor Research & Development, Eli Lilly and Company, “Francesco Angelini”, Glaxo Smith & Kline, Italfarmaco, Novartis, Pfizer Inc., Roche, Sanofi Aventis, Schwarz Biosciences GmbH, Xoma, Wyeth Pharmaceuti, 2, Astrazeneca, Bristol Myers and Squibb, Janssen Biologics B.V., Roche, Wyeth, Pfizer, 8; N. Wulffraat, Novartis, Pfizer, Roche, 5; P. Quartier, Abbvie, Chugai-Roche, Novartis, Pfizer, 2, Abbvie, BMS, Chugai-Roche, Novartis, Pfizer, Servier, Sweedish Orphan Biovitrum, 5, ChugaiRoche, Novartis, Pfizer, 8; R. Brik, Novartis, 2, Novartis, 5; L. McCann, None; H. E. Foster, Pfizer, BioMarin, 2, Pfizer Inc, 8; M. Frosch, None; V. Gerloni, None; L. Harel, None; C. Len, None; K. Houghton, None; R. Joos, None; K. Abrams, Novartis Pharmaceutical Corporation, 3, Novartis, 1; K. Lheritier, Novartis Pharma AG, 3, Novartis, 1; S. Kessabi, Novartis Pharma AG, 3; A. Martini, Abbott, Astrazeneca, Bristol Myers and Squibb, Janssen Biologics B.V., Ely Lilly and Company, “Francesco Angelini”, Glaxo Smith & Kline, Italfarmaco, Novartis, Pfizer Inc., Roche, Sanofi Aventis, Schwarz Biosciences GmbH, Xoma, Wyeth Pharmaceuticals Inc., 2, Abbott, Astrazeneca, Bristol Myers and Squibb, Janssen Biologics B.V., Ely Lilly and Company, “Francesco Angelini”, Glaxo Smith & Kline, Italfarmaco, Novartis, Pfizer Inc., Roche, Sanofi Aventis, Schwarz Biosciences GmbH, Xoma, Wyeth Pharmaceuticals Inc., 5, Astellas, Astrazeneca,

The median CRP level of 158 mg/L at BL decreased by a median of 82% and 94% by Weeks 2 and 12, respectively. There was a rapid improvement in the median number of active joints from 10 at BL to 2.5 at Week 2 and 0 at Week 12. Similarly, for joints with limitation of motion, median values decreased from 9 at BL to 2.5 at Weeks 2 and 1 at Week 12, respectively. While 94% pts had fever due to SJIA at BL, only 13% at Week 2 and 2% at Week 12 had fever. Notably, CAN therapy resulted in marked improvements in patient-reported outcomes: parent/ patient assessment of pain (0–100 mm, VAS) decreased from a mean of 67 mm at BL to 22 mm at Week 2 and 11 mm at Week 12. The median CHAQ disability score decreased from 1.8 at BL to 0.6 at Week 2 and 0.3 at Week 12. Between BL and Week 12, the median parent’s/patient’s assessment of overall well-being improved from 63 mm to 4.5 mm and the physician’s global assessment of SJIA activity (0–100 mm, VAS) decreased from 70 mm to 3 mm. Conclusion: Based on this post-hoc analysis, response of the SJIA patients studied in the CAN program showed rapid and clinically important improvements in patient reported outcomes and disease activity by Week 12

S112

of therapy, with aACR 50 or higher responses reached by the majority of SJIA pts within 2 weeks of the initial CAN dose.

271 Baseline Characteristics Of Patients With Active Systemic JIA On Canakinumab Therapy Successfully Discontinuing Corticosteroids: Secondary Analyses From A Pivotal PHASE 3 Study. Hermine Brunner1, Nicolino Ruperto2, Tama´s Constantin2, Nico Wulffraat2, Gerd Horneff2, Jordi Anton2, Reinhard Berner2, Fabrizia Corona2, Rubeń J. Cuttica2, Marine Desjonqueres2, Michel Fischbach2, Maria Alessio2, Alice Chieng2, Wolfgang Emminger2, Elie Haddad1, Karine Lheritier3, Ken Abrams4, Josef Hruska3, Daniel J. Lovell1 and Alberto Martini2. 1Pediatric Rheumatology Collaborative Study Group (PRCSG), Cincinnati, OH, 2Pediatric Rheumatology International Trials Organization (PRINTO)-Istituto Gaslini, Genova, Italy, 3 Novartis Pharma AG, Basel, Switzerland, 4Novartis Pharmaceuticals Corporation, East Hanover, NJ Background/Purpose: Interleukin-1␤ (IL-1␤) is a key cytokine in the pathogenesis of systemic juvenile idiopathic arthritis (SJIA). Canakinumab (CAN), a selective fully human anti-IL-1␤ monoclonal antibody, is efficacious in the treatment of SJIA.1 Limitation of use of corticosteroids (CS), a current mainstay of therapy for SJIA, is desirable given their well-known long-term side effects. The present study was aimed to determine patient characteristics that are associated with CS discontinuation with CAN therapy and provide details for the CS-reducing potential of CAN in SJIA. Methods: As part of the phase 3 program of CAN, patients aged 2–19 years with active SJIA plus fever received CAN (4 mg/kg; max:300 mg) every 4 weeks subcutaneously.1 During a CS-tapering phase of up to 20 weeks, CS were reduced as per pre-specified rules, provided patients achieved an adapted JIA ACR50.1 Here, we present patient baseline features pertinent to CS tapering successes, defined as at least a 25% reduction of the baseline CS dose (primary endpoint), reaching a low-dose CS requirement, i.e. ⱕ0.2 mg/kg/day and CS-free status (secondary endpoints). Results: At baseline, 128/177 patients used daily CS [median (range) mg/kg dose: 0.27 (0.02–1.00)] of whom 72% (92/128) entered the CStapering phase. Upon completion of the CS-tapering phase 57/128 patients (44.5%; p⬍0.0001; 90%CI: 37.1–52.2) qualified as CS-tapering successes (primary endpoint). Of note, the majority of the patients either discontinued CS entirely (n⫽42) or required only low-dose CS (n⫽24) (secondary endpoints). Compared to patients still on CS (n⫽86), the 42 patients achieving CS-free status [values are all medians (1st, 3rd quartile)] had fewer joints with active arthritis [15 (8, 29) vs. 7 (3, 13)], fewer joints with limitation on motion [14 (7, 33) vs. 5.5 (2, 12)] at baseline. The groups were no different in terms of gender, race, SJIA duration, CRP, number of flares in 6-months period preceding baseline, or specific types of systemic features (hepatosplenomegaly, lymphadenopathy or serositis). Additional associations of CS-free status, initial CAN response, and select laboratory baseline features will be provided.

References: 1. Ruperto N. et al. N Engl J Med 2012;367:2396–406. Disclosure: H. Brunner, Consulting fees, 5; N. Ruperto, Research grants, 2, Speakers’ bureau, 8; T. Constantin, None; N. Wulffraat, Research grants, 2, Consulting fees, 5; G. Horneff, Research grants, 2, Speakers’ bureau, 8; J. Anton, Research grant, 2, Consulting fees, 5, Speakers’ bureau, 8; R. Berner, Research grants, 2; F. Corona, None; R. J. Cuttica, Roche, Abbott, Pfizer, Novartis, BMS, 8; M. Desjonqueres, None; M. Fischbach, None; M. Alessio, None; A. Chieng, None; W. Emminger, None; E. Haddad, None; K. Lheritier, Full-time, Novartis, 3, Novartis Pharmaceutical Corporation, 1; K. Abrams, Novaris Pharmaceutical corporation, 3, Novartis Pharmaceutical Corporation, 1; J. Hruska, Novartis, 3; D. J. Lovell, Research grants, 2, Consulting fees, 5, Employment, 3, Speakers’ bureau, 8; A. Martini, Research grants, 2, Speakers’ bureau, 8.

272 Predictive Markers Of Joint Damages Of Children With Systemic-Onset Juvenile Idiopathic Arthritis In Long-Term Course Of Treatment With Tocilizumab. Tomo Nozawa, Taichi Kanetaka, Kenichi Nishimura, Masako Kikuchi, Tomomi Sato, Nodoka Sakurai, Ryoki Hara, Kazuko Yamazaki and Shumpei Yokota. Yokohama City University School of Medicine, Yokohama, Japan Background/Purpose: Systemic-onset juvenile idiopathic arthritis (sJIA) is a subtype of chronic childhood arthritis characterized by spiking fever, rash, and arthritis. Tocilizumab (TCZ), an anti-IL-6 receptor monoclonal antibody, has dramatically improved patients’ inflammatory symptoms/signs as well as their prognosis. In the long-term treatment of s-JIA patients with TCZ, however, there found joint damages in some patients on X-ray examination although their clinical inflammatory symptoms/signs and laboratory findings were well-controlled. The objective of this study was to determine predictive markers of joint damages of s-JIA patients under the TCZ treatment. Methods: Although 62 patients have been treated with TCZ, 33 patients were excluded from the study because 17 patients were effectively treated with TCZ to be drug-off, and the duration of TCZ treatment in 16 patients was shorter than 48 months and/or progression of radiographic joint damages were not detectable. Twenty-nine patients (16 boys and 13 girls) with s-JIA were eligible in this study who have been treated with TCZ over 48 months. The following blood tests were determined before and 6, 12, 24, 36, 48 months after the TCZ administration: CBC, CRP/ESR, D-dimer, liver function tests, creatinine, and MMP-3. The radiographs were serially examined before and after TCZ treatment. Results: The mean age of disease onset was 5.2⫾2.9 years old. Of the 29 patients, 12 patients (41.3%) had radiographic progression of joint damages during the long-term TCZ treatment (JD (⫹)), and 17 patients (58.7%) had no joint damages on X-ray examinations (JD (⫺)) although there were no systemic inflammatory symptoms/signs demonstrable in both groups. Before TCZ treatment there were no significant differences of laboratory data between JD (⫹) and JD (⫺). By 12 months after TCZ treatment WBC counts were apparently more increased to normal range in JD (⫹) group than JD (⫺) group (p⬍0.05). By 24 months after TCZ treatment MMP-3 levels were still higher in JD (⫹) group than in JD (⫺) group (p⬍0.05). Taken together, although TCZ completely blocked clinically the inflammatory symptoms/ signs as well as laboratory changes of inflammation such as CRP and ESR, there found a part of patients who had progressive joint damages even under the TCZ treatment, suggesting some biological mechanisms which were IL-6-independent will play an important role in damaging joints. For the patients who complained of something different on their joints, the tapering of prednisolone tended to be delayed. As the results, the mean doses of prednisolone at 12 months were significantly higher in JD (⫹) group than JD (⫺) group. Conclusion: The long-lasting elevation of MMP-3 levels and higher WBC counts in patients with s-JIA treated with TCZ will be predictive of radiographic progression of joint damages. Disclosure: T. Nozawa, None; T. Kanetaka, None; K. Nishimura, None; M. Kikuchi, None; T. Sato, None; N. Sakurai, None; R. Hara, None; K. Yamazaki, None; S. Yokota, None.

S113

Sunday, October 27

Disclosure: H. Brunner, Novartis, Genentech, Medimmune, EMD Serono, AMS, Pfizer, UCB, Jannsen, 5, Genentech, 8; P. Quartier, Abbvie, Chugai-Roche, Novartis, Pfizer, 2, Abbott/Abbvie, BMS, Chugai-Roche, Novartis, Pfizer, Servier, SOBI, Speakers Bureau: Chugai-Roche, Novartis, Pfizer, 5; T. Constantin, None; S. Padeh, None; I. Calvo, None; M. Erguven, None; L. Goffin, None; M. Hofer, Novartis, 2; T. Kallinich, Novartis, 2, Roche, Novartis, ALK, 8; S. Oliveira, Novartis, 2; Y. Uziel, Novartis, 5; S. Viola, To Gaslini Hospital: Abbott, Astrazeneca, BMS, Centocor Research & Development, Eli Lilly and Company, “Francesco Angelini”, Glaxo Smith & Kline, Italfarmaco,Novartis, Pfizer Inc., Roche, Sanofi Aventis, Schwarz Biosciences GmbH, Xoma, Wyeth Pharmaceutical, 5; K. Nistala, None; C. Wouters, Novartis Belgium, Roche Belgium, GSK ImmunoInflammation USA, 2; K. Lheritier, Novartis, 1, Novartis, 3; J. Hruska, Novartis, 3; K. Abrams, Novaris Pharmaceutical corporation, 3, Novartis Pharmaceutical Corporation, 1; A. Martini, Abbott, Astrazeneca, Bristol Myers and Squibb, Janssen Biologics B.V., Ely Lilly and Company, “Francesco Angelini”, Glaxo Smith & Kline, Italfarmaco, Novartis, Pfizer Inc., Roche, Sanofi Aventis, Schwarz Biosciences GmbH, Xoma, Wyeth Pharmaceuticals Inc., 2, Abbott, Astrazeneca, Bristol Myers and Squibb, Janssen Biologics B.V., Ely Lilly and Company, “Francesco Angelini”, Glaxo Smith & Kline, Italfarmaco, Novartis, Pfizer Inc., Roche, Sanofi Aventis, Schwarz Biosciences GmbH, Xoma, Wyeth Pharmaceuticals Inc., 5, Astellas, Astrazeneca, Bristol Myers and Squibb, -Glaxo Smith & Kline, Italfarmaco, MedImmune, Novartis, 8; N. Ruperto, To Gaslini Hospital: Abbott, Astrazeneca, BMS, Centocor Research & Development, Eli Lilly and Company, “Francesco Angelini”, Glaxo Smith & Kline, Italfarmaco, Novartis, Pfizer Inc., Roche, Sanofi Aventis, Schwarz Biosciences GmbH, Xoma, Wyeth Pharmaceuti, 2, Astrazeneca, Bristol Myers and Squibb, Janssen Biologics B.V., Roche, Wyeth, Pfizer, 8; D. J. Lovell, NIH, 2, Astra-Zeneca, Centocor, Jansen, Wyeth, Amgen, Bristol-Meyers Squibb, Abbott, Pfizer, Regeneron, Hoffman La-Roche, Novartis, Genentech, 5, Genentech, Roche, 8.

Conclusion: Findings of the Phase III program of CAN in SJIA suggest that CS requirements are substantially reduced after as few as 4 injections. The CS-sparing potential of CAN appears to be great as CS were discontinued entirely in a sizeable proportion of patients with active SJIA in this study.

Sunday, October 27

273

274

Neutropenia With Tocilizumab Treatment Is Not Associated With Increased Infection Risk In Patients With Systemic Juvenile Idiopathic Arthritis. Fabrizio De Benedetti1, Hermine I. Brunner2, Eileen M. Baildam3, Ruben Burgos-Vargas3, Gerd Horneff4, Hans-Iko Huppertz3, Kirsten Minden5, Barry L. Myones2, Karen Onel2, Jianmei Wang6, Kamal N. Bharucha7, Daniel J. Lovell2, Alberto Martini8 and Nicolino Ruperto3. 1IRCCS Ospedale Pediatrico Bambino Gesu´, Rome, Italy, 2PRCSG, Cincinnati, OH, 3PRINTO, Genoa, Italy, 4Pediatric Rheumatology International Trials Organization (PRINTO)-Istituto Gaslini, Genova, Italy, 5pediatric rheumatology, Berlin, Germany, 6Roche, Welwyn Garden City, United Kingdom, 7Genentech, South San Francisco, CA, 8Istituto Giannina Gaslini, Genova, Italy.

Reported Macrophage Activation Syndrome In Patients With SystemicOnset Juvenile Idiopathic Arthritis Treated With Tocilizumab. Shumpei Yokota, Yasuhiko Itoh, Tomohiro Morio, Naokata Sumitomo and Seiji Minota. The Safety Evaluation Committee of Actemra® for JIA, Tokyo, Japan.

Background/Purpose: In the phase 3 TENDER trial of tocilizumab (TCZ) in patients with systemic juvenile idiopathic arthritis (sJIA), decreases in neutrophil count were commonly observed. The purpose of this analysis was to determine if neutropenia was associated with increased risk of infection and to investigate variables associated with development of neutropenia in patients treated with TCZ for up to 2 years in TENDER. Methods: One hundred twelve children with active, persistent sJIA were randomized 2:1 to receive TCZ by body weight (12 mg/kg ⬍30 kg or 8 mg/kg ⱖ30 kg) or placebo IV every 2 weeks for 12 weeks and continued in an ongoing TCZ open-label extension.1 Worst Common Toxicity Criteria (CTC) neutropenia grade (grade 1, ⱖ1.5 and ⬍2.0⫻109/L; grade 2, ⱖ1.0 and ⬍1.5⫻109/L; grade 3, ⱖ0.5 and ⬍1.0⫻109/L; grade 4, ⬍0.5⫻109/L) and lowest observed neutrophil count (109/L) were identified for each patient. Univariate linear regression analysis was performed to investigate association of patient characteristics with lowest observed neutrophil count. Rates of infections and serious infections (per 100 patient years [PY]) in periods ⫾15 days around grade1–2 neutropenia (22.9 PY) and around grade 3–4 neutropenia (5.5 PY) were compared to corresponding rates in periods with normal neutrophil count (173.6 PY). Results: Up to week 104, 64/112 patients (57.1%) had at least 1 episode of grade 1–4 neutropenia; worst grade: 1 (n⫽2), 2 (n⫽34), 3 (n⫽26), and 4 (n⫽2). Rates of infections and serious infections during period of normal neutrophil counts (276.5/100 PY [95% CI: 252.3, 302.3] and 11.5/100 PY [95% CI: 7.0, 17.8], respectively) were comparable to those observed ⫾15 days around grade 1–2 neutropenia (226.7/100 PY [95% CI: 169.3, 297.3]; 8.7/100 PY [95% CI: 1.1, 31.5]) and grade 3–4 neutropenia (292.5/100 PY [95% CI: 167.2, 475.0]; 0/100 PY), with no trend toward increased risk with higher grade neutropenia. Methotrexate (MTX) use (Yes/No) was significantly associated with lowest observed neutrophil count (difference: –0.575 [95% CI: –1.02, –0.13], p⫽0.012), with 62% of 77 patients receiving MTX vs 46% of 35 patients not receiving MTX having grade 1–4 neutropenia. Younger age was borderline associated with lowest observed neutrophil count (␤⫽0.04661, p⫽0.047). Concurrent use of glucocorticoids (GC) and TCZ exposure were not associated with lowest observed neutrophil count (p⬎0.3). Conclusion: No trend for association between neutropenia and increased risk of infections was observed in the TENDER trial. Background MTX, and somewhat younger age, was associated with increased risk of neutropenia, while TCZ exposure and concurrent GC use were not. Reference: 1. De Benedetti F et al. N Engl J Med 2012;367:2385. Disclosure: F. De Benedetti, Abbott, Pfizer, BMS, Roche, Novimmune, Novartis, SOBI, 2; H. I. Brunner, Novartis, Genentech, Medimmune, EMD Serono, AMS, Pfizer, UCB, Janssen, 5, Genentech, 8; E. M. Baildam, None; R. Burgos-Vargas, Abbott, BMS, Janssen, Pfizer, Roche, 5, Abbott, BMS, Janssen, Pfizer, Roche, 8, Abbott Immunology Pharmaceuticals, 2; G. Horneff, Abbott, Pfizer, 2; H. I. Huppertz, Abbott, Chugai, Pfizer, Roche, Swedish Orphan, 5; K. Minden, Pfizer, Abbvie, Roche, Chugai, Medac, 5, Pfizer, Abbvie, 2; B. L. Myones, None; K. Onel, None; J. Wang, Roche Pharmaceuticals, 3; K. N. Bharucha, Genentech and Biogen IDEC Inc., 3; D. J. Lovell, NIH, 2, AstraZeneca, Centocor, Janssen, Wyeth, Amgen, Bristol-Meyers Squibb, Abbott, Pfizer, Regeneron, Hoffmann-La Roche, Novartis, Genentech, 5, Roche, Genentech, 8; A. Martini, Abbott, AstraZeneca, BMS, Centocor, Lilly, Francesco Angelini, GSK, Italfarmaco, MerckSerono, Novartis, Pfizer, Regeneron, Roche, Sanofi Aventis, Schwarz Biosciences, Xoma, Wyeth, 2, Abbott, AstraZeneca, BMS, Centocor, Lilly, Francesco Angelini, GSK, Italfarmaco, MerckSerono, Novartis, Pfizer, Regeneron, Roche, Sanofi Aventis, Schwarz Biosciences, Xoma, Wyeth, 5, Abbott, Boehringer, BMS, Novartis, Astellas, Italfarmaco, MedImmune, Pfizer, Roche, 8; N. Ruperto, To Gaslini Hospital: Abbott, Astrazeneca, BMS, Centocor Research & Development, Eli Lilly and Company, “Francesco Angelini”, Glaxo Smith & Kline, Italfarmaco, Novartis, Pfizer Inc., Roche, Sanofi Aventis, Schwarz Biosciences GmbH, Xoma, Wyeth Pharmaceuti, 2, Astrazeneca, Bristol Myers and Squibb, Janssen Biologics B.V., Roche, Wyeth, Pfizer, 8.

Background/Purpose: Systemic-onset juvenile idiopathic arthritis (s-JIA) is a subtype of chronic childhood arthritis that is characterized by a spiking fever, rash, and arthritis. About 7% of s-JIA patients develop macrophage activation syndrome (MAS), a potentially fatal complication. Although several biologic therapies have dramatically improved the prognosis of s-JIA, the risk of developing MAS under biologic therapy seems not vanished because there are several reports of MAS associated with etanercept and canakinumab in patients with s-JIA. To evaluate the prevalence of MAS in patients with s-JIA under the treatment of tocilizumab. Methods: During the post-marketing surveillance of tocilizumab conducted by Chugai Pharmaceutical Co. Ltd. in Japan from 2008 to 2012, a total of 394 patients was registered. Tocilizumab, 8 mg/kg, was administered to the patients every 2 weeks, and routine investigations including CBC, CRP/ESR, D-dimer, liver function tests, creatinine, urinary beta 2-microglobulin, total cholesterol/triglyceride, and serum ferritin were determined before each infusion. Among them, 23 patients (5.8%) were diagnosed by the pediatrician in attendance with MAS, suspected MAS, or hemophagocytic syndrome; and then clinical course and laboratory data of the patients were reported as severe adverse events. Data were analyzed by the members of the Safety Evaluation Committee. Data were provided by Chugai Pharmaceutical Co. Ltd. Results: The Committee investigated the clinical manifestations and the changes of laboratory data along with the time course of each patient reported. Three patients were defined as definitive MAS (0.76%), 2 patients as EB virus-associated hemophagocytic syndrome (0.51%). Among other 18 patients, 11 patients (2.8%) were defined as probable MAS because the patients’ information including laboratory data was incomplete. Seven patients (1.8%) were judged as possible MAS or non-MAS because the treatment intervention for MAS was too early to define as MAS, or other diseases such as infectious disease, and hemolytic-uremic syndrome were suspected. Among 8 out of 14 patients defined as definitive or probable MAS, clinical manifestations were quite different from those seen in MAS on conventional therapy were. One patient died due to multi-organ failure, and none experienced sequelae. Conclusion: The frequency of definitive MAS under the tocilizumab therapy is estimated as 0.76%, and possible MAS as 2.8% (total 3.6%). Since the development of MAS in s-JIA patients under the conventional therapy was reported to be 6.8% ⬃ 13%, tocilizumab may have reduced the frequency of MAS developing in s-JIA. However, the clinical symptoms/signs such as fever, rash and an anguish look were not apparent in about half of the patients, and the inflammatory markers such as CRP and ESR were suppressed owing to the inhibition of IL-6 by tocilizumab. Taken together, physical examination and laboratory tests should be carefully taken at each visit in patients on tocilizumab in order to detect deadly MAS incipiently. Disclosure: S. Yokota, Chugai, 7; Y. Itoh, Chugai, 5; T. Morio, Chugai, 5; N. Sumitomo, Chugai, 5; S. Minota, Chugai, 7.

275 Changes In Serum IL-18 Level In Systemic Juvenile Idiopathic Arthritis Patients Who Attained Drug-Free Remission By Tocilizumab. Tomohiro Kubota1, Syuji Takei1, Yuichi Yamasaki1, Junko Yasumura2, Naomi Kuwada3, Yukiko Nonaka1, Tomoko Takezaki1, Tsuyoshi Yamatou1, Tomokazu Nagakura4, Yasuhito Nerome1, Hiroyuki Imanaka1 and Harumi Akaike1. 1Kagoshima University, Kagoshima, Japan, 2Hiroshima University, Hiroshima City, Japan, 3Kumamoto University, Kumamoto City, Japan, 4 House of Meguminoseibo, Usuki, Japan Background/Purpose: Tocilizumab (TCZ), anti-human interleukin-6 receptor monoclonal antibody, was the first biologic agent used in the treatment of systemic juvenile idiopathic arthritis (sJIA) in Japan. We have previously reported that long-term TCZ therapy could induce drug-free remission in sJIA patients resistant to conventional prednisolone (PSL) therapy1). On the other hand, extremely high levels of serum IL-18 (⬎10,000

S114

Conclusion: TCZ had the potential to induce drug-free remission in 1/4 of sJIA patients who had been refractory to the conventional therapy. Serum IL-18 level could be a useful bio-marker in tapering the TCZ therapy in sJIA. 1) Kubota T, et al: #267, 2011 ACR meeting 2) Shimizu M, et al 䡺FRheumatology 2010;49:1645–1653 Disclosure: T. Kubota, None; S. Takei, Chugai,Eisai, Takeda, Bristol-Myers, 2, Caugai, Eisai, Takeda, Abbvie, Astellas, Teijin, Novartis, Pfizer, Asahi Kasei, 8; Y. Yamasaki, None; J. Yasumura, None; N. Kuwada, None; Y. Nonaka, None; T. Takezaki, None; T. Yamatou, None; T. Nagakura, None; Y. Nerome, None; H. Imanaka, None; H. Akaike, None.

276 Assessment Of Radiographic Progression In Patients With Systemic Juvenile Idiopathic Arthritis Treated With Tocilizumab: 2-Year Results From The Tender Trial. Clara Malattia1, Nicolino Ruperto2, Elena Palmisani2, Silvia Pederzoli2, Angela Pistorio2, Hermine I. Brunner3, Rubeń J. Cuttica4, Inmaculada Calvo2, Stella Maris Garay2, Despina Eleftheriou2, Carine Wouters2, Jianmei Wang5, Clare Devlin5, Daniel J. Lovell3, Alberto Martini6, Fabrizio De Benedetti7 and Angelo Ravelli6. 1 Pediatria 2, Genoa, Italy, 2PRINTO, Genoa, Italy, 3PRCSG, Cincinnati, OH, 4 Pediatric Rheumatology International Trials Organization (PRINTO)-Istituto Gaslini, Genova, Italy, 5Roche, Welwyn Garden City, United Kingdom, 6Istituto Giannina Gaslini, Genova, Italy, 7IRCCS Ospedale Pediatrico Bambino Gesu´, Rome, Italy Background/Purpose: A phase 3 trial (TENDER) demonstrated the efficacy of the interleukin-6 receptor inhibitor tocilizumab (TCZ) in patients with systemic juvenile idiopathic arthritis (sJIA).1,2 The purpose of this analysis was to investigate progression of radiographic joint damage in patients with sJIA treated with TCZ for up to 2 years in TENDER. Methods: One hundred twelve patients 2–17 years old with active, refractory sJIA of ⱖ6 months’ duration and an inadequate response to previous non-steroidal anti-inflammatory drugs and oral corticosteroids were enrolled in TENDER. Patients were randomized 2:1 to receive TCZ according to body weight (12 mg/kg ⬍30 kg or 8 mg/kg ⱖ30 kg) or placebo IV every 2 weeks for 12 weeks. Patients then received open-label TCZ in the ongoing long-term extension. Radiographic progression was calculated as change in adapted Sharp/van der Heijde score (aSH) score and/or Poznanski score, assessed on hand and wrist radiographs, from baseline to weeks 52 and 104. Radiographic progression was indicated by a positive aSH score change or negative Poznanski score change. Clinical efficacy endpoints included American College of Rheumatology (ACR) Pediatric (Pedi) 70/90 responses. Results: Baseline and ⱖ1 postbaseline aSH and Poznanski scores were available for 47 and 33 patients, respectively (reasons for missing x-rays: early withdrawal, no consent, unreadable x-rays). Baseline characteristics for patients with radiographic data were similar to the whole TCZ population.1 Patients with assessable aSH/Poznanski scores had 5.2/4.8-year disease duration, 21.3/19.2 active joints, 20.0/18.2 joints with limitation of movement, and erythrocyte sedimentation rates of 53.9/59.2 mm/h. At weeks 52 and 104, 20 and 19 patients, respectively, had aSH progression, and 8 and 6 patients, respectively, had Poznanski score progression. Median change in aSH score from baseline to weeks 52 and 104 were 0 and 0.5, respectively (Table). Median change in Poznanski score from baseline to weeks 52 and 104 were 0.3 and 0.17, respectively (Table).

aSH score (n ⫽ 47), median (IQR) Poznanski score (n ⫽ 33), median (IQR) ACR Pedi 70 (n ⫽ 112), n/N (%) ACR Pedi 90 (n ⫽ 112), n/N (%)

Week 52

Week 104

0.00 (–8.70: 4.00) 0.30 (–0.02: 1.03) 92/106 (86.8) 67/106 (63.2)

0.50 (–7.50: 12.00) 0.17 (0.01: 1.04) 57/65 (87.7) 46/65 (70.8)

IQR, interquartile range.

Conclusion: Though changes in radiographic scores over time were seen in many patients, on average, patients with sJIA did not experience

S115

Sunday, October 27

pg/ml) was reported in active phase of sJIA2). Therefore, serum IL-18 levels were monitored in order to find the way how to taper and discontinue TCZ in sJIA patients. Methods: sJIA patients treated with 8 mg/kg of TCZ for more than 1 year who were initially resistant to 3 times of consecutive weekly methyl-PSL pulse therapy in active phase and/or had been refractory to long-term oral steroid therapy were recruited. Our tapering protocol for patients who maintained clinical remission by TCZ was as following; decrease PSL dose gradually to less than 0.2mg/kg/day, prolongation of TCZ interval from every 2w to 3w, discontinuation of PSL, and discontinuation of TCZ, in tern. Serum levels of IL-6/IL-18 were measured by ELISA. Results: A total of 28 sJIA patients were recruited in this study; their mean disease duration was 4.1 years, and mean PSL dose was 0.5mg/kg/day at initiating TCZ. Twenty-seven patients (96%) attained clinical remission with combination therapy of PSL⬍0.2mg/kg/day and TCZ every 3w. Thereafter, 10 patients (36%) completed PSL-free remission, and 7 (25%) attained drug-free remission by discontinued TCZ (Figure 1). No significant difference was found in TCZ treatment period between the two patient groups who achieved the drug-free remission (mean 7.0 yrs) and who maintained clinical remission (mean 5.5 yrs). Serum IL-6 decreased quickly in correlating with improvement of clinical symptoms after starting TCZ. On the contrary, serum IL-18 did not decrease in a patient who was resistant to TCZ therapy, or in patients who flared or developed to macrophage activation syndrome during the TCZ therapy. In patients who attained clinical remission by TCZ, however, IL-18 steadily decreased with treatment course (figure 2). In all patients maintained the drug-free remission for more than 2 years, serum IL-18 was ⬍ 1,000 pg/ml when the TCZ was discontinued.

Sunday, October 27

noticeable progression of radiographic damage over 2 years of treatment with TCZ.

Disclosure: S. Ringold, None; P. F. Weiss, None; R. A. Colbert, None; E. Morgan DeWitt, NIAMS-NIH, 2; T. C. Lee, None; K. Onel, None; S. Prahalad, None; R. Schneider, None; S. Shenoi, None; R. K. Vehe, None; Y. Kimura, None.

References: 1. De Benedetti F et al. N Engl J Med 2012;367:2385. 2. De Benedetti F et al. Ann Rheum Dis 2012;71(Suppl 3):425.

278

Disclosure: C. Malattia, None; N. Ruperto, To Gaslini Hospital: Abbott, Astrazeneca, BMS, Centocor Research & Development, Eli Lilly and Company, “Francesco Angelini”, Glaxo Smith & Kline, Italfarmaco, Novartis, Pfizer Inc., Roche, Sanofi Aventis, Schwarz Biosciences GmbH, Xoma, Wyeth Pharmaceuti, 2, Astrazeneca, Bristol Myers and Squibb, Janssen Biologics B.V., Roche, Wyeth, Pfizer, 8; E. Palmisani, None; S. Pederzoli, None; A. Pistorio, None; H. I. Brunner, Novartis, Genentech, Medimmune, EMD Serono, AMS, Pfizer, UCB, Jannsen, 5, Genentech, 8; R. J. Cuttica, Roche, Abbott, Pfizer, Novartis, BMS, 8; I. Calvo, None; S. M. Garay, None; D. Eleftheriou, None; C. Wouters, None; J. Wang, Roche Pharmaceuticals, 3; C. Devlin, Roche Pharmaceuticals, 3; D. J. Lovell, NIH, 2, AstraZeneca, Centocor, Janssen, Wyeth, Amgen, Bristol-Meyers Squibb, Abbott, Pfizer, Regeneron, Hoffmann-La Roche, Novartis, Genentech, 5, Roche, Genentech, 8; A. Martini, Abbott, AstraZeneca, BMS, Centocor, Lilly, Francesco Angelini, GSK, Italfarmaco, MerckSerono, Novartis, Pfizer, Regeneron, Roche, Sanofi Aventis, Schwarz Biosciences, Xoma, Wyeth, 2, Abbott, AstraZeneca, BMS, Centocor, Lilly, Francesco Angelini, GSK, Italfarmaco, MerckSerono, Novartis, Pfizer, Regeneron, Roche, Sanofi Aventis, Schwarz Biosciences, Xoma, Wyeth, 5, Abbott, Boehringer, BMS, Novartis, Astellas, Italfarmaco, MedImmune, Pfizer, Roche, 8; F. De Benedetti, Abbott, Pfizer, BMS, Roche, Novimmune, Novartis, SOBI, 2; A. Ravelli, None.

277 Childhood Arthritis and Rheumatology Research Alliance (CARRA) Standardized Consensus Treatment Plans for New Onset Polyarticular Juvenile Idiopathic Arthritis. Sarah Ringold1, Pamela F. Weiss2, Robert A. Colbert3, Esi Morgan DeWitt4, Tzielan C. Lee5, Karen Onel6, Sampath Prahalad7, Rayfel Schneider8, Susan Shenoi9, Richard K. Vehe10 and Yukiko Kimura11. 1Seattle Children’s Hospital, Seattle, WA, 2The Children’s Hospital of Philadelphia, Philadelphia, PA, 3NIAMS NIH, Bethesda, MD, 4 Cincinnati Children’s Hospital Medical Center, Cincinnati, OH, 5Stanford University School of Medicine, Stanford, CA, 6PRCSG, Cincinnati, OH, 7 Emory Children’s Center, Atlanta, GA, 8Pediatric Rheumatology Collaborative Study Group (PRCSG), Cincinnati, OH, 9Seattle Children’s Hospital, University of Washington, Seattle, WA, 10University of Minnesota, Minneapolis, MN, 11Joseph M Sanzari Children’s Hospital, Hackensack University Medical Center, Hackensack, NJ Background/Purpose: There is no standardized approach to the initial treatment of polyarticular juvenile idiopathic arthritis (pJIA) among North American pediatric rheumatologists. Understanding the comparative effectiveness of the diverse therapeutic options available will result in better health outcomes for children with pJIA. The Childhood Arthritis and Rheumatology Research Alliance (CARRA) developed consensus treatment plans (CTP) for use in clinical practice to facilitate such studies. Methods: A case-based survey was administered to CARRA members to identify the most common treatment approaches for new-onset pJIA. Two face-to-face consensus meetings employed modified nominal group technique to agree on the treatment plans, operational case definitions, endpoints and data elements to be collected. A core workgroup reviewed the literature, refined the plans and developed medication dosing and monitoring recommendations. The final CTPs were presented for approval at a face-to-face meeting in April 2013. Results: Sixty percent of 230 members answered the survey, which identified significant variability in treatment for new onset pJIA. 72 (2011) and 58 (2012) CARRA members attended consensus meetings to develop the 3 CTPs based on treatment strategies that focus on timing of introduction of medication categories (biologic and non-biologic DMARDs) for the first 12 months of therapy. Case definitions and clinical and laboratory monitoring schedules were established. The 3 CTPs include a Step-Up CTP (non-biologic with the addition of a biologic DMARD after 3–6 mos), Early Combination CTP (non-biologic plus biologic DMARD as initial treatment), and a Biologic Only CTP. This approach was approved by 96% of the 72 CARRA JIA Committee members attending the 2013 CARRA meeting. Conclusion: Standardized CTPs based on 3 treatment timing strategies were developed for new-onset pJIA. Coupled with data collection at defined intervals, use of these CTPs will enable studying their comparative effectiveness in an observational setting like the CARRA Registry to optimize initial management of pJIA. Based on a high level of agreement among pediatric rheumatologists during the development process, sufficient participation and enrollment in all CTPs is anticipated.

Results Of a 24 Month Extension Study In Patients Who Participated In The Trial Of Early Aggressive Therapy In Polyarticular Juvenile Idiopathic Arthritis. Carol A. Wallace1, John Bonsack2, Steven J. Spalding3, Hermine Brunner4, Kathleen M. O’Neil5, Diana Milojevic6, Sarah Ringold7, Laura E. Schanberg8, Gloria C. Higgins9, Beth S. Gottlieb10, Joyce J. Hsu11, Marilynn G. Punaro12, Yukiko Kimura13 and Audrey F. Hendrickson14. 1Seattle Childrens Hosp & Research Institute, Seattle, WA, 2University of Utah Health Sciences Center, Salt Lake City, UT, 3The Cleveland Clinic, Cleveland, OH, 4Cincinnati Children’s Hospital Medical Center, Cincinnati, OH, 5Riley Hospital for Children, Indianapolis, IN, 6University of California, San Francisco, San Francisco, CA, 7Seattle Children’s Hospital, Seattle, WA, 8Duke University Medical Center, Durham, NC, 9Nationwide Childrens Hosp, Columbus, OH, 10Cohen Children’s Medical Center of New York, New Hyde Park, NY, 11Stanford University, Palo Alto, CA, 12Texas Scottish Rite Hospital, Dallas, TX, 13Joseph M Sanzari Children’s Hospital, Hackensack University Medical Center, Hackensack, NJ, 14 Seattle Children’s Research Institute, Seattle, WA Background/Purpose: The Trial of Early Aggressive Therapy in Polyarticular Juvenile Idiopathic Arthritis (TREAT-JIA) was a double-blind, randomized, placebo-controlled trial in 85 patients with onset of polyarticular JIA ⬍ 12 mos to determine if clinical inactive disease (CID) could be achieved by 6 months. Two aggressive treatment arms were used; (i) (MTX arm) SQ methotrexate (MTX; 0.5 mg/kg/wk) or (ii) (M-E-P arm ) etanercept (ETN 0.8 mg/kg/wk), prednisolone (0.5 mg/kg/d tapered to zero by 17 wks) and MTX at the dose above. The purpose of this investigation was to follow these children beyond the original study for 2 additional years to understand the longer-term outcomes. Methods: Patients who completed a minimum of 6 mos in the TREAT study were eligible to enroll in this extension study, regardless of response during the original trial and whether or not they continued to receive the same medications. Patients were treated as per standard of care. Physician, patient/parent and laboratory reported measures of disease status and safety information were collected at clinic visits every 3 mos for up to 2 years. Information regarding disease status and safety during the time period between studies was collected from chart review. Adverse events grade 3 and higher as well as infections requiring systemic therapy were reported. Results: Twelve of the 15 original TREAT study sites participated and enrolled 52 of 77 (67.5%) eligible patients, 48 of whom returned for follow-up visits. TREAT baseline demographic and disease characteristics as well as disease state at the end of the TREAT trial did not differ between those who participated in the extension and those who did not. At enrollment into the extension study, 21 (44%) were receiving ETN and MTX, 13 (27%) MTX alone, 6 (12%) ETN alone, 7 (15%) no meds or NSAIDs alone, 1 patient each on prednisone, adalimumab and abatacept. Twenty-five (52%) entered the extension study in CID, while 23 (48%) had active disease. Patients were followed for a mean of 21.4 mos (range 9 to 24) and 27 (56%) patients spent more than 50% of their follow up time in CID. Eight patients were in CID ⬎12 months and 2 were in CID off meds for the entire study. Disease activity during periods of AD tended to be low with means of MD global of 2.4; active joint count of 3.5; parent global evaluation of 2.4; CHAQ of 0.32; ESR 19; and morning stiffness of 23 minutes. Patients who were RF(⫺) tended to spend more study time in CID than RF(⫹) patients (52% vs 43%), as did patients who were ANA (⫺) 58% vs 47% ANA (⫹). Patients who achieved CID at 6 months in the TREAT study tended to spend more time in CID (58% vs 45%) than did those who ended the TREAT study in AD (55% vs 43%). There were no serious adverse events or adverse events grade 3 or higher reported. Four patients had 6 infections requiring systemic antibiotics. Conclusion: Early aggressive therapy in this group of polyarticular JIA patients, with high initial disease activity and proportion with RF positivity, was associated with prolonged periods of CID in the majority of patients during this 24 month extension study. Those not in CID had low levels of disease activity. Disclosure: C. A. Wallace, Amgen, 2, Pfizer Inc, 2, Novartis Pharmaceutical Corporation, 5; J. Bonsack, Novartis Pharmaceutical Corporation, 5; S. J. Spalding, None; H. Brunner, Novartis, Genentech, Medimmune, EMD, Serono, AMS, Pfizer, UCB, Janssen, 5, Genentech and Biogen IDEC Inc., 8; K. M. O’Neil, UCB, 5; D. Milojevic, None; S. Ringold, None; L. E. Schanberg, Novartis Pharmaceutical Corporation, 9, Lilly, 5, UCB, 5, Amgen, 9, BMS, 9, SOBI, 9, Pfizer Inc, 9; G. C. Higgins, None; B. S. Gottlieb, None; J. J. Hsu, None; M. G. Punaro, None; Y. Kimura, None; A. F. Hendrickson, None.

S116

280

Safety and Effectiveness Of Adalimumab In Children With Polyarticular Juvenile Idiopathic Arthritis Aged 2 To 4 Years Weighing normal at screening)‡ C4 (subj > normal at screening)§

Atacicept 75 mg (n ⴝ 112) 24-wk FU

Placebo (n ⴝ 111) Wk 52 24-wk FU 3.51 1.16 ⫺3.13 ⫺5.63 9.30 ⫺34.85

⫺7.01 ⫺16.22 ⫺29.27 ⫺66.67 ⫺50.00 ⫺55.00

Atacicept 150 mg (n ⴝ 62) 24-wk FU

Wk 52

⫺37.62 ⫺57.89 ⫺68.75 ⫺81.47 ⫺28.21 ⫺67.08

⫺9.24 ⫺21.96 ⫺24.04 ⫺66.67 ⫺40.00 0.00

Wk 52

12-wk FU

Wk 52

12-wk FU

Wk 52

12-wk FU

13.89

41.42

ⴚ31.46

ⴚ20.85

ⴚ37.86

ⴚ11.78

4.13

3.56

7.15

3.01

15.38

5.43

⫺0.44

0.00

42.71

18.18

49.50

48.94

* Flow cytometry population: placebo, n ⫽ 37; atacicept 75 mg, n ⫽ 38; atacicept 150 mg, n ⫽ 25; †Anti-dsDNA ⬎ ⫽ 30 IU/ml; placebo, n ⫽ 71; atacicept 75 mg, n ⫽ 62; atacicept 150 mg, n ⫽ 43; ‡C3 ⬍0.9 g/L; placebo, n ⫽ 33; atacicept 75 mg, n ⫽ 53; atacicept 150 mg, n ⫽ 28; §C4 ⬍0.1 g/L; placebo, n ⫽ 28; atacicept 75 mg, n ⫽ 32; atacicept 150 mg, n ⫽ 17; Wk ⫽ week; FU ⫽ follow-up; IgG ⫽ immunoglobulin G; IgA ⫽ immunoglobulin A; IgM ⫽ immunoglobulin M.

Conclusion: Atacicept caused declines in IgG, IgA and IgM, as well as naı¨ve and memory B cells. With the exception of memory B cells, these levels returned towards baseline after therapy. Atacicept also had favorable effects on anti-dsDNA and complement levels. The exploratory post-hoc analysis of potential predictive biomarkers identified a subgroup of patients with higher BLyS and APRIL levels at baseline who were more likely to benefit from treatment. In addition, patients with the largest decrease in Ig levels and/or naı¨ve B cells demonstrated a greater clinical response. Further studies are required to clarify the associations between biomarkers and clinical response to atacicept. Disclosure: D. A. Isenberg, Merck Serono, 5; D. Wofsy, Merck Serono, BristolMyers Squibb, Genentech, 5, GlaxoSmithKline, 9; Y. Li, Employed by EMD Serono, 3; D. Licu, Merck Serono S.A., 3; S. D. Wax, Employed by EMD Serono, 3; C. Gordon, Merck Serono, 5, Merck Serono, 9; C. Pena Rossi, Employed by Merck Serono S.A., 3.

2552 Predictive Significance Of ANTI-Prothrombin Antibodies For Second and Third Trimester Obstetric Complications In Patients With Systemic LUPUS Erythematosus and ANTI-Phospholipid Syndrome. Marta Mosca1, Chiara Tani1, Martina Fabris2, Francesca Strigini3, Linda Carli4, Sabrina Vagnani1, Rosaria Talarico4, Chiara Baldini4, Alessandra Della Rossa4, Elio Tonutti5 and Stefano Bombardieri1. 1Rheumatology Unit, University of Pisa, Pisa, Italy, 2Institute of Clinical Pathology, Udine, Italy, 3 University of Pisa, Pisa, Italy, 4Rheumatology Unit, Pisa, Italy, 5Santa Maria Hospital, Udine, Italy. Background/Purpose: Anti-prothrombin antibodies (aPT) either alone or in association with other anti-phospholipid antibodies have been associated with an increased risk of thrombosis as well as of adverse obstetric outcomes particularly, recurrent early miscarriages. To assess whether anti-prothrombin antibodies have a predictive significance for the occurrence of late obstetric complications and maternal events in patients with SLE and APS. Methods: Sera from pregnant patients with SLE and APS collected during the first trimester of pregnancy were examined for the presence of anti-prothrombin antibodies (aPT), anticardiolipin antibodies (aCL), anti-B2GPI antibodies, Lupus anticoagulant (LA). Pregnancies were prospectively followed. The following obstetric complications were considered: preterm delivery, pre-eclampsia, eclampsia, intrauterine growth retardation, intrauterine fetal death. In addition the occurrence of maternal thrombotic complications was recorded. Results: Seventy six pregnancies in 60 patients (18 APS and 52 SLE) with a duration ⬎10 weeks were included in the analysis. Antiphospholipid antibodies were positive in 26 patients (aCL 25, LA 17, aPT 16, anti- B2GPI 6). Pregnancy outcomes are reported in the table 1. A significant correlation was observed between the presence of anti-prothrombin antibodies and the occurrence of obstetric complications (p⬍0.05). The positivity of anti-prothrombin antibodies in addition to either positive aCL, anti-B2GPI and/or LA increased the risk of developing late obstetric complications (p⬍ 0.001). No correlations were observed between the presence of anti-prothrombin antibodies alone or multiple anti-phospholipid antibodies and the occurrence of maternal thrombotic events.

S1089

Tuesday, October 29

Pharmacodynamics and Predictive Biomarkers In Patients Treated With Atacicept: Data From The APRIL-SLE Trial. David A. Isenberg1, David Wofsy2, Yong Li3, Daiana Licu4, Stephen D. Wax3, Caroline Gordon5 and Claudia Pena Rossi4. 1University College London, London, United Kingdom, 2University of California San Francisco and NIAID Autoimmunity Centers of Excellence, San Francisco, CA, 3EMD Serono, Rockland, MA, 4 Merck Serono S.A., Geneva, Switzerland, 5University of Birmingham, Birmingham, United Kingdom.

Table. Median percent change from baseline; Treatment Completer population

IUFD Perinatal death Preterm delivery IUGR Pre-eclampsia HELLP Maternal thrombotic manifestations

SLE

APS

Whole group

2 0 17 7 7 0 2

3 1 12 1 4 2 3

5 1 29 8 11 2 5

Conclusion: The presence of aPT antibodies appears associated with an increased risk for late obstetric complications in SLE and APS and therefore their assessment is suggested in these patients at the beginning of pregnancy. Disclosure: M. Mosca, None; C. Tani, None; M. Fabris, None; F. Strigini, None; L. Carli, None; S. Vagnani, None; R. Talarico, None; C. Baldini, None; A. Della Rossa, None; E. Tonutti, None; S. Bombardieri, None.

2553

Tuesday, October 29

The Clinical Value Of Testing For Aphl, a New ELISA Kit With a Unique Phospholipid Mixture In Patients With Systemic Lupus Erythematosus (SLE). Savino Sciascia1, Giovanni Sanna2, Veronica Murru3, Munther A. Khamashta3 and Maria Laura Bertolaccini3. 1Centro di Immunopatologia e Documentazione su Malattie rare, Torino, Italy, 2Louise Coote Lupus Unit, St. Thomas’ Hospital, London, United Kingdom, 3Lupus Research Unit, The Rayne Institute, Kings College London School of Medicine, London, United Kingdom. Background/Purpose: Antibodies directed to APhL (a mixture of phospholipids) have been reported to predict APS more reliably than aCL tests. We designed this study to evaluate the performance characteristics of the IgG and IgM APhL ELISA, in an attempt to establish the diagnostic utility of these new aPL antibody tests in a wide cohort of SLE patients. Methods: This study included 158 patients (152 women, mean age 41.8⫾11.6 years, mean disease duration 11.7⫾7.8years), all fulfilling the 1982 criteria for SLE. All the patients were tested for the routinely used aPL, including LA, aCL and ab2GPI. Data were compared with that obtained by the APhL ELISA (Louisville APL Diagnostic, Inc, Louisville, KY, USA). The diagnostic accuracy for each test was assessed by ROC and their area under the curve (AUC) analysis. Results: APhL were found in 77% of the patients. IgG and IgM APhL were more frequently found along with routine aPL than in isolation (p⫽0.007 and p⫽0.04, respectively). APhL were more frequently found in patients with definite APS than in those without (p⫽0.001). aPhL were not only more frequent in patients with thrombosis than in those without (p⫽0.02) but their titres were also significantly higher in thrombosis, when compared to those without (p⫽0.02). APhL showed an OR for thrombosis of 1.46 [95%CI1.08–3.19] (AUC 0.533). Twenty-seven patients (17.09%) were positive for isolated APhL. Of those 8 (29.6%) had a thrombotic event (6 arterial, 1 venous, 1 both). No associations were found between the presence of APhL and pregnancy loss. The presence of aPhL did not correlate with that of any of the other aPL tested. Multivariate analysis confirmed the associations with APS, particularly with thrombosis. Conclusion: APhL are frequent in patients with SLE. Their presence is associated with thrombosis, making these antibodies a novel marker for APS. The finding of isolated APhL in thrombosis (particularly arterial) deserves further investigation. Disclosure: S. Sciascia, None; G. Sanna, None; V. Murru, None; M. A. Khamashta, None; M. L. Bertolaccini, None.

2554 In Vitro Fertilization In Systemic Lupus Erythematosus and Antiphospholipid Syndrome: A Series Of 82 Cycles. Pauline Orquevaux1, Agathe Masseau2, Ve´ronique le Guern3, Vanessa Gayet3, Danie`le Vauthier-Brouzes4, Du Boutin5, Bertrand Wechsler6, Nathalie Morel7, Jean Loup Pennaforte1, Jean-Charles Piette8 and Nathalie Costedoat-Chalumeau9. 1CHU Reims, Reims, France, 2Nantes University Hospital, Nantes, France, 3Cochin Hospital, Paris, France, 4Pitie´ Salpe´trie`re Hospital, Paris, France, 5Assistance Publique-Hoˆpitaux de Paris, Hopital Pitie´-Salpe´trie`re, Paris, France, 6Hoˆpital Pitie´-Salpeˆtrie`re, AP-HP, UPMC Univ Paris 06 & French National Reference Center For Systemic Lupus and Antiphospholipid Syndrome, Paris, France, 7 COCHIN, Paris, France, 8CHU Pitie´-Salpeˆtrie`re, Paris, France, 9Hopital Cochin, Paris, France.

Background/Purpose: Patients with systemic lupus erythematosus (SLE) and antiphospholipid syndrome (APS) undergoing in vitro fertilization (IVF) are at increased risk of both hormone-associated flare and thrombosis. The literature is scarce with two series of IVF performed in 19 and 21 patients respectively [1; 2]. We report our experience with in vitro fertilization in 34 women with SLE and/or APS. Methods: Retrospective study of women followed in four French centers (Pitie´-Salpeˆtrie`re, Cochin, Nantes and Reims) who (1) had a diagnosis of SLE (ACR criteria) and/or APS (Sydney criteria), and (2) underwent at least one cycle of IVF between 2003 and 2012. Results: The diagnosis of the 34 included women was: SLE alone (n⫽9, including one case diagnosed during the IVF), SLE associated with antiphospholipid antibodies (n⫽9), SLE associated with APS (n⫽5), and primary APS (n⫽11, including one case diagnosed during the IVF). These women underwent 82 cycles of IVF. Underlying causes of infertility were of female origin (42%), male origin (33%), mixed (21%) or unexplained (4%). There was no premature ovarian insufficiency due to cyclophosphamide. Median age at IVF was 34.7 years (range, 22–45). Median number of IVF cycles was 2.4 (1–7). 72 cycles (88%) of IVF were allowed and supervised by an internist. Women were treated with hydroxychloroquine (52%), steroids (61%), aspirin (79%) and/or low-molecular-weight heparin (67%). Ovulation induction protocols varied according to the centers. Eight IVF cycles (10%) resulted in complications: SLE flare in 4 (three joint flares and one lupus enteritis) and thrombosis in 4. Interestingly, one SLE flare occurred in a patient with unknown SLE and 2 flares and 2 thromboses were explained by poor adherence to treatment. No ovarian hyperstimulation syndrome was observed. 24 pregnancies (29%) occurred, including four twin pregnancies, and lead to 22 live births (92% of pregnancies), 1 miscarriage and one medical termination for trisomy 13. In addition, during the follow-up, eight spontaneous pregnancies occurred. Eventually, a total of 24 patients (70%) delivered at least one healthy child. Conclusion: SLE flare and thrombosis were low (10%) and were often explained by poor adherence to treatment or absence of treatment. These preliminary results confirm that IVF can be successfully performed in SLE and/or APS women providing they have adequate treatment. The new protocols using GnRH antagonists may further decrease those risks. [1] Le Thi Huong D et al. Semin Arthritis Rheum 2002; 32: 174–188 [2] Guballa N et al., Arthritis Rheum 2000; 43: 550–556 Disclosure: P. Orquevaux, None; A. Masseau, None; V. le Guern, None; V. Gayet, None; D. Vauthier-Brouzes, None; D. Boutin, None; B. Wechsler, None; N. Morel, None; J. L. Pennaforte, None; J. C. Piette, None; N. Costedoat-Chalumeau, None.

2555 Development and Validation Of a Questionnaire To Screen For The Presence Of Autoimmune Disease. Alan Boroway1, Aileen M. Davis2, Carolina Landolt-Marticorena1 and Joan E. Wither2. 1Toronto Western Hospital, University of Toronto, Toronto, ON, 2Toronto Western Research Institute, Toronto Western Hospital, University of Toronto, Toronto, ON. Background/Purpose: Healthy family members of patients with systemic autoimmune rheumatic diseases (SARDs) are at higher risk of developing autoimmune disease than the general population. It is postulated that serological changes precede the onset of clinical autoimmune disease in affected individuals. Our aim is to determine if these serological changes are present in the relatives of patients with SARD, and if so, whether they can be used to predict the development of autoimmune disease. This would permit introduction of treatment early in the disease course thereby preventing damage. We have previously obtained blood for serologic testing and performed a cellular analysis on the first-degree relatives of patients with systemic lupus erythematosus. At the time of blood draw, a questionnaire was administered seeking evidence of autoimmune disease. In total 893 medical histories were obtained via questionnaire, with 178 reporting an autoimmune disease. However, less than 50% were able to be confirmed as accurate (Cooper et al, J Rheumatol. 2008 35 (10): 2001–4). In this study, we developed a validated questionnaire that more accurately reflects the presence of autoimmune disease, that can be used for a follow-up of the first-degree relatives of patients with SARDs. Methods: A questionnaire was developed in consultation with an expert for distribution to patients for validation. In contrast to our previous questionnaire it incorporates questions that have been developed to confirm the

S1090

presence of disease through questions regarding specific clinical manifestions of disease and/or treatment. After obtaining informed consent, the questionnaire was administered to patients with each autoimmune disease of interest, with those having other autoimmune diseases or osteoarthritis serving as controls. Accuracy at identifying the presence or absence of a particular autoimmune disease was confirmed against data from patients’ medical records. False positive (FP) and false negative (FN) rates were compared for each autoimmune disease. Results: 63 patients were consented and completed the questionnaire, including at least 5 patients with each autoimmune disease of interest. Results are shown below: Comparison of Accuracy of Original Questionnaire (Cooper et al) and New Questionnaire

Systemic Lupus Erythematosus Rheumatoid Arthritis Systemic Sclerosis Polymyositis/Dermatomyositis Sjogren’s Syndrome Antiphospholipid Syndrome Hemolytic Anemia Multiple Sclerosis Thyroid Disease Type 1 Diabetes Idiopathic Thrombocytopenic Purpura

Cooper et al Questionnaire FP Rate N (%)

Questionnaire FP Rate (Screening Question) N (%)

Questionnaire FP Rate (Confirmatory Questions) N (%)

Questionnaire FN Rate (Confirmatory Questions) N (%)

9/24 (32.5%) 45/60 (75%) 1/4 (25%) 2/2 (100%) 3/9 (33.3%) 4/5 (80%) 12/13 (92.3%) 2/4 (50%) 45/96 (46.9%) N/A N/A

0/18 (0%) 6/12 (50%) 0/13 (0%) 6/10 (60%) 3/9 (33.3%) 2/5 (40%) 2/7 (28.6%) 1/6 (16.7%) 5/12 (41.7%) 0/10 (0%) 3/7 (43%)

— 1/12 (8.3%) — 1/10 (10%) 1/9 (11.1%) 1/5 (20%) 2/7 (28.6%) 1/6 (16.7%) 0/12 (0%) — 1/7 (14.3%)

— — — — — — 4/56 (7.1%) 1/57 (1.8%) — — 1/56 (1.8%)

Disclosure: A. Boroway, None; A. M. Davis, None; C. Landolt-Marticorena, None; J. E. Wither, None.

2556 Concordance Of Indirect Immunofluorescence and Multiplex Immunoassay For Measurement Of Antinuclear Antibodies. Caroline D’Souza1, Donald L. Kimpel1, Walter Oliveira2 and Janet E. Lewis1. 1University of Virginia School of Medicine, Charlottesville, VA, 2University of Virginia, Department of Pathology, Medical Laboratories, Charlottesville, VA. Background/Purpose: The indirect immunofluorescence assay (IIF) is the gold standard for ANA testing. However, many laboratories now use multiplex assay as an ANA screening test. Multiplex ANA assay tests for a limited number of autoantibodies (antibodies to dsDNA, chromatin, ribosomal P, SSA-52, SSA-60, SSB, Sm, SmRNP, RNP-A, RNP-68, Scl-70, Jo-1, centromere B) while IIF allows the detection of antibodies to a much larger number of autoantigens. There is a concern that reliance on multiplex analysis as a screening test might lead to cases where a positive ANA, and hence a diagnosis, is missed. Our study aims to evaluate the concordance between IIF and multiplex assay for ANA screening. Methods: ANA is tested by IIF at UVA. Samples are considered positive at a titer of 1:80. Thirteen specific autoantibodies are tested by an automated multiplex platform. For this study, samples with at least one autoantibody positive on multiplex analysis were considered ANA ⫹ by multiplex. A total of 545 blood samples sent to the UVA lab between June and November 2012 to be tested for ANA and specific autoantibodies were analyzed. Results: About 25.3% are ANA positive by IIF and 24.7% are ANA positive by multiplex. Only 13.8% of samples are positive by both methods (concordant). Kappa coefficient of agreement is 0.39 (fair). The most common antibody in all multiplex⫹ samples is that to RNP followed by SSA and dsDNA. In the subset of multiplex ⫹ samples that were IIF negative, RNP was the most common antibody (40%). The most common pattern overall was homogenous. When the threshold for a positive ANA by IIF was increased to 1:160, the number of total samples that were IIF positive is reduced as expected (20.4%). Kappa coefficient of agreement was improved only to a value of 0.428 (moderate). A large number (74%) of the low titer (1:80) IFA assays were also multiplex negative. The majority (84.7%) of strongly positive IIF samples (1:640 and above) were

Multiplex I I F

ⴚ ⴙ

ⴚ 347 (63.6%) 63 (11.5%)

ⴙ 60 (11%) 75 (13.8%)

Fig 1: Concordance chart for ANA tested by IIF and Multiplex with a positive cutoff for IIF at a titer of 1:80.

Conclusion: We found a high rate of ANA positivity by both IIF and multiplex. Most of the discordant samples were either low titer IIF or had autoantibodies that might be less specific like anti-RNP. The agreement between the two tests as measured by a kappa co-efficient was only fair at a cutoff of 1:80 and improved to moderate at 1:160. This suggests that there is at best moderate agreement between the two tests. Also, in some cases, multiplex did not identify samples with high titer ANAs. We conclude that there is still a role for IIF to be used for ANA screening in patients suspected to have an autoimmune disease. Disclosure: C. D’Souza, None; D. L. Kimpel, None; W. Oliveira, None; J. E. Lewis, None.

2557 Fluctuations In sVCAM-1 and Adiponectin Mirror Fluctuations In Disease Activity In Lupus, But Cannot Be Use To Accurately Predict Impending Changes In Disease State. Carolina landolt-Marticorena1, Stephenie Prokopec2, Stacey Morrison3, Heather Reich3, James Scholey3, Dafna Gladman4, Murray B. Urowitz4, Paul Boutros2, Paul R. Fortin5 and Joan E. Wither6. 1Toronto Western Hospital, University of Toronto, Toronto, ON, 2Ontario Institute for Cancer Research, Toronto, ON, 3The Toronto Western Hospital, Toronto, ON, 4University of Toronto, Toronto Western Hospital, Toronto, ON, 5University of Laval, Quebec, QC, 6 Toronto Western Research Institute, Toronto Western Hospital, University of Toronto, Toronto, ON. Background/Purpose: Previous reports have identified a number of pro-inflammatory cyto/chemokines as candidate activity-specific SLE biomarkers. In our preliminary studies we found that out of a panel of 20 common cyto/chemokines examined, only MCP-1, IP-10, sVCAM and adiponectin were preferentially elevated in patients with active as compared to inactive disease. In this study, the relationship between fluctuations in disease activity and these four cyto/chemokines was examined to determine their utility in predicting fluctuations (exacerbation or improvement) in disease activity. Methods: SLE patients (n ⫽ 55) satisfying 4 or more ACR criteria were recruited and followed over a 14-month period with a minimum of 3 clinical and biochemical assessments over that time. The plasma concentration of the 4 analytes was determined by Luminex assay. Disease activity was determined by the SLEDAI-2K (S-2K). A modified SLEDAI-2K (mS-2K) was calculated by subtracting the contribution of anti-dsDNA antibodies and complement (C3) from the global score. Variation between visits was examined with a change in S-2K of ⫾ 4 between two consecutive visits deemed a clinically significant event. Event classification was compared with analyte levels using a one-way ANOVA followed by visualization. The ability of analyte levels to forecast disease activity was evaluated using a leave-one-out crossvalidation analysis. Results: Analysis of all available assessments showed a statistically significant positive correlation between adiponectin (Spearman’s rho (r) ⫽ 0.25, p ⬍ 1 ⫻ 10⫺3), VCAM (r ⫽ 0.24, p ⬍ 1 ⫻ 10⫺3) and IP-10 (r ⫽ 0.21, p ⫽ 2 ⫻ 10⫺3) and global disease activity as defined by the mS-2K. These correlations were comparable to the performance of traditional biomarkers; anti-dsDNA antibodies (r ⫽ 0.22, p ⬍ 1 ⫻ 10⫺3) and complement (C3, r ⫽ ⫺0.24, p ⬍ 1 ⫻ 10⫺3). The relationship between meaningful fluctuations in disease status between consecutive visits and biomarker levels was examined. Changes in disease state were exclusively reflected in changes in sVCAM-1 (p ⫽ 0.04) and adiponectin (p ⫽ 0.05). Both traditional biomarkers (anti-dsDNA antibodies and C3) and other tested cytokines (MCP-1, IP-10) did not reflect fluctuations in disease state. The ability of these biomarkers to predict a clinically meaningful change in disease status within 3 to 6 months of assessment was also assessed. In our analysis ⬍ 10% of events were accurately predicted.

S1091

Tuesday, October 29

There were no FP results in the general screening questions of both SLE and Type 1 DM. In those SARDs that had FP responses to the original screening question, accuracy was was improved in many by the confirmatory questions. In general FN rates were low. Conclusion: The new questionnaire improved overall accuracy over our original questionniare in screening for most autoimmune diseases and can be used to more accurately identify family members with these conditions.

also multiplex positive. However, 9 of the 59 high titer IIF samples were multiplex negative.

Conclusion: These results suggest that despite their early promise, and evidence that some of the selected cyto/chemokines fluctuate with disease activity, none of the analytes examined could accurately forecast clinically significant changes in disease state. Disclosure: C. landolt-Marticorena, None; S. Prokopec, None; S. Morrison, None; H. Reich, None; J. Scholey, None; D. Gladman, Abbott, Bristol-Myers Squibb, Celgene, Johnson & Johnson, MSD, Novartis, Pfizer, UCB Pharma, 2, Abbott, Bristol-Myers Squibb, Celgene, Johnson & Johnson, MSD, Novartis, Pfizer, UCB Pharma, 5; M. B. Urowitz, None; P. Boutros, None; P. R. Fortin, None; J. E. Wither, None.

ACR/ARHP Poster Session C Systemic Sclerosis, Fibrosing Syndromes, and Raynaud’s II Tuesday, October 29, 2013, 8:30 AM–4:00 PM

2558

Tuesday, October 29

Bone Morphogenic Protein Receptor 2 (BMPR2) Gene Mutations Are Associated To The Development Of Isolated Pulmonary Arterial Hypertension (PAH) In Systemic Sclerosis (SSc) Patients. Tatiana Sofia Rodriguez-Reyna1, Jose Luis Hernandez-Oropeza1, Tomas Rene PulidoZamudio2, Felipe Masso´2, Jessica Gutierrez-Manjarrez1, Alexandra Rueda de Leon-Aguirre1, Julio Sandoval-Zarate2 and Carlos Rodriguez-Osorio1. 1Instituto Nacional de Ciencias Medicas y Nutricion Salvador Zubiran, Mexico City, Mexico, 2Instituto Nacional de Cardiologia Ignacio Chavez, Mexico City, Mexico. Background/Purpose: Isolated pulmonary arterial hypertension (PAH) secondary to Systemic Sclerosis (SSc) is a severe life-threatening complication. Several pathogenic pathways have been implicated in its development. Bone morphogenic protein receptor 2 (BMPR2) is a protein of the TGF-beta receptor superfamily; mutations in this gene seem to be implicated in the proliferation of pulmonary artery smooth muscle cells, a major component of pulmonary arteriolar remodeling in PAH, and they have been associated to the susceptibility to develop idiopathic and familial PAH; exon 5 has higher frequency of mutations according to previous analyses. The aim of this study was to evaluate if BMPR2 gene exon 5 mutations are associated to isolated PAH in Mexican Mestizo SSc patients. Methods: DNA samples from 38 Mexican Mestizo SSc patients with isolated PAH (mean resting pulmonary artery pressure (PAP) ⬎25 mmHg, pulmonary wedge pressure ⬍ 15 mmHg and pulmonary vascular resistance ⬎ 3 Wood units by right side catheterization; or systolic PAP ⬎40 mmHg, tricuspid annular plane systolic excursion ⬍20 mmHg and dilatation and hypertrophy of right ventricle by transthoracic echocardiogram; patients were also required to have no or minimal interstitial lung disease by computed tomography) and 38 Mexican Mestizo SSc patients without PAH, matched by SSc subtype, age (⫹/⫺ 5 years), gender, time of evolution of the disease and SSc-associated autoantibody were analyzed. High resolution melting analysis (HRMA) of exon 5 of the BMPR2 gene was performed in previously amplified DNA (sense primer: 5⬘TCA TGC TAT TCT GCA TTC ATC.3⬘, antisense primer: 5⬘CAG GTC TAG TAT CAC AGT AGA.3⬘). Aberrant curves were considered as positive for mutation in the exon 5 of the BMPR2 gene. Severity of the organ involvement was evaluated using the Medsger severity scale. Statistical analysis was performed using SPSS v.19 and Epi-Info 7.0. Chi square was used to compare mutation frequencies between groups; Student’s t test and Wilcoxon rank test were used to compare numeric variables. Values were considered significant when two-tailed p values were ⬍0.05. Results: Mean age was 50 years, mean time of SSc evolution was 12.25 years, 80% of patients were female, with no differences between PAH and non-PAH patients. Mean PAP was 57 mmHg in PAH-SSc patients and 25 mmHg in non-PAH-SSc patients (p⬍0.001); BMPR2 gene exon 5 mutations were detected in 7 (18%) of PAH-SSc patients and in 1 (3%) non-PAH-SSc patient (p⫽0.001, OR: 31, 95% CI 5.1–1263). From the 7 patients with PAH that had a mutation, 6 were female, 6 had limited cutaneous SSc, 6 had severe vascular involvement (digital ulcers and/or gangrene at any time of the evolution of the disease), 1 had anti-Topoisomerase I and 2 had anti-U1-RNP antibody. Prevalence of BMPR2 gene exon 5 mutation in our sample was 11%. Conclusion: Prevalence of BMPR2 gene exon 5 mutations in PAHSSc patients is higher than in non-PAH-SSc patients; prevalence of BMPR2 gene exon 5 mutations in this SSc sample is similar to the prevalence reported in some idiopathic PAH studies; our results suggest

that BMPR2 gene exon 5 mutations increase the risk to develop PAH in Mexican Mestizo SSc patients. Disclosure: T. S. Rodriguez-Reyna, None; J. L. Hernandez-Oropeza, None; T. R. Pulido-Zamudio, None; F. Masso´, None; J. Gutierrez-Manjarrez, None; A. Rueda de Leon-Aguirre, None; J. Sandoval-Zarate, None; C. Rodriguez-Osorio, None.

2559 Effect Of NOX4 Overexpression On The Levels Of Micro RNAs Relevant To Systemic Sclerosis Fibrotic Process. Sonsoles Piera-Velazquez, Alma Makul and Sergio A. Jimenez. Jefferson Institute of Molecular Medicine, Division of Connective Tissue Diseases and Scleroderma Center, Thomas Jefferson University, Philadelphia, PA. Background/Purpose: Systemic sclerosis (SSc) is characterized by the excessive deposition of collagen and other connective tissue components in skin and multiple internal organs. Although transforming growth factor ␤ (TGF-␤) has been shown to play a crucial role in the development of tissue fibrosis in SSc, recently, other factors such as excessive oxidative stress have been implicated in the pathogenesis of the disease. Oxidative stress is caused by cellular overproduction of ROS (reactive oxidative species). NOX4 is one of seven NADPH isoforms which plays a crucial role in the generation of ROS. Numerous recent studies have described an important role of microRNAs (miRNA) in the pathogenesis of the fibrotic process in various fibrotic diseases including renal, pulmonary, and liver fibrosis as well as in SSc. The purpose of these studies was to induce overexpression of NOX4 in normal human dermal fibroblasts and to examine the effect of increased NOX4 levels on the expression of various miRNA that participate in the regulation of tissue fibrosis. Methods: Two different cell lines of normal human dermal fibroblasts were isolated from skin biopsies, and were expanded in monolayer cultures to approximately 70% of confluence and then they were transiently transfected with either a NOX4 protein expression construct (pCI-NOX4) to induce in NOX4 expression or with the vector control pCI. The effects of NOX4 overexpression on global gene expression levels and on the expression of relevant miRNA were examined employing microarrays using Human gene 1.0 ST array Affymetrix gene chips. Results: Induced NOX4 expression did not cause apparent morphological changes or cytotoxicity. Remarkable changes in the expression levels of numerous transcripts associated with the phenotypic activation of fibroblasts into myofibroblasts were observed. Also, NOX4 overexpression caused significant changes in the levels of several relevant miRNAs. It was of substantial interest that the predicted gene targets for these miRNA included highly relevant genes such as those encoding members of the TGF-␤ pathway, as well as genes encoding numerous ECM proteins, and several crucial regulatory/transcription factor genes including Fli1, Gli3, and Hif1A. Conclusion: Induced NOX4 expression resulted in remarkable changes in the levels of expression of numerous transcripts associated with the phenotypic activation of myofibroblasts as well as changes in the levels of several miRNAs that exert important regulatory effects on various genes that participate in the fibrotic process. These observations provide strong support to the notion that targeting NOX4 expression may be a novel therapeutic approach for SSc and other systemic fibrotic disorders. Disclosure: S. Piera-Velazquez, None; A. Makul, None; S. A. Jimenez, None.

2560 Analysis Of Global Gene Expression Of Pulmonary Endothelial Cells From Caveolin-1 Knock-Out Mice. Sonsoles Piera-Velazquez1, Zhaodong Li1, Sankar Addya2, Peter J. Wermuth1 and Sergio A. Jimenez1. 1Jefferson Institute of Molecular Medicine, Division of Connective Tissue Diseases and Scleroderma Center, Thomas Jefferson University, Philadelphia, PA, 2Kimmel Cancer Center, Thomas Jefferson University, Philadelphia, PA. Background/Purpose: Caveolin-1 (Cav-1) deficiency has recently been shown to participate in the pathogenesis of tissue fibrosis in Systemic Sclerosis (SSc). Although the mechanisms involved have not been fully elucidated, it has been shown that Cav-1 deficiency can induce the phenotypic conversion of quiescent fibroblasts into activated myofibroblasts, the effector cells responsible for the fibrotic process in SSc. It has also been shown that Cav-1 deficiency mediates the phenotypic conversion of endothelial cells (EC) into myofibroblasts through endothelial-mesenchymal transition (EndoMT). Furthermore, Cav-1 deficiency causes a synergistic stimulation of TGF-␤-induced EndoMT. The objective of this study was to

S1092

Disclosure: S. Piera-Velazquez, None; Z. Li, None; S. Addya, None; P. J. Wermuth, None; S. A. Jimenez, None.

2561 Multiplex Cytokine Analysis Of Dermal Interstitial Blister Fluid In Systemic Sclerosis Defines Potential Pathogenic Pathways and Differentiates Clinical Subsets. Kristina E.N. Clark1, Henry Lopez2, Joanna Nikotorowicz-Buniak1, Xu Shiwen1, Korsa Khan1, George Martin3, David J. Abraham1, Christopher P. Denton1 and Richard J. Stratton1. 1UCL Medical School, London, United Kingdom, 2Murigenics, Vallejo, CA, 3Aero Dap, Vallejo, CA. Background/Purpose: Clinical diversity in systemic sclerosis (SSc) is likely to reflect multifaceted pathogenesis and the effect of key growth factors or cytokines operating within a disease-specific microenvironment. Dermal interstial fluid sampling offers the potential to examine local biological mechanisms and define protein expression within lesional tissue. We propose multiplex cytokine analysis could be a pragmatic method to define the inflammatory and immune activity in the lesions of SSc patients. Methods: Interstitial fluid samples from forearm skin of patients (n⫽25; DcSSc ⫽19, LcSSc⫽6) and comparable sites on healthy controls (HC) (n⫽10) were collected using the dermal suction blister method. These were profiled by Luminex array for inflammatory cytokines, chemokines, and growth factors. Permutation analysis (SAM in EXCEL) was used to compare cytokine levels in SSc and HC samples. Results: Luminex array profiling of the dermal blister fluid showed increased inflammatory cytokines (mean IL-6 in SSc 77.2 pg/ml versus 17.8 pg/ml in HC p⫽0.009, mean IL-17 in SSc 0.61 pg/ml versus 0 pg/ml, p⫽0.03), and vascular growth factors (VEGF 21.7 pg/ml in SSc, 13.5 pg/ml in HC (p⫽NS) and PDGF-aa 16.4 pg/ml in SSc versus 0.97 pg/ ml in HC, p⫽0.049). Additionally MCP-3 (CCL7), IL-15, and IFN-g were all found to be significantly increased in SSc compared to HC (p⬍0.05) (Figure 1). Subanalysis highlighted a correlation between IL-6 and skin score (r⫽0.44, p⫽0.024) in SSc, and MCP-3 with disease duration (r⫽0.54, p⫽0.005). There was also a significant correlation between IL-6 and IL-10 (r⫽0.59, p⫽0.002) IFN-g, IL-17, PDGF-bb were largely undetectable in the blister fluid of HC, but were present in a subgroup of SSc patients. IL-17 was only detectable in DcSSc (5/19), and in 0/6 LcSSc and 0/10 HC, while IFN-g was present in the blister fluid of 1/10 HC, and 15/25 SSc. IFN-g levels

were higher in the diffuse subset compared to those with limited disease (mean 1.76 pg/ml and 0.67pg/ml respectively). IL-6 showed a trend towards increased concentrations in DcSSc compared to LcSSc, but this was not statistically significant (mean 71.7 pg/ml in DcSSc, 32.3 pg/ml in LcSSc, p⫽0.07).

Figure 1. Graph to show mean concentration of MCP-3, PDGF-AA, IL-6, IL-17a and IFN-gamma in SSc patients and healthy controls.

Conclusion: Our results confirm the potential utility of dermal blister fluid to non-invasively define local biological processes in SSc, and identify profibrotic, angiogenic and T-cell derived factors expressed locally within the skin lesions. This technique of profiling patients using blister fluid has the potential to complement clinical and gene expression based classification to facilitate targeted therapy, as well as providing potential markers of disease activity or treatment effect. Disclosure: K. E. N. Clark, None; H. Lopez, Murigenics, 4; J. Nikotorowicz-Buniak, None; X. Shiwen, None; K. Khan, None; G. Martin, Murigenics, 3; D. J. Abraham, None; C. P. Denton, None; R. J. Stratton, None.

2562 Inhibition Of SOX9 Phosphorylation Abrogates The Increased Expression Of Profibrotic Genes In Systemic Sclerosis Dermal Fibroblasts. Sonsoles Piera-Velazquez, Alma Makul and Sergio A. Jimenez. Jefferson Institute of Molecular Medicine, Division of Connective Tissue Diseases and Scleroderma Center, Thomas Jefferson University, Philadelphia, PA. Background/Purpose: SOX9, a high mobility group transcription factor is a master regulator of chondrogenesis and plays a crucial role in the regulation of chondrocyte gene expression. However, recent studies on hepatic fibrosis and segmental glomerulosclerosis have suggested that SOX9 may participate in tissue fibrosis. Furthermore, we previously demonstrated that human normal dermal fibroblasts contained high levels of SOX9 phosphorylated at serine residue 181 and that TGF␤1 stimulates Ser181 SOX9 phosphorylation mediated by RhoA kinase and to a lesser extent by PI3K. We explored here the role of Ser181 phosphoSOX9 in the fibrotic process of Systemic Sclerosis (SSc) employing cultured SSc dermal fibroblasts treated in vitrowith specific inhibitors of the kinases that may be responsible for Ser181 SOX9 phosphorylation and examined the effect of these inhibitors on the increased expression of profibrotic genes and exaggerated production of extracellular matrix proteins characteristics of SSc fibroblasts. Methods: Dermal fibroblasts obtained from normal skin and from clinically affected forearm skin from patients with diffuse SSc of recent onset were studied. Ser181 phosphoSOX9 levels were assessed by Western blot analysis of cell lysates of confluent dermal fibroblast cultures employing a phospho-specific antibody that recognizes a SOX9 epitope containing a phosphorylated Ser 181 residue. Gene expression analyses were performed employing real time PCR. Collagen production was assessed by Western blots of fibroblast culture media. The effects of kinase inhibitor treatment on Ser181 phosphoSOX9 were assessed in confluent cultures in the presence or absence of TGF-␤1 (10ng/mL) for 24h. The potential kinases involved were identified employing Kinexus phosphoproteome databases. The intracellular kinases PIM1, PIM2 and PKC␦ were identified as being responsible for Ser181 SOX9 phosphorylation. The role of these kinases was examined by inhibition with specific small molecule kinase inhibitors. For PKC␦ inhibition studies two novel specific inhibitors developed by CompleGen (CG1037 and CG1056) were employed.

S1093

Tuesday, October 29

examine the differences in global gene expression patterns of Cav-1 knockout (KO) and wild-type murine lung EC regarding genes participating in the EndoMT process. Methods: Pulmonary EC were isolated from wild-type (WT) and cav-1 knockout (cav-1 KO) mice employing immunomagnetic methods with sequential anti-CD31 and anti-CD102 antibody selection followed by in vitroculture and treatment with TGF-␤1. To assess the differences in gene expression we performed microarray analysis of RNA isolated from cultured early passage lung EC from wild-type and cav-1 KO mice, employing the Affymetrix mouse cDNA array. Pathway analysis was performed employing Ingenuity software. Results: A large number of differentially expressed genes were identified in lung EC from cav-1 KO mouse in comparison to EC from wild-type controls, including several unexpected extracellular matrix proteins such as decorin (58-fold higher), dermatopontin (28-fold higher), and versican (17fold higher), as well as fibroblast growth factor 7 (30-fold higher), and some metalloproteinases, such as, matrix metalloproteinase 3 which was increased 12-fold. Of interest was also the observation that podoplanin a small glycoprotein considered to be specifically expressed in EC of lymphatic origin and which is highly increased in a variety of malignant cells was elevated 17–20 fold in samples from the cav-1 KO mouse lung EC. Numerous genes involved in the EndoMT process were also found to be upregulated in the cav-1 KO mice lung EC compared with the wild-type (WT) controls. Cav-1 deficiency caused a synergistic stimulation of the TGF-␤-induced gene expression patterns although some unique genes appeared to be induced by TGF-␤ treatment of the Cav-1 KO cells but not by treatment of the WT cells. Conclusion: Cav-1 deficiency results in remarkable changes in the global gene expression patterns of murine lung EC and potentiates the stimulatory effects of TGB-␤1 inducing the differential expression of numerous profibrotic genes and of genes involved in EndoMT. These results provide novel information about the mechanisms responsible for the participation of Cav-1 in SSc pathogenesis and further suggest that modulation of Cav-1 expression or activity/function may be a novel therapeutic target for SSc. Supported by NIH grant RO1 AR055660

Results: Previous results showed that dermal fibroblasts from SSc patients displayed marked elevation of Ser181 phosphoSOX9 levels in comparison with normal fibroblasts. Here, we show that TGF-␤ caused potent stimulation of Ser181 SOX9 phosphorylation which was abrogated at nM concentrations by the two small molecule inhibitors targeting PKC␦ as well as the PIM1 and PIM2 specific kinase inhibitors. The kinase inhibitors did not cause morphological changes or detectable cytotoxicity at the concentrations employed. The levels of type I collagen production were reduced in parallel with the changes in Ser181 phosphoSOX9 levels. Conclusion: The results indicate that Ser181 phosphoSOX9 participates in the molecular mechanisms responsible for the exaggerated fibrotic process in SSc and demonstrate that PKC␦ and PIM1/2 kinases are responsible for Ser181 SOX9 phosphorylation. Thus, the kinases involved in Ser181 SOX9 phosphorylation provide novel therapeutic targets for SSc and other fibrotic disorders. Disclosure: S. Piera-Velazquez, None; A. Makul, None; S. A. Jimenez, None.

2563

Tuesday, October 29

Modulating Myofibroblast Transition Of Human Scleroderma Fibroblasts Through Inhibition Of Rho Guanine Nucleotidase-Regulated Gene Transcription. Andrew Haak1, Pei-Suen Tsou2, Dinesh Khanna2, David A. Fox2, Scott D. Larsen3 and Richard R. Neubig4. 1Department of Pharmacology, University of Michigan Medical Center, Ann Arbor, MI, 2 Division of Rheumatology, University of Michigan Medical Center, Ann Arbor, MI, 3Vahlteich Medicinal Chemistry Core, College of Pharmacy, University of Michigan, Ann Arbor, MI, 4Department of Pharmacology & Toxicology, Michigan State University, East Lansing, MI. Background/Purpose: Systemic sclerosis (SSc) or scleroderma, like many fibrotic disorders, has no effective therapy. Deposition of collagen and excess extracellular matrix depends on the transition of dermal fibroblasts into myofibroblasts. Current drug development has focused on targeting the initial inflammatory stimuli and specific receptors involved in fibrosis. Recent evidence, however, indicates that a transcriptional program regulated by Rho GTPase is critical for myofibroblast transition. This could be a more effective convergent target for interrupting pathological fibrosis regardless of the initial inflammatory or fibrotic stimulus. We recently identified novel inhibitors of this transcription mechanism including CCG-203971 (Bell et al, Bioorg Med Chem Lett. 23:3826, 2013). Here we assess their effectiveness in human SSc dermal fibroblasts. Methods: Primary human dermal fibroblasts were obtained from normal individuals and patients with diffuse SSc. Cells were passaged three times in DMEM containing 10% fetal bovine serum prior to studies which were carried out at passage 4 or 5. mRNA for connective tissue growth factor (CTGF), alpha-smooth muscle actin (ACTA2) and collagen type 1 alpha 1 (COL1␣1) were quantified by qRT-PCR. The fraction of cells positive for the myofibroblast maker alpha-smooth muscle actin (␣-SMA) was determined by immunocytochemistry. Inhibitors of Rho/ MRTF/SRF-regulated gene expression (CCG-203971) and pirfenidone, the only approved antifibrotic therapy, were tested for their ability to suppress these markers of myofibroblast transition and fibrosis. Results: SSc dermal fibroblasts express significantly more mRNA for CTGF (3.7-fold) and ACTA2 (4.6-fold) than do normal control fibroblasts. Similarly the percentage of cells staining positively for ␣-SMA was significantly greater for SSc (77⫾9%) than for normal (27⫾18%) dermal fibroblasts (p⬍0.001, n⫽6). CTGF, ACTA2, and COL1␣1 mRNA in SSc fibroblasts were inhibited at 24 hours by CCG-203971 with an IC50 of ⬃10 ␮M. Treatment of SSc fibroblasts for 72 hours with 10 ␮M CCG-203971 reduced the percentage of ␣-SMA positive cells to control levels (26⫾18%) while pirfenidone at 300 ␮M was only marginally effective (41⫾16%). Conclusion: Targeting the Rho/MRTF/SRF transcriptional pathway strongly suppresses myofibroblast activation and fibrosis-related gene expression in human SSc dermal fibroblasts. This could represent a novel targeted approach to disrupt a key genetic switch involved in fibrosis mechanisms and may provide a broad spectrum approach to treatment for systemic sclerosis and other disorders of fibrosis. Disclosure: A. Haak, None; P. S. Tsou, Arthritis Foundation, 2; D. Khanna, NIH, 2, Scleroderma Foundation, 2; D. A. Fox, None; S. D. Larsen, None; R. R. Neubig, None.

2564 Functional Activation Of The TRPV1 and TRPV2 Non-Selective Cation Channels Potentiates TGF-␤1-Induced Endothelial-ToMesenchymal Transition In Murine Pulmonary Endothelial Cells Suggest a Potential Role Of Trpv Channels In The Pathogenesis Of Systemic Sclerosis. Peter J. Wermuth and Sergio A. Jimenez. Jefferson Institute of Molecular Medicine, Division of Connective Tissue Diseases and Scleroderma Center, Thomas Jefferson University, Philadelphia, PA. Background/Purpose: Transient receptor potential (Trp) cation channel, subfamily V (TRPV) members can impact a diverse range of vascular functions, including vascular tone, vascular permeability, mechanosensing, angiogenesis and endothelial cell (EC) proliferation, apoptosis, and death. Recently, it has been shown that endothelial-to-mesenchymal transition (EndoMT) may play a role in generating myofibroblasts responsible for the uncontrolled production of extracellular matrix in many fibrotic diseases, including Systemic Sclerosis (SSc). Thus the purpose of these studies was to investigate the effect of TRPV1 and TRPV2 inhibition on the EndoMT process. Methods: Pulmonary EC were isolated from three C57BL/6J mice employing sequential immunomagnetic selection with anti-CD31 and antiCD102 antibodies followed by in vitro culture and treatment with TRPV1 and TRPV2 agonists or antagonists in the presence and absence of TGF-␤1. EndoMT was assessed by ␣-smooth muscle actin (␣-SMA) immunofluorescence and by Western blot analysis for ␣-SMA and type I collagen. Induction of type I, type III and type IV collagens, ␣-SMA, fibronectin, as well as various mesenchymal cell specific genes, including the transcription factor Twist1, and the transcriptional repressors Snail1 and Snail2 as well as expression of EC-specific genes (Pecam1 and VE-cadherin) was assessed by semi-quantitative RT-PCR in triplicate for two replicates per cell line. Results: Exposure of lung ECs to either the TRPV1 antagonist capsazepine or the TRPV2 antagonist tranilast caused downregulation of expression of EC-specific genes and upregulation of mesenchymalspecific genes. These agents did not induce noticeable changes in cell morphology. Treatment of cells with either capsazepine or tranilast in combination with TGF-␤1 showed synergisitic stimulation of TGF-␤1 effects on gene expression. Capsazepine in combination with TGF-␤1 produced a pronounced morphological change from EC to fibroblast morphology cells which was of greater magnitude than those induced by TGF-␤1 alone. In contrast, neither the TRPV1 agonist capsaicin nor the TRPV2 agonist probenecid alone or in combination with TGF-␤1 induced EndoMT in murine lung ECs. Capsaicin and probenecid, however, abrogated TGF-␤1-induced EndoMT, and reversed TGF-␤1-induced morphological changes in the lung EC. Conclusion: Functional downregulation of TRPV1 and TRPV2 causes strong synergistic potentiation of TGF-␤1-induced EndoMT and may play a pathogenetic role in pulmonary fibrosis and other fibrotic diseases, including SSc. The abrogation of TGF-␤1-induced EndoMT by capsaicin and probenecid suggests that induction or activation of TRPV1 and TRPV2 activation could represent an important and novel strategy to prevent EndoMT-mediated generation of activated myofibroblasts in fibrotic diseases. Supported by NIH grant R01AM19616 to SAJ. Disclosure: P. J. Wermuth, None; S. A. Jimenez, None.

2565 Synergistic Effects Of Endothelin-1 On Transforming Growth Factor-␤1 (TGF-␤1) Induced Endothelial-To-Mesenchymal Transition. A Novel Mechanism For The Fibrogenic Effects Of Endothelin. Peter J. Wermuth, Zhaodong Li and Sergio A. Jimenez. Jefferson Institute of Molecular Medicine, Division of Connective Tissue Diseases and Scleroderma Center, Thomas Jefferson University, Philadelphia, PA. Background/Purpose: The process of endothelial-to-mesenchymal transition (EndoMT) may be a crucial pathway in the generation of activated myofibroblasts, cells that play a pivotal role in the development of tissue and organ fibrosis in fibrotic diseases such as Systemic Sclerosis

S1094

Disclosure: P. J. Wermuth, None; Z. Li, None; S. A. Jimenez, None.

2566 Nilotinib Treatment Effect In The Skin As Measured By DNA Microarray In Patients With Diffuse Systemic Sclerosis. Jessica K. Gordon1, Tammara A. Wood2, Robert F. Spiera1 and Michael L. Whitfield3. 1Hospital for Special Surgery, New York, NY, 2Dartmouth Medical School, Hanover, NH, 3Geisel School of Medicine at Dartmouth, Hanover, NH. Background/Purpose: Gene expression profiling by DNA microarray is used to identify potential biomarkers in Systemic Sclerosis (SSc). Discrete gene expression signatures have been observed to subset patients and correlate with clinical parameters. We tested the hypothesis that treatment modulates gene expression in SSc skin in the context of an open label study of nilotinib to treat early diffuse cutaneous (dc)SSc. Methods: Ten adult patients with early progressive dcSSc of ⬍3 yrs since the initial SSc symptom were treated with nilotinib 400 mg PO twice daily. Three mm punch biopsies of lesional forearm skin were performed at baseline and after 6 and 12 mos of treatment, and the Modified Rodnan Skin Score (MRSS) was recorded. Clinical improvement was defined as improvement in MRSS by 20% from baseline. Using this criterion, we classified 4 patients as improvers and 4 patients as non-improvers. RNA was extracted from the whole skin, converted to cRNA, and hybridized to Agilent 8⫻60K whole human genome microarrays. Significance Analysis of Microarrays (SAM) was used to identify genes differentially expressed between pre- and post-treatment skin biopsies for both improvers and non-improvers separately and together. Patients before and after 6 and 12 months of treatment were assigned to intrinsic subsets using a gene expression classifier algorithm.

Results: Eight out of 10 patients from the nilotinib trial had at least baseline and 6 month biopsies that were included in the gene expression analysis. Patient characteristics (n⫽8) were: 75% female; median age 48.5 (IQR 41,53); median disease duration from the first non-Raynaud’s symptom of SSc: 0.67 (IQR 0.42, 0.71); baseline Modified Rodnan Skin Score (MRSS): 27.5 (IQR 22, 32); 62.5% RNA Pol3 positive; 25% Scl70 positive. SAM identified 185 genes with significantly decreased expression post treatment in improvers. These genes were significantly enriched for inflammatory response, hematopoiesis and cell-cell adhesion. In contrast, we did not find significant changes in gene expression post-treatment in non-improvers. Analysis of patient intrinsic subsets showed that 3 out of 4 non-improvers showed stable or increasing inflammatory signatures. Improvers showed either rapid decrease in the inflammatory signature or absence of the inflammatory signature with concomitant stable or increasing fibroproliferative signatures. Conclusion: We observed modulation of gene expression in skin as measured by DNA microarray in improvers but not in non-improvers during the course of the nilotinib trial. Patients that did not show improvement had an inflammatory signature that was stable or increasing, whereas patients that improved show decreasing inflammatory and increasing fibroproliferative gene signatures. Whether these changes in gene-expression signify biological effect of nilotinib or natural history of scleroderma will require an untreated control group to determine. Disclosure: J. K. Gordon, None; T. A. Wood, None; R. F. Spiera, Roche Pharmaceuticals, 2, Novartis Pharmaceutical Corporation, 2, GlaxoSmithKline, 2; M. L. Whitfield, None.

2567 Cadherin-11 Regulates Pulmonary Fibrosis In Bleomycin-Induced Lung Injury. Mesias Pedroza, Anuh T. George and Sandeep K. Agarwal. Baylor College of Medicine, Houston, TX. Background/Purpose: Pulmonary fibrosis (PF) is the leading cause of death in systemic sclerosis. Pathologically, PF is characterized by an aberrant wound healing repair mechanism leading to excessive fibroblast proliferation, myofibroblast differentiation, and extensive matrix deposition in the alveolar airways. Epithelial-to-mesenchymal transition (EMT) is the differentiation of fibroblast-like phenotype from epithelial cells that results in myofibroblast accumulation. Cadherin-11 (Cad11), a mesenchymal, homophillic adhesion molecule, has been reported to be expressed on hyperplastic alveolar epithelial cells (AEC), a populations of cells postulated to be undergoing EMT during the development of fibrosis. However it is not known if Cad11 is a regulator of EMT and if this contributes to the development of PF. We hypothesized that Cad11 contributes to the development of pulmonary fibrosis through the regulation of EMT. Methods: Cad11 deficient mice and neutralizing Cad11 monoclonal antibodies were used to determine the role of Cad11 in the intraperitoneal (IP) bleomycin (BLM) model of pulmonary fibrosis that exhibits EMT. AEC cell lines (A549, MLE-12) and primary murine AEC induced to undergo EMT by TGF-beta were used to determine if Cad11 regulates EMT in vitro. Results: Cad11 deficient mice had reduced fibrosis in the IP BLM model as assessed by improved arterial oxygen saturation, decreased collagen deposition, diminished alpha-smooth muscle actin (aSMA) accumulation, decreased TGF-b production and reduced b-catenin expression. These findings were confirmed using neutralizing antiCad11 monoclonal antibodies in wild type mice. In vitrostudies demonstrated that activation of Cad11 using immobilized Cad11-Fc fusion protein increased Col1a expression by A549 and MLE-12 AEC. In contrast, siRNA knockdown of Cad11 expression in A549 and MLE-12 AEC decreased the expression of fibrotic genes such as Col1a, aSMA, and CTGF. In addition, primary AECs from Cad11 deficient mice had decreased capacity to undergo EMT induced by TGF-beta relative to primary AECs from wild type mice. Finally, Cad11 inhibition with soluble Cad11-Fc fusion protein reduced EMT markers in wild type AECs treated with TGF-beta. Conclusion: These findings demonstrate that Cad11 is an important mediator of EMT in vitro and pulmonary fibrosis in vivo. Furthermore, these data suggest that Cad11 may be a therapeutic target in the treatment of pulmonary fibrosis. Disclosure: M. Pedroza, None; A. T. George, None; S. K. Agarwal, None.

S1095

Tuesday, October 29

(SSc). It has been previously demonstrated that TGF-␤1 induces EndoMT in vitro in murine lung endothelial cells (ECs). Since extensive previous studies demonstrated a potent profibrotic role for Endothelin-1 (ET-1), the role of ET-1 in the induction of EndoMT was investigated in cultured murine lung ECs. Methods: Pulmonary EC were isolated from three C57Bl/6J mice employing trypsin/collagenase tissue digestion followed by sequential immunomagnetic selection with anti-CD31 and anti-CD102 antibodies. After in vitro culture, the purified EC were treated with ET-1 in vitro in the presence and absence of TGF-␤1. EndoMT was assessed by immunofluorescence for ␣-smooth muscle actin (␣-SMA) and by Western blot analysis for ␣-SMA and type I collagen. Induction of type I, type III and type IV collagens, ␣-SMA, fibronectin, as well as several mesenchymal specific genes, including the transcription factor Twist1 and the transcriptional repressors Snail1 and Snail2 was assessed by semi-quantitative RT-PCR triplicate assays for two replicates per cell line. Results: Treatment of murine pulmonary ECs with TGF-␤1 induced high levels of ␣-SMA expression in comparison to saline-treated control ECs. ET-1 alone did not affect ␣-SMA levels but produced a synergistic effect with TGF-␤1 by potentiating TGF-␤1-induced EndoMT as indicated by increased ␣-SMA production. A quantitative assessment of the number of ␣-SMA expressing EC in TGF-␤1-treated cultures was 27% compared to 52% in cultures treated with both TGF-␤1 and ET-1. These results were confirmed by Western blot analysis. Semi-quantitative RT-PCR analysis demonstrated that ET-1 synergistically potentiated TGF-␤1-mediated increased expression levels of types I and III collagens, ␣-SMA, fibronectin, Twist1, Snail1 and Snail2. Expression of EC-specific VE-cadherin (Cdh5) was downregulated in TGF-␤1-treated cultures compared to saline controls and ET-1 also synergistically potentiated this TGF-␤1-mediated effect. Conclusion: ET-1 plays an important role in regulating EndoMT by causing a synergistic potentiation of TGF-␤1-induced EndoMT-mediated generation of activated myofibroblasts and of EndoMT-mediated increased expression of extracellular matrix components including Types I and III collagens. Furthermore, ET-1 also potentiated the TGF-␤1-induced increase in expression of various genes involved in the phenotypic conversion of EC into myofibroblasts including Twist1, Snail1 and Snail2. Since ET-1 plays a crucial role in the pathogenesis of SSc-associated pulmonary arterial hypertension and may play a profibrotic role in skin and lung fibrosis, the results described here identify a novel mechanism supporting the concept that ET-1 plays a key pathogenetic role in SSc-associated pulmonary fibrosis. Supported by NIH Grant R01AM19616 to SAJ.

2568

Tuesday, October 29

Involvement Of Collagen-Binding Heat Shock Protein 47 In The Scleroderma-Associated Fibrosis. Haiyan Chu1, Ting Wu1, Wenyu Wu2, Wenzhen Tu3, Yanyun Ma1, Qingmei Liu1, Hejian Zou4, Li Jin1 and Jiu-Cun Wang1. 1Ministry of Education Key Laboratory of Contemporary Anthropology, School of Life Sciences, Fudan University, Shanghai, China, 2Division of Dermatology, Huashan Hospital, Fudan University, Shanghai, Shanghai, China, 3Shanghai Traditional Chinese Medicine-Integrated Hospital, Shang hai, China, 4Huashan Hospital, Shanghai, China. Background/Purpose: Scleroderma or systemic sclerosis (SSc) is characterized by the fibrosis of skin and visceral organs, especially the uncontrolled fibrosis of multiple organs. Collagen is a major extracellular matrix (ECM) protein that deposited in the fibrotic organs. Heat shock protein 47 (HSP47) has been identified as a collagen-specific chaperon, and plays an important role in the development of fibrosis. Our previous study demonstrated that HSP47 was significantly up-regulated in the skin of dermal sclerosis mice induced by bleomycin. The aim of the present study is to investigate the role of HSP47 in the pathogenesis of scleroderma. Methods: For in vivo studies, C57BL/6 female mice of 6–8 weeks (n ⫽ 5 for each treatment) were injected with bleomycin subcutaneously into the same site of the shaved upper back daily for 3 weeks. The mRNA and protein level of HSP47 in the lesion skins from the mouse model was assessed by real-time PCR and western blot. In clinical study, skin biopsies and fibroblasts, peripheral blood mononuclear cells (PBMC) and plasma of SSc patients were obtained to study the role of HSP47 during the pathogenesis of SSc. Immunohistochemical staining was performed to identify the localization of HSP47 in the skin of SSc patients, real-time PCR and western blot were conducted to respectively test the gene and protein expression levels of HSP47 in the skin fibroblasts, and ELISA was used to detect the protein level of HSP47 in the plasma of SSc patients. For in vitro study, silencing and over-expression of HSP47 in NIH/3T3 fibroblast cells were performed using HSP47 siRNA and HSP47 expression plasmid respectively to investigate the effect of HSP47 on collagen production. Results: For in vivo study, HSP47 was significantly up-regulated in the skin lesion of the mice treated by bleomycin. Meanwhile, the expression of ␣-SMA increased in the skin lesion of the mice in response to bleomycin. In clinical study, the level of HSP47 increased in the fibroblasts cultured from SSc patients’ skin, the number of HSP47 positive cells increased in the skin of SSc patients, and the localization of HSP47-positive cells were observed in accordance with the ␣-SMA-positive cells. Additionally, the protein level of HSP47 in the plasma and the mRNA level of HSP47 in the peripheral blood mononuclear cells were both elevated in the SSc samples. In vitrostudy found that over-expression of HSP47 in the fibroblasts increased the level of collagen, whereas knockdown of HSP47 gene decreased its expression. Conclusion: The production of collagen is affected by HSP47, and HSP47 might be involved in the pathogenesis of scleroderma. Further investigations may be necessary to test this collagen chaperone as an important therapeutic approach. Disclosure: H. Chu, None; T. Wu, None; W. Wu, None; W. Tu, None; Y. Ma, None; Q. Liu, None; H. Zou, None; L. Jin, None; J. C. Wang, None.

2569 Serum S100A4 Levels Correlate With Skin Fibrosis and Lung Involvement In Systemic Sclerosis. Michal Tomcik1, Lucie Andres Cerezo1, Simona Skacelova1, Martin Komarc2, Radim Becvar1, Mariam Grigorian3, Joerg HW Distler4 and Ladislav Senolt1. 1Institute of Rheumatology, Department of Rheumatology, 1st Faculty of Medicine, Charles University, Prague, Czech Republic, 2Institute of Biophysics and Informatics of the First Faculty of Medicine, Charles University, Prague, Czech Republic, 3Institute of Cancer Biology, Danish Cancer Society, Copenhagen, Denmark, 4Department of Internal Medicine III and Institute for Clinical Immunology, University of Erlangen-Nuremberg, Erlangen, Germany. Background/Purpose: Our previous study demonstrated that S100A4 is overexpressed in scleroderma (SSc) skin, fibroblasts and preclinical models of SSc in a TGF-␤ dependent manner. Furthermore, we showed that S100A4 is a new regulator of canonical TGF-␤ signalling and its inhibition prevents the stimulatory effects of TGF-␤ on collagen synthesis. Similarly, deficit of S100A4 prevented dermal fibrosis induced by bleomycin or in Tsk-1 mice. The aim of this study was to evaluate S100A4 in the circulation of SSc patients and characterize its potential association with skin changes and SSc-related features.

Methods: A total of 33 patients (29 females; mean age 52.8; disease duration 4.2 years; dcSSc/lcSSc ⫽ 8/25) who met the ACR classification criteria for SSc and 20 healthy individuals matched by age and sex were included in this study. Serum S100A4 levels were measured using ELISA (CycLex Co., Ltd., Nagano, Japan). CRP, ANA and ENA complex were evaluated. SSc-related manifestations were obtained from the Czech Registry Database of SSc patients. Interstitial lung disease, pulmonary arterial hypertension, oesophageal dysmotility, cardiac and renal involvement, nailfold capillaroscopy and Raynaud’s phenomenon were recorded. Skin changes were assessed using the modified Rodnan skin score and EUSTAR SSc activity score was determined. Data are presented as mean ⫾ SEM. Results: S100A4 serum levels were significantly increased in SSc patients compared with healthy controls (119.2 ⫾ 23.4 vs. 43.9 ⫾ 3.3 ng/ml, p ⫽ 0.011). Patients with diffuse cutaneous SSc had significantly higher levels of serum S100A4 compared with patients with limited cutaneous SSc or healthy controls (201.8 ⫾ 53.1 vs. 92.7 ⫾ 24.0 ng/ml, p ⫽ 0.017 and 201.8 ⫾ 53.1 vs. 43.9 ⫾ 3.3 ng/ml, p ⫽ 0.001, respectively). The levels of S100A4 positively correlated with the modified Rodnan skin score (r ⫽ 0.556, p ⫽ 0.001). In addition, S100A4 levels negatively correlated with forced vital capacity (FVC) and saturation of peripheral oxygen (SpO2) (r ⫽ ⫺ 0.362, p ⫽ 0.038 and r ⫽ ⫺ 0.414, p ⫽ 0.029, respectively). Of particular interest, S100A4 levels positively correlated with EUSTAR SSc activity score (r ⫽ 0.750, p ⫽ 0.0001). However, only relations between S100A4 and the modified Rodnan skin score, and S100A4 and EUSTAR SSc activity score were approved at corrected level of statistical significance after Bonferroni’s correction (p ⫽ 0.001 ⬍ 0.01, p ⫽ 0.0001 ⬍ 0.01, respectively). The presence of autoantibodies (ANA, anti-centromere, anti-Scl70), a pathological capillaroscopic pattern (early, active, late), the administration of low dose glucocorticoids (or immunosuppressive treatment) and exhibition of the main individual clinical symptoms of SSc did not significantly affect levels of serum S100A4. Conclusion: We demonstrate that S100A4 serum levels are significantly increased in SSc patients compared with healthy controls. Higher levels of S100A4 are associated with dcSSc subset, skin involvement, deteriorated parameters of lung involvement and higher disease activity. These data support our previous findings on the role of S100A4 as a regulator of TGF-␤ induced fibroblast activation and dermal fibrosis in SSc. Acknowledgements: This study was supported by IGA NT 13698–4. Disclosure: M. Tomcik, None; L. Andres Cerezo, None; S. Skacelova, None; M. Komarc, None; R. Becvar, None; M. Grigorian, None; J. H. Distler, None; L. Senolt, None.

2570 Tenofovir But Not Adefovir Prevents Liver and Skin Fibrosis In Two Models Of Adenosine-Mediated Injury. Jessica L. Feig1, Doreen Tivon1, Miguel perez Aso2, Timothy Cardozo1 and Bruce N. Cronstein3. 1New York University School of Medicine, New York, NY, 2NYU Univ Medical Center, New York, NY, 3NYU School of Medicine, Division of Rheumatology, New York, NY. Background/Purpose: Acyclic nucleoside phosphonates are a key class of antivirals commonly used in the treatment of both DNA and retroviral infections. Adefovir and tenofovir are AMP analogues that resemble substrates of CD73. We have previously reported that adenosine, generated by the CD73-mediated dephosphorylation of AMP, acting at A2Areceptors, plays a critical role in development of both hepatic and dermal fibrosis in murine models of cirrhosis and scleroderma, respectively. A recent clinical trial demonstrated that tenofovir, but not other antiviral agents, reverses hepatic fibrosis/cirrhosis in patients with hepatitis B. We therefore proposed the hypothesis that tenofovir’s antifibrotic effects are mediated by inhibition of adenosine production by CD73-mediated dephosphorylation of AMP. Methods: In silico modeling and docking studies were performed using an ICM-Browser www.molsoft.com. CD73 enzyme activity was quantitated by malachite green. Alkaline Phosphatase activity (Abcam) was performed. Thioacetamide(TAA,100 mg/kg IP)-treated mice were treated with vehicle, Adefovir, or Tenofovir (75mg/kg, SubQ) [n⫽5–10 per group]. Bleomycin (0.25 U, SubQ)-treated mice were treated with vehicle, Adefovir, or Tenofovir (75mg/kg, IP) [n⫽5–10 per group]. Adenosine levels were determined by HPLC. Skin breaking strength was via tensiometer. H&E or picrosirus red-stained slides were imaged, and pixel quantification was performed with SigmaScan software. Scar index was determined as the ratio of red/green pixels representing compact/filamentous fibers; higher numbers indicate more fibrosis.

S1096

Results: In silico modeling data suggested that both adefovir and tenofovir bound to the enzymatic pocket of CD73. Tenofovir (Sequoia), but not adefovir (Sequoia), inhibited CD73 activity of 293T cells overexpressing CD73 (38⫹7.4%, at 10 uM) and of recombinant enzyme (72⫹1.0%, at 10uM). Yet, the inhibition of CD73 by Tenofovir (Gilead) was not pharmacologically relevant with an IC50⬎100uM. Alkaline phosphatase activity wasn’t modulated by Adefovir or Tenofovir. Adefovir decreased adenosine levels in the skin of bleomycin-challenged mice though this trend was not significant. Tenofovir significantly decreased adenosine levels in the skin of bleomycin-challenged mice (273.95⫹8.41 vs. 432.58⫹24.34nM adenosine/ 12mm punch biopsy, n⫽8–10, [p⬍0.05]). Tenofovir (75mg/kg), but not Adefovir (75mg/kg), diminished hepatic fibrosis in thioacetamide-treated mice (fibrotic area/hepatic slide area 1.00⫹0.04% vs 4.45⫹0.37%) Tenofovir (75mg/kg), but not Adefovir (75mg/kg), diminished bleomycin-induced dermal fibrosis in bleomycin-treated mice (73.7⫹3.1% reduction of hydroxyproline content [p⬍0.05]; 33.5⫹3.8% reduction of dermal thickness [p⬍0.06] and reduction of breaking tension by 66.8⫹1.4% [p⬍0.05]). Picrosirius red staining showed dramatic altering of dermal collagen quality (scar index of 1.2⫹0.1 vs 22.2⫹0.7 [p⬍0.001], normal skin is 2.5) in Tenofovir-treated mice. Conclusion: Tenofovir reduces fibrosis via inhibition of adenosine production. Tenofovir may have therapeutic potential in treating fibrosis in patients suffering from non-viral fibrosing diseases such as scleroderma. Disclosure: J. L. Feig, None; D. Tivon, None; M. perez Aso, None; T. Cardozo, None; B. N. Cronstein, Canfite Pharma, 1, NIH, Gilead, Takeda, AstraZeneca, 2, NYU School of Medicine, 3, Merck-SeronoBristol-Myers Squibb, Novartis, CanFite Biopharmaceuticals, Cypress Laboratories, Regeneron (Westat, DSMB), Endocyte, Protalex, Allos, Inc., Savient, Gismo Therapeutics, Antares Pharmaceutical, Medivector, 5, Multiple patents on adenosine receptors and bone metabolism, pharmacology, 9.

2571

Background/Purpose: Systemic Sclerosis (SSc) is an autoimmune disorder characterized by tissue fibrosis and vasculopathy. This latter comprises both neointima proliferation and defective angio and vasculogenesis. Although some of the clinical manifestations of the vasculopathy often precede the onset of fibrosis, there is scanty of data investigating the molecular mechanisms eventually linking the profibrotic activation of fibroblasts (FBs) and the poor angio/vasculogenesis in Scleroderma. Recently, a proteomic analysis of the secretome of SSc dermal FBs (SSc FBs), identified among other proteins a consistent increased secretion of Pigment Epithelium Derived Factor (PEDF) associated with the profibrotic phenotype (AJP, 2010). PEDF has been described as produced by retinal-pigmented epithelium and melanocytes, and is a major endogenous inhibitor of intraocular angiogenesis. Here we aimed to validate the increased expression of PEDF in SSc and test the hypothesis that PEDF might play an important role in establishing or perpetuating SSc vasculopathy. Methods: PEDF expression was investigated in the involved skin and FBs of 4 SSc patients in the early phase of the diffuse form of the disease and 4 healthy controls (HC) by immunohistochemistry (IHC) and real time-PCR analysis. Functional effects of PEDF on angio/vasculogenesis were examined by Matrigel assays. Organotypic co-culture assays were performed seeding HUVECs or microvascular endothelial cells (MVECs), on monolayers of either primary healthy FBs (HC FBs) or SSc FBs or HC FBS silenced for Caveolin-1 (Cav-1). Endothelial cells were evidenced by IHC staining for CD31 in organotypic co-culture assays. Vascular tubule number, length and junctions were identified and analyzed by Angiosys software (TCS CellWorks). Results: Both Healthy and Scleroderma skin biopsies showed high PEDF protein expression on melanocytes, as expected. Nevertheless, in SSc skin 52% (⫹/⫺5.9) of the FBs showed a strong expression of PEDF whereas only 13% (⫹/⫺0.68) of the FBs in HC skin were positive (p⬍0.0006). Double

Disclosure: V. Liakouli, None; G. Mavria, None; J. Gillespie, None; M. Scarcia, None; P. Cipriani, None; R. Giacomelli, None; P. Emery, None; F. Del Galdo, None.

2572 An Atypical Cyclin-Dependent Kinase Mediates Fibrosis and Is a Novel Target In Scleroderma. Jun Wei1, Roberta G. Marangoni1, Wenxia Wang1, Jingang Huang2, Joerg H. W. Distler2 and John Varga3. 1Northwestern University, Chicago, IL, 2Department of Internal Medicine III and Institute for Clinical Immunology, University of Erlangen-Nuremberg, Erlangen, Germany, 3Northwestern University Medical School, Chicago, IL. Background/Purpose: Cyclin-dependent kinase 5 (CDK5), expressed primarily in the central nervous system, plays important roles in axonal guidance, dopaminergic signaling, neuronal migration, and pain sensing. Aberrant CDK5 function is implicated in neurodegenerative diseases including Alzheimer’s disease. In contrast to other cyclin-dependent kinases, binding of the non-cyclin activator p35 is sufficient to induce CDK5 activity. Recent studies reveal novel extra-neuronal role for in inflammation and metabolism. The nuclear receptor PPAR-␥, a master regulator of adipogenesis and adipokine production, is impaired in scleroderma. Since PPAR-␥ is a novel CDK5 substrate, we hypothesized that the CDK5/p35 pathway might be responsible for impaired PPAR-␥ function in SSc and play a role in the development and persistence of fibrosis. Methods: Expression of p35 was examined in skin biopsies from bleomycin-injected mice and explanted scleroderma fibroblasts. Regulation of p35/CDK5 expression and activity in vitro was examined in human and mouse skin fibroblasts, progenitor cells and mature adipocytes by real-time qPCR, Western analysis and in vitro kinase assays. Effects of CDK5/p35 loss-of-function and gain-of-function were evaluated in normal and scleroderma skin fibroblasts using monolayer cultured fibroblasts and ex vivo using human skin organ cultures. CDK5 inhibition was examined in vivo using mouse models of fibrosis induced by bleomycin or constitutively active Type I TGF-␤ receptor. Results: Levels of p35 mRNA were markedly elevated in explanted scleroderma fibroblasts (n⫽6, p⬍0.005). p35 was also elevated in lesional skin from mice with bleomycin-induced scleroderma. Both p35 expression and CDK5 activity were strongly stimulated by TGF-␤ in human and mouse skin fibroblasts, mesenchymal progenitor cells and mature adipocytes. Ectopic p35 and CDK5 caused suppression of adiponectin expression and simultaneous stimulation of collagen synthesis in these cells, whereas RNAimediated knockdown of p35/CDK5 abrogated TGF-␤-induced fibrotic gene expression. Pharmacological inhibitors of CDK5 not only prevented but even reversed TGF-␤-induced fibrotic responses in monolayer cultures and in skin organ cultures, and ameliorated collagen overproduction in scleroderma fibroblasts. Moreover, CDK5 inhibitor prevented and reversed skin fibrosis in complementary inflammatory and TGF-␤-driven mouse models of scleroderma. Conclusion: The CDK5/p35 axis has a previously unrecognized important non-neuronal function in modulating fibrotic responses. Elevated p35 expression and CDK5 activity is an unexpected feature of scleroderma that might contribute to development of fibrosis. Pharmacological targeting CDK5/p35 might be novel treatment for fibrosis. Disclosure: J. Wei, None; R. G. Marangoni, None; W. Wang, None; J. Huang, None; J. H. W. Distler, None; J. Varga, None.

S1097

Tuesday, October 29

Pigment Epithelium Derived Factor Secreted By Activated Fibroblasts Can Contribute To Impaired Angio and Vasculogenesis In Scleroderma. Vasiliki Liakouli1, Georgia Mavria2, Justin Gillespie3, Margherita Scarcia2, Paola Cipriani1, Roberto Giacomelli1, Paul Emery4 and Francesco Del Galdo4. 1Rheumatology Unit, University of Aquila, L’Aquila, IT, L’Aquila, Italy, 2Signal Transduction and Angiogenesis group, Leeds Institute of Cancer and Pathology, University of Leeds, Leeds, UK, Leeds, United Kingdom, 3 Leeds Institute of Rheumatic and Musculoskeletal Medicine, University of Leeds, Leeds, United Kingdom, 4Leeds Institute of Rheumatic and Musculoskeletal Medicine, University of Leeds and NIHR Leeds Musculoskeletal Biomedical Research Unit, Leeds Teaching Hospitals NHS Trust, Leeds, United Kingdom.

IHC studies indicated that FBs positive for PEDF showed a decreased expression of Cav-1 in both HC and SSc skin. In vitro studies confirmed that SSc FBs showed on average a 5-fold increased in PEDF expression compared to HC FBs (p⬍0.05). Functional studies confirmed that recombinant PEDF protein had a direct inhibitory effect on vasculogenesis, suppressing both tubule length and number. Consistently, organotypic co-culture assays indicated that SSc FBs or HC FBs silenced for Cav-1 inhibited tubulogenesis both on MVECs or HUVECs, respectively. Conclusion: PEDF expression is increased in SSc biopsies and SSc FBs. PEDF expression is associated with decreased Cav-1 expression in vivo and it is induced by silencing Cav-1 in vitro. Functionally, PEDF can suppress vasculogenesis both in Matrigel and co-culture assays. This suggests that the decreased expression of Cav-1 observed in SSc FBs may contribute to the vasculopathy of Scleroderma. Further studies unraveling the mechanisms of the antiangiogenic effect of PEDF may shed light in understanding the molecular events linking the profibrotic phenotype and SSc vasculopathy.

2573

Tuesday, October 29

The Role Of STAT-3 In The Development Of Pulmonary and Dermal Fibrosis. Mesias Pedroza, Sarah To, Anuh T. George, David J. Tweardy and Sandeep K. Agarwal. Baylor College of Medicine, Houston, TX. Background/Purpose: Fibrosis is the accumulation of excessive extracellular matrix in tissues, leading to tissue damage. In systemic sclerosis, the trigger is postulated to be an autoimmune response that leads to tissue injury, production of growth factors, pro-inflammatory and profibrotic cytokines, and accumulation of myofibroblasts. Two potential sources of myofibroblasts are the differentiation of local fibroblasts and the process of epithelial-tomesenchymal transition (EMT). IL-6 is a proinflammatory and profibrotic cytokine that is increasingly recognized as an important mediator of fibrosis and may contribute to the accumulation of myofibroblasts. After engaging its receptor, IL-6 signals through the STAT-3. STAT-3 has been shown to be elevated in skin and pulmonary fibrosis. The extent to which STAT-3 is involved in the development of fibrosis and the mechanisms by which it leads to fibrosis are not known. We hypothesizes that STAT-3 signaling contributes to the development of tissue fibrosis in the lung and skin in part through the modulation of EMT. Methods: Fibrotic tissue from systemic sclerosis patients, idiopathic pulmonary fibrosis patients, and mouse models of lung and skin fibrosis was processed for phosphor-STAT-3 staining by immunohistology. To determine if STAT-3 signaling contributes to the development of fibrosis, C-188-9, a novel small molecule STAT-3 inhibitor, was administered to C57/Bl-6 mice in both the intraperitoneal (IP) bleomycin mouse model of lung fibrosis and the subcutaneous (SC) bleomycin mouse model of skin fibrosis. To determine the role of STAT-3 in EMT and myofibroblast differentiation, C-188-9 was used in tissue culture experiments with alveolar epithelial cells (AEC; MLE-12 and primary AEC) and murine lung fibroblasts. Results: Phospho-STAT-3 expression was increased in fibrotic tissue from systemic sclerosis patients, idiopathic pulmonary fibrosis patients, and mouse models of lung and skin fibrosis. STAT-3 inhibition by C-188-9 decreased fibrotic endpoints (collagen deposition by Sircol, expression of alpha-smooth muscle actin (SMA), and improved arterial oxygen saturation) in the IP bleomycin pulmonary fibrosis model. C-188-9 also decreased the development of dermal fibrosis in the SC bleomycin model as assessed by decreased dermal thickness, a reduction of alpha-SMA accumulation, and decreased collagen deposition. In vitro studies show that TGF-beta or IL-6 trans-signaling (IL-6/sIL-6R-alpha) were able to induce 1) EMT on primary AEC and MLE-12 cell line and 2) myofibroblast differentiation from fibroblasts. C-188-9 prevented TGF-beta and IL-6/sIL-6R-alpha induced EMT assessed by Col1a, alpha-SMA, Twist, and Snail mRNA levels and reduced myofibroblast differentiation as assessed by Col1a and alpha-SMA mRNA levels. Conclusion: These findings demonstrate that STAT-3 contributes to the development of tissue fibrosis in the skin and the lung and plays a role in the development of myofibroblasts in vitro. Furthermore, these data suggest that STAT-3 may be a therapeutic target in the treatment of dermal and pulmonary fibrosis. Disclosure: M. Pedroza, None; S. To, None; A. T. George, None; D. J. Tweardy, StemMed LtdP, 1; S. K. Agarwal, None.

transplant-free survival in patients with SSc-PAH, both with and without ILD, treated aggressively with PAH therapies. Methods: All SSc patients who had a right heart catheterization (RHC) diagnostic for PAH (mean pulmonary artery pressure (mPAP) greater than or equal to 25 mm Hg, pulmonary capillary wedge pressure (PCWP) less than or equal to 15 mm Hg, pulmonary vascular resistance (PVR) greater than or equal to 240 dynesxsecond/cm5), between 2001–2012 were enrolled. ILD was defined as greater than 30% disease extent on high-resolution computed tomography (HRCT) or when disease extent 10–30%, forced vital capacity less than 70%. Kaplan-Meier and Cox proportional hazards models were used to analyze survival and identify prognostic variables. Results: Of the 99 patients with SSc-PAH, 71 also had ILD. Patients with PAH⫹ILD were younger than patients with PAH alone (Mean 55 years vs. 60 years, respectively, p⫽0.07) and a smaller percentage were woman (70% vs. 93%, respectively, p⫽0.02). SSc type/disease duration, ethnicity, comorbidities, hemoglobin, creatinine, mPAP, PCWP, PVR, use of supplemental oxygen, and six-minute walk distance were similar between patients with PAH alone and PAH⫹ILD. Twenty-four percent of all patients started prostacyclin therapy within 6 months of the RHC, while 24% started prostacyclin therapy after 6 months of the RHC. The 1-, 2-, 3-year survival estimates were 72%, 59%, 50%, and 82%, 66%, 60%, for the PAH⫹ILD and PAH alone groups, respectively, p⫽0.5 (Figure 1). In the multivariate model, after controlling for potentially confounding variables, male gender (hazard ratio 0.6, p⫽0.008) and prostacyclin therapy initiation within 6 months of the RHC (hazard ratio 1.4, p⫽0.007) were the only factors significantly associated with transplant-free survival.

Figure 1. Kaplan-Meier survival curves demonstrating no significant difference in transplant-free survival for SSc-PAH patients with ILD (blue line) and without ILD (red line) (Log rank p-value 0.5).

Conclusion: Survival of SSc-PAH and ILD has improved compared with historical series (1); this may in part be due to aggressive use of prostacyclin therapy. References: (1) Le Pavec J, et al. Arthritis Rheum 2011;63:2456. Disclosure: E. Volkmann, None; R. Saggar, Gilead, 9, United Therapeutics, 9, Actelion Pharmaceuticals US, 9; B. Torres, None; L. Yoder, None; R. Elashoff, None; R. Saggar, Actelion Pharmaceuticals US, 9, United Therapeutics, 9, Gilead, 9; H. Agrawal, None; N. Borazan, None; S. Thomas, None; D. Furst, AbbVie, 2, Actelion Pharmaceuticals US, 2, Amgen, 2, BMS, 2, Gilead, 2, GlaxoSmithKline, 2, NIH, 2, Novartis Pharmaceutical Corporation, 2, Pfizer Inc, 2, Roche Pharmaceuticals, 2, Genentech and Biogen IDEC Inc., 2, UCB, 2, AbbVie, 8, Actelion Pharmaceuticals US, 8, UCB, 8.

ACR/ARHP Poster Session C Systemic Sclerosis, Fibrosing Syndromes, and Raynaud’s Clinical Aspects and Therapeutics II Tuesday, October 29, 2013, 8:30 AM–4:00 PM

2574 Early Use of Prostacyclin Therapy Improves Transplant-Free Survival in Patients With Systemic Sclerosis-Related Pulmonary Arterial Hypertension Plus Interstitial Lung Disease. Elizabeth Volkmann1, Rajan Saggar1, Bryant Torres1, Lynne Yoder1, Robert Elashoff1, Rajeev Saggar2, Harsh Agrawal1, Nabeel Borazan3, Sarah Thomas1 and Daniel Furst1. 1University of California, Los Angeles, Los Angeles, CA, 2St. Joseph’s Hospital and Medical Center, Phoenix, AZ, 3Rheumatology UCLA, Los Angeles, CA. Background/Purpose: The leading causes of death in systemic sclerosis (SSc) are pulmonary arterial hypertension (PAH) and interstitial lung disease (ILD). Use of PAH therapy in patients with both SSc-PAH and ILD is controversial and may not improve survival (1). This study investigates

2575 Utility Of Autoantibody Testing For Predicting Risk Of Pulmonary Arterial Hypertension: A Retrospective Analysis In Routine Autoantibody Laboratory. Masataka Kuwana1, Yuichiro Shirai1, Hidekata Yasuoka1, Tsutomu Takeuchi1 and Kenichi Masui2. 1Keio University School of Medicine, Tokyo, Japan, 2National Defense Medical College, Tokorozawa, Japan. Background/Purpose: Pulmonary arterial hypertension (PAH) is an intractable complication of connective tissue disease (CTD). Current guidelines recommend early detection and intervention for improvement of

S1098

Disclosure: M. Kuwana, None; Y. Shirai, None; H. Yasuoka, None; T. Takeuchi, None; K. Masui, None.

2576 Epoprostenol Rescue Therapy In Systemic Sclerosis-Associated Pulmonary Arterial Hypertension and Idiopathic Pulmonary Arterial Hypertension. Adrienne M. Roos1, Christopher Pasarikovski1, Amie T. Kron1, John T. Granton2, Peter Lee3, John Thenganatt4 and Sindhu R. Johnson5. 1 Toronto Scleroderma Research Program, Toronto Western Hospital, Mount Sinai Hospital, and University of Toronto, Toronto, ON, 2Toronto Pulmonary Hypertension Programme, Toronto General Hospital and University of Toronto, Toronto, ON, 3Mt. Sinai Hospital, Toronto, ON, 4University Health Network Pulmonary Hypertension Programme, Toronto General Hospital, Divisions of Respirology and Critical Care Medicine, Toronto, ON, 5Division of Rheumatology, Toronto Western Hospital, University Health Network Pulmonary Hypertension Programme, Toronto General Hospital, Mount Sinai Hospital and University of Toronto, Toronto, ON. Background/Purpose: Epoprostenol has been demonstrated to improve hemodynamics, functional class, and six-minute walk distance (6MWD) in systemic sclerosis-associated pulmonary arterial hypertension (SSc-PAH) and idiopathic PAH (IPAH) patients. In contemporary practice, it is usually reserved for patients who have failed treatment with endothelin receptor antagonists and/or phosphodiesterase-5 inhibitors. The effect of epoprostenol rescue therapy on survival has not been evaluated. The objective of this study was to evaluate the role of intravenous epoprostenol as rescue therapy in the SSc-PAH and IPAH patients. Methods: Patients attending the University Health Network Pulmonary Hypertension Program between 1998 and 2012 were included if they had a diagnosis of SSc-PAH and IPAH based on a mean pulmonary artery pressure

(mPAP) of ⬎ 25 mmHg and a pulmonary capillary wedge pressure of ⬍15 mmHg on cardiac catheterization, and had been treated with intravenous epoprostenol after treatment with endothelin receptor antagonists and/or phosphodiesterase-5 inhibitors for PAH. The primary outcome was survival. Survival was defined as the time from initiation of epoprostenol to death from any cause. Patients were censored as of May 1, 2012. Survival was evaluated using Kaplan Meier curves. Results: 1140 patients were reviewed to identify 36 patients with SSc-PAH and 24 patients with IPAH treated with epoprostenol after failure with oral pulmonary hypertension specific therapies. 83% of SScPAH and 75% of IPAH patients were female. The mean (standard deviation) PAH duration prior to initiation of epoprostenol was 3.3 (5.7) years for SScPAH, and 2.1 (2.1) years for IPAH patients. Median 1-, 2-, 3-, 4-, 5-year survival for SSc patients was 85.7%, 60.7%, 53.6%, 46.1%, 42.3%; and for IPAH patients was 83.3%, 70.8%, 65.8%, 59.2%, 59.2%. There was no significant difference in survival between the SScPAH and IPAH patients treated with epoprostenol (p⫽0.13). Conclusion: Our findings demonstrate desirable long-term survival and support the use of epoprostenol as rescue therapy for SSc-PAH and IPAH patients. Disclosure: A. M. Roos, None; C. Pasarikovski, None; A. T. Kron, None; J. T. Granton, Support respirology program at the hospital foundation., 9, Pfizer support of research study via CIHR grant., 9; P. Lee, None; J. Thenganatt, None; S. R. Johnson, None.

2577 Sex Disparities In Survival Of Systemic Sclerosis-Associated Pulmonary Arterial Hypertension and Idiopathic Pulmonary Arterial Hypertension Patients. Christopher Pasarikovski1, John T. Granton2, Peter Lee3, Adrienne M. Roos1, Amie T. Kron1, Cathy Chau4 and Sindhu R. Johnson5. 1Toronto Scleroderma Research Program, Toronto Western Hospital, Mount Sinai Hospital, and University of Toronto, Toronto, ON, 2Toronto Pulmonary Hypertension Programme, Toronto General Hospital and University of Toronto, Toronto, ON, 3Mt. Sinai Hospital, Toronto, ON, 4Toronto Scleroderma Research Program, Toronto Western Hospital, Mount Sinai Hospital, University of Toronto, Toronto, ON, 5Toronto Western Hospital, Toronto General Hospital, Mount Sinai Hospital, and University of Toronto, Toronto, ON. Background/Purpose: Systemic sclerosis (SSc) associated pulmonary arterial hypertension (PAH) and idiopathic PAH (IPAH) are conditions with poor survival. There is evidence to suggest that sex affects survival. The primary objective of this study was to evaluate the effect of sex on survival in SSc-PAH and IPAH. We secondarily evaluated the effect of sex on disease onset, time to diagnosis, disease progression and treatment. Methods: Patients were included if they attended the Toronto Scleroderma Program or the University Health Network Pulmonary Hypertension Program; had a diagnosis of SSc-PAH or IPAH defined as a mean pulmonary artery pressure ⬎25mmHg and age ⬎ 16 years. Sex was defined as self-reported biological and physiological characteristics at birth (male, female). The primary outcome was the time from diagnosis to death from all causes. Secondary outcomes were sex differences in age of diagnosis, disease duration and SSc manifestations. Cox proportional hazards model were used to evaluate survival. Results: 52 male and 267 female SScPAH; and 47 male and 107 female IPAH patients were identified. Male SSc patients had a shorter mean (standard deviation) time from SSc diagnosis to PAH diagnosis (5.6 (8.7)) versus (8.4 (9.6)), p⫽0.047), increased frequency of renal crisis (19% versus 9%, p⫽ 0.04), interstitial lung disease (67% versus 49%, p⫽0.02), and digital ulcers (29% versus 19%, p⬍0.001). Male IPAH patients had a higher frequency of diabetes (30% versus 12%). Despite adjusting for these differences, male SScPAH patients have decreased 1-, 2-, 3-, and 5-year survival (82.6%, 70.6%, 60.8%, 48.2%) compared to females (84.4%, 73.4%, 64.2%, 52.8%). Similarly, male IPAH patients have decreased 1-, 2-, 3-, and 5-year survival (93.4%, 87.9%, 84.8%, 77.7%) compared to females (94.5%, 91.0%, 88.7%, 83.2%). Conclusion: Sex disparities appear to exist in survival of SSc-PAH and IPAH patients. Further investigation is needed to evaluate this disparity, mechanisms for disparity, and the role of a targeted screening and treatment approach. Disclosure: C. Pasarikovski, None; J. T. Granton, Support respirology program at the hospital foundation., 9, Pfizer support of research study via CIHR grant., 9; P. Lee, None; A. M. Roos, None; A. T. Kron, None; C. Chau, None; S. R. Johnson, None.

S1099

Tuesday, October 29

outcomes. To achieve this goal, it is imperative to identify subgroups with high risk for developing PAH. Patients with systemic sclerosis (SSc) are known to have the highest risk for PAH. Serum autoantibodies, such as anticentromere, anti-U1RNP, anti-U3 RNP, and anti-Th/To, were also associated with PAH in certain CTD subgroups. However, little information is available for utility of autoantibody profiles for predicting PAH risk in patients with CTD. Here, we examined a role of autoantibody testing in identifying patients with PAH risk using a large-scale database of a routine autoantibody laboratory. Methods: This study enrolled 6,162 patients, whose sera were sent to our autoantibody laboratory between 1995 and 2008 because of clinical suspicion of CTD. They were selected from 9,872 consecutive patients based on observation period ⬎3 years, and availability of full clinical information on medical charts. Indirect immunofluorescence, double immunodiffusion, and RNA immunoprecipitation assay were routinely conducted to identify autoantibodies to centromere, topoisomerase I, Sm, U1RNP, SSA, SSB, Th/To, U3RNP, SRP, aminoacyl tRNA synthetase, and ribosome. Results: During 6.6 ⫾ 5.6 years of follow-up, 71 patients (1.2%) were diagnosed as having PAH confirmed by right heart catheterization. Mixed connective tissue disease (MCTD), SSc, and systemic lupus erythematosus (SLE) were clinical diagnosis associated with PAH (unadjusted odds ratio [OR] 10.8, 8.4, and 2.0, respectively). PAH also occurred in a small population of patients with primary Sjo¨gren’s syndrome (n ⫽ 7; 0.8%), rheumatoid arthritis (n ⫽ 3; 0.2%), or dermatomyositis (n ⫽ 1; 0.3%). Autoantibodies associated with PAH included those to centromere, U1RNP, Sm, SSA, SSB, and Th/To (unadjusted OR 4.5–6.6). Female gender and Raynaud’s phenomenon were also identified as PAH risk factors (unadjusted OR 8.0 and 10.5, respectively). Multivariate logistic regression analysis revealed that MCTD, SSc, and SLE were independent PAH risk, indicating that the overall PAH risk could be explained primarily by clinical diagnosis. Interestingly, anti-SSA antibody without diagnosis of MCTD, SSc, or SLE was another independent PAH risk, indicating limited utility of the autoantibody status. When we further developed a multivariate logistic regression model by combining the diagnosis and autoantibody profile, PAH risk can be explained by 6 independent variables, including SSc with anticentromere (OR 578, 95% confidence interval [CI] 119–10409), MCTD (OR 397, 95% CI 81–7174), SLE with anti-U1RNP (OR 150, 95% CI 29–2743), SSc without anticentromere (OR 103, 95% CI 18–1923), anti-SSA without diagnosis of MCTD, SSc, or SLE (OR 56, 95% CI 11–1028), and SLE without anti-U1RNP (OR 42, 95% CI 6.8–813). Conclusion: A combination of clinical diagnosis and autoantibody profiles effectively stratifies PAH risk in patients suspected to have CTD, and may aid in selection of patients who benefit from active screening program for PAH detection.

Tuesday, October 29

2578

2579

Pulmonary Hypertension In Systemic Sclerosis: Clinical Classification and Pulmonary Hypertension Subtypes. Monica Mohile1, Mary Lucas2, Virginia D. Steen3, Thomas A. Medsger Jr.2 and Robyn T. Domsic2. 1University of Pittsburgh Medical Center, Pittsburgh, PA, 2University of Pittsburgh, Pittsburgh, PA, 3Georgetown University Medical Center, Washington, DC.

Recommendations For Screening and Detection Of Connective-Tissue Disease Associated Pulmonary Arterial Hypertension. Dinesh Khanna1, Heather Gladue2, John D. Fitzgerald3, Richard N. Channick4, Lorinda Chung5, Oliver Distler6, Daniel Furst7, Eric Hachulla8, Marc Humbert9, David Langleben10, Stephen C. Mathai11, Rajeev Saggar12, Scott H. Visovatti13 and Vallerie McLaughlin2. 1University of Michigan Medical Center, Ann Arbor, MI, 2University of Michigan, Ann Arbor, MI, 3 UCLA School Med Rehab #32–59, Los Angeles, CA, 4Massachusetts General Hospital, Boston, MA, 5Stanford Univ Medical Center, Palo Alto, CA, 6Department of Rheumatology, University Hospital Zurich, Zurich, Switzerland, 7University of California in Los Angeles, Los Angeles, CA, 8 Claude Huriez University Hospital, Lille, France, 9Centre Chirurgical Marie-Lannelongue, Le Plessis-Robinson, France, 10McGill University, Montreal, QC, 11Johns Hopkins University, Baltimore, MD, 12St. Joseph’s Hospital and Medical Center, Phoenix, AZ, 13The University of Michigan, Ann Arbor, MI.

Background/Purpose: Pumonary hypertension (PH) is a significant complication of systemic sclerosis (SSc), with prevalence reports of 10–25%. Predictors of PH remain somewhat elusive. Our objective was to assess the pattern of PH timing and subtype development in limited versus diffuse cutaneous SSc. Methods: From our prospectively enrolled institutional scleroderma databank we performed a 10-year cross-sectional study of consecutive patients seen between January 1, 2000 and December 31, 2009. PH was defined as a mean pulmonary artery pressure (PAP) ⬎ 25 mmHg on right heart catheterization (RHC) or transthoracic echocardiogram (TTE) with PAP ⬎ 45 and PH diagnosed by a cardiologist. Descriptive statistics were used for baseline and PH characteristics, Kaplan Meier plots for time to development of PH and Cox proportional hazards for adjusted analysis. Results: Of the 1,156 SSc patients included, 80% were female, the average age at first SSc symptom was 43.7 ⫾ 14.2 years, and 44% had diffuse SSc. Two hundred thirteen (18%) had PH, of whom 122 (10% of total) had pulmonary arterial hypertension (PAH) and 90 (8% of total) had PH secondary to either ILD (PH-ILD) or cardiac involvement (PH-cardiac). 75% were confirmed by RHC; 25% did not have available RHC numbers in the hospital system but were confirmed as PH by cardiology evaluation. Prevalence of all PH was more common in limited (22%) vs diffuse SSc (14%; p⫽0.001). The profile of PH subtype was different, with 67% of limited SSc having PAH, compared to 37% in diffuse SSc (p ⬍ 0.0001). As shown in Table 1 approximately 5% of diffuse patients developed PAH with more secondary PH, whereas limited SSc patients developed predominantly PAH (p⫽⬍ 0.0001). Patients with limited SSc developed all PH later in disease at a median 12.0 (IQR 5.7, 20.7), compared to 7.7 (3.5, 14.7) years since first SSc symptom in diffuse SSc (p⫽0.02), and this persisted after adjustment for age and gender (p⫽0.02). However there was no difference in the time to development of PAH compared to secondary PH in either diffuse or limited SSc (Figure 1). Table 1. Prevalence of PH subtypes in limited and diffuse SSc

Limited cutaneous SSc Diffuse cutaneous SSc

NoPH 78% 86%

PAH 15% 5%

Secondary PH PH-ILD 5% 6%

PH-cardiac 2% 3%

Background/Purpose: Pulmonary arterial hypertension (PAH) affects up to 15% of patients with connective tissue diseases (CTD) and is one of the leading causes of mortality in systemic sclerosis (SSc) and mixed connective tissue disease (MCTD). Previous recommendationswere developed as part of larger efforts in PAH and did not provide detailed recommendations in CTD-PAH. To develop recommendations for screening and early detection of CTD-PAH using rigorous data-driven and consensus-building methodology. Methods: We performed a systematic review for the screening and diagnosis of PAH in CTD by searching available databases. Using the RAND/UCLA methodology, we developed case scenarios followed by 2 stages of voting—first one was voted anonymously by 10 international experts on 1 (inappropriate) ⫺9 (appropriate) scale and 2nd voting after face-to-face meeting. Results: The key recommendations include: 1. All patients with SSc should be screened for PAH. 2. MCTD or other CTD’s with scleroderma features (referred hereon as scleroderma-spectrum disorders) should be screened similar to patients with SSc. 3. Screening of asymptomatic patients is not recommended for MCTD or other CTD patients without features of scleroderma. 4. RHC is mandatory for diagnosis of PAH. 5. Acute vasodilator testing is not required as part of the evaluation of PAH in patients with SSc, SSc-spectrum disorders, or other CTDs. 6. Initial screening evaluation in patients with SSc and sclerodermaspectrum disorders include pulmonary function test (PFT) including diffusion capacity carbon monoxide (DLCO), Transthoracic echocardiogram (TTE), NT- Pro BNP, and DETECT algorithm if DLCO% ⬍ 60% and ⬎3 years disease duration. 7. In SSc and SSc-spectrum disorders, TTE and PFT should be performed on an annual basis or TTE, PFT, and NT-Pro BNP if new signs or symptoms develop. Recommendations for RHC for SSc and scleroderma-spectrum disorders Non-invasive test TTE

PFTs

Threshold for RHC TR velocity of 2.5–2.8 m/s TR velocity of ⬎2.8 m/s Right atrial (RA major dimension ⬎53 mm) or right ventricular enlargement (Mid cavity RV dimension ⬎35 mm), irrespective of TR velocity FVC/DLCO ratio ⬎1.6 &/or DLCO ⬍60%** FVC/DLCO ratio ⬎1.6 &/or DLCO ⬍60% & NT-Pro BNP ⬎2 times upper limit of normal** Meets DETECT algorithm in patients with DLCO ⬎60% & disease duration ⬎ 3 years

Signs or symptoms* required for RHC Yes No No

Yes No

Conclusion: These data suggest that although patients with limited SSc are more likely to develop PAH, the rate of PAH and secondary PH development is similar regardless of clinical subset. All SSc patients, including late diffuse SSc, should continue to be screened for all types of PH.

Composite measure

Disclosure: M. Mohile, None; M. Lucas, None; V. D. Steen, Gilead Science,, 2, Gilead Science, 5, Actelion Pharmaceuticals US, 2, Actelion Pharmaceuticals US, 8, Roche Pharmaceuticals, 2, Celgene, 2, Sanofi-Aventis Pharmaceutical, 2; T. A. Medsger Jr., None; R. T. Domsic, Actelion Pharmaceuticals US, 5.

* Symptoms: dyspnea on rest or exercise, fatigue, pre-syncope/syncope, chest pain, palpitations, dizziness, lightheadedness. Signs: Loud pulmonic sound, peripheral edema **TTE without overt systolic dysfunction, greater than grade I diastolic dysfunction or greater than mild mitral or aortic valve disease or evidence of PH.

S1100

No

Conclusion: We provide consensus-based and evidence-driven recommendations for screening and early detection of CTD-PAH. It is our hope that these recommendations will lead to early detection of CTD-PAH and ultimately improve patient outcomes. Disclosure: D. Khanna, NIH, 2, Scleroderma Foundation, 2, Actelion Pharmaceuticals US, 5, Actelion Pharmaceuticals US, 8, Gilead, 5, United Therapeutics, 5, United Therapeutics, 8, Roche Pharmaceuticals, 5, BMS, 5, DIGNA, 5, Merck Pharmaceuticals, 5; H. Gladue, None; J. D. Fitzgerald, None; R. N. Channick, Actelion, United Therapeutics, Bayer, 5, Actelion, United Therapeutics, 2; L. Chung, None; O. Distler, Sanofi, Active Biotech, Pfizer, Actelion, and Novartis, 2, Actelion, Sanofi-Aventis, Roche, Genentech, Active Biotec, medac, Novartis, Sinoxa, 4D Science, 5; D. Furst, AbbVie, Actelion, Amgen, BMS, Gilead, GSK, NIH, Novartis, Pfizer, Roche/ Genentech, UCB, 2, AbbVie, Actelion, Amgen, BMS, BiogenIdec, Janssen, Gilead, GSK, NIH, Novartis, Pfizer, Roche/Genentech, UCB, 5, AbbVie, Actelion, UCB, 8; E. Hachulla, Actelion, GSK, Pfizer, Lilly, United Therapeutics, 5; M. Humbert, None; D. Langleben, None; S. C. Mathai, None; R. Saggar, Gilead, Actelion, United Therapeutic, 9; S. H. Visovatti, None; V. McLaughlin, Actelion, Bayer, Gilead, Merck, United Therapeutics., 5, Gilead, United Therapeutics, 8, Actelion, Ikaria, Novartis, and United Therapeutics., 2.

2580 Does Mycophenolate Mofetil (MMF) Have An Effect On Pulmonary Hemodynamics? Observations From The Pulmonary Hypertension Assessment and Recognition Of Outcomes In Scleroderma (PHAROS) Cohort. Lesley Ann Saketkoo1, Matthew R. Lammi1, Jessica K. Gordon2, Paula Lauto3 and Virginia D. Steen4. 1LSU Scleroderma and Sarcoidosis Patient Care and Research Center, New Orleans, LA, 2Hospital for Special Surgery, New York, NY, 3LSU Health Sciences Center, New Orleans, LA, 4 Georgetown University Medical Center, Washington, DC.

Table 1. Comparison between SSc patients on No IS medications and MMF at time of first RHC. Continuous variables are reported as median, interquartile range; categorical values are proportional. No IS n Age Sex (% female) N (%) Limited SSc Time from 1st SSc Symptom (years) mPAP PVR PCWP FVC TLC FEV/FVC DLCO FVC:DLCO 6MWD

MMF

203 39 60 (52,68) 54 (47,63) 84% 65% 144 (71%) 14 (35%) 11.2 (5.3,21.0) 6.9 (3.0,10.3) 33 (23,44) 29 (25,35) 355 (242,692) 222 (162,344) 11 (3,14) 11 (9,15) 73.1 (64,88) 63.9 (48,84) 76.5 (66,93) 62.8 (47,80) 81 (75.37) 87 (81,93) 37.5 (30,50) 34.3 (28,39) 1.94 (1.6,2.4) 1.85 (1.4,2.5) 338.3 (238,428) 396.2 (343,475)

p Value *0.0143 0.6491 *⬍0.0001 *0.0004 *0.0016 *0.0006 0.7036 *0.0298 *0.0002 *0.0043 0.0929 0.6908 *0.0298

n Age Sex (% female) Time from 1st SSc symptom (years) Duration of MMF prior to RHC mRSS PH Meds at Time of RHC mPAP PVR PCWP FVC TLC FEV/FVC DLCO FVC:DLCO 6MWD

IcSSc on MMF

dcSSc on MMF

14 57.5 (50,53.8) 64% 8.4 (6.9,15.3)

23 51 (40,63) 67% 4.1 (2.6,8.8)

0.47 (0.18,1.6) 3 (2,5) 36% 29 (26,36) 234 (140,466) 11.5 (9.8,16.5) 75.9 (57.2,90.0) 68,9 (54,7, 46,8) 84.5 (79.0,88.5) 36.2 (31.4,40.2) 2.04 (1.4,2.6) 417.1 (362,502)

0.55 (0.15,1.2) 16 (7,26) 65% 29.0 (25.35) 222 (166,340) 11.0 (9.15) 61.7 (45.5,76.2) 60.3 (46.8,77.9) 87.0 (81.8,93) 32.5(25.2,38.6) 1.73 (1.2,2.3) 388.8 (241,466)

p Value 0.2185 1.0 *0.0176 0.9694 *⬍0.0001 0.1014 0.7507 0.7495 0.7621 0.1039 0.1898 0.3181 0.2862 0.5168 0.3384

Conclusion: Patients treated with MMF compared to the No IS group had lower mPAP and PVR at time of the diagnosis of PH with no difference between groups in PCWP. Differences in mPAP between groups were not explained by differences in age, FVC, or disease duration. These data suggest that MMF could potentially play a role in pulmonary artery remodeling and modifying the severity of PH. These findings warrant prospective controlled investigations of MMF in SSc PH. Disclosure: L. A. Saketkoo, Gilead Pharmaceuticals, 2, United Therapeutics, 2, Actelion Pharmaceuticals US, 2; M. R. Lammi, None; J. K. Gordon, None; P. Lauto, None; V. D. Steen, Gilead Science,, 2, Gilead Science, 5, Actelion Pharmaceuticals US, 2, Actelion Pharmaceuticals US, 8, Roche Pharmaceuticals, 2, Celgene, 2, SanofiAventis Pharmaceutical, 2.

2581 Can Changes In NT-Probnp Predict Early Response To Therapy and Prognosis In Systemic Sclerosis Associated Pre-Capillary Pulmonary Hypertension? Vincent Sobanski1, Bernadette Lynch2, Benjamin E. Schreiber3, Clive Handler2, Christopher P. Denton4 and John G. Coghlan5. 1 Royal Free Hospital, London, United Kingdom, 2The Royal Free Hospital, London, United Kingdom, 3National Pulmonary Hypertension Service, London, United Kingdom, 4Royal Free and University College Medical School, London, United Kingdom, 5The Royal Free Hospital NHS Foundation Trust, London, United Kingdom. Background/Purpose: Pulmonary hypertension (PH) is a severe complication of systemic sclerosis (SSc), affecting 5–12% of patients. Despite recent progress in treatment, prognosis remains poor. Early therapeutic management and goal-oriented approach can improve long-term prognosis. Response to therapy is usually assessed by functional and hemodynamic parameters between 3–6 months after initiation of treatment. This study aimed to compare the changes in NT-proBNP with functional and hemodynamic parameters between baseline and 3–6 months after initiation of therapy. Methods: A retrospective study, undertaken in a National Pulmonary Hypertension Centre, identified patients diagnosed with SSc associated pre-capillary PH on right heart catheterisation (RHC) (mean pulmonary arterial pressure (mPAP) ⱖ 25mmHg and pulmonary capillary wedge pressure (PCWP) ⱕ 15mmHg) between 1998 and 2012 (n⫽600). Patients were included if they had a second RHC between 3 and 6 months after the initial RHC diagnosing PH and if NT-proBNP results were available within 30 days prior to each RHC. 53 patients were identified. 6 patients were excluded (glomerular filtration rate ⬍ 30 mL/min/1.73m2). Changes in variables (⌬) between baseline and 3–6 months were calculated in absolute value, percentage of variation and logarithm of each value. Pearson or Spearman methods were used to estimate correlation coefficient, where appropriate. Patients were divided into two groups: D⫽“NT-proBNP decreasing” and I/S⫽“NT-proBNP increasing or stable” according to the difference in NT-proBNP levels between baseline and repeat RHC. Survival analyses were performed using Kaplan-Meier method and log-rank test. Results: 47 patients (42 female and 5 male) were included. 83% had limited cutaneous SSc and 53% were anti-centromere antibody positive. The mean age at diagnosis of PAH was 62.1 ⫾ 11.5 years; the mean time between

S1101

Tuesday, October 29

Background/Purpose: Systemic sclerosis (SSc) related pulmonary hypertension (PH) carries a high mortality; with SSc pulmonary arterial hypertension (PAH) having a 4x higher mortality then idiopathic PAH. It is unknown whether immunosuppressant (IS) drugs, particularly mycophenoloate mofetil (MMF), have any effect on vascular remodeling in SSc PH since it has not been formally studied. This analysis looks at the possible effects of MMF in SSc patients who have developed PH. Methods: PHAROS is a prospective registry designed to provide substantive data to recognize aspects of PH unique to SSc. Patients were stratified by history of MMF or No IS use (no MMF or other immunosuppressant drugs) at the time of the diagnosis of PH by right heart catheterization (RHC). Calculations are derived from non-parametric analyses using Mann-Whitney and Fisher’s Exact as applicable followed by regression analyses. Results: There were 39 SSc patients who had received MMF (mean duration 0.92 years) and 203 patients receiving No IS prior to diagnosis of PH. Patients treated with MMF when compared to the No IS group, were more likely to be younger, have diffuse SSc and have shorter disease duration. Patients treated with MMF had a significantly lower mean pulmonary artery pressure (mPAP) and pulmonary vascular resistance (PVR) with no difference in pulmonary capillary wedge pressure (PCWP) (Table 1). However, stratified analyses between diffuse and limited SSc patients on MMF revealed no significant differences in mPAP, PVR, PCWP nor in FVC, TLC, FEV/FVC nor DLCO (Table 2). In the group as a whole, MMF (p⫽0.046) and SSc subtype (p⫽0.01) were the only independent determinants of mPAP when adjusted for differences in age, FVC and disease duration.

Table 2. Comparison of patients on MMF by limited and diffuse subtypes.

both RHCs was 3.6 ⫾ 1.0 months and the mean follow-up after first RHC was 28.0 ⫾ 15.4 months. Six-minute walking distance (6MWD ⫺ p⫽0.002), mPAP (p⫽0.003), cardiac index (p⫽0.034) and pulmonary vascular resistances (PVR – p⬍0.0001) significantly improved in the group D. There was a trend in decreasing for PVR in the I/S group (p⫽0.094). In the total population, ⌬NT-proBNP was negatively correlated with ⌬6MWD in percentage (R⫽ ⫺0.305, p⫽0.050). ⌬NT-proBNP in percentage tended to correlate with ⌬mPAP in percentage (p⫽0.090) or log (p⫽0.073). No correlation was found between ⌬NT-proBNP in percentage or in log with PVR, cardiac output, cardiac index, venous oxygen saturation, right arterial pressure and PCWP. Survival was significantly better in patients with decreasing NT-proBNP in all patients with pre-capillary PH (p⫽0.045) and in patients with PAH (p⫽0.002). There was no difference for PH-ILD patients (p⫽0.658). Conclusion: Patients with decreasing NT-proBNP presented a significant improvement in hemodynamic parameters at 3–6 months. No strong correlation was found between ⌬NT-proBNP and changes in functional or hemodynamic variables. However, early changes in NT-proBNP seem to be associated with prognosis – especially in PAH patients. Disclosure: V. Sobanski, None; B. Lynch, None; B. E. Schreiber, None; C. Handler, None; C. P. Denton, None; J. G. Coghlan, None.

2582

Tuesday, October 29

Treatment Of Pulmonary Hypertension In Scleroderma Patients With Restricitive Lung Disease. Observations From The Pulmonary Hypertension Assessment and Recognition Of Outcomes In Scleroderma Cohort. Virginia D. Steen1 and Robyn T. Domsic2. 1Georgetown University Medical Center, Washington, DC, 2University of Pittsburgh, Pittsburgh, PA. Background/Purpose: Trials of therapy in pulmonary hypertension (PH) have generally excluded patients with significant interstitial lung disease, but many patients with systemic sclerosis(SSc) and PH have some interstitial lung disease. There are concerns that PH drugs may cause ventilation-perfusion mismatch and worsen PH, and it remains unclear if PH drugs improve outcomes for these patients. PHAROS, Pulmonary Hypertension Assessment and Recognition of Outcomes in Scleroderma, is a prospective observational multi-center cohort study of SSc-PH. Our objective is to describe the real-world experience of SSc-PH patients with restrictive lung disease treated with a single PH medication. Methods: We included patients with PH defined as a mean pulmonary artery pressure (mPAP) ⱖ25mmHg on catheterization who had a FVC ⬍70% predicted at the time of the diagnosis. Patients were grouped into 3 groups, based on the treatment during the first year following the PH diagnosis: No drug treatment, treatment with an endothelin receptor blocker antagonist (ERA), or andphosphodiesterase inhibitor (PDE). For this initial look at the data, patients treated with combination therapy and/or prostacyclins (PCA) were excluded. The primary outcome examined was one-year survival, with secondary outcomes of hospitalization, change in functional class, 6 minute walk and patient questionnaires at one year. Results: Of the 266 patients with PH in PHAROS, 98 had a FVC ⬍70% and were included. There were 26 receiving no therapy (No Drug), 19 receiving ERA and 33 treated with PDE. We excluded 18 treated with combination and 2 treated with PCAs alone. There were no significant differences in the demographics or the cardiopulmonary parameters between the 3 groups. These patients had striking restrictive disease with a mean FVC between 50 and 55% predicted. Although over survival was not statistically different between the 3 groups, patients in the PDE group had fewer deaths and a better 1 and 2 year survival than the No Drug group. (p⬍0.05). There were more patients in the PDE group who had an improvement compared to the ND group and neither treatment group had an increase in hospitalizations after starting the drugs (Table 1). Age–yr, mean Disease Duration yr, mean % Female sex % White race % Diffuse cutaneous FVC% predicted DLCO % predicted RHC m PAP PVR (Woods) Response in 1st year Hospitalizations Deaths in 1st year Survival, 1, 2 year

No Drug n ⴝ 26 54.5 10.1 70 76 42 52.2 33.6 31.85 293 23% 2 7 69%, 61%

ERA n ⴝ 19 57.2 8.3 79 63 63 49.9 25.9 33.3 343 37% 1 3 82%, 64%

PDE n ⴝ 33 56.8 9.2 62 64 42 55.5 42.0 36 419 43% 2 1 96%, 76%

Conclusion: SSc patients with PH and restrictive lung disease are being treated with PH drugs. Our observations suggest that some patients may respond favorably, without evidence of worsening PH. We strongly encourage further study of PH drugs in SSc-ILD patients in a controlled setting. Disclosure: V. D. Steen, Gilead Science,, 2, Gilead Science, 5, Actelion Pharmaceuticals US, 2, Actelion Pharmaceuticals US, 8, Roche Pharmaceuticals, 2, Celgene, 2, Sanofi-Aventis Pharmaceutical, 2; R. T. Domsic, None.

2583 Long-Term Treatment With Endothelin Receptor Antagonist Increases Peripheral Blood Perfusion In Systemic Sclerosis Patients. Maurizio Cutolo1, Barbara Ruaro1, Elena Bernero1, Francesca Ravera1, Giuseppe Zampogna1, Elisa Alessandri1, Vanessa Smith2 and Alberto Sulli1. 1Research Laboratory and Academic Unit of Clinical Rheumatology, Department of Internal Medicine, University of Genova, Genoa, Italy, 2Department of Rheumatology, Ghent University Hospital, Ghent, Belgium. Background/Purpose: Systemic sclerosis (SSc) is characterized by progressive impairment of the microvascular system and decrease of peripheral blood perfusion (PBP) (1–3). The vasoactive peptide endothelin-1 (ET-1) seems to be implicated in these events (4). Laser Doppler flowmetry (LDF) is a technique to evaluate blood perfusion at peripheral sites, such as the central area of the fingertips (2). The aim of this study was to investigate long-term effects of ET-1 receptor antagonism on PBP evaluated by LDF technique in SSc patients. Methods: Twenty-six SSc patients (mean age 62⫾12SD years, mean SSc duration 8⫾4 years) were enrolled during their programmed standard treatment protocols for digital ischaemia. At baseline (T0), 13 patients already receiving cyclic intravenous infusion of iloprost (average 80 mcg/day, for 5 continuous days, every three months), continued the treatment for further 3 years (ILO group: T1,T2,T3). The remaining 13 patients, although they continued the same cyclic intravenous iloprost treatment as the previous group, also received bosentan 125 mg twice a day for 3 years (ILO⫹BOS group: T1,T2,T3) since complaining digital ulcers. PBP was yearly evaluated in all SSc patients using LDF at both basal temperature and after heating of the LDF probe at 36°C to test microvascular dilation capacity. PBP was measured at the central area of the fingertips of the 2nd to 5thfinger bilaterally, scoring the average value as perfusion units (PU) (2). Non-parametric tests were used for the statistical analysis. Results: A progressive and statistically significant increase of PBP was observed in the ILO⫹BOS group at basal temperature during the follow-up (median PU T0 51, T1 74, T2 70, T3 85, respectively, p⫽0.007); at 36°C, PBP significantly increased only during the first two years of follow-up (median PU T0 81, T1 104, T2 110, respectively, p⫽0.0003; T3 105). In contrast, not statistically significant PBP changes were observed in ILO group at both basal temperature (median PU T0 104, T1 78, T2 55, T3 44, p⫽0.70) and 36°C (median PU T0 126, T1 108, T2 109, T3 87, p⫽0.45). Gradient of PU between the evaluations at basal temperature and at 36°C was found progressively decreased during the followup in the ILO group, but not in the ILO⫹BOS group. Two SSc patients experienced new digital ulcers in the ILO⫹BOS group (15%), and four in the ILO group (31%). No serious side effects were observed. Transient increase of liver transaminases was managed by temporary discontinuation of bosentan treatment in two cases. Conclusion: Long-term treatment with ET-1 receptor antagonist in combination with iloprost was found to significantly increase PBP in SSc patients, in contrast to the treatment with iloprost alone. This seems in agreement with the recent observation that long-term bosentan treatment reduces microvascular damage progression, as assessed by naifold capillaroscopy (5). References: 1. Cutolo M, et al. Nat Rev Rheumatol 2010; 6, 578–87. 2. Cutolo M, et al. J Rheumatol 2010; 37:1174–80. 3. Rosato E, et al. J Rheumatol 2010; 37: 2531–9. 4. Vancheeswaran R, et al. J Rheumatol 1994; 21:1838–44. 5. Cutolo M et al. J Rheumatol. 2013;40:40–5. Disclosure: M. Cutolo, Actelion Pharmaceuticals US, 2, BMS, 2, Mundipharm, 2, Sanofi-Aventis Pharmaceutical, 2; B. Ruaro, None; E. Bernero, None; F. Ravera, None; G. Zampogna, None; E. Alessandri, None; V. Smith, Actelion Pharmaceuticals US, 5; A. Sulli, None.

S1102

2584 Screening and Diagnostic Modalities For Systemic Sclerosis-Associated Pulmonary Arterial Hypertension: A Systematic Review. Heather Gladue1, Nezam I. Altorok1, Whitney Townsend1, Vallerie McLaughlin1 and Dinesh Khanna2. 1University of Michigan, Ann Arbor, MI, 2University of Michigan Medical Center, Ann Arbor, MI.

Single studies looked at EKG findings, cardiac MRI, Chest X-ray, CT, autoantibodies, laboratory values and physical exam findings. Screening Modality

Number of Studies

TTE (TR velocity)

12 (11 SSc, 1SLE)

NT-ProBNP

5 studies (2 cohort, 3 case control) All SSc

PFT’s

4 studies (3 cohort, 1 case control) All SSc 3 cohort studies All SSc

Composite Measures

Conclusion Studies had various inclusion criterion. VTR ranges to prompt RHC • 2.73–3.16 m/s without PAH-associated symptoms • 2.5–3.16 m/s with symptoms Various Cut offs ⬎236pg/ml sensitivity (45–93%) specificity (83–100%) Decreased DLCO was associated with PAH. 1. DLCO ⬎70.3% and FVC/DLCO ⱖ1.82 and NTProBNP ⱖ 209.8pg/ml ⫺100% sensitive and 100% specific. 2. DLCO/VA⬍70% and NT-ProBNP ⬎97th percentile for age 75% sensitive and 97% specific. 3. Cochin RPS score (RPS⫽0.0001107(age) ⫹0.0207818 (100-FVC) ⫹0.04905 (150-DLCO/ alveolar volume). A cut off of 2.73 sensitivity of 89.5 and specificity of 74.1% for PAH

Conclusion: Our systematic review shows that most evidence exists for use of TTE, pulmonary function tests, and NT-ProBNP for screening and diagnosis of PAH in CTD. These data will be used to develop guidelines for screening and diagnosis of PAH in CTD. Disclosure: H. Gladue, None; N. I. Altorok, None; W. Townsend, None; V. McLaughlin, Actelion, Bayer, Gilead, Merck, United Therapeutics., 5, Gilead, United Therapeutics, 8, Actelion, Ikaria, Novartis, and United Therapeutics., 2; D. Khanna, BMS, DIGNA, Roche, Actelion, Gilead, Merck, United Therapeutics, 5, National Institutes of Health, PHA, scleroderma foundation, 2.

2585 Left Ventricular Dysfunction Reflected By Higher Serum Brain Natriuretic Peptide Accounts For Poorer Prognosis Of Pulmonary Arterial HypertensionAssociatedWithSystemicSclerosis. SumiakiTanaka1,Yoshiyuki Arinuma1, Tatsuhiko Wada1, Tatsuo Nagai1, Jun Okada2 and Shunsei Hirohata1. 1Kitasato University School of Medicine, Sagamihara, Japan, 2 Kitasato Junior Collage of health and Hygienic Sciences, Minami-Uonuma, Japan.

Table. Serum levels of BNP in PAH patients with CTDs

WHO-FC

serum levels of BNP(pg/ml)* PAH patients with PAH patients non-SSc CTDs with SSc

I II III IV

35.0 (27.2–44.8) 67.1 (52.3–85.9) 108.3 (33.9–139.7) 224.4 (170.0–297.9)

54.0 (41.5–70) 103.5 (80.3–133.3) 167.1 (128.9–216.8) 346.4 (261.6–458.6)

*Means (95% CI) were estimated with mixed effects models (fixed effect: SSc, WHO-FC(4), random effect; patient). SSc p ⫽ 0.013, WHO-FC: p⬍0.0001

Conclusion: Our results demonstrate that left ventricular dysfunction coexists with PAH associated with SSc, accounting for poor survival of PAH associated with SSc. Thus, special attention to this complication would be required in the management of patients with PAH associated with SSc using PAH-specific drugs. Disclosure: S. Tanaka, None; Y. Arinuma, None; T. Wada, None; T. Nagai, None; J. Okada, None; S. Hirohata, None.

S1103

Tuesday, October 29

Background/Purpose: Pulmonary arterial hypertension (PAH) affects patients with connective tissue diseases (CTD), especially systemic sclerosis (SSc) and MCTD. Despite this, there continues to be delay in screening and diagnosis of these patients. We undertook a systematic review to summarize the best evidence for screening and diagnosis of PAH in CTD. Methods: A systematic search was performed in Pubmed, EMBASE, Web of Science and Scopus databases up to June 2012 with an experienced librarian, related to PH and CTD. We only focused on WHO group I PAH and excluded manuscripts that did not rule out interstitial lung disease (WHO group III) or left heart disease (WHO group II), or if the diagnosis of pulmonary arterial hypertension (PAH) was not made by RHC. Results: We started with 2805 titles, 838 abstracts, and included 21 manuscripts. Twelve studies assessed the tricuspid regurgitation velocity (VTR) or equivalent right ventricular systolic pressure (RVSP) using transthoracic echocardiogram (TTE) as a threshold for RHC in patients suspected as having PAH. 11 of these studies were in SSc, and one pertained to SLE. The screening threshold for RHC was VTR ⬎2.73 to ⬎3.16 m/s without symptoms or 2.5 to 3.0 m/s with symptoms and resulted in 20–67% of patients having RHC proven PAH (Table). In the SLE study, VTR ⬎3.0 m/s on TTE led to RHC in 3 patients and none had PAH. Three studies looked at pulmonary function tests and various DLCO% predicted cut-offs. These studies suggest that a low DLCO% 45–70% is associated with a 5.6–7.4% development of PAH, and the decline in DLCO% is associated with an increase in the specificity (For 60% cut off, spec⫽ 45%, and for 50% cut off, spec⫽90%]) for PAH. Five studies looked at NT-ProBNP and a cut-off ⬎209pg/ml has a high sensitivity (90–100%) and a specificity ranging from 45–95%.

Background/Purpose: Recentry, development of potent effective newer drugs for pulmonary arterial hypertension (PAH) have resulted in improving survival of the patients. However, the prognosis of patients with PAH associated with SSc remains still poor compared with that of patients without SSc. To explore the characteristics relevant to the poor survival of patients with PAH associated with SSc, we analyzed serum levels of brain natriuretic peptide (BNP). Methods: We analyzed 2786 measurements of serum levels of BNP (pg/ml) obtained from 83 patient with PAH associated with CTDs, including 41 patients with SSc, and 42 patients with CTDs other than SSc (16 patients with SLE, 20 patients with MCTD and 6 patients with other CTDs), who had been diagnosed as PAH based on right heart catheterization test and followed between April 2001 and March 2013 in our hospital. For statistical analysis, we converted serum levels of BNP into logarithmic values based on the distribution. Comparison of serum levels of BNP between PAH with SSc and that with non-SSc CTDs was carried out using mixed effects model in which we set presence or absence of SSc, WHO-functional class (WHO-FC) as fixed effects, and patient as a random effect. Comparison of hemodynamics between PAH with SSc and that with non-SSc CTDs was similarly carried out. Results: Significant deviation of the distribution of serum levels of BNP was observed between SSc and non-SSc CTDs (Figure). Thus, median (IQR) of serum level of BNP in SSc and non-SSc CTDs were 104.6 (52.5, 214) pg/ml, and 48.9 (19.9, 149.4) pg/ml, respectively. Mixed effects model showed that serum levels of BNP of each WHO-FC in SSc were significantly higher than those in non-SSc CTDs (Table). On right heart catheterization examinations, pulmonary capillary wedge pressure in SSc was significantly higher than that in non-SSc CTDs (8 (5.6, 11.0) mmHg, 6.3(5, 8) [median (IQR)], respectively; p⫽0.04), although there were no significant differences in mean pulmonary artery pressure, cardiac output, or pulmonary vascular resistance between both group.

2586

Tuesday, October 29

Antinuclear Antibody Negative Systemic Sclerosis Patients Have Less Vasculopathic Disease Manifestations. Gloria Salazar1, Shervin Assassi1, Fredrick M. Wigley2, Daniel E. Furst3, Laura K. Hummers2, Dinesh Khanna4, John Varga5, Elena Schiopu6, Virginia D. Steen7, Murray Baron8, Marie Hudson8, Janet E. Pope9, Monique E. Hinchcliff5, Marvin J. Fritzler10, David B. Robinson11, Robert W. Simms12, Richard M. Silver13, Tracy M. Frech14, Barri J. Fessler15, Jerry A. Molitor16, Sara Zamanian1, Niall Jones17, John D. Reveille1, Frank C. Arnett1 and Maureen D. Mayes1. 1University of Texas Health Science Center at Houston, Houston, TX, 2Johns Hopkins University School of Medicine, Baltimore, MD, 3UCLA Medical School, Los Angeles, CA, 4Division of Rheumatology, University of Michigan Medical Center, Ann Arbor, MI, 5Northwestern University Medical School, Chicago, IL, 6 University of Michigan Medical School, Ann Arbor, MI, 7Georgetown University Medical Center, Washington, DC, 8McGill University, Montreal, QC, 9St Joseph Health Care, London, ON, 10University of Calgary, Calgary, AB, 11University of Manitoba, Winnipeg, MB, 12Boston University School of Medicine, Boston, MA, 13Medical University of SC, Charleston, SC, 14 University of Utah School of Medicine, SLC, UT, 15University of Alabama at Birmingham, Birmingham, AL, 16Univ of MN MMC108, Minneapolis, MN, 17Rheumatology Clinic at 124th Street Medical Clinic, Edmonton, AB. Background/Purpose: Systemic Sclerosis (SSc) is an autoimmune disease characterized by fibrosis of skin and internal organs, as well as vasculopathy and immune dysregulation. Autoantibody formation is one of the hallmarks of SSc. Autoantibodies in patients with SSc carry considerable value in diagnosis and in predicting various clinical outcomes; however their role in the pathogenesis of SSc is unclear. While the great majority of patients with SSc have circulating antinuclear antibodies (ANA) (90–95%), a small percentage is ANA negative (5–10%). The detailed demographic and clinical characteristics of patients without ANA have not been clearly explored. The objective of this study was to examine the demographic and clinical characteristics of systemic sclerosis (SSc) patients without antinuclear antibodies (ANA) compared to ANA positive patients. Methods: SSc patients enrolled in the Scleroderma Family Registry and DNA Repository were included. Relevant demographic and clinical data were entered directly by participating sites or by chart review. Autoantibodies were determined at one site utilizing commercially available kits. Results: This study included 3249 patients, of whom 208 (6.4%) were ANA negative. The proportion of male patients was higher in the ANA negative group (OR 1.65 p⫽0.008). ANA negative patients experienced significantly less vasculopathic manifestations of SSc. The percent predicted diffusion capacity of carbon monoxide (DLco) was higher in ANA negative patients (p⫽0.03). Seven ANA negative patients had pulmonary arterial hypertension (PAH) per right heart catheterization (RHC) versus 213 ANA positive (OR⫽ 0.23 p⬍0.001) indicating that PAH was significantly less common in the ANA negative group. They also had less often telangiectasias and digital ulcers/ pits (p⫽0.01 and p⬍0.001, respectively). Although, diffuse cutaneous involvement was more common, the modified Rodnan Skin Score (mRSS) was lower in the ANA negative group (2.4 points lower, p⫽0.018). Furthermore, they experienced more malabsorption (p⫽0.003). There was no difference in the frequency of pulmonary fibrosis and scleroderma renal crisis. All-cause mortality was also not different between the two groups (p⫽0.28). The above observations remained significant after adjusting for potential confounders (age, disease duration, gender, disease type) (Table 1). Table 1. Multivariable analysis of clinical parameters in systemic sclerosis (SSc) patients who are ANA negative compared with ANA positive patients. Telangiectasias Digital ulcers and pits MPAP per RHC PAH per RHC

OR/coef

95% CI

p

0.59 0.38 ⫺5.58 0.28

0.38, 0.91 0.24, 0.59 ⫺10.85, ⫺0.32 0.11, 0.70

0.01 ⬍0.001 0.04 0.006

Conclusion: ANA negative patients constitute a distinct subset of SSc characterized by fewer vasculopathic features of the disease, a higher proportion of men, more frequent lower gastrointestinal involvement as well as higher proportion of diffuse cutaneous disease but less severity of skin fibrosis overall.

Disclosure: G. Salazar, None; S. Assassi, None; F. M. Wigley, None; D. E. Furst, Abbott, Actelion, Amgen, BMS, Gilead, GSK, NIH, Novartis, Pfizer, Roche/Genentech, UCB, 2, Abbott, Actelion, Amgen, BMS, Janssen, Gilead, GSK, NIH, Novartis, Pfizer, Roche/Genentech, UCB, 5, Abbott, Actelion, UCB, 8; L. K. Hummers, Actelion Pharmaceuticals US, 2, medimmune, 2; D. Khanna, NIH, 2, Scleroderma Foundation, 2; J. Varga, None; E. Schiopu, None; V. D. Steen, Gilead Science,, 2, Gilead Science, 5, Actelion Pharmaceuticals US, 2, Actelion Pharmaceuticals US, 8, Roche Pharmaceuticals, 2, Celgene, 2, Sanofi-Aventis Pharmaceutical, 2; M. Baron, None; M. Hudson, None; J. E. Pope, None; M. E. Hinchcliff, None; M. J. Fritzler, Inova Diagnostics, Inc., 5; D. B. Robinson, None; R. W. Simms, None; R. M. Silver, Genentech and Biogen IDEC Inc., 8, Actelion Pharmaceuticals US, 9; T. M. Frech, None; B. J. Fessler, None; J. A. Molitor, None; S. Zamanian, None; N. Jones, None; J. D. Reveille, None; F. C. Arnett, None; M. D. Mayes, None.

2587

SEE ABSTRACT #2919

2588 Pulmonary Hypertension In Patients With Anti-U1-RNP Antibodies. Vincent Sobanski1, Bernadette Lynch2, Benjamin E. Schreiber3, Svetlana I. Nihtyanova4, Jennifer Harvey1, Clive Handler2, Christopher P. Denton4 and John G. Coghlan5. 1Royal Free Hospital, London, United Kingdom, 2The Royal Free Hospital, London, United Kingdom, 3National Pulmonary Hypertension Service, London, United Kingdom, 4Royal Free and University College Medical School, London, United Kingdom, 5The Royal Free Hospital NHS Foundation Trust, London, United Kingdom. Background/Purpose: Pulmonary hypertension (PH) is a leading cause of morbidity and mortality in patients with connective tissue diseases (CTD). Patients with anti-U1-RNP antibodies belong to a heterogeneous group, including systemic sclerosis (SSc), systemic lupus erythematosus (SLE) or mixed connective tissue disease (MCTD). This study aimed to (i) describe the prevalence, clinical and hemodynamic characteristics of PH in U1-RNP positive patients and (ii) analyse survival and compare prognosis between the different CTDs associated with these antibodies. Methods: In this National Pulmonary Hypertension database, more than 2000 patients had a right heart catheterisation (RHC) performed between 1998 and 2012. One hundred and twelve patients were tested anti-U1-RNP positive. Twenty-two patients did not have PH (mean Pulmonary Arterial Pressure [mPAP] ⬍ 25 mmHg). In the 90 patients with PH, 10 had post-capillary PH (pulmonary capillary wedge pressure [PCWP] ⬎ 15 mmHg). Eighty had pre-capillary PH (PCWP ⱕ 15 mmHg). Six patients were excluded because of incomplete RHC data. Thirty-seven had PH due to lung disease (PH-ILD) and 37 had pulmonary arterial hypertension (PAH). Anti-U1-RNP positive patients with PH were compared with unselected U1-RNP negative patients also having PH: 387 SSc and 21 SLE. For binary variables, comparison between 3 PH groups was performed using Monte-Carlo method with Fischer test for comparison one by one. ANOVA was used to compare continuous variables with a normal distribution. Kruskal-Wallis test was used for non-parametric comparisons. Kaplan-Meier method with log-rank test was used for survival analysis. All statistical analyses were performed using SPSS for Microsoft Windows®. Results: The proportion of patients with SSc, SLE or MCTD was not significantly different between the 3 groups of PH. Anti-Sm antibodies were only seen in PH-ILD (p⫽0.038). Six-minute walking distance and WHO FC were significantly better in PAH vs. PH-ILD (p⫽0.003 and p⫽0.007 respectively). Mean PAP was similar in 3 groups. In U1-RNP negative patients, mPAP was significantly higher in SLE than in SSc patients (p⫽0.003). In PAH patients, SLE patients had higher mPAP than others groups (p⫽0.023). When comparing U1-RNP positive vs. negative in SSc or SLE patients, no difference was observed in hemodynamic values. Among PH-ILD patients, U1-RNP negative SSc had lower DLCO than U1-RNP positive (p⫽0.001). MCTD patients had lower CI (p⫽0.007) and higher PVR (p⫽0.05) than U1-RNP negative SSc patients. SLE and MCTD had a better survival than SSc (p⫽0.005 and p⫽0.037, respectively). PH-SSc patients carrying the U1-RNP antibody tended to have a better survival than U1-RNP negative (p⫽0.065). This trend was found in PAH patients (p⫽0.068) but not in PH-ILD patients (p⫽0.225). Conclusion: These data confirm that patients with anti-U1-RNP antibodies are heterogeneous. To our knowledge, this is the largest cohort of patients with PH and MCTD. It provides hemodynamic data for all patients at PH diagnosis. In this group, patients with SSc have the worse

S1104

prognosis. Interestingly, survival of patients fulfilling MCTD classification criteria is intermediate between SSc and SLE. Disclosure: V. Sobanski, None; B. Lynch, None; B. E. Schreiber, None; S. I. Nihtyanova, None; J. Harvey, None; C. Handler, None; C. P. Denton, None; J. G. Coghlan, None.

2589 Scleroderma Patients With Pulmonary Hypertension and Increased Pulmonary Capillary Wedge Pressure In The Pulmonary Hypertension Assessment and Recognition Of Outcomes In Scleroderma (PHAROS) Cohort. Matthew R. Lammi1, Lesley Ann Saketkoo2, Jessica K. Gordon3, Paula Lauto4 and Virginia D. Steen5. 1Louisiana State University Health Sciences Center, Pulmonary and Critical Medicine, New Orleans, LA, 2LSU HSC - New Orleans, Sections of Rheumatology and Pulmonary Medicine, New Orleans, LA, 3Hospital for Special Surgery, New York, NY, 4LSU Health Sciences Center, New Orleans, LA, 5Georgetown University Medical Center, Washington, DC.

Table. Baseline characteristics in patients dying prior to 2 years of follow-up compared to those who survived for more than 2 years Parameter Age (years) Sex (% female) SSc disease duration (years) Scleroderma type (%) Limited Diffuse NYHA functional class I/II III/IV Immunosuppressant use at baseline (% yes) 6-minute walk distant (meters) Berg Dyspnea Score Mean pulmonary artery pressure (mmHg) Pulmonary capillary wedge pressure (mmHg)

Death within 2 years 58.3 ⫾ 15.5 56% 8.9 ⫾ 6.8

Survival > 2 years 54.1 ⫾ 12.3 85% 8.2 ⫾ 8.4

40% 50%

55% 35%

0.71

30% 70% 20%

73% 27% 44%

p ⫽ 0.05

186 ⫾ 96 5.7 ⫾ 3.2 40.0 ⫾ 9.8

366 ⫾ 107 2.7 ⫾ 2.0 32.7 ⫾ 9.1

p ⫽ 0.002 p ⫽ 0.03 p ⫽ 0.05

22.9 ⫾ 4.3

18.6 ⫾ 2.4

p ⫽ 0.001

p value 0.68 0.16 0.83

0.40

Disclosure: M. R. Lammi, None; L. A. Saketkoo, Gilead Pharmaceuticals, 2, United Therapeutics, 2, Actelion Pharmaceuticals US, 2; J. K. Gordon, None; P. Lauto, None; V. D. Steen, Gilead Science,, 2, Gilead Science, 5, Actelion Pharmaceuticals US, 2, Actelion Pharmaceuticals US, 8, Roche Pharmaceuticals, 2, Celgene, 2, Sanofi-Aventis Pharmaceutical, 2.

2590 Laser Speckle Contrast Analysis A New Method To Evaluate Blood Perfusion in Different Skin Areas Of Systemic Sclerosis Patients: Comparison With Laser Doppler Flowmetry. Alberto Sulli1, Barbara Ruaro1, Giorgia Ferrari1, Elena Bernero1, Carmen Pizzorni1, Vanessa Smith2 and Maurizio Cutolo1. 1Research Laboratory and Academic Unit of Clinical Rheumatology, Department of Internal Medicine, University of Genova, Genoa, Italy, 2Department of Rheumatology, Ghent University Hospital, Ghent, Belgium. Background/Purpose: In systemic sclerosis (SSc), blood perfusion (BP) may be evaluated by both laser speckle contrast analysis (LASCA) and laser Doppler flowmetry (LDF) techniques, as well as microangiopathy may be assessed by nailfold videocapillaroscopy (NVC) (1–3). This study aimed at investigating BP by LASCA in different skin areas of SSc patients, looking for any correlations with the extent of the nailfold microvascular damage. Correlations between LASCA and LDF analysis at fingertips were also checked. Methods: Sixty-eight SSc patients and 68 sex and age matched healthy subjects were enrolled. The BP was assessed in the area of fingertips, periungual areas, dorsum and palm of both hands, as well as of face, forehead, tip of nose, zygomas and perioral region by LASCA, and scored as perfusion units (PU) (1). LDF was performed at the level of 2nd, 3rd, 4th, and 5th fingertip bilaterally (2), and average blood perfusion recorded as PU (2). NVC was performed in SSc patients and images classified and scored as previously reported (4–6). Non-parametric tests were used for the statistical analysis. Results: Median BP was significantly lower in SSc patients when compared to healthy subjects at fingertips (86 vs 189 PU, respectively, p⬍0.0001), periungual areas (69 vs 140 PU, respectively, p⬍0.0001) and palms (77 vs 111 PU, respectively, p⬍0.0001). Conversely, both groups showed similar BP values at dorsum of hands and different areas of the face. The median BP gradient between fingertips and palm was lower in SSc patients than in healthy subjects (11 vs 67 PU, respectively, p⬍0.0001), as well as was the gradient between dorsum and periungual areas (25 vs 69 PU, respectively, p⫽0.0009). There was a statistically significant progressive decrease of BP in SSc patients with different NVC pattern of microangiopathy (“early”, “active”, or “late”), as well as a statistically significant negative correlation between microangiopathy evolution score (MES) and BP, at the level of fingertip areas (p⬍0.004), periungual areas (p⬍0.01) and palms (p⬍0.02), but not in the other areas. A positive correlation was detected between LASCA and LDF values at fingertip level, in all subjects (p⬍0.0001). LASCA, since evaluating large skin areas, is significantly less time consuming, is more accepted by patients and shows lower intra-operator variability than LDF (95% vs 88%). Conclusion: This study shows that LASCA technique detects differences of BP at fingertips, periungual and palm areas between SSc patients and healthy subjects. Statistically significant correlations were observed between nailfold microangiopathy damage extent and BP at the level of fingertips, periungual and palm areas. LASCA positively correlates with LDF values at fingertips, but brings an higher reproducibility. References: 1. Draijer M, et al. Laser Med Sci (2009); 24: 639–51. 2. Cutolo M, et al. J Rheumatol 2010; 37:1174–80. 3. Cutolo M, et al. Nature Rev Rheumatol 2010; 6, 578–87. 4. Cutolo M, et al. Rheumatology 2004; 43:719–26.4. 5. Sulli A, et al. Ann Rheum Dis 2008;67:885–7. 6. Smith V, et al. Ann Rheum Dis 2010; 69:1092–6. Disclosure: A. Sulli, None; B. Ruaro, None; G. Ferrari, None; E. Bernero, None; C. Pizzorni, None; V. Smith, Actelion Pharmaceuticals US, 5; M. Cutolo, Actelion Pharmaceuticals US, 2, BMS, 2, Mundipharm, 2, Sanofi-Aventis Pharmaceutical, 2.

S1105

Tuesday, October 29

Background/Purpose: Systemic sclerosis (SSc) commonly leads to pulmonary hypertension (PH), which may be associated with left heart disease and an elevated pulmonary capillary wedge pressure (PCWP). Patients in this PH subgroup may have pulmonary artery pressures (PAP) in proportion to their elevated PCWP (pulmonary venous hypertension, “PVH”) or higher than expected given their PCWP (“out-of-proportion PH”). It is not known what causes this difference in patients with SSc. Methods: Baseline characteristics from 48 patients in the PHAROS registry (a prospective observational cohort of SSc patients at high risk for or with PH) who had a PCWP⬎15 on their inclusion right heart catheterization (RHC) were retrospectively analyzed. Using unpaired t-tests, characteristics of those who died before 2 years of follow-up (n⫽10) were compared to those who were alive at 2 years (n⫽20). Patients were divided into 2 groups based on their initial RHC diastolic pressure gradient (DPG⫽diastolic PAP minus PCWP): PVH (DPG ⱕ5mmHg) or out-of-proportion PH (DPG䡺5mmHg). Comparisons were made between groups using unpaired t-tests or Chi square. Kaplan-Meier analysis compared survival and time to first hospitalization between the 2 groups. Results: At baseline, the mean PAP (mPAP) was 36.8⫾11.8 mmHg and the PCWP was 19.4⫾3.3mmHg. Univariate factors associated with death prior to 2 years are shown in the table. In multivariate analysis, the only independent factors associated with death prior to 2 years were lower 6MWD (p⫽0.01) and higher PCWP (p⫽ 0.01). The out-of-proportion PH group (n⫽26) had higher baseline mPAP (42.7⫾13.0 vs. 29.7⫾3.7mmHg, p⬍0.0001), DPG (12.7⫾8.6 vs. 2.9⫾1.5mmHg, p⬍0.0001), and pulmonary vascular resistance (376⫾235 vs. 204⫾101 dynes/sec/cm5, p⫽0.003) compared to the PVH group (n⫽22). Although there was no difference in baseline immunosuppression use overall, mycophenolate use was less common in the out-of-proportion PH group (8% vs. 37%, OR 0.15, p⫽0.027). There were no differences between the PVH and the out-of-proportion PH groups in age, sex, disease duration, pulmonary function, SSc subtype, or 6MWD (all p⬎0.05). There was no difference in 3-year survival between the 2 subgroups (PVH: 1-year⫽95%, 3-year⫽61%; out-of-proportion PH: 1-year⫽85%, 3-year⫽63%, p⫽0.73). There was a trend towards a shorter time to first hospitalization in the out-of-proportion PH group (p⫽0.13).

Conclusion: In patients with SSc-PH and a PCWP⬎15, lower 6MWD and higher PCWP were independently associated with an increased risk for death within 2 years. In SSc patients with a PCWP⬎15, those with out-of-proportion PH were less likely to be on mycophenolate (MMF) compared to those with pulmonary venous hypertension. The relationship between MMF and PH in SSc needs further investigation, as MMF’s anti-fibrotic effects may theoretically decrease pulmonary artery remodeling in these patients.

2591

Tuesday, October 29

Scleroderma Renal Crisis and Pulmonary Arterial Hypertension Are Very Rare In The Same Patient. Bernadette Lynch1, Vincent Sobanski1, Clive Handler1, Benjamin E. Schreiber2, John G. Coghlan3, Voon H. Ong4, Aine Burns1 and Christopher P. Denton5. 1The Royal Free Hospital, London, United Kingdom, 2National Pulmonary Hypertension Service, London, United Kingdom, 3The Royal Free Hospital NHS Foundation Trust, London, United Kingdom, 4The Royal Free and University College Medical School, London, United Kingdom, 5Department of Rheumatology, London, United Kingdom. Background/Purpose: Systemic Sclerosis (SSc) is associated with occlusive vasculopathy resulting in digital ischaemia, telangiectasia, scleroderma renal crisis (SRC) and pulmonary arterial hypertension (PAH). SRC and PAH have similar histological features with neointimal proliferation and medial hyperplasia and shared pathogenic mechanisms are plausible. Pulmonary hypertension (PH) after SRC is frequently observed but generally due to post capillary mechanisms with elevated post capillary wedge pressure (WHO Group II). We have explored the association of PAH (WHO Group I) with SRC based upon shared histopathological features and possible similarities in pathogenesis. Methods: Of more than 2000 SSc patients attending our institution between 1990–2013, 150 patients had a confirmed diagnosis of SRC of which six (4%) patients had a diagnosis of PAH (mean pulmonary arterial pressure ⱖ25mmHg and pulmonary capillary wedge pressure ⬍15mmHg) following right heart catherisation. Estimated glomerular filtration rate (eGFR) was calculated using the 4-variable Modified Diet in Renal Disease (MDRD) equation. Demographic, clinical and haemodynamic parameters are expressed as mean ⫾ SD. Results: Two patients had diffuse cutaneous SSc (dcSSc) and 4 patients had limited cutaneous SSc (lcSSc). 5/6 (83%) demonstrated an immunofluoresecence and ENA reactivity consistent with anti-RNA polymerase III reactivity, confirmed by specific ELISA. All patients with SRC and PAH were female. PAH was diagnosed after SRC in all patients (49 vs 44 years old). 4/6 (66.7%) patients required dialysis. 2/6 (33.3%) patients who did not require dialysis had an eGFR of 26 and 44mls/min/1.73m2. 2 patients recovered renal function at 6 and 14 months after the diagnosis of SRC. The mean mPAP (mean pulmonary arterial pressure) was 42 ⫾ 12mmHg with a mean pulmonary capillary wedge pressure (PCWP) 9.3 ⫾ 2.6mmHg and a cardiac index of 5.2 ⫾ 0.8 L/min. 5 patients were treated with Bosentan and 1 patient with intravenous Iloprost. 3/6 (50%) patients died with a mean survival from date of diagnosis of SRC of 112 ⫾ 30 months and a mean survival from date of diagnosis of PAH of 47 ⫾ 17 months. Conclusion: The presence of SRC and PAH in the same patient is very rare and PAH occurred after SRC in all patients at a mean of 5 years later. Temporal separation of SRC and PAH suggests that these complications may occur independently, but the frequent association with ARA suggests common susceptibility factors may be relevant. Disclosure: B. Lynch, None; V. Sobanski, None; C. Handler, None; B. E. Schreiber, None; J. G. Coghlan, None; V. H. Ong, None; A. Burns, None; C. P. Denton, None.

2592

tion. Correlations of RHI with various clinical symptoms, disease duration, laboratory data and disease subtypes were examined. Results: RHI values were inversely correlated with mPAP (r ⫽ ⫺0.47, p ⬍ 0.01), while positively correlated with %DLco (r ⫽ 0.35, p ⬍ 0.05). Furthermore, the values of RHI inversely correlated with disease duration in dcSSc patients (r ⫽ ⫺0.45, p ⬍ 0.05), but not in lcSSc patients. Of note, RHI values were increased after one month of oral prednisolone in 4 patients with progressive skin sclerosis in whom skin stiffness was improved in response to the treatment. Conclusion: Although mPAPs of all the patients enrolled in this study were less than 25 mmHg, the values of RHI were significantly and inversely correlated with mPAP, suggesting that the decreased RHI value reflects the reduction of the pulmonary vascular bed prior to the development of definite pulmonary arterial hypertension in SSc. Since skin sclerosis may affect the values of RHI, we have to be cautious to interpret the impact of prednisolone on endothelial function evaluated by EndoPAT in this disease. Disclosure: N. Aozasa, None; Y. Asano, None; M. Hatano, None; R. Saigusa, None; K. Takakuwa, None; T. Takahashi, None; T. Toyama, None; H. Sumida, None; A. Yao, None; K. Kinugawa, None; H. Maki, None; T. Inaba, None; H. Muraoka, None; S. Minatsuki, None; I. Komuro, None; S. Sato, None.

2593 The Prognosis Of Scleroderma Renal Crisis In RNA-Polymerase III Antibody (ARA) Positive Compared To ARA Negative Patients. Bernadette Lynch1, Henry Penn2, Jennifer Harvey3, Aine Burns1 and Christopher P. Denton4. 1The Royal Free Hospital, London, United Kingdom, 2Northwick Park Hospital, Harrow, United Kingdom, 3Royal Free Hospital, London, United Kingdom, 4UCL Medical School, London, United Kingdom. Background/Purpose: Scleroderma renal crisis (SRC) usually presenting with accelerated hypertension and acute kidney injury (AKI) is one of the most severe complications of Systemic Sclerosis (SSc). The presence of RNA-polymerase III auto-antibodies (ARA) is recognized as a strong risk factor for SRC but studies have not explored long-term outcomes in ARA positive cases compared to ARA negative cases. Methods: Of more than 2000 SSc patients attending our institution between 1990–2013, 150 patients had a confirmed SRC. 80% patients had diffuse cutaneous SSc (dcSSc) and 20% patients had limited cutaneous SSc (dcSSc). ARA was measured by commercial ELISA or radio-immunoprecipitation. Patients were divided into two groups: ARA positive or ARA negative. Demographic and clinical parameters were compared between groups using Student’s t-test or Chi-squared analyses where appropriate. Results: 61/150 (41%) patients were ARA positive and significantly more likely to have dcSSc (88.3% vs 73.8%, p⫽0.032) than lcSSc compared to ARA negative patients. There was no significant difference in age of onset of SRC (51.2 vs 51.9 years) or the number of females (73% vs 79%) between the two groups. 50.8% of ARA positive patients required dialysis compared to 29.2% of ARA negative patients (p⫽0.07). The mean time to recovery of renal function was significantly longer in ARA positive patients (14.25 vs 8.27 months, p⫽0.032). Significantly more ARA positive patients were able to discontinue dialysis compared to ARA negative patients (53.3% vs 25.5%, p⫽0.01). ARA positive patients had a significantly better survival outcome (figure 1).

Clinical Significance Of Endothelial Vasodilator Function Evaluated By Endopat In Patients With Systemic Sclerosis. Naohiko Aozasa, Yoshihide Asano, Masaru Hatano, Ryosuke Saigusa, Kouta Takakuwa, Takehiro Takahashi, Tetsuo Toyama, Hayakazu Sumida, Atsushi Yao, Koichiro Kinugawa, Hisataka Maki, Toshiro Inaba, Hironori Muraoka, Shun Minatsuki, Issei Komuro and Shinichi Sato. University of Tokyo Graduate School of Medicine, Tokyo, Japan. Background/Purpose: Systemic sclerosis (SSc) is a multisystem autoimmune disease characterized by initial vascular injuries and resultant fibrosis of skin and certain internal organs. Evidence has shown that vascular impairment in SSc may be a sign of endothelial dysfunction. The peripheral arterial tonometry (EndoPAT) is a device that is used to assess endothelial vasodilator function in a rapid and non-invasive fashion. Since its clinical significance has not been well established in SSc, we evaluated the association between the values of reactive hyperemia measured by EndoPAT and clinical features of SSc. Methods: Thirty-three consecutive patients with SSc were studied, including 15 patients with limited cutaneous SSc (lcSSc) and 18 with diffuse cutaneous SSc (dcSSc). Clinical symptoms, such as swollen fingers, nailfold bleeding (NFB), pitting scars, digital ulcers, and Raynaud’s phenomenon, were meticulously recorded. Reactive hyperemia peripheral arterial tonometry index (RHI) was measured using EndoPAT on all patients. Mean pulmonary artery pressure (mPAP) was measured by right heart catheteriza-

Figure 1. Kaplan Meier survival estimate showing the effect of ARA positive auto-antibodies in patients with SRC.

S1106

Conclusion: In SRC, although more ARA positive patients required dialysis they also had significantly greater capacity for long-term recovery and survival compared to ARA negative patients. Disclosure: B. Lynch, None; H. Penn, None; J. Harvey, None; A. Burns, None; C. P. Denton, None.

2594 Stabilisation Of Microcirculation In Early Systemic Sclerosis Patients With Diffuse Skin Involvement Following Rituximab Treatment. Vanessa Smith1, Carmen Pizzorni2, Valeria Riccieri3, Saskia Decuman4, Yves P. Piette1, Ellen Deschepper5, Filip De Keyser1 and Maurizio Cutolo6. 1Department of Rheumatology, Ghent University Hospital, Ghent, Belgium, 2Research Laboratory and Academic Unit of Clinical Rheumatology, Department of Internal Medicine, University of Genova, Genoa, Italy, 3Department of Internal Medicine and Medical Specialities, University Sapienza, Rome, Italy, 4 Department of Internal Medicine, Ghent University, Ghent, Belgium, 5Biostatistics Unit, Department of Public Health, Ghent University, Ghent, Belgium, 6University of Genova, Genova, Italy.

References: 1. Koenig M, Joyal F, Fritzler MJ, Roussin A, Abrahamowicz M, Boire G et al. Autoantibodies and microvascular damage are independent predictive factors for the progression of Raynaud’s phenomenon to systemic sclerosis: A twenty-year prospective study of 586 patients, with validation of proposed criteria for early systemic sclerosis. Arthritis Rheum 2008; 58: 3902–12. 2. Sulli A, Secchi ME, Pizzorni C, Cutolo M. Scoring the nailfold microvascular changes during the capillaroscopic analysis in systemic sclerosis patients. Ann Rheum Dis 2008; 67: 885–7. Disclosure: V. Smith, Actelion Pharmaceuticals US, 5; C. Pizzorni, None; V. Riccieri, None; S. Decuman, None; Y. P. Piette, None; E. Deschepper, None; F. De Keyser, Abott, 5, UCB, 5, MSD, 5, Roche Pharmaceuticals, 5, GSK, 5; M. Cutolo, Actelion Pharmaceuticals US, 2, BMS, 2, Mundipharm, 2, Sanofi-Aventis Pharmaceutical, 2.

2595 Microvascular Abnormalities In Patients With Early Systemic Sclerosis: Less Severe Morphological Changes Compared To Patients With Definite Disease. Cintia Camargo, Juliana Sekiyama, Maria I. Arismendi and Cristiane Kayser. Universidade Federal de Saõ Paulo, Saõ Paulo, Brazil. Background/Purpose: To analyze the morphological and functional abnormalities of the microcirculation through widefield nailfold capillaroscopy (NFC), videocapillaroscopy and cutaneous microvascular blood flow measurement before and after cold stimulus, associated with vascular injury markers in patients with early systemic sclerosis (SSc), compared to primary Raynaud’s phenomenon (PRP), definite SSc, and healthy controls.

Disclosure: C. Camargo, None; J. Sekiyama, None; M. I. Arismendi, None; C. Kayser, None.

2596 Association Between Nail Fold Capillaroscopy Abnormalities and Thermographic Assessment Of Peripheral Microvascular Dysfunction In An Unselected Cohort Of Patients Under Investigation For Symptoms Of Raynaud’s Phenomenon. Bhavisha Vasta1, Marina Scolnik2, Darren Hart1, Jacqueline A. Shipley1, Sue Brown1, Eleanor Korendowych1, Neil J McHugh3 and John D. Pauling3. 1Royal National Hospital for Rheumatic Diseases, Bath, United Kingdom, 2Rheumatology Section, Hospital Italiano de Buenos Aires, Buenos Aires, Argentina, 3Department of Pharmacy and Pharmacology, University of Bath, Bath, United Kingdom. Background/Purpose: Nail fold capillaroscopy (NC) and infrared thermography (IRT) allow objective assessment of digital microvascular abnormalities in patients with Raynaud’s phenomenon (RP) and have an important role in disease classification, particularly in systemic sclerosis (SSc). The relationship between NC appearances and assessment of digital microvascular function with IRT in an unselected population of patients with RP symptoms has not previously been explored. Methods: A retrospective review of all patients investigated with both NC and IRT for symptoms of RP between August 2010 and November 2012 was undertaken. Thermographic assessment of the resting longitudinal thermal gradient of all fingers was undertaken by calculating the mean distal dorsal difference (DDD) at 23oC (without prior knowledge of the clinical diagnosis or NC findings). NC images were reviewed (BV and MS) and categorized as normal, non-specific abnormalities and SSc-pattern abnormalities (Cutolo, Clin Exp Rheumatol. 2007). Indeterminate cases were reviewed by a 3rd assessor (JP) and consensus reached. A subsequent case note review was undertaken to categorize patients according to the following diagnoses; primary RP (no clinical features of connective tissue disease and ANA negative), SSc (sclerodactyly in conjunction with a SSc-specific autoantibody), fibromyalgia (FMS) (1990 criteria) and other diagnoses. Results: One hundred and forty one patients were identified. NC appearances, IRT analysis and diagnoses are summarized in the accompany-

S1107

Tuesday, October 29

Background/Purpose: Microangiopathy in systemic sclerosis is progressive over time [1, 2]. This study assesses microangiopathic evolution by nailfold videocapillarocopic (NVC) analysis after two treatment course (month 0/6) with rituximab in early diffuse systemic sclerosis (dcSSc) patients. Methods: Twelve months follow-up (open-label study) of six consecutive patients with early dcSSc. Patients received an infusion of two times 1000 mg rituximab at month 0 and 6, together with 100 mg methylprednisolone. All patients were on a stable dose methotrexate (10–25 mg/week) as background therapy since at least 12 weeks. Capillaroscopic assessment, clinical read outs (modified Rodnan skin score [mRSS], lung function and echocardiography) and disease activity score (DAS) were performed at 0, 3, 6 and 12 months. Results: There was a clinical significant change in skin score with a mean (SD) mRSS of 24.8 (5.95) at baseline and 10.2 (1.17) at month 12 (Mixed Model Analyses [MMA] p⬍0.001) and a significant decrease in DAS, with a median of 3.8 at baseline (range: 2.0–7.0) and 0.5 at month 12 (range: 0.0–2.0) ( MMA p⬍0.001). Indices of internal organ involvement remained stable. Semi-quantitatively scored NVC parameters remained stable showing no progression of the microvascular damage during the twelve months follow-up: mean score (SD) of capillary loss at baseline/12 months: 2.170 (0.408)/ 1.830 (0.408) (MMA p⫽0.341), mean score (SD) of giants at baseline/ 12 months: 0.670 (0.516)/ 1.17 (0.408) (MMA p⫽0.093), mean score of haemorrhages at baseline/12 months: 0.670 (0.516)/1.00 (0.000) (MMA p⫽0.529) and mean score of neoangiogenesis at baseline/12 months: 0.830 (0.408)/ 0.830 (0.753) (MMA p⫽0.383). Conclusion: This is the first open pilot study to show that two immunosuppressive treatment courses with rituximab may not only have potential efficacy for skin and stabilisation of internal organ involvement but also additional stabilisation of microangiopathic parameters in early dcSSc.

Methods: Eighty patients with definite SSc (American College of Rheumatology criteria), 46 patients with early SSc (2001 LeRoy’s criteria), 40 PRP patients, and 45 controls were included. The morphological abnormalities were evaluated through widefield nailfold capillaroscopy (NFC) (10–25x magnification), and videocapillaroscopy (200x magnification). The following parameters were analyzed in eight digits of the hands (excluding the thumb) by widefield NFC and videocapillaroscopy: number of capillaries/ mm, enlarged, giant and ramified capillaries, microhaemorrhages, and avascular score. A score based on a semiquantitative rating scale (score 0–3) was used for each videocapillaroscopy parameter. Patients with scleroderma pattern were distributed into three patterns: early, active and late. Fingertip blood flow (FBF) was measured using laser Doppler imaging (LDI) at baseline and during 30 minutes after cold stimulus. Serum endothelin-1 (ET-1), von Willebrand factor (vWF) and transforming growth factor beta-1 (TGF-␤1) were measured by ELISA. Results: Patients with early SSc showed significantly higher number of capillaries/mm, and lower enlarged, giant capillaries and avascular score compared to patients with definite SSc in widefield NFC and videocapillaroscopy (p⬍0.001). Besides the number/score of enlarged capillaries, microhaemorrhages and avascular score were higher in patients with early SSc compared to patients with PRP and controls in both methods (p⬍0.001). Ninety-six percent of patients with definite SSc, and 76% of early SSc patients presented the scleroderma pattern using videocapillaroscopy. The early pattern was most frequently found among early SSc patients (37% in early SSc vs 17% in definite SSc, p⬍0.01). The active pattern was found in 63% of early SSc and in 60% of definite SSc. Interestingly, the late pattern was not observed in patients with early SSc, but was observed in 23% of definite SSc patients. FBF values before and after cold stimulus were significantly higher in controls compared to PRP, early SSc and definite SSc (p⬍0.001), with no difference between early and definite SSc. Serum levels of ET-1, vWF activity and TGF-␤1 were similar between early and definite SSc. Conclusion: Early SSc patients showed functional changes and serum vascular markers levels similar to patients with established disease. Nonetheless, the morphological changes were less severe in early SSc, suggesting a progression of the peripheral microangiopathy along different phases of the disease. These findings may have important implications in the management of patients with early disease.

ing table. NC abnormalities were associated with a significantly lower median thermographic DDD indicating more severe peripheral microvascular function (p⫽0.017). There was no difference between thermographic findings of patients with non-specific and SSc-pattern NC abnormalities (possibly due to small patient numbers with SSc-pattern abnormalities). NC abnormalities were present in all patients with SSc and the majority (54.5%) had SScpattern changes. Patients with primary RP were significantly less likely to have NC abnormalities compared with SSc (p⬍0.0001). Non-specific NC abnormalities were identified in 15/46 (32.6) patients with primary RP but SSc-pattern abnormalities were not identified in any patients with primary RP. NC appearances were generally normal in FMS with only a small proportion of patients (3/18, 16.7%) exhibiting non-specific NC abnormalities.

Tuesday, October 29

Median average DDD across all fingers, ° C, median (IQR) p value vs. normal NC appearance (Mann Whitney U) Proportion of patients with a mean DDD ⬍ ⫺1, n (%) Clinical diagnosis, n Primary RP SSc FMS Other rheumatic Diseases

Normal NC appearance (n ⫽ 83)

Nonspecific NC changes (n ⫽ 48)

SSc-pattern NC changes (n ⫽ 10)

⫹0.15 (4.16)

⫺1.98 (4.21)

⫺2.09 (3.71)

0.017

0.099

35 (42.2)

25 (52.1)

7 (70)

31 0 15 37

15 5 3 25

0 6 0 4

Conclusion: The presence of NC abnormalities is associated with thermographic evidence of more profound digital microvascular dysfunction in patients under investigation for symptoms of RP. The presence of NC abnormalities differs between primary RP and SSc. Despite a high reported prevalence of RP symptoms in FMS, the majority of patients have normal NC and thermographic findings. Disclosure: B. Vasta, None; M. Scolnik, None; D. Hart, None; J. A. Shipley, None; S. Brown, None; E. Korendowych, None; N. J. McHugh, None; J. D. Pauling, None.

statistical difference in the baseline characteristics between the acupressure groups vs. the control group. At the end of study, there were no statistical differences between the acupressure vs. education groups. However, there were trends in the patient reported severity of RP favoring acupressure groups (Table). In addition there were no significant differences in EndoPAT measurements or serum markers of vasculopathy. Sensitivity analysis using the completers showed similar results. Table: Change from week 1 compared to week 8* Variable

Group AⴙB

Control

P value

No of attacks Mean (SD) Pain Mean (SD) Tingling Mean (SD) Numbness Mean (SD) Average duration of attacks Mean (SD) RCS average for difficulty Mean (SD) RCS average for pain Mean (SD) Endopat Mean (SD) Patient VAS Mean (SD) MD VAS Mean (SD)

N: 12 ⫺6.3 (8.0) N: 12 ⫺13.6 (30.1) N: 12 ⫺5.3 (13.9) N: 12 ⫺10.5 (37.0) N: 12 ⫺9.2 (18.7) N: 12 ⫺2.0 (2.4) N: 12 ⫺1.4 (2.0) N: 14 ⫺0.1 (0.4) N: 16 ⫺1.6 (2.2) N: 16 ⫺2.1 (2.0)

N: 6 ⫺7.2 (12.8) N: 6 9.6 (27.6) N: 6 ⫺1.2 (7.5) N: 6 1.1 (22.1) N: 6 0.8 (11.2) N: 5 ⫺0.6 (2.9) N: 5 ⫺0.4 (3.0) N: 7 ⫺0.2 (0.8) N: 7 ⫺0.7 (2.0) N: 7 ⫺1.9 (1.3)

1.0 0.12 0.64 0.89 0.08 0.72 0.63 0.55 0.39 1.0

*Using LOCF.

Conclusion: Our pilot RCT showed that acupressure groups showed trends in improvement in symptoms associated with RP. However, there were no differences in the endothelial function and serum markers of vasculopathy. The parameters used to evaluate patients with RP have marked variability and supports the need for a composite measure to be developed for RP trials.

2597 Acupressure For The Treatment Of Raynaud’s Phenomenon: A Pilot Randomized Controlled Trial. Heather Gladue1, Richard E. Harris1, Veronica Berrocal1, Pei-Suen Tsou2, Gautam Edhayan3, Ray Ohara3 and Dinesh Khanna4. 1University of Michigan, Ann Arbor, MI, 2Division of Rheumatology, University of Michigan Medical Center, Ann Arbor, MI, 3University of Michigan Medical School, Ann Arbor, MI, 4University of Michigan Medical Center, Ann Arbor, MI. Background/Purpose: Raynaud’s phenomenon (RP) affects approximately 10% of the US population. The high cost, lack of efficacy, and side effects of conventional medical therapies necessitates the need for complementary or alternative options. Objective: A pilot, randomized control trial (RCT) of acupressure vs. patient education for the treatment of RP. Methods: A pilot single-center RCT of vasodilation acupressure, relaxation acupressure vs. RP education obtained from the Raynaud’s Association (control). Patients with either primary (N ⫽ 15) or secondary (N ⫽ 8) RP were randomized from January through April by block randomization to the 3 groups for an 8 week period. Patients randomized to acupressure were instructed on how to self-perform at home by a single investigator and a DVD was provided with instructions. The primary endpoint was a decrease in the severity, frequency and duration of RP. All patients kept a daily Raynaud’s diary, (recording the number and duration of attacks, pain, tingling and numbness on a 0–100 scale), and daily Raynaud’s condition score. At baseline and 8 weeks, EndoPAT was performed to determine endothelial function, and serum was collected for biomarker analysis (VEGF, tPA, sE-Selectin, BFGF, VCAM-1, ICAM). Data analysis was conducted using the last observation carried forward and paired statistical analyses were used to assess difference. Results: 23 patients were randomized and 9 discontinued prematurely (5 patients withdrew due to time restraints, 2 each for unrelated medical problems and lost to follow-up). Since there was no statistical difference between acupressure groups, they were combined and compared to the education group. 78% of patients were female, 96% were Caucasian, the mean age was 49.8 (SD⫽16) yrs; 5/16 patients in the acupressure group had secondary RP and 1/7 in the control group had secondary RP.. There was no

Disclosure: H. Gladue, None; R. E. Harris, None; V. Berrocal, None; P. S. Tsou, None; G. Edhayan, None; R. Ohara, None; D. Khanna, BMS, DIGNA, Roche, Actelion, Gilead, Merck, United Therapeutics, 5, National Institutes of Health, Scleroderma Foundation, PHA, Actelion, 2.

2598 Timing and Outcome Of Transition From Primary To Secondary Raynaud’s Phenomenon: A Capillaroscopic Based Study. Alberto Sulli1, Giorgia Ferrari1, Elena Bernero1, Carmen Pizzorni1, Vanessa Smith2, Barbara Ruaro1 and Maurizio Cutolo1. 1Research Laboratory and Academic Unit of Clinical Rheumatology, Department of Internal Medicine, University of Genova, Genoa, Italy, 2Department of Rheumatology, Ghent University Hospital, Ghent, Belgium. Background/Purpose: Raynaud’s phenomenon (RP) is classified as primary (PRP) or secondary (SRP) depending on its association with an underlying disease (1,2). PRP can evolve to SRP during followup (1,3). This is a prospective study to investigate timing and outcome of transition from PRP to SRP in a large cohort of patients, as well as to assess the progression of the nailfold scleroderma-patterns of microangiopathy in patients with isolated RP. Methods: 2,853 patients with RP were investigated. Patients achieving a diagnosis of PRP at first visit (normal nailfold videocapillaroscopy (NVC), normal or negative laboratory findings including autoantibodies, absence of other clinical findings and exclusion of an underlying disease) were followed to monitor the appearance of specific morphological alterations at NVC, autoantibodies positivity, or clinical symptoms for a mean period of 52⫾40SD months (1,3). Nailfold microangiopathy was detected by NVC, and capillary abnormalities were classified according to different patterns of microangiopathy (4–5). Results: At first visit, 797 (28%) patients out of 2,853 showed a NVC scleroderma-pattern, and got the diagnosis of SRP. Among the remaining 2,056 patients showing a normal capillaroscopy at baseline, 1,677 patients were lost during follow-up or not prospectively evaluable, whereas 379 patients were followed for a mean time of 52⫾40SD months. Among these last, 180 patients (47%) were positive for ANA, and 199 (53%) were ANA

S1108

negative (PRP patients). During the follow-up, the NVC scleroderma-pattern was diagnosed in 62 patients out of 379 (16%). In particular, it was found in 48 (27%) of the ANA-positive patients (27 “early”, 4 “active”, 9 “late”, 8 “like” NVC scleroderma-pattern at the end of the follow-up): these patients achieved the diagnosis of systemic sclerosis (46 patients), UCTD (1 patient) and Sjogren syndrome (1 patient). NVC was found positive also in 14 (7%) of the ANAnegative patients (transition to SRP) (13 “early”, and 1 “like” NVC sclerodermapattern at the end of the follow-up): these patients developed systemic sclerosis (9 patients) and UCTD (5 patients). The time of transition from normal/not specific capillary alterations to “early” scleroderma-pattern was 37⫾29 months. In particular, in ANA-positive patients NVC became positive within 43⫾36SD months, while in ANA-negative patients in 18⫾16 months. The 132 ANApositive patients with a not defined NVC pattern developed respectively systemic sclerosis (7 patients), UCTD (104 patients), MCTD (8 patients), autoimmune thyroiditis (6 patients), Sjogren syndrome (4 patients), antiphospholipid syndrome (1 patients), LES (1 patients). Conclusion: This large cohort study demonstrates that 7% of patients initially diagnosed as affected by PRP may transit to SRP in a mean time of 37⫾29 months. The transition from normal NVC pattern to sclerodermapattern was more frequent in ANA-positive patients (27%). References. 1. Cutolo M, et al. Arthritis Rheum 2007;56:2102–3. 2. LeRoy EC, et al. Clin Exp Rheumatol 1992;10:485–8. 3. Hirschl M, et al. Arthritis Rheum. 2006;54:1974– 81. 4. Cutolo M, et al. Rheumatology 2004; 43:719–26.4. 5. Smith V, et al. Ann Rheum Dis 2010; 69:1092–6. Disclosure: A. Sulli, None; G. Ferrari, None; E. Bernero, None; C. Pizzorni, None; V. Smith, Actelion Pharmaceuticals US, 5; B. Ruaro, None; M. Cutolo, Actelion Pharmaceuticals US, 2, BMS, 2, Mundipharm, 2, Sanofi-Aventis Pharmaceutical, 2.

2599

Background/Purpose: Mycophenolate reduces chronic allograft nephropathy and interstitial fibrosis by inhibiting TGF-␤, which is an important molecule in the pathogenesis of systemic sclerosis (SSc). The main side effects observed are gastrointestinal disturbance, myelosuppression, and increased risk of infection. This maybe a limitation of its use in SSc patients since gastrointestinal involvement is very common. The objective of this study is to evaluate the safety, in particular gastrointestinal adverse events, of mycophenolate in SSc. Secondarily we evaluated the effectiveness of mycophenolate in SSc skin and lung disease. Methods: A literature search of Medline, Embase, Cochrane Central Register of Controlled Trials, and CINAHL (inception - September 2012) was performed. Titles and abstracts were screened to identify studies that described the use of mycophenolate in SSc patients. Inclusion criteria included exposure to mycophenolate, and reporting of modified Rodnan skin score (MRSS), forced vital capacity (FVC), diffusing capacity of carbon monoxide (DLCO); or adverse events. The primary outcome was gastrointestinal events occurring after the initiation of mycophenolate. Secondary safety outcomes included myelosuppression, infection, malignancy, and death occurring after the initiation of mycophenolate. Results: 616 citations were identified and 20 were included in the analysis. 477 patients have been exposed to mycophenolate. The mean disease duration ranged between 0.8–14.1 years. There were 89 non-lethal adverse events, of which 43 (48%) were gastrointestinal and 46 (52%) were non-gastrointestinal adverse events. The most commonly reported gastrointestinal events included diarrhea (n⫽18 (20%)), nausea (n⫽12 (13%)), and abdominal pain (n⫽3 (3%)). The most commonly reported non-gastrointestinal adverse events were infections (n⫽33 (37%), and 6 cytopenias (n⫽6, (7%)). The reported rate of discontinuation ranged between 8–40%. Seven observational studies report mycophenolate is effective improvement or stabilization in FVC, and 5 observational studies report stabilization or improvement in MRSS. Conclusion: Observational studies report mycophenolate is effective in improving or stabilizing interstitial lung disease, and may be effective for skin involvement. However, gastrointestinal adverse events are common in SSc patients. Disclosure: M. Omair, None; A. Alahmadi, None; S. R. Johnson, None.

Shear Wave Elastography: A Novel Quantitative Approach For Evaluating Scleroderma Skin. Ingeborg Sacksen, Mark H. Wener, P. Scott Pollock and Manjiri K. Dighe. University of Washington, Seattle, WA. Background/Purpose: Systemic sclerosis (SSc) is a multi-system disease with both visceral and cutaneous fibrosis. Dermal elasticity is reduced and stiffness increased due to excessive dermal and subcutaneous deposition of collagen leading to increased skin thickness as well as hardening and tightening of the skin. Shear wave elastography (SWE) is an emerging, operator-independent technique, which can obtain quantitative stiffness values. We describe a novel quantitative approach for evaluating the skin in scleroderma using shear wave elastography Methods: We evaluated the skin of thirteen patients with diffuse systemic sclerosis and ten healthy controls using an Aixplorer shear wave ultrasound instrument (SupersonicImagine, Bothell, WA). Each patient underwent a SWE exam of the right arm at multiple sites (3rd finger, extensor forearm, upper arm), sternum, and face. The images consist of a spectrum of translucent colored bands superimposed on B-mode ultrasonographic images indicative of soft and highly elastic tissue to hard and minimally elastic tissue. The mean stiffness values of skin (measured in kPa units) were obtained at each site by averaging 3 measurements made per image obtained. Results: The skin of seven men and five women with diffuse systemic sclerosis as well as 10 healthy controls was evaluated. The mean and 1SD SWE stiffness measures were (in kPa): finger –SSc 98.3 (SD-56) vs control – 32.1, (SD- 11.8); forearm – SSc 89.6 (SD- 58.9) vs control – 18.3 (SD-6.9), upper arm – SSc 34.5 (SD-19.2) vs control 17.7(SD8.6), face: SSc 19.3 (SD-17.7) vs control 8.4(SD-2.7), chest –SSc 23.6 (SD-17.1) vs control 10.9(SD-6.0). Maximum (out of range high) values of 300kPA were obtained for 1 patient at the finger and two patients at the forearm. Two-tailed t –tests were performed between the means of patients and controls at each site. Statistical significance (p⬍0.01) was achieved at the finger, forearm, all sites combined, and (p⬍0.05) at the upper arm. Statistical significance was not reached for the chest (p⫽0.056) and the face (p⫽0.08). Z scores (based on the mean and SD of each site in controls) were calculated for patient measurements at each individual site and combined sites compared to controls. Z scores at all sites were above the mean for controls (see figure 1).

Conclusion: Shear wave elastography is a valid and feasible quantitative method for assessing abnormal elasticity of the skin in scleroderma patients. It adds a new dimension to the assessment of skin involvement in SSc, and is an objective, non-invasive tool for potential use in clinical trials. Studies comparing its performance to stage of disease, skin thickness, and mRss are ongoing. Disclosure: I. Sacksen, None; M. H. Wener, None; P. S. Pollock, None; M. K. Dighe, None.

2601 Modified Rodnan Ultrasound Skin Score: An Ultrasound Skin Scoring Approach In Systemic Sclerosis. Ingeborg Sacksen, P. Scott Pollock and Mark H. Wener. University of Washington, Seattle, WA. Background/Purpose: Systemic sclerosis (SSc) is a multi-organ disease characterized by thickening, hardening and tightening of the skin. Skin

S1109

Tuesday, October 29

Safety and Effectiveness Of Mycophenolate In Systemic Sclerosis: A Systemic Review. Mohammed Omair1, Abdulaziz Alahmadi2 and Sindhu R. Johnson3. 1Mount Sinai Hospital/University of Toronto, Toronto, ON, 2Toronto Scleroderma Research Program, Division of Rheumatology, Mount Sinai Hospital, Toronto Western Hospital, University Health Network, University of Toronto, Toronto, ON, 3Toronto Western Hospital, Toronto General Hospital, Mount Sinai Hospital, and University of Toronto, Toronto, ON.

2600

Tuesday, October 29

thickening is described by physical exam and measured semi-quantitatively as the modified Rodnan skin score (mRSS). Ultrasound (US) has been investigated to measure skin thickening, however, approaches to convert and relate US findings to mRSS’s have not been adopted, in part because of the variability of normal skin thickness at different sites (as well as by age and gender). We present preliminary data reproducing a 17-point mRSS in scleroderma patients and controls by using mid frequency ultrasound (15mHz) and correlating it to physical exam findings. We propose development of an ultrasound adaptation of the mRSS: the Modified Rodnan Ultrasound Skin Score (mRUSS), which is easy to use, quantifiable, and allows ultrasound-based categorization of the skin in scleroderma patients. Methods: We examined the skin of 16 patients (8 women) with diffuse cutaneous SSc and 10 healthy controls using a Sonosite M-Turbo 15mHz ultrasound machine operated by a single examiner. Each patient underwent a 17-point mRSS physical exam, and 12 patients had a second mRSS performed by an experienced rheumatologist. All patients then underwent a 17-point, comprehensive ultrasound skin exam at reproducible, predefined sites corresponding to the anatomic sites examined in the mRSS. The sites were: single sites on the face, chest wall, and abdomen, and bilateral 3rd finger, dorsum of the hand, extensor forearm, upper arm, thighs, lower legs, and the dorsal surface of the feet. Two dermal thickness measurements were taken from each image and averaged. The mean and standard deviation (SD) was calculated for controls at each site, and the z-score for each site calculated as [(skin thickness)-(mean control thickness at that site)]/(SD of control at that site)]. Results: Various cut-points were tested for translating the z-score to mRUSS’s assigned to each anatomic site, and the best concordance between the overall mRSS and the overall mRUSS were obtained with the following criteria: z-score ⬍1⫽ mRUSS 0; z-score 1–3⫽mRUSS 1; z-score 3–5 ⫽mRUSS 2; z-score ⬎5 ⫽mRUSS 3. Total mRUSS’s were then correlated to the mRSS (physical exam). The correlation between observer 1 mRSS compared to US score (assuming this to be the gold standard) was.90 and.73 for observer 2 (less experienced provider) (see figure 1). The r-value comparing observer 1 to observer 2 was 0.97.

severity as an increase in this score is associated with involvement of internal organs and increased mortality. Recent studies have found that high levels of the profibrotic type-2 cytokine IL-13 are produced by peripheral blood effector CD8⫹ T cells from SSc patients and that IL-13-producing circulating CD8⫹ T cells express skin-homing receptors and induce a profibrotic phenotype in normal dermal fibroblasts. Although one study describes a reduction of the circulating CD8⫹ T cell subset in dcSSc, the clinical implications of this finding are unknown. Methods: The frequencies of circulating CD4, CD8, CD20 and CD14 cells were determined by multi-parameter fluorescence-activated cell sorting (FACS) in peripheral blood mononuclear cells isolated from newly diagnosed, untreated dcSSc patients. Samples from patients with early onset, untreated seropositive rheumatoid arthritis were tested as controls. Clinical parameters recorded at the time of sample collection included: mRss, global assessment of activity, global assessment of severity, forced vital capacity and DLCO. Results: Percentages and absolute counts of cell subsets isolated from dcSSc patients (n⫽10) with a mean age of 58⫾8.24 years and a mean disease duration of 2.7⫾1.8 years were compared with age matched early onset RA patients (n⫽10). The frequencies of the studied cell subpopulations was similar between RA and dcSSc except for the absolute number and percentage of the CD8⫹ fraction, which was lower among dcSSc (p⫽.007 and.03, respectively). Lower frequencies of CD8⫹ T cells were associated with higher mRss and higher global assessments of activity (p⫽.0037 and.0052, respectively). For an increase of 1 unit in the mRSS score (range 0–51) there was a decrease of 0.79 in absolute CD8⫹ T cells and for an increase of 1 unit in the global assessment of activity (range 0–10), there was a decrease of 0.75 in absolute CD8⫹ T cells. Conclusion: Our findings suggest that early dcSSc have lower frequencies of peripheral blood circulating CD8⫹ T cells and that those are inversely associated with skin scores and disease activity. These findings are particularly relevant in light of recent evidence suggesting that skin homing – relocation IL-13-producing CD8⫹ T cells are directly involved in modulating dermal fibrosis in SSc, and emphasizes the need to prospectively assess the utility of reduced CD8⫹ T cell frequencies as a biomarker of dcSSc severity. Disclosure: M. Hudson, None; M. Lora, None; C. Di Ioia, None; S. Tatibouet, None; S. Bernatsky, None; I. Colmegna, None.

2603 Are There Differences In Limited Systemic Sclerosis According To Extension Of Skin Involvement?. Marina Scolnik1, Eliana Lancioni2, Luis J. Catoggio1, Mirtha Sabelli2, Zaida Bedran2, Carla Saucedo2, Josefina Marin2 and Enrique Soriano1. 1Rheumatology Section, Hospital Italiano de Buenos Aires, Buenos Aires, Argentina, 2Rheumatology Unit, Internal Medical Services, Hospital Italiano de Buenos Aires, Buenos Aires, Argentina.

Conclusion: Use of mid frequency US (15-mhz) is a readily available, reproducible, objective method for evaluating skin thickness in scleroderma. The newly proposed mRUSS scoring system correlates well with the mRSS based on physical exam findings. These findings will require verification with more patients and by independent examiners, but they have promise for use of precise ultrasound measurements to generate a mRSS equivalent. Disclosure: I. Sacksen, None; P. S. Pollock, None; M. H. Wener, None.

2602 Reduced Frequencies Of Circulating CD8 T Cells In Early Diffuse Cutaneous Systemic Sclerosis Is Associated With Worse Skin Scores. Marie Hudson1, Maximilien Lora2, Christopher Di Ioia1, Solene Tatibouet1, Sasha Bernatsky3 and Ines Colmegna4. 1McGill University, Montreal, QC, 2 McGill University Health Centre, Montreál, QC, 3McGill University Health Center, Montreal, QC, 4McGill University Health Centre, Montreal, QC. Background/Purpose: Diffuse cutaneous systemic sclerosis (dcSSc) is associated with significant morbidity and mortality. Measurement of skin thickness (modified Rodnan skin score (mRss)) is a surrogate of dcSSc

Background/Purpose: There is consensus in classifying Systemic Sclerosis (SSc) according to extension of skin involvement as limited and diffuse, using the elbows and knees as “limits” to distinguish between them. Decades ago, Barnett classified SSc as Type 1(only sclerodactyly), Type 2 (acrosclerosis - distal but may reach up to elbows and/or knees plus face) and Type 3 (diffuse skin involvement). Patients with Type 2 had an intermediate degree of organ involvement compared to Type 1 (less) and Type 3 (more). This issue has not been recently addressed. We examined the characteristics of our patients with limited disease to see if we could find differences between Barnett Type 1 and Type 2 subsets. Methods: electronic medical records of patients registered between years 2000–2011 with the problem: scleroderma, SSc or CREST and those with anti Scl-70, anticentromere or anti nucleolar antibodies in laboratory database were reviewed. Cases fulfilling ACR 1980 criteria were included and were classified as diffuse or limited according to LeRoy‘s criteria with limited being separated into sclerodactyly (only fingers) and acrosclerosis (fingers and up to elbows and/or knees) (Barnett⬘s Types 1 and 2). Results: 234 SSc patients (216 females) fulfilled criteria. Female/male ratio was 12:1; 24% had diffuse SSc and 76% limited (64% sclerodactyly and 12% acrosclerosis). Total follow up was 688 patients-years. Over half (55.1%) are still under our care and 17 died during this period. Ten year survival rate was 80% for limited and 70% for diffuse variants respectively (HR: 0.88 95% CI: 0.7–1.1). Table 1 shows clinical and serological profile of this cohort.

S1110

Anti Scl-70 was present in 16%, anticentromere in 53% and nucleolar ANA in 7% of overall patients. Within the limited group, several characteristics in the acrosclerosis (Type 2) group were more similar to the diffuse than the Type 1(sclerodactyly) patients. Duration of Raynaud was shorter, and they had significantly more anti Scl-70 and less anti centromere antibodies than those with Type 1. In particular, interstitial lung disease (ILD) was significantly more prevalent in Type 2 group, and similar to Type 3. Other characteristics did not reach statistical differences. Table 1. Clinical and serological profile of this cohort Type of systemic sclerosis Females, n (%) Age at diagnosis, years mean (SD) Duration of raynaud before diagnosis, years median (SD) Anti Scl-70, % Anticentromere, % Nucleolar ANA, % Other autoantibodies, % GI involvement, % Interstitial lung disease, % Pulmonary hypertension with ILD,% Pulmonary hypertension (without ILD), % Echocardiographic abnormalities (other than PH), % Digital ulcers, % Renal Crisis, n Pregnancy after diagnosis, n Patient-year, years (SD) Currently followed, n (%) Died under our care, n

Limited (n ⴝ 178) Sclerodactyly Acrosclerosis (n ⴝ 149) (n ⴝ 29)

p (sclero vs acro)

142 (95.3) 59.8 (15.2) 9.2 (3.4)

25 (86.2) 54.9 (16.2) 5.5 (1)

0.06 0.14 0.018

6.4 82.3 5.7 26.1 65.9 17.1 5.5 9.1

39.3 17.9 3.6 23.5 81.3 50 0 4.8

⬍0.001 ⬍0.001 0.54 0.54 0.22 ⬍0.001 0.34 0.45

5.7

4.5

0.65

26.9 0 3 462 (3.1) 88 (59.5) 11

34.5 0 1 81.2 (3.1) 15 (51.7) 2

0.44

Diffuse (n ⴝ 56) 49 (87.5) 53 (18.3) 1.9 (0.8) 43.9 4.8 20.9 25.9 64.3 65.3 15.4 2.6 10 32.1 1 3 144.8 (3.6) 26 (46.4) 4

p (limited vs diffuse) 0.13 0.038 0.008 ⬍0.001 ⬍0.001 0.001 0.76 0.45 ⬍0.001 0.03 0.39 0.26 0.86

Scl-70: anti-topoisomerase I; ANA: antinuclear antibodies; GI: gastrointestinal; ILD: interstitial lung disease; PH: pulmonary arterial hypertension

Disclosure: M. Scolnik, None; E. Lancioni, None; L. J. Catoggio, None; M. Sabelli, None; Z. Bedran, None; C. Saucedo, None; J. Marin, None; E. Soriano, None.

2604 Longitudinal Assessment Of Scleroderma Skin By Optical Coherence Tomography: Preliminary Validation Of Sensitivity To Change Over-Time. Giuseppina Abignano1, Lesley-Anne Bissell1, Jason Britton2, Daniel Woods3, Maya H. Buch1, Dennis McGonagle1, Paul Emery1 and Francesco Del Galdo1. 1Leeds Institute of Rheumatic and Musculoskeletal Medicine, University of Leeds and NIHR Leeds Musculoskeletal Biomedical Research Unit, Leeds Teaching Hospitals NHS Trust, Leeds, United Kingdom, 2Medical Physics Department, Leeds Teaching Hospitals NHS Trust, Leeds, United Kingdom, 3Michelson Diagnostics Ltd, Kent, United Kingdom. Background/Purpose: Optical coherence tomography (OCT) has been shown to be a quantitative reliable tool to assess skin involvement in Systemic Sclerosis (SSc) (1). However the sensitivity to change over-time has not been evaluated. The present study aimed to compare skin assessment by OCT over-time in patients with SSc. Methods: We performed 24 OCT scans of dorsal forearms on 12 sites of analysis from 7 SSc patients (6 with diffuse, 1 with limited subset according to Le Roy et al; mean disease duration at baseline⫽ 8 ⫾ 2.3 years) at 0 and 24 months. Clinical skin involvement was assessed using the modified Rodnan skin score (mRss). Minimum and Maximum Optical Density (Min and Max OD) of the mean-A scans were calculated employing Matlab software as previously described (1). Comparison of the local mRss and Min and Max OD at the 2 time-points was performed by two-tailed paired t-test employing GraphPrism software. Results: Five sites of analysis with local mRss⫽0 did not change over 24 months. Accordingly, both Min and Max OD showed only an average ⫺1.81% and ⫺2.31% change, respectively (p⬎0.05) (1). On the contrary,

References: 1. Abignano et al. Ann Rheum Dis 2013. Disclosure: G. Abignano, None; L. A. Bissell, None; J. Britton, None; D. Woods, Michelson Diagnostics Ltd, 3; M. H. Buch, None; D. McGonagle, None; P. Emery, None; F. Del Galdo, None.

2605 Quality Of Life and Psychosocial Aspects In Juvenile Localized Scleroderma. Roberta Culpo, Marco Ricca, Fabio Vittadello, Giuseppina Sequi, Francesco Zulian and Giorgia Martini. University of Padua, Padua, Italy. Background/Purpose: Juvenile Localized Scleroderma (JLS) is a chronic, autoimmune disease, characterized by skin and subcutaneous tissues fibrosis, which can cause a poor quality of life and psychosocial and behavioural problems in affected children particularly when severe deformities such as face asymmetry, joint contractures, and growth disturbances of limbs develop. To date, quality of life and psychological aspects in JLS have been poorly investigated. Purpose: to evaluate quality of life and psychosocial aspects of patients with JLS as compared with healthy peers and identify specific disease characteristics possibly related to quality of life impairment and psychosocial problems. Methods: Two types of questionnaires (Pediatric Quality of Life Inventory 4.0™ Generic Core Scales and Child Behaviour Check List (CBCL) 6–18/Youth Self Report (YSR) 11–18) were administered to 40 consecutive patients with JLS aged 6 to 18 years and their parents. Patients’ demographic and clinical data were collected during medical examination and through the review of clinical records. Same questionnaires were administered to a control group of 44 healthy children and their parents. Results: In PedsQL™ (children forms) no difference was found between JLS group and control group. In PedsQL™ (parents forms) children with JLS showed poorer quality of life as compared to control group (76.8 vs 84.8, p⫽0.017), especially in emotional area (64.5 vs 79, p⫽0.004). In CBCL/6–18 mean scores were lower in activity scale (35.2 vs 41.1, p⫽0,006) and higher in internalizing problems scale (58 vs 53.2, p⫽0.026) and depression scale (59.9 vs 55.9, p⫽0.038) in JLS group compared to control group. In YSR/11–18 mean scores were lower in social competence scale (44.2 vs 49.7, p⫽0.007) and in total competence scale (40.9 vs 42.4, p⫽0.028) and higher in internalizing problems scale (54.7 vs 50.9, p⫽0.031) in JLS group compared to healthy controls. Disease relapses, longer delay in correct diagnosis, onset of disease in adolescence and shorter disease duration significantly correlated with lower quality of life and psychosocial and behavioural problems. Conclusion: Our study shows that quality of life is poorer in children with JLS compared to healthy peers. Emotional area and social activities are the most affected ambits and patients show also depressive and internalizing problems. Among patients with JLS, a greater need for psychological support is mainly related to disease relapses, longer diagnostic delay, shorter disease duration and onset in adolescence or pre-adolescence ages. Disease severity in terms of lesion extension or deformities and therapy related issues do not seem related to impairment in the investigated areas. Disclosure: R. Culpo, None; M. Ricca, None; F. Vittadello, None; G. Sequi, None; F. Zulian, None; G. Martini, None.

2606

S1111

SEE ABSTRACT #2917

Tuesday, October 29

Conclusion: These results appear to confirm that extension of skin involvement within limited SSc may identify two different subsets with clinical and serologic characteristics. Indeed, Type 2 as defined by Barnett appears to have intermediate organ involvement, and serology may be more similar to the diffuse type.

in 5 sites of analysis mRss improved by 2 points (two sites with local score “2” improved to “0”, three sites with local score of “3” improved to “1”). In these sites Min OD showed an increase of 29% (range⫽23–37%; p⫽0.0005) and Max OD of 29.8% (range⫽18–41%; p⫽0.0027). Furthermore, both Min and Max OD showed a trend forward a decrease (⫺3.54%, ⫺5.41% respectively) at the 2 sites of analysis with worsening mRSS (one point increase) but the low sample size did not allow to perform a statistical evaluation. Conclusion: Although preliminary for the low number of observations, this study provides the first evidence suggesting that OCT of the skin is sensitive to change over time and it changes consistently with mRss. Studies including a larger number of patients and sites of analysis with different grades of skin involvement and improvement/deterioriation of clinical score are needed to reach a definitive validation.

2607

Tuesday, October 29

Evaluation Of The Construct Validity Of The Patient-Reported Outcomes Measurement Information System (PROMIS®) Gastrointestinal (GI) Symptoms Measures In Systemic Sclerosis (SSc). Dinesh Khanna1, Puja Khanna1, Brennan Spiegel2, Lin Chang3, Gil Y. Melmed4, Roger Bolus5 and Ron Hays2. 1University of Michigan Medical Center, Ann Arbor, MI, 2 University of California, Los Angeles, Los Angeles, CA, 3David Geffen School of Medicine at University of California, Los Angeles, Los Angeles, CA, 4Cedar-Sinai Medical Center, Los Angeles, CA, 5Research Solutions Group, Encinitas, CA. Background/Purpose: As part of the National Institutes of Health PROMIS¨ roadmap initiative, we developed GI Symptoms measures that assess 8 domains: Gastroesophageal reflux (13 items), disrupted swallowing (7 items), diarrhea (5 items), bowel incontinence/soilage (4 items), nausea and vomiting (4 items), constipation (9 items), belly pain (6 items), and gas/bloat/ flatulence (12 items). All scales are calibrated using a two-parameter IRT graded response model and scored on a T-score metric with a mean of 50 and SD of 10 in the U.S. general population. This paper evaluates the construct validity of the GI measures in patients with SSc. Methods: 165 patients with SSc were administered the PROMIS GI Symptoms measures and UCLA SCTC GIT 2.0 instrument. GIT 2.0 has 5 symptom scales: reflux, distention/ bloating, diarrhea, constipation, and fecal incontinence. Product-moment correlations of the PROMIS GI measures with the GIT 2.0 symptoms scales were used to evaluate construct validity. In a subset of patients (N⫽37), both instruments were administered at 2 time points. F-statistics was calculated from one-way ANOVAs to assess responsiveness to change Results: Patients with SSc GI involvement had scale scores 0.2–0.7 SD worse than US population. Hypothesized correlations were larger than other scales and in the right direction (Table). F-statistics were greater for 6 of 8 PROMIS scales (range 0.45 for belly pain to 3.21 for reflux scale) vs. GIT 2.0 except for diarrhea scale (0.67 vs. 0.98 for GIT 2.0) and constipation scale (1.37 vs. 1.79 for GIT 2.0). Table. Product-moment correlations between PROMIS GI Symptoms scales and UCLA SCTC GIT scales

Reflux Disrupted swallowing Nausea and vomiting Belly pain Gas/bloat/flatulence Diarrhea Constipation Fecal incontinence

Reflux

Distention/ bloating

Diarrhea

Constipation

Fecal incontinence

0.77 0.61 0.66 0.45 0.46 0.25 0.37 0.12

0.44 0.39 0.44 0.49 0.73 0.25 0.32 0.11

0.13 0.16 0.20 0.23 0.30 0.65 0.05 0.43

0.25 0.21 0.22 0.34 0.29 0.02 0.76 ⫺0.18

⫺0.03 0.13 0.18 0.04 0.10 0.54 ⫺0.01 0.87

Methods: This is a retrospective study of well-characterised cohort of 1920 SSc patients from 2008 to 2013 with data collected through clinical database and patient records including all SSc patients diagnosed with CD by duodenal biopsy (n⫽12), and 96 of 236 SSc patients with GI symptoms (chronic diarrhoea, bloating, constipation, discomfort, dysphagia, heartburn and faecal incontinence) and negative coeliac antibodies. 4 patients in the SSc-CD group had confirmed positive coeliac antibodies. Key demographic and clinical features including small bowel symptoms (chronic diarrhoea, bloating, constipation and discomfort) were examined. Results: The prevalence of CD with positive coeliac antibodies in our SSc cohort with GI symptoms was 4/236 (1.69%). 92.6% were female, with median (interquartile range, IQR) of age at SSc onset of 45.0 (33.0 – 55.7) years, and 49.3 (41.0 – 55.7) years at CD diagnosis. 18.5% patients had diffuse subset and 66.7% of the patients were Caucasians. 33 (30.6%) had an overlap with another systemic autoimmune disease (17 had polymyositisdermatomyositis, 8 rheumatoid arthritis, 5 Sjo¨gren’s syndrome, 3 others). A higher frequency of psoriasis was identified in 25% of SSc-CD patients compared with 2.1% (p⫽0.009) in negative-coeliac antibodies patients. SSc-CD patients had increased frequency of myopathy (33.3% vs 10.4%, p⫽0.04), and anti-Pm/Scl antibody (25.0% vs 3.1%, p⫽0.01). 88% of entire cohort had small bowel symptoms with no statistical differences between the two groups. Chronic diarrhoea was the most common symptom in 62.0% of patients. Upper endoscopy was performed in all SSc-CD patients and in 41.7% with negative-coeliac antibodies, with no statistical differences in frequency of esophagitis, gastritis, gastric antral vascular ectasia or macroscopic duodenitis. There were no differences in the prevalence of bacterial overgrowth or requirement for enteral/parenteral nutrition. Among all patients 44.4% had vitamin D deficiency, 40.7% iron deficiency and 7.4% osteoporosis. After gluten-free diet all SSc-CD patients had an improvement of small bowel symptoms. 50% of these patients achieved complete remission and the symptoms recurred in the remaining half with median (IQR) 5.0 (2.5 – 8.25) years. There was significant improvement in chronic diarrhoea (75% to 41.7%, p⫽0.04), abdominal distension or bloating (50% to 16.7%, p⫽0.04), and in weight loss (33.3% to 0%, p⫽0.04). Conclusion: The clinical presentation of CD may be indistinguishable to SSc-GI disease and our study indicates that CD in this subgroup of SSc patients with GI involvement is no more common than the general population. The positive response to gluten-free diet suggests that screening for CD in selected SSc patients may be helpful. Disclosure: A. Guilleń-Del Castillo, None; V. Sobanski, None; J. Harvey, None; C. Denton, None; V. H. Ong, None.

2609

GIT 2.0 Reflux scale asks about reflux, dysphagia to solid foods, and nausea/ vomiting.

Conclusion: PROMIS GI Symptoms scales are significantly correlated with the hypothesized GIT 2.0 scales and 6 of 8 scales showed greater responsiveness to change than the GIT 2.0. Disclosure: D. Khanna, NIH, 2, Scleroderma Foundation, 2, Actelion Pharmaceuticals US, 5, Actelion Pharmaceuticals US, 8, Gilead, 5, United Therapeutics, 5, United Therapeutics, 8, Roche Pharmaceuticals, 5, BMS, 5, DIGNA, 5, Merck Pharmaceuticals, 5; P. Khanna, None; B. Spiegel, None; L. Chang, None; G. Y. Melmed, Janssen, given imagine, Abbvie, 5, Abbott, prometheus labs, 9, Pfizer Inc, 2; R. Bolus, NIH, 5; R. Hays, None.

2608 Coeliac Disease In Scleroderma - Clinical Features, Frequency and Impact Of Screening In Scleroderma. Alfredo Guilleń-Del Castillo1, Vincent Sobanski2, Jennifer Harvey2, Christopher Denton3 and Voon H. Ong4. 1 Hospital Universitari Vall d’Hebron, Barcelona, Spain, 2Royal Free Hospital, London, United Kingdom, 3Royal Free and University College Medical School, London, United Kingdom, 4The Royal Free and University College Medical School, London, United Kingdom. Background/Purpose: Recent studies suggest that coeliac disease (CD) affects 4–7% of Scleroderma (SSc) patients. This association however has not been well characterized in a large cohort of patients. We evaluate the key clinical features, nutritional complications, and clinical response after gluten free diet in SSc patients with CD (SSc-CD), compared with a group of SSc patients with gastrointestinal (GI) symptoms and negative coeliac antibodies.

Dutch Translation and Validation Of The University of California, Los Angeles scleroderma Clinical Trial Consortium Gastrointestinal Tract Instrument 2.0. Jessica Meijs, Daisy Pors, Theodora P.M. Vliet Vlieland, Tom W.J. Huizinga and Annemie J.M. Schuerwegh. Leiden University Medical Center, Leiden, Netherlands. Background/Purpose: Gastrointestinal tract (GIT) involvement occurs in approximately 90% of the patients with systemic sclerosis (SSc) and leads to a decrease in health-related quality of life. To identify and evaluate GIT involvement in patients with SSc, the University of California Scleroderma Clinical Trial Consortium Gastrointestinal Tract Instrument 2.0 (UCLA GIT 2.0) was developed. The UCLA GIT 2.0 consists of 34 items on 7-multi-item subscales. The total score averages 6 of 7 scales (excluding constipation) and ranges from 0 (no GIT problems) to 3 (most severe GIT problems) [1]. The aim of our study was to translate the UCLA GIT 2.0 from English into Dutch and validate the Dutch version. Methods: First, the UCLA-GIT 2.0 questionnaire was translated according to international guidelines [2]. Secondly, the questionnaire was fieldtested among 17 SSc patients. Then, in order to test internal consistency and validity, the final Dutch version was administered to SSc patients participating in a standardized, annual comprehensive medical assessment, compromising visits to health care professionals, laboratory and cardiopulmonary investigations, the SSc Health Assessment Questionnaire (SHAQ) and Short Form-36 (SF-36). The internal consistency was tested by computing Cronbach’s alpha. For convergent and construct validity, Pearson correlations were computed between the Dutch UCLA-GIT and the SF-36 and SHAQ. (rⱕ 0.29 is considered a small correlation, r 0.30 –0.49 a moderate, and r ⱖ 0.50 a strong correlation). To determine the reliability, the instrument was re-administered with an interval of two weeks to 27 of the patients, and the intraclass-correlation coefficient (ICC) was computed.

S1112

Results: Ninety-two patients were included. Patients were on average 53.8 (SD 15) years, mostly women (76%), Caucasian (82%) and 53% of the patients had limited cutaneous SSc. The median total UCLA GIT 2.0 score was 0.18 and a floor effect was seen (17% scored 0). Cronbach’s alpha was ⱖ0.73 for all scales, except diarrhea (alpha⫽0.42). The total GIT score was weakly or moderately associated with the SF-36 mental component summary scale (r⫽⫺0.247, p⬍0.05), the SF-36 physical component summary scale (r⫽⫺0.348, p⬍0.01) and the SHAQ score (r⫽0.287, p⬍0.01). Correlations between the GIT subscale scores and the SF-36 subscale scores were strong with respect to the corresponding emotional wellbeing and social functioning scale (r⫽⫺0.515, p⬍0.01). The test-retest reliability was acceptable (ICC: 0.788). Conclusion: The Dutch UCLA GIT 2.0 questionnaire showed good internal consistency, reliability and an acceptable construct validity according to comparisons with SF-36 and SHAQ. The Dutch translation will now be further validated with objective measures of GIT involvement. References: 1. Khanna D, et al. Reliability and validity of the University of California, Los Angeles Scleroderma Clinical Trial Consortium Gastrointestinal Tract Instrument. Arthritis Rheum 2009;61:1257–63. 2. Beaton, D.E., et al. Guidelines for the process of cross-cultural adaptation of self-report measures. Spine 2000;25:3186–91. Disclosure: J. Meijs, Actelion Pharmaceuticals, 2; D. Pors, None; T. P. M. Vliet Vlieland, None; T. W. J. Huizinga, None; A. J. M. Schuerwegh, Actelion Pharmaceuticals, 5.

secreted by infiltrating neutrophils, was detected in the bronchoalveolar lavage and in lung extracts. Inhibition of BAFF using BAFF-R-Ig or gene depletion (BAFF⫺/⫺ mice) resulted in a marked decrease of lung fibrosis, assessed by histology, and quantification of TGF-beta and collagen by qPCR and ELISA. Interestingly, levels of BAFF were significantly decreased in IL-1R1⫺/⫺, IL-17A⫺/⫺ and IL-17 RA⫺/⫺ mice, in which lung fibrosis was significantly reduced after BLM challenge, compared to WT mice. In vitro, recombinant BAFF induced a dramatic increase in IL-17 secretion by lung cells isolated after BLM challenge. Conclusion: These results confirm the implication of B lymphocytes and sheds light to a new pathogenic role of BAFF in fibrosis according to consistent results in the bleomycin model, SSc and IPF. Taken together, these results suggest the following scenario: i) BAFF is induced by IL-1␤ and IL-17, which are potent pro-fibrogenic cytokines; ii)BAFF in turn induces IL-17 secretion by Th17 cells in an amplification loop. This study adds to the rationale of evaluating the therapeutic interest of BAFF inhibition in systemic sclerosis, inflammatory arthritis-related lung fibrosis and idiopathic pulmonary fibrosis. Disclosure: A. Francois, None; P. Schneider, None; A. Davidson, None; E. Chatelus, None; J. Avouac, None; Y. Allanore, None; B. Villeret, None; A. Gombault, None; P. Gasse, None; S. Marchand Adam, None; B. Ryffel, None; S. Bahram, None; P. Georgel, None; J. Sibilia, None; I. Couillin, None; J. E. Gottenberg, None.

2611

WITHDRAWN

2612

A New Pathogenic Role Of BAFF As a Critical Mediator Of Skin and Lung Fibrosis In Experimental Bleomycin-Induced Pulmonary Fibrosis, Systemic Sclerosis and Idiopathic Pulmonary Fibrosis. Antoine Francois1, Pascal Schneider2, Anne Davidson3, Emmanuel Chatelus4, Je´roˆme Avouac5, Yannick Allanore5, Be´renge`re Villeret6, Aure´lie Gombault6, Pame´la Gasse7, Sylvain Marchand Adam8, Bernhard Ryffel9, Siamak Bahram10, Philippe Georgel11, Jean Sibilia12, Isabelle Couillin7 and Jacques-Eric Gottenberg4. 1 University of Strasbourg, Illkirch-Strasbourg, France, 2University of Lausanne, Lausanne, Switzerland, 3Feinstein Institute for Medical Research, Manhasset, NY, 4Strasbourg University Hospital, Strasbourg, France, 5Paris Descartes University, Rheumatology A department, Cochin Hospital, Paris, France, 6University of Orleans and National Center for Scientific Research, Orleáns, France, 7University of Orleans and National Center for Scientific Research, Orleans, France, 8National institute of the health and the medical research, Tours, France, 9University and CNRS, 3b rue de la Ferollerie, Orleans, France, 10Immunorhumatology Molećular, Strasbourg, France, 11Laboratoire d’ImmunoGeńe´tique Molećulaire Humaine, Strasbourg, France, 12CHU Hautepierre, Strasbourg, France.

Epidemiology Of Cancer In Systemic Sclerosis. Systematic Review and Meta-Analysis Of Cancer Incidence, Predictors and Mortality. Tatiana Nevskaya1, Shelly Chandran2, Adrienne M. Roos1, Christopher Pasarikovski1, Amie T. Kron1, Cathy Chau3 and Sindhu R. Johnson4. 1Toronto Scleroderma Research Program, Toronto Western Hospital, Mount Sinai Hospital, and University of Toronto, Toronto, ON, 2Toronto Scleroderma Research Program, Toronto Western Hospital, University of Toronto, Toronto, ON, 3Toronto Scleroderma Research Program, Toronto Western Hospital, Mount Sinai Hospital, University of Toronto, Toronto, ON, 4Toronto Western Hospital, Toronto General Hospital, Mount Sinai Hospital, and University of Toronto, Toronto, ON.

Background/Purpose: Interstitial pneumonitis and lung fibrosis are frequent systemic complications of inflammatory arthritides, including systemic sclerosis (SSc), rheumatoid arthritis, or primary Sjo¨gren’s syndrome. B lymphocytes are involved in the pathogenesis of such lung involvement. BAFF (B-cell activating factor of the TNF family) plays a crucial role in autoreactive B-cell activation and survival. We therefore investigated the pathogenic role of BAFF in fibrosis Methods: Levels of BAFF were assessed using ELISA in serum of 150 patients with SSc and 80 healthy controls and in bronchoalveolar lavage of 12 patients with idiopathic pulmonary fibrosis (IPF) and 7 controls. Cocultures of B lymphocytes, stimulated or not with BAFF, and skin fibroblasts from patients with SSc and controls were performed. Lung fibrosis induced by bleomycin (BLM) was compared in BAFF⫺/⫺ knock-out, BAFF-R-Ig treated mice and wild type (WT) mice. Lung BAFF expression was compared in BAFF IL-1R1⫺/⫺, IL-17A⫺/⫺, IL-17RA⫺/⫺ mice after BLM challenge. Levels of IL-17 secreted by lung cells isolated from BLM-treated mice were analyzed after stimulation with anti-CD3 and BAFF Results: Serum BAFF level was significantly increased in SSc patients compared to controls (median 1.5 vs. 0.5; p⬍ 0.0001). Patients with SSc and increased BAFF levels had a significantly higher incidence of pulmonary fibrosis assessed by lung CT scan (63% vs. 37%; p⬍ 0.005). A significant increase in BAFF bronchoalveoar levels was observed in patients with IPF, especially in those with clinical exacerbation, compared to controls. Coculture of B lymphocytes and skin fibroblasts induced collagen secretion, which was further enhanced after BAFF stimulation. In the BLM model of lung fibrosis, a marked increase of BAFF, mainly

Background/Purpose: To improve our understanding of the epidemiology of cancer in systemic sclerosis (SSc) by evaluating the incidence, prevalence, relative risk of overall and site-specific malignancies in comparison with the general population, and cancer-attributable mortality. Methods: MEDLINE, CINAHL, EMBASE and Cochrane Library (inception-May 2012) were searched. Estimates were combined using a random effects model. Consistency was evaluated using the I2statistic. Results: 4,876 citations were searched to identify 59 articles. The average incidence of malignancy in SSc was 14 cases/1000 person-years; the prevalence ranged between 4%-22%. Cancer was the leading cause of non-SSc related deaths with a mean of 38%. Overall SIR for all-site malignancy risk was 1.85 (95% CI 1.52, 2.25; I276%).There was a greater risk of lung (SIR 4.69, 95% CI 2.84, 7.75; I293%) and haematological (SIR 2.58, CI 95% 1.75, 3.81; I20%) malignancies, including nonHodgkin’s lymphoma (SIR 2.55, 95% CI 1.40, 4.67; I20%). SSc patients were at a higher risk of leukemia (SIR 2.79, 95% CI 1.22, 6.37; I2 0%), liver (SIR 4.75, 95%CI 3.09, 7.31; I20%), cervical (SIR 2.28, 95% CI 1.26, 4.09; I254%) and oropharyngeal (SIR 5.0, 95% CI 2.18, 11.47; I258%) cancers. Risk factors include a-RNAP I/III positivity, male sex, and late onset SSc. Smoking and longstanding interstitial lung disease (ILD) increase the risk of lung cancer; longstanding gastroesophageal reflux disease with Barrett’s esophagus and a positive family history of breast cancer, respectively, increase the risk of esophageal adenocarcinoma and breast cancer. Conclusion: SSc patients have a two-fold increase in malignancy, and greater risk of lung and haematological malignancies that contribute significantly to mortality. Vigilance should be considered in SSc patients with a-RNAP I/III antibodies, male sex, smokers, late disease onset, a positive family history of breast cancer, long duration of ILD, Barrett’s esophagus. Disclosure: T. Nevskaya, None; S. Chandran, None; A. M. Roos, None; C. Pasarikovski, None; A. T. Kron, None; C. Chau, None; S. R. Johnson, None.

S1113

Tuesday, October 29

2610

ACR/ARHP Poster Session C Vasculitis III Tuesday, October 29, 2013, 8:30

AM–4:00 PM

2613

Tuesday, October 29

NF-␬B Pathway Is Depleted In Phagocytes From Behçetⴕs Disease Patients Secondarily To Constitutive Phosphorylation Of The p65 Subunit. Sandro F. Perazzio1, Paulo Vitor Soeiro Pereira2, Alexandre W.S. Souza3, Antonio Condino-Neto2 and Luis Eduardo C. Andrade1. 1Escola Paulista de Medicina - Universidade Federal de Saõ Paulo, Sao Paulo, Brazil, 2 ICB IV - Universidade de Saõ Paulo, Saõ Paulo, Brazil, 3University Medical Center Groningen, Groningen, Netherlands. Background/Purpose: Increased neutrophil activation has been previously shown in Behçet’s disease (BD) patients and it is unclear whether neutrophil activation occurs constitutively or if it is secondary to a yet unknown stimulus or some serum or tissue soluble factor. NF-kB seems to modulate the immune response in BD and has been associated to apoptosisrelated factors leading to apoptosis resistance in T cell subsets. The present study investigated the NF-kB pathway and its activation induced by TLR. Methods: Neutrophils and peripheral blood mononuclear cells (PBMC) were obtained from patients with active BD (aBD; n⫽30), inactive BD (iBD; n⫽31), septic patients (SP; n⫽25), and healthy controls (HC; n⫽30). BD activity was established as Behçet’s Disease Current Activity Form simplified (BDCAFs) scoreⱖ2. The functional analysis of the NF-kB pathway was performed by: 1) determining CD62L shedding after stimulation with specific ligands for TLR-2, -3, -4, -5, -7, and with Streptococcus pneumoniae, Streptococcus sanguinis, and Candida albicans; 2) the intracellular expression of phosphorylated NF-kB-p65 before and after stimulation with PMA, TLR-3, TLR-7, plasma from HC, aBD or iBD; 3) the intracellular expression of STAT3 before and after stimulation with PMA, IL-10, or plasma from HC or aBD. Results: BD patients homogeneously presented very low CD62L shedding with all forms of stimulus, in contrast to the large dispersion observed in the other groups. The percent increase in CD62L shedding was more evident after activation by TLR3 in iBD (31⫾28%: p⫽0.022) and aBD (27⫾20%; p⫽0.029) than in SP (3.4⫾25%). In contrast, the activation by TLR7 was lower in iBD (27⫾23%; p⫽0.022) and aBD (32⫾27%; p⫽0.029) than in SP (74⫾39%). Neutrophils from aDB presented higher expression of phosphorylated NF-kB-p65 before stimuli (mean of intensity of fluorescence [MFI]⫽123.54⫾47.63 versus HC⫽69.43⫾39.29, p⫽0.050) and after PMA (MFI⫽194.10⫾97.18 versus HC⫽92.24⫾42.34, p⫽0.030) and TLR7 (MFI⫽135.76⫾45.92 versus HC⫽77.51⫾42.34, p⫽0.050). Monocytes from aDB also presented higher expression of phosphorylated NF-kB-p65 before stimuli (MFI⫽313.40⫾110.81 versus HC⫽135.98⫾87.61, p⫽0.018) and after PMA (MFI⫽372.80⫾145.01 versus HC⫽175.51⫾98.35, p⫽0.030) and TLR3 (MFI⫽320.40⫾117.39 versus HC⫽162.15⫾74.47, p⫽0.030). Phosphorylated NF-kB-p65 expression and CD62L shedding after human plasma and microbial stimuli, respectively, was equivalent for phagocytes from aBD, iBD, and HC. STAT3 expression in neutrophils and monocytes showed no difference among the aBD, iBD, and HC. There was no difference in medication use between aBD and iBD. Conclusion: We originally showed that the NF-kB pathway of phagocytes is constitutively activated in BD patients and that there is additional increment after TLR stimuli. The low CD62L shedding of BD phagocytes may be due to exhaustion of the NF-kB pathway. These findings may underlie the characteristic hyperactivity of neutrophils in BD, represented by activation of the final pathway for several stimuli, like cytokines, chemoattractants and microorganisms, potentially involved in the pathophysiology of this disease. Disclosure: S. F. Perazzio, None; P. V. S. Pereira, None; A. W. S. Souza, None; A. Condino-Neto, None; L. E. C. Andrade, None.

2614 Plasma Of Active Behçetⴕs Disease Increases Oxidative Metabolism Profile Of Normal and Patients Phagocytes. Sandro F. Perazzio1, Paulo Vitor Soeiro Pereira2, Alexandre W.S. Souza3, Antonio Condino-Neto2 and Luis Eduardo C. Andrade1. 1Escola Paulista de Medicina - Universidade Federal de Saõ Paulo, Sao Paulo, Brazil, 2ICB IV - Universidade de Saõ Paulo, Saõ Paulo, Brazil, 3University Medical Center Groningen, Groningen, Netherlands.

Background/Purpose: Behçet’s disease (BD) exhibits several features suggestive of neutrophil hiperactivity. It is unclear whether neutrophil activation occurs constitutively or if it is secondary to yet unknown stimuli. Previous studies suggested a possible role for infectious agents and for unknown plasma factors in BD pathogenesis. The present study investigated the oxidative burst in phagocytes of patients with active BD (aBD) and inactive BD (iBD) in the presence of diverse stimuli. Methods: Neutrophils and peripheral blood mononuclear cells (PBMC) were obtained from patients with aBD (n⫽30), iBD (n⫽31), septic patients (SP; n⫽25), and healthy controls (HC; n⫽30). BD activity was established as Behçet’s Disease Current Activity Form Simplified (BDCAFs) scoreⱖ2. Oxidative burst was assessed by dihidrorhodamine (DHR) oxidation before and after stimulation with phorbol myristate acetate (PMA) or human plasma (from HC or aBD). H2O2 and O2⫺ production was determined in neutrophils and PBMC by luminol/ lucigenin luminescence intensity with or without stimulation with PMA, Streptococcus pneumoniae, Streptococcus sanguinis, Candida albicans or human plasma. Results: There was no significant difference in medication use between aBD and iBD patients. Resting phagocytes from the four groups presented equivalent oxidative burst (DHR assay and H2O2/O2⫺ production). However, aBD neutrophils produced more O2⫺ when exposed to aBD plasma (median/range⫽96,297/19,202–298,941) compared to HC plasma (48,831/4,001–105,848; p⫽0.028) or non-stimulated (5,721/551– 23,838; p⬍0.01). HC neutrophils also produced more O2⫺ when stimulated by aBD plasma (93,452/12,242–226,932) versus HC plasma (36,225/8,432–79,461; p⫽0.028) versus non-stimulated (3,387/1,870– 13,935, p⬍0.01). aBD neutrophils also produced more H2O2 when exposed to aBD plasma (404,045/24,825-1,408,347) compared to HC plasma (338,238/5,042–745,653; p⫽0.02) or non-stimulated (9,300/ 2,706–27,193; p⬍0.01). The same occurred with neutrophils from HC: aBD plasma (355,079/51,354–513,977); non-stimulated (15,092/4,95947,814; p⬍0.01). PBMC from aBD produced more O2⫺ when exposed to aBD plasma (39,208/9,656-315,306) compared to HC plasma (10,135/ 3,394-41,873; p⫽0.046) or non-stimulated (10,052/3,262-37,352; p⫽0.028). The same occurred with HC PBMC: 24,531/4,931–73,813; 3,661/1,697-28,992 (p⫽0.03); 1,994/928-5,532, respectively (p⫽0.02). Finally, H2O2 production in aBD PBMC was enhanced by aBD plasma (55,223/17,186–253,194) versus HC plasma (35,193/2,081–331,239; p⫽0.05) versus non-stimulated (18,690/2,340–78,728; p⫽0.046). The same was observed with HC PBMC: aBD plasma (10,357/4,024–75,036); HC plasma (2,502/975-7,810, p⫽0.028). H2O2 and O2⫺ production after microbial stimuli was equivalent for phagocytes from aBD, iBD, and HC. Conclusion: Phagocytes from BD patients were not constitutively activated and responded normally to microbial and PMA stimuli. Plasma from patients with active BD exerted a strong stimulus for H2O2 and O2⫺ production. These findings warrant further studies in order to identify possible metabolic pathways involved in neutrophil activation in BD. Disclosure: S. F. Perazzio, None; P. V. S. Pereira, None; A. W. S. Souza, None; A. Condino-Neto, None; L. E. C. Andrade, None.

2615 Serum Beta 2 Microglobulin and Its Association With Disease Activity in Patients With Behcet’s Disease. Meltem Alkan Melikoglu1 and Mehmet Melikoglu2. 1Ataturk University Medical School, Rheumatology, Erzurum, Turkey, 2Health Ministry Erzurum Regional Training and Research Hospital, Dermatology, Erzurum, Turkey. Background/Purpose: ␤2-microglobulin (␤2M) is a low-molecular weight protein, that is a constant component of human leukocyte antigen (HLA), which is secreted from all nucleated cells. Serum ␤2M levels increase in lymphoproliferative and autoimmune diseases related to the activities of B lymphocytes (1). The number of B cells secreting immunoglobulins increased in the active phase of Behçet’s disease (2). The aim of this study was to evaluate a possible relation between B2M and Behcet’s Disease (BD) clinical disease activity. Methods: Seventy eight patients with BD were included in our study. Disease activity was evaluated with “BD Current Activity Form” (BDCAF) which offers an easy-to-complete, valid and reliable method of

S1114

assessing disease activity in BD. With this activity form, headache, mouth ulceration, genital ulceration, erythema, skin pustules, arthralgia, arthritis, new involvements in gastrointestinal system (GIS), eye, nervous system, major vessels and patient’-physician’s assessments of the overall disease activity scores were analyzed. The levels of serum B2M of the patients were also measured. The independent samples t test and Pearson’s correlation test were used to analyze the data. Results: In the comparison of serum B2M levels of the patients with or without the components of BDCAF, no significant relation could be found between headache, mouth ulceration, genital ulceration, erythema nodosum, superficial thrombophlebitis, skin pustules, arthralgia, arthritis, new involvements in gastrointestinal system (GIS), nervous system, major vessels and patient’-physician’s overall perception of disease activity. However, we found significantly higher levels of B2M in patients with uveitis than without eye involvement (p⫽0,032). Although each component of BDCAF did not show an association with B2M levels except from eye involvement, total BDCAF score was also found to be correlated with B2M levels (p䡺0,05). Conclusion: Presence of a new developed eye involvement may be associated with higher levels B2M. Correlation between total disease activity score and B2M levels may contribute the overall disease activity perception in BD. Disclosure: M. A. Melikoglu, None; M. Melikoglu, None.

2616

Background/Purpose: The etiology and pathogenesis of Behçet’s disease (BD) are not yet well understood, but immunoregulatory abnormalities have been proposed as pathogenic mechanisms. Ischemiamodified albumin (IMA) is increased in diseases associated with oxidative stress, which plays an important role in the development of Behçet’s disease. To investigate oxidative status and IMA levels in patients with Behçet’s disease and their association with clinical and laboratory parameters Methods: Thirty-five patients with Behçet’s disease and 31 healthy controls matched for age and gender were enrolled. Patients’ clinical and demographic characteristics were recorded. Serum IMA, malondialdehyde (MDA), total antioxidative status (TAS), total oxidative status (TOS) and oxidative stress index (OSI) levels were then evaluated. Results: Serum IMA, MDA, TOS and OSI levels were significantly higher in patients with Behçet’s disease than in the healthy control group. Serum TAS was significantly lower in patients with Behçet’s disease than in the healthy controls. There was a statistically significant positive correlation between IMA levels and MDA, TOS, erythrocyte sedimentation rate and C-reactive protein levels. IMA levels were statistically significantly higher in patients with vascular involvement (p⫽0.016). Conclusion: Our results demonstrate increased oxidative stress and decreased antioxidant capacity in patients with Behçet’s disease. IMA levels were higher in patients with Behçet’s disease than in healthy controls. IMA may be produced through an adaptive response to chronic hypoxia and oxidative stress, which may play a role in the systemic inflammation seen in Behçet’s disease. IMA can be a useful guiding marker, particularly in patients with vascular involvement. Disclosure: E. Capkin, None; M. Karkucak, None; M. Kola, None; A. Karaca, None; S. C. Karahan, None; A. Sumer, None; A. A. Capkin, None; F. Gokmen, None; R. A. Sari, None.

2617 Clinical Features Of 127 Patients With Behcet’s Disease In Japan. Keisuke Nishimura1, Jun Saegusa1, Sho Sendo1, Yoshinori Kogata1, Goichi Kageyama1, Seiji Kawano1, Shunichi Kumagai2 and Akio Morinobu1. 1Kobe University Graduate School of Medicine, Kobe, Japan, 2Shinko hospital, Kobe, Japan.

Disclosure: K. Nishimura, None; J. Saegusa, None; S. Sendo, None; Y. Kogata, None; G. Kageyama, None; S. Kawano, None; S. Kumagai, None; A. Morinobu, None.

2618 Bipolar Disorders May Represent a Primary Feature Of Behçet’s Disease. Rosaria Talarico1, Laura Palagini2, Elena Elefante1, Claudia Ferrari1, Chiara Stagnaro1, Chiara Baldini1, Chiari Tani1, Marta Mosca3 and Stefano Bombardieri1. 1Rheumatology Unit, Pisa, Italy, 2Psychiatry Unit, Department of Neuroscience, Pisa, Italy, 3Rheumatology Unit, University of Pisa, Pisa, Italy. Background/Purpose: Frequency of psychiatric disorders in BD is a debated issue: while some experts attribute their presence to the chronicity of the illness, others think that they may be imputable to disease activity or to intrinsic features of the disease. The primary aims were to determine the frequency of psychiatric disorders in BD patients, both with neurological involvement or without; the secondary aims were: to investigate a possible association between disease activity/organ involvement and psychiatric profile of the BD patients and to compare the distribution of psychiatric disorders of patients with BD with those in patients with other chronic diseases. Methods: One hundred and twenty BD patients with a diagnosis of BD according the ISG criteria were studied. Demographic profile of the cohort studied are summarized in Table 1. Psychiatric disorders evaluated were: bipolar disorder, obsessive-compulsive disorder, depression and sleep disorder. Age and sex matched disease controls of systemic lupus erythematosus (SLE) and chronic arterial hypertension were included. BD disease activity was evaluated by means of BD current activity form 2006 and clinician’s overall perception of disease activity. Results: Prevalence of psychiatric disorders are shown in Table 2. No correlations were found between the presence of psychiatric disorders and

S1115

Tuesday, October 29

Ischemia-Modified Albumin: A Novel Marker of Vascular Involvement in Behcet’s Disease? Erhan Capkin1, Murat Karkucak2, Mehmet Kola3, Adem Karaca2, su¨leyman Caner Karahan4, Aysegul Sumer4, Arzu Aydin Capkin5, Ferhat Gokmen2 and Refik Ali Sari6. 1Rheumatolgy, Trabzon, Turkey, 2Rheumatology, Trabzon, Turkey, 3Oftalmology, Trabzon, Turkey, 4 BioChemistry, Trabzon, Turkey, 5Dermatology, Trabzon, Turkey, 6Immunology and Allergy, Trabzon, Turkey.

Background/Purpose: Behcet’s disease (BD) is characterized by recurrent oral aphthae and any of several systemic manifestations including genital ulcer, ocular disease, skin lesions, gastrointestinal involvement, neurologic disease, vascular disease, or arthritis. BD is more common along the ancient silk road, including Japan. We present the disease profile of Japanese BD patients at Kobe University Hospital. Methods: We retrospectively investigated the clinical features of 127 patients (63 males and 64 females) who fulfilled the International Criteria for Behcet’s Disease (ICBD) and were treated in our hospital from April 2008 to May 2013. We assessed sex, age of onset, symptoms, complications, type of HLA and medications. Results: Median age at diagnosis was 37.6 ⫾12.7 years old. Oral ulcers were the most common manifestation (97.6%), followed by genital ulcer (52.0%), ocular involvements (52.0%), papulopustular lesion (45.7%), erythema nodosum (43.3%), gastrointestinal manifestations (22.0%), vascular involvements (11.0%), and neurogenic diseases (5.5%). Ocular involvements were found to be significantly more frequent in males (male 63.5%, female 40.6%, P⫽0.01). The frequency of gastrointestinal manifestations was significantly higher in patients with vascular involvements (P⫽0.007), while patients with gastrointestinal manifestations showed a lower association rate with ocular involvements (P⫽0.001). Of the 127 BD patients, 2 patients were associated with spondyloarthropathy, one patient with giant cell arteritis and one with systemic sclerosis. The relationship of HLA and disease manifestations was studied in 54 patients (32.7% with HLA-B51 and 23.1% with HLA-A26). Ocular manifestations were significantly higher in patients with B51 than those without B51 (P⫽0.03). Colchicine was most commonly used for the treatment of BD (49.6%). TNF␣ inhibitor, infliximab, was used for patients with refractory ocular manifestations (n⫽21) and gastrointestinal manifestations (n⫽8). Only 3 patients had to be discontinued because of ineffectiveness or adverse events. The cumulative proportion of patients continuing infliximab therapy at 1 and 2 years was 95.8% and 84.3%, respectively. Pneumonia was the only severe adverse event. Conclusion: A higher incidence of gastrointestinal manifestations was observed in Japanese patients with BD in our hospital. The frequency of gastrointestinal manifestations was significantly higher in patients with vascular involvements. Of the 127 BD patients, 25 patients (19.7%) were treated with infliximab, and the long-term infliximab treatment persistence rate was high.

disease activity; specifically either BD patients with an activity index ⱕ 7 and ⱖ 8 were equally characterized by a high prevalence of psychiatric disorders. Moreover, the occurrence of psychiatric disorders did not result correlating with a specific organ involvement. Comparing the frequency of bipolar disorder with the disease controls, the results have shown a significant difference in favor of bipolar disorder in BD (p ⬍ 0.001). Table 1. Demographic profile.

Number of patients M/F Mean age ⴞ SD (min-max) (years) Mean disease duration ⴞ SD (min-max) (years)

Neuro- BD

BD without neurological involvement

46 38/8 43 ⫾ 7 (15–68)

64 42/22 42 ⫾ 8 (18–71)

9 ⫾ 2 (3–28)

10 ⫾ 2 (4–28)

M:F Mean age ⫾ SD

Table 2. Prevalence of psychiatric disorders

Tuesday, October 29

and 2/5 with more than one type of major involvement had planned to end their lives. BS patients with eye involvement had the highest frequency of answering positively to these 2 questions (p⬍0.001 for each). None of the drugs that were used seemed to be associated with suicidal ideation. For each of the questions regarding suicidal ideation, the BDCAF, BDQoL and BDI scores of BS patients were higher among those who answered positively. Regardless of suicidal ideation, the BDI scores of BS patients with active major organ involvement were higher than AS patients and healthy controls.

psychiatric disorders

Neuro- BD n (%)

BD without neurological involvement n (%)

bipolar disorder obsessive-compulsive disorder depression sleep disorder

41 (65) 29 (46) 20 (32) 5 (11)

28 (64) 20 (43) 16 (36) 10 (16)

BS major organ (nⴝ63)

BS mucocutaneous (nⴝ240)

AS (nⴝ50)

Healthy controls (nⴝ106)

p

36/27

83/157

14/36

77/29

⬍ 0.001

36 ⫾ 12

38.2 ⫾ 11

37.9 ⫾ 10.3

36 ⫾ 10.1

0.218

Thought about ending their lives

16/63 (25.4%)

21/240 (8.8%)

4/50 (8%)

7/ 106 (6.6%)

⬍ 0.001

Ever planned to end their lives

16/63 (25.4%)

22/240 (9.2%)

4/50 (8%)

9/ 106 (8.5%)

0.002

Beck depression inventory scores

16.9 ⫾ 11.4

10.9 ⫾ 9

10.8 ⫾ 10.1

9 ⫾ 7.7

⬍ 0.001

Conclusion: The frequency of suicidal ideation is increased among BS patients who have major organ involvement. This increase in suicidal ideation is correlated with higher disease activity and increased depression scores. Longitudinal studies are required for determining whether the frequency of suicidal ideation decreases with treatment and improvement in disease activity. Disclosure: D. Uzunaslan, None; C. Saygin, None; G. Hatemi, None; K. Tascilar, None; H. Yazici, None; V. Hamuryudan, None.

Conclusion: Our results show a high frequency of psychiatric disorders in BD patients. This elevated prevalence both in BD patient with or without neurological involvement, independently from disease activity and significantly than in disease controls, strongly suggest that psychiatric disorders may represent a primary feature of BD. Disclosure: R. Talarico, None; L. Palagini, None; E. Elefante, None; C. Ferrari, None; C. Stagnaro, None; C. Baldini, None; C. Tani, None; M. Mosca, None; S. Bombardieri, None.

2619 Suicidal Ideation Among Patients With Behcet’s Syndrome. Didem Uzunaslan, Caner Saygin, Gulen Hatemi, Koray Tascilar, Hasan Yazici and Vedat Hamuryudan. Istanbul University, Cerrahpasa Medical Faculty, Rheumatology, Istanbul, Turkey. Background/Purpose: An increased frequency of depression has been reported in Behcet’s syndrome (BS). While an increased suicidal ideation has been reported in other chronic rheumatologic conditions, this has not been studied in patients with BS. We aimed to evaluate the frequency of suicidal ideation among BS patients and to delineate the factors predicting an increase in suicidal ideation. Methods: The frequency of suicidal ideation was evaluated with a standard questionnaire that was previously used (2), among BS patients with and without major organ involvement, and ankylosing spondylitis (AS) patients who attended our outpatient clinic for their routine visits and a group of healthy controls. We questioned whether the subjects 1) thought that life was not worth living, 2) thought about ending their lives within the previous year, 3) ever planned to end their lives and if yes, whether they had planned to end their lives before the onset of their disease. In addition, Beck depression inventory (BDI) and Behcet’s disease quality of life (BDQoL) questionnaires were filled. Disease activity was assessed using the Behcet’s disease current activity form (BDCAF) for BS patients and BASDAI and BASFI for AS patients. Results: We surveyed 240 BS patients who had only mucocutaneous symptoms, 63 BS patients with active major organ involvement (30 eye, 31 vascular, 7 neurologic involvement), 50 patients with AS, and 106 healthy controls (Table). Number of subjects who answered affirmatively to the second and third questions were significantly higher among BS patients with active major organ involvement compared to the other groups (p⬍0.001 and p⫽0.002). Among BS patients with active major organ involvement, 9/27 with eye, 1⁄4 with neurologic, 3/27 with vascular, and 3/5 with more than one type of major involvement had thought of ending their lives. 9/27 with eye, 2/4 with neurologic, 3/27 with vascular

2620 Development Of De Novo Major Involvement During The Follow-Up In Behçet’s Disease. Claudia Ferrari1, Rosaria Talarico1, Chiara Stagnaro1, Anna d’Ascanio1, Chiara Tani1, Chiara Baldini1, Marta Mosca2 and Stefano Bombardieri1. 1Rheumatology Unit, Pisa, Italy, 2Rheumatology Unit, University of Pisa, Pisa, Italy. Background/Purpose: Behçet’s disease (BD) is globally characterized by a variable spectrum of disease profile: while prevalent mucocutaneous lesions and arthritis represent the only clinical features in patients with a benign disease subset, there are other patients who develop potentially sight or life-threatening manifestations, due to ocular, neurological or major vascular involvement. The primary aim of the study was to evaluate the incidence of the novo major involvement during the follow-up in a cohort of patients with BD; the secondary aim was to analyze the epidemiological profile and the long-term outcome of those patients who developed de novo major involvement. Methods: One hundred and twenty patients were evaluated. The male/female ratio was 1.6:1, with a mean disease duration of 11⫾5 years. Their mean age was 42⫾9 years (min:18, max:77), while the mean age at disease onset was 25⫾4 years (min:10, max:58). The mean ⫾ SD duration of follow-up at our centre was 8⫾2 (min:2, max:12) years. We have defined the development of de novo major involvement during the follow-up as the occurrence of severe ocular (anterior uveitis, posterior uveitis, retinal vasculitis), vascular (deep vein thrombosis, superficial vein thrombosis, arterial thrombosis, arterial aneurysm) or CNS (ischaemic pons-mesencephalon lesions and meningoencephalitis) involvement after a latency period from the diagnosis of at least 3 years Results: At the time of diagnosis, the 52% of the cohort (n⫽62) presented a prevalent muco-cutaneous involvement. Among this sub group of patients, we observed that after at least 3 years from the diagnosis, 21 patients (34%) was characterized by the occurrence of de novo major involvement (i.e. ocular: 3; CNS: 9; vascular: 9). The demographic profile of this subgroup (male: 19, female: 2) was characterized by a mean age of 33⫾4 years (min 24-max 40) and a young age at disease onset. The long term outcome after a mean follow-up of 8 years has shown that the majority of these subjects presented during the disease course also relapsing attacks. Moreover, they were characterized by a bigger number of DMARDs (both traditional and anti TNF-alpha) compared to the other part of the cohort Conclusion: As awaited, younger age and male sex represent predictive factors of poor long-term clinical outcome. We have found that ocular disease seems the organ involvement that more frequently has an onset before the first years of disease. Globally, he development of the novo major involvement during the course of BD represents an entity that may

S1116

confirm that a tight control is strongly recommended during the course of BD. Disclosure: C. Ferrari, None; R. Talarico, None; C. Stagnaro, None; A. d’Ascanio, None; C. Tani, None; C. Baldini, None; M. Mosca, None; S. Bombardieri, None.

2621 Venous Claudication Is A Severe and Frequent Symptom In BEHCET’S Syndrome. SerdaL Ugurlu1, Emire Seyahi1, Veysel Oktay2, Zerrin Yigit2, Serdar Kucukoglu2 and Hasan Yazici1. 1Cerrahpasa Medical Faculty, University of Istanbul, Istanbul, Turkey, 2Cardiology Institute, University of Istanbul, Istanbul, Turkey.

Table. Severity and frequency of claudication Healthy BS patients with venous BS patients without controls thrombosis n ⴝ 59 thrombosis n ⴝ 42 n ⴝ 55 Age, mean ⫾ SD, years Claudication as assessed by questionnaire, n (%) Leg pain during the treadmill exercise (the patient continues to walk) n (%) Claudication necessating the termination of the treadmill exercise, n (%)

Comparison Of Different Methods Of Skin Pathergy Test In Patients With Behçet’s Syndrome. Aysegul Lacin, Cigdem Atan Uzun, Zafer Gunendi and Feride Gogus. Gazi University Faculty of Medicine, Ankara, Turkey. Background/Purpose: Skin pathergy test (SPT) has a diagnostic value in Behçet’s syndrome (BS). There are different descriptions of the test which may effect its positivity rate. We aimed to compare different methods of SPT in patients with Behçet’s syndrome. Methods: Patients with BS who fulfilled the International Study Group Criteria for Behçet’s Disease and healthy volunteers were involved in the study. At 90% power (alpha⫽ 0.5) for a 47% positivity rate for the oral method 20 patients and healthy volunteers were recruited (1). Pricks at five different sites were performed with a 20 gauge needle. i)subcutaneous ii) intradermal iii) intradermal with 0.1 cc of 0.09% saline solution iv) intravascular v) oral. The skin pathergy test was considered positive if there was a papule, pustule or oral aphtae (for the oral test) at the prick site after 48 hours. Sensitivity, specificity, positive and negative predictive values of SPT were calculated. Results: All patients with Behçet’s syndrome were positive for at least one SPT method where none of the healthy volunteers had a positive SPT. Among the five SPT methods subcutaneous and oral methods were most sensitive (70% and 60% respectively). All SPT methods showed 100% specificity and positive predictive value. Sensitivity and negative predictive values of SPTs are shown in Table 1. Table 1. Sensitivity and negative predictive value of skin pathergy test methods Skin Pathergy Methods

Sensitivity

Negative predictive value

Subcutaneous Oral Intradermal Intradermal with saline Intravascular

%70 %60 %50 %50 %20

%77 %71 %67 %67 %55

Conclusion: Subcutaneous method is the most sensitive test among SPTs. Oral method may be an alternative for subcutaneous SPT. Reference: 1. Sharquie KE, Al-Araji A, Hatem A.Oral pathergy test in Behçet’s disease Br J Dermatol. 2002 Jan;146(1):168–9 Disclosure: A. Lacin, None; C. A. Uzun, None; Z. Gunendi, None; F. Gogus, None.

2623 A CT Evaluation Of Pulmonary and Cardiac Lesions In BEHÇET’S Syndrome Patients Without Pulmonary Symptoms. Emire Seyahi1, Deniz Cebi Olgun1, SerdaL Ugurlu1, Idil Hanci2, Reona Takahashi1 and Hasan Yazici1. 1Cerrahpasa Medical Faculty, University of Istanbul, Istanbul, Turkey, 2Krankrenhaus Nordwest, Frankfurt am Main, Frankfurt, Germany.

p

37 ⫾ 7 18 (31)

34 ⫾ 7 2 (5)

36 ⫾ 9 0

10 (17)

3 (7)

1 (2)

0.016

6 (10)

0

0

0.006

0.346 ⬍ 0.001

Conclusion: Venous claudication seems to be a severe and frequent symptom being present in up to 1/3 of BS patients with major vein involvement. It clearly limits the walking capacity in 10 % of these patients even when tested in a treadmill set at low pace. References: 1) Ugurlu S, et al.. Rheumatology (Oxford). 2008;47:472–5. Disclosure: S. Ugurlu, None; E. Seyahi, None; V. Oktay, None; Z. Yigit, None; S. Kucukoglu, None; H. Yazici, None.

Background/Purpose: In Behçet’s syndrome (BS) patients with symptomatic pulmonary artery involvement (PAI) varying and multiple pulmonary parenchymal and cardiac lesions can be seen in thorax CT examinations (1). These lesions are pulmonary nodules of ⱖ 1 cm (85 %), cavities (47 %), ground-glass opacities (45 %), mild pleural (45 %) and pericardial effusions (21 %), mediastinal lymphadenopathies (21 %) and intracardiac filling defects (28 %). Whether these pulmonary and cardiac abnormalities are exclusively observed in symptomatic BS with PAI or in vascular involvement without PAI have not been adequately studied. We now report the frequency of such lesions among BS patients with vascular disease but no PAI along with BS patients with no vascular disease at all. Methods: Consecutive BS patients seen in the outpatient clinic between December 2011 and November 2012 were studied. Those with a disease duration of ⬎ 5 years were excluded. Contrasted thorax CT scans were obtained in BS patients with vascular disease along with BS patients with no vascular disease at all. Only patients with no prior history relevant with PAI or those with no pulmonary symptoms such as hemoptysis,

S1117

Tuesday, October 29

Background/Purpose: In a previous cross-sectional questionnaire survey, we had shown that intermittent claudication was significantly more common among BS patients when compared to healthy controls, and had proposed that this was a “venous claudication” (1) rather than due to atherosclerotic vascular disease since a. a history of myocardial infarction or angina pectoris were not increased in frequency in the same group of patients as compared to age and sex matched healthy controls; and b. this intermittent claudication was specifically more common among males with venous thrombosis. The so called venous claudication is thought to be an exercise induced pain resulting from venous outflow impairment (1). With this study we aimed to 1. To reassess the frequency of venous claudication by a questionnaire survey and 2. To further study this venous claudication prospectively by a formal treadmill exercise in BS patients with and without venous thrombosis along with healthy controls. Methods: We studied 59 BS patients with lower extremity venous thrombosis (LEVT), 42 BS patients without venous disease and 55 healthy controls. All patients and controls were male. Patients and controls with peripheral arterial disease were excluded. Intermittent claudication was assessed initially by Rose questionnaire. After this, patients were asked to walk in the treadmill at a set speed of 3.5 km/h and 10% inclination for 10 minutes. Patients who first experienced persistent symptoms consistent with venous claudication but still able to walk and those who had to give up the treadmill were noted. Pre and post-exercise ankle brachial pressure indices (ABPIs) were also measured. Results: The mean ages of the patients and controls were similar (Table). Pre and post-exercise ABPIs did not differ between patients and controls. There were significantly more patients who described claudication in the questionnaire among those with LEVT (31 %) compared to those with no venous disease (5 %) and healthy controls (0 %) (P⬍0.001) (Table). Similarly, the number of patients who experienced claudication but still continued to walk on the treadmill were significantly more among those with LEVT. Finally, only those with LEVT had to stop the treadmill challenge due to claudication. Pre and post exercise ABPI’s were similar among BS patients with LEVT. There was no relation between the presence of vena cava, iliac or femoral vein involvement and the presence of claudication or limitation of walking capacity.

2622

cough, chest pain and dyspnea were studied. CT scans were analyzed formally using a checklist by a radiologist who was blinded to the clinical diagnoses of the patients. Results: We studied 49 (43 M/ 6 F) BS patients with vascular involvement and 35 (32 M/3 F) BS patients with no vascular involvement. Lower extremity deep or superficial vein thrombosis was present in all 49 patients with vascular involvement. The mean age (32 ⫾ 7 years vs 33 ⫾ 8 year, respectively) and disease duration (4 ⫾ 3 years vs 3 ⫾ 2 years, respectively) among patients with and without vascular involvement were similar. None of the patients had pulmonary artery aneurysms or thrombosis. Similarly, cavities, pleural or pericardial effusions, and intracardiac filling defects were not observed in any of the scans. The most common parenchymal lesions were nodules of ⬍ 1 cm, which were present in 61 % and 37 % patients with and without vascular involvement, respectively, (P ⫽ 0.029). Nodules of ⱖ 1 cm were only present in 4 patients with vascular disease (8 %) and none of those with no vascular disease. Ground-glass opacities were present in 2 and 3 patients with and without vascular involvement, respectively. The frequency of pulmonary and cardiac lesions in the current study was much less when compared to that found in PAI. Conclusion: Pulmonary parenchymal and cardiac lesions were rarely observed in BS patients with no pulmonary artery involvement. Although small pulmonary nodules (⬍1 cm) seemed to be more frequent among patients with vascular disease compared to those without, their significance remains to be further studied. References: 1) Seyahi E et al. Pulmonary artery involvement and associated lung disease in Behçet disease: a series of 47 patients. Medicine (Baltimore). 2012;91:35–48.

Tuesday, October 29

Disclosure: E. Seyahi, None; D. Cebi Olgun, None; S. Ugurlu, None; I. Hanci, None; R. Takahashi, None; H. Yazici, None.

2624 Evaluation of Asymptomatic Venous Disease By Venous Doppler Ultrasonography in Patients With Behcet’s Disease. Fatma AlibazOner, Emrah Karatay, Ihsan Nuri Akpinar, Tu¨lin Ergun and Haner Direskeneli. Marmara University, School of Medicine, Istanbul, Turkey. Background/Purpose: One of the major causes of mortality and morbidity in Behcet’s disease (BD), especially in young males of Mediterranean origin, is vascular involvement. A limited data suggests also a high prevelance of venous insufficiency (VI) and some cases of asymptomatic thrombosis in BD. In this study, we aimed to investigate prospectively asymptomatic venous disease by venous doppler ultrasonography (US) in patients with BD without known vascular disease. Methods: The study included 93 patients with BD (M/F: 45/48, age: 36.4⫾10 years), 97 patients with ankylosing spondylitis (AS)(M/F: 50/47, age: 37.5⫾9.5 years) and 43 healthy control subjects (M/F: 25/18, age: 34.7⫾4.5 years). Vessels of both upper and lower extremities were examined while the subjects were in supine position by venous doppler US. CEAP severity score was used to evaluate the severity of the venous insuficiency. Results: Upper extremity veins were totally normal in all patients and controls. We also did not detect any “silent thrombosis” in study groups. Venous insufficiency findings in lower extremity were detected in 32.2% (n⫽30) in the BD group, 28.8% (n⫽28) in AS group and 9.3% (n⫽4) in the healthy control group. Both BD and AS patients had significantly higher VI rates than healthy controls (p⫽0.007 and 0.015). The rate of VI was similar between patients with BD and AS (p⫽0.64). Similarly, CEAP severity score in BD (0.34 (0–3)) was significantly higher than controls (0)(p⫽0.008), but similar to AS (0.18 (0–39)(p⫽0.18). No correlations were present between C-reactive protein elevations (⬎5 mg/L) and VI in patients with both BD (p⫽0.546) and AS (p⫽0.754). Conclusion: A high prevelance of venous insufficiency was present in both BD and AS patients without asymptomatic thrombosis. Presence of VI also in AS, a disease without a major tendency to venous thrombosis, suggests that chronic inflammation might cause a mild insufficiency detected only by Power US in venous vessels. Long-term consequences of this finding requires further follow-up studies. Disclosure: F. Alibaz-Oner, None; E. Karatay, None; I. N. Akpinar, None; T. Ergun, None; H. Direskeneli, None.

2625 Biologic Therapy In Refractory Uveitis Of Behcet’s Syndrome: Switching and Dose Modification. Multicenter Study Of 124 Patients. F. OrtizSanjuań1, Vanesa Calvo-Rıó1, Ricardo Blanco2, Emma Beltrań3, Juan SańchezBursoń4, Marina Mesquida5, Alfredo M. Adan5, M Hernandez Grafella3, E Valls Pascual6, L Martıńez-Costa6, A Sellas-Ferna`ndez7, Miguel Cordero-Coma8, Manuel Diaz-llopis9, David Salom9, Jl Garcıá Serrano10, Norberto Ortego10, JM Herreras11, Alejandro Fonollosa12, A Aparicio13, O Maı´z14, A Blanco14, I Torre15, Cruz Fernańdez-Espartero16, V Jovani17, D Peitado-Lopez18, Esperanza Pato19, J Cruz20, J. Carlos Fernandez-Cid20, Elena Aurrecoechea21, M Garcıá22, M Caracuel23, Carlos Montilla24, A Atanes25, F Francisco26, S Insua27, S Gonza´lez-Sua´rez28, A Sańchez-Andrade29, F Gamero30, Luis Linares31, F Romero-Bueno32, J Garcıá33, AJ Garcıá Gonza´lez33, Raquel Almodovar34, E Minguez35, C Carrasco Cubero36, Alejandro Olive´ Marque´s37, J Va´zquez38, O Ruiz Moreno39, F Jimenez-Zorzo39, J Manero39, Javier Loricera1 and Miguel Angel Gonza´lez-Gay1. 1Hospital Universitario Marque´s de Valdecilla. IFIMAV. Santander. Spain, Santander, Spain, 2Hospital Universitario Marque´s de Valdecilla. IFIMAV, Santander, Spain, 3Hospital General Universitario, Valencia, Valencia, Spain, 4Hospital de Valme. Sevilla, Sevilla, Spain, 5Hospital Clıńic of Barcelona, Barcelona, Spain, 6Hospital Peset Valencia, Valencia, Spain, 7Hospital Val d⬘Hebron. Barcelona, Barcelona, Spain, 8Hospital de Leoń, Leoń, Spain, 9Hospital Universitario La Fe de Valencia, Valencia, Spain, 10Hospital San Cecilio. Granada, Granada, Spain, 11Hospital Universitario, IOBA. Valladolid, Valladolid, Spain, 12Hospital de Cruces. Bilbao, Baracaldo, Spain, 13Hospital de Toledo., Toledo, Spain, 14Hospital Donosti San Sebastian, San Sebastiań, Spain, 15Hospital Basurto. Bilbao, Bilbao, Spain, 16Hospital Universitario de Mo´stoles. Madrid, Madrid, Spain, 17Hospital General de Alicante., Alicante, Spain, 18Hospital Universitario La Paz Madrid, Madrid, Spain, 19Hospital Clıńico San Carlos. Madrid, Madrid, Spain, 20Hospital de Pontevedra, Pontevedra, Spain, 21Hospital Sierrallana. Torrelavega, Torrelavega, Spain, 22Hospital La Princesa. Madrid, Madrid, Spain, 23Hospital de Co´rdoba., Co´rdoba, Spain, 24Hospital Universitario de Salamanca, Salamanca, Spain, 25HUCA La Corunã., A Corunã, Spain, 26 Hospital Doctor Negrıń Canarias., Canarias, Spain, 27Hospital Universitario Santiago de Compostela, Santiago de Compostela, Spain, 28Hospital Cabuen˜es, Gijoń, Gijoń, Spain, 29Hospital Lucus Augusti Lugo, Lugo, Spain, 30Hospital San Pedro Alcantara Caceres, Caceres, Spain, 31Hospital Universitario Virgen de la Arrixaca. Murcia, Murcia, Spain, 32Fundacioń Jimenez Dıáz. Madrid, Madrid, Spain, 33Hospital 12 de Octubre. Madrid, Madrid, Spain, 34Hospital Universitario Fundacioń Alcorcoń. Madrid, Alcorcon. Madrid, Spain, 35Hospital Clıńico de Zaragoza, Zaragoza, Spain, 36Hospital de Me´rida, Me´rida, Spain, 37Hospital Germans Trias i Pujol. Badalona, Barcelona, Spain, 38Hospital de Ferrol. A Corunã, A Corunã, Spain, 39Hospital Universitario Miguel Servet. Zaragoza, Zaragoza, Spain. Background/Purpose: Several situations can be observed in patients undergoing biologic therapy in uveitis associated to Behçet⬘s syndrome (BS): a) Patients are switched to another therapy because of insufficient response (IR), toxicity, or change in the route of administration, b) remission is achieved and discontinuation or reduction of dose is performed. Our aim was to study these situations in a large series of BS patients receiving biologic therapy. Methods: Multicenter study of 124 patients with uveitis refractory to conventional therapy who required at least one biologic agent. Standard dose of Infliximab (IFX), generally 5 mg/kg/i.v. was given at 0, 2, 6 and then every 4–8 weeks, Adalimumab (ADA) 40 mg/sc/every other week, golimumab (Goli) 50 mg/4 weeks, tocilizumab (TCZ) 8 mg/kg/i.v./4 weeks and rituximab (RTX) 1g/i.v./15 days (2 doses) every 6 months. Results: The biological agent used as the first choice was either IFX in 77 patients or ADA in 47. All the IFX-treated patients received the standard dose at weeks 0, 2, 6 and then every 4–8 weeks. Initial IFX dosage was: a) IFX at 5 mg/k/i.v. (69 cases) with a maintenance dose every: 4 weeks (n⫽15 cases), 6 weeks (16), 7 weeks (1) or every 8 weeks (37). b) 3 mg/kg (7 cases) with a maintenance dose every: 4 weeks (1 case), 6 weeks (1) or every 8 weeks (5). c) IFX 4 mg/kg (1 case) with a maintenance dose every 4 weeks. Shortening infusion times for IFX administration as the maintenance treatment was required in 5 cases because of IR. Initial IFX therapy was changed to another single biologic agent in 32 cases; 30 to ADA (because of IR [16 cases], decision of changing from i.v. to subcutaneous route of administration [5 cases] or toxicity [9 cases]), 1 to RTX because of infusional reaction, and 1 to etanercept because of toxicity. Initial ADA therapy was changed to another single biologic agent in 5 cases; 2 to Goli (because of IR or toxicity in 1 case each) and 3 to IFX (2 cases because of IR and 1 because of toxicity). In 3 cases there was a double biologic switching, 1 case from ADA to IFX and to Goli, 1 case from IFX to ADA and to Goli and 1 case from ADA to IFX and to TCZ. Improvement was achieved in: a) 14 of the 16 patients switched from

S1118

IFX to ADA due to IR; b) 3 of the 3 patients switched from ADA to another biologic due to IR. Persistent clinical remission was achieved in 66 (53.2%) patients and, the dose was reduced or the agent was discontinued. IFX was decreased from 5 to 3 mg/k/i.v. in 4 patients and the maintenance dose interval was increased in 23 cases. ADA maintenance dose interval was increased for more than two weeks in 13 patients. The biologic agent was discontinued in 26 cases (21 with IFX and 5 with ADA) that had clinically persistent remission. After a mean follow-up of 13.1 ⫾ 9.2 months after biologic agent discontinuation, 21 of these 26 patients remained in remission while 5 of the 26 experienced a flare that led to the resumption of the therapy with same agent achieving remission again in all of them. Conclusion: Switching of biologic agents may be useful. Once clinical remission is achieved, dose reduction or in some cases discontinuation of the biologic agent may be obtained. Disclosure: F. Ortiz-Sanjuań, None; V. Calvo-Rıó, None; R. Blanco, None; E. Beltrań, None; J. Sańchez-Bursoń, None; M. Mesquida, None; A. M. Adan, None; M. Hernandez Grafella, None; E. Valls Pascual, None; L. Martıńez-Costa, None; A. SellasFerna`ndez, None; M. Cordero-Coma, None; M. Diaz-llopis, None; D. Salom, None; J. Garcıá Serrano, None; N. Ortego, None; J. Herreras, None; A. Fonollosa, None; A. Aparicio, None; O. Maı´z, None; A. Blanco, None; I. Torre, None; C. FernańdezEspartero, None; V. Jovani, None; D. Peitado-Lopez, None; E. Pato, None; J. Cruz, None; J. C. Fernandez-Cid, None; E. Aurrecoechea, None; M. Garcıá, None; M. Caracuel, None; C. Montilla, None; A. Atanes, None; F. Francisco, None; S. Insua, None; S. Gonza´lez-Sua´rez, None; A. Sańchez-Andrade, None; F. Gamero, None; L. Linares, None; F. Romero-Bueno, None; J. Garcıá, None; A. Garcıá Gonza´lez, None; R. Almodovar, None; E. Minguez, None; C. Carrasco Cubero, None; A. Olive´ Marque´s, None; J. Va´zquez, None; O. Ruiz Moreno, None; F. Jimenez-Zorzo, None; J. Manero, None; J. Loricera, None; M. A. Gonza´lez-Gay, None.

2626

Background/Purpose: Uncontrolled studies suggest a beneficial effect of infliximab in the treatment of severe uveitis of BS. The majority of these studies had short observation periods. Methods: The charts of 43 BS patients (33 men; age at the initiation of infliximab: 31⫾8.4 SD years) treated with infliximab (5 mg/kg) were reviewed retrospectively. All patients had severe, sight-threatening posterior uveitis of long-duration (82⫾50 SD months) refractory to previous treatments with multiple immunosuppressives (azathioprine⫽43, Cyclosporin A⫽42, Interferon alfa⫽38; duration of previous treatment: 62⫾43 SD months) and steroids. Efficacy was assessed by the change of visual acuity (LogMAR VA) and the decrease in the number of attacks under infliximab. Results: The duration of infliximab treatment was 31⫾18 SD months (median 29 months). In 37 (86%) patients infliximab was combined with azathioprine (n⫽ 26), azathioprine and cyclosporine A (n⫽10) or interferon alfa (n⫽1). Twenty-nine (67%) patients had no useful vision (LogMAR ⬎1) in at least one eye at the time of initiation of infliximab. Five (17%) had gained useful vision of both eyes under infliximab. The mean VA (all patients) was maintained under infliximab (right eye: 0.83⫾ 0.74 vs 0.84 ⫾0.79; left eye: 1.27⫾0.83 vs 1.13⫾ 0.88). VA of the worse eye improved in 11 patients, remained stable in 24 and worsened in 8 patients under infliximab. Eleven patients (26%) had no attacks during previous treatment and were prescribed infliximab to protect the already compromised vision. The VA remained stable or improved in 10 patients and 9 patients remained attack free under infliximab. Thirty-two patients (74%) had attacks before infliximab. Eleven (34%) of them became attack free under infliximab. The number of uveitis attacks per year (all patients) under previous immunosuppressive treatment dropped from 1.9 ⫾3 to 0.3 ⫾0.6 in the right eye (p⫽0.0001) and from 2.5⫾ 2.9 to 0.3 ⫾0.6 in the left eye (p⫽0.0001). There was also a significant decrease in the numbers of patients with hypopyon (11 and 3; x2⫽ 5.46, p⫽0.019), vascular infiltrations (23 and 12; x2⫽ 5.83, p⫽0.016), macular edema (8 and 1; x2⫽ 6.08, p⫽0.014), and retinal hemorrhages (7 and 0; x2⫽ 7.62, p⫽0.006) under infliximab. Infliximab treatment was terminated in 16 patients (adverse events ⫽5, inefficacy ⫽1, terminal eye disease ⫽ 4, unrelated reasons ⫽ 6). Fifteen of these patients received further treatment (biologics ⫽ 3, azathioprine ⫽12) for 22⫾13 SD months and 12 of them remained attack free during this period. The adverse events causing discontinuation were pulmonary tuberculosis in 3 (at 17., 32., and 46. month of infliximab treatment, respectively), depression in 1, pneumonia in 1 patient.

Disclosure: V. Hamuryudan, None; G. Hatemi, None; Y. Ozyazgan, None; D. Ucar, None; S. Yurdakul, None; E. Seyahi, None; K. Tascilar, None; S. Ugurlu, None; H. Yazici, None.

2627 Prognostic Factors Of Visual Function In The Treatment With Infliximab For Uveitis Of Behçet’s Disease. Akihiko Nakabayashi1, Toru Hirano1, Yoshihiro Hishitani1, Keisuke Hagihara1, Kei Nakai2, Kouji Nishida2 and Atsushi Kumanogoh1. 1Osaka University Graduate School of Medicine, Suita, Japan, 2Osaka University Graduate School of Medicine, Suita, Japan. Background/Purpose: Infliximab, a chimeric monoclonal antibody against TNF-␣, is expected to improve prognosis of visual function in patients with severe and refractory uveitis of Behçet’s disease. However, prognostic factors remain to be determined. Methods: Observational study in one institute was conducted in the cohort of patients with Behçet’s disease treated with infliximab. Outcome was defined as failure to recover above 20/50 of visual acuity (VA) at 2 year after the first administration of infliximab. Associations between outcome and baseline characteristics were assessed by bivariable analysis and logistic regression analysis to adjust confounding factors. Significance of each factor was assessed by likelihood ratio test. Results: Infliximab was initiated in 28 patients between February 2007 and May 2012. Nineteen patients continued to receive infliximab beyond 2 year and could be assessed. Among 38 eyes of 19 patients, 18 eyes could not achieve 20/50 of VA (Figure). Bivariable analysis between the outcome and baseline characteristics demonstrated p-value (p) ⫽ 0.53 for sex, p ⫽ 0.19 for disease duration, p ⫽ 0.43 for HLA-B51, p ⫽ 0.97 for C-reactive protein (CRP), p ⬍ 0.0001 for VA at the first administration, p ⫽ 0.088 for deviation of VA during 6 months before the first administration, p ⫽ 0.48 for concomitant use of corticosteroids and p⫽ 0.042 for concomitant use of immunosuppressants such as cyclosporine or methotrexate. Adjusted analysis with confounding factors such as sex, age, disease duration, VA at the first administration and concomitant use of immunosuppressants by logistic regression analysis revealed disease duration (p⫽0.024) and visual acuity (p ⬍ 0.0001) as significant factors (Table). Receiver operating characteristic (ROC) analysis demonstrated 16.5 months of disease duration and 20/140 of visual acuity (Area under the curve: 0.95) for candidate cutoff values.

Figure. Change of visual acuity

S1119

Tuesday, October 29

Infliximab For Sight-Threatening and Refractory Uveitis Of Behcet’s Syndrome. Vedat Hamuryudan1, Gulen Hatemi2, Yilmaz Ozyazgan2, Didar Ucar2, Sebahattin Yurdakul2, Emire Seyahi2, Koray Tascilar2, Serdal Ugurlu2 and Hasan Yazici1. 1Istanbul University, Cerrahpasa Medical Faculty, Rheumatology, Istanbul, Turkey, 2Behcet’s Syndrome Research Center, Cerrahpasa Medical Faculty University of Istanbul, Istanbul, Turkey.

Conclusion: Infliximab combined with immunosuppressives was useful in preserving visual acuity and in controlling the attacks of BS patients with refractory, sight-threatening uveitis at the long-term. Relapses were infrequent after discontinuation. Tuberculosis appears as a concern in prolonged treatment.

2629

Table. Contributable factors for visual acuity at 2 year Factor Sex Age Disease duration Visual acuity Immunosuppressants

p-value 0.41 0.42 0.024 ⬍0.0001 0.64

*regression logistic analysis and likelihood ratio test

Conclusion: Disease duration and VA at the first administration of infliximab were the most contributable factors for the prognosis of visual function after 2 years. Cutoff values, 16.5 months of disease duration and 20/140 of VA suggest the importance of ‘windows of opportunity’ in treatment for uveitis of Behçet’s disease. Disclosure: A. Nakabayashi, None; T. Hirano, None; Y. Hishitani, None; K. Hagihara, None; K. Nakai, None; K. Nishida, None; A. Kumanogoh, None.

2628

Tuesday, October 29

Immunogenicity Of Infliximab Modulates Efficacy and Safety In Behcet’s Disease Patients With Uveitis. Mitsuhiro Takeno, Kayo Terauchi, Yohei Kirino, Ryusuke Yoshimi, Nobuhisa Mizuki, Etsuko Shibuya and Yoshiaki Ishigatsubo. Yokohama City University Graduate School of Medicine, Yokohama, Japan. Background/Purpose: Infliximab (IFX) suppresses ocular attacks in Behcet’s disease (BD) with uveitis, resulting in favorable long-term visual prognosis. However, some patients had ocular attacks which accumulate one or two weeks before the next IFX infusion, suggesting that the efficacy of IFX depends on the concentration. This study investigates IFX through levels and antibody toward IFX (ATI) in BD patients receiving IFX and analyzes the relationship of the pharmacokinetics with clinical efficacy and safety. Methods: We retrospectively examined clinical courses of 20 BD patients (female 7, male 22 age 41.9 ⫹ 14.4 yo) who met the Japanese revised Behcet’s disease Diagnositic criteria (2003) and received IFX because of refractory uveitis to conventional therapies including cyclosporine A. In principle, immunosuppressants were discontinued before introduction of IFX therapy. In the original regimen, IFX (5mg/kg) was given at 0, 2, 6 week, and thereafter every 8 weeks, but the intervals were shortened to 5 weeks after major ocular attacks occurred. The blood samples were drawn prior to the next infusion. IFX concentrations and ATI in the sera were determined by ELSIA. Results: Mean duration was 6.4 ⫹ 5.8 years from the disease onset to initiation of infliximab therapy. A. Duration of IFX was 7 to 80 months. The therapy was discontinued in 2 patients because of infusion reaction and/or insufficient efficacy. Frequencies of ocular attacks (/6 months) were 2.6 ⫹ 2.1 and 0.4 ⫹ 0.5 before and after therapy, indicating that IFX suppresses ocular attacks significantly. During the therapy, total 29 ocular attacks occurred at 6.87 ⫹ 1.12 weeks after the last infusion, 1.43 ⫹ 1.53 weeks before the next infusion. The infusion interval was shortened from 5 to 7 weeks in 8 patients who experienced major ocular attacks. Thereafter, frequency of ocular attacks was reduced from 0.84 to 0.37/6months, while that was 0.10 in patients who continued to receive the infusions every 8 weeks. The mean IFX through level was 5.0 ⫹ 6.1 ␮g/ml but it was undetectable (less than 0.1␮g/ml) in 7 patients. Of the 7 patients, 3 had recent ocular attacks, 6 showed extraocular symptoms, 6 had infusion reaction, and 6 had ATI. Shortening the infusion interval was associated with increased through level, leading to another remission in a patient. Besides 6 patients, ATI was also positive in one patient who had infusion reaction but no ocular attacks. Three of 4 patients who required admission due to infusion reaction were positive for ATI. Conclusion: The present study suggests that low IFX trough level is also associated with ocular attacks, extraocular manifestations, and ATI, which is partially responsible for serious infusion reaction. Therefore, shortening the infusion interval and concurrent usage of immunosuppressants appear reasonable strategies to circumvent the issues. Disclosure: M. Takeno, None; K. Terauchi, None; Y. Kirino, None; R. Yoshimi, None; N. Mizuki, None; E. Shibuya, None; Y. Ishigatsubo, None.

Is Complete Remission a Realistic Target With Current Therapeutic Options in Behçet’s Disease? Fatma Alibaz-Oner1, Gonca Mumcu2, Zeynep Kubilay1, Gu¨lsen Ozen3, Gu¨lce Celik1, Aslı Karadeniz1, Meryem Can1, Sibel Yılmaz Oner1, Nevsun Inanc4, Pamir Atagunduz4, Tu¨lin Ergun1 and Haner Direskeneli1. 1Marmara University, School of Medicine, Istanbul, Turkey, 2Marmara University, Faculty of Health Sciences, Istanbul, Turkey, 3Marmara University School of Medicine., Istanbul, Turkey, 4Marmara University School of Medicine, Istanbul, Turkey. Background/Purpose: The clinical course of Behcet’s disease (BD) as a multi-systemic disorder with a remitting-relapsing nature is unsufficiently explored. As complete remission should be aimed in all inflammatory diseases, we investigated the frequency of complete remission in patients with BD in routine practice. Methods: In this retrospective study, 258 patients with BD (F/M: 130/128, mean age: 41.1⫾11,5 years) classified according to ISG criteria were included. The demographic and clinical data for active organ manifestations and treatment protocols were evaluated, both for the current visit and in the last month. Patients having at least one of any disease manifestations were categorized as active. Results: A total of 1757 visits of 258 patients were overviewed. Mean visit number was 6,8⫾ 2,7 (range: 1–10) and mean follow-up duration was 45.8⫾36.5 months (2–165). One hundred twenty-five patients (48.4%) were of mucocutaneus type, whereas 133 patients (51.6%) had major organ involvement. When all visits combined, 19.8– 43.9% of the patients were using immunosuppressives (IS), whereas 35.3–59.3% was under non-IS therapies such as colchicine or NSAIDs. There was also a group of noncompliant patients (6.4–45%) without any treatment in some visits. Patients were clinically active in 67.2% (n⫽1182) of the total visits (n⫽1757). Frequency of clinical activity increased to 75.6% (68.1– 90.3) when the month before the visit was also included. The major cause of the activity was aphthous ulcers (39.4–63.2%) with other mucocutaneous manifestations also commonly present (Genital ulcer: 3.5–27.1 %, erythema nodosum: 8.2–22.5%, papulopustular lesions: 18.2–33.7%, arthritis: 21.3–33.5%, uveitis: 0.5– 8.5% and vascular involvement: 2.5–10.8%). No difference was observed between the frequency of activity of patients having ISs or non-IS therapies. Conclusion: Although complete remission is the current, primary target in inflammatory rheumatological diseases such as rheumatoid arthritis or vasculitides, it is fairly difficult to achieve complete remission in BD with current therapeutic regimens. The reluctance of the clinicians to be aggressive for some BD manifestations with low morbidity, such as mucocutaneous lesions, might be influencing the continuous, low-disease activity state in BD patients. Disclosure: F. Alibaz-Oner, None; G. Mumcu, None; Z. Kubilay, None; G. Ozen, None; G. Celik, None; A. Karadeniz, None; M. Can, None; S. Yılmaz Oner, None; N. Inanc, None; P. Atagunduz, None; T. Ergun, None; H. Direskeneli, None.

2630 Low Medication Adherence Is Observed In Behcet’s Disease. Gonca Mumcu1, Ali Taze1, Esra Kula1, Semiha Yemez1, Silay Eksi1, Leyla Ko¨ksal1, Fatma Alibaz-Oner2, Sibel Yilmaz Oner2, Pamir Atagu¨ndu¨z2, Nevsun Inanc2, Tu¨lin Ergun2 and Haner Direskeneli2. 1Marmara University, Faculty of Health Sciences, Istanbul, Turkey, 2Marmara University, School of Medicine, Istanbul, Turkey. Background/Purpose: Symptom-control with reduction of mortality and morbidityare the main treatment goalsin Behcet’s Disease (BD). However, similar to other chronic disorders, increased unnecessary visits, hospitalisation and medication costs are thought to originate from lack of adherence to medications of the treatment protocols in BD. Therefore, the aim of this study was to evaluate self-reported medication adherence in BD patients. Methods: The study group was composed of 118BD patients(F/M:62/ 56, mean age:37.7⫾10.8 years) and disease control groups regarding 50

S1120

patients with Rheumatoid Arthritis (RA, F/M:37/13, mean age:39.9⫾10.4 years) and 58 patients with Familial Mediterranean Fever (FMF, F/M:39/ 19, mean age:34.4⫾11.9 years). Medication adherence was evaluated by the Morisky scale that is a brief 4-item structured questionnaire. Each patient has a scale score ranging from “0” to “4” with low scores indicating better adherence. Trained interviewers (n⫽4) who were not involved in any disease assessment or treatmenthelped to individuals with visual impairment or illiterates in filling the questionnaire. Results: The Moriskyscale score was similar in BD (2.05⫾0.99) and FMF (2.01⫾1.1) (p⫽0.99), whereas it was significantly lower in RA (1.54⫾1.05) than BD (p⫽0.021).In BD, no significant difference was present in the Morisky scale score according to gender (females:2.09⫾0.9vsmales: 2.0⫾1.02, p⫽0.60).The score of patients with major organ involvement (2.1⫾1.1) was also similar to mucocutaneous ones (1.9⫾0.9) (p⫽0.57) in males. In contrast, the score was lower in patients with major organ involvement (1.7⫾1.1) compared to patients with mucocutaneous involvement (2.29⫾0.8) in female BD patients (p⫽0.04). Moreover, a weak correlation was observed between the Morisky scale score and the number of medication (4.8⫾2.4, r:0.3 p⫽0.029). Conclusion: Self-reported medication adherence was found to be low in BD patients. Medication adherence seems to be different according to gender in the frame of major organ involvement, suggesting that either the perception or consequences of disease-associated morbidity is associated to gender-related features in BD. Disclosure: G. Mumcu, None; A. Taze, None; E. Kula, None; S. Yemez, None; S. Eksi, None; L. Ko¨ksal, None; F. Alibaz-Oner, None; S. Yilmaz Oner, None; P. Atagu¨ndu¨z, None; N. Inanc, None; T. Ergun, None; H. Direskeneli, None.

2631

Background/Purpose: Behçet’s Syndrome (BS) follows an active course during the child-bearing years in both men and women. Whether fertility is decreased among the BS patients due to the condition itself or to the frequently used medications like cyclophosphamide (CYC), azathioprine or colchicine is not clear. We aimed to determine the infertility rate, and the effect of drugs and types of organ involvement on fertility in BS patients. Methods: We included BS patients with and without major organ involvement, familial Mediterranean fever (FMF), ankylosing spondylitis (AS) patients and healthy controls recruited from hospital staff. In order to show a 20% increase in infertility in BS with 0.05 alpha and 80% power, we calculated that each group should contain at least 125 individuals. Among patients who visited the clinic for routine controls, individuals with an even waiting list number were selected. A structured interview was performed by two physicians and medical records were reviewed. Infertility was defined as the inability to conceive after one year of unprotected intercourse. We compared the differences in the proportion of individuals who had never conceived in their lifetime, who had a successful conception but became infertile after disease onset, and individuals who conceived late or with assisted reproductive technology (ART), between BS and control groups. Finally infertility was separately assessed in a group of 62 patients who had used CYC and was compared to that observed among patients with major organ involvement who were CYC naı¨ve. Multivariate logistic regression analysis was used to determine the association of infertility with involved organs and the drugs which were used. Results: The numbers of subjects who were not able to ever conceive, who were not able to conceive only after disease onset, and who were able to conceive late or only with ART were not increased among patients with BS (Table). There were more FMF patients who conceived late or only with ART. Average number of children, miscarriages, terminations and ectopic pregnancies were not significantly different in patients with BS. Univariate logistic regression showed an increased risk of infertility with CYC (OR 6.1, %95 CI 0.7–54.2), however this effect was not confirmed in multivariate analysis. This was attributed

BS major BS organ mucocutaneous FMF (nⴝ190) (nⴝ135) (nⴝ126) AS (nⴝ129) 29.2⫾8.6 30.8⫾9.4 28.7⫾11.9* 31.4⫾10.5

Healthy (nⴝ125) NA

Mean age at diagnosis Never conceived 4/156 (%2.6) 0/108 3/91 (%3.3) 3/100 (%3) 3/103 (%2.9) despite attempts Failed to conceive 7/88 (%8) 5/49 (%10.2) 6/80 (%7.5) 6/66 (%9.1) NA after disease onset Conceived late or 10/156 (%6.4) 9/108 (%8.3) 15/91 (%16.5) 6/100 (%6) 4/103 (%3.9) only with ART

p 0.012* 0.502 0.95

0.014

Conclusion: Infertility rate is not appreciably increased among BS patients as compared to FMF patients, AS patients and healthy controls. Major organ involvement does not seem to affect this. CYC is the only drug which seems to decrease fertility in BS. Disclosure: C. Saygin, None; D. Uzunaslan, None; G. Hatemi, None; K. Tascilar, None; H. Yazici, None.

2632 Long Term Outcome Of Neuro-Behçet’s Disease. Nicolas Noel1, Remy Bernard2, Bertrand Wechsler1, Matthieu Resche-Rigon2, Du Boutin1, Jean-Charles Piette1, Aure´lie Drier1, Didier Dormont1, Patrice P. Cacoub Sr.1 and David Saadoun3. 1Assistance Publique-Hoˆpitaux de Paris, Hopital Pitie´-Salpe´trie`re, Paris, France, 2Hopital Saint-Louis, Paris, France, 3Pitie´-Salpeˆtrie`re Hospital, APHP, Paris, France. Background/Purpose: Neurological involvement occurs in 5.3 to up to 59% of patients with Behçet’s disease (BD). Although the clinical and imaging features of neuro-Behçet’s disease (NBD) have been extensively described, few studies have reported on the long term outcome, treatment and prognosis of NBD by multivariate analyses. Moreover, no large study has focused on the impact of immunosuppressants on the outcome of NBD. In this study, we sought to report the long-term outcome of neurological involvement in patients with Behçet’s disease (BD). Methods: Retrospective analysis of 115 patients fulfilling the International Criteria for BD [57% male with median [Q1-Q3] age of 37 [30–46] years] with neuro-Behçet’s disease (NBD) after exclusion of cerebral venous thrombosis. Factors associated with relapses of NBD, dependence and mortality were assessed by multivariate analysis. The event-free survival was calculated using Kaplan-Meier curves and a multivariate Cox proportional hazard ratio model was performed. Results: Seventy eight (68%) patients presented with an acute onset of NBD and 37 (32%) with a progressive course. The HLA B51 allele was carried by 49% of patients. Overall, 46/115 (40%) patients had a severe initial disability status, represented by a Rankin score ⱖ 3. The 5 and 7 years event-free survival rate were of 65% and 53%, respectively. In multivariate analysis, HLA-B51 positivity was independently associated with the risk of relapses of NBD (OR⫽4.2 [1.6–10.9]). After a median follow-up of 73 [59–102] months, 21 (25.2%) patients became dependent or died. Factors independently associated with poor outcome were a paresis at onset (OR⫽6.47 [1.73–24.23]) and a brainstem location of inflammatory lesions on MRI (OR⫽8.41 [1.03–68.43]). All the 115 patients were treated with glucocorticosteroids, including 53/115 (46.1%) with cyclophosphamide and 40/115 (34.8%) with azathioprine. A trend towards a longer event-free survival was observed in severe NBD patients (i.e. Rankin score ⱖ 3 at onset) receiving intravenous cyclophosphamide compared with those treated with azathioprine (p ⫽0.06). Conclusion: NBD is a severe condition in which HLA-B51 carriers seems to have a worse prognosis. Disclosure: N. Noel, None; R. Bernard, None; B. Wechsler, None; M. RescheRigon, None; D. Boutin, None; J. C. Piette, None; A. Drier, None; D. Dormont, None; P. P. Cacoub Sr., None; D. Saadoun, None.

S1121

Tuesday, October 29

Fertility In Behçet’s Syndrome: Structured Interview In A Multidisciplinary Center. Caner Saygin, Didem Uzunaslan, Gulen Hatemi, Koray Tascilar and Hasan Yazici. Istanbul University, Cerrahpasa Medical Faculty, Rheumatology, Istanbul, Turkey.

to a type II error caused by the low number of patients who had used CYC. Finally the infertility was higher among the separate group of 62 CYC users as compared to the patients with major organ disease who were CYC naive (7/20 vs 3/60 among those who atttempted to conceive, p⫽0.002).

2633 1

Tuesday, October 29

Clinical Spectrum Of Cutaneous Vasculitis. Javier Loricera , Vanesa Calvo-Rıó1, Francisco Ortiz-Sanjuan1, Hećtor Fernańdez-Llaca1, Marcos A. Gonza´lez-Lo´pez1, Lino A´lvarez2, M. Carmen Gonza´lez-Vela1, Domingo Gonza´lez-Lamunõ2, Cristina Mata1, Javier Rueda-Gotor3, Vıćtor M. Martıńez-Taboada4, Miguel Angel Gonza´lez-Gay1 and Ricardo Blanco3. 1 Hospital Universitario Marque´s de Valdecilla. IFIMAV. Santander. Spain, Santander, Spain, 2Hospital Universitario Marque´s de Valdecilla. IFIMAV. Santander, Santander, Spain, 3Hospital Universitario Marque´s de Valdecilla. IFIMAV, Santander, Spain, 4Hospital Universitario Marque´s de Valdecilla. IFIMAV., Santander, Spain. Background/Purpose: Cutaneous vasculitis (CV) encompasses a large and heterogeneous set of syndromes characterized by inflammation of the skin vessels. The most common clinical manifestation is palpable purpura with leukocytoclastic vasculitis in biopsy. Our objective was to study clinical associations of a large series of patients with CV. Methods: Study of 817 consecutive and unselected patients with CV of the same university hospital. The diagnosis required histological confirmation except in obvious cases like Henoch-Scho¨nlein Purpura (HSP) in childhood. Primary vasculitic syndromes were classified according to the classification of Chapel Hill (Jennette J et al. Arthritis Rheum 1994; 37:187).Secondary vasculitis were considered those due to malignancies, major infections and connective tissue diseases. The different clinical associations were studied according to different age groups, considering adults patients older than 20 years (Michel et al. J Rheumatol 1992; 19:721–8). Results: Of 817 patients (457 men/360 women), 459 were adults (mean age⫾SD, 55.3⫾17.5 years) and 358 childrenyoung (mean age ⫾ SD, 7.5 ⫾ 4.2 years).Of the 358 children-young, 355 had primary VC: HSP (279 cases), cutaneous leukocytoclastic angiitis (CLA) (68) and urticarial vasculitis (8).Of the 459 adults, 369 had a primary CV: CLA (239), HSP(105), urticarial vasculitis (12) and mixed cryoglobulinemia (13 patients).The CV was a manifestation of a systemic necrotizing vasculitis in 13 adults; polyarteritis nodosa (3), microscopic polyangiitis (4), Wegener’s disease (3) and ChurgStrauss syndrome (3).In addition, the CV was secondary to other processes in 77 adults: connective tissue diseases (34), severe infection (25), neoplasia (18).There were only three children-young patients, who had a secondary CV, a major infection (2 cases) and a connective tissue disease (1 case).Patients in which the CV was secondary to systemic necrotizing vasculitis, connective, serious infections or malignancies had clinical and laboratory data suggestive of this underlying disease. Conclusion: CV in children in most cases (99.2%) is primary, usually HSP or CLA, and has a good prognosis. By contrast, in adults a 22.4% is due to malignancies, major infections, connective tissue diseases, systemic necrotizing vasculitis or cryoglobulinemic vasculitis. Therefore, clinical assotiations depends of age, being more serious in adults. Disclosure: J. Loricera, None; V. Calvo-Rıó, None; F. Ortiz-Sanjuan, None; H. Fernańdez-Llaca, None; M. A. Gonza´lez-Lo´pez, None; L. A´lvarez, None; M. C. Gonza´lez-Vela, None; D. Gonza´lez-Lamunõ, None; C. Mata, None; J. RuedaGotor, None; V. M. Martıńez-Taboada, None; M. A. Gonza´lez-Gay, None; R. Blanco, None.

2634 Henoch-Scho¨nlein Purpura: Clinical Spectrum Of The Disease In 417 Patients From A Single Center. F. Ortiz-Sanjuań1, Vanesa CalvoRıó1, Javier Loricera1, C Mata1, L Martıń Penagos1, L Alvarez1, M. Carmen Gonza´lez-Vela1, D Gonza´lez-Lamunõ1, Javier Rueda-Gotor2, Hećtor Fernańdez-Llaca1, Marcos A. Gonza´lez-Lo´pez1, Susana Armesto1, M. Enriqueta Peiro´1, M. Arias1, Miguel Angel Gonzalez-Gay1 and Ricardo Blanco2. 1 Hospital Universitario Marque´s de Valdecilla. IFIMAV. Santander. Spain, Santander, Spain, 2Hospital Universitario Marque´s de Valdecilla. IFIMAV, Santander, Spain. Background/Purpose: The severity of clinical features and the outcome in the different series of Henoch-Scho¨nlein Purpura (HSP) shows great variability, probably due to selection-bias. Our aim was to establish the actual clinical spectrum of HSP in all age groups using an unselected and wide series of patients diagnosed at a single center. Methods: Study of all consecutive patients classified as having HSP at a single center according to the criteria proposed by Michel et al (J Rheumatol 1992; 19: 721–8). HSP was pathologically confirmed in 110 cases by a skin biopsy showing the characteristic histological findings consistent with leukocytoclastic vasculitis. The remaining 307 patients without skin biopsy (mainly

children) had typical non thrombocytopenic symmetric palpable purpura. In addition, all of them fulfilled the criteria proposed by Michel et al. Results: We performed a retrospective review of 417 patients (240 men/177 women), with a median age at the time of disease diagnosis of 7.5 years (interquartile range-IQR: 5.3–20.1). Three-quarters of them (n⫽315) were children or young patients (age ⬍20 years) and a quarter (n⫽102) were adults. The most frequent precipitating events were a previous infection (38%), usually an upper respiratory tract infection, and/or drug intake (18.5%) shortly before the onset of HSP. At disease onset the most common manifestations were skin lesions (55.9%), nephropathy (24%) gastrointestinal involvement (13.7%), joint symptoms (9.1%), and fever (6.2%). When the disease was fully established, the main clinical features were skin involvement (100%), mainly palpable purpura, gastrointestinal (64.5%), joint (63.1%), renal involvement (41.2%), fever (20.4%), constitutional syndrome (2.9%) and peripheral neuropathy (1.9%). The main gastrointestinal features were the typical colicky abdominal pain (64.5%), nausea and vomiting (14.4%), melena and or rectorrhagia (12.9%), and positive stool guaiac test (10.3%). Renal involvement (41.2%), usually behaving as a mild nephropathy; nephrotic syndrome (4.8%), nephritic syndrome (2.9%) and renal insufficiency (4.8%). The main laboratory data were leukocytosis (36.7%), anemia (8.9%) and increased of serum IgA levels (31.7%). The most frequent treatments used were corticosteroids (35%), nonsteroidal anti-inflammatory drugs (14%), and cytotoxic agents (5%). After a median follow-up of 12 (IQR: 2–38) months, complete recovery was observed in 346 patients (83.2%) and persistent, usually mild, nephropathy was observed in 32 cases (7.7%). Relapses were observed in 133 patients (31.9%). Conclusion: Although HSP is a typical vasculitis affecting children and young people, it is not uncommon in adults. The prognosis is favorable in most cases depending mostly on renal involvement. Disclosure: F. Ortiz-Sanjuań, None; V. Calvo-Rıó, None; J. Loricera, None; C. Mata, None; L. Martıń Penagos, None; L. Alvarez, None; M. C. Gonza´lez-Vela, None; D. Gonza´lez-Lamunõ, None; J. Rueda-Gotor, None; H. Fernańdez-Llaca, None; M. A. Gonza´lez-Lo´pez, None; S. Armesto, None; M. E. Peiro´, None; M. Arias, None; M. A. Gonzalez-Gay, None; R. Blanco, None.

2635 IgA Vasculitis In Adults – a Rare and Benign Disease? Alojzija Hocevar1, Jaka Ostrovrsnik1, Vesna Jurcic2 and Matija Tomsic3. 1University Medical Centre Ljubljana, Ljubljana, Slovenia, 2University of Ljubljana, Faculty of Medicine, Ljubljana, Slovenia, 3University Medical Centre Ljubjana, Ljubljana, Slovenia. Background/Purpose: IgA vasculitis (IgAV) is the most common vasculitis in children, with a defined epidemiology (14 cases/ 105children), clinical picture and generally favourable prognosis. In adults IgAV is less well characterized. It is assumed to be rare, with a higher incidence of renal involvement but overall a benign disease. Our aim was to better define IgAV in our adult patient population. Methods: A retrospective chart analysis of adult patients diagnosed with IgAV at our rheumatological department between June 2010 and June 2013. The demographics, clinical data, treatment and outcome were recorded. Results: 205 new cases of vasculitis were identified (Table 1). In 71 (34.6%) IgAV was histologically confirmed, 42 males (59.2%), 29 females (40.8%), mean age 58.2 ⫾ 21.5. Prior infection was detected in 26/71 (36.6%) cases. Leukemia and primary biliary cirrhosis was coincidently diagnosed in one and two patients respectively. In 11 cases a new drug was introduced before the IgAV episode (mostly antibiotics). Skin involvement was seen in all patients (purpura 71/71 cases, skin necrosis 32/71 (45.1%), bullous lesions 7/71 (9.8%)) and there was an isolated manifestation in 15 cases (21.1%). Arthritis was present in 16/71 (22.5%) and gastrointestinal involvement in 29/71 (40.8%; severe with bleeding or ileus in 9/71 (12.6%). 43/71 (60.6%) had renal involvement. Lung, heart and testicular involvement was rare - one case of each. The most common combination of symptoms was skin, gastrointestinal and renal involvement (18/71 cases; 25.3%). No treatment (13/71) or local glucocorticoids (6/71) were sufficient in the isolated purpura. 52 patients (73.2%) received systemic glucocorticoid, 10/71 (14.1%) needed additional immunosuppression (9 cyclophosphamide (CyC), 1 mycophenolate mofetil), four received IVIG and two patients plasmapheresis in addition to CyC. None needed dialysis. Two patients died due to active vasculitis, lung and gastrointestinal manifestation respectively and one due to cytomegalovirus infection. Two patients died shortly thereafter due to heart failure and pneumonia respectively.

S1122

2637

Table 1. Newly diagnosed vasculitis from June 2010 to June 2013 TYPE

number

%

Takayasu arteritis GCA Isolated aortitis PAN Localized PAN AAV (GPA and MPA) EGPA Cryoglobulinemic vasculitis IgA vasculitis Anti-GBM disease Behçet disease Cogan syndrome SVV secondary to infection Undetermined SVV PCNSV Vasculitis secondary to RA TOTAL

1 77 3 2 3 18 3 12 71 1 5 2 3 2 1 1 205

0.5 37.6 1.5 1.0 1.5 8.8 1.5 5.9 34.6 0.5 2.4 1.0 1.5 1.0 0.5 0.5

GCA - giant cell arteritis; PAN - polyarteritis nodosa; AAV - ANCA associated vasculitis; GPA -granulomatosis with polyangiitis; MPA - microscopic polyangiitis; EGPA - eosinophilic granulomatosis with polyangiitis; SVV - small vessel vasculitis; PCNSV - primary central nervous system vasculitis; RA - rheumatoid arthritis

Conclusion: In our adult patient population IgAV is the second most common vasculitis. Adult IgAV patients frequently need systemic immunosuppression. The most common clinical manifestation is concomitant skin, gastrointestinal and renal involvement. Heart and lung involvement is rare. Disclosure: A. Hocevar, None; J. Ostrovrsnik, None; V. Jurcic, None; M. Tomsic, None.

IgA Vasculitis – The Second Most Common Systemic Vasculitis In Adults In Slovenia. Alojzija Hocevar1, ŽIga Rotar2, Jaka Ostrovrsnik1, Vesna Jurcic3, Jelka Lindic1 and Matija Tomsic4. 1University Medical Centre Ljubljana, Ljubljana, Slovenia, 2University Medical Centre Ljubljana, 1000 Ljubljana, Slovenia, 3University of Ljubljana, Faculty of Medicine, Ljubljana, Slovenia, 4University Medical Centre Ljubjana, Ljubljana, Slovenia. Background/Purpose: IgA vasculitis (IgAV) is assumed to be rare in adults. Reported annual incidence rates vary between 0.8 and 1.8 cases per 100,000 adults.1 Our aim was to determine the incidence rate of histologically confirmed IgAV in our country. Methods: A retrospective patient chart review of adult patients diagnosed with IgAV was performed at the departments of rheumatology, nephrology and dermatovenerology at an integrated secondary/tertiary university teaching hospital, which is the only hospital serving a region representing approximately a quarter of the national population. In order not to miss any cases managed by the few private dermatovenerology and nephrology outpatient clinics, the attached medical faculty’s Institute of Pathology provided a list of all histologically proven cases of IaAV in the region. The annual incidence rate of histologically confirmed IgAV was calculated. Results: From June 2010 to June 2013 48, 3, and 7 new cases of IgAV were identified at the departments of rheumatology, nephrology, and dermatovenerology, respectively in a well-defined region with a population of 517,445 white Caucasians aged 20 years or above. No additional cases were identified from records retrieved at the Institute of Pathology. The estimated annual incidence rate of IgAV is 3.7 (95% CI 2.2–5.7) per 100,000 adults. The largest proportion of cases were identified in the winter (31.0% cases) and the smallest in the spring (17.2% cases). Male to female ratio was 1.6. Conclusion: The annual incidence rate of IgAV of 3.7 per 100,000 adults may be underestimated due to retrospective case ascertainment and inclusion of histologically confirmed cases only, yet IgAV none the less represents the second most common systemic vasculitis among adults in our population, second only to giant cell arteritis with an estimated annual incidence rate of 4.7 per100,000 adults, which is in striking discord with previous reports.1 Reference: 1 Penny K, Fleming M, Kazmierczak D, Thomas A. An epidemiological study of Henoch-Scho¨nlein purpura. Paediatr Nurs. 2010;22:30–5. Disclosure: A. Hocevar, None; Rotar, None; J. Ostrovrsnik, None; V. Jurcic, None; J. Lindic, None; M. Tomsic, None.

Background/Purpose: preliminary Classification Criteria for cryoglobulinemic vasculitis (CV) have been developed in 2011 by an European cooperative study, with an adequate methodology in a large number of real cases and controls (1). The aim of this study is to validate these classification criteria for CV. Methods: Centres from Europe, United States, Japan and Egypt, were involved. A dedicated chart included: l) a validated questionnaire for CV (1); 2) the pattern of organ involvement (4 items: constitutional, articular, vascular and neurologic involvement); 3) laboratory tests (3 items: rheumatoid factor, complement C4 and serum monoclonal component), according to the preliminary criteria (1). New patients with CV (Group A) and controls (Group B), i.e., subjects with cryoglobulins but lacking CV based on the golden standard clinical judgment, were studied. A sample size of at least 140 patients for each group was estimated in order to obtain a sensitivity and a specificity of at least 90⫾5%, based on the previous results (1). Sensitivity and specificity were calculated by comparing Group A versus Group B. Finally, not for classification purposes, but to disclose whether the Criteria may be also clinically helpful in patients lacking serum cryoglobulins, but where CV is suspected (1), Group A was also compared with Group C, including patients with diseases mimicking CV, but without serum cryoglobulins. Results: Six hundred forty-three patients were enrolled in 22 Centres (from Italy, Spain, France, Greece, Slovenia, Japan and Egypt). Major

S1123

Tuesday, October 29

2636

Validation Study Of The International Classification Criteria For The Cryoglobulinemic Vasculitis. Luca Quartuccio1, Miriam Isola2, Laura Corazza3, Soledad Retamozo4, Manal Abdel-Moneim El-Menyawi5, Elisa Gremese6, Marco Sebastiani7, Nicolo Pipitone8, Teresa Urraro9, Vincenza Conteduca10, Christos Koutsianas11, Benjamin Terrier12, Mostafa Naguib Zoheir13, Alessandra Ghinoi14, Davide Filippini15, Francesco Saccardo16, Mohamed Nabil Salem17, Salvatore Scarpato18, Paolo Fraticelli19, Antonio Tavoni20, Eleonora Catarsi21, Cesare Mazzaro22, Pietro Pioltelli23, Mervat Matar5, Patrizia Scaini24, Matija Tomsic25, Norihiro Nishimoto26, Dimitrios Vassilopoulos27, Michael Voulgarelis28, Gaafar M. Ragab29, Carlo Salvarani30, Armando Gabrielli31, Patrice Cacoub32, Loic Guillevin33, Domenico Sansonno34, Anna Linda Zignego9, Gianfranco Ferraccioli6, Athanasios G. Tzioufas35, Manuel Ramos-Casals36, Clodoveo Ferri7, Maurizio Pietrogrande37, Giuseppe Monti16, Massimo Galli38, Stefano Bombardieri39 and Salvatore De Vita1. 1Rheumatology Clinic, DSMB, University of Udine, Udine, Italy, 2Institute of Statistics, DSMB, University of Udine, Udine, Italy, 3 Rheumatology Clinic, University of Udine, Udine, Italy, 4Laboratorio de Enfermedades Autoinmunes Josep Font, IDIBAPS, Hospital Clıńic, Barcelona, Spain, 5Faculty of Medicine, Cairo, Egypt, 6Division of Rheumatology, Institute of Rheumatology and Affine Sciences, Catholic University of the Sacred Heart, Rome, Italy, 7University of Modena and Reggio Emilia, Modena, Italy, 8Arcispedale S Maria Nuova, IRCCS, Reggio Emilia, Italy, 9 Center for Systemic Manifestations of Hepatitis Viruses (MASVE), Department of Experimental and Clinical Medicine, University of Florence, Florence, Italy, 10Department of Internal Medicine and Clinical Oncology, University of Bari, Bari, Italy, 11Department of Pathophysiology, Medical School of Athens, Athens, Greece, 12Cochin University Hospital, Paris, France, 13Faculty of Medicine, Cairo University, Cairo, Egypt, 14Arcispedale S Maria Nuova, Reggio Emilia, Italy, 15Rheumatology Unit, Ospedale Niguarda,, Milan, Italy, 16Internal Medicine Unit, Saronno Hospital, Azienda Ospedaliera di Busto Arsizio, Saronno (VA), Italy, 17Faculty of Medicine, Beni Swafe University, Beni Swafe, Egypt, 18Rheumatology Unit, M. Scarlato Hospital, Scafati, Salerno, Italy, 19Istituto di Clinica Medica, Universita` Politecnica delle Marche, Ancona, Italy, 20Immunoallergology Unit, Pisa, Italy, 21University of Pisa, Pisa, Italy, 22Department of Internal Medicine, Pordenone General Hospital, Pordenone, Italy, 23Hematology, S. Gerardo Hospital, Monza, Italy, 24Nephrology, Spedali Civili di Brescia, Brescia, Italy, 25University Medical Centre Ljubjana, Ljubljana, Slovenia, 26 Osaka Rheumatology Clinic, Osaka, Japan, 27Athens University School of Medicine, Athens, Greece, 28Department of Pathophysiology, Athens, Greece, 29Faculty of Medicine, Cairo University, Giza, Egypt, 30Arcispedale S Maria Nuova-IRCCS, Reggio Emilia, Italy, 31Universita` Politecnica delle Marche, Ancona, Italy, 32Hopital Pitie´-Salpe´trie`re, Paris, France, 33Division of Internal Medicine, Hoˆpital Cochin, University Paris Descartes, Paris, France, 34Section of Internal Medicine and Clinical Oncology, Department of Biomedical Sciences and Human Oncology, University of Bari, Medical School, Bari, Italy, 35School of Medicine, National University of Athens, Athens, Greece, 36Hospital Clıńic, Barcelona, Spain, 37Internal Medicine Unit, Policlinico San Marco, Bergamo, Italy, 38Istituto di Malattie Infettive e Tropicali, Universita` di Milano c/o Ospedale L. Sacco, Milano, Italy, 39 Rheumatology Unit, University of Pisa, Pisa, Italy.

organizative/local issues did not allow American experts to participate. Group A comprised 268 patients with CV, Group B 182 controls with serum cryoglobulins without CV, and Group C 193 controls without serum cryoglobulins. Notably, 20 patients showed type I cryoglobulinemia, 13 in Group A, and 7 in Group B. Group C included 108/193 (55.9%) systemic vasculitides, 100/108 (92.6%) were small vessel vasculitides. The classification criteria [positivity of at least 2/3 items among questionnaire (ⱖ2/3 positive questions), clinical item (ⱖ3/4 clinical manifestations), laboratory (ⱖ2/3 tests)] showed 89.9% (95% CI 86.1–93.6) of sensitivity and 93.5% (95% CI 89.7–97.2) of specificity, replicating previous results (1). Sensitivity of 91.7% and specificity of 100% were observed in the subgroup of type I cryoglobulinemia. By the comparison of Group A vs. Group C, the Criteria showed a specificity 92.6% (88.8–96.5) and a sensitivity of 77.8% (72.6– 83.0) when the laboratory item was positive (questionnaire ⫹ laboratory item; or clinical ⫹ laboratory item). Conclusion: the International Classification Criteria for the CV have been validated in a new real cohort. High specificity and sensitivity were confirmed. Notably, in patients where CV is suspected on clinical grounds, but where cryoglobulins are negative by initial testing, or not yet available (patients who cannot be classified as CV, as positive serum cryoglobulinemia is a conditio sine qua non for classification) (1), the Criteria appear relevant to strengthen the suspicion for CV, and to optimize the follow-up.

Tuesday, October 29

1. De Vita S, et al. Ann Rheum Dis 2011;70(7):1183–90. Disclosure: L. Quartuccio, None; M. Isola, None; L. Corazza, None; S. Retamozo, None; M. A. M. El-Menyawi, None; E. Gremese, None; M. Sebastiani, None; N. Pipitone, None; T. Urraro, None; V. Conteduca, None; C. Koutsianas, None; B. Terrier, None; M. N. Zoheir, None; A. Ghinoi, None; D. Filippini, None; F. Saccardo, None; M. N. Salem, None; S. Scarpato, None; P. Fraticelli, None; A. Tavoni, None; E. Catarsi, None; C. Mazzaro, None; P. Pioltelli, None; M. Matar, None; P. Scaini, None; M. Tomsic, None; N. Nishimoto, Bristol-Myers Squibb Japan, 2; D. Vassilopoulos, None; M. Voulgarelis, None; G. M. Ragab, None; C. Salvarani, None; A. Gabrielli, None; P. Cacoub, None; L. Guillevin, None; D. Sansonno, None; A. L. Zignego, None; G. Ferraccioli, None; A. G. Tzioufas, None; M. Ramos-Casals, None; C. Ferri, None; M. Pietrogrande, None; G. Monti, None; M. Galli, None; S. Bombardieri, None; S. De Vita, None.

2638 Cryoglobulinemia Vasculitis With Or Without Associated Cryofibrinogenemia: A Different Phenotype? Martin Michaud1, Guillaume Moulis2, Jacques Pourrat1, Benedicte Puissant1, Antoine Blancher1 and Laurent Sailler3. 1Toulouse University Hospital, University of Toulouse, Toulouse, France, 2Toulouse University Hospital, Clinical Pharmacology Department, University of Toulouse, UMR INSERM-UPS 1027, Toulouse, France, 3 Toulouse University Hospital, Toulouse, France. Background/Purpose: Cryoglobulin (CryoGl) and cryofibrinogen (CryoFg) are cryoproteins, both responsible for well-described systemic vasculitis. Nevertheless, cryoglobulinemia may be associated with cryofibrinogenemia. Such mixed cryoprotein vasculitis (MCV) has not been properly described yet. This work aimed at determining whether the phenotype of MCV is different from CryoGl vasculitis. Methods: Concomitant CryoGl and CryoFg dosages performed at our University Hospital from January 2011 to December 2012 were extracted and the corresponding medical files reviewed. We included all adult patients with systemic vasculitis associated with CryoGl with or without CryoFg. As in the CryoVas study, systemic vasculitis was defined by purpura or cutaneous ulcers, or other symptoms biopsy-proven associated with a detectable cryoprotein. Qualitative variables were compared with Fisher or chi2 tests and quantitative variables with Student t-test or Wilcoxon-Mann-Withney test. Results: Among 107 patients with cryoprotein positive dosage, 21 had MCV and 16 CryoGl vasculitis. Male:female sex ratio was respectively 1/2 and 5/3 (p⫽0.8). Mean age at diagnosis was respectively 58 ⫾ 20 and 59 ⫾ 16 years (p⫽0.9). Cutaneous manifestations (purpura, skin necrosis or ulceration and acrocyanosis) were similar in both groups as well as kidney (38 vs 31%) and peripheral nerve involvement (28 vs 32%). The frequency of rheumatic symptoms (arthralgia and/or myalgia) or of venous/arterial thrombotic events (43 vs 25%, p⫽0.4) was not different betweens both groups (51 vs 32%, p⫽0.2). Estimated glomerular filtration rate and cryoglobulin dosage were similar in both groups. As expected, ␣1 and ␣2globulin levels were higher in the MCV group (p⬍0.05). Decreased values of C3 and C4 were more frequent in the MCV group albeit not significantly (respectively, 29 vs 12% and 53 vs 35%). Regarding associated diseases, neoplasms tended to be 3 times more frequent in the MCV group (43 vs 12%, p⫽0.07) whereas the frequency of auto-immune disorders was similar in both groups (38% vs 50%,

p⫽0.5). The use of systemic corticosteroids was more frequent in the MCV group (67 vs 31%, p⫽0.03), as well as exposure to immunosuppressives (47 vs 18%, p⫽0.07). Conclusion: Cryofibrinogemnemia is associated with a more serious disease in cryoglobulinemic patients as they are associated with an increased use of corticosteroids and immunosuppressive drugs. Neoplasms should also be carefully searched in this group. Disclosure: M. Michaud, None; G. Moulis, None; J. Pourrat, None; B. Puissant, None; A. Blancher, None; L. Sailler, None.

2639 Long-Term Outcome of Monoclonal (type 1) Cryoglobulinemia. Antoine Neél1, François Perrin1, Olivier Decaux2, Thomas Dejoie1, Maxime Halliez1, Beátrice Mahe´1, Thierry Lamy2, Fadi Fakhouri1, Patrick Jego2, Christian Agard1, Cećile Vigneau2, Lucienne Guenet2, Bernard Grosbois2, Philippe Moreau1 and Mohamed Hamidou1. 1Nantes University Hospital, Nantes, France, 2Rennes University Hospital, Rennes, France. Background/Purpose: To investigate long-term outcome of symptomatic type 1 cryoglobulinemia and its determinants. Methods: Retrospective analysis of a prospective cohort from 2 French university hospitals. Patients with type 1 cryoglobulinemia were identified using laboratory databases. Inclusion criterion was the presence of persistent symptoms of cryoglobulinemia. Results: Among 227 screened patients, 36 were included. Skin or vasomotor symptoms were the most frequent features (75%). Nephropathy and neuropathy occurred in 30% and 47% of cases, respectively. The underlying B cell lymphoproliferative disorder (LPD) was a gammopathy of unknown significance (MGUS) in 13 (36%) and a hematologic malignancy (HM) in 23 (64%; Waldenstrom macroglobulinemia (WM) in 12, low grade non-Hodgkin lymphoma (NHL) in 6, multiple myeloma (MM) in 4, and chronic lymphocytic leukemia (CLL) in 1). Severe manifestations affected half the patients and were more frequent with IgG cryoglobulins (82 vs 30% (p⫽0,006)). At last follow-up, 64% of patients had suffered no hematologic manifestation. Potent chemotherapeutic regimens were mainly used in HM. For patients with MGUS, WM or NHL, fludarabine or rituximab-based regimens appeared to yield better therapeutic responses. Five-year actuarial survival rate was 82 %. Older age and hemoglobin level ⬍ 12g/dL at diagnosis correlated with a poorer survival (p⬍0.05, Log-rank test). Nephropathy, infections, Richter’s transformation and second malignancies were important sources of morbi-mortality. Conclusion: Despite its limitations, this series provide novel information regarding type 1 cryoglobulinemia. Further studies are needed to improve its management. To date, therapeutic strategy should be tailored according to patient’s characteristics (age, comorbidities, underlying LPD), and therapeutic target. Disclosure: A. Neél, None; F. Perrin, None; O. Decaux, None; T. Dejoie, None; M. Halliez, None; B. Mahe´, None; T. Lamy, None; F. Fakhouri, None; P. Jego, None; C. Agard, None; C. Vigneau, None; L. Guenet, None; B. Grosbois, None; P. Moreau, None; M. Hamidou, None.

2640 VERY LONG-TERM Effects Of The “4 PLUS2 Infusion PROTOCOL” Of Rituximabalone In Patients With HCV-Associatedmixed Cryoglobulinemia With Diffusemembranoproliferative Glomerulonephritis, Severe Polyneuropathy and Necrotic Ulcersof Skin. Dario Roccatello1, Savino Sciascia1, Simone Baldovino2 and Daniela Rossi1. 1Centro di Immunopatologia e Documentazione su Malattie rare, Torino, Italy, 2Centro di Ricerche di Immunologia Clinica ed Immunopatologia e Documentazione su Malattie Rare (CMID), Universita` di Torino, Torino, Italy. Background/Purpose: Mixed cryoglobulinemia syndrome (MCs) is a systemic vasculitis characterized by multiple organ involvement due to the vascular deposition of IgMk/IgG cryoglobulins. B cells expansion usually triggered by HCV infection plays a central role in MCs. Objectives: to evaluate the long term effects of B-cells depletion in MCs Methods: Twenty seven patients, (mean age 60.2 [range 35–78] years, HCV infection in 96% of cases) with symptomatic type-II MCs with systemic manifestations, including renal involvement (diffuse membranoproliferative glomerulonephritis in 15 cases), peripheral neuropathy (26 cases) and large skin ulcers (9 case, in 7 necrotizing) were considered

S1124

eligible for Rituximab (RTX) therapy. RTX was administered at a dose of 375 mg/m2 on days 1, 8, 15 and 22. Two more doses were administered 1 and 2 months later. No other immunosuppressive drugs were added. Response was evaluated by assessing the changes in clinical signs, symptoms, laboratory parameters and electromyographic indices for a very long term follow-up (mean 54.3 months [12–96]) Results: Complete remission of pre-treatment active manifestations was observed in all the cases of skin purpuric lesions and nonhealing vasculitic leg ulcers, and in 80% of cases of peripheral neuropathy. A significant improvement in the clinical neuropathy disability score was observed. Electromyography revealed that the amplitude of compound motor action potential had increased. Cryoglobulinemic glomerulonephritis, observed in 15 patients, significantly improved during the follow-up starting from the second month after RTX (serum creatinine from 2.2⫾1.9SD to 1.6⫾1.2SD mg/dl, pⱕ.05; 24-hour proteinuria from 2.3⫾2.1SD to 0.9⫾1.9SD g/24h, pⱕ.05). Significant improvement of serological hallmarks, such as cryocrit and low complement C4, were also detected (pⱕ.05). The safety of RTX was confirmed by the absence of side effects recorded during the mean 54-month follow-up. Reinduction was performed in 9 relapsed cases (after a mean of 31.1 months, range 12–54) with resolutive beneficial effects. Conclusion: In this open prospective study, the “4 plus 2 infusion protocol” of RTX appeared to be very effective and safe in the treatment of patients with MCs-associated membranoproliferative nephritis, polyneuropathy and severe skin involvement Disclosure: D. Roccatello, None; S. Sciascia, None; S. Baldovino, None; D. Rossi, None.

2641

Background/Purpose: Compared to rheumatoid arthritis and other arthritis-associated rheumatic diseases, autoimmune connective tissue disorders and vasculitides are still in the early stages of biological therapies. The study investigated whether health care and treatment with disease specific immunosuppressive drugs have changed in these indications. Furthermore, disease activity and outcomes of patients with autoimmune connective tissue diseases and vasculitides have been examined. Methods: On average, a total of 3,900 patients with systemic lupus erythematodes (SLE), systemic sclerosis (SSc), primary Sjo¨gren‘s syndrome (PSS), poly- and dermatomyositis (PM/DM), mixed connective tissue disease (MCTD), other connective tissue diseases (other CTD), polymyalgia rheumatica (PMR), primary systemic vasculitis, and other primary vasculitides (ICD10: I67.7, I77.6, L13.1, L95.0/8/9, M31.0/1/8/9, M35.2/6) were documented in the National Database of the German Collaborative Arthritis Centers in each of the years 1995 to 2011. Cross-sectional data of these years were analyzed to detect time trends. Results: Between 1995 and 2011, the number of patients consulting a rheumatologist within six months after symptom onset increased from about 50% to 60%, especially patients with SLE and systemic sclerosis were seen earlier in specialized rheumatologic care. The overall percentage of patients receiving immunosuppressive drugs increased from 52.7% to 62.3%. High dose steroid use (⬎7.5 mg/d) decreased from 26.1% to 13.4%. In 2011, 3.5% of all patients received a biologic therapy, the most frequent use was seen in patients with other primary vasculitides (9.2%). Since 2005, the mean disease activity assessed by the physician (NRS 0–10) decreased from 3.0 to 2.2 and the percentage of patients with a high disease activity (7–10) from 7.6% to 1.8%. Overall, work participation (patients ⬍65 years) remained stable (46.2% 1995, 49.4% 2011), but increased considerably in female patients with PSS, SSc and PMR (see Table 1).

Mean age

N

Female

Immunosuppressive therapy

female

male

1995

2011

1995

2011

1995

2011

1995

2011

1995

2011

1995

2011

1995

2011

1,432

946

43.1

46.3

89%

89%

69.6%

81.4%

3.0

2.2

45.6%

46.4%

55.3%

52.4%

SSc

317

317

55.7

57.5

86%

84%

42.9%

48.5%

3.8

3.1

36.5%

44.0%

–

–

PSS

320

200

53.1

57.9

93%

94%

43.7%

51.0%

3.2

2.5

44.2%

64.8%

–

–

DM/PM

143

94

50.7

54.9

76%

65%

65.9%

86.0%

3.1

2.4

43.2%

(41.7%)

–

–

MCTD/ overlap

329

165

47.4

50.0

89%

90%

68.8%

71.7%

3.4

2.4

47.0%

47.5%

–

–

other CTD

343

261

48.1

53.5

85%

86%

52.5%

54.5%

2.6

2.2

50.5%

63.0%

–

–

PMR

862

855

68.7

71.4

77%

67%

15.2%

28.6%

3.1

1.7

25.8%

40.0%

59.6%

60.0%

primary systemic vasculitis

292

435

50.6

60.3

52%

61%

72.0%

78.5%

3.1

2.3

31.1%

29.0%

63.0%

52.9%

other primary vasculitides

202

256

47.0

45.9

64%

45%

46.0%

57.6%

3.3

2.5

41.9%

(50.0%)

69.2%

(76.1%)

SLE

#For case numbers ⬍50, percentages are given in brackets, for male patients with SSc, PSS, DM/PM, MCTD and other CTD case numbers were to low in both years for valid comparison. Conclusion: Early rheumatologic care and immunosuppressive therapy have increased in patients with autoimmune connective tissue diseases and vasculitides during the past 15 years. Advancement in disease control and employment status in these rather infrequent chronic autoimmune diseases encourages to further improve disease outcomes and work participation.

Disclosure: D. Huscher, None; K. Albrecht, None; K. Thiele, None; S. Bischoff, None; J. G. Richter, None; I. Ko¨tter, None; W. Ochs, None; A. Zink, None.

2642 A Novel Genetic Basis For Systemic Vasculitis: Polyarteritis Nodosa Caused By Recessive Mutations In An Immune-Related Gene. Reeval Segel1, Pnina Elkan-Navon2, Sarah B. Pierce3, Tom Walsh3, Judith Barash4, Shay Padeh5, Avraham Zlotogorski6, Yackov Berkun7, Isabel Voth8, Philip Hashkes2, Liora Harel9, Eduard Ling10, Fatos Yalcinkaya11, Ozgur Kasapcopur11, Paul F. Renbaum2, Ariella Weinberg-Shukron2, Barbara Schormair12, Mordechai Shohat13, Alan A. Rubinow14, Elon Pras5, Juliane Winkelmann15, Mustafa Tekin16, Yair Anikster5, Mary-Claire King3 and Ephrat Levy-Lahad2. 1Hebrew University Medical School, Jerusalem, Israel, 2 Shaare Zedek Medical Center, Jerusalem, Israel, 3University of Washington, Seattle, WA, 4Kaplan Medical Center, Rehovot, Israel, 5Sheba Medical Center, Ramat Gan, Israel, 6Hadassah-Hebrew University Medical Center, Jerusalem, Israel, 7Hadassah Medical Center, Mount Scopus, Jerusalem, Israel, 8Technische Universitat Munchen Klinikum rechts der Isar, Munich, Germany, 9Schneider Children’s Medical Center, Tel Aviv University, Petach Tikvah, Israel, 10Soroka University Medical Center and Ben-Gurion University, Beer-Sheva, Beer Sheva, Israel, 11Ankara University, Ankara, Turkey, 12Helmholtz Zentrum Munchen, Munich, Germany, 13Rabin Medical Center, Beilinson Hospital, Petach Tikvah, Israel, 14Hadassah Medical Center, Jerusalem, Israel, 15Stanford University, San Francisco, CA, 16Miller School of Medicine, University of Miami, Miami, FL. Background/Purpose: Polyarteritis nodosa (PAN) is a systemic necrotizing vasculitis. Disease pathogenesis and possible genetic factors are poorly understood. We identified familial, mostly pediatric PAN, in twenty patients from six families, five of Georgian Jewish and one of German origin. Disease segregation was consistent with autosomal recessive inheritance. Methods: Clinical data was collected from eight medical centers in Israel, Germany and Turkey. Exome sequencing was performed in six patients with familial PAN, four Georgian Jewish and two Germans. Targeted sequencing was done in another thirteen patients with familial PAN and in seventeen single cases: three of Gerogian Jewish and fourteen of Turkish origin. Mutations were assessed by impact on enzymatic activity in patient sera, analysis of protein structure, expression in mammalian cells, and biophysical analysis of purified protein. Results: In most patients, disease onset was in early childhood and was most severe in those with onset in infancy. Skin manifestations almost universal: livedo reticularis, nodules, recalcitrant leg ulcers and acral necrosis. Raynaud’s phenomenon, myalgia, arthralgia or arthritis were also common. Systemic involvement included gastrointestinal disease with significant weight loss and intestinal perforation, testicular pain, coronary and renal vasculitis with severe hypertension, peripheral neuropathy and brain strokes. Angiography demonstrated mesenteric, celiac, hepatic, or renal aneurysms, and infarcts in the renal cortex. There was a dramatically beneficial response to anti-TNF alpha blockade; life-saving in some cyclophosphamide refractory patients. All patients had recessive mutations in an immune-related gene, to be named at the meeting. Mutations in this gene have not been previously associated with any human phenotype. All Georgian Jewish cases were homozygous for this mutation, with variable phenotype; The German and the Turkish patients

S1125

Tuesday, October 29

Earlier Admission To Specialized Care, Intensified Treatment and Improved Outcome In Patients With Autoimmune Connective Tissue Disorders and Vasculitides In Germany 2011 Compared To 1995. Do¨rte Huscher1, Katinka Albrecht2, Katja Thiele2, Sascha Bischoff2, Jutta G. Richter3, Ina Ko¨tter4, Wolfgang Ochs5 and Angela Zink6. 1Charite´ University Hospital and German Rheumatism Research Centre, a Leibniz Institute, Berlin, Germany, 2German Rheumatism Research Centre, Berlin, Germany, 3 Heinrich-Heine-University, Duesseldorf, Germany, 4ZIRS, Centre for Interdisciplinary Rheumatology Stuttgart, Stuttgart, Germany, 5Rheumatology Practice Bayreuth, Bayreuth, Germany, 6German Rheumatism Research Centre and Charite´ University Medicine, Berlin, Germany.

Work participation (only patients 10 pack-years vs.< 10 pack-years) BMI (>25 kg/m2 [overweight]) vs. 10 pack-years vs. < 10 pack-years) RERI* ROR** IL6 IL6- per one unit increase Interaction between IL6 and BMI (overweight vs. normal) RERI* ROR** Interaction between IL6 and Smoking (> 10 pack-years vs. < 10 pack-years) RERI* ROR**

Odds Ratio (95%CI) 1.78 (1.25, 2.52) 1.55 (1.11, 2.17) 1.42 (0.78, 2.56)

0.24 (⫺0.37, 0.85) 4.11 (1.29, 13.09)

⫺0.11 (⫺2.95, 2.73) 1.65 (0.52, 5.29) 1.32 (1.04, 1.68) 0.24 (⫺0.48, 0.97) 1.08 (0.68, 1.74)

0.89 (⫺0.17, 1.95) 1.43 (0.88, 2.33)

TNFRII and IL6 were log-transformed in all models. * RERI⫽additive interaction, relative excess risk due to interaction ** ROR⫽ multiplicative interaction, ratios of odds ratios

Conclusion: In this large prospective cohort study of women with blood collected up to 14 years prior to the diagnosis of RA, a multiplicative interaction was found between high BMI and elevated TNFRII level in increasing the risk of RA. This suggests that BMI and TNF␣ elevation may belong to the same biologic pathway in RA pathogenesis. Disclosure: E. V. Arkema, None; S. Malspeis, None; B. Lu, None; L. T. Hiraki, None; E. W. Karlson, None; K. H. Costenbader, None.

S1132

2658

2659

Dipeptidyl Peptidase-4 Inhibitors In Type 2 Diabetes May Reduce The Risk Of Autoimmune Diseases. Seoyoung C. Kim1, Sebastian Schneeweiss1, Robert J. Glynn1, Michael Doherty1, Allison Goldfine2 and Daniel H. Solomon3. 1Brigham and Women’s Hospital, Boston, MA, 2Joslin Diabetes Center, Boston, MA, 3Harvard Medical School, Brigham and Women’s Hospital, Division of Rheumatology, Division of Pharmacoepidemiology, Boston, MA.

Antibodies To Citrullinated Enolase, Fibrinogen, and Vimentin Are Associated With Markers Of Endothelial Dysfunction In First-Degree Relatives Of Patients With Rheumatoid Arthritis: The Studies Of The Etiology Of Rheumatoid Arthritis. Jan M. Hughes-Austin1, Kendra A. Young2, Kevin D. Deane3, Michael H. Weisman4, Jane H. Buckner5, Ted R. Mikuls6, James R. O’Dell7, Richard M. Keating8, Peter K. Gregersen9, Jeremy Sokolove10, William H. Robinson11, V. Michael Holers3 and Jill M. Norris2. 1University of California, San Diego, La Jolla, CA, 2Colorado School of Public Health, Aurora, CO, 3University of Colorado School of Medicine, Aurora, CO, 4Cedars-Sinai Med Ctr, Los Angeles, CA, 5Benaroya Research Institute, Seattle, WA, 6Omaha VA and University of Nebraska Medical Center, Omaha, NE, 7University of Nebraska Medical Center, Omaha, NE, 8The University of Chicago, Chicago, IL, 9Feinstein Institute for Medical Research, Manhasset, NY, 10VA Palo Alto Health Care System and Stanford University, Palo Alto, CA, 11Stanford University School of Medicine, Stanford, CA.

Table. Risk of autoimmune diseases associated with type 2 DM treatments: PS-matched ‘as treated’ analysis DPP4i (nⴝ58,275)

Non-DPP4i (nⴝ58,275) Personyears IR* (PY) (95% CI)

HR

Outcome

Cases

Personyears (PY)

HR

Cases

RA Other autoimmune diseases Composite: RA or other autoimmune diseases

47 59

39,379 39,361

1.19 (0.88–1.59) 1.50 (1.14–1.93)

0.64 (0.45–0.91) 0.53 (0.39–0.73)

87 129

48,665 48,615

1.79 (1.43–2.21) 2.65 (2.22–3.15)

Ref Ref

105

39,325

2.67 (2.18–3.23)

0.57 (0.45–0.73)

214

48,539

4.41 (3.84–5.04)

Ref

IR* (95% CI)

*Per 1,000 PY The logistic model for PS includes age, sex, comorbidities, smoking, obesity, non-DM medications, number of DM meds, number of primary and specialist visits, and other health care utilization.

Conclusion: In this large cohort of T2DM patients, initiating DPP4i combination therapy was associated with a decreased risk of incident RA or other AD compared to those initiating non-DPP4i combination therapy. These results suggest possible pharmacologic pathways for reducing the incidence of AD. Disclosure: S. C. Kim, Pfizer Inc, 2, Pfizer and Asisa, 9; S. Schneeweiss, Pfizer Inc,, 2, Novartis Pharmaceutical Corporation, 2, Boehringer Ingelheim, 2, WHISCON, LLC, 5, BOOZ and Company, 5; R. J. Glynn, AstraZeneca, 2, Novartis Pharmaceutical Corporation, 2; M. Doherty, None; A. Goldfine, None; D. H. Solomon, Lilly, Amgen, CORRONA, 2, Lilly, Novartis, BMS, Pfizer, 6, Lilly, BMS, Novartis, 9.

Background/Purpose: Among rheumatoid arthritis (RA) patients, antibodies to citrullinated protein antigens (ACPA) have been detected in atherosclerotic plaques. In particular, antibodies to citrullinated fibrinogen (cit-fib) have been associated with prevalent coronary artery calcium (Sokolove, Arth Rheum 2013). In the absence of RA, ACPA have been associated with cardiovascular disease (CVD) (Cambridge, Atherosclerosis 2013), suggesting that ACPA themselves may directly contribute to accelerated CVD recognized in RA. Therefore, we hypothesized that among RA-free first-degree relatives (FDRs) of patients with RA, ACPA are associated with markers of endothelial dysfunction, an indicator of initial vascular injury in CVD. Methods: From the Studies of the Etiology of RA (SERA) (a multicenter prospective study of preclinical RA), 113 FDRs who had been positive for any of 5 RA-related antibodies (Abs): rheumatoid factor (RF), RF isotypes – IgM, IgG, IgA, or anti-cyclic citrullinated peptide (anti-CCP2) on at least one of their visits, and 100 FDRs who had never been Ab positive were selected, frequency matched on age, sex, and field center site. In cross-sectional testing of single samples obtained at the baseline exam, the following were measured: endothelial activation/injury markers: soluble intracellular adhesion molecule-1 (sICAM), soluble vascular cell adhesion molecule-1 (sVCAM) and E-selectin; and a panel of 15 ACPA (using Bio-Plex bead-based assay). ACPA were dichotomized as positive/negative based on cut-offs in 200 RA patients and 98 blood bank controls, and were developed using receiver operating characteristic (ROC) curves giving ⬎90% specificity. Soluble VCAM was analyzed as a continuous variable; sICAM and E-selectin were log-transformed to approximate normality and were analyzed in log units. ANCOVA was used to evaluate associations between ACPA positivity and markers of endothelial dysfunction, adjusting for age, sex, race, body mass index, pack-years of smoking, and high sensitivity C-reactive protein. Results: Among 193 FDRs with complete data and presented in detail in Table 1, sICAM was significantly higher in FDRs who were positive for ACPA to cit-enolase compared with those who were negative, with anti-citvimentin showing similar magnitude and marginal significance. sVCAM was significantly higher in those who were positive for antibodies to cit-fibrinogen and cit-vimentin. No significant associations were observed between ACPA and E-selectin. Table. 1. Proportion of SERA FDRs with ACPA and associations of individual ACPA with markers of endothelial dysfunction n (%) positive nⴝ193 Apolipo E (277–296) cit2 cyclic Biglycan (247–266) cit cyclic Clusterin (221–240) cit Enolase (5–21) cit Fibrinogen A (211–230) cit cyclic Fibrinogen A (41–60) cit3 cyclic Fibrinogen A (556–575) cit cyclic Fibrinogen A (616–635) cit3 cyclic Fibrinogen A cit Filaggrin (48–65) cit2 cyclic Histone 2A (1–20) cit cyclic Histone 2B (62–81) cit cyclic Histone 2B–cit Vimentin cit Vimentin (58–77) cit3 cyclic

33 (17)

Log ICAM B p 0.0006 (0.09) 0.99

VCAM B 37.29 (44.8)

p 0.41

27 (14) 25 (13) 19 (10) 32 (17)

0.04 (0.10) ⫺0.03 (0.10) 0.24 (0.11) ⫺0.04 (0.09)

75.11 (48.9) 62.49 (50.0) 79.73 (56.9) 39.85 (45.9)

0.13 0.21 0.16 0.39

⫺0.03 (0.11) ⫺0.06 (0.11) 0.18 (0.12) ⫺0.002 (0.10)

30 (16)

⫺0.02 (0.09)

0.84

75.10 (47.0)

0.11

⫺0.004 (0.10)

0.97

32 (17)

⫺0.04 (0.09)

0.70

89.49 (45.7)

0.05

⫺0.08 (0.10)

0.41

0.68 0.76 0.03 0.69

Log E-selectin B p ⫺0.04 (0.10) 0.65

21 (11)

0.06 (0.11)

0.60

130.59 (53.2)

0.02

0.0007 (0.12)

1.00

25 (13) 26 (14) 40 (21) 32 (17) 8 (4) 10 (5) 8 (4)

⫺0.02 (0.10) 0.02 (0.10) ⫺0.0003 (0.08) ⫺0.07 (0.09) 0.04 (0.17) ⫺0.12 (0.15) 0.32 (0.17)

0.86 0.83 1.00 0.41 0.80 0.43 0.05

15.96 (51.0) 76.05 (49.0) 55.68 (41.9) 67.76 (45.2) 116.93 (84) 16.20 (76.9) 204.17 (83.4)

0.75 0.12 0.19 0.14 0.17 0.83 0.02

⫺0.03 (0.11) ⫺0.01 (0.11) ⫺0.03 (0.09) 0.02 (0.10) 0.22 (0.18) 0.26 (0.17) 0.31 (0.18)

0.77 0.91 0.76 0.87 0.22 0.12 0.09

*adjusted for age, sex, race, BMI, pack-years of smoking and high sensitivity C-reactive protein

S1133

0.75 0.57 0.15 0.98

Tuesday, October 29

Background/Purpose: Dipeptidyl peptidase-4 inhibitors (DPP4i), such as linagliptin, saxagliptin, and sitagliptin, are oral glucose-lowering drugs for type 2 diabetes mellitus (T2DM). DPP4 is a transmembrane glycoprotein widely expressed in various cells including fibroblasts, T lymphocytes, and macrophages, and has a co-stimulatory function in the immune response. Altered levels of DPP4 activity were noted in patients with autoimmune diseases (AD) such as rheumatoid arthritis (RA), systemic lupus erythematosus (SLE), inflammatory bowel disease (IBD), psoriasis, and multiple sclerosis (MS). The objective of this study was to estimate the incidence rate (IR) of systemic AD such as RA, SLE, psoriasis, MS, and IBD in patients with T2DM initiating a DPP4i drug compared to those initiating non-DPP4i oral hypoglycemic agents. Methods: We conducted a population-based cohort study using commercial insurance claims data (2005–2011). Among patients aged ⱖ40 years with T2DM, two mutually exclusive exposure groups were selected: 1) DPP4i combination therapy (DPP4i and at least 1 other oral non-DPP4i drugs) and 2) non-DPP4i combination therapy (2 or more oral non-DPP4i drugs). Patients with a diagnosis of systemic AD, HIV, and cancer, and use of insulin-containing drugs or immunosuppressive drugs at baseline were excluded. Incidence rates (IRs) were calculated for RA, other AD and composite AD. RA and other AD were defined with ⱖ2 diagnosis codes that were ⱖ7 days apart and ⱖ1 prescription for disease-specific immunosuppressive drugs or steroids. Propensity score (PS)-matched Cox regression models compared the risk of RA or AD in DPP4i initiators compared to non-DPP4i initiators, controlling for baseline demographic factors, comorbidities, medications, and health care utilization. Sensitivity analyses matched on PS compared DPP4i combination therapy initiators separately to sulfonylurea combination therapy initiators and to thiazolidinediones combination therapy initiators. Results: We included 58,275 patients starting DPP4i combination therapy 1:1 PS-matched to those starting non-DPP4i combination therapy. Risks of RA and other AD were significantly lower in the DPP4i group vs. non-DPP4i with the HR of 0.64 (95% CI 0.45–0.91) for RA, 0.53 (95% CI 0.39–0.73) for other AD, and 0.57 (95% CI 0.45–0.73) for composite AD (Table). In sensitivity analysis, the risk of other AD and composite AD was significantly reduced in initiators of DPP4i combination therapy compared to sulfonylurea combination therapy and thiazolidinediones combination therapy, but the risk of RA was not.

Conclusion: In FDRs without classified articular RA, sICAM and sVCAM were higher in those who were ACPA positive. While cit-fib and cit-vimentin have been implicated in subclinical CVD in RA, and antibodies to a-enolase have been associated with vasculitis, our findings support the potential role of these ACPA in the initial insult in the development of CVD, even in the absence of clinical RA. Disclosure: J. M. Hughes-Austin, None; K. A. Young, None; K. D. Deane, None; M. H. Weisman, ACR/EULAR, 2, ACR, 2, Rigel, 2, SanofiAventis, 2, NIH, 2, FDA, 2, Cedars-Sinai Medical Center, 3, UCB, 5; J. H. Buckner, None; T. R. Mikuls, None; J. R. O’Dell, None; R. M. Keating, None; P. K. Gregersen, None; J. Sokolove, None; W. H. Robinson, None; V. M. Holers, None; J. M. Norris, None.

Conclusion: CCP2 positivity is inversely associated with self-reported n-3 supplement use and increasing levels of erythrocyte membrane n-3 fatty acid concentrations. These data suggest intake of longer chain n-3 fatty acids (EPA⫹DHA) may reduce the likelihood of developing anti-CCP2 antibodies, and potentially, the future development of RA. Only a small percentage of the association between n-3 supplement use and CCP2 positivity is explained by increasing levels of membrane n-3 fatty acids, suggesting other beneficial mechanisms, yet to be explored. Disclosure: R. W. Gan, None; K. A. Young, None; G. O. Zerbe, None; M. K. Demoruelle, None; M. H. Weisman, None; J. H. Buckner, None; P. K. Gregersen, None; T. R. Mikuls, None; J. R. O’Dell, None; R. M. Keating, None; M. J. Clare-Salzler, None; K. D. Deane, None; V. M. Holers, None; J. M. Norris, None.

2660

Tuesday, October 29

Omega-3 Supplement Use Is Associated With a Reduced Risk Of AntiCyclic Citrullinated Protein Positivity In a Population Without Rheumatoid Arthritis, But At Risk For Future Disease. Ryan W. Gan1, Kendra A. Young1, Gary O. Zerbe2, M. Kristen Demoruelle3, Michael H. Weisman4, Jane H. Buckner5, Peter K. Gregersen6, Ted R. Mikuls7, James R. O’Dell8, Richard M. Keating9, Michael J. Clare-Salzler10, Kevin D. Deane3, V. Michael Holers3 and Jill M. Norris1. 1Colorado School of Public Health, Aurora, CO, 2Colorado School of Public Health / University of Colorado Anschutz Medical Campus, Aurora, CO, 3University of Colorado School of Medicine, Aurora, CO, 4CedarsSinai Medical Center, Los Angeles, CA, 5Benaroya Research Institute, Seattle, WA, 6Feinstein Institute for Medical Research, Manhasset, NY, 7Omaha VA and University of Nebraska Medical Center, Omaha, NE, 8University of Nebraska Medical Center, Omaha, NE, 9The University of Chicago, Chicago, IL, 10University of Florida, College of Medicine, Gainesville, FL. Background/Purpose: Rheumatoid arthritis (RA) is characterized by systemic inflammation and circulating autoantibodies, which can be present in serum years prior to RA diagnosis and can define higher risk for future development of classified disease. In particular, the anti-cyclic citrullinated protein (anti-CCP) autoantibody is specific for RA, and its presence is strongly associated with future disease onset. While dietary intake of anti-inflammatory omega-3 (n-3) fatty acids may protect against development of RA and decrease the need for drugs to treat RA symptoms, the exact roles that n-3 fatty acids play prior to disease onset is unknown. We investigated the relationship between anti-CCP positivity and n-3 fatty acids, as measured by n-3 supplement use and erythrocyte membrane levels, in a population without RA, but increased risk for future RA. Methods: The Studies of the Etiology of RA (SERA) is a multisite study of a cohort of first-degree relatives of RA probands (FDRs) and a cohort enriched with the HLA-DR4 genetic variant, both of which are RA-free, but at-risk for future RA. A nested case-control design was employed, with 30 cases defined as positive for anti-CCP2 autoantibody (Axis-Shield), and 48 controls always negative for any RA-related autoantibodies, frequency matched on age at study visit, sex, race, and study site. Antibody status, self-reported n-3 supplement use, and erythrocyte membrane levels of n-3 fatty acids (a biomarker for long-term fatty acid status measured by GC-MS) were obtained from a single crosssectional visit. Logistic regression was used to assess the association between n-3 supplement use and CCP2 positivity, and test the association between a 1SD increase in membrane n-3 levels and CCP2 positivity. Mediation analyses assessed the percent of the association between self-report n-3 supplement use and CCP2 positivity mediated by membrane levels. Results: Demographic characteristics were similar between CCP2 positive cases and controls. Subjects with CCP2 positivity were less likely than controls to self-report n-3 supplement use (OR: 0.14; 95% CI: 0.03–0.68). In addition, CCP2 positive subjects had lower erythrocyte membrane levels of certain n-3 fatty acids when compared to controls (Table). In mediation analyses, membrane n-3s explained ⬃9% of the association between supplement use and CCP2. Table. The association between anti-CCP2 positivity and increasing levels of erythrocyte membrane n-3 fatty acids. Odds ratios reported correspond to a 1 SD increase in erythrocyte membrane n-3 fatty acid levels. Self-Reported Use of n-3 Supplement* Erythrocyte Membrane n-3 Total n-3 fatty acid* ALA(18:3n-3)* EPA(20:5n-3)* DPA(22:5n-3)* DHA(22:6n-3)* EPA ⫹ DHA* EPA ⫹ DHA ⫹ DPA*

Odds Ratio 0.14

95% CI 0.03–0.68

p-value 0.01

0.48 0.90 0.58 0.65 0.53 0.51 0.48

0.26–0.87 0.56–1.45 0.30–1.11 0.40–1.05 0.31–0.92 0.29–0.92 0.26–0.88

0.02 0.67 0.10 0.67 0.02 0.03 0.02

* Each row represents a separate model (i.e. the fatty acid variables are not adjusted for each other).

2661 Physical Activity, Adiposity, and The Risk Of Gout In Women: The Nurses Health Study. Hyon K. Choi1, Lindsay C Burns2, Yuqing Zhang1, Sharan Rai3 and Gary Curhan4. 1Boston University School of Medicine, Boston, MA, 2University of British Columbia, Vancouver, BC, 3Arthritis Research Centre of Canada, Richmond, BC, 4Harvard Medical School, Boston, MA. Background/Purpose: There is a remarkable, increasing disease burden of gout and its associated cardiovascular (CV)-metabolic comorbidities in the US. While the benefits of physical exercise on CV-metabolic outcomes have been shown to extend beyond weight loss, such an independent impact on the risk of gout is unknown. We evaluated the potential impact of physical activity on the risk of incident gout in a large prospective cohort of women in the Nurses Health Study. Methods: We examined the relation between physical activity, body mass index, and the risk of incident gout in 94,389 female participants with no history of gout at baseline. We used the American College of Rheumatology criteria to ascertain gout. Cox proportional hazards models were used to estimate the relative risk (RR) of incident gout after adjusting for the following variables in a time-varying manner: age, total energy intake, alcohol, body-mass index, use of diuretics, history of hypertension, history of chronic renal failure, and daily mean intake of meats, seafood, dairy foods, coffee, and total vitamin C. Results: During 26 years of follow-up, we documented 846 confirmed incident cases of gout. An increasing level of physical activity was independently associated with a decreasing risk of gout. The multivariate relative risks (RRs) of gout were 1.00, 0.99, 0.90, 0.81, and 0.75 (95% CI, 0.58 to 0.97) (P for trend ⬍ 0.001) for physical activity levels from the lowest to the highest quintile (Table). In contrast, compared with BMI ⬍ 21, the multivariate RRs of gout were 1.71, 1.81, 2.91, 5.00, 6.77 (95% CI, 4.60 to 10.0) (P for trend ⬍ 0.001) for BMI categories of 21–22.9, 23–24.9, 25–29.9, 30–34.9, and ⱖ35, respectively. These findings persisted in subgroup analyses stratified by major factors such as intake of alcohol, dairy products, and sugary soda (all P values for interaction ⬎ 0.05). Table 1. Relative Risk of Incident Gout According to Physical Activity Physical Activity (hr/wk) No. of Cases Person-Years Age-Adjusted RR (95% CI) Multivariate* RR (95% CI) Multivariate* ⫹ BMI-adjusted RR (95% CI)

0

0.03–0.50

0.51–1.50

1.51–3.74

ⱖ3.75

P for trend

249 394096 1.0

257 251287 0.84 (0.66, 1.08)

104 315926 0.70 (0.57, 0.86)

138 299321 0.56 (0.44, 0.71)

98 318665 0.44 (0.34, 0.56)

– – ⬍ 0.001

1.0

0.94–0.74, 1.20)

0.82–0.67, 1.01)

0.70–0.55, 0.89)

0.59–0.46, 0.76)

⬍ 0.001

1.0

0.99–0.78, 1.26)

0.90–0.73, 1.11)

0.81–0.64, 1.03)

0.75–0.58, 0.97)

0.012

Table 2. Relative Risk of Incident Gout According to BMI Body Mass Index ⬍ 21 21–22.9 23–24.9 25–29.9 30–34.9 ⱖ35 P for trend (kg/m2) No. of Cases 33 75 99 304 248 173 – Person-Years 322346 419510 441973 678320 262748 120362 – Age Adjusted RR 1.0 1.80 (1.20, 2.71) 2.07 (1.40, 3.08) 3.80 (2.65, 5.45) 7.81 (5.42, 11.2) 12.8 (8.83, 18.7) ⬍ 0.001 (95% CI) Multivariate RR 1.0 1.70 (1.13, 2.57) 1.82 (1.22, 2.70) 2.94 (2.05, 4.23) 5.14 (3.55, 7.45) 7.07 (4.81, 10.4) ⬍ 0.001 (95% CI)* Multivariate*⫹ 1.0 1.71 (1.13, 2.58) 1.81 (1.22, 2.69) 2.91 (2.02, 4.18) 5.00 (3.45, 7.26) 6.77 (4.60, 10.0) ⬍ 0.001 Physical Activityadjusted RR (95% CI) Abbreviations: RR⫽ Relative Risk; CI ⫽ confidence intervals * Adjusted for age, total energy intake, menopause, use of hormonal replacement, diuretic use, history of hypertension, and intake of alcohol, sugar-sweetened soft drinks, coffee, total meats, seafood, dairy foods, and total vitamin C

S1134

Conclusion: Our findings provide prospective evidence that increased physical activity is independently associated with a decreased risk of incident gout among women, beyond its impact on adiposity. In contrast, increased adiposity was independently associated with an increased risk of incident gout. These findings support the fundamental role of increased physical activity and weight control in the prevention of gout. Disclosure: H. K. Choi, None; L. C. Burns, None; Y. Zhang, None; S. Rai, None; G. Curhan, None.

ACR Concurrent Abstract Session Health Services Research, Quality Measures and Quality of Care Innovations in Health Care Delivery Tuesday, October 29, 2013, 2:30 PM–4:00 PM

Number of health care visits in baseline year (range 1–41) Baseline BILD score Baseline SLAQ score Disease duration, per 10 years

1.0 (1.0, 1.0)

1.0 (1.0, 1.1)

1.2 (1.1, 1.3) 0.9 (0.9, 1.0) 0.9 (0.7, 1.1)

1.1 (1.0, 1.1) 1.1 (1.0, 1.1) 1.9 (1.5, 2.5)

BILD⫽Brief Index of Lupus Damage, SLAQ⫽Systemic Lupus Activity Questionnaire. The model is also adjusted for age, sex, race/ethnicity, education, health insurance source and type, and the providers seen during the baseline year (rheumatologists vs. generalists).

Conclusion: In this community-based cohort of individuals with SLE, we illustrate for the first time a strong link between processes of care, defined by SLE quality measures, and the subsequent accumulation of disease damage, an important outcome. This suggests that improving quality of care has the potential to reduce accumulated damage in SLE even in advance of the development of new treatment options. Disclosure: J. Yazdany, None; L. Trupin, None; P. P. Katz, None; G. Schmajuk, None; E. H. Yelin, None.

2662 Quality Of Care Predicts Disease Outcomes Among Patients With Systemic Lupus Erythematosus. Jinoos Yazdany1, Laura Trupin1, Patricia P. Katz1, Gabriela Schmajuk1 and Edward H. Yelin2. 1University of California, San Francisco, San Francisco, CA, 2UC San Francisco, San Francisco, CA.

Moving Towards Personalized Healthcare: A Patient Reported Outcome Based Algorithm Can Aid Rheumatologists and Patients In Monitoring Rheumatoid Arthritis In Daily Clinical Practice. Jos Hendrikx1, Jaap Fransen1, Alessandro Toniolo2 and Piet L.C.M. van Riel1. 1Radboud University Nijmegen Medical Centre, Nijmegen, Netherlands, 2Pfizer Pharmaceuticals, Rome, Italy. Background/Purpose: Several Patient Reported Outcome (PRO)-based instruments to measure disease activity in Rheumatoid Arthritis (RA) exist, though an evidence base for their use in monitoring strategies is lacking. The objective of this study was to evaluate alerts generated by a PRO-based algorithm for monitoring patients with RA in the second year follow-up after diagnosis. Methods: The algorithm (figure 1) monitors an easily attainable, equally weighted PRO-score of VAS general health, VAS disease activity and VAS pain. Red flags are generated in two instances: 1. the target level of disease activity is not met (set at a Patient Acceptable Symptom State value of 40mm) 2. change in disease activity (current level versus best moving average of 3 past visits) surpasses the early alert threshold (set at a Minimal Important Change value of 20mm) Three consecutive red flags trigger an alert to the physician. Data from patients with complete second year follow-up, included consecutively in the Nijmegen Early RA cohort between 2003–2011, were used. Weekly PRO-scores were interpolated, with addition of a random error component, based on the empirical individual patient data during the second year follow-up after diagnosis. Alerts were evaluated against DMARD/ Biologic medication escalation registered in daily clinical practice. Escalation was defined as an increase in dose/frequency or start of new drug. Analyses were stratified according to DAS28 level at 12 months.

Table. Longitudinal relationship between higher performance on SLE quality measures and disease activity and damage, with adjustment for other health services and sociodemographic factors. Increase in SLAQ by 4 or more points OR (95% CI)

Increase in BILD by 2 or more points OR (95% CI)

0.9 (0.5, 1.4)

0.4 (0.2, 0.7)

. Ref 1.2 (0.8, 1.9) 1.6 (0.9, 2.9)

. Ref 1.1 (0.7, 1.8) 1.9 (1.0, 3.6)

Figure 1. Preliminary PRO-based algorithm

Characteristic Higher performance (at least 85% fulfillment of eligible quality measures) Eligible services in baseline year 2–3 4–6 7–11

Results: Data of 158 patients with RA were available for analysis. An overview of the algorithm performance is shown in table 1. The Negative Predictive Values (NPVs) represent the proportion of visits without medication escalation in all cases that the algorithm did not generate an alert. Overall, in 92.5% of visits where the algorithm did not generate an alert medication was also not escalated.

S1135

Tuesday, October 29

Background/Purpose: Although measures to assess quality of care in systemic lupus erythematosus (SLE) are available, their relationship to long-term disease outcomes is unknown. Using a prospective, longitudinal cohort study, we examined the associations between high quality care and risks of increased SLE disease activity and damage. Methods: Data derive from a longitudinal study of individuals with SLE, the UCSF Lupus Outcomes Study (LOS). Participants were followed from 2009 through 2012, responding to yearly telephone surveys. The primary outcomes in this study were clinically meaningful increases (⬎0.5 standard deviations) in 1) disease activity, defined as greater than or equal to 4 points on the Systemic Lupus Activity Questionnaire (SLAQ) between the 1st and 4th year of observation, and 2) disease damage, defined as greater than or equal to 2 points on the Brief Index of Lupus Damage (BILD) over the same time period. Both SLAQ and BILD have been validated for self-report in telephone surveys. Our primary independent variable was performance on 13 previously validated SLE quality measures, with a pass rate of greater than or equal to 85% (corresponding to the top quintile of performance, with pass rate defined as proportion of eligible recommended services received) considered indicative of high quality care. Using multivariable logistic regression, we examined the relationship between higher performance on quality measures and disease outcomes, adjusting for baseline disease activity and damage, age, gender, race/ethnicity, education, insurance source, health care utilization, specialties of providers seen, number of eligible quality measures, and follow-up time. Results: Among 737 participants, the mean age at baseline was 50 years (SD 13), 93% were female, 63% were white; 41% had college degrees, and 15% had poverty-level incomes. Respondents were eligible for a mean of 5 quality measures (SD 2, range 2–11). There were 155 participants who had a clinically meaningful increase in SLAQ and 162 with an increase in BILD. In our models, we found no statistically significant relationship between performance on quality measures and changes in SLAQ score. However, receiving higher quality SLE care was significantly protective against increased disease damage (adjusted OR 0.4, 0.2–0.7; Table), even after adjusting for covariates, including baseline damage, disease activity, sociodemographic characteristics, insurance, health care utilization and number of eligible quality measures.

2663

Conclusion: Rheumatologists started AOP therapy significantly more often in patients who would be recommended for therapy based on NOF guidelines. However, only 67.2% of rheumatology and 54.9% of nonrheumatology FRAX® high-risk with osteoporosis patients were started on AOP by 1-yr. In FRAX® high-risk with osteopenia patients only 40.5% of those treated by a rheumatologist and 27.8% of those not treated by a rheumatologist were started on therapy at 1-yr. These data indicate a care gap in osteoporosis treatment in both physician groups which needs to be addressed to improve quality of care.

Table 1. preliminary algorithm results

Total DAS28 ⬍2.6 DAS28 ⱕ3.2 & ⱖ2.6 DAS28 ⬍5.1 & ⬎3.2 DAS28 ⱖ5.1

Visits PPV NPV Sens 693 14.0 92.5 55.7 Stratified according to DAS2812 months 277 6.0 95.9 38.5 101 16.7 95.4 66.7 282 18.3 86.8 54.5 33 13.8 100 100

Spec 61.5

A priori chance of escalation 10.1

70.5 67.4 55.0 13.8

4.7 8.9 15.6 12.1

PPV: Positive Predictive Value, NPV: Negative Predictive Value, Sens: Sensitivity, Spec: Specificity

Conclusion: The high NPVs indicate that, when using the algorithm, the chance of missing patients in need of medication escalation is very low. These findings provide evidence that an off-site monitoring system could aid in optimizing the number and timing of face-to-face consultations of patients with their rheumatologists. Further investigation into the optimal PRO-score, measurement frequency and algorithm parameter values to be used for off-site monitoring is warranted. Disclosure: J. Hendrikx, Pfizer Inc, 2; J. Fransen, Pfizer Inc, 2; A. Toniolo, Pfizer Inc, 3; P. L. C. M. van Riel, Pfizer Inc, 2.

2664

Tuesday, October 29

Evaluation Of Osteoporosis Medication Starts In Patients Based On T-Score and FRAX® Absolute Fracture Risk Model In a Large Health Care System. Robert A. Overman and Chad L. Deal. Cleveland Clinic, Cleveland, OH. Background/Purpose: The National Osteoporosis Foundation (NOF) 2008 guidelines recommend treatment for postmenopausal women (PMW) and men ⱖ 50 if the T-score is ⱕ⫺2.5 at the hip or spine and in patients with osteopenia (T-score ⫺1.0 to ⫺2.5) if the WHO FRAX® 10-yr fracture risk is ⱖ3% for hip or ⱖ20% for major osteoporotic fractures. We evaluated treatment initiation in patients after DXA and compared treatment by rheumatologists to non-rheumatologists. Methods: The Cleveland Clinic DXA registry was linked with the patient’s electronic medical record using Explorys Inc. PMW and men ⱖ50 in the registry between 7/2009 and 12/2012, who were anti-osteoporosis medication (AOP) naı¨ve, and had at least one office visit in the years pre and post-DXA were included. New use of AOPs; bisphosphonates, teriparatide, denosumab, raloxifene, calcitonin, and estrogen started within 90, 180, and 365 days post-DXA were collected through 2/2013. Subjects who did not exceed each post-DXA time period were not included in the analysis. Subjects were stratified into 6 groups based on T-score (osteoporosis or osteopenia); FRAX® 10-yr risk of major osteoporotic fracture or hip fracture, ⱖ20% and/or ⱖ3% (high-risk) or ⬍20% and ⬍3% (low-risk); and treatment by a rheumatologist or non-rheumatologist. Results are presented as % difference in treatment starts. Group comparisons were made using chi-square with pⱕ0.05 demonstrating statistical significance. Results: Study subjects had a mean age of 70.9 (SD 10.5) and 80.8% (3456/4280) were female. The difference in treatment starts at 90, 180 and 365 days after initial DXA for rheumatologists and non-rheumatologists are presented in Table 1. The groups were osteoporosis at either spine or hip and FRAX® high-risk; osteoporosis and FRAX® low-risk; osteopenia and FRAX® high-risk. Rheumatologists were compared to non-rheumatologists. Treatment would be recommended for all 6 groups based on NOF guidelines. Rheumatologists started significantly more patients on AOP than non-rheumatologists at 180 and 365 days in patients with osteoporosis and FRAX ® high-risk, and at all-time points in patients with osteopenia and FRAX® high-risk. The greater number of AOP starts in rheumatologists indicate closer adherence to NOF guidelines for treatment. Table 1. Percentage of osteoporosis treatment naı¨ve patients started on osteoporosis therapy by Rheumatologists v Non-Rheumatologists based on time periods post-DXA Osteoporosis Osteoporosis and FRAX® and FRAX® High-Risk High-Risk (Rheum) (Non-Rheum) 42.4 (70/165) 40.3 (546/1355)

Osteoporosis and FRAX Low-Risk® (Rheum) 32.4 (24/74)

Osteoporosis Osteopenia Osteopenia and FRAX® and FRAX® and FRAX® Low-Risk % High-Risk High-Risk (Non-Rheum) diff (Rheum) (Non-Rheum) 42.4 (378/891) ⫺10.0 24.1 (58/241) 17.3 (269/1554)

Treatment % Initiation diff % within 90 2.1 Days (n/ group n) % within 180 55.6 (85/153) 48.6 (613/1262) 7.0 45.7 (32/70) 51.2 (429/838) Days (n/ group n) % within 365 67.2 (90/134) 54.9 (594/1082) 12.3** 58.3 (35/60) 58.7 (423/721) Days (n/ group n)

% diff 6.8*

⫺5.5 32.2 (75/233) 21.9 (316/1440) 10.3** ⫺0.4 40.5 (81/200) 27.8 (340/1225) 12.7**

Subjects who contributed at least the period days to the analysis were included in percentages for each row Rheum ⫽ subjects who had at least one visit with a rheumatologist in the year before and after DXA Non-Rheum ⫽ subjects not seen by a Rheumatologist in the year before and after DXA FRAX® High-Risk ⫽ 10-yr risk of hip fracture ⱖ3% and/or 1-yr risk of major osteoporotic fracture ⱖ20% FRAX® Low-Risk ⫽ 10-yr risk of hip fracture ⬍3% and 10-yr risk of major osteoporotic fracture ⬍20% % diff ⫽ the difference in percent treated by rheumatologists vs. a non-rheumatology physician *: pⱕ0.05 **: pⱕ0.01.

Disclosure: R. A. Overman, None; C. L. Deal, None.

2665 Efficiency Gains For Rheumatology Consultation From An Electronic Referral System In a Safety Net Health System. Meghan M. Scheibe1, John B. Imboden2, Gabriela Schmajuk2, Mary Margaretten2, Jonathan D. Graf2, Alice Chen2, Edward H. Yelin2 and Jinoos Yazdany2. 1California Pacific Medical Center, San Francisco, CA, 2University of California, San Francisco, San Francisco, CA. Background/Purpose: Rheumatology care is a scarce resource in some settings. Health information technology holds promise in increasing the quality and efficiency of rheumatology referrals, but little information about its use for these purposes is available. The aim of this study was to determine the proportion of electronic referrals that are appropriate for pre-consultation exchange as well as to assess potential efficiency gains in a safety net health system. Methods: The Electronic Referral System (eReferral) allows two-way electronic communication between referring and specialty providers for new patient referrals. Once a referral is initiated, several possible outcomes may result: either the patient is scheduled in rheumatology clinic, the consult is deemed not appropriate for rheumatology, more information is requested from the referrer to gauge the appropriateness of the consultation, or the question is answered entirely within the eReferral system without scheduling a clinic visit. The latter 3 constitute “pre-consultation exchange.” We performed a retrospective study of a random sample of eReferrals for ambulatory rheumatology consultation between 2007–2012. A sample of approximately 10% of eReferrals (n⫽257) was reviewed. The primary clinical question and disposition for these referrals was recorded. Using a blinded adjudicated, structured review performed independently by two rheumatologists, we rated each eReferral for pre-consultation exchange using the categorization above. Results: Between 2007–2012, 2383 eReferrals were sent to rheumatology. Of the 257 referrals examined in more detail, the top reasons for consultation are shown (Table). 156 (61%) eReferrals were rated as appropriate for pre-consultation exchange. The primary reasons pre-consultation exchange was recommended were: additional testing was needed to clarify or expedite the consultation request (37%), the clinical question was unclear (14%), or the consult was deemed not appropriate for rheumatology consultation (10%). Referrals for pain were most likely to be rated as appropriate for pre-consultative exchange and those for crystal arthropathies were the least likely (Table). Following the exchange, 49 (32%) eReferrals were able to be resolved without a clinic visit. Table. Appropriateness of Rheumatology eReferrals for Pre-consultation Exchange in a Safety Net Setting.

Diagnosis

N (%) Top Reasons for eReferral

N (%) eReferrals recommended for immediate scheduling

Potential inflammatory arthritis Rheumatoid arthritis Pain Systemic lupus erythematosus Crystal Arthropathy

42 (17%) 35 (14%) 31 (13%) 20 (8%) 17 (7%)

15 (35%) 15 (43%) 2 (6%) 9 (45%) 11 (65%)

N (%) eReferrals deemed appropriate for pre-consultation exchange 27 (65%) 20 (57%) 29 (94%) 11 (55%) 6 (35%)

Conclusion: We evaluated eReferral, an innovative program designed to improve the quality and efficiency of specialty referrals. Over half of the referrals to rheumatology were rated as appropriate for pre-consultative exchange. In addition, one-third of the referrals were able to be resolved without a clinic visit, suggesting that using an iterative referral system that facilitates communication between referrers and specialists has the potential to improve the triage and efficiency of new patient visits in a busy

S1136

rheumatology practice, particularly in settings where rheumatologists are a scarce resource. Disclosure: M. M. Scheibe, None; J. B. Imboden, None; G. Schmajuk, None; M. Margaretten, None; J. D. Graf, None; A. Chen, None; E. H. Yelin, None; J. Yazdany, None.

10

65

NHW Improve diet 6 3 8 [10]

11

53

AA

12

53

13

63

NHW Increase physical activity 5 3 7 [10] AA Improve diet 5 3 6 [10]

Increase physical activity 5 3 5 [10]

2666 Implementing a Personalized Health Plan To Improve Therapeutic Outcomes In Veterans With Gout. Astrud Lorraine Leyva1, Una E. Makris2 and Salahuddin Kazi1. 1UT Southwestern Medical Center, Dallas, TX, 2UT Southwestern Medical Center, VA Medical Center, Dallas, TX.

Table 1. Patient Age Race

Goal Pre 3 Post [desired]

1

66

NHW Increase physical activity 2 3 2 [10]

2

73

H

Improve diet 3 3 5 [8]

3

56

AA

Improve diet 7 3 10 [10]

4

57

NHW Improve diet 4 3 5 [8]

5

43

AA

Stop alcohol use 3 3 8 [10]

6

65

AA


7

60

AA

8

45

AA

9

68


Increase physical activity 4 3 4 [10] NHW Increase physical activity 7 3 8 [10]

Pre/Post Uric Acid Quotes illustrating feedback for PHP 7.5/7.7

“Would have to change my expectation with 8, a little lower. I’d have a better chance of success with lower expectation” 12.6/6.1 “Makes me more aware to stay healthy and I’m eating better than I had been” 4.3/4.1 “Having a doctor calling me and taking time to show their concern is really good” 8.0/6.0 “The weekly phone calls helped by re adjusting medications as problems came up” 12.7/10.9 “The program is a good reminder to stop drinking and source of education” 8.6/5.3 “Having someone to talk to on a regular basis was good. Provided encouragement and answer questions as they come up” 6.2/5.6 “How many doctors call you and ask if you are doing ok? It made me accountable to get out of bed.” 6.8/7.7 “Helpful to have someone concerned about you calling every week” 7.8/5.9 “Helps me stay an track”

“The program may be better calls were every 2 weeks” 11.0/10.4 “Constant reminder to keep on working on my goal and to stay on medications” 8.4/5.9 “I like hearing from the doctor every week.” 6.8/4.1 “The depression is bad.” The phone calls “it’s like an uplift”. “I have a better understanding of what gout brings on.”

NHW ⫽ non-Hispanic white, AA ⫽ African American, H ⫽ Hispanic

Conclusion: This pilot study suggests that Veterans had a positive response to the gout PHP. Veterans indicated satisfaction with physician follow-up phone calls; however, this may not be feasible in a “real world” setting. Future studies evaluating the efficacy of the PHP in gout patients should include a control group, utilize technology for continuous reinforcement, measure unanticipated clinical encounters, and lastly, extend the duration of follow-up to allow behavioral, physical activity, or psychosocial change. Disclosure: A. L. Leyva, None; U. E. Makris, None; S. Kazi, None.

2667 Fast-Track Pathway In Giant Cell Arteritis: A Cost-Effectivenss Analysis. Katerina Achilleos1, Pravin Patil1, Win Win Maw1, Laura Bown2, David Halsall2, Charles Dobson2, Christian Dejaco1, Frances Borg1, Sunil Gupta3 and Bhaskar Dasgupta1. 1Southend University Hospital, Westcliffon-sea, United Kingdom, 2National Health Service England, Leeds, United Kingdom, 3Clinical Commissioning Group, Southend-on-sea, United Kingdom. Background/Purpose: With incidence of 2.2 per 10,000 patient years, it is estimated that 12,000 new cases of Giant Cell Arteritis (GCA) are diagnosed every year in the UK of whom 20% lose sight permanently. This has significant implications on personal, social care and socioeconomic costs. A fast-track GCA pathway was introduced in 2012 aimed at reducing multiple referral routes, delayed review and treatment and improving outcomes. We report the cost effectiveness of this pathway. Methods: A retrospective data analysis of 138 patients investigated for GCA (Jan 2009-Dec 2012), comparing costs of fast-track pathway (57 patients; Jan – Dec 2012) with the conventional referral route (81 patients; pre 2012) using the incremental cost effectiveness ratio (ICER). Data was collected from patient records, general practice and pathology databases along with telephone questionnaires. Direct costs included: GP appointments, investigations such as blood tests, scans and biopsies, outpatient appointments (Rheumatology, Eye Clinic and Neurology), A&E attendances, Inpatient stays, readmissions, drugs for treatment of GCA. Costing data for treatment and diagnosis costs was gathered from Reference Costs 2011/12, Prescription Cost Analysis 2012, and Unit Costs of Health and Social Care 2012. QOL was measured in 66 patients using the EQ5D. Health gains from diagnosis and treatment of GCA were quantified using Quality Adjusted Life Years (QALY’s). Results: The fast-track pathway has seen a reduction in number of GP appointments and cost of diagnosis and treatment. The difference in QALYs between patients with and without sight loss due to GCA was 0.2. Each patient that didn’t lose vision gained on average 2.6 QALY’s. The average cost of diagnosing and treating a patient with suspected GCA in the conventional pathway was ⱕ2,600, whilst in the fast-track; this was ⱕ1,675, a difference of ⱕ925 per patient. The ICER of implementing the fast-track pathway is -ⱕ1,950 per QALY. Thus, there is an average cost saving to the NHS of ⱕ925 for each patient treated for suspected GCA. Table 1. Patients with sight loss and the costs of treatment/diagnosis in the conventional and fast-track pathway

Conventional pathway Fast-track pathway

S1137

Total number of patients suspected of having GCA

Total number of patients diagnosed with GCA

Number of patients with Sight loss

Average cost of diagnosis and treatment

81 57

46 33

17 (37%) 3 (9%)

ⱕ2,600 ⱕ1,675

Tuesday, October 29

Background: Gout is a chronic crystal-induced arthritis that predominantly occurs among men with rising prevalence with advancing age. Patients with uncontrolled gout present with recurrent flares leading to unanticipated clinical encounters, i.e. phone calls, ER visits, and/or hospitalization. Given the patient population at the Dallas VAMC, it is appropriate to evaluate novel strategies for improving gout management. Outcomes in gout may be improved by eliciting goals of care from the patient. The Dallas VA has adapted a personalized health plan (PHP) derived from Duke University’s Integrative Medicine program that assesses patients’ desired goals in 8 different domains, initially tested in diabetic patients. The purpose of this study is to pilot the PHP to facilitate achieving goal uric acid ⬍6mg/dL in patients with uncontrolled gout. Methods: We recruited consecutive patients at the VA rheumatology clinic with uncontrolled gout, defined by uric acid ⬎ 6mg/dL AND ⱖ 1 acute gout flare(s) from July to September 2012. This project was reviewed by the VA IRB and considered to be a quality improvement project. The initial interview focused on formulating a goal (congruent with gout management). Patients selected a goal, indicating their starting and desired status using a numerical scale (1–10). Continuous reinforcement was achieved by weekly phone calls (ALL). Uric acid levels were evaluated at baseline and at the end of the study period. An exit interview was conducted after 3 months; which included open ended questions regarding feedback for the PHP (see Table 1). Clinical encounters over 1 year were evaluated for trends and compared to non-PHP patients using a student’s t-test. Results: Thirty-six veterans meeting study criteria were approached; 13 agreed to participate and completed the 3 month study. Participants’ average age was 59 ⫾ 8.9 years, 5/13 were Non-Hispanic White, 7/13 African-American and 1/13 Hispanic. All but one selected a goal to increase physical activity or adjust diet. At completion of the study, 1 patient achieved his desired goal; 3/13 remained at the same level; 9/13 did not achieve, but approached, the goal, as seen in Table 1. 10/13 patients had a reduction in uric acid level, of which 8/10 reached the goal of ⱕ6mg/dL. Clinical encounter data did not significantly differ between PHP participants and non-participants.

8.5/5.6

Conclusion: The fact-track pathway leads to a reduction in irreversible sight loss and is associated with reduced diagnosis and treatment costs. It results in 2.6 QALY’s gained for each patient that does not suffer sight loss, suggesting that the fast-track pathway is more cost effective than the conventional pathway for management of early GCA. The ICER of implementing the fast-track pathway is -ⱕ1,950 per QALY with an average cost saving to the NHS of ⱕ925 for each patient. Our results do not include the small educational/refresher costs and are preliminary but indicate that the Fast track GCA pathway should be ‘rolled out’ globally. Disclosure: K. Achilleos, None; P. Patil, None; W. W. Maw, None; L. Bown, None; D. Halsall, None; C. Dobson, None; C. Dejaco, None; F. Borg, None; S. Gupta, None; B. Dasgupta, None.

ACR Concurrent Abstract Session

Table 1. Secondary efficacy endpoints at Week 8 comparing FX006 40 mg vs. TCA IR 40 mg

Osteoarthritis I: Therapeutics in Osteoarthritis Tuesday, October 29, 2013, 2:30 PM–4:00 PM

Tuesday, October 29

2668 A Randomized, Double-Blind, Dose Ranging Study Comparing FX006, An Intra-Articular (IA) Sustained-Release Formulation Of Triamcinolone Acetonide (TCA), To An Approved Injectable Suspension Of TCA In Patients With Osteoarthritis (OA) Of The Knee. Neil Bodick1, Joelle Lufkin1, Christina Willwerth1, Pierre Lachance2, Gregory Jasey3, Anil Gupta4, Anthony Chris5, Marc Russo6, Michael O’Mahony7, Sam Henein8, Louise Murdoch9, Ferdinandus de Looze10, David Hunter11 and Michael Clayman1. 1Flexion Therapeutics, Burlington, MA, 2Clinique St-Louis (Recherche) Inc, Quebec, QC, 3North Walkerville Orthopaedic Associates, Windsor, ON, 4Anil K. Gupta Medicine Professional Corporation, Toronto, ON, 5Aim Health Group, Waterloo, ON, 6Hunter Clinical Research, Broadmeadow, Australia, 7London Road Diagnostic Clinic and Medical Centre, Sarnia, ON, 8SKDS Research Inc., Newmarket, ON, 9Emeritus Research, Malvern East, Australia, 10AusTrials Pty Limited, Sherwood, Australia, 11Royal North Shore Hospital, St. Leonards, Australia. Background/Purpose: FX006 is a novel sustained-release formulation of TCA in poly(lactic-co-glycolic) acid microspheres intended to maintain therapeutic concentrations in the joint up to 3 months following IA injection. In patients (pts) with OA of the knee, FX006 was compared to an approved injectable suspension of TCA (TCA IR) to identify a dose of FX006 to move forward into Phase 3. Methods: Pts with baseline average daily pain intensity (ADP) of ⱖ5 to ⱕ9 on an 11 point Numeric Rating Scale (NRS) ranging from 0 (no pain) to 10 (pain as bad as you can imagine) were randomized (1:1:1:1) and treated with a single IA injection of FX006 (containing 10, 40 or 60 mg of TCA) or 40 mg of TCA IR. Pts were evaluated at 7 visits over 12 wks for efficacy, safety and PK. Primary endpoint: change from Baseline to each of Wks 8, 10 and 12 in the weekly mean of ADP. Secondary endpoints: time to onset of pain relief, WOMAC (pain, stiffness and function), responder status, and global impression of change. Results: Pts (228) were randomized and treated (FX006 10 mg, n⫽58, FX006 40 mg, n⫽59, FX006 60 mg, n⫽60; TCA IR, n⫽51). Treatment arms were well balanced across demographic and baseline characteristics (mean baseline ADP: 6.5). FX006 40 mg was superior to TCA IR on weekly mean of ADP at Wk 5 to Wk 10 (p⬍0.05) (average pain reduction ⫺0.9) (Figure 1). The 40 mg dose also demonstrated significant improvement over TCA IR (p⬍0.05) in secondary outcomes at Wk 8 (Table 1). Time to onset of pain relief was the same for FX006 40 mg and TCA IR. The 10 mg dose produced effects that were consistently improved relative to TCA IR but of lessor magnitude than those of the 40 mg dose. The 60 mg dose did not produce an improvement relative to the 40 mg dose. There were no related SAEs. AEs were generally mild and unrelated to study drug. Local knee-related AEs, lab assessments, ECGs and vital signs were unremarkable and similar across all treatments. Plasma PK across the FX006 doses were linear and dose-dependent. Peak plasma concentrations achieved with TCA IR were 10⫻ of that seen with FX006 40 mg.

WOMAC A (pain) (0–4 Likert scale) LSM change from baseline (SE) LSMD vs TCA IR (90% CI) 1 sided p-value WOMAC A1 (pain on walking) (0–4 Likert scale) LSM change from baseline (SE) LSMD vs TCA IR (90% CI) 1 sided p-value WOMAC B (stiffness) (0–4 Likert scale) LSM change from baseline (SE) LSMD vs TCA IR (90% CI) 1 sided p-value WOMAC C (function) (0–4 Likert scale) LSM change from baseline (SE) LSMD vs TCA IR (90% CI) 1 sided p-value OMERACT-OARSIa n (%) p-value OR (90% CI) Responder ⬎30% Improvementb n (%) p-value OR (90% CI) PGIC (1–7 Likert scale) LSM change from baseline (SE) LSMD vs TCA IR (90% CI) 1 sided p-value CGIC (1–7 Likert scale) LSM change from baseline (SE) LSMD vs TCA IR (90% CI) 1 sided p-value

FX006 40 mg (nⴝ59) ⫺1.33 (0.098) ⫺0.37 (⫺0.61, ⫺0.13) 0.0058

TCA IR 40 mg (nⴝ51) ⫺0.96 (0.108)

⫺1.2 (0.12) ⫺0.4 (⫺0.7, ⫺0.1) 0.0098

⫺0.8 (0.13)

⫺1.49 (0.112) ⫺0.49 (⫺0.77, ⫺0.22) 0.0018

⫺0.99 (0.124)

⫺1.31 (0.096) ⫺0.37 (⫺0.61, ⫺0.14 0.0049

⫺0.94 (0.106)

53 (89.8%) 0.0118 3.9 (1.6, 9.3)

32 (69.6%)

47 (85.5%) 0.0119 3.5 (1.5, 7.9)

27 (62.8%)

1.8 (0.16) ⫺0.7 (⫺1.1, ⫺0.3) 0.0013

2.6 (0.17)

1.8 (0.16) ⫺0.7 (⫺1.1, ⫺0.3) 0.0013

2.6 (0.17)

a n⫽59 for FX006 40 mg; n⫽46 for TCA IR b n⫽55 for FX006 40 mg; n⫽43 for TCA IR LSM ⫽ least squares mean; LSMD ⫽ least square means difference; WOMAC ⫽ Western Ontario and McMaster Osteoarthritis Index; OMERACT-OARSI ⫽ Outcome Measures in Rheumatoid Arthritis Clinical Trials- Osteoarthritis Research Society International; PGIC ⫽ Patient Global Impression of Change; CGIC ⫽ Clinical Global Impression of Change

Conclusion: In pts with OA of the knee, the 40 mg dose of FX006 demonstrated an amplified and prolonged therapeutic effect relative to TCA IR. A Phase 3 study of the 40 mg dose of FX006 is being planned. Disclosure: N. Bodick, Flexion Therapeutics, 1, Flexion Therapeutics, 3; J. Lufkin, Flexion Therapeutics, 1, Flexion Therapeutics, 3; C. Willwerth, Flexion Therapeutics, 1, Flexion Therapeutics, 3; P. Lachance, None; G. Jasey, None; A. Gupta, None; A. Chris, None; M. Russo, None; M. O’Mahony, None; S. Henein, None; L. Murdoch, None; F. de Looze, None; D. Hunter, None; M. Clayman, Flexion Therapeutics, 1, Flexion Therapeutics, 3.

S1138

2669 Treatment Of Knee Osteoarthritis Patients With Strontium Ranelate Reduces The Loss Of Cartilage Volume and Bone Marrow Lesions As Assessed By Magnetic Resonance Imaging: Data From The Phase III Strontium Ranelate Efficacy In Knee Osteoarthrtis Trial. Johanne Martel-Pelletier1, Camille Roubille1, Jean-Pierre Raynauld1, François Abram2, Marc Dorais3, Philippe Delorme1 and Jean-Pierre Pelletier1. 1 Osteoarthritis Research Unit, University of Montreal Hospital Research Centre (CRCHUM), Montreal, QC, 2Imaging Research & Development, ArthroLab Inc., Montreal, QC, 3StatSciences Inc., Notre-Dame de l’Iˆle Perrot, QC.

Disclosure: J. Martel-Pelletier, ArthroLab, 4, Servier, 5, Servier, 2; C. Roubille, None; J. P. Raynauld, ArthroLab, 5; F. Abram, ArthroLab, 3; M. Dorais, ArthroLab, 5; P. Delorme, None; J. P. Pelletier, ArthroLab, 4, Servier, 5, Servier, 2.

2670 Does Change In Femorotibial Cartilage Thickness Differ Between Acutely Anterior-Cruciate Ligament Injured Knees Treated With and Without Reconstructive Surgery. Wolfgang Wirth1, Felix Eckstein1, Martin Hudelmaier1, Stefan Lohmander2 and Richard Frobell3. 1 Paracelsus Medical University, Salzburg, Austria, 2Department of Orthopedics, Clinical Sciences Lund, Lund University, Lund, Sweden, 3 Lund University, Lund, Sweden. Background/Purpose: An ACL tear is a serious knee injury, involving chronic alterations in joint biomechanics. The risk of developing knee OA after an ACL tear is elevated, but the driving mechanisms are not known. We tested the hypothesis that surgical reconstruction of an acute ACL tear would influence change in femorotibial cartilage thickness over the first five years after injury. Methods: In a treatment RCT (the KANON-trial), 121 young (32 women, mean age 26.1 years) active adults with an acute ACL tear in a

Disclosure: W. Wirth, Chondrometrics GmbH, Ainring, Germany, 4, Chondrometrics GmbH, Ainring, Germany, 3; F. Eckstein, Chondrometrics GmbH, 4, MerckSerono, 5, Sanofi Aventis, 5, Abbot, 5, Synarc, 5, GlaxoSmithKline, 4, Genzyme, 5, Medtronic, 5, Synthes, 5, Pfizer, 2, Eli Lilly and Company, 2, MerckSerono, 2, GlaxoSmithKline, 2, Centocor, 2, Wyeth Pharmaceuticals, 2, Novartis Pharmaceutical Corporation, 2, Stryker, 2; M. Hudelmaier, None; S. Lohmander, None; R. Frobell, None.

2671 The Relationship Between Toe-Out Angle During Walking and Risk Of Medial Knee Osteoarthritis Incidence: The Multicenter Osteoarthritis Study. K. Douglas Gross1, Yuqing Zhang2, Emily K. Quinn1, Michael C. Nevitt3, Neil A. Segal4, Cora E. Lewis5 and David T. Felson2. 1 Boston University, Boston, MA, 2Boston University School of Medicine, Boston, MA, 3University of California, San Francisco, San Francisco, CA, 4 University of Iowa, Iowa City, IA, 5University of Alabama, Birmingham, Birmingham, AL. Background/Purpose: A previous study suggests a linear relationship between increased toe-out angle during walking and reduced risk of medial knee OA progression. Yet, gait lab findings indicate that both highly positive (toe-out) and negative (toe-in) angles can reduce medial knee load. The relationship of these angles with medial knee OA incidence has not been studied. We assessed the dose-response relationship of toe-out angle during walking with 2-year risk of incident medial knee OA. Methods: The NIH-funded Multicenter Osteoarthritis Study (MOST) includes adults aged 50–79 years that have or are at risk of knee OA. Among 60-month participants, mean toe-out angle was measured during 4 self-paced walking trials using a 4.9 meter GAITRite walkway (14-day

S1139

Tuesday, October 29

Background/Purpose: This study aimed to evaluate in a subpopulation of osteoarthritis (OA) patients from the SEKOIA Phase III trial the disease-modifying (DMOAD) effect of strontium ranelate (SrRan) on the changes in knee OA cartilage volume and bone marrow lesions (BML) using magnetic resonance imaging (MRI). Methods: Patients with knee OA (n⫽300) received placebo (n⫽112) or SrRan at 1 g/day (n⫽113) or at 2 g/day (n⫽105). The study included all randomised patients who received at least one dose of treatment and had at least two MRI examinations. MRI was performed at baseline and at 12, 24, and 36 months. The changes in cartilage volume loss and BML were assessed in the global knee and subregions. Missing values were imputed and the analyses were adjusted according to Bonferroni. Results: Data showed no between-group differences at baseline with regard to demographics, clinical symptoms, or imaging characteristics. Treatment with SrRan at 2 g/day significantly decreased cartilage volume loss on the tibial plateau as early as 12 months (p⫽0.002), which persisted up to 36 months (p⫽0.003). The urinary CTX-II was found to be significantly decreased in both SrRan groups at 36 months. At baseline, the BML were detected mainly in the medial compartment where they were found at increased prevalence in the femur compared to the plateau. At 36 months, SrRan treatment at both 1 and 2 g/day significantly reduced the BML score change in the medial compartment (p⫽0.002 and p⫽0.001, respectively). Interestingly, patients treated with SrRan 2 g/day showed no increase in BML score in the medial femur, central condyle or plateau, and those treated with SrRan 1g/day showed no increase in the medial central condyle or plateau, while an increase (39%, 40%, and 17%, respectively) in BML score was found in the placebo group. In those patients with BML at baseline, SrRan 2 g/day significantly (p⫽0.023) decreased the cartilage loss at 36 months in the medial plateau. Conclusion: In knee OA patients, treatment with SrRan 2 g/day was found to have a beneficial effect on both cartilage and subchondral bone by significantly reducing the cartilage volume loss in the tibial plateau and the progression of BML in the medial compartment. In turn, the decrease in BML in the medial tibial plateau was associated with a marked and significant reduction in cartilage volume loss at SrRan 2 g/day. These results from the MRI study support the DMOAD effect of SrRan reported using X-rays.

previously un-injured knee received similar rehabilitation. Baseline sagittal MR images were available for 117 participants, with year 2 and year 5 follow-up for each 112 of the 121 participants. 59 of those underwent an early ACL reconstruction (ACLR), 29 had a delayed ACLR, and 29 were treated with rehab alone. Cartilage thickness (ThC) was measured manually with blinding to time point and treatment using a dedicated software (Chondrometrics GmbH, Ainring, Germany). Primary outcome was the mean change in ThC for the entire femorotibial joint (FTJ); secondary outcomes were mean change in ThC for the medial and the lateral femorotibial compartments (MFTC/LFTC), the subregion with the largest ThC loss (ordered value 1 ⫽ OV1) and gain (OV16). ThC changes were analyzed using the t-test (crude differences) and analysis of covariance (differences with adjustment for age, sex & BMI). Results: The cartilage thickness change over 5 years in the FTJ did not differ significantly between knees treated with early ACLR (⫹148␮m, 95%CI: [⫹38, ⫹258]␮m), with delayed ACLR (⫹174 [⫹42, ⫹306]␮m), or with rehabilitation alone (⫹121 [⫺3,⫹246]␮m, crude/adjusted pⱖ0.56/0.65). In addition, no significant differences were found for the periods between BL3 Y2 and Y23 Y5. The change in FTJ ThC was largely driven by change in MFTC ThC (early ACLR: ⫹132 [⫹77, ⫹187] ␮m, delayed ACLR: ⫹128 [⫹40, ⫹216] ␮m, rehab alone: ⫹82 [⫹19, ⫹146] ␮m), but without significant differences between treatment groups over the entire 5 year period (pⱖ0.28/0.20) or for the sub-periods (pⱖ0.19/0.20). Smaller changes, not significantly different between treatment groups (pⱖ0.60/0.70), were observed in LFTC (early ACLR: ⫹16 [⫺53, ⫹85] ␮m, delayed ACLR: ⫹46[⫺25, ⫹116] ␮m, rehab alone: ⫹39 [⫺46, ⫹124] ␮m). OV 1 was significantly more negative and OV 16 was significantly more positive over 5 years in knees treated with early ACLR than in knees treated with rehab alone (OV1: p⫽0.04/0.03; OV16: p⫽0.02/0.01). These differences were predominantly driven by changes occurring between BL3 Y2 with OV1 being significantly more negative in knees with early (p⫽0.01/0.01) or delayed ACLR (p⫽0.04/0.04) compared to knees treated with rehab alone. No other significant differences were found for any period. Conclusion: Change in mean femorotibial cartilage thickness over a five year period after acute ACL injury did not differ between knees treated with early or delayed ACL reconstruction or knees treated with rehabilitation alone. An early ACL reconstruction may induce greater magnitudes of subregional cartilage thickness change as compared to knees treated with rehabilitation alone but the clinical relevance of such change remains to be determined.

retest ICC ⫽ 0.95). Readers scored medial joint space narrowing (JSN) on semiflexed knee x-rays using OARSI grades (weighted ␬ ⫽ 0.81). Among knees with medial JSN grade 0 at 60 months, incident cases had medial JSN ⱖ 1 and ⬎ lateral JSN grade at 84 months. With the middle quintile as a reference, logistic regression estimated the relative odds of medial OA incidence in each case-based quintile of toe-out angle while adjusting for age, BMI, walking speed, sex, race, clinic site, and non-independent knees of a single subject. We used a quadratic spline to smooth the curve and obtain a clear picture of the dose-response relationship. In sensitivity analysis, knees with pain during the walking exam were excluded. Results: 1112 participants (mean age 66.6 ⫹/⫺ 7.6 yrs, BMI 29.6 ⫹/⫺ 5.3 kg/m2, walking speed 1.19 ⫹/⫺ 0.19 m/sec, 61.1% female, 88.0% white, 55.7% Iowa clinic) contributed 1856 knees with a mean toe-out angle of 6.7 ⫹/⫺ 5.4° (range ⫺11.0, 30.0°). After adjustments, the relative odds of incident medial knee OA had an n-shaped relationship with toe-out angle (p for trend ⬍ 0.01), such that the highest (9.8 to 30.0°) and lowest (⫺11.0 to 2.7°) quintiles had 11–25% reduced odds compared to the middle quintile (5.1 to 7.5°) (see table and figure). Results were unchanged in sensitivity analysis (p for n-shaped trend ⫽ 0.03). Toe-In ⴚ11.0, 2.7°

4———————————3 2.8, 5.0° 5.1, 7.5° 7.6, 9.7°

# knees 424 289 % incident 3.3% 3.5% Adj OR* 0.89 0.89 (95% CI) (0.41, 1.92) (0.38, 2.06)

356 3.9% 1.00 (ref)

Toe-Out p for n 9.8, 30.0° shape trend

284 503 6.7% 3.4% 1.69 0.75 (0.85, 3.36) (0.36, 1.60)

Methods: 297 adults with no significant knee pain, injury, or history of clinical knee disease were recruited. Participants underwent knee magnetic resonance imaging at baseline and 2 years later. Popliteal artery wall thickness, knee cartilage volume and bone marrow lesions (BML) were assessed. Results: Of 278 participants with valid popliteal artery wall thickness measurement, 254 (91.4%) completed the follow-up. After adjusting for age, gender, body mass index and tibial bone area, increased popliteal artery wall thickness was associated with reduced medial tibial cartilage volume (B ⫽ ⫺6.7, 95% CI ⫺12.9, ⫺0.6, p ⫽ 0.03) and increased rate of medial tibial cartilage volume loss (B ⫽ 0.06, 95% CI 0.01, 0.12, p ⫽ 0.03). There was a trend for medial tibiofemoral BML getting worse in relation to increased popliteal artery wall thickness (odds ratio 1.07, 95% CI 0.99, 1.15, p ⫽ 0.07). No significant associations were observed in lateral tibiofemoral compartment. Conclusion: Increased popliteal artery wall thickness was associated with adverse changes in knee structure, as evidenced by reduced medial tibial cartilage volume, increased rate of cartilage volume loss and a trend for BML worsening over 2 years. These findings suggest an association between vascular pathology and early knee structural changes, supporting the hypothesis that vascular health may play a role in the development of knee osteoarthritis. Disclosure: Y. Wang, None; D. Novera, None; A. Wluka, None; J. Fairley, None; G. Giles, None; R. O’Sullivan, None; A. Teichtahl, None; F. Cicuttini, None.

0.01

2673

Tuesday, October 29

*Adjusted for age, BMI, walk speed, sex, race, clinic, and non-independent knees

Figure. Quadratic spline depicting relative odds of medial knee osteoarthritis incidence by continuous measurement of toe-out angle during walking.

Conclusion: These results suggest that both highly positive (toe-out) and negative (toe-in) angles during walking are protective against medial knee OA incidence. Clinical trials should determine if gait training to alter toe-out angle is effective in preventing medial OA onset in at-risk knees. Disclosure: K. D. Gross, None; Y. Zhang, None; E. K. Quinn, None; M. C. Nevitt, None; N. A. Segal, None; C. E. Lewis, None; D. T. Felson, None.

2672 Associations Between Popliteal Artery Wall Thickness and Knee Structure In Adults Without Clinical Knee Disease: A Prospective Cohort Study. Yuanyuan Wang1, Diaz Novera1, Anita Wluka1, Jessica Fairley1, Graham Giles2, Richard O’Sullivan3, Andrew Teichtahl1 and Flavia Cicuttini4. 1Monash University, Melbourne, Australia, 2University of Melbourne, Melbourne, Australia, 3Epworth Hospital, Melbourne, Australia, 4Department of Epidemiology and Preventive Medicine, Monash University, Melbourne, 3004, Australia. Background/Purpose: There is evidence for a vascular contribution to the pathogenesis of osteoarthritis. The aim of this study was to examine the association between popliteal artery wall thickness, previously shown to be associated with risk of generalized osteoarthritis, and knee structural changes in an asymptomatic cohort.

Mortality In Clinically Relevant Osteoarthritis and Rheumatoid Arthritis Compared With The General Population. Aleksandra Turkiewicz1, Tuhina Neogi2, George Peat3 and Martin Englund1. 1Department of Orthopedics, Clinical Sciences Lund, Lund University, Lund, Sweden, 2 Boston University School of Medicine, Boston, MA, 3Research Institute for Primary Care & Health Sciences, Keele University, Keele, United Kingdom. Background/Purpose: There is strong evidence in support of increased mortality in patients with rheumatoid arthritis (RA), while the relation is more controversial in osteoarthritis (OA). Our aim was to assess mortality rates in patients with clinically relevant OA and RA, respectively, compared with those in the general population who have sought health care to minimize confounding. Methods: We used a cohort study design within the Skåne Health Care Register in Sweden, which is a legislative, mandatory register based on physicians’ International Classification of Diseases (ICD) 10 diagnostic codes. The register covers all primary and specialist health care as well as hospitalizations in southern Sweden (population 1.3 million). The Swedish population register contains information about vital events (births, deaths) and changes in residential address. We identified all patients aged ⱖ45 years having received the diagnosis of knee OA (ICD-10: M17), hip OA (M16), hand OA (M18, M15.1, M15.2, M19.0D, M19.1D, M19.2D), or RA (M05, M06) during 1999 to 2006. For RA patients we required at least two health care visits with the specific code registered, with at least one from a specialist in rheumatology or internal medicine. For OA we required at least one visit with the specific code registered. For the reference population aged ⱖ45, we required at least one visit with any diagnostic code (90% of the population had sought care during the period). Using the population register we followed all subjects from Jan 1st 2007 until relocation outside of the region, death, or Dec 31st 2011. We calculated mortality rates using the Cox proportional hazard model adjusted for age and sex, disposable income, marital status, and highest level of education reached (provided by Statistics Sweden). In a sensitivity analysis, we considered subjects who received a diagnosis of pain in a joint (M25.5, location unspecified) at age 55 or older as OA subjects in addition to the definition above. Results: The hazard ratio of death for RA patients compared to general population was 1.58 (95% CI: 1.35–1.84). Patients with knee, hip or hand OA did not have increased risk of mortality compared with the general population, adjusted for age, sex and socioeconomic variables (Figure). The risk of death remained similar when OA in any joint (i.e., any combination of knee, hip, or hand) was assessed, or when pain in a joint at age ⱖ55 was additionally considered as OA. Effect estimates were similar in men and women (Table).

S1140

C-reactive protein (hsCRP), and proteinuria, and (3) traditional cardiovascular risk factors including fasting lipids and baseline CIMT. Results: 201/221 APPLE subjects had available samples and were included in the analysis; 61/201 (30%) had vitamin D deficiency at baseline. There was no change in 25(OH)D levels after 1 year of atorvastatin or placebo. In univariable analysis, baseline 25(OH)D deficiency was associated with season (p ⬍ 0.01), minority status (p ⬍ 0.01), body mass index (p ⫽ 0.04), duration of SLE (p ⬍ 0.01), SLICC damage index (p ⫽ 0.04), hsCRP (p ⬍ 0.01), mean-max IMT (p ⫽ 0.01), LDL-cholesterol (p ⫽ 0.03), and timed urine protein (p ⫽ 0.03). In multivariable modeling, vitamin D deficiency was associated with age, latitude, season, minority status, proteinuria, and hsCRP (see Table). Table. Logistic modeling of vitamin D deficiency (25OHD ⬍ 20ng/mL)

Knee OA nⴝ26 825 69.5 (11.8) 15 820 (590 122 982 4 166 3 388

Hip OA nⴝ14 557 72.6 (10.8) 8 449 (58) 65 498 2 771 4 231

Hand OA nⴝ7 279 65.8 (10.2) 5 666 (78) 34 532 660 1 911

RA nⴝ4 838 67.3 (11.0) 3 459 (72) 21 399 1 002 4 682

Reference population nⴝ511 115 63.2 (12.3) 269 118 (53) 2 386 221 56 373 2 362

Conclusion: We found no evidence of increased risk of mortality over 5 years in patients with doctor-diagnosed knee, hip or hand OA, compared with the general population, while increased mortality in RA was confirmed. Disclosure: A. Turkiewicz, None; T. Neogi, None; G. Peat, None; M. Englund, None.


Odds ratio 1.28 0.87

95% CI 1.09, 1.50 0.76, 0.99

P-value 0.002 0.034

2.83 1.23 0.57 17.47 2.47 1.40

0.87, 9.23 0.45, 3.36 0.19, 1.75 5.22, 58.48 0.96, 6.34 1.07, 1.83

0.084 0.685 0.327 ⬍0.001 0.060 0.015

Conclusion: Vitamin D deficiency is common in pediatric lupus and is independently associated with elevated hsCRP, a marker of inflammation which predicts higher general cardiovascular disease risk. Observed association differences between univariable and multivariable modeling may be related to confounding or loss of power and requires further study. The association between vitamin D deficiency and hsCRP is novel in the SLE population and suggests that vitamin D deficiency may independently contribute to heightened inflammation and cardiovascular risk in this population. Grant support: ClinicalTrials.gov identifier: NCT00065806. APPLE supported by the NIH (NIAMS contract N01-AR-2–2265), the Edna and Fred L. Mandel Jr. Center for Hypertension and Atherosclerosis, and Pfizer, which provided atorvastatin and matching placebo. Secondary analysis supported by the Rainbow Babies and Children’s Hospital Pediatrics Pilot Award, and the NIH (NIAMS contract 5P30-AR-047363– 12). Disclosure: A. B. Robinson, None; V. Tangpricha, None; E. Yow, None; R. Gurion, None; G. McComsey, None; L. E. Schanberg, Pfizer Inc, 2, Pfizer Inc, 5.

Pediatric Rheumatology - Clinical and Therapeutic Aspects III: Systemic Lupus Erythematosus and Other Disease Outcomes Tuesday, October 29, 2013, 2:30 PM–4:00 PM

2675 2674 Vitamin D Deficiency Is Common and Associated With Increased C-Reactive Protein In Children With Lupus: An Atherosclerosis Prevention In Pediatric Lupus Erythematosus Substudy. Angela B. Robinson1, Vin Tangpricha2, Eric Yow3, Reut Gurion1, Grace McComsey1 and Laura E. Schanberg4. 1Rainbow Babies and Children’s Hospital / Case Medical Center, Cleveland, OH, 2Emory University School of Medicine, Atlanta, GA, 3Duke Clinical Research Institute, Durham, NC, 4Duke University Medical Center, Durham, NC. Background/Purpose: Epidemiologic associations suggest vitamin D may play a role in inflammation and progression of atherosclerosis. Using frozen serum, carotid intima medial thickness (CIMT) measurements, and other data from the Atherosclerosis Prevention in Pediatric Lupus Erythematosus (APPLE) trial, we assessed associations between 25-hydroxyvitamin D [25(OH)D] and measures of systemic lupus erythematosus (SLE) disease activity and cardiovascular risk. Methods: Participants in the 3-year APPLE trial were randomized to placebo or atorvastatin and CIMT progression was measured. Serum collected at baseline and 1 year was used to measure 25(OH)D levels. Logistic regression models for vitamin D deficiency [25(OH)D levels ⬍ 20 ng/mL] at baseline were constructed to evaluate univariable and multivariable associations with baseline variables. Variables collected as part of the APPLE trial included (1) known risk factors for vitamin D deficiency, (2) SLE-specific factors including duration of illness, disease activity, high sensitivity

Phase 2 Trial On Triptorelin For Ovary Protection In ChildhoodOnset Systemic Lupus Erythematosus. Rina Mina1, Andreas Reiff2, Clovis A. Silva3, Patricia Vega Fernandez4, Gloria C. Higgins5, Lisa F. Imundo6, Marisa S. Klein-Gitelman7, Calvin Williams8, Carol A. Wallace9, Nadia E. Aikawa10, Shannen L. Nelson11, Jun Ying12, Susan R. Rose13 and Hermine I. Brunner14. 1Cincinnati Children’s Hospital Medical Center/University of Cincinnati School of Medicine, Cincinnati, OH, 2 Children’s Hospital Los Angeles, Los Angeles, CA, 3Faculdade de Medicina da Universidade de Saõ Paulo, Sao Paulo, Brazil, 4Cincinnati Children’s Hospital Medica Center, CINCINNATI, OH, 5Nationwide Childrens Hosp, Columbus, OH, 6Morgan Stanely Children’s Hospital of New York-Presbyterian, Columbia University Medical Center, New York, NY, 7Children’s Memorial Hospital, Chicago, IL, 8Medical College of Wisconsin, Milwaukee, WI, 9Seattle Childrens Hosp & Research Institute, Seattle, WA, 10Faculdade de Medicina, Universidade de Saõ Paulo, Saõ Paulo, Brazil, 11Cincinnati Children’s Hospital Medical Center, Cincinnati, OH, 12University of Cincinnati, Cincinnati, OH, 13 Cincinnati Children’s Hospital Medical Center, Cincinnati, OH, 14 PRCSG, Cincinnati, OH. Background/Purpose: Gonadotoxicity is a known side effect of cyclophosphamide (CYC) therapy in childhood-onset systemic lupus erythematosus (cSLE). Gonadotropin releasing hormone (GnRH) agonists, such as triptorelin, can induce complete ovarian suppression (COS), which makes ovaries resistant to CYC gonadotoxicity. The aims of this research are to determine (1) the dose of triptorelin sufficient for reliable

S1141

Tuesday, October 29

Age, mean (SD) Women, n (%) Person time, years Deaths, n Mortality rate, per 100 000 person-years HZ (95% CI)—adjusted for age, marital status, income and highest level of education reached Men 0.89 (0.79–1.00) 0.97 (0.85–1.10) 0.89 (0.65–1.21) 1.36 (1.04–1.79) Women 0.90 (0.82–0.98) 0.90 (0.80–1.01) 0.95 (0.79–1.15) 1.71 (1.41–2.06)

Age Latitude Season 1st quarter 2nd quarter 3rd quarter Minority status Log timed urine proteinuria Log hsCRP

Tuesday, October 29

COS in cSLE patients receiving CYC, and (2) the safety profile of triptorelin in cSLE. Methods: This double-blind, placebo-controlled, dose-escalation RCT (phase I/II) recruited cSLE patients from 7 international centers (NCT00124514). All patients were female, aged ⱕ 21 years and Tanner breast stage ⱖ 2. They were randomized in a 4:1 fashion to receive either intramuscular triptorelin [␮g/kg/dose: 25 (T1), 50 (T2), 75 (T3), 100 (T4)] or placebo every 4 weeks during CYC induction. The primary efficacy endpoint was the monthly dose of triptorelin that resulted in maintenance of COS which was defined a priori as (A) basal level of Follicle Stimulating Hormone (FSH) ⬍ 2 mIU/ml and of Luteinizing Hormone (LH) ⬍ 1 mIU/ml; or (B) FSH ⬍ 3 mIU/ml and of LH ⬍ 2 mIU/ml during GnRH Stimulation Testing (GAST). FSH and LH levels were drawn at the expected trough levels of triptorelin (27 days post administration). The triptorelin dose was escalated by 25% or at least 20 ␮g/kg until maintenance of COS was achieved (maximum⫽150 ␮g/kg/dose). Adverse events (AE) were graded according to the National Cancer Institute Common Toxicity Criteria, Version 2. Results: Twenty-nine females were randomized (Placebo⫽6, T1⫽ 9, T2⫽8, T3⫽5, T4⫽1). Of those who received triptorelin, 4 patients did not complete the study due to CYC discontinuation or withdrawal of consent. As expected, none of those in the placebo group achieved COS. A dose of 75 and 110 ␮g/kg maintained COS in 75% and 90% of the patients, respectively. There was no relationship between the triptorelin dose and the patients’ age, Tanner stage, global disease activity, renal and hepatic function. None of the patients in T3 and T4 arms needed dose escalation of triptorelin while the converse is true for majority of patients in T1 and T2. AE were mild and the most common were hot flashes (24%), headache (14%), and insomnia (10%). Serious AE included hospitalization due to dehydration (Placebo⫽1), worsening of SLE disease activity (Placebo⫽1), and cutaneous vasculitis (T3⫽1). Conclusion: This interim analysis supports that higher doses of triptorelin than previously suggested are needed for COS in adolescent females with cSLE who require CYC. Doses of intramuscular triptorelin at 110 ␮g/kg induce continuous COS between injections given every 28 days in almost all females, and thus likely provide ovarian protection. These high doses of triptorelin appear well tolerated. Disclosure: R. Mina, None; A. Reiff, None; C. A. Silva, None; P. Vega Fernandez, None; G. C. Higgins, None; L. F. Imundo, None; M. S. Klein-Gitelman, None; C. Williams, None; C. A. Wallace, Amgen, 2, Pfizer Inc, 2, Novartis Pharmaceutical Corporation, 5; N. E. Aikawa, None; S. L. Nelson, None; J. Ying, None; S. R. Rose, None; H. I. Brunner, Novartis, Genentech, Medimmune, EMD Serono, AMS, Pfizer, UCB, Jannsen, 5, Genentech, 8, Research grant from the FDA 5R01FD002369 and 2M01RR008084, 2.

2676 Increased Pulse Wave Velocity In Juvenile Idiopathic Arthritis Patients Compared To Controls From The General Population. Hanne A. Aulie1, Anne Marit Selvaag1, Vibke Lilleby1, Oyvind Molberg1, Anders Hartmann2, Hallvard Holdaas2 and Berit Flato1. 1Department of Rheumatology, Oslo University Hospital, Rikshospitalet, Oslo, Norway, 2Department of Nephrology (and specialised Endocrinology), Oslo University Hospital, Rikshospitalet, Oslo, Norway. Background/Purpose: The aim of this study was to compare markers of subclinical cardiovascular disease (CVD) in adults with juvenile idiopathic arthritis (JIA) and controls, and assess the relation between arterial properties and JIA characteristics, received therapy, and traditional cardiovascular risk factors. Methods: Eighty-seven JIA patients (median age 38.4 years), with persistently active disease at least 15 years after disease onset, registered by longitudinal follow-up, were reexamined after median 29 years and compared to 87 matched controls from the Norwegian population register. Pulse wave velocity (PWV), a direct measure of large arterial stiffness and augmentation index (AIx), a marker of arterial stiffness and wave reflection, were estimated. Linear regression analyses adjusted for age and gender were used to analyze the associations between arterial stiffness and traditional cardiovascular risk factors as well as disease characteristics. Linear multiple regression analyses were performed to identify predictors of arterial stiffness. Results: PWV, systolic and diastolic blood pressures (SBP and DBP) were significantly higher in patients than controls (table). In patients, increased PWV was mainly related to DBP (p ⬍0.001), but was also associated with higher SBP (p⬍0.001), glucose (p⫽0.001), and pulse rate (p⫽0.003), lower high density lipoprotein cholesterol (p⫽0.050), and male

gender (p⫽0.013). Higher AIx was determined by increased DBP (p⬍0.001), age (p ⫽ 0.039), female gender (p ⬍ 0.001), daily smoking (p⫽0.024), high sensitivity C- reactive protein (CRP) (p⫽0.014), and physician‘s global assessment of disease activity (p⫽0.001). Furthermore, AIx was associated with longer duration of active disease (p⫽0.012), higher CRP area under the curve (p⫽0.006), less vigorous physical activity (p⫽0.004), joint erosions (p⫽0.016), and therapy with prednisolone (p⬍0.001), nonsteroidal antiinflammatory drugs (p⫽0.036), and methotrexate (p⫽0.015). Hypertension, daily smokers, and insulin resistance were more frequent in patients than controls (p⫽0.039, p⫽0.043 and p⫽0.034 respectively), but the traditional cardiovascular risk factors were more extensively related to increased AIx in controls. Table.

Systolic blood pressure (mmHg) Diastolic blood pressure (mmHg) Pulse (beats/ minute) Pulse wave velocity (PWV) (m/s) Augmentation index (AIx)

n

JIA patients

Controls

87 87 87 78 79

119.4 ⫾ 14.5 115.7 ⫾ 9.8 75.7 ⫾ 10.3 72.7 ⫾ ⫾8.2 62.7 ⫾ 10.7 61.8 ⫾ 9.7 7.2 ⫾ 1.0 6.9 ⫾ 0.8 14.5 ⫾ 10.8 12.0 ⫾ 12.2

P (paired t-test) 0.050 0.029 0.564 0.035 0.154

Conclusion: JIA patients with long-term active disease experienced increased arterial stiffness assessed by PWV. Traditional cardiovascular risk factors and the degree of disease activity and severity, treatment factors and physical activity influenced the arterial stiffness measured by PWV and AIx in these patients. Disclosure: H. A. Aulie, None; A. M. Selvaag, None; V. Lilleby, None; O. Molberg, None; A. Hartmann, None; H. Holdaas, None; B. Flato, None.

2677 Disease Progression Into Adulthood In Patients With Juvenile Idiopathic Arthritis – a Longitudinal 30 Year Follow-Up Study. Anne Marit Selvaag, Hanne Aulie, Vibke Lilleby and Berit Flatø. Department of Rheumatology, Oslo University Hospital, Rikshospitalet, Oslo, Norway. Background/Purpose: The aim of the study was to assess disease activity and health status in a previously studied cohort of patients with juvenile idiopathic arthritis (JIA) 30 years after disease onset and reveal predictors of active disease. Methods: A total of 254 patients with JIA, first referred to our hospital from 1980 to1985, were reexamined clinically after median 15 years of disease duration, and by mailed questionnaires after median 23 years. These patients were invited to attend the present study. All patients were assessed by questionnaires, and those with signs of active disease after 15, 23 and/or 30 years were invited to a clinical reexamination. Health status was measured by HAQ and SF-36, and disease activity by the clinical version of the juvenile arthritis disease activity score (cJADAS). Logistic regression analyses were used to assess predictors of persistently active disease and an unacceptable symptom state (cJADAS ⬎ 4.5). Results: One hundred and seventy-one patients (67%) were included in the study. They were examined after a median of 30 (range 21–40) years of disease duration, median age 39 (range 28–45) years, 74% females. After 30 years, 101 patients (59%) were in clinical remission off medication, 12 (7%) were in remission on medication and 58 (34%) had persistently active disease. Thirty-seven of 57 patients (65%) with active disease at 15 years follow-up had active disease at 30 years follow-up, and 20 patients (35%) went into remission on/off medication. Eighty-four of 97 patients (87%) in remission off medication at 15 years were in remission at 30 years follow-up. Patients in remission on medication at 15 years (n⫽17) tended to flare (n⫽9) or go into remission off medication (n⫽6). The cJADAS score was median 1.9 for the total study group, and 43% (n⫽73) of the patients had a cJADAS score ⱕ 1.0, 10 % (n⫽16) had cJADAS from 1.1 to 2.0, 18% (n⫽30) had cJADAS from 2.1 to 4.5, and 30% (n⫽ 50) had cJADAS ⬎ 4.5. Predictors of persistently active disease at 30 years follow-up were: being diagnosed with a JIA subgroup other than persistent oligo articular and systemic JIA (OR 4.1, 95%CI 1.5–11.5), being DR1 positive (OR 8.3, 95%CI 2.3–30.3), a short total time in remission (OR 9.0, 95% CI 3.0–26.7), and not being in remission at 15 years follow-up (OR 13.7, 95%CI 4.9–38.4). R2⫽65%. Predictors of a cJADAS score ⬎ 4.5 at 30 years follow-up were physician’s global score of disease activity at 15 years (OR 2.4, 95%CI 1.7–3.5) and daily smoking (OR 2.5, 95%CI 1.1–5.9). R2⫽23%.

S1142

We compared disease activity at 15 years with that of 30 years follow-up for 90 patients. There were significant improvements in physicians global assessment of disease activity (p⫽0.003), number of active joints (p⫽0.010), ESR (p⫽0.041) and CRP (p⬍0.001), but no significant change in patient’s global assessment, number of joints with limitation of motion, HAQ, or SF-36. Conclusion: The overall remission rates were stable between 15 and 30 years. After 30 years, a third of the JIA patients had an unacceptable symptom state and a third was not in remission on or off medication. Physician’s assessment of disease activity and inflammatory markers improved over the years, but patient reported disability and health status did not change. Disclosure: A. M. Selvaag, The Norwegian Women’s Public Health Association, 2; H. Aulie, None; V. Lilleby, None; B. Flatø, None.

2678 Medically Significant Infections Are Increased In Patients With Juvenile Idiopathic Arthritis Treated With Etanercept. Results From The British Society For Paediatric and Adolescent Rheumatology Etanercept Cohort Study. Rebecca Davies1, Taunton R. Southwood2, Lianne Kearsley-Fleet1, Mark Lunt1, Kimme L. Hyrich3 and on Behalf Of The BSPAR Etanercept Cohort Study1. 1Arthritis Research UK Epidemiology Unit, University of Manchester, Manchester, United Kingdom, 2 Institute of Child Health, University of Birmingham and Birmingham Children’s Hospital, Birmingham, United Kingdom, 3Arthritis Research UK Epidemiology Unit, Manchester Academic Health Sciences Centre, Institute of Inflammation and Repair, The University of Manchester, Manchester, United Kingdom.

Table. Patient characteristics and incidence of serious infection. MTX registered Subjects (n) 182 Exposure (person-years) 466 Total serious infection, n 21 First serious infection, n 21 Upper respiratory tract, n 4

On ETN 677 1498 134 99 23

On ETN monotherapy .. 827 67 47 13

On ETN & MTX in combination .. 671 67 52 10

0 3 5 6 1 0 2 4.5 [2.8, 6.9] ref ..

15 9 12 5 14 7 11 3 6 2 2 1 16 7 8.9 [7.5, 10.6] 8.1 [6.3, 10.3] 2.03 [1.12, 3.65] 1.96 [0.96, 4.00] .. ref

6 7 7 8 4 1 9 10.0 [7.7, 12.7] 2.14 [1.12, 4.10] 1.24 [0.83, 1.86]

* Variables in PD: age, gender, Childhood Health Assessment Questionnaire (CHAQ), Juvenile Arthritis Disease Activity Score (JADAS), disease duration, co-morbidities, concurrent oral steroid use, ILAR category.

Conclusion: ETN therapy is associated with an increased risk of SI in JIA patients. This risk does not increase further for ETN-MTX patients compared to ETN monotherapy patients. Disclosure: R. Davies, None; T. R. Southwood, None; L. Kearsley-Fleet, None; M. Lunt, None; K. L. Hyrich, None; O. B. O. T. BSPAR Etanercept Cohort Study, Pfizer Inc, 2.

2679 Incidence Rates Of Serious Infections and Infection Subtypes Among Pediatric Systemic Lupus Erythematosus Patients Enrolled In Medicaid, According To Medication Use. Linda T. Hiraki1, Candace H. Feldman2, Mary Beth Son3, Jessica M. Franklin4, Michael A. Fischer4, Daniel H. Solomon5, Seoyoung C. Kim6, Wolfgang C. Winkelmayer7 and Karen H. Costenbader8. 1Brigham and Women’s Hospital, Harvard School of Public Health, Boston, MA, 2Division of Rheumatology, Immunology and Allergy, Brigham and Women’s Hospital, Harvard Medical School, Boston, MA, 3Children’s Hospital Boston, Boston, MA, 4Division of Pharmacoepidemiology and Pharmacoeconomics, Brigham and Women’s Hospital, Boston, MA, 5Harvard Medical School, Brigham and Women’s Hospital, Division of Rheumatology, Division of Pharmacoepidemiology, Boston, MA, 6Brigham and Women’s Hospital, Boston, MA, 7Stanford University School of Medicine, Stanford, CA, 8Brigham and Women’s Hospital, Harvard Medical School, Boston, MA. Background/Purpose: We investigated incidence rates of serious infections and infection subtypes among children with SLE enrolled in Medicaid, the U.S. health insurance program for low-income children and parents. Methods: We identified all children aged 3 to ⬍18 years with SLE (ⱖ3 ICD-9 codes of 710.0, each ⬎30 days apart) in the Medicaid Analytic eXtract (MAX) from 2000–2006. This dataset contains all outpatient and inpatient Medicaid claims for enrollees in 47 U. S. states and the District of Columbia. Filled prescriptions were documented and patients were classified as ever/never users of corticosteroids (CS) and immunosuppressants (IS). We identified serious infections from hospital discharge diagnosis codes for all infections, and for specific subtypes of infections (bacterial, fungal and viral). We calculated incidence rates (IR) per 1000 person-years for number of infections ⱖ7 days following first prescription. Incidence rate ratios (IRR) (95% CI) were calculated comparing: 1) receipt of neither CS nor IS, 2) receipt of CS alone and 3) receipt of both CS and IS. We employed Poisson models, adjusted for age, sex and duration of enrollment in Medicaid. Results: Of the 4,068 children identified with SLE. A total of 839 serious infections occurred in 457 children during 8,854 person-years. 2.9% of the children had ⱖ3 infections during their Medicaid enrollment. (Table) Incidence rates for all serious infections requiring hospitalization varied between 43 and 109 per 1000 children per year. Among children with SLE receiving CS alone, incidence rates were approximately twice as high for both all infections and for bacterial infections, and 5.2 times higher for viral infections, compared to those children not receiving CS or IS. Among children receiving both CS and IS, overall serious infection incidence rates were approximately 2.5 times higher, bacterial infections were 2.3 times higher, and viral infections were 7.5 times higher, compared to children not receiving either medication. Fungal infection incidence rates did not appear to be increased among users of CS or IS.

S1143

Tuesday, October 29

Background/Purpose: The association between anti-TNF therapy and increased rate of infection are widely documented in adults with rheumatoid arthritis. Findings in children with juvenile idiopathic arthritis (JIA) have been less consistent and limited by small sample size and restricted follow up times. Dutch and German national registers documented low rates of serious infection (SI) in etanercept (ETN) treated patients with JIA, although one analysis has suggested more frequent SIs when ETN was used in combination with methotrexate (MTX). The aims of this analysis were (1) to compare rates of serious infection in JIA patients treated with ETN vs. MTX and (2) to compare the rates between ETN-MTX combination and ETN monotherapy. Methods: To 31/5/2013, 677 ETN and 182 biologic-naive MTX control patients had been recruited to the BSPAR Etanercept Cohort Study, a UK national register established in 2004 to monitor the safety and effectiveness of ETN in children with JIA. All patients were followed by regular hospital questionnaires. Serious infections were defined as any infection regarded as medically significant by the patient’s clinican. This on-drug analysis followed all patients until first SI, death, treatment discontinuation or last follow-up date, whichever came first. Cox proportional hazards models were used to compare rates of SI between cohorts. Adjustments were made for potential confounders including age, gender, co-morbidities, oral steroid use, disease duration, ILAR category (systemic versus non-systemic) and disease severity, using a propensity score stratified into deciles (PD). Missing baseline data were accounted for using multiple imputation. Results: The ETN cohort were older (mean 10.6 v 7.8 years), with longer disease duration (4.6 v 1.2 years), and had a higher proportion of children with systemic arthritis (15% v 4%). Disease activity was similar between the cohorts. The mean duration of follow up was 2.4 years in the ETN cohort, and 2.6 years in the MTX cohort. A total of 120 first SI’s were reported (99 ETN, 21 MTX). Patients on ETN had higher rates of SI than controls (PD adjusted HR 2.03 (1.12, 3.65)). The risk of SI did not differ between patients receiving ETN as monotherapy and those receiving ETN in combination with MTX (PD adjusted HR 1.24 (95 % CI 0.83, 1.86)) (Table).

Skin & soft tissue, n Lower respiratory tract, n Varicella & zoster, n Other Viral, n Urinary tract, n Bone/joint, n Other or Not specified, n SI incident rate/100 pyrs SI PD adjusted HR*

Table. Incidence Rates and Incidence Rate Ratios for Serious Infections among Children with SLE enrolled in Medicaid, 2000–2006 All Infections IRR (95% CI)

IR (95%CI)

Bacterial Infections IR IRR (95% CI) (95% CI)

Fungal Infections IR IRR (95%CI) (95% CI)

Viral Infections IR IRR (95%CI) (95% CI) 2.1 (2.1, 2.2)

SLE No CS* or IS** nⴝ763

43.6 (43.2, 44.0)

37.9 1.0 (ref) (37.5, 38.2)

3.6 1.0 (ref) (3.5, 3.7)

CS* alone nⴝ1211

86.5 1.97 (86.2, 86.8) (1.55, 2.53)

1.0 (ref)

71.2 1.90 (70.9, 71.5) (1.49, 2.41)

3.7 0.98 (3.7, 3.8) (0.71, 1.35)

11.2 5.16 (11.1, 11.3) (3.37, 7.90)

1.0 (ref)

CS* and IS** 109.0 2.48 nⴝ1872 (108.8, 109.2) (1.92, 3.22)

88.5 2.32 (88.3, 88.7) (1.79, 3.00)

4.4 1.25 (4.3, 4.4) (0.92, 1.70)

15.7 7.51 (15.6, 15.8) (4.32, 13.06)

Incidence rate ratios (IRRs) adjusted for age, sex and Medicaid enrollment duration. Incidence rates (IRs) and IRRs reported per 100,000 person years. *CS: Corticosteroids: oral/IV prednisone, methylprednisolone, dexamethasone, hydrocortisone, prednisolone, cortisone. **IS: Immunosuppressants: mycophenolate mofetil, oral/IV cyclophosphamide, azathioprine, cyclosporine, tacrolimus. Hydroxychloroquine not included. Bacterial infections: cellulitis, endocarditis, pneumonia, pyelonephritis, septic arthritis, osteomyelitis, bacteremia, listerosis. Fungal infections: systemic candidiasis, cryptococcosis, aspergillosis, histoplasmosis, pneumocystic carinii, . Viral infections: cytomegaloviral disease, influenza, herpes zoster, varicella zoster.

Table. Number of events, crude incidence rates (IRs) and adjusted Hazard Ratios (HR) for subsequent hospitalized infection by biologic*

Conclusion: We observed significant variation in incidence rates of serious infections among children with SLE. Those children with SLE who were receiving CS alone or in combination with IS, had much higher rates of bacterial and viral infections, compared to those children receiving neither CS nor IS. Viral infection rates were over 5 times higher among the children receiving these mediations than among those who were not. Disclosure: L. T. Hiraki, None; C. H. Feldman, None; M. B. Son, None; J. M. Franklin, None; M. A. Fischer, None; D. H. Solomon, Lilly, Amgen, CORRONA, 2, Lilly, Novartis, BMS, Pfizer, 6, Lilly, BMS, Novartis, 9; S. C. Kim, Pfizer Inc, 2, Pfizer and Asisa, 9; W. C. Winkelmayer, ACUMEN, 5, Amgen, Bayer, GlaxoSmithKline, 9, Keryx, 9, Medtronic, 9, AJKD, JAMA, 9, American Society of Nephrology Public Policy Board, 6; K. H. Costenbader, None.

Tuesday, October 29

Results: During follow-up of 10,794 hospitalized infections while exposed to anti-TNF therapy, we identified 7,807 person-years of exposure to target biologics and 2,666 subsequent hospitalized infections; of this exposure time 4% was on abatacept, 2% on rituximab and 94% on anti-TNFs, including 23% on etanercept, 18% on adalimumab and 53% on infliximab. Abatacept users had the lowest crude incidence rate of subsequent infection, and etanercept users had the highest. After adjusting for infection risk score decile, the original anti-TNF medication and other potential confounders, abatacept (hazard ratio (HR): 0.80, 95% CI: 0.64–0.99) and etanercept (HR: 0.83, 95% CI: 0.72–0.96) users had significantly lower risks of infection compared to infliximab users.

ACR Concurrent Abstract Session Rheumatoid Arthritis - Clinical Aspects IV: Comorbidities in Rheumatoid Arthritis Tuesday, October 29, 2013, 2:30 PM–4:00 PM

2680 Risk of Subsequent Infection among Rheumatoid Arthritis Patients Using Biologics. Huifeng Yun1, Fenglong Xie1, Elizabeth S. Delzell1, Lang Chen1, Emily Levitan1, James Lewis2, Kenneth G. Saag1, Timothy Beukelman1, Kevin L. Winthrop3, John Baddley1, Paul M. Muntner1 and Jeffrey R. Curtis1. 1University of Alabama at Birmingham, Birmingham, AL, 2University of Pennsylvania, Philadelphia, PA, 3Oregon Health & Science University, Portland, OR. Background/Purpose: Much has been written about infections associated with biologic agents in patients with rheumatoid arthritis (RA). However, less is known about the risk of subsequent infections in RA patients who resume biologic therapy after a serious infection. To compare the subsequent risk of hospitalized infections associated with specific biologic agents among RA patients previously hospitalized for infection while receiving anti-TNF therapy. Methods: Using Medicare data from 2006–2010 for 100% of beneficiaries with RA, we identified patients who were hospitalized with infection while on anti-TNF agents and who had US Medicare fee-forservice hospital, physician and prescription drug coverage continuously in the 6 months before the hospitalization discharge date and throughout follow up. Follow-up began 60 days after hospital discharge and ended at the earliest of subsequent infection, loss of Medicare coverage or after 18 months. We determined biologic exposure on each person-day during follow-up and treated exposure as time varying. Confounding was controlled through a person-specific infection risk score that was separately derived among new users of anti-TNF and non-biologic DMARDs. We calculated the incidence rate of subsequent hospitalized infection for each biologic and used cluster adjusted Cox regression to evaluate the association between specific biologics and subsequent infection, controlling for the decile of the infection risk score, types of anti-TNF use, steroid use during baseline, non-biologic DMARD use during baseline and coexisting biologic exposures.

Biologic Exposure Abatacept Rituximab Etanercept Adalimumab Infliximab

Events 88 38 661 497 1382

PYs† 333 133 1,831 1,423 4,087

IR/100 PYs 26.5 28.5 36.1 34.9 33.8

Crude HR (95% CI) 0.88 (0.71–1.09) 0.93 (0.68–1.28) 1.07 (0.98–1.17) 1.03 (0.93–1.14) 1.0 (ref)

Adjusted HR (95% CI)‡ 0.80 (0.64–0.99) 0.85 (0.62–1.18) 0.83 (0.72–0.96) 0.92 (0.79–1.07) 1.0 (ref)

*Biologic exposure was defined as the days’ supply field from filled prescriptions to which was added a 30-day ‘extension’ period. † Person years. ‡Adjusted for the decile of disease risk score, types of anti-TNF use before the index hospitalization, steroid dose during baseline, methotrexate use during baseline, infection type for the index hospitalization and concurrent biologic exposures during follow up.

Conclusion: Among RA patients who experienced a hospitalized infection while on anti-TNF therapy, the risk of subsequent hospitalized infection was lower for abatacept and etanercept than for other commonly prescribed biologic therapies. Disclosure: H. Yun, None; F. Xie, None; E. S. Delzell, Amgen, 2; L. Chen, None; E. Levitan, Amgen, 2; J. Lewis, Pfizer, Prometheus, Lilly, Shire, Nestle, Janssen, AstraZeneca, Amgen, 5, Centocor, Shire, Takeda, 2; K. G. Saag, Ardea; Regeneron; Savient; Takeda, 5, Ardea; Regeneron; Savient: Takeda, 2; T. Beukelman, Genentech and Biogen IDEC Inc., 5, Novartis Pharmaceutical Corporation, 5, Pfizer Inc, 2; K. L. Winthrop, Pfizer Inc, 2, Pfizer, UCB, Genentech, Regeneron, 5; J. Baddley, BMS, 2, Pfizer Inc, 2; P. M. Muntner, Amgen, Inc, 2, Amgen, Inc, 5; J. R. Curtis, Roche/Genentech, UCB, Janssen, CORRONA, Amgen, Pfizer, BMS, Crescendo, AbbVie, 2, Roche/Genentech, UCB, Janssen, CORRONA, Amgen, Pfizer, BMS, Crescendo, AbbVie, 5.

2681 Reduced Fertility In Women With Rheumatoid Arthritis: Influence Of Disease Activity and Medication Use. Jenny Brouwer, Johanna MW Hazes, Joop SE Laven and Radboud JEM Dolhain. Erasmus University Medical Center, Rotterdam, Netherlands. Background/Purpose: Many female rheumatoid arthritis (RA) patients who try to conceive have a time to pregnancy (TTP) longer than 12 months. During this period RA often cannot be treated optimally. As a result, a longer TTP may result in higher chance of permanent joint damage, with a negative impact on social roles. Thus far, no studies on possible causes of the prolonged TTP in RA patients have been reported. In order to optimize care for these women, we aimed to identify clinical factors associated with a prolonged TTP in RA patients. Methods: From 2002 until 2010, a nationwide prospective cohort study on pregnancy in RA patients (PARA study) was performed. Women with RA according to the 1987 American College of Rheumatology (ACR) criteria were included preconceptionally or during first trimester. All preconceptionally included women were actively trying to become pregnant. To study the effect of various disease characteristics and medications on the TTP, we performed a multivariable Cox regression analysis. Results: Two-hundred-forty-five patients were included. The mean age was 31.3⫾3.9 years. The TTP was longer than 12 months in 42% of patients, of whom 40 women(16%) had not gotten pregnant during the follow up period. A longer TTP was related to various variables. Hazard ratios for occurrence of pregnancy were 0.96 (95%CI 0.92–1.00) per year increase in age; 0.52 (0.38–0.70) for nulliparity; 0.81 (0.71–0.93) per

S1144

point increase in disease activity score (DAS28); 0.61 (0.45–0.83) for preconceptional prednisone use; and 0.66 (0.46–0.94) for non-steroidal anti-inflammatory drug (NSAID) use. Eighty-five patients used prednisone during the preconceptional period. The impact of prednisone use on TTP was dose-dependent, with a significant longer TTP when the daily dose was higher than 7.5 mg. TTP was not affected by smoking, disease duration, rheumatoid factor, anti-citrullinated protein antibodies, past methotrexate use, and preconceptional sulfasalazine use.

naire (HAQ). This register was linked to a regional health care register, which contains information on all inpatient and outpatient procedures in the area, to the national population register to add information on vitality and residential address, and to a regional register of patients with arthritis treated with biologics, which covers ⬎90% of such patients in the region. The total follow-up was calculated for each calendar year from 1998 through 2011, and the annual incidence rate of orthopedic surgery procedures was estimated. The incidence rate for 1998–2001 was compared to those of 2002–2006 and 2007–2011. The impact of demographics and patient reported outcomes on the risk of future orthopedic surgery was analyzed using Cox proportional hazard models. Results: The incidence of all orthopedic surgery procedures during the whole study period was 82,3/1000 person-years (95 % confidence interval (CI) 78,7–86,0). The incidence of all procedures declined significantly over time, with a decrease also for large joint surgery and small joint surgery (hand, wrist, foot and ankle surgery), when studied separately (Table). There was a decline in hip surgery, but not in knee surgery (Table). The incidence of large joint surgery was reduced already in 2002–2006 compared to 1998–2001, whereas a decline in small joint surgery was apparent only in the final period (2007–2011). Female sex was a predictor of orthopedic surgery [age adjusted hazard ratio (HR) 1.50; 95 % CI 1.23–1.83). Greater disability, measured by HAQ, was associated with a higher rate of orthopedic surgery (HR per standard deviation 1.37; 95 % CI 1.25–1.50 adjusted for age, sex and duration of RA), and similar, although weaker, associations were observed for VAS pain and VAS global. Table. Incidence of orthopaedic surgery procedures per 1000 person-years (95% CI)

Conclusion: TTP in female RA patients is longer if patients are older or nulliparous, have higher disease activity scores, use NSAIDs or use prednisone in a daily dose higher than 7.5 mg daily. Treatment strategies during the preconceptional period should try to maximize suppression of disease activity, taking account of possible negative effects of NSAIDs use and use of higher doses of prednisone. Disclosures: This study was funded by the Dutch Arthritis Foundation (Reumafonds). J. Brouwer, None; J. M. Hazes, None; J. S. Laven, None; R. J. Dolhain, None.

2682 Incidence Trends and Predictors Of Orthopedic Surgery In Patients With Rheumatoid Arthritis – Results From a Well Defined Population. Korosh Hekmat1, Lennart Jacobsson1, Jan-Åke Nilsson1, Minna Willim1, Martin Englund2, Ingemar F. Petersson3 and Carl Turesson1. 1 Lund University, Malmo¨, Sweden, 2Lund University, Lund, Sweden, 3 Musculoskeletal Scienes, Department of Orthopedics, Clinical Sciences, Lund, Sweden. Background/Purpose: Orthopedic surgery is used effectively in many patients with severe rheumatoid arthritis (RA). The aim of modern pharmacologic treatment is to prevent joint destruction and reduce the need for orthopedic surgery. Our purpose was to investigate trends in the incidence as well as predictors for such procedures in a population-based sample of patients with RA. Methods: The study was based on a dynamic cohort of all known patients from a defined geographical area with a clinical diagnosis of RA who fulfilled the 1987 American College of Rheumatology criteria for RA. A total of 2342 patients (68.7% women) were included. Questionnaires were sent to the RA patients in the register in 1997, 2002, 2005 and 2009(response rates 62–74 %) including visual analogue scales (VAS) for general health and pain and the health assessment question-

2002–2006 82.6 (76.7–88.8) 16.5 (14.0–19.4) 13.2 (10.9–15.8) 43.4 (39.2–47.9) 37.9 (33.9–42.1)

2007–2011 71.8 (66.1–77.8) 17.6 (14.8–20.7) 12.9 (10.6–15.6) 30.5 (26.9–34.5) 39.3 (35.1–43.8)

P for trend ⬍0.001 ⬍0.001 0.759 ⬍0.001 0.009

Conclusion: The overall incidence of orthopedic surgery declined over time which coincides with increasing use of more intensive pharmacological treatment including the use of biologics. The decrease in large joint surgery predated that of small joint surgery. Improved management may reduce the need for orthopedic interventions in patients with RA. Disclosure: K. Hekmat, None; L. Jacobsson, None; J. Nilsson, None; M. Willim, None; M. Englund, None; I. F. Petersson, None; C. Turesson, None.

2683 Application Of a Multi-Biomarker Disease Activity (Vectra® DA) Score For Assessing Rheumatoid Arthritis Patients With Fibromyalgia Or Low C-Reactive Protein. Yvonne C. Lee1, James Hackett2, Claire Alexander3, Michelle A. Frits1, Christine K. Iannaccone1, Nancy A. Shadick1, Michael E. Weinblatt1, Oscar Segurado3 and Eric H. Sasso3. 1 Brigham and Women’s Hospital, Boston, MA, 2Hackett & Associates, Inc., San Diego, CA, 3Crescendo Bioscience Inc., South San Francisco, CA. Background/Purpose: Clinical assessment of rheumatoid arthritis (RA) may be challenging if patients have fibromyalgia (FM) or if C-reactive protein (CRP) is low (ⱕ1 mg/dL). A multi-biomarker disease activity (MBDA) blood test has been developed to assess RA disease activity with a score (range: 1–100) that is calculated using a validated algorithm for 12 serum protein biomarkers (VCAM-1, EGF, VEGF-A, IL-6, TNF-RI, MMP-1, MMP-3, YKL-40, leptin, resistin, SAA, CRP). The present study evaluated the role of the MBDA score for assessing disease activity in a cohort of established RA patients, including patients with concomitant FM or low CRP. Methods: 208 RA patients from a prospective observational cohort were randomly selected for a substudy of pain in RA. For the present cross-sectional study, DAS28-CRP components, the Widespread Pain Index (to diagnose FM by a modified version of the 2010 ACR Diagnostic Criteria for FM), and the MBDA (Vectra‰ DA) score were evaluated for

S1145

Tuesday, October 29

Any orthopedic surgery Hip surgery Knee surgery Any small joint surgery Any large joint surgery

1998–2001 94.6 (87.3–102.2) 27.8 (23.9–32.0) 12.3 (9.8–15.3) 43.8 (38.9–49.0) 48.5 (43.4–54.0)

the initial substudy visit. 198 patients with non-missing baseline MBDA score and DAS28-CRP components were included. Measures of disease activity were compared between patients with RA⫹FM vs. RA without FM using: t-test or Wilcoxon rank sum tests; multivariate adjustment for age, sex, race, BMI, RF-positivity and non-biologic DMARD use by linear regression or poisson regression (for SJC, TJC); and cumulative probability plots. CRP was compared to MBDA scores by crossclassification. Results: Characteristics of the overall group (N⫽198) included: mean age 58.1 years, 84.8% female, 15.9 years mean duration RA, and 61.6%/60.6% taking a non-biologic/biologic DMARD. Patients with RA⫹FM (N⫽25) vs. RA alone (N⫽173) had similar CRP levels, MBDA scores and swollen joint counts (SJC), whereas the 25 RA⫹FM patients had significantly greater unadjusted values for patient global assessment (PGA) (median 50 vs. 15, p⬍0.001), DAS28-CRP (mean 3.6 vs. 2.8, p⬍0.01) and tender joint counts (TJC) (median 4.0 vs. 1.0, p⫽0.04). Multivariate adjustment gave similar results, but with a larger p-value for TJC (p⫽0.30). CRP levels were ⱕ1, ⬎1 to 3, or ⬎3 mg/dL in 93%, 6%, and 1% of the 198 subjects, respectively. Among those with low CRP (ⱕ1 mg/dL), MBDA scores were low (ⱕ29) in 51%, moderate (30–44) in 36% and high (⬎44) in 13%, with similar findings seen in the RA⫹FM and RA-alone groups. For those with low CRP, the TJC and SJC increased across low to high MBDA categories, suggesting that MBDA differentiated levels of joint inflammation when CRP was low.

Tuesday, October 29

Table 1. Disease activity measures for RA patients who met 2010 ACR diagnostic criteria for FM versus those who did not. Disease Activity Index MBDA score C-reactive protein (mg/dL) Swollen joint count Tender joint count Patient global assessment DAS28-CRP

RA ⴙ FM (N ⴝ 25) 33 2.0 1.0 4.0 50 3.6

RA without FM (N ⴝ 173) 32 1.6 1.0 1.0 15 2.8

Unadjusted P-value1 0.65 0.84 0.38 0.04 ⬍0.001 ⬍0.01

Adjusted P-value2 0.86 0.72 0.40 0.30 ⬍0.001 ⬍0.01

Values for disease activity measures are unadjusted for covariates and are medians, except for MBDA score and DAS-28, which are means. 1 P-values for unadjusted means by t-test; for unadjusted medians by Wilcoxon test. 2 P-values for multivariate (covariate-adjusted) analysis.

resulting in seroconversion in 77–94% of adults after administration of a single dose. The aim of the present study was to investigate impact of different anti-rheumatic treatments on antibody response to pH1N1 vaccination in patients with rheumatoid arthritis (RA) and spondylarthropathy (SpA). Methods: Patients with arthritis (n⫽291; mean age 57 years, 64% women) participated. Hemagglutination inhibition (HI) assay was performed on blood samples drawn before and after a mean (SD) of 8.3 (4) months following vaccination. A positive immune response i.e. seroconverstion was defined as negative prevaccination serum and postvaccination HI titer ⱖ40 or a ’4-fold increase in HI titer. There were 7 treatment groups: 1) RA on methotrexate (MTX); 2) RA on anti-TNF monotherapy; 3) RA on anti-TNF⫹MTX; 4) RA on other biologics (abatacept, rituximab, tocilizumab); 5) SpA on anti-TNF monotherapy; 6) SpA on anti-TNF⫹MTX and 7) SpA on NSAIDs/analgesics. Predictors of positive immune response were studied using logistic regression analysis. Results: The percentage of patients with positive immune response in the different treatment groups were: 1) 42% 2) 53% 3) 43% 4) 20% 10%, 50% 5) 76% 6) 47% and 7) 59%, respectively. RA patients on rituximab had significantly lower (p⬍0.001) and SpA on anti-TNF monotherapy significantly better response rates compared to other treatment groups (p 0.001–0.033). Higher age (p⬍0.001) and current smoking predicted impaired immune response (p⫽0.025). Antibody titers 3–6 months after vaccination were generally lower compared to those within first 3 months but no further decrease in titers were observed 6–22 months after vaccination. Conclusion: Rituximab treatment severely reduced antibody response to pH1N1 influenza vaccine. All other treatments groups showed lower although acceptable antibody response. Serological immunity seems to be stable for 22 months in the current patient population, with the exception of rituximab (and possibly abatacept) treated patients. Disclosure: M. C. Kapetanovic, None; L. E. Kristensen, None; T. Saxne, None; T. Aktas, None; A. Mo¨rner, None; P. Geborek, None.

2685

Conclusion: Patients with RA⫹FM, vs. those with RA alone, had similar MBDA scores and CRP values, but significantly greater DAS28CRP, mostly due to greater PGA. However, MBDA score differed from CRP because MBDA score detected moderate or high disease activity in nearly half of patients with low CRP (ⱕ1 mg/dL). Further study is needed to determine the clinical meaning of discordance between CRP and MBDA scores. Disclosure: Y. C. Lee, Forest Research Insitutute, 2, Merck Pharmaceuticals, 1, Novartis Pharmaceutical Corporation, 1, Cubist, 1, Elan Corporation, 1; J. Hackett, Crescendo Bioscience, Inc., 5; C. Alexander, Crescendo Bioscience, Inc., 1, Crescendo Bioscience, Inc., 3; M. A. Frits, None; C. K. Iannaccone, None; N. A. Shadick, MedImmune, 2, Crescendo Biosciences, 2, Amgen, 2, Abbott Immunology Pharmaceuticals, 2, Genentech and Biogen IDEC Inc., 2; M. E. Weinblatt, MedImmune, 2, Crescendo Bioscience, Inc., 2, Bristol Myers Squibb, 2, MedImmune, 5, Crescendo Bioscience, Inc., 5, Bristol Myers Squibb, 5; O. Segurado, Crescendo Bioscience, Inc., 1, Crescendo Bioscience, Inc., 3; E. H. Sasso, Crescendo Bioscience Inc., 1, Crescendo Bioscience Inc., 3.

2684 Impact Of Anti-Rheumatic Treatment On Immunogenicity Of Pandemic H1N1 Influenza Vaccine In Patients With Arthritis. Meliha C. Kapetanovic1, Lars-Erik Kristensen2, Tore Saxne3, Teodora Aktas4, Andreas Mo¨rner4 and Pierre Geborek3. 1Dept of Clinical Sciences Lund, Section of Rheumatology, Lund University, Lund, Sweden, 2Department of Clinical Sciences, Section of Rheymatology, University Hospital of Skåne, Lund, Sweden, 3Lund University, Lund, Sweden, 4 Vaccinology Unit, Department of Diagnostics and Vaccinology, Swedish Institute for Communicable Disease Control, Solna, Sweden, Solna, Sweden. Background/Purpose: An adjuvanted pandemic H1N1 influenza (pH1N1) vaccine (Pandemrix®) was reported as highly immunogenic

Anti-Peptidylarginine Deiminase 3/4 Cross-Reactive Antibodies Are Associated With Radiographic Interstitial Lung Disease In Rheumatoid Arthritis, An Effect Potentiated By Smoking. Jon T. Giles1, Erika Darrah2, Sonye K. Danoff2, Cheilonda Johnson2, Felipe Andrade3, Antony Rosen2 and Joan M. Bathon4. 1Columbia University, College of Physicians & Surgeons, New York, NY, 2Johns Hopkins University, Baltimore, MD, 3Division of Rheumatology, Johns Hopkins University School of Medicine, Baltimore, MD, 4Columbia University, New York, NY. Background/Purpose: Antibodies targeting citrullinated proteins (ACPA) are implicated in the pathogenesis of interstitial lung disease (ILD) in rheumatoid arthritis (RA). Citrullinated proteins and the enzymes that catalyze citrullination, the peptidylarginine deiminases (PADs), are detected in RA-ILD lung tissue and in the lungs of heavy smokers. A subset of RA patients demonstrates cross-reactive antibodies against PADs 3 and 4 that have been shown to lower the calcium threshold required for PAD activation, an effect potentially relevant to the pathogenesis of RA-ILD. We sought to explore the association between anti-PADs and radiographic RA-ILD. Methods: RA patients participating in a study of subclinical cardiovascular disease underwent multi-detector computed tomography (MDCT) of the chest with interpretation by a pulmonary radiologist for ILD features. A semi-quantitative ILD Score (ILDS; range 0–32) was calculated. Concurrent serum samples were assessed for antibodies against PAD by immunoprecipitation of S35-labeled PAD3 and 4. Results: Among the 176 RA patients studied [60% female, 86% Caucasian, mean age 59⫾9 years, 11% current smokers, median RA duration⫽8 years, median DAS28⫽3.7], any CT-ILD was observed in 58 (33%). Anti-PAD3/4 cross-reactive antibodies were detected in 19 (11%) and 37 (21%) had antibodies directed against PAD4 alone. In univariate analysis, the frequency of any CT-ILD among those with anti-PAD3/4 was 68%, vs. 27 and 29% among those with anti-PAD4 only and neither reactivity, respectively (crude OR⫽5.39; p⫽0.001 for the comparison of anti-PAD3/4 vs. no anti-PAD). The association was stronger after adjustment for relevant demographic and RA disease/treatment confounders

S1146

(Fig 1A: adjusted OR⫽7.22; p⫽0.001). Anti-PAD3/4 antibodies were significantly associated with all CT-ILD predominant patterns (i.e. ground glass, honeycombing, etc. . .) and quantitative ILD scores were significantly higher for the anti-PAD3/4 vs. the no anti-PAD group (median ILD score 2 vs. 0 units, respectively; p⫽0.020). The association of anti-PAD3/4 antibodies with CT-ILD was stronger in ever smokers than never smokers. Among never smokers, the adjusted frequency of any ILD was 39 vs. 17%, respectively, for those with vs. without anti-PAD3/4 (Fig 1B, never smokers: OR⫽3.01; p⫽0.19) compared with 93 vs. 30%, respectively, for ever smokers with vs. without anti-PAD3/4 (Fig 1B, ever smokers: OR⫽29.5; p⫽0.004, p-value for interaction⬍0.05).

Methods: MUSICA was a double-blind, randomized, parallel-arm study to examine the impact of MTX dose on disease outcomes and ultrasonographic signs in moderately to severely active RA pts who have failed prior synthetic DMARDs. The study enrolled 309 pts taking MTX ⱖ 15 mg/week (wk) for ⱖ 12 wks prior to enrollment. Patients blindly received either high (20 mg/wk) or low dose (7.5 mg/wk) MTX; all pts received open-label 40 mg ADA every other wk for 24 wks. Non-inferiority was assessed using the 95% confidence interval of the difference between high and low dose wk-24 outcomes. A 15% margin (0.56) of the high dose mean 24-wk 28-joint disease activity score (DAS28) was used as the primary endpoint. Ultrasound images acquired every 4 wks were independently read and scored blindly by 4 ultrasoundexperienced rheumatologists. A 10-joint semi-quantitative scoring system incorporating OMERACT definitions for pathology (1) measured synovial hypertrophy, vascularity, and bony erosions. Results: The study populations for both MTX dosages were well balanced for baseline demographics and disease characteristics and had overall age 54.8, 5.3 years RA disease duration, and a DAS28 of 5.8 (all means). Rapid improvement in clinical indices was seen in both groups after addition of ADA. After 24 wks of ADA combination therapy, pts receiving combination therapy with ADA displayed improvements consistent with other trials including low dose MTX. Differences in clinical and ultrasonographic outcomes comparing low vs high dose MTX were minimal. The primary endpoint, mean DAS28, did not meet noninferiority criteria. Although outcomes favored maintaining 20 mg MTX, no statistically significant differences were detected for most clinical, functional, and ultrasound outcomes (Table). Statistically significant differences were only detected in wk-24 swollen joint count and physician’s global assessment of disease activity. The number of adverse events (AEs), serious AEs, and infectious AEs were fewer in the low dose MTX arm.

Conclusion: The prevalence of CT-ILD was markedly higher among RA patients with anti-PAD3/4 antibodies, even after accounting for relevant confounders, particularly among ever smokers. Further mechanistic studies are needed to determine whether there is a biologic interaction between smoking and pulmonary hyper-citrullination facilitated by anti-PAD3/4 antibodies that contributes to ILD pathogenesis. Disclosure: J. T. Giles, None; E. Darrah, None; S. K. Danoff, None; C. Johnson, None; F. Andrade, None; A. Rosen, Sanofi-Aventis Pharmaceutical, 5, Inova Diagnostics, Inc., 7; J. M. Bathon, None.

ACR Concurrent Abstract Session Rheumatoid Arthritis Treatment - Small Molecules, Biologics and Gene Therapy: Novel Treatment Strategies in Rheumatoid Arthritis Tuesday, October 29, 2013, 2:30 PM–4:00 PM

Non-Inferiority Parameter a DAS28(CRP) Clinical and Functional Outcomes* ACR20, n/N (%) ACR50, n/N (%) ACR70, n/N (%) HAQ-DI HAQ-DI improvement ⱖ 0.22, n/N (%) TJC (0–68) SJC (0–66) PGA disease activity (100 mm VAS) PtGA pain (100 mm VAS) PtGA disease activity (100 mm VAS) Ultrasound Outcomes Synovial hypertrophy Synovial vascularity Synovial vascularity improvement by 30%, n/N (%) Synovial vascularity change from baseline Bony erosions

ADA ⴙ 7.5 mg MTX (N ⴝ 154) 4.11 (3.88, 4.34)

ADA ⴙ 20 mg MTX (N ⴝ 155) 3.75 (3.52, 3.97)

86/151 (57.0) 45/151 (29.8) 20/151 (13.2) 0.98 ⫾ 0.75 96/151 (63.6) 15.1 ⫾ 16.0 9.6 ⫾ 12.8 23.7 ⫾ 21.3 40.0 ⫾ 27.9 38.2 ⫾ 27.5

95/154 (61.7) 58/154 (37.7) 31/154 (20.1) 0.95 ⫾ 0.78 101/154 (65.6) 13.5 ⫾ 15.3 7.7 ⫾ 10.5 19.0 ⫾ 17.4 36.4 ⫾ 27.7 33.8 ⫾ 26.8

Difference (Low - High) 0.37 (0.07, 0.66) p value 0.395 0.145 0.114 0.476 0.707 0.264 0.028 0.035 0.141 0.094

32.6 ⫾ 7.2 4.43 ⫾ 4.86 65/143 (45.5)

32.9 ⫾ 6.6 4.09 ⫾ 4.53 77/147 (52.4)

0.955 0.779 0.221

⫺1.52 ⫾ 4.18 1.36 ⫾ 2.43

⫺1.46 ⫾ 3.43 1.41 ⫾ 2.01

0.779 0.598

a

2686 Impact of Methotrexate Dose Reduction Upon Initiation of Adalimumab On Clinical and Ultrasonographic Parameters in Patients With Moderate to Severe Rheumatoid Arthritis. Gurjit S. Kaeley1, Amy M. Evangelisto2, Midori Jane Nishio3, Shufang Liu4 and Hartmut Kupper5. 1University of Florida, Jacksonville, FL, 2Arthritis, Rheumatic and Back Disease Associates, Voorhees, NJ, 3Diablo Clinical Research, Walnut Creek, CA, 4AbbVie Inc., North Chicago, IL, 5AbbVie Deutschland GmbH & Co. KG, Ludwigshafen, Germany. Background/Purpose: Methotrexate (MTX) is the recommended first-line disease-modifying antirheumatic drug (DMARD) for the treatment of moderately to severely active rheumatoid arthritis (RA). Whether patients (pts) with inadequate response to MTX that begin combination therapy with adalimumab (ADA) can reduce MTX dose remains unclear. Clinical and ultrasonographic outcomes of MTX dose reduction when initiating ADA were examined.

Primary endpoint, mean (95% CI). *Mean ⫾ SD unless noted otherwise. Means were calculated using last observation carried forward (LOCF). ADA, adalimumab; MTX, methotrexate; DAS28, 28-joint disease activity score; CRP, Creactive protein; ACR, American College of Rheumatology; HAQ-DI, disability index of the health assessment questionnaire; TJC, tender joint count; SJC, swollen joint count; PGA, physician’s global assessment; VAS, visual analogue scale; PtGA, patient’s global assessment.

Conclusion: Addition of ADA to MTX inadequate responders led to robust results consistent with prior studies. Compared to 20 mg MTX, lowering MTX weekly dose to 7.5 mg in combination with ADA resulted in small differences in clinical and ultrasonographic outcomes in RA pts. Statistically, non-inferiority was not met for mean DAS28. Based on the small differences in most outcomes, MTX dose reduction may be considered when initiating ADA therapy in MTX inadequate responders. Disclosure: G. S. Kaeley, AbbVie Inc., 5; A. M. Evangelisto, AbbVie Inc., 2, AbbVie Inc., 8; M. J. Nishio, None; S. Liu, AbbVie Inc., 3, AbbVie Inc., 1; H. Kupper, AbbVie, 3, AbbVie, 1.

S1147

Tuesday, October 29

Table. Week 24 Clinical, Functional, and Ultrasonographic Outcomes

2687

Tuesday, October 29

Very High Remission Rates Are Achieved By Methotrexate and Intraarticular Glucocorticoids Independent Of Induction Therapy With Adalimumab; Year 2 Clinical Results Of An InvestigatorInitiated Randomised, Controlled Clinical Trial Of Early, Rheumatoid Arthritis (OPERA). Kim Hørslev-Petersen1, Merete L. Hetland2, Peter Junker3, Jan Pødenphant4, Torkell Ellingsen5, Palle Ahlqvist6, Hanne M. Lindegaard7, Asta Linauskas8, Annette Schlemmer9, Mette Y. Dam10, Ib Hansen11, Tine Lottenburger6, Anette Jørgensen10, Sophine B. Krintel12, Johnny Raun1, Christian G. Ammitzbøll10, Julia Johansen12, Mikkel Østergaard13 and Kristian Stengaard-Pedersen10. 1University of Southern Denmark, Graasten, Denmark, 2DANBIO, Center for Rheumatology and Spine Diseases, Glostrup Univ Hospital, Glostrup, Denmark, 3 University of Southern Denmark, Odense, Denmark, 4Copenhagen University at Gentofte, Hellerup, Denmark, 5Silkeborg Regional Hospital, Silkeborg, Denmark, 6University of Southern Denmark, Vejle, Denmark, 7 Odense University Hospital, Odense, Denmark, 8Vendsyssel Hospital, Hjørring, Denmark, 9Aalborg University Hospital, Aalborg, Denmark, 10 Arhus University Hospital, Aarhus, Denmark, 11Viborg Hospital, Viborg, Denmark, 12Copenhagen University and Glostrup Hospital, Copenhagen, Denmark, 13Copenhagen University Hospital Glostrup, Copenhagen, Denmark. Background/Purpose: In a double-blind placebo-controlled 2-year investigator-initiated trial of patients with early rheumatoid arthritis (RA), we investigated if additional adalimumab (ADA) for 1 year on top of an aggressive treat-to-target strategy with methotrexate (MTX) and intraarticular (i.a.) corticosteroid reduced disease activity. Previously published results after 1 year showed that ⬎75% of patients in both groups achieved low disease activity, but remission rates and physical function were higher in the ADA group1. Here we present results from the 2nd year. Methods: DMARD-naı¨ve early RA patients (n⫽180) were randomized 1:1 to receive i.a. triamcinolone (40 mg/ml) in any swollen joint and MTX (20 mg/wk) for two years in combination with placebo-ADA (MTX⫹PLA) or MTX⫹ADA (40 mg eow) during the first year. Oral glucocorticoids were not allowed. ADA/PLA was withdrawn after 1 yr. During year 2, in both treatment arms, ADA (40 mg eow) was only (re)initiated in patients with recurrence of active disease (DAS28 CRP⬎3.2). Clinical response was assessed by DAS28(CRP), clinical disease activity index (CDAI), simplified disease activity index (SDAI) and ACR/EULAR remission criteria. Physical function was assessed by HAQ (health assessment questionnaire). Results: Baseline characteristics were similar in the two groups: MTX⫹PLA/MTX⫹ADA: DAS28(CRP) 5.6(3.8–7.3)/5.5(3.8–7.8), p⫽0.53. After 2 years the median MTX dose and cumulated dose of i.a. triamcinolone were similar in the two groups (Table), and biologics were (re)initiated in 15%/17% of patients. During the 2ndyear, disease activity (DAS28CRP, CDAI and SDAI) and HAQ scores decreased and ACR/ EULAR remission rates increased in the MTX⫹PLA group. After 2 years, remission rates in MTX⫹PLA/MTX⫹ADA groups were: DAS28CRP remission: 69%/66%; CDAI remission: 55%/57%; SDAI remission: 54%/50%; ACR/EULAR(28 joints):44%/45% with no significant differences between MTX⫹PLA/MTX⫹ADA in any clinical outcome measure (p⫽0.36–1.00). Table. Doses, disease activity, and remission rates at one and two years Year 1 Treatment MTX dose mg/wk I.a. triamcinolone (ml cumulated) Triple DMARD Biologics (open trial) DAS28CRP CDAI SDAI DAS28CRP⬍3.2 DAS28CRP⬍2.6 CDAIⱕ2.8 SDAI⬍3.3 ACR/EULAR Boolean (28) ACR/EULAR Boolean (40) HAQ

MTXⴙ 1st YR PLACEBO

MTXⴙ 1st YR ADA

20 (15–20) 7 (2–18.8) 32 % 7% 2.6 (1.7–4.7) 3.9 (0–13.6) 5.0 (0.8–20.2) 76% 49% 41 % 36 % 30 % 29 % 0.25 (0–1.44)

Year 2 P

MTXⴙ 1st YR PLACEBO

MTXⴙ 1st YR ADA

P

20 (7.5–20) 5.4 (1.8–17.4)

0.17 0.08

20 (10–20) 0 (0–7)

20 (7.6–20) 0 (0–7.7)

0.33 0.19

16 % 7% 2.0 (1.7–5.2) 1.9 (0–15.4) 2.7 (0.7–30.4) 80% 74% 61 % 57 % 48 % 49 % 0.13 (0–1.5)

0.018 1.00 0.009 0.01 0.006 0.65 0.001 0.01 0.007 0.017 0.014 0.40

31 % 15 % 2.0 (1.7–4.5) 1.9 (0.1–14.5) 2.8 (0.7–19.0) 84% 69% 55 % 54 % 44 % 44 % 0.13 (0–1.63)

20 % 17 % 2.0 (1.7–4.4) 2.2 (0–15.2) 3.3 (0.7–17.7) 83% 66% 57 % 50 % 45 % 42 % 0.13 (0–1.5)

0.15 0.97 0.97 0.75 0.36 1.00 0.79 0.87 0.66 1.00 0.91 0.37

Values are medians (5%/95% percentiles) or percentage. We used Mann-Whitney or Pearson’s chisquare tests. Analysis was by ITT with last observation carried forward. Completer analysis and ITT without imputations gave similar results.

Conclusion: An aggressive treat-to-target strategy of i.a. triamcinolone and methotrexate in early RA provided excellent disease control at 2 years’ follow-up independent of 1 year induction therapy with adalimumab. 1

Hørslev-Petersen K et al. Ann Rheum Dis Online First 7 mar 2013

Disclosure: K. Hørslev-Petersen, UCB, 8; M. L. Hetland, None; P. Junker, None; J. Pødenphant, None; T. Ellingsen, None; P. Ahlqvist, None; H. M. Lindegaard, Lilly, MSD, Nordpharma, Roche, 8; A. Linauskas, None; A. Schlemmer, MSD, UCB, 8; M. Y. Dam, None; I. Hansen, None; T. Lottenburger, None; A. Jørgensen, None; S. B. Krintel, None; J. Raun, None; C. G. Ammitzbøll, None; J. Johansen, None; M. Østergaard, None; K. StengaardPedersen, None.

2688 Methotrexate Dose Has Minimal Effects On Methotrexate-Related Toxicity In Patients With Early Rheumatoid Arthritis Treated In Combination With Adalimumab – Results Of Concerto Trial. Gerd R. Burmester1, Alan J. Kivitz2, Ronald F. van Vollenhoven3, Stefan Florentinus4, Piyalal M. Karunaratne5, Hartmut Kupper6, Maxime Dougados7 and Roy M. Fleischmann8. 1Charite´ - Universita¨tsmedizin Berlin, Berlin, Germany, 2Altoona Center for Clinical Research, Duncansville, PA, 3The Karolinska Institute, Stockholm, Sweden, 4AbbVie, Rungis, France, 5 AbbVie Inc., North Chicago, IL, 6AbbVie Deutschland GmBH & Co KG, Ludwigshafen, Germany, 7Rene´ Descartes University, Paris, France, 8 University of Texas Southwestern Medical Center, Dallas, TX. Background/Purpose: Adalimumab is usually used in combination with 15–20 mg methotrexate (MTX) weekly in early rheumatoid arthritis (RA). Lower doses of MTX in combination with biologics have not been evaluated in controlled clinical trials. The objective of this study was to evaluate the MTX dose-response of MTX-related toxicities in patients (pts) with early RA receiving ADA in combination with MTX. Methods: CONCERTO was a 26-week (wk), phase 3, double-blind, parallel-arm study in MTX-naive pts with active RA ⬍1 year in duration. Pts were randomized 1:1:1:1 to open-label ADA 40 mg every other wk ⫹ blinded weekly oral MTX doses of 2.5, 5, 10, or 20 mg. All pts took 5 mg folic acid weekly throughout the study. Pts in the 20 mg arm started with 10 mg MTX, with bi-weekly increases of 2.5 mg through wk 8. MTX-toxicity related adverse events (AEs) were defined according to MTX prescribing information and recorded as AEs at each visit through wk 26. Laboratory data were summarized through wk 26. Results: Of the 395 randomized pts, 358 (91%) completed 26 wks. The safety population included 98, 100, 99, and 98 pts in the 2.5, 5, 10, and 20 mg arms, respectively. 9 pts (2.3%) discontinued study drug due to an AE that was not necessarily MTX-related. Of the total population, the percentage of pts with any AE was 62, 59, 67, and 69%, any serious AE was 5, 2, 3 and 7%, and any infection was 20, 17, 24, and 35% in the 2.5, 5, 10, and 20 mg arms, respectively. Two infections in the 5 mg arm were serious AEs, appendicitis and sepsis, but did not result in study discontinuation and were not associated with MTX. The overall incidence of MTX-related AEs was low (Table). Infection and abnormal hair loss appeared to have a MTX dose relationship. No differences in mean change from baseline in hematocrit, neutrophil count, platelet count, ALT, or AST values were observed at wk 26 among the 4 doses of MTX. The percentage of pts with ALT values ⬎upper limit of normal (ULN) that were associated with MTX at wk 26 were 3, 1, 6, and 6% in the 2.5, 5, 10, and 20 mg arms, respectively. Regarding AST, 4, 0, 3, and 3% in the 2.5, 5, 10, and 20 mg arms, respectively, had AST values ⬎ULN. Table. Adverse Events Reported by Investigators to be Associated with MTX

Infection Nausea &/or vomiting Stomach pain/ discomfort Abnormal hair loss Excessive fatigue &/ or malaise Dizziness Oral ulcers Fever &/or chills

S1148

ADAⴙ2.5 mg MTX (Nⴝ98) n (%) 6 (6.1) 6 (6.1) 5 (5.1)

ADAⴙ5 mg MTX (Nⴝ100) n (%) 7 (7.0) 3 (3.0) 4 (4.0)

ADAⴙ10 mg MTX (Nⴝ99) n (%) 11 (11.1) 13 (13.1) 7 (7.1)

ADAⴙ20 mg MTX (Nⴝ98) n (%) 13 (13.3) 8 (8.2) 7 (7.1)

Total (Nⴝ395) n (%) 37 (9.4) 30 (7.6) 23 (5.8)

1 (1.0) 4 (4.1)

5 (5.0) 1 (1.0)

5 (5.1) 3 (3.0)

8 (8.2) 2 (2.0)

19 (4.8) 10 (2.5)

4 (4.1) 0 0

0 1 (1.0) 1 (1.0)

4 (4.0) 5 (5.1) 3 (3.0)

1 (1.0) 2 (2.0) 0

9 (2.3) 8 (2.0) 4 (1.0)

Conclusion: Overall, the combination of ADA plus varying doses of MTX was well tolerated, and MTX-related AEs were infrequent. The frequency of infections and abnormal hair loss appeared to have a MTX-dose relationship. Serious infections were very rare and had no relationship to MTX. No differences in mean change from baseline in blood cell counts or transaminases were observed at wk 26 among the 4 doses of MTX. Disclosure: G. R. Burmester, AbbVie, Essex/Schering-Plough, Novartis, Roche, and Wyeth, 2, AbbVie, Essex/Schering-Plough, Novartis, Roche, and Wyeth, 5, AbbVie, Essex/Schering-Plough, Novartis, Roche, and Wyeth, 8; A. J. Kivitz, AbbVie, Amgen, AstraZeneca, BMS, Celgene, Genentech, Janssen, Pfizer, and UCB, 2, AbbVie, BMS, Genentech, Pfizer, and UCB, 5, Pfizer and BMS, 8; R. F. van Vollenhoven, AbbVie, BMS, GSK, Human Genome Sciences, Merck, Pfizer, Roche, and UCB, 2, AbbVie, BMS, GSK, Human Genome Sciences, Merck, Pfizer, Roche, and UCB, 5; S. Florentinus, AbbVie, 1, AbbVie, 3; P. M. Karunaratne, AbbVie, 1, AbbVie, 3; H. Kupper, AbbVie, 1, AbbVie, 3; M. Dougados, AbbVie, BMS, Merck, Novartis, Pfizer, SanofiAventis, and UCB, 2, AbbVie, BMS, Merck, Novartis, Pfizer, Sanofi-Aventis, and UCB, 5; R. M. Fleischmann, AbbVie, Pfizer, Merck, Roche, UCB, Celgene, Amgen, Astra-Zeneca, BMS, Janssen, Lilly, and Novartis, 2, AbbVie, Pfizer, Merck, Roche, UCB, Celgene, Amgen, Astra-Zeneca, BMS, Janssen, Lilly, and Novartis, 5.

2689

Background/Purpose: In the PRIZE study patients (pts) with early (mean 6 mos. since symptom onset) moderate-severe rheumatoid arthritis (RA), who achieved remission after open-label treatment with 50 mg etanercept (ETN) ⫹ 10–25 mg MTX (Phase 1)1 were randomized to a double-blind 39-wk period of (25mg) ETN25⫹ MTX, MTX alone, or placebo (PBO, Phase 2)2; the study drugs were then withdrawn (randomized treatment remained blinded) and pts were monitored for an additional 26 weeks (Phase 3). Using patients who achieved remission in Phase 1, the objective of this analysis was to assess sustained remission and other clinical and safety outcomes in randomized pts (Phase 2) and subsequent withdrawal of drug treatments (Phase 3). Methods: Pts with DAS28 remission (⬍2.6, n⫽193) began Phase 2 at wk 52; those maintaining DAS28 LDA by wk 91 (ⱕ3.2, n⫽131) continued and had study drugs withdrawn (MTX tapered to week 95) through wk 117 (end of study). Remission and other standard clinical outcomes were assessed during Phase 2 and 3. LOCF was used for missing data. Fisher’s exact test was used for significance. Results: Between wks 52 and 91, the proportion of pts in the ETN25⫹MTX group who maintained DAS28 remission declined slower than those treated with MTX or PBO; at wk 117, significantly more pts in the ETN25⫹MTX group maintained remission relative to PBO following withdrawal of randomized treatment (Figure). Other outcomes had similar results. At the beginning of Phase 2, the proportions of pts in the ETN25⫹MTX group with DAS28 LDA, Boolean remission, ACR50/70 and normal HAQ (ⱕ0.5) were 100%, 71%, 92%/83%, and 84%, respectively; MTX group: 100%, 72% 95%/87% and 86%, respectively; PBO group: 100%, 63%, 95%/82%, and 75%, respectively. At wk 91, the proportions declined to 89%, 68%, 78%/71%, and 78% (ETN25⫹MTX); 69%, 46%, 71%/62%, and 72% (MTX); 46%, 23%, 45%/37%, and 45% (PBO), respectively. After treatment withdrawal at wk 117, the proportions of pts who maintained these outcomes were 56%, 47%, 46%/41%, and 67% (ETN25⫹MTX), 43%, 25%, 38%/29%, and 59% (MTX), 37%, 16%, 35%/23%, and 40% (PBO), respectively. There were no unexpected safety findings.

Disclosure: P. Emery, Abbott Laboratories, Bristol-Myers Squibb, Merck, Novartis, Pfizer Inc, Roche-Chugai, UCB Pharma Ltd, 5; W. Spieler, None; M. Stopinska-Polaszewska, None; N. I. Korshunov, Pfizer Inc, Merck, Roche, Novarits, 5; J. Bukowski, Pfizer Inc, 1, Pfizer Inc, 3; R. Pedersen, Pfizer Inc, 3; T. Williams, Pfizer Inc, 1, Pfizer Inc, 3; S. Gaylord, Pfizer Inc, 1, Pfizer Inc, 3; B. Vlahos, Pfizer Inc, 1, Pfizer Inc, 3.

2690 Survival Of The Second Biologic After The First Tumor Necrosis Factor Alpha Inhibitor’s Failure In The Treatment Of Rheumatoid Arthritis: Prospective Observational Data From Biorx. Si Registry. ŽIga Rotar1 and Matija Tomsic2. 1University Medical Centre Ljubljana, 1000 Ljubljana, Slovenia, 2University Medical Centre Ljubjana, Ljubljana, Slovenia. Background/Purpose: Current recommendations for management of RA propose a tumor necrosis factor alpha inhibitor (TNFi) for patients failing to achieve the treatment target with synthetic DMARDs. If 1st TNFi fails, a different TNFi, abatacept, rituximab (RTX) or tocilizumab (TCZ) are equally recommended. It has been shown that non TNFi biologics are at least as effective as the TNFi in this setting. Since there is little is known about the survival of 2nd biologic agent, we decided to investigate the survival of the 2nd biologic after switching from the 1st TNFi.

Figure 1.

S1149

Tuesday, October 29

Assessing Maintenance Of Remission After Withdrawal Of Etanercept Plus Methotrexate, Methotrexate Alone, Or Placebo In Early Rheumatoid Arthritis Patients Who Achieved Remission With Etanercept and Methotrexate: The Prize Study. Paul Emery1, Wolfgang Spieler2, Maria Stopinska-Polaszewska3, Nikolay I Korshunov4, Jack Bukowski5, Ronald Pedersen5, Theresa Williams6, Stefanie Gaylord6 and Bonnie Vlahos6. 1Institute Rheumatic and Musculoskeletal Medicine University of Leeds, Leeds, United Kingdom, 2ZeFOR GmbH Zentrum fu¨r Forschung, Zerbst, Germany, 3Family Physicians Specialists Clinic, Torun, Poland, 4State Budgetary Healthcare Institution of Yaroslavl Region, Yaroslavl, Russia, 5Pfizer Inc, Collegeville, PA, 6Pfizer Inc., Collegeville, PA.

Conclusion: Of the early RA pts with DAS28 remission at the beginning of Phase 2, those randomized to a reduced dose of ETN had a modest loss of efficacy in Phase 2, but randomization to PBO and withdrawal of ETN in Phase 3 both led to steep declines in all outcomes.

(as a group), RTX and TCZ are presented in Figure 2. 2nd TNFi failed due to insufficient efficacy in 90%, and adverse events in 8%.

Table 1.

Tuesday, October 29

Line of biologic DMARD Biologic DMARD Patients iTNF used % adalimumab % etanercept % infliximab % certolizumab % golimumab Gender (% female) mean age at starting bio-DMARD [yr] (SD) mean age at diagnosis [yr] (SD) mean time to bioDMARD [yr] (SD) % RF positive %ACPA positive % with history of smoking % current % past prior synth-DMARDs median (Q1;Q3) % concomitatnt DMARD % methotrexate mean methotrexate dose [mg] (SD) % leflunomide % concomitant prednisolone mean prednisolone dose [mg] (SD) Mean baseline disease activity DAS28ESR (SD) DAS28CRP (SD) Promis HAQ (SD) ESR [mm/h] (SD) CRP [mg/L] (SD) PGA (0–100) (SD) Pain (0–100) (SD) EGA (0–100) (SD) TJC28 (SD) SJC28 (SD)

1st iTNF

2nd iTNF

Rituximab

Tocilizumab

688

133

34

75

43.6 30.7 10.7 9.9 5.1 81.5 54.7 (11,5)

33.8 39.9 9.8 15.0 1.5 78.2 55.0 (11.4)

/ / / / / 76.5 56.9 (12.2)

/ / / / / 90.7 57.6 (11.3)

p⫽ 0.056‡ p⫽ 0.272*

45.2 (11.8)

45.2 (11.4)

45.5 (11.8)

45.7 (12.0)

p⫽ 0.948*

9.8 (8.1)

9.9 (6.9)

10.6 (7.0)

11.9 (8.4)

p⫽ 0.286†

81,8 77.0 26.2

81.2 73.7 28.6

91.2 96.5 29.4

77.3 74.5 28.0

p⫽ 0.227‡ p⫽0.008‡ p⫽0.988‡

15.1 11.1 3 (3;4)

18.9 9.8 4 (3;4)

23.5 5.9 4 (3;5)

18.7 9.3 4 (3;5)

p⫽ 0.808‡ p⫽ 0.777‡ p⫽ 0.313†

82.0

72.2

79.4

80.0

p⫽ 0.290‡

71.1 17.4 (4.2)

55.6 16.8 (3.6)

79.4 17.6 (3.3)

68.0 16.1 (3.3)

p⫽0.020‡ p⫽ 0.252*

10.5 45.3

15.8 48.1

0.0 41.2

12.0 52.0

p⫽0.044‡ p⫽ 0.576‡

6.9 (3.4)

7.3 (3.3)

7.7 (3.0)

5.7 (1.9)

p⫽0.011*

6.4 (1.0) 5.9 (1.0) 44.7 (24.3) 39.4 (22.6) 23.5 (28.2) 66.9 (22.9) 67.7 (22.0) 62.0 (22.5) 14.3 (6.6) 12.8 (5.5)

6.1 (1.1) 5.7 (1.1) 41.1 (22.7) 38.1 (24.2) 24.7 (27.1) 66.0 (23.8) 66.7 (21.5) 59.9 (21.5) 12.6 (6.9) 12.4 (6.7)

6.3 (1.0) 5.8 (1.1) 42.1 (25.0) 42.2 (25.5) 35.1 (37.7) 64.1 (18.7) 62.2 (21.9) 63.3 (19.7) 12.7 (6.9) 13.2 (5.3)

6.5 (1.0) 6.0 (1.0) 48.6 (23.7) 21.7 (25.1) 23.0 (2.4) 70.5 (20.9) 63.9 (22,3) 65.7 (19.5) 14.8 (6.2) 12.9 (5.8)

p⫽0.043* p⫽ 0.139* p⫽ 0.094* p⫽ 0.469† p⫽ 0.404† p⫽ 0.151† p⫽ 0.678† p⫽ 0.132† p⫽0.033† p⫽ 0.632†

Conclusion: After the 1st TNFi fails, a 2nd TNFi is more likely to fail earlier than RTX or TCZ (p⫽0.000). There is a trend of better survival of RTX vs TCZ, which did not reach statistical significance (p⫽0.057). Disclosure: Rotar, None; M. Tomsic, None.

Figure 2.

Methods: Data was extracted from the mandatory nationwide registry of patients with rheumatoid arthritis treated with biologics on Dec 15th, 2012. Kaplan Meier survival analysis was performed. Statistical significance was determined using the Log-Rank and Wilcoxon tests. Results: From the dataset collected between February 2007 and December 2012 we identified 688 RA patients who received TNFi as a first biologic. Flow of patients is depicted in Figure 1. Baseline patient characteristics are shown in Table 1. The 1st TNFi was stopped for inefficiency, adverse events, other reasons, and death in 73,4%, 16.7%, 8.7%, and 1.0%, respectively. Kaplan Meier survival curves for 2nd TNFi

2691 Results Of The Strass Trial Regarding Impact Of Progressive Spacing Of Tnf-Blocker Injections In Rheumatoid Arthritis Patients In Das28 Remission: Is There a Difference Between Drugs - Adalimumab and Etanercept - Or Their Mode Of Use - Monotherapy Or Combination? Bruno Fautrel1, Thao Pham2, Jacques Morel3, Toni Alfaiate4, Emmanuelle Dernis5, Philippe Gaudin6, Olivier Brocq7, Elisabeth Solau-Gervais8, JeanMarie Berthelot9, Jean-Charles Balblanc10, Xavier Mariette11 and Florence Tubach12. 1Paris 6 – Pierre et Marie Curie University; AP-HP, Rheumatology, Pitie´-Salpeˆtrie`re Hospital, - GRC-UPMC 08 – EEMOIS, Paris, France, 2 Sainte Marguerite Hospital, Marseille, France, 3Lapeyronie Hospital, Montpellier, France, 4AP-HP, Paris, France, 5Le Mans Hospital, Le Mans, France, 6 CHU Hoˆpital Sud, Grenoble Teaching Hospital, Echirolles, France, 7Hospital of Princesse Graˆce de Monaco, Monaco, France, 8University Hospital of Poitiers, Poitiers, France, 9Nantes University Hospital, Nantes, France, 10 Centre Hospitalier Geńe´ral de Belfort, Belfort, France, 11Biceˆtre University Hospital, Le Kremlin Bicetre, France, 12Universite Paris Diderot, Paris, France. Background/Purpose: The STRASS trial was an 18-month randomized controlled trial, conducted in established RA patients in DAS28 remission with etanercept (ETA) or adalimumab (ADA), comparing the impact of a DAS28-driven step-down strategy based on TNF-blocker injection spacing (S arm) to a maintenance strategy (M arm). It demonstrated that, although relapses were more frequent with the spacing strategy arm, no substantial increase in disease activity or structural damage progression occurred throughout the 18-month follow-up. We aimed to assess in a subgroup analysis if the feasibility of TNF-blocker injection spacing differs by molecule –ADA or ETA – and/or its use as monotherapy (MONO) or in combination with synthetic DMARD (COMBO). Methods: Inclusion criteria were: ETA or ADA ⬎ 1 year, DAS28 remission ⬎ 6 months, no progression of structural damage on X-rays. Patients were randomized and followed every 3 months for 18 months. In the S arm, the inter-injection interval was increased every 3 months up to complete interruption at 4th step. In these stratified analyses, disease activity was assessed by DAS28 repeated measures and analyzed in a mixed linear model (GLM). Relapse, defined by DAS28 ⬎2.6 and ⌬DAS28 ⬎0.6, was in a Cox model and statistical comparison by Wald chi-square test. Results: 137 patients were included, 64 and 73 in the S and M arm (mean/%: age 55 yrs, female 78%, RA duration 9.5 yrs, ACPA⫹ 78%, erosive 88%, DAS28 1.8, HAQ 0.5). Sixty-three patients were treated with ADA and 74 with etanercept; 33 received the TNF-blocker as MONO (ADA 11, ETA 22) and 104 in combination with either methotrexate or leflunomide (ADA 52, ETA 52). There were no significant differences in either trial arm in disease activity between the ADA- and the ETA-treated populations, with mean DAS28 of 2.7 ⫾1.2 and 2.3 ⫾1.2 respectively (p⫽0.55). The relapse incidence was higher in the ADA than in ETA group: 77.4% vs. 60.3% (p⫽0.03). The median time to relapse was 12 months in the S arm and 18 months in the M-arm for ETA, and 9 months in the S arm, 18 months in the M-arm for ADA. Comparisons between the use of TNF-blockers as MONO or COMBO showed no significant differences in disease activity: mean DAS28 of 2.7 ⫾1.3 and 2.4 ⫾1.1 respectively (p⫽0.88), nor in relapse rates: 69.7% vs. 67.6% (p⫽0.8). The median time to relapse was 12 months in the S arm and 18 months in the M-arm for MONO, and 9 months in the S arm, 18 months in the M-arm for COMBO. Conclusion: The molecule and the way TNF-blockers are prescribed do not influence RA disease activity control. However, the risk of relapse appears higher with ADA compared to ETA in this subgroup analysis. (ClinicalTrials.gov n°: NCT00780793). Disclosure of Interest: None Declared. B. Fautrel, Abbott, Pfizer, 2; T. Pham, None; J. Morel, Abbott Laboratories, 2; T. Alfaiate, None; E. Dernis, None; P. Gaudin, None; O. Brocq, None; E. Solau-Gervais, None; J. M. Berthelot, None; J. C. Balblanc, None; X. Mariette, None; F. Tubach, None.

S1150

2693

ACR Concurrent Abstract Session Spondylarthropathies and Psoriatic Arthritis Pathogenesis, Etiology Tuesday, October 29, 2013, 2:30 PM–4:00 PM

2692 Evidence That Fat Metaplasia Is a Key Intermediary In The Development Of Sacroiliac Joint Ankylosis Following Repair Of Erosions In Patients With Spondyloarthritis. Walter P. Maksymowych1, Stephanie Wichuk1, Praveena Chiowchanwisawakit2, Robert GW Lambert1 and Susanne Juhl Pedersen3. 1University of Alberta, Edmonton, AB, 2 Mahidol University, Bangkok, Thailand, 3Copenhagen University Hospital at Glostrup, Copenhagen, Denmark.

Age BL SSS Backfill score Change in SSS Erosion score Change in SSS Fat score

␤ coefficient ⫺0.022 0.073 ⫺0.14 0.20

SE 0.010 0.027 0.045 0.058

t ⫺2.25 2.75 ⫺3.11 3.36

p value 0.026 0.0068 0.0023 0.001

Conclusion: Ankylosis in the SIJ develops following repair of erosion and fat metaplasia is a key intermediary step in this pathway. Disclosure: W. P. Maksymowych, None; S. Wichuk, None; P. Chiowchanwisawakit, None; R. G. Lambert, None; S. J. Pedersen, None.

Background/Purpose: Spondyloarthritides (SpA) are characterised by both peripheral and axial arthritis. Hallmarks of peripheral SpA are the development of enthesitis, most typically of the Achilles tendon and plantar fascia, and new bone formation. This study was undertaken to unravel the mechanisms leading towards enthesitis and new bone formation in preclinical models of SpA. TNF⌬ARE mice are an established model for SpA, characterized by an enhanced TNF messenger RNA stability, which leads to the development of peripheral and axial arthritis (sacroiliitis, spondylitis), and Crohn⬘s like ileitis. Methods: To study the development of enthesitis in relation to mechanical strain, hind limb unloading of TNF⌬ARE mice was performed, followed by histological analysis. Activation of extracellular signal-regulated kinase (Erk1/2) and p38 pathways in response to mechanical strain was studied on Achilles tendon lysates. In addition, TNF⌬ARE mice were treated with small molecular inhibitors of Erk1/2 and p38. Since new bone formation does not occur in TNF⌬ARE mice, the collagen antibody-induced arthritis model, which is also characterized by enthesitis, was used to study this particular feature. Hind limb unloading was again performed, followed by histological analysis and micro-CT. Results: We demonstrated that the first signs of inflammation in TNF⌬ARE mice were found at the entheses. Importantly, enthesitis occurred equally in the presence or absence of mature T and B cells, underscoring the importance of stromal cells. Hind limb unloading in TNF⌬ARE mice significantly suppressed inflammation at the synovio-entheseal complex of the Achilles tendon compared to weight bearing controls. Erk1/2 signalling played a crucial role in mechanotransduction associated inflammation. Furthermore, new bone formation was strongly promoted at entheseal sites by biomechanical stress and correlated with the degree of inflammation. Conclusion: These findings provide a formal proof of the concept that mechanical strain drives both entheseal inflammation and new bone formation in SpA. Disclosure: P. Jacques, None; S. Lambrecht, None; E. Verheugen, None; E. Pauwels, None; M. Verhoye, None; A. Van der Linden, None; G. Kollias, None; R. J. Lories, Pfizer Inc, 2, Celgene, 2, Abbvie, 5, Merck Pharmaceuticals, 5, Janssen Pharmaceutica Product, L.P., 5; D. Elewaut, None.

2694 HLA-B27 Expression Prevents Tnf␣-Induced Inhibition Of Bone Formation in Vitro In IFN␥/Tnf␣-Treated Osteoblasts. Eva Yang, Grace Kwon, Robert A. Colbert and Gerlinde Layh-Schmitt. NIAMS NIH, Bethesda, MD. Background/Purpose: HLA-B27 predisposes to ankylosing spondylitis (AS), an immune-mediated inflammatory disease associated with osteitis, bone loss, and dysregulated bone formation, most notably in the axial skeleton. Although the role of HLA-B27 is incompletely understood, it is thought to exert upstream pro-inflammatory effects. Recently however, it was shown that HLA-B27 promotes osteoclast development from monocytes in HLA-B27 transgenic rats. Given that MHC class I molecules can also be expressed in osteoblasts (OBs), we examined whether HLA-B27 alters the response of OBs to pro-inflammatory cytokines IFN␥ and TNF␣ during differentiation and mineralization in vitro. Methods: Wild type (WT), HLA-B7 (B7), and HLA-B27 (B27) transgenic rat calvarial OBs were harvested and differentiated in osteogenic medium for up to 3 weeks. Cultures were treated with rat IFN␥ (100 ng/mL), TNF␣ (30 ng/mL) or both cytokines for up to 5 days of the 3-week culture period upon pre-calcified nodule formation, and collected for evaluation of gene (real-time PCR) and protein (immunoblotting) expression, and mineralization (alizarin red staining). Results: IFN␥ had no effect on mineralization while TNF␣ inhibited mineralization similarly for all 3 genotypes in a time- and dose-dependent

S1151

Tuesday, October 29

Background/Purpose: Fat metaplasia is detected as bright signal on T1W MRI and has been shown to develop after resolution of inflammation in spine and sacroiliac joints (SIJ). Tissue with bright signal on T1W MRI, termed backfill (BF), may also fill areas of excavated sacral and iliac bone along the joint space and is thought to reflect repair of erosion. Imaging studies in the spine have indicated that fat metaplasia predicts development of new bone. We hypothesized that ankylosis in the SIJ develops following repair of erosion and that fat metaplasia is a key intermediary step in this pathway. Methods: We used the SPARCC SIJ Structural Score (SSS) method to assess fat metaplasia (FAT), erosion (ER), BF, and ankylosis (ANK). This score relies on the T1W sequence and assesses 5 consecutive coronal slices anteriorly through the cartilaginous portion of the joint from the transitional slice. Lesions are scored dichotomously (present/ absent) in SIJ quadrants (fat, erosion) or halves (backfill, ankylosis). Scoring ranges are: FAT (0–40), ER (0–40), BF (0–20), ANK (0–20). Four readers assessed 45 pairs of MRI scans blinded to time point (baseline, 2 years) from 45 cases in a prospective cohort receiving either standard (n⫽22) or anti-TNF (n⫽23) therapies. In a second study, two readers assessed 147 pairs of scans blinded to time point (baseline, 2 years) from cases either on standard (n⫽69) or anti-TNF (n⫽78) therapies. Univariate analyses and multivariate linear regression focused on identifying significant MRI predictors of change in BF and ANK scores, adjusted for age, sex, symptom duration, treatment, CRP (baseline and 2-year change), SPARCC SIJ inflammation score (baseline and 2-year change), and baseline SSS scores for FAT, ER, BF, and ANK. Results: Using mean SSS scores for 4 readers in the 45 cases, resolution of ER was significantly associated with the development of BF (p ⫽ 0.0082) and new ANK (p⫽0.045) at 2 years. Using mean scores of two readers in the 147 cases, resolution of ER was significantly associated with the development of BF (p⬍0.0001), FAT (p⬍0.0001) and new ANK (p⫽0.0001) at 2 years. New ANK was also significantly associated with development of FAT (p⫽0.0005). Associations were also significant in both treatment groups. A decrease in ER score was a significant predictor for development of new BF in the multivariate regression model (adjusted R2 ⫽0.44, F ratio 14.6, p⬍0.0001) (change in SSS erosion: ␤⫽ ⫺0.74, t⫽ ⫺4.1, p⫽0.0001). 31 (21.1%) of patients developed new ANK and these had significantly more resolution of ER than patients without new ANK (p⫽0.014, Mann-Whitney). Significant independent predictors of new ANK in the multivariate model (adjusted R2⫽ 0.24, F ratio ⫽ 10.0, p ⬍0.0001) were baseline BF score, decreased ER score and development of new FAT (Table).

Proof Of Concept: Enthesitis and New Bone Formation In Spondyloarthritis Are Driven By Mechanical Strain and Stromal Cells. Peggy Jacques1, Stijn Lambrecht1, Eveline Verheugen1, Elin Pauwels2, Marleen Verhoye3, Annemie Van der Linden3, George Kollias4, Rik J. Lories5 and Dirk Elewaut6. 1Laboratory for Molecular Immunology and Inflammation, Ghent University, Ghent, Belgium, 2UGCT, Ghent University, Ghent, Belgium, 3University of Antwerp, Antwerp, Belgium, 4Biomedical Sciences Research Center ’Alexander Fleming’, Vari, Greece, 5Laboratory for Skeletal Development and Joint Disorders, KU Leuven, Leuven, Belgium, 6Department of Rheumatology Ghent University Hospital, Ghent, Belgium.

manner. TNF␣ also inhibited mineralization of IFN␥-treated WT and B7 OBs, similar to TNF␣ alone. In contrast, B27 OBs were refractory to the inhibitory effects of TNF␣ when cells were co-treated with IFN␥, exhibiting 3-fold higher mineralization than WT and B7 controls. HLA-B was synergistically upregulated at the mRNA and protein level by co-treatment with IFN␥ and TNF␣ in both B7 and B27 OBs. Despite comparable upregulation of HLA-B27 and HLA-B7, only B27-expressing OBs exhibited activation of the unfolded protein response (UPR) as evidenced by induction of GRP78/ BiP and CHOP, along with increased XBP1 mRNA splicing. B27-expressing OBs were responsive to TNF␣ as shown by RUNX2 degradation. However, RUNX2 degradation was incomplete in IFN␥/TNF␣-treated B27-expressing OBs and RUNX2 binding to STAT1 was reduced compared to both WT and B7 controls. Conclusion: This is the first demonstration to our knowledge that HLA-B27 expression can alter OB function as demonstrated in this transgenic animal model. Upregulation of HLA-B27 caused ER stress and activation of the UPR. In addition to its traditional role to maintain ER homeostasis, the UPR is also involved in development and differentiation and is activated during osteogenesis. Additionally, the ER chaperone BiP has been shown to affect mineralization processes, and IFN␥-activated STAT1 can inhibit mineralization through cytosolic sequestration of RUNX2. While the precise mechanism remains unclear, our studies strongly suggest that HLA-B27induced UPR activation can affect OB function. The possibility that HLAB27 expression and upregulation could interfere with TNF␣-induced inhibition of bone formation might explain why patients with AS exhibit axial progression and develop ankylosis despite active inflammation. Disclosure: E. Yang, None; G. Kwon, None; R. A. Colbert, None; G. Layh-Schmitt, None.

Tuesday, October 29

2695 Host Genetic Background Disrupts The Relationship Between Microbiota and Gut Mucosal Tolerance Leading To Spondyloarthritis and Ileitis After a Dectin-1 Trigger. Linda Rehaume1, Stanislas Mondot2, Daniel Aguirre de Ca´rcer2, Jared Velasco1, Helen Benham1, Sumaira Hasnain3, Jaclyn Bowman1, Merja Ruutu1, Philip Hansbro4, Michael McGuckin3, Mark Morrison2 and Ranjeny Thomas1. 1University of Queensland Diamantina Institute, Brisbane, Australia, 2CSIRO Livestock Industries, Brisbane, Australia, 3Mater Medical Research Institute, Brisbane, Australia, 4Center for Asthma and Respiratory Disease and Hunter Medical Research Institute, Newcastle, Australia. Background/Purpose: Chronic inflammatory diseases known as spondyloarthropathies (SpA) including ankylosing spondylitis, psoriatic arthritis, reactive arthritis, and arthritis associated with inflammatory bowel disease, collectively affect 3% of the population, and are strongly heritable. Consistent with shared clinical features and genetic associations, the pathology of each disease is associated with microorganisms, and flares in bowel symptoms correlate with arthritic and spondylitic activity. However, it is not clear how the microbiome influences disease expression in SpA and how this relates to host genetic background. We sought to understand this in BALB/c wild type and ZAP-70W163C (SKG) point-mutant mice with reduced T cell receptor signaling. Methods: SKG and BALB/c mice housed under specific pathogen-free (SPF) or germ-free (GF) conditions were injected intraperitoneally with microbial 1,3-D beta-glucan (curdlan). Arthritis, spondylitis and ileitis were assessed histologically. Microbiome composition in serial fecal samples of mice cohoused from weaning was analyzed by 454 pyrosequencing. Results: By analysis 8 weeks after curdlan, pathological severity of arthritis, spondylitis and ileitis depended on both genetic background and microbiome. Under SPF conditions, SKG mice developed severe spondylitis, arthritis and ileitis whereas curdlan-treated BALB/c developed mild arthritis and spondylitis, but no ileitis. Under GF conditions, SKG mice had reduced spondyloarthritis incidence and no ileitis. Thus development of ileitis was most sensitive to genetic background and associated microbiome i.e. absent in GF SKG mice and SPF BALB/c mice, highest incidence and severity in SPF SKG and intermediate in GF SKG mice colonized with a limited bacterial consortium. Initiation of ileal IL-23 expression, ER stress, depletion of goblet cells and IL-17 response in draining lymph nodes depended on presence of the microbiome. By pyrosequencing, microbiome content in turn depended on the genetic background of the host and the microbial response to beta-glucan over time. Consistent with transmissible suppressive microbes in BALB/c feces, ileitis but not arthritis or spondylitis severity was reduced in SKG cohoused with BALB/c mice.

Conclusion: Our data are consistent with impaired microbial homeostasis in SKG hosts where T cells express ZAP70W163C, and provide a molecular basis for understanding the relationship between immune genetic susceptibility and development of SpA in response to an inflammatory environmental trigger, through host genetic effects on the gut microbiome. Modification of the microbiome presents a novel prophylatic strategy to attenuate genetic risk of SpA. Disclosure: L. Rehaume, None; S. Mondot, None; D. Aguirre de Ca´rcer, None; J. Velasco, None; H. Benham, None; S. Hasnain, None; J. Bowman, None; M. Ruutu, None; P. Hansbro, None; M. McGuckin, None; M. Morrison, None; R. Thomas, Janssen Pharmaceutica Product, L.P., 2, UCB, 5, Abbott Laboratories, 5.

2696 Impaired Bacterial Clearance and An Exuberant Inflammatory Response Promote Chlamydia-Induced Reactive Arthritis In SKG Mice. Athan Baillet1, Linda Rehaume2, Helen Benham2, Connor O’Meara3, Charles Armitage4, Marina Harvie4, Geraldine Brizard5, Jared Velasco2, John V. Forrester6, Mariapia Degli-Esposti5, Kenneth Beagley4 and Ranjeny Thomas2. 1The University of Queensland, Brisbane, Australia, 2University of Queensland Diamantina Institute, Brisbane, Australia, 3Institute of Health & Biomedical Innovation, Biomedical Sciences, Faculty of Health, Queensland University of Technology, Brisbane, Australia, 4Institute of Health & Biomedical Innovation, Brisbane, Australia, 5Lions Eye Institute, Nedlands, Australia, 6Lions Eye Institute, Nedlands, Bahrain. Background/Purpose: The sterile inflammatory arthritis associated with spondylitis, uveitis and rash, known as reactive arthritis, commences weeks after certain gastrointestinal or genitourinary infections, including Salmonella and Chlamydia in genetically-susceptible patients. Several potential mechanisms are proposed, including auto-immunity by molecular mimicry and persistent inflammation in response to intra-cellular live Chlamydia. ZAP70W163C mutant BALB/c mice (known as SKG), which exhibit reduced T cell receptor signaling, an autoreactive CD4⫹ T cell repertoire, relative lymphopenia and increased regulatory T cells are susceptible to spondyloarthritis after bacterial beta-glucan. Chlamydia muridarum (Cmu) is an obligate intracellular pathogen normally controlled by macrophages. We studied whether Cmu genital infection triggered reactive arthritis in SKG mice to explore the relationship between host immunity, bacterial clearance and joint inflammation. Methods: SKG and BALB/c mice were genitally infected with mouseadapted Cmu (5⫻102⫺106 IFUs). Conjunctivis, lid swelling/thickening and arthritis were assessed weekly for 12 weeks post infection (wpi). Eye, skin and joint sections were scored after sacrifice. Bacterial load was quantified in genital swabs. Cmu Major Outer Membrane Protein (MOMP) antigenspecific cytokine production was assessed in splenocytes 1 wpi; cytokines were measured in serum, skin and joint explants at 12 wpi. Regulatory T cells (Treg) were depleted from FoxP3-DTR BALB/c or SKG mice. Cmu genomic ompA DNA was quantified by PCR in spleens, iliac lymph nodes and joints. Non parametric tests assessed statistical significance. Results: SKG but not BALB/c mice developed typical histological features of chronic reactive arthritis, with and remained autoantibodynegative, from 5 weeks after genital infection. Cmu load in genital tract was significantly increased in SKG relative to BALB/c mice in the first week after infection, associated with impaired Cmu MOMP-antigen-specific T cell interferon (IFN)-gamma and increased TNF response. Arthritis severity was correlated with prior Cmu load and live Cmu infection was required for reactive arthritis, as no disease occurred after bacterial inactivation with UV or antibiotics. While depletion of regulatory T (Treg) cells from Cmu-infected Foxp3.DTR SKG mice prompted rapid bacterial clearance, it also hastened onset of severe arthritis, conjunctivitis and dermatitis, accompanied by increased MOMP-specific IFN-gamma production. Cmu DNA was detected in myeloid cells derived from spleen and lymph nodes draining the genital tract of both strains, confirming their capacity to transport Chlamydial inclusions from the site of infection to other organs. The proportion of myeloid cells was 2-fold higher in SKG relative to BALB/c mice and this, together with the increased bacterial load, increases the likelihood of systemic delivery of Cmu DNA to peripheral sites. Conclusion: Cmu induced reactive arthritis results from an exuberant inflammatory response to a deficient early control of infection and incomplete inflammatory suppression by Treg in the setting of host genetic T cell receptor-ZAP70 signalling deficiency. Disclosure: A. Baillet, None; L. Rehaume, None; H. Benham, None; C. O’Meara, None; C. Armitage, None; M. Harvie, None; G. Brizard, None; J. Velasco, None; J. V. Forrester, None; M. Degli-Esposti, None; K. Beagley, None; R. Thomas, Janssen Pharmaceutica Product, L.P., 2, UCB, 5, Abbott Laboratories, 5.

S1152

2697 Identification Of Genetic and Epigenetic Alterations In Spondyloarthritis. Tibor A. Rauch1, Beata Tryniszewska1, Andras Vida1, Timea Ocsko1, Sandor Szanto2, Holly L. Rosenzweig3, Matthew A. Brown4, Gethin P Thomas5, Katalin Mikecz1 and Tibor T. Glant1. 1Rush University Medical Center, Chicago, IL, 2 University of Debrecen, Debrecen, Hungary, 3Oregon Health & Science University, Portland, OR, 4University of Queensland Diamantina Institute, Brisbane, Australia, 5University of Queensland Diamantina Institute, Woolloongabba, Australia.

Disclosure: T. A. Rauch, None; B. Tryniszewska, None; A. Vida, None; T. Ocsko, None; S. Szanto, None; H. L. Rosenzweig, None; M. A. Brown, None; G. P. Thomas, None; K. Mikecz, None; T. T. Glant, None.

ACR Concurrent Abstract Session Systemic Lupus Erythematosus - Human Etiology and Pathogenesis II Tuesday, October 29, 2013, 2:30 PM–4:00 PM

2698 The Treg/Th17 Imbalance Of Patients With Systemic Lupus Erythematosus Were Mediated By Mir-663 Through Down-Regulating TGF-␤1 Secretion Of Bone Marrow-Derived Mesenchymal Stem Cells. Lingyu Geng, Xia Li, Xuebing Feng and Lingyun Sun. Department of Rheumatology and Immunology, the Affiliated Drum Tower Hospital of Nanjing University Medical School, Nanjing, China. Background/Purpose: Systemic lupus erythematosus (SLE) patients exist an imbalance between CD4⫹CD25⫹FoxP3⫹ T regulatory (Treg)

Disclosure: L. Geng, None; X. Li, None; X. Feng, None; L. Sun, None.

2699 Transcription Factor RFX1 Regulates Th17 Differentiation and Its Role In The Pathogenesis Of Systemic Lupus Erythematosus. Ming Zhao, Gongping Liang, Qian Tang, Yang Yang, Yixin Tan and Qianjin Lu. Second Xiangya Hospital, Central South University, Changsha, China. Background/Purpose: Recently, evidence is emerging that abnormal regulation of Th17 differentiation plays an important role in the pathogenesis of systemic lupus erythematosus (SLE). However, the molecular mechanisms are poorly understood. In our previous study, we found that transcription factor RFX1 expression was inhibited in SLE CD4⫹ T cells, which leads to overexpression of autoimmune-related genes CD11a and CD70. Here, we further investigate whether RFX1 regulates Th17 differentiation in SLE. Methods: 40 SLE patients and 30 healthy controls were recruited. CD4⫹ T cells were isolated by magnetic beads. All patients fulfilled at least 4 of the SLE classification criteria of the American College of Rheumatology. RFX1 and cytokines expression levels were detected by real-time PCR, western blot and ELISA. CD4⫹ T cells were transfected with RFX1 expression plasmid (pSG-RFX1) or siRNA-RFX1 by nucleofector device in combination with the human T-cell nucleofector kit. Luciferase report gene assay and Chromatin Immunoprecipitation (ChIP) and Electrophoretic Mobility Shift Assay (EMSA) were used to confirm the target gene of RFX1. H3 acetylation levels and H3 lys9 (H3K9) tri-methylation levels in the promoter region of IL17A were measured by ChIP-qPCR.

S1153

Tuesday, October 29

Background/Purpose: Ankylosing spondylitis (AS) is the prototypic spondyloarthritis and affects approximately 0.2–0.5% of the human population. It is estimated that genetic risk factors contribute ⬎90% of disease susceptibility, and the major genetic risk factor is HLA-B27. Approximately the other half of the genetic risk is composed by non-MHC traits. By generating interval-specific congenic (IVSC) strains utilizing the genetic combination of proteoglycan-induced SpA (PGISpA) in susceptible BALB/c mice and SpA-resistant DBA/2 mice (both carry the same MHC) we can identify non-MHC causative genes by excluding the effect of the MHC. These IVSC strains with overlapping chromosome (Chr) regions were used to narrow the (Pgis2) risk locus on mouse Chr2 (mChr2) to a small region syntenic with AS risk loci on human chromosomes 4 and 9. Methods: PGISpA-resistant DBA/2 and –susceptible BALB/c strains were intercrossed, and then backcrossed with BALB/c mice until the entire Pgis2 locus (mChr2) was from DBA/2 on a full BALB/c background. Congenic mice were tested for PGISpA, and then backcrossed to BALB/c strain. Recombination events were genotyped until partially overlapping DBA/2 intervals (n⫽12) completely covered the Pgis2 locus. Heterozygous females and males with the same genomic intervals were intercrossed, homozygous offspring tested for PGISpA. Based on the overlapping Chr2 intervals, we narrowed a resistant (DBA/2 origin) to ⬃3.5 Mbp in size (mChr2:30.5–34.0 Mbp; syntenic with hChr9:129.⫺132.8 Mbp). All IVSC mice were tested with near infrared OsteoSenseTM 750 (Visen Medical) probe, spine histology, serum cytokines and cytokines in antigen-stimulated in vitro spleen cell cultures. The corresponding 3 Mbp genomic regions of PGISpA-resistant and –susceptible IVSC and parent strains (n⫽10) were sequenced by high-throughput methods and analyzed using the GeneSpring NGS program. Results: We identified a total of 4,416 mutations within the 3.5 Mbp-long genomic region. The 3.5 Mbp region contains 70 protein-coding genes, 12 antisense transcripts (RNA) and 25 different forms (small nucloelar, miRNA) of non-coding RNAs. Mutations/indels were detected (93.38%) in three relatively small genomic regions affecting 3 (Gpr107-Nsc1-Hmcn2) genes and their intergenic regions, and two other genes (St6galnac6 and Lmx1b). PGISpA was significantly reduced (p⬍0.001) in each IVSC strain carrying any of these three DBA/2 alleles. The Gpr107-Nsc1-Hmcn2 triplet contained mutations affecting multiple transcription binding sites, the Lmx1b transcription factor has an in-frame 18-nucleotide deletion in exon 1, and the promoter region of the St6galnac6 was hypermethylated in genomic regions of DBA/2 origin. Conclusion: Although all three alleles affected PGISpA individually, it appears that an “epistatic triangle” exists among these alleles that protect against PGISpA. Overall, all mutations with predicted functional consequence are located in non-coding sequences underlining the critical role of epigenetic alterations in PGISpA (and perhaps in AS).

and IL-17-producing cells (Th17). Correction of this Treg/Th17 imbalance may have therapeutic impact for SLE patients. Our previous study demonstrated that bone marrow derived mesenchymal stem cells (MSCs) from SLE patients are defective in immune modulation, which might be involved in Treg/Th17 imbalance, but the mechanisms are not clear yet. As a computer predicted target of miR-663, Transforming growth factor-beta1 (TGF-␤1) is an important negative immune regulatory factors secreted by MSCs which could modulate SLE Treg/Th17 imbalance. The aim of this study is to investigate whether miR-663 could contribute to SLE Treg/Th17 imbalance though directly down-regulating TGF-␤1 mRNA of SLE BMSCs, to further understand the pathogenesis of SLE. Methods: BMSCs were isolated, cultured and expanded from iliac crest bone marrow of four healthy controls and five SLE patients. MicroRNA expressions of BMSCs from were determined by MicroRNA array analysis. Real-time PCR was used to further determine the miR-663 expression and TGF-␤1 mRNA level in MSCs, secretion of TGF-␤1 was detected by ELISA. As a computer predicted target of miR-663, TGF-␤1 was determined using the luciferase reporter assay system. MSCs were transfected with miR-663a, pre-miR-663a, and anti-miR-663a, and then co-cultured with PBMCs from SLE patients for 3 days respectively with PHA pre-stimulated, flow cytometry was used to detect their effect on percentage of Treg and Th17 cells. Results: Mir-663 was significantly up-regulated(2.52- fold higher, P⬍0.05) in MSCs from SLE patients(n⫽5) compared to normal controls(n⫽4) by microRNA array analysis. The expression of miR-663 was further determined by RT-PCR, and the synthesis of TGF-␤1 mRNAs was significantly lower(3.88- fold lower, P⬍0.05) in MSCs from SLE patients. Transfection of SLE MSCs with pre-miR-663a caused significant upregulation of miR-663a expression(8.33- fold higher, n⫽3, P⬍0.01) and markedly lower synthesis of TGF-␤1 mRNA(1.52-fold lower, n⫽3, P⬍0.05), while transfection with anti-miR-663a markedly decreased the miR-663a expression (3.01- fold lower, n⫽3, P⬍0.05). The mean value of Treg/Th17 was significantly decreased in PBMCs from SLE patients compared to normal controls (0.48⫾0.12vs0.65⫾0.09, n⫽8, P⬍0.01). Compared to SLE MSCs, MSCs from normal controls exhibited a better immune suppression effect through up-regulating Treg/Th17 of SLE patients (0.68⫾0.15vs0.54⫾0.14, n⫽3, P⬍0.05), and transfection of normal MSCs with pre-miR-663a caused significant downregulation of Treg/Th17 compared to miR-663-control group(0.38⫾0.07vs0.68⫾0.15, n⫽3, P⬍0.01), while transfection with anti-miR-663a led to an opposite effect (0.76⫾0.08vs0.68⫾0.15, n⫽3, P⬍0.05). Conclusion: SLE patients exist an imbalance between Treg and Th17 cells, which might be associated with down-regulated TGF-␤1 secretion of bone marrow-derived mesenchymal stem cells mediated by miR-663.

Results: Compared with normal controls, RFX1 protein levels were decreased and IL17A mRNA levels were increased significantly in SLE CD4⫹ T cells. The IL-17A protein levels in serum of SLE patients were increased significantly. A negative correlation was observed between IL17A mRNA levels and RFX1 protein in SLE CD4⫹ T cells. Luciferase report gene assay, ChIP and EMSA showed that RFX1 can repress promoter activity of IL17A through binding the promoter of IL17A. Transfection of siRNARFX1 leads to up-regulated expression of IL17A through increasing H3 acetylation level and reducing H3K9 tri-methylation level in normal CD4⫹ T cells. In contrast, transfection of pSG-RFX1 inhibits expression of IL17A through reducing H3 acetylation level and increasing H3K9 tri-methylation level of IL17A promoter in SLE CD4⫹ T cells. Conclusion: RFX1 is involved in repressing IL17A expression through regulating the epigenetic modifications in the promoter region of IL17A in CD4⫹T cells. Decreased RFX1 expression contributes to abnormal regulation of Th17 cells in SLE patients. Disclosure: M. Zhao, None; G. Liang, None; Q. Tang, None; Y. Yang, None; Y. Tan, None; Q. Lu, None.

2700

Tuesday, October 29

Abnormal Mitochondrial Electron Transport Chain Activity At Complex I Is Regulated By Nitric Oxide and N-Acetylcysteine In Lupus Lymphocytes. Edward Doherty1 and Andras Perl2. 1SUNY Upstate, Syracuse, NY, 2SUNY Upstate Medical University, Syracuse, NY. Background/Purpose: Systemic lupus erythematosus (SLE) peripheral blood lymphocytes (PBL) show mitochondrial dysfunction, characterized by elevated mitochondrial transmembrane potential (⌬⌿m) and mass and low ATP, attributed to elevated production of nitric oxide (NO) and reduced glutathione (GSH). To understand the molecular bases of mitochondrial dysfunction we tested activity of the electron transport chain (ETC) complexes I and IV. As N-acetylcysteine (NAC) reversed GSH depletion and improved disease activity in SLE, its target of impact in mitochondria was also investigated. Methods: PBL from 65 SLE subjects and 30 healthy controls, matched for patients’ age within ten years, gender, and ethnicity were studied. Mitochondrial respiration was measured using a Clark-type O2 electrode (Oxygraph, Hansatech, Norfok, UK) in 1) freshly isolated PBL, 2) PBL rested overnight, 3) PBL exposed to NO donors NOC-9 or NOC-18, 4) PBL exposed to NAC, 5) PBL exposed to ␤-mercaptoethanol (BME). Snitrosylation was assessed with a biotin switch kit (Cayman, Ann Arbor, MI). ⌬⌿m, mass, NO, ONOO⫺, Ca2⫹, and H2O2 stress were assessed by flow cytometry (Meth. Mol. Biol. 900:61–89, 2012). Results: SLE PBL have increased respiration upon T cell activation (SLE: 4.157nmol/ml/min ⫾0.186, control: 3.655nmol/ml/min ⫾0.167; p⫽0.012). Digitonin-permeabilized SLE PBL also show increased respiration without exogenous substrates (SLE: 2.492 nmol/ml/min ⫾ 0.196, control: 2.137 nmol/ml/min ⫾ 0.153; p ⫽ 0.027) and with substrates of ETC complex IV (SLE: 7.722 nmol/ml/min ⫾ 0.419, control: 7.006 nmol/ml/min ⫾ 0.505; p ⫽ 0.028). SLE PBL had elevated mitochondrial mass (⫹10% ⫾ 3% p ⫽ 0.002). When normalized to mitochondrial mass, SLE PBL exhibit 33% reduced respiration through complex I (p⫽0.036) and 81% enhanced respiration through complex IV (p ⫽ 0.036). As expected, exposure to NOC-18 increased NO (⫹158% ⫾ 51% p ⫽ 0.005) and mitochondrial mass in normal PBL (⫹162% ⫾ 68% p ⫽ 0.016). NO exposure selectively increased O2 consumption by SLE PBL through complex I relative to healthy controls (SLE: 1.405 nmol/ml/min ⫾ 0.206; Controls: 1.277 nmol/ml/min ⫾ 0.150; p ⫽ 0.026). In contrast, NAC decreased both ONOO⫺ (⫺37% ⫾0.04 p⫽0.0001) and H2O2 levels without (⫺45% ⫾ 5% p⫽0.0005) or with co-incubation with NOC-18 (ONOO⫺: ⫺38% ⫾7% p⫽0.002; H2O2: ⫺71% ⫾2% p⫽0.000004). Compared to NOC-18 alone, co-incubation with NAC also decreased NO (⫺92% p⫽0.014), ONOO⫺ (⫺38% p⫽0.006), mitochondrial mass (⫺87% p⫽0.017), and Ca2⫹ (⫺52% p⫽0.042). Direct addition of NAC into the Oxygraph chamber inhibited respiration by 53% through complex I (p⫽0.004) but not complex IV. In contrast, BME failed to affect respiration at complex I, but it reduced respiration by 80% at complex IV (p⫽0.031). NAC also blocked respiration through complex I by 56% in SLE PBL (p⫽0.0001). Upon NO exposure, S-nitrosylation is increased in freshly isolated lupus PBL relative to matched healthy control PBL (p⫽0.002). Conclusion: The results of this data suggest that SLE PBL have defective ETC activity at complex I. NAC helps maintain a reducing environment, possibly by directly blocking respiration at complex I. Disclosure: E. Doherty, None; A. Perl, None.

2701 Spleen Tyrosine Kinase (Syk) Regulates Systemic Lupus Erythematosus T Cell Signaling. Alexandros P. Grammatikos1, Debjani Ghosh2 and Vasileios C. Kyttaris3. 1Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, MA, 2The Brody School of Medicine, East Carolina University, Greenville, NC, 3BIDMC, Harvard Medical School, Boston, MA. Background/Purpose: Systemic Lupus Erythematosus (SLE) T cells display an aberrant response to CD3/ TCR stimulation. The CD3/TCR complex in SLE is characterized by the substitution of the CD3 associated kinase Zap70 by Syk. The objective of this study is to better understand the role of Syk in the dysregulated SLE T cell phenotype. Methods: A two-step approach was followed: (i) to examine whether increased Syk expression creates an SLE-like phenotype Syk was overexpressed in healthy blood-donor T cells; and (ii) to examine whether aberrant gene expression changes in SLE T cells can be corrected by Syk downregulation, SYK was silenced using a SYK-specific siRNA. Cells were lysed following 72h of incubation and expression levels of 36 genes previously associated with SLE measured in real-time PCR. Protein was also collected and expression changes confirmed in flow cytometry and western blot. Expression was normalized towards CD3E and GPADH for PCR and ␤-actin for western blot. Results: Forced expression of Syk in normal T cells reproduced many of the aberrant gene expression changes seen in SLE. Expression of cytokine IL-21, cell surface molecule CD44, and intracellular molecules PP2A and OAS2 were found to increase in cells overexpressing SYK (fold increase over control transfected ⫾SEM: IL-21, 5.7⫾1.5; CD44, 4.2⫾2.2; PP2A, 1.5⫾1; OAS2, 1.5⫾0.3). Silencing of Syk in SLE T cells normalized the expression of the above molecules (fold decrease over control transfected ⫾SEM: IL-21, ⫺1.9⫾0.2; CD44, ⫺2.8⫾0.4; PP2A, ⫺2.7⫾1.9; OAS2, ⫺1.1⫾0.8). Findings were confirmed at the protein level using flow cytometry (mean positive cells ⫾SEM: (i) empty vector vs. SYK overexpressing, IL-21, 7.9⫾1 vs. 13⫾2.3, p⫽0.05; CD44v6, 10.7⫾1.2 vs. 16⫾2.1, p⫽0.05; (ii) control vs. SYK siRNA transfected, IL-21, 18.6⫾2.4 vs. 11.4⫾1.6, p⫽0.03; CD44v6, 11.7⫾0.7 vs.7.6⫾1.2, p⫽0.05) and western blot (mean relative expression ⫾SEM: (i) empty vector vs. SYK overexpressing, OAS2, 0.91⫾0.1 vs. 1.39⫾0.2, p⫽0.05; PP2A, 0.49⫾0.1 vs. 1.23⫾0.3, p⫽0.05; (ii) control vs. SYK siRNA transfected, OAS2, 1.39⫾0.3 vs. 1.06⫾0.2; PP2A, 1.14⫾0.3 vs. 0.69⫾0.1). Conclusion: Our data show that overexpression of Syk in healthy T cells recapitulates at least part of the SLE T cell phenotype. Syk overexpressing T cells may provide more help to B cells through IL21 secretion, have enhanced migration to tissues through CD44 upregulation and produce pro-inflammatory rather than counter-inflammatory cytokines. Inhibiting Syk in SLE T cells leads to the opposite effect, further underscoring Syk’s potential as a therapeutic target in SLE. Disclosure: A. P. Grammatikos, None; D. Ghosh, None; V. C. Kyttaris, None.

2702 Molecular Pathogenesis and Genetics Of Tartrate-Resistant Acid Phosphatase Deficiency In Systemic Lupus Erythematosus. Jie An1, Tracy A. Briggs2, Audrey Dumax-vorzet2, Alice Wiedeman1, Laurence Chaperot3, Joel Plumas3, Yanick J. Crow2 and Keith B. Elkon1. 1University of Washington, Seattle, WA, 2University of Manchester, Manchester, United Kingdom, 3Immunobiology & Immunotherapy of Cancers, La Tronche, France. Background/Purpose: Biallelic mutations in ACP5, encodng tartrateresistant acid phosphatase (TRAP), result in the immuno-osseous disorder spondyloenchondrodysplasia (SPENCD), characterized by a variety of autoimmune phenotypes – most particularly systemic lupus erythematosus (SLE). Importantly, patients with SPENCD demonstrate an upregulation of type 1 interferon (IFN) stimulated genes (ISGs – an “interferon signatures”) similar to that observed in SLE. Since very little is known about the function of TRAP in immune cells, the objectives of our study were: a) to determine the consequences of TRAP deficiency in human immune cells, B) to identify substrates of TRAP, and c) to determine whether ACP5 mutations occur in ‘idiopathic’ SLE. Methods: Unbiased substrates of TRAP were queried using a yeast 2 hybrid (Y2H) screen in a human macrophage cDNA library, and interaction of a candidate substrate, osteopontin (OPN) with TRAP was determined by confocal microscopy and immunoprecipitation-western blot analysis (IPwestern). TRAP overexpression / knockdown was performed by transfection with cDNA / lentivirus shRNA, respectively. Expression of ISGs was

S1154

determined by quantitative PCR (qPCR). Phosphatase activity was quantified by colorimetry, and dephosphorylation of substrates by Liquid Chromatogram-tandem Mass Spectrometer (LC-MS/MS). Sequencing of ACP5 coding exons was undertaken in patients with SLE. Results: OPN is a substrate for TRAP in Macrophase and pDC. TRAP interacted and co-localized with OPN as determined by Y2H, confocal microscopy and by IP-western. Consistent with these data, recombinant human TRAP (rhTRAP) dephosphorylated recombinant human OPN (rhOPN) by the release of free phosphate in an in vitro assay. LC-MS/MS demonstrated rhTRAP dephosphorylated rhOPN at two serine residues in vitro. To relate the functional significance of TRAP deficiency to IFN-␣ production, we knocked down TRAP expression in pDC and observed that TRAP specific shRNA, but not scrambled shRNA, increased the expression of ISGs as well as IL-6 following TLR9 stimulation. This was associated with increased nuclear translocation of IRF7 and P50 in TRAP KD pDC cells compared to control cells. Sequencing of ACP5 in 865 SLE patients and 511 controls revealed an excess of heterozygous ACP5 possibly pathogenic missense variants in SLE patients (11 adults in the lupus cohort compared to 2 in controls). These variants were predicted as pathogenetic since they were: coding, non-synonymous, occurred at a frequency of less than 1/300 in controls, were in residues conserved in mammalian species and were predicted to destabilize protein on in silico testing. Transient transfection of several mutants and patient serum assays revealed a significant reduction in TRAP enzyme activity. Conclusion: Our findings indicate that TRAP and OPN co-localize, and that OPN is a substrate for TRAP in immune cells. Significantly, TRAP deficiency in pDCs leads to increased IFN-␣ production, providing at least a partial explanation for why ACP5 mutations cause lupus in the context of SPENCD. Detection of ACP5 missense variants in lupus patients suggests that impaired function of TRAP may play a role increasing susceptibility to adult-onset idiopathic lupus.

2703 An Intronic CR2 Polymorphism Associated With Systemic Lupus Erythematosus Alters CTCF Binding and CR1 Expression. Jian Zhao1, Brendan M. Giles2, Rhonda L. Taylor3, Gabriel A. Yette2, Kara M. Lough2, Lawrence J. Abraham3, Hui Wu1, Patrick M. Gaffney4, Jennifer A. Kelly4, Kenneth M. Kaufman5, John B. Harley6, Carl D. Langefeld7, Elizabeth E. Brown8, Jeffrey C. Edberg9, Robert P. Kimberly8, Daniela Ulgiati3, Betty P. Tsao1 and Susan A. Boackle2. 1David Geffen School of Medicine University of California Los Angeles, Los Angeles, CA, 2University of Colorado School of Medicine, Aurora, CO, 3University of Western Australia, Perth, Western Australia, Australia, 4Oklahoma Medical Research Foundation, Oklahoma City, OK, 5Cincinnati Children’s Hospital Medical Center and the University of Cincinnati, Cincinnati, OH, 6US Department of Veterans Affairs Medical Center, Cincinnati, OH, 7Wake Forest School of Medicine, Winston-Salem, NC, 8University of Alabama at Birmingham, Birmingham, AL, 9Department of Medicine, University of Alabama at Birmingham, Birmingham, AL. Background/Purpose: Systemic lupus erythematosus (SLE) is characterized by the production of antibodies to nuclear antigens, which form immune complexes that deposit in tissues and cause damage. Complement receptor 2 (CR2/CD21) is primarily expressed on B cells and follicular dendritic cells (FDC), and is required for normal humoral immune responses. Complement receptor 1 (CR1/CD35) is associated with CR2 on the B cell surface, and its cofactor activity is required for the conversion of iC3b to C3dg, the specific ligand for CR2. We previously identified CR2 variants associated with decreased risk of SLE. To identify the causal variant(s) for the association, we conducted trans-ancestral fine-mapping of CR2 and the surrounding genomic region. Methods: Genotyped and imputed genetic variants located in a 57.6kb region spanning 5’ of CR2 to intron 1 of CR1 were assessed for association with SLE in 15,750 unrelated case-control subjects from four ancestral groups using a logistic regression model adjusted for gender and global ancestry. Allele-specific functional effects of associated variants were determined using electrophoretic mobility shift assay (EMSA), chromatin immunoprecipitation (ChIP)-PCR, quantitative real-time PCR, and quantitative flow cytometry. Results: The strongest association signal was detected at rs1876453, located in intron 1 of CR2 (Pmeta⫽4.2⫻10⫺4, OR 0.85), specifically when subjects with lupus were stratified based on the presence of dsDNA autoantibodies (case-control Pmeta ⫽7.6⫻10⫺7 , OR 0.71; case-only

Disclosure: J. Zhao, None; B. M. Giles, None; R. L. Taylor, None; G. A. Yette, None; K. M. Lough, None; L. J. Abraham, None; H. Wu, None; P. M. Gaffney, None; J. A. Kelly, None; K. M. Kaufman, None; J. B. Harley, None; C. D. Langefeld, None; E. E. Brown, None; J. C. Edberg, None; R. P. Kimberly, None; D. Ulgiati, None; B. P. Tsao, None; S. A. Boackle, None.

ACR Concurrent Abstract Session Systemic Sclerosis, Fibrosing Syndromes, and Raynaud’s II: Pathogenesis, Animal Models, Genetics: Novel Signaling Pathways Mediating Fibrosis Tuesday, October 29, 2013, 2:30 PM–4:00 PM

2704 Twist1 Regulates Transforming Growth Factor Beta Dependent Activation Of Fibroblasts In Fibrosis. Katrin Palumbo-Zerr1, Andrea Liebl1, Pawel Zerr1, Michal Tomcik2, Alfiya Distler1, Christian Beyer1, Oliver Distler3, Georg Schett4 and Joerg H. W. Distler1. 1Department of Internal Medicine III and Institute for Clinical Immunology, University of ErlangenNuremberg, Erlangen, Germany, 2Institute of Rheumatology, Department of Rheumatology, 1st Faculty of Medicine, Charles University, Prague, Czech Republic, 3Department of Rheumatology, University Hospital Zurich, Zurich, Switzerland, 4University of Erlangen-Nuremberg, Erlangen, Germany. Background/Purpose: Twist1 is a member of the tissue-restricted class B of basic-helix-loop-helix (bHLH) transcription factors acting as master regulators on different tissues. Twist1 is controlled by spatial-temporal expression, dimer choice and cellular localization. Twist1 was shown to be a regulator of mesenchymal transition (EMT). Moreover, Twist1 regulates pathways such as Wnt/␤catenin signalling and responses to hypoxia. The aim of the present study was to investigate the role of Twist1 in TGF␤ signaling and in fibroblast activation in systemic sclerosis (SSc). Methods: Twist1 expression was quantified in SSc patients and different mouse models of fibrosis by qPCR, Western Blot and IF. Collagen synthesis was quantified by qPCR and SirCol in fibroblasts overexpressing or lacking Twist1. Interaction of Twist1, E12 and Id was shown by Co-IP. ChIP assays were performed to analyze Twist1transcriptional binding. The role of Twist1 in vivo was evaluated using inducible, conditional knockout mice with either ubiquitinous (Ubc CreER) or fibroblast-specific (col1a2 CreER) depletion of Twist1. Mice were either challenged with bleomycin or infected with constitutively active TGF␤ receptor I adenovirus (TBRICA). Results: Twist1 mRNA was elevated by 330 % in the skin of SSc patients (p ⫽ 0.002). Consistently, Twist1 was increased by 490 % (p ⫽ 0.008) and by 349 % (p ⫽ 0.001) in bleomycin-induced and TBRICA mouse models, respectively. Induction of Twist1 was mediated by TGF␤ in a Smad3 dependent manner in vitro and in vivo, as demonstrated using the selective TGF␤ receptor I inhibitor SD208 and Smad3/4 siRNA. Twist1 overexpression fostered the pro-fibrotic effects of TGF␤ and enhanced fibroblast differentiation and release of collagen. In contrast, fibroblasts lacking Twist1 ameliorated TGF␤ signalling. Chronic stimulation with TGF␤ upregulated Twist1 and E12 in fibroblasts, but resulted in an even more pronounced induction of Id1 and Id3. Id proteins have great affinity for E12 and compete against class bHLH factor Twist1 for E12 binding. Co-IP revealed a time dependent shift of Twist1/E12 heterodimers to Id/E12 and Twist1/Twist1 complexes. In addition, binding of Twist1 homodimers to the col1a1 and col1a2 promoter was demonstrated using ChIP assay. Mice lacking Twist1 in

S1155

Tuesday, October 29

Disclosure: J. An, None; T. A. Briggs, None; A. Dumax-vorzet, None; A. Wiedeman, None; L. Chaperot, None; J. Plumas, None; Y. J. Crow, None; K. B. Elkon, None.

Pmeta⫽1.9⫻10⫺4, OR 0.75). Using EMSA, we demonstrated that the minor allele at rs1876453 altered the formation of several DNA-protein complexes, including one containing the highly conserved 11 zinc-finger protein CCCTCbinding factor (CTCF), which has pleiotropic effects on genomic organization, gene regulation, and alternative splicing. The allele-specific effects of rs1876453 on CTCF binding were confirmed by ChIP-PCR, with three-fold higher enrichment of the region surrounding this variant in the presence of the major allele (p ⫽ 0.0178). Although allele-specific effects of rs1876453 on B cell CR2 mRNA or protein levels were not identified, levels of complement receptor 1 (CR1/CD35) mRNA and protein were significantly higher on the B cells of subjects harboring the minor allele (p ⫽ 0.0248 and p ⫽ 0.0006, respectively). Conclusion: These data suggest that rs1876453 in CR2 has long-range effects on gene regulation that decrease susceptibility to lupus. Since the minor allele at rs1876453 is preferentially associated with reduced risk of the highly specific dsDNA autoantibodies that are present in preclinical, active, and severe lupus, understanding its mechanisms will have important therapeutic implications.

fibroblasts were less sensitive to bleomycin induced dermal fibrosis with diminished dermal thickening (⫺69,3 %, p ⬍ 0.0001), myofibroblast counts (⫺62,3 %, p ⬍ 0.0001) and hydroxyproline content (⫺58,1%, p ⬍ 0.0001). Mice with ubiquitous Twist1 knockout were protected to a similar degree, highlighting that fibroblasts are the key-effector cells for Twist1 signalling in experimental skin fibrosis. Twist1 deficiency also protected from TBRICA induced dermal fibrosis. Conclusion: We identified Twist1 as an important downstream mediator of TGF␤ in SSc. TGF␤ induces Twist1 expression and stimulates its promoter binding, which triggers myofibroblast differentiation and release of collagen. Knockdown of Twist1 ameliorates the pro-fibrotic effects of TGF␤ and ameliorates experimental fibrosis in different murine models of SSc. Thus our study characterizes Twist1 as a key-regulator of fibroblast activation in SSc. Disclosure: K. Palumbo-Zerr, None; A. Liebl, None; P. Zerr, None; M. Tomcik, None; A. Distler, None; C. Beyer, None; O. Distler, Sanofi, Active Biotech, Pfizer, Actelion, and Novartis, 2, Actelion, Sanofi-Aventis, Roche, Genentech, Active Biotec, medac, Novartis, Sinoxa, 4D Science, 5; G. Schett, None; J. H. W. Distler, None.

2705

Tuesday, October 29

Inhibition Of Microrna-155 Ameliorated Experimental Fibrosis By Suppressing Wnt and Akt Pathways. Qingran Yan1, Qiong Fu1, Jie Chen1, Xinfang Huang1, Nan Shen2 and Chunde Bao1. 1Renji Hospital, Shanghai Jiaotong University School of Medicine, Shanghai, China, 2Shanghai Institute of Rheumatology, Renji Hospital, Shanghai Jiaotong University School of Medicine, Shanghai, China. Background/Purpose: MicroRNA-155 plays a critical role in various physiological and pathological processes including cell proliferation and migration, epithelial- mesenchymal transition, as well as inflammation. Previous research found that miR-155 could promote the development of murine pulmonary fibrosis. The aim of this study was to evaluate the role of miR-155 in a bleomycin induced mouse model of systemic sclerosis (SSc) and to investigate the potential mechanisms involved. Methods: miR-155 knock-out (KO) C57BL/6 and background controlled mice were given bleomycin subcutaneously every other day for 21days to induce experimental fibrosis. Dermal thickness and collagen deposition were determined histologically by Sirius red dying. Tissue fibroblast number was measured through ␣-smooth muscle actin (␣-SMA) staining by immunohistochemistry. For in vitro study, primary murine skin fibroblasts were cultured and stimulated with TGF-␤. siRNA was applied to inhibit miR-155 expression. Real-time PCR was used to compare collagen synthesis and activation marker expressions in miR-155 knockdown fibroblasts with controls. Collagen release was analyzed by Sircol assay. Major TGF-␤ induced profibrotic pathways were analyzed by Western Blot. Targets of miR-155 were predicted by bio-informatics and were confirmed by using dual luciferase report assay and Western blot. Results: The expression level of miR-155 was significantly elevated in mice injected with bleomycin, as well as in fibroblasts stimulated by TGF-␤. Compared with control mice, miR-155 KO mice displayed significantly less fibrosis in both skins and lungs, as well as less number of activated fibroblasts in situ. Consistent with the in vivo findings, miR-155-knockdown primary fibroblasts showed significantly reduced collagen synthesis and lower activation markers (fibronectin and ␣-SMA) upon TGF-␤ stimulation. Analysis of major TGF-␤ downstream pathways including Smad2/3, JNK, ERK, Wnt and Akt revealed that the activation of Wnt and Akt pathway were notably suppressed in miR-155-knockdown fibroblasts. These two pathways were also attenuated in the skin of miR-155 KO mice injected with bleomycin. Further bio-informatics and literature research suggested that miR-155 could directly target casein kinase I (CK-1) and Src homology 2-containing inositol phosphatase-1 (SHIP-1), which are negative regulators of Wnt and Akt pathway, respectively. We found miR-155 inhibition would result in an increased level of these two proteins in primary fibroblast. Further studies using dual luciferase reporter assay confirmed the direct interaction between miR-155 and 3’-UTR of CK-1 and SHIP-1. Conclusion: Inhibition of miR-155 could ameliorate experimental fibrosis in a bleomycin induced SSc mouse model. miR-155 modulates profibrotic Wnt and Akt pathway by directly targeting CK-1 and SHIP-1, respectively. miR-155 is a potential target for therapeutic intervention in SSc. Disclosure: Q. Yan, None; Q. Fu, None; J. Chen, None; X. Huang, None; N. Shen, None; C. Bao, None.

2706 Fli1 Haploinsufficiency Exacerbates Dermal Fibrosis Via Activation Of Fibroblasts, Endothelial Cells and Macrophages In Bleomycin-Treated Mice. Takashi Taniguchi1, Yoshihide Asano1, Kaname Akamata1, Shinji Noda1, Takehiro Takahashi1, Yohei Ichimura1, Tetsuo Toyama1, Maria Trojanowska2 and Shinichi Sato1. 1University of Tokyo Graduate School of Medicine, Tokyo, Japan, 2Boston University, Boston, MA. Background/Purpose: Systemic sclerosis (SSc) is a multisystem autoimmune disorder with clinical manifestations resulting from immune activation, vascular injuries and fibrosis development. Previous reports suggest that deficiency of the transcription factor Fli1 (Friend leukemia integration-1) has a pivotal role in the pathogenesis of SSc. Although Fli1 deficiency activates fibroblasts and endothelial cells toward an SSc phenotype in vitro, Fli1⫹/⫺ mice show no clinical symptoms similar to SSc, suggesting that some additional factors are required to develop SSc in those mice. To address this issue, we generated bleomycin (BLM)-induced SSc murine model using Fli1⫹/⫺mice and evaluated their phenotype by focusing on fibroblasts, endothelial cells, and immune cells. Methods: Wild type and Fli1⫹/⫺ mice were used in the BLM model of scleroderma. Degree of dermal thickness and fibrosis were determined by histological analyses. The quantity of the collagen-specific amino acid hydroxyproline was measured. Immunohistochemistry and real-time PCR were conducted to evaluate the degree of inflammation and the expression of cytokines, growth factors, chemokines, and cell adhesion molecules. The influence of Fli1 deficiency on the phenotypical changes in dermal fibroblasts, endothelial cells, and macrophages were also evaluated in vitro by real-time PCR and TGF-b bioassay. Results: BLM induced dermal fibrosis to a much greater extent in Fli1⫹/⫺ mice than in wild type mice. In addition, upon BLM treatment, Fli1⫹/⫺ mice exhibited higher mRNA levels of CCN2 in the lesional skin than wild type mice, while mRNA levels of TGF-b were comparable. On the other hand, Fli1 haploinsufficiency greatly activated dermal fibroblasts in response to BLM treatment partly due to the elevated expression of integrin aVb3 and aVb5 in those cells, leading to the activation of latent TGF-b on their cell surface. Upon BLM treatment, the lesional skin of Fli1⫹/⫺ mice showed higher expression of IL-4, IL-6, IL-10, IFN-g, TNF-a, iNOS, arginase1, Fizz1, and Ym1 than that of wild type mice. The numbers of myofibroblasts, macrophages, and mast cells and the ratio of CD4⫹/CD8⫹ T cells were increased in the lesional skin of BLM-treated Fli1⫹/⫺ mice relative to that of BLM-treated wild type mice. Moreover, Fli1 haploinsufficiency promoted M2 macrophage infiltration in the lesional skin of BLM-treated mice and also promoted M2 differentiation of peritoneal macrophages by IL-4 or IL-13 stimulation in vitro. As for endothelial cells, Fli1 haploinsufficiency modulated the expression of cell adhesion molecules toward the induction of Th2 skewed inflammation by BLM treatment, as shown by the lower expression of E-selectin and P-selectin and the higher expression of ICAM-1 and GlyCAM-1 in the lesional skin of Fli1⫹/⫺mice than in that of wild type mice. Conclusion: With BLM treatment, Fli1⫹/⫺ mice exhibited exacerbated dermal fibrosis due to the phenotypical changes of fibroblasts, endothelial cells, and immune cells toward an SSc phenotype to a much greater extent than wild type mice. Some additional factors induced by BLM treatment may be required for Fli1⫹/⫺ mice to develop an SSc phenotype. Disclosure: T. Taniguchi, None; Y. Asano, None; K. Akamata, None; S. Noda, None; T. Takahashi, None; Y. Ichimura, None; T. Toyama, None; M. Trojanowska, None; S. Sato, None.

2707 Downregulation Of KLF5 and Fli1 Cooperatively Contribute To Distinct Manifestations Of Systemic Sclerosis In Vitro and In Vivo. Shinji Noda1, Yoshihide Asano1, Katsuhito Fujiu1, Ichiro Manabe1, Ryozo Nagai1, Kaname Akamata1, Takashi Taniguchi1, Takehiro Takahashi1, Yohei Ichimura1, Tetsuo Toyama1, Daisuke Tsuruta2, Maria Trojanowska3 and Shinichi Sato1. 1 University of Tokyo Graduate School of Medicine, Tokyo, Japan, 2Osaka City University Graduate School of Medicine, Osaka, Japan, 3Boston University, Boston, MA. Background/Purpose: Systemic sclerosis (SSc) is a multisystem autoimmune disease characterized by initial vascular injuries and resultant fibrosis of skin and certain internal organs. Although the pathogenesis of SSc still remains unknown, our latest data have demonstrated that epigenetic suppression of transcription factor Fli1 is deeply involved in the constitutive activation of fibroblasts and endothelial cells in this disease. In murine

S1156

Disclosure: S. Noda, None; Y. Asano, None; K. Fujiu, None; I. Manabe, None; R. Nagai, None; K. Akamata, None; T. Taniguchi, None; T. Takahashi, None; Y. Ichimura, None; T. Toyama, None; D. Tsuruta, None; M. Trojanowska, None; S. Sato, None.

2708 The Hedgehog Transcription Factor Gli-2 Is a Novel Downstream Mediator Of The Pro-Fibrotic Effects Of Transforming Growth FactorBeta. Barbora Sumova1, Cinzia Cordazzo2, Katrin Palumbo-Zerr3, Clara Dees4, Pawel Zerr3, Oliver Distler5, Georg Schett4, Ladislav Senolt6 and Joerg H. W. Distler3. 1Department of Rheumatology of the First Faculty of Medicine, Institute of Rheumatology and Connective Tissue Research Laboratory, Prague, Czech Republic, 23Dipartimento Cardiotoracico e Vascolare, Laboratory of Respiratory Cell Biology, University of Pisa, Pisa, Italy, 3 Department of Internal Medicine III and Institute for Clinical Immunology, University of Erlangen-Nuremberg, Erlangen, Germany, 4University of Erlangen-Nuremberg, Erlangen, Germany, 5Department of Rheumatology, University Hospital Zurich, Zurich, Switzerland, 6Institute of Rheumatology, Department of Rheumatology, 1st Faculty of Medicine, Charles University, Prague, Czech Republic. Background/Purpose: Systemic sclerosis (SSc) is characterized by aberrant activation of fibroblasts with increased release of extracellular matrix components. TGF␤ is a key-mediator of fibroblast activation in SSc. More recently, hedgehog signaling (Hh) has also been shown to induce tissue fibrosis in SSc. Gli-2 is activated by Hh signaling and induces the transcription of hedgehog target genes. Since the TGF␤ and Hh signaling are both key drivers of fibroblast activation in SSc, the aim of this study was to investigate potential crosstalk between those two pathways in SSc. Methods: The expression of Gli2 in the fibroblasts and tissue was analyzed by real- time PCR and IF staining. Collagen synthesis was quantified by real-time PCR and hydroxyproline assay. Selective knockdown strategies were used to evaluate the role of Gli2 in TGF␤ signaling in vitro and in vivo. The interaction between Gli-2 and TGF␤ signaling was analyzed using mice with inducible, fibroblast-specific knockout of Gli2 (Gli-2; Col1a2 CreER) and fibroblast-specific overexpression of a constitutively active TGF␤ receptor I TBR (TBR; Col1a2 CreER). Cre mediated recombination was induced postnatally by injection of tamoxifen. Results: The expression of Gli-2 was significantly increased in the skin of SSc patients compared to healthy controls. In particular, SSc fibroblasts

expressed high levels of Gli-2. A prominent overexpression of Gli-2 in fibroblasts was also observed in mice overexpressing TBR. Consistently, stimulation with TGF␤ increased the mRNA and the protein level of Gli2 in a time-dependent manner with maximal effects after 3 hours of stimulation (6-fold increase, p⬍0.05) and induced nuclear translocation of Gli2. Gli-2 knockout fibroblasts expressed lower levels of TGF␤ targeted genes col1a2 (⫺47 %), CTGF (⫺30 %), Smad7 (⫺44 %), PAI-1 (⫺37 %) upon stimulation with TGF␤ compared to control fibroblasts (p⬍0.05 each). The role of Gli-2 as a downstream mediator of TGF␤ in fibroblasts was further analyzed in vivo using mice with fibroblast-specific depletion of Gli-2 and simultaneous overexpression of TBR. Fibroblast-specific overexpression of TBR in TBR; Col1a2 CreER mice induced progressive skin fibrosis. Fibroblast-specific, postnatal depletion of Gli-2 in Gli-2; Col1a2 CreER did not result in a basal skin phenotype. However, fibroblast-specific knockdown of Gli-2 protected from TBR induced fibrosis. Gli-2 ⫻ TBR; Col1a2 CreER mice treated with tamoxifen showed reduced dermal thickening (⫺30 %), decreased myofibroblast counts (⫺41 %) and lower hydroxyproline levels (⫺45 %) (p⬍0.05 each) compared to TBR; Col1a2 CreER mice. Conclusion: We present here first evidence for a novel interaction of TGF␤ and Hh signaling in fibrosis. TGF␤ stimulates Hh signaling in a non-canonical manner by activation of the downstream transcription factor Gli-2. Fibroblast-specific knockdown of Gli-2 reduces the stimulatory effects of TGF␤ on fibroblasts and ameliorates TBR-induced skin fibrosis, thereby demonstrating Gli-2 is as a novel downstream mediator of the pro-fibrotic effects of TGF␤. Disclosure: B. Sumova, None; C. Cordazzo, None; K. Palumbo-Zerr, None; C. Dees, None; P. Zerr, None; O. Distler, Sanofi, Active Biotech, Pfizer, Actelion, and Novartis, 2, Actelion, Sanofi-Aventis, Roche, Genentech, Active Biotec, medac, Novartis, Sinoxa, 4D Science, 5; G. Schett, None; L. Senolt, None; J. H. W. Distler, None.

2709 Secreted Frizzled-Related Protein 4 Can Be Induced By Transforming Growth Factor-␤, Is Regulated By Caveolin-1 and Can Induce NonCanonical Wnt Signaling In Fibroblasts. Justin Gillespie1, Giuseppina Abignano1, Michael McDermott1, Paul Emery1 and Francesco Del Galdo2. 1 Leeds Institute of Rheumatic and Musculoskeletal Medicine, University of Leeds, Leeds, United Kingdom, 2Leeds Institute of Rheumatic and Musculoskeletal Medicine, University of Leeds and NIHR Leeds Musculoskeletal Biomedical Research Unit, Leeds Teaching Hospitals NHS Trust, Leeds, United Kingdom. Background/Purpose: Systemic Sclerosis (SSc) is a heterogeneous disease characterized by autoimmune activation, fibroproliferative vasculopathy, and tissue fibrosis of skin and multiple internal organs. Several studies have indicated that both caevolin-1 (CAV-1) and WNT/␤-catenin signaling play important roles in the pathogenesis of tissue fibrosis. Indeed, CAV-1 is downregulated by 40% in SSc skin compared to healthy controls and, intriguingly, tissue expression studies with SSc skin biopsies show both upregulation of canonical WNT ligands1,2 and consistent upregulation of Frizzled-Related Protein 4 (SFRP4), a putative WNT antagonist, at both mRNA at protein level3,4. To evaluate the role CAV-1 and TGF␤ on the expression of SFRP4, and the effects of SFRP4 upon WNT signaling in fibroblasts. Methods: Immortalized primary healthy (HC) and SSc fibroblasts were cultured in 10% DMEM and starved in 0.5% DMEM for 24hrs prior to stimulation with recombinant TGF␤ (10ng/ml), WNT-3a/5a (100ng/ml) and/or SFRP4 (100–500ng/ml). Gene expression was quantified by SYBRgreen RTPCR and by western blot. CAV-1 siRNA was transfected at 5␮M. Canonical WNT signaling was assessed by TOPFlash luciferase reporter activity. ELISA was used to measure both the level of Phospho-c-JUN from whole cell lysates, and the level of SFRP4 from SSc patient sera. Results: In SSc fibroblasts, the basal expression of SFRP4 was increased at protein level and also by 264% at mRNA level compared to HC [p⬍0.001]. TGF␤ stimulation upregulated SFRP4 mRNA by 170% [P⬍0.01] at 48hrs and by 348% [P⬍0.01] at 72hrs. TGF␤ also induced a time-dependent increase of both SFRP4 and ␣-SMA protein expression, while reducing CAV-1. The siRNA-mediated silencing of CAV-1 was sufficient to induce a time-dependent increase in SFRP4 protein expression. WNT-3a induced a 600% increase in TOPFlash activity, while co-treatment with SFRP4 suppressed this activation by 59%. In contrast, SFRP4 induced a 260% increase [p⬍0.001] in c-JUN phosphorylation, a non-canonical WNT target, at 10 min in both HC and SSc fibroblasts. This was similar to the activity of the non-canonical WNT-5a ligand. Interestingly, basal Phospho-c-JUN was

S1157

Tuesday, October 29

models, Fli1 haploinsufficiency results in increased expression of type I collagen at mRNA and protein levels in the back skin, but fibrosis does not occur at least partly because the expression of CTGF, which plays a critical role to establish and maintain tissue fibrosis, remains unchanged. Since our pilot studies have revealed that transcription factor KLF5 is a potent repressor of the CTGF gene and its expression is epigenetically suppressed in SSc dermal fibroblasts, we generated mice with heterozygous deletions of Fli1 and KLF5 and evaluated whether they recapitulated the clinical and histopathological features of SSc. Methods: mRNA and protein levels of target molecules were determined by quantitative real-time PCR and by immunoblotting, immunohistochemistry and ELISA, respectively. Vascular structure in the back skin was visualized by FITC-dextran intravenously injected. Ultrastructue of collagen fibers was evaluated by electron microscopy. Auto-antibodies were detected by indirect immunofluorescence with Hep2 cells. Results: Klf5⫹/⫺;Fli1⫹/⫺mice (C57BL/6) were fertile, born in Mendelian ratio, and did not display any apparent early developmental abnormalities. Notably, these mice spontaneously recapitulated cardinal features of SSc, including peripheral vasculopathy, fibrosis of skin and lung, autoimmunity and inflammation. Dermal vascular abnormalities, including stenosis of arterioles with a moth-eaten pattern and bushy capillary tips, occurred around the age of 1 month prior to the development of dermal fibrosis with ultrastructural changes of collagen fibrils similar to those seen in SSc. Interstitial lung disease with B cell lymphoid aggregates and vascular changes characteristic of pulmonary arterial hypertension and pulmonary venoocclusive disease appeared around the age of 2–4 months and progressed along with disease duration. Serum IL-6 levels and IL-6 mRNA levels in the skin and lung were elevated around the age of 2 months and splenic B cells secreted a greater amount of IL-6 in response to toll-like receptor 4 stimulation with LPS and/or T cell help in the form of an anti-CD40 antibody. Importantly, anti-nuclear antibodies were also detected. Conclusion: These studies underscore the concept of epigenetic reprogramming underlying pathogenic changes in SSc and implicate the Fli1 and KLF5 pathways as central mediators of this disease.

increased by 180% [P⬍0.005] in SSc compared to HC fibroblasts. Further, SFRP4 treatment increased COL1A1 expression by 26% in HCs at 500ng/ml, while in SSc fibroblasts SFRP4 induced both COL1A1 and ␣-SMA by 49–50% with 100–500ng/ml SFRP4, respectively. Additionally, SSc sera levels of SFRP4 inversely correlated with diffusion lung capacity of carbon monoxide % (r⫽⫺0.29, P⫽0.01) and positively with mRSS in patients with normal lung function (r⫽0.40, p⫽0.006). Conclusion: The increased expression of SFRP4 observed in SSc skin biopsies may be a direct consequence of CAV-1 downregulation by TGF␤ in tissue fibroblasts. Given the non-canonical WNT activity of SFRP4, a TGF␤ primed microenvironment may be responsible for shaping the phenotype of both fibroblasts and neighboring cells, through aberrant WNT pathway activation. Additionally, circulating SFRP4 may have potential as a biomarker of fibrosis. Investigation of the mechanisms linking CAV-1 expression and SFRP4 function will improve our understanding of the pathogenic role that aberrant WNT activation plays in SSc. Disclosure: J. Gillespie, None; G. Abignano, None; M. McDermott, None; P. Emery, None; F. Del Galdo, None.

ACR Concurrent Abstract Session T-cell Biology and Targets in Autoimmune Disease Tuesday, October 29, 2013, 2:30 PM–4:00 PM

2710

Tuesday, October 29

Deficiency Of The Protein Kinase Ataxia Telangiectasia Mutated Accelerates T Cell Aging In Rheumatoid Arthritis. Zhen Yang1, Hiroshi Fujii2, Eric L. Matteson3, Jorg J. Goronzy1 and Cornelia M. Weyand1. 1Stanford University School of Medicine, Stanford, CA, 2Tohoku University, Sendai, Japan, 3Mayo Clinic, Rochester, MN. Background/Purpose: T lymphocytes hold a pinnacle position in the pathogenesis of rheumatoid arthritis (RA) and through their longevity contribute to disease chronicity. The protein kinase Ataxia telangiectasia mutated (ATM) is traditionally considered critical in DNA damage repair and cell cycle checkpoint regulation. Recent studies have extended ATM function from the nucleus to the cytoplasm where it acts as an oxidative stress sensor and regulates cellular metabolism and apoptosis, affecting cellular fate decisions and survival. CD4 T cells from RA patients express low levels of ATM protein and are inefficient in repairing damaged DNA. Whether ATM deficiency affects reactive oxygen species (ROS) sensing, metabolic competence, apoptotic sensitivity as well as T cell functional differentiation is unknown. Methods: Naı¨ve T cells (CD4⫹CD45RO⫺) were purified from the blood of RA patients (n⫽59) and age-matched controls (n⫽62). T cell proliferation, cell cycle progression and IL-2 production was compared 48 and 72 hours following TCR ligation. Naı¨ve-to-Memory conversion was monitored by cytometric phenotyping. Resting and activated T cells were loaded with fluorogenic dyes (DCF and DHE) to quantify H2O2 and O2-production. The G2/M cell cycle checkpoint was measured with anti-phospho-Histone H3. Expression of ATM, p-ATM (Ser1981), AMPKa1, and p-AMPKa1 (Thr172) were quantified by immunoblotting. ATM signaling was blocked through the inhibitor ku-55933 or siRNA-mediated knockdown. Results: Compared to age-matched control T cells, RA T cells produced low levels of ATM transcripts, ATM monomers and ATM dimers. In search for upstream signals triggering ATM activity, control and RA T cells were compared for the intracellular levels of H2O2 and O2-. Both ROS were diminished in RA T cells. Reduced ATM signaling in RA T cells resulted in delayed histone H2Ax phosphorylation. ATM-deficient RA T cells passed through the G2/M cell cycle checkpoint without proper arrest, accelerating their differentiation into IL-2-producing effector cells. The modified cell cycle behavior led to the rapid conversion of naı¨ve into memory T cells, depleting the naı¨ve T cell reserve in RA patients. Abnormalities in cell cycle progression and T cell differentiation were reproduced in healthy T cells by blocking ATM signaling with ku-55933 or by siRNA-mediated ATM knockdown. Further downstream effects of insufficient ATM signaling in RA T cells included decreased activation of AMPK, a kinase critically involved in energy sensing and metabolic regulation. Conclusion: Deficiency of ATM redirects cell cycle behavior, metabolic fitness and functional differentiation of RA T cells. ATM-low RA T cells fail to adhere to cell cycle checkpoints, proliferate faster and differentiate prematurely. Also, ATM deficiency impairs triggering of the energy sensor AMPK, leaving the cell without sufficient energy supply. Consequences

include the depletion of the naı¨ve T cell reserve, accelerated T cell aging and excessive production of T cell effector cytokines. Overall, these abnormalities in T cell homeostasis and function bias the immune system toward chronic non-resolving inflammation. Disclosure: Z. Yang, None; H. Fujii, None; E. L. Matteson, None; J. J. Goronzy, None; C. M. Weyand, None.

2711 Self-Specific Polyclonal CD4 T Cells Enter a FR4hi CD73hi Anergic State To Prevent Immunopathology and Serve As a Reservoir For The Differentiation Of Protective Foxp3ⴙ Tregs. Lokesh Kalekar and Daniel L. Mueller. University of Minnesota Medical School, Minneapolis, MN. Background/Purpose: Clonal anergy has been proposed as a mechanism to allow for peripheral recognition of self antigens, without the consequence of immunopathology. Model systems have confirmed that CD4 T cells can become unresponsive to self antigen stimulation, but there is currently no evidence for the existence of anergic CD4 T cells in the normal polyclonal repertoire. Methods: Gene arrays were used to identify mRNAs specifically overexpressed in anergic ovalbumin (OVA)-specific OT-II TCR-transgenic CD4 T cells after adoptive transfer into healthy OVA-transgenic mice. Overexpression of anergy molecules was confirmed by multiparameter flow cytometry in anergic glucose-6-phosphate isomerase (GPI)-specific KRN TCR-transgenic CD4 T cells after adoptive transfer to self antigen-expressing healthy mice. Antigen-experienced polyclonal CD4 T cells expressing these anergy molecules were characterized by flow cytometry, purified, and tested for their capacity to cause immunopathology in lymphopenic hosts. Finally, adoptive transfer recipients of anergic polyclonal CD4 T cells were tested for evidence of autoantibody production and weight loss versus the generation of protective Foxp3⫹ Tregs. Results: Using microarrays to compare mRNA expression in naive, effector, and anergic OT-II CD4 T cells, we discovered that genes for folate receptor 4 (Folr4; FR4) and ecto-5’-nucleotidase (Nt5e; CD73) were greatly over-expressed following the induction of clonal anergy. Examination of KRN CD4 T cells following their initial adoptive transfer to GPI-expressing hosts confirmed up-regulation of FR4 and CD73 during the induction of anergy, and a reversal of this anergic state following adoptive transfer of FR4hi CD73hi anergic KRN T cells to TCRa-deficient lymphopenic hosts. Making use of these FR4 and CD73 anergy markers, 2–3% of the peripheral mature polyclonal CD4 repertoire was discovered to be anergic. Foxp3 was not expressed; however, CD69, PD-1, CTLA4, CD5, and Nur77 up-regulation suggested continuous engagement by high affinity TCRs on these tolerant T cells. Consistent with in vivo anergy induction in response to self antigen recognition, antigen-experienced (CD44hi) insulin/I-Ag7 tetramerbinding polyclonal CD4 T cells in NOD mice demonstrated high expression of FR4 and CD73. Adoptive transfer of FR4hi CD73hi polyclonal CD4 T cells to syngeneic TCRa-deficient lymphopenic hosts led to reconstitution of the Foxp3⫹ Treg compartment. Remarkably, elimination of newly differentiated Tregs in these anergic CD4 T cell recipients led to fatal wasting disease and the production of autoantibodies. Conclusion: Peripheral self antigen recognition in normal hosts drives autoreactive CD4 T cells into an anergic state that is associated with the up-regulation of FR4 and CD73. Anergic T cells remain in the peripheral repertoire, both as a threat to health upon anergy reversal as well as a reservoir of precursors for protective Foxp3⫹ Tregs. A greater understanding of the role of anergy induction and reversal in the development of rheumatic diseases will offer new insights into strategies designed to restore self tolerance. Disclosure: L. Kalekar, None; D. L. Mueller, None.

2712 Increased Soluble PD-1: A Link Between Generation Of Immunological Memory and Risk Of Disease Flare In Early RA. Stinne Greisen1, Tue W. Kragstrup1, Kristian Stengaard-Pedersen2, Merete Lund Hetland3, Kim Hørslev-Petersen4, Malene Hvid1 and Bent Deleuran1. 1Aarhus University, Aarhus, Denmark, 2Arhus University Hospital, Aarhus, Denmark, 3The Danish Rheumatologic Database (DANBIO), Glostrup Hospital., Copenhagen, Denmark, 4University of Southern Denmark, Graasten, Denmark. Background/Purpose: Disturbances in the regulatory mechanisms of the immune system play a major role in rheumatoid arthritis (RA). Programmed Death 1 (PD-1) on lymphocytes negatively regulate T cell receptor (TCR)-

S1158

Disclosure: S. Greisen, None; T. W. Kragstrup, None; K. Stengaard-Pedersen, None; M. L. Hetland, None; K. Hørslev-Petersen, None; M. Hvid, None; B. Deleuran, None.

2713 IL-21 Inhibited Follicular Regulatory T Cells To Promote Autoreactive Germinal Center Development In Autoimmune BXD2 Mice. Yanna Ding1, Hui-Chen Hsu2, Jun Li1, PingAr Yang1, Qi Wu1, Allan J. Zajac1 and John D. Mountz1. 1University of Alabama at Birmingham, Birmingham, AL, 2 Birmingham VA Medical Center, Birmingham, AL. Background/Purpose: Follicular T helper (Tfh) cells have been correlated with germinal center (GC) formation, autoantibody production and disease severity in human systemic lupus erythematosus (SLE). Follicular regulatory T cells (Tfr) are a regulatory CD4 T (Treg) subset that counteract Tfh cells and suppress GC B cell differentiation. We recently identified that interleukin 21 (IL-21) promotes Tfh differentiation in autoimmune BXD2 mice that develop spontaneous GCs. However, the role of IL-21 in regulating Tfr is unknown. The aim of this study is to determine the modulatory effects of IL-21 on Tfr/Tfh balance in BXD2 mice. Methods: Real-time PCR analysis was used to determine the expression of Treg-related genes in sorted Tfr cells after in vitro cultured with IL-21 (50ng/ml) stimulation. Exogenous IL-21 was provided to BXD2-Il21⫺/⫺ mice via adenovirus (Ad)-IL21 administration to investigate the effect of IL-21 on Tfr in vivo. Sorted Tfr cells from BXD2-Il21⫺/⫺ mice were co-cultured with CD4 T cells and B cells in vitro as well as transferred into young BXD2 mice in vivo to determine the function of Tfr. FACS staining was used to determine the percent and phenotype of Tfh and Tfr. Confocal imaging analysis was used to visualize the distribution of Tfr cells and GCs. ELISA was used to determine serum levels of IL-21 and TGF-␤1. ELISPOT was used to determine the generation of antibody producing B cells.

Results: GL-7⫹Fas⫹ GC B cells and CXCR5⫹ICOS⫹ Tfh cells were significantly reduced, but the frequency of FoxP3⫹CXCR5⫹ICOS⫹ Tfr cells was 2-fold higher in BXD2-Il21⫺/⫺ mice than in wild-type BXD2 mice, although the frequencies of conventional Treg cells were equivalent. The Tfr cells in BXD2-Il21- /- mice expressed high level of other Treg markers including CTLA4, TGF-␤1 and GITR. FoxP3⫹Tfr cells were localized in spleen follicular areas of BXD2-Il21⫺/⫺ mice. Transfer of Tfrs from BXD2Il21⫺/⫺ mice into BXD2 mice although decreased GC size and reduced IgM, it did not suppress IgG autoantibody producing B cells in the spleen. AdIL-21 administration to BXD2-Il21⫺/⫺ mice increased GC B cells but decreased the serum TGF-␤1, membrane TGF-␤1 expression in Tfrs and the ratio of Tfr/Tfh in the spleen. In vitro, IL-21 decreased FoxP3⫹ and significantly reduced Tgfb, Gitr and Il2 expression in Tfrs. IL-21 also counteracted Tfr-induced B cells death and inhibition of Tfhs. Conclusion: The present study suggests that IL-21 positively promotes autoreactive GC reactions in BXD2 mice at least partially by decreasing Tfrs and compromising the suppressive function of Tfrs on B cells and Tfhs. The results further suggest that IL-21 can skew the balance between Tfr and Tfhs to favor Tfh differentiation. Thus, IL-21 blockage is necessary when developing a Tfr transferring therapeutic strategy to inhibit autoimmune GC response. Disclosure: Y. Ding, None; H. C. Hsu, None, 2; J. Li, None, 2; P. Yang, None; Q. Wu, None; A. J. Zajac, None, 2; J. D. Mountz, None, 2.

2714 Commensal Intestinal Microbiota Drives Spontaneous Interleukin-1and T Helper 17-Mediated Arthritis In Mice. Shahla Abdollahi-Roodsaz1, Rebecca Rogier2, Tom Ederveen2, Harm Wopereis3, Raish Oozeer3, Marije I. Koenders2 and Wim B. van den Berg1. 1Rheumatology Research and Advanced Therapeutics, Radboud University Nijmegen Medical Centre, Nijmegen, Netherlands, 2Rheumatology Research and Advanced Therapeutics, Radboud University Nijmegen Medical Center, Nijmegen, Netherlands, 3 Danone Research, Wageningen, Netherlands. Background/Purpose: Altered composition of intestinal microbiota in recent-onset rheumatoid arthritis (RA) and possible efficacy of oral antibiotics suggest a role of intestinal microbiota in RA. This study aimed to investigate the involvement of commensal intestinal microbiota in T cell-dependent experimental arthritis. Methods: IL-1 receptor antagonist deficient (IL-1Ra⫺/⫺) mice spontaneously developing T cell-driven IL-17-dependent autoimmune arthritis were used. Intestinal and systemic T cell differentiation and arthritis development were studied in conventional and germ-free (GF) mice. Contribution of intestinal microbiota was investigated using oral broad-spectrum and selective antibiotic treatments via drinking water, combined with recolonization by specific microbiota. Multiplex 454 pyrosequencing of V5 and V6 hypervariable regions of fecal bacterial 16S rRNA was used to identify specific microbiota associated with arthritis. Results: Compared to wild-type mice, small intestinal lamina propria of IL-1Ra⫺/⫺ mice contained increased Th17 and to lower extent Th1 percentages, both of which significantly correlated with arthritis severity. Importantly, GF IL-1Ra⫺/⫺ mice had a marked abrogation of arthritis along with reduced intestinal Th1 and in particular Th17. GF IL-1Ra⫺/⫺ mice exhibited a notable decrease in IL-1␤ and IL-17 production by splenocytes upon CD3 and Toll-like receptor stimulations, suggesting abolishment of systemic Th17 response. Interferon ␥ was only detectable upon CD3 stimulation and was reduced as well. Relevance of intestinal microbiota was underlined by significant longterm suppression of arthritis by one-week oral treatment with Metronidazole, Neomycin and Ampicillin (each 1g/l). Interestingly, recolonization of antibiotictreated IL-1Ra⫺/⫺ mice by segmented filamentous bacteria, previously reported as a prominent intestinal Th17 inducer, was sufficient to cause full-blown arthritis. Elimination of distinct intestinal microbiota subsets showed that arthritis is suppressed when Gram-negative, but not Grampositive, bacteria are selectively eradicated, indicating members of intestinal Gram-negative commensals may drive arthritis. High-throughput pyrosequencing revealed lower microbiota abundance (operational taxonomic units) and reduced species richness and diversity (Chao and Shannon indices, resp.) in arthritic IL-1Ra⫺/⫺ compared to wild-type mice. The genus Helicobacter, belonging to Gram-negative bacteria, was found associated with arthritis severity (0.0% in wild-type versus 1.1 % in arthritic mice). The disease was further associated increased Anaeroplama and Lactobacillus accompanied by decreased Bacteroidetes. Valida-

S1159

Tuesday, October 29

induced activation through interaction with two ligands (PD-L1 and PD-L2) on activated antigen-presenting cells, including fibroblasts. PD-1 exists in a soluble form (sPD-1), and is increased in chronic RA. Here, we investigate whether sPD-1 plays a role in the generation of immunological memory in early RA. Methods: In a longitudinal set of steroid- and DMARD naı¨ve RA patients (eRA) (n⫽76, ⬍3 months of disease) we measured plasma levels of sPD-1 by ELISA (R&D systems) at baseline and after 3, 6 and 12 months of treatment with methotrexate (MTX) ⫹ placebo (PLA) or MTX ⫹ adalimumab (ADA). After 12 months of treatment, ADA/PLA was discontinued and patients were followed every third month for flare. sPD-1 was also measured in chronic RA (cRA) in plasma and synovial fluid (SF) (n⫽24, ⬎8 years of disease). Data were expressed as median (IQR) and correlations were assessed by Spearman’s rho. T cells were stimulated with anti-CD3/CD28 and co-cultured with RA fibroblast-like synoviocytes (FLS) for 3 days with addition of sPD-1 (20ng/ml), sPD-L1 (20ng/ml) or TNF␣ (10ng/ml). T cells were examined by flow cytometry for expression of PD-1, CD25, CD69 and proliferation, and FLS were examined for PD-L1 expression. Results: The plasma levels of sPD-1 was increased in eRA (0.52 ng/ml (0.3–1.1 ng/ml)) compared with cRA (0.14 ng/ml (0.07 ng/ml ⫺0.33 ng/ml), p⬍0.001) and decreased following treatment (0.26 ng/ml (0.2–0.6 ng/ml), p⬍0.001)). Treatment with ADA did not influence this decrease. Baseline sPD-1 correlated tightly with anti-CCP (r⫽0.42) and IgM-RF (r⫽0.60, both p⬍0.001). The correlation persisted at 3, 6 and 12 months. Patients with high baseline sPD-1 showed increased risk of disease flares (p⬍0.05). In cRA sPD-1 levels were increased in SF (0.31ng/ml, (0.14 ng/ml ⫺0.73 ng/ml)) compared with plasma (p⬍0.01). T cell survival in FLS co-cultures was 76% when adding sPD-1 compared with 61% in controls. Stimulation with sPD-1 did not affect proliferation, CD25, CD69 or PD-1 expression by T cells. Co-culture with stimulated T cells increased PD-L1 expression on FLS from 2% to 45%. In co-cultures with TNF␣, PD-L1 was expressed on nearly all FLS. Conclusion: Plasma levels of sPD-1 were increased in eRA and decreased in response to treatment suggesting sPD-1 to be important for the initial phase of RA. The high concentrations found in SF points to a function of sPD-1 locally in the joint. The close association with IgM-RF and anti-CCP, the increased risk of flare with high sPD-1 in eRA plasma and the ability of sPD-1 to promote T cell survival in co-cultures with FLS support a role for sPD-1 in adaptive immunity and generation of immunological memory. Taken together, our findings suggest that sPD-1 is capable of blocking the PD-L1 binding-site on FLS, inhibiting the effects of the PD-1-induced TCR regulation and thereby increasing the risk of generating auto-reactive cells.

tion of microbiota alterations and studies on T cell-modulatory and diseaseinducing characteristics in GF mice are in progress. Conclusion: The presence of commensal intestinal microbiota is critical for the development of autoimmune T cell-driven arthritis, probably via shaping the T cell differentiation by Gram-negative bacteria. Understanding the molecular and cellular mechanisms linking the intestinal T cell response with extra-intestinal disease may help identify novel therapeutic targets in RA.

Conclusion: Discrepancy between PGA and EGA, which is supposed to reflect disease activity of RA, may be partially explained by the difference in psychological state, and support for psychological state may be effective for improvement of PGA. At present, clinical, radiological and functional remissions have been proposed as goal of treatment of RA. As PGA remission is a limiting factor for Boolean remission and PGA relates to psychological state, psychological remission might be important to attain true remission.

Disclosure: S. Abdollahi-Roodsaz, None; R. Rogier, None; T. Ederveen, None; H. Wopereis, None; R. Oozeer, None; M. I. Koenders, None; W. B. van den Berg, None.

Disclosure: M. Fusama, None; K. Yukioka, None; T. Kuroiwa, None; C. Yukioka, None; M. Inoue, None; T. Nakanishi, None; N. Murata, None; N. Takai, None; K. Higashi, None; T. Kuritani, None; K. Maeda, None; Y. Miura, None; H. Sano, None; M. Yukioka, None; H. Nakahara, None.

ARHP Concurrent Abstract Session Clinical Practice/Patient Care Tuesday, October 29, 2013, 2:30 PM–4:00

2715

2717 Fatigue in Rheumatoid Arthritis Is Associated With Modifiable Lifestyle Factors. Patricia P. Katz1, Vladimir Chernitskiy2 and David I. Daikh1. 1 University of California, San Francisco, San Francisco, CA, 2University of California San Francisco, San Francisco, CA.

PM

WITHDRAWN

Tuesday, October 29

2716 Support For Psychological State May Be Effective to Attain Remission Of Boolean-Based Definition Of Patient Global Assessment In Patients With Rheumatoid Arthritis. Mie Fusama1, Kumiko Yukioka2, Takanori Kuroiwa2, Chikako Yukioka2, Miyako Inoue2, Tae Nakanishi2, Norikazu Murata2, Noriko Takai3, Kayoko Higashi4, Taro Kuritani4, Keiji Maeda4, Yasushi Miura1, Hajime Sano5, Masao Yukioka2 and Hideko Nakahara4. 1 Kobe University Graduate School of Health Sciences, Kobe, Japan, 2 Yukioka Hospital, Osaka, Japan, 3Yukioka College of Health Science, Osaka, Japan, 4NTT West Osaka Hospital, Osaka, Japan, 5Hyogo College of Medicine, Nishinomiya-city, Japan. Background/Purpose: The patients with rheumatoid arthritis (RA) often do not meet patient global assessment (PGA) remission criterion despite a good clinical disease state and, therefore, PGA is not solely influenced by RA disease activity. It has been reported that PGA is a limiting factor for reaching remission. We evaluated whether psychological state influences on Booleanbased definition of PGA remission for patients with RA. Methods: The patients with RA (n⫽112) were recruited. Disease activity was evaluated with swollen joint counts (SJC) and tender joint counts (TJC) and evaluator’s global assessment (EGA), PGA, and patient pain (Pain VAS) were assessed with visual analog scale. Depression and anxiety were examined utilizing the Hospital Anxiety and Depression Scale-Depression (HADS-D) and the State-Trait Anxiety Inventory (STAI), respectively. General health status was evaluated with Sort Form-36 (SF-36). Comparison analyses were performed between remission group (VAS⫽⬍1cm) and non-remission group (VAS⬎1cm) in PGA, EGA, and pain VAS. Data analyses were performed utilizing Wilcoxon rank-sum test and Spearman correlation analysis. Results: There was no significant difference in age, duration and PSL dosage between PGA remission group (n⫽35) and non-remission group (n⫽77) or EGA remission group (n⫽38) and non-remission group (n⫽74). SJC and TJC of remission groups were significantly lower in PGA and EGA (p⬍0.0001, p⬍0.0001, respectively). Among TJC, SJC and pain VAS, PGA was most significantly related to pain VAS (r⫽0.9232, p⬍0.0001), while EGA was related to SJC (r⫽0.7725, p⬍0.0001). STAI (State) and HADS-D were significantly lower in pain VAS remission group (p⫽0.0018, p⫽0.0005, respectively). STAI (State) was significantly lower in PGA remission group (p⬍0.05), while no significant difference was found between two groups in EGA (p⫽0.2463). Similarly in HADS-D, the number of patients with depression was significantly fewer in PGA remission group (p⬍0.05), while no significant difference was found between two groups in EGA (p⫽0.2522). In SF-36, all of the eight components improved significantly in PGA remission group. There was no significant difference in mental health between two groups in EGA (p⫽0.1602). PGA is related to pain VAS, in contrast EGA is related to SJC. The patients with pain VAS remission are less depressive and less anxious. The patients with PGA remission have depression or anxiety less than those with PGA non-remission. Moreover, mental health of SF-36 was significantly better in PGA remission group. On the contrary, there is no significant difference in anxiety, depression and mental health of SF-36 between two groups in EGA.

Background/Purpose: Fatigue has been identified as a major concern for individuals with rheumatoid arthritis (RA) and has been endorsed as a core outcome measure. However, in order to develop adequate treatments for fatigue, its sources need to be identified. Methods: During home visits (n⫽160), assessments were made of fatigue (Fatigue Severity Index; FSI), self-reported sleep quality (Pittsburg Sleep Quality Index; PSQI), depression (Patient Health Questionnaire-9; PHQ9), usual levels of physical activity (International Physical Activity Questionnaire; IPAQ), RA disease activity (RA Disease Activity Index; RADAI), muscle strength (by hand-held dynamometer), functional limitations (Short Physical Performance Battery; SPPB), body composition (body mass index and percent fat mass by bioelectrical impedance), and pulmonary function (by spirometry). Blood was drawn to measure inflammatory markers (C-reactive protein [CRP], tumor necrosis factor-a [TNF], and interleukin-6 [IL-6]), RA-specific markers, anemia, and thyroid function, and information was collected on demographics, medication use, and smoking. The FSI average fatigue level over the past 7 days, rated on a 0–10 (no-severe fatigue) scale, was used as the outcome measure. Analyses were first conducted to evaluate bivariate relationships with fatigue. Multivariate analyses, including all variables significantly associated with fatigue in the bivariate analyses, were then conducted to identify independent predictors of fatigue. Results: Mean age (⫾SD) was 59 (⫾11), mean disease duration was 21 (⫾13) years, and 85% were female. Mean FSI rating was 3.8 (⫾2.0; range 0–10). Significant bivariate relationships with fatigue are shown in the Table. No significant associations were found for age, sex, ethnicity, disease duration, knee extension or grip strength, TNF, IL-6, pulmonary function, or prednisone or biologic use (not shown). In multivariate analyses, obesity, RADAI, depression, and smoking remained significantly and independently associated with fatigue. Table. Factors associated with fatigue

BMI obesity Disease activity (RADAI) Depressive symptoms (PHQ) Sleep quality (PSQI) Hip flexor strength CRP Physical activity ⱖ150 minutes/week in moderate/vigorous activity Ever smoke

Bivariate b (p) 1.56 (⬍.0001) 0.68 (⬍.0001) 0.25 (⬍.0001) 1.21 (⬍.0001) ⫺0.03 (.004) 0.22 (.008) ⫺1.20 (.0005)

1.16 (.0008)

Multivariate* b (p) 1.54 (.006) 0.30 (.005) 0.12 (.002) 0.36 (.07) ⫺0.006 (.52) ⫺0.009 (.89) ⫺0.23 (.41)

0.69 (.009)

* Multivariate model included all variables significantly associated with fatigue in bivariate analyses

In additional analyses, performing ⱖ150 minutes/week of moderate/ vigorous physical activity was associated with lower odds of obesity (OR⫽0.29 [95% CI 0.15, 0.55]), lower depression score (p⫽.002), and better sleep quality (p⫽.04), suggesting that that physical activity may have an indirect association with fatigue, mediated by these factors.

S1160

Conclusion: In this cross-sectional study, much of RA fatigue appeared to arise from factors that are secondary to the disease process itself (obesity, poor sleep quality, and depression), but are commonly present in RA. In addition, physical inactivity appeared to play a role, although its effects were indirect. Further research is warranted to determine the time-ordering of the relationships and if interventions that target modifiable lifestyle factors such as physical activity improve fatigue. Disclosure: P. P. Katz, None; V. Chernitskiy, None; D. I. Daikh, None.

2718 Exploring E-Health Ethics and Multi-Morbidity: A Qualitative Study Of Patient and Clinician Experiences Using Digital Media For Health Purposes. Anne F. Townsend1, Paul Adam2, Linda C. Li3, Jenny Leese4, Michael McDonald1, Sheila Kerr5, Gordon Whitehead5 and Catherine L. Backman4. 1University of British Columbia, Vancouver, BC, 2Mary Pack Arthritis Centre, Vancouver, BC, 3Arthritis Centre of Canada, Richmond, BC, 4 Arthritis Research Centre of Canada, Richmond, BC, 5Consumer Advisory Board, Vancouver, BC.

Disclosure: A. F. Townsend, None; P. Adam, None; L. C. Li, None; J. Leese, None; M. McDonald, None; S. Kerr, None; G. Whitehead, None; C. L. Backman, None.

2719 Urinary Angiostatin As Alternative To Biopsy In lupus Nephritis Patients Among Egyptian. Shereen Algergawy1, Osama Alshaar2 and Rania Zakaria3. 1associate prof at benha university faculty of medicine rheumatology and rehabilitation department Egypt, cairo, Egypt, 2benha university faculty of medicine, benha, Egypt, 3benha university faculty of medicine, benha, Egypt. Background/Purpose: Lupus nephritis (LN) is one of the most frequent manifestations of SLE and can be present in 60% of SLE patients. Kidney biopsy remains the basis of LN diagnosis. There is urgent need to identify biomarkers that alternative to renal pathology in lupus.

Disclosure: S. Algergawy, None; O. Alshaar, None; R. Zakaria, None.

2720 Providing Comprehensive, Comparative Post-TJR Outcome Feedback To Surgeons For Quality Monitoring and Value Decisions. Patricia D. Franklin1, Bruce Barton1, Leslie R. Harrold1, Wenjun Li1, Regis O’Keefe2, Jeroan Allison1 and David Ayers1. 1University of Massachusetts Medical School, Worcester, MA, 2University of Rochester Medical Center, Rochester, NY. Background/Purpose: With the CMS decision to publicly report hospital-specific post-operative total joint replacement (TJR) events and readmissions, surgeons and hospitals need comprehensive posthospital data by which to manage and monitor outcomes. Currently, hospitals lack complete data on post-discharge events because patients seek care at multiple emergency rooms and hospitals. In addition, more than 40% of US TJR patients are under 65 years of age and are not captured in CMS data. To address this data void, we developed methods to capture comprehensive post-TJR medical and surgical events, as well as longitudinal patient-reported outcomes, and to report comparative data to participating surgeons and sites. Methods: We established a national consortium of ⬎120 diverse orthopedists representing all regions of the US with varied hospital and surgeon practices to ensure that data reflect typical US practice. We are enrolling ⬎10,000 patients annually. Web-based and scannable paper data collection formats support efficient data gathering. Patients consent to participate and submit annual outcome assessments directly to us to assure complete reporting of hospital and emergency care. Post-TJR medical events are validated through chart review and CMS data (for those ⬎65 years). Quarterly post-operative event rates and PROs are calculated for the national cohort, the individual surgeon, and site to deliver comparative feedback. Results: A secure, HIPAA compliant MD website was established that presents summary and comparative descriptive statistics for primary TKR and THR for all of enrolled patients, specific to the site, and to the surgeon. A secure downloadable and printable report includes an Executive Summary of key outcome comparisons, as well as comprehensive tables of patient demographics, pre-operative medical and musculoskeletal comorbidities, post-operative events, and post-operative PROs enabling the providers to compare their outcomes to the other participating surgeons. Patient-level data are provided on request by the surgeon Conclusion: In an era of public reporting of outcomes, surgeons and hospitals need a single comprehensive source of post-discharge medical events, readmissions, and PROs to manage and monitor all patient outcomes. We developed a secure web-site to return comparative data to providers. These data are used to demonstrate value in payer negotiations and to guide quality improvement efforts. Disclosure: P. D. Franklin, NIAMS-NIH, NLM-NIH, AHRQ, Zimmer,, 2; B. Barton, None; L. R. Harrold, CORRONA, Inc., 5; W. Li, AHRQ, 2; R. O’Keefe, None; J. Allison, AHRQ, 2; D. Ayers, AHRQ, Zimmer, 2.

S1161

Tuesday, October 29

Background/Purpose: E-Health can potentially transform and enhance health care delivery and empower patients. E-health includes a spectrum of digital health services and information communicated through the Internet and mobile technologies. As digital tools are embraced it is vital to examine emerging ethical issues. We apply a relational ethics lens to situations arising in e-health for patients with arthritis and their health care providers (HCP). Relational ethics emphasizes context, interdependence and relationships that facilitate or constrain meaningful self-direction. The prevalence of comorbidities in patients with arthritis amplifies the complexity of managing diseases concurrently for both patients and HCP. We explore their perspectives on the use of e-health technologies and how e-health use impacts the patient-physician relationship, and in so doing make ethical issues explicit. Methods: This is Phase 1 of a 2-phase qualitative focus group (FG)/ interview study, informed by narrative and phenomenology to understand the ‘lived experience’ of e-health use. Eligible participants were: adults with multi-morbidity including arthritis and HCP with relevant caseloads, recruited via online ads, notices, and word of mouth. The FG guides were consistent across groups and organized around 4 areas: 1) E-health tools/devices; 2) descriptions of e-health experiences; 3) impact of e-health use on actions and decisions including patient-provider consultations; 4) a recap to check alignment and range of views. An iterative, constant comparative analysis began with independent open coding of transcribed data by at least 2 researchers; other team members discussed and clustered emerging codes; an ethical lens was then applied to clusters and key categories were identified and agreed upon by the wider team. Results: 36 participants (18 patients, 18 HCPs) participated in 7 FG, 4 with patients and 3 with rehabilitation professionals and physicians (HCP). We interviewed 4 HCPs who were unable to attend a FG. Patients and HCP expressed similar views about e-health, though examples, emphasis and priorities varied. Predominant themes were: 1) Changing notions of trust (e.g. privacy was a concern but less so than expected; 2) Responsibilities (e.g. patients used e-health for tasks to prepare for consultations; 3) Partnerships (e.g. concordance); 4) Burden (e.g. searching for relevant information was time consuming and could be overwhelming. Conclusion: There was evidence that fundamental aspects of patient-HCP relationships are shifting. It is critical to make the ethical issues in e-health explicit, as we track the transition towards empowered patients and receptive HCPs.

Aim: To study urinary angiostatin in SLE and investigate their possible role as indicative of renal involvement to be used as alternative to biopsy. Methods: Angiostatin levels were measured in urine samples from 45 lupus patients and 14 healthy volunteers, renal biopsy was obtained from all patients Results: SLE patients had elevated urinary angiostatin as compared to controls (P ⬍ 0.001). Levels of urinary angiostatin were higher in patients with an active LN (lupus nephritis ) than those with inactive LN (P ⬍ 0.001). In patients with LN urinary angiostatin correlated with the renal score of the Systemic Lupus Erythematosus Disease Activity Index. Urinary angiostatin levels varied significantly and there is significant positive correlation (P ⬍0.05) levels with the activity and choronicty scores of the examined renal biopsies among the histopathological groups. Conclusion: Urinary angiostatin is potentially useful markers of lupus activity. urinary angiostatin able to differentiate patients with active SLE from those with inactive disease.urinary levels are indicative of renal activity. Patients with Class IV lupus nephritis exhibited the highest levels of urinary angiostatin. Urinary angiostatin can be used as non invasive marker to detect renal involvement in lupus nephritis alternative to renal biopsy

2721

Tuesday, October 29

Exercise Therapy and Ultrasound Guided Glucocorticoid Injection In Patients With Painful Shoulder: A Randomised Controlled Trial. Karen Ellegaard1, Robin Christensen2, Sara Rosager Mortensen1, Cecilie Bartholby3, Søren Torp-Pedersen4, Thomas Bandholm5, Bente DanneskioldSamsøe4, Henning Bliddal4 and Marius Henriksen6. 1The Parker Institute, Copenhagen, Denmark, 2Musculoskeletal Statistics Unit, The Parker Institute, Department of Rheumatology, Copenhagen University Hospital, Bispebjerg and Frederiksberg, Denmark, 3The Parker Institute, Copenhangen, Denmark, 4 The Parker Institute, Department of Rheumatology, Copenhagen University Hospital, Bispebjerg and Frederiksberg, Denmark, Frederiksberg, Denmark, 5 Hvidovre Hospital, Copenhagen, Denmark, 6The Parker Institute, Copenhagen University Hospital at Frederiksberg, Copenhagen F, Denmark. Background/Purpose: SubAcromial Impingement (SAI) accounts for approximately 50% of all shoulder pain. The most commonly used treatments for SAI are glucocorticoid injection, exercise therapy or a combination of both. However the effect of exercise added to glucocorticoid injections is inadequately described in the peer reviewed literature. The aim of this study was to investigate the effectiveness of a standardized combination of exercise added to glucocorticoid injection compared to the glucocorticoid alone as the therapy for treatment of SAI. Methods: Patients with unilateral SAI of minimum 4 weeks duration and thickened subacromial bursa (⬎2mm assessed by ultrasound) were included (NCT: 01506804). At baseline patients were randomized to two steroid injections into the painful shoulder separated by one week with subsequent 10 weeks exercise therapy of the involved shoulder (intervention group, IG), or two steroid injections into the painful shoulder separated by one week with subsequent 10 weeks exercise therapy of the un-involved shoulder (defined as sham-intervention, control group; CG). The patients were re-examined after the exercise program (at week 12) and again at week 26. Primary outcomes were change from baseline in shoulder pain at rest and during active shoulder abduction analyzed using intention-to-treat population with a non-responder imputation for missing data (i.e., baseline observation carried forward). Results: 99 patients (58 females) with an average age of 49 years were randomised (49 IG/50 CG). In the IG, 17 participants were lost to follow-up; in the CG 18 were lost. At the 12 week follow-up there were no statistically significant group difference in pain at rest (MD⫽ 1.677 [95% CI ⫺3.644 to 6.001], P⫽0.533); and during abduction (MD⫽2.230 [95% CI ⫺6.462 to 10.921], P⫽0.612); both groups had decreased pain.. At the second follow up (26 weeks) the positive effect on pain at rest was maintained in the CG but decreased in the IG group; there was a trend towards a difference between groups (MD⫽ 5.613 [95% ⫺0.867 to 12.093], P⫽0.088). The beneficial effect on pain during active abduction was not different between groups at week 26 (MD⫽ 2.234 [95% CI ⫺6.754 to 11.224], P⫽0.623). Conclusion: We found no beneficial effects of combined glucocorticoid injection and exercise therapy of the involved shoulder compared to injections combined with exercise of the uninvolved shoulder. In fact, the CG experienced a prolonged effect on pain at rest compared to the IG. This finding is in contradiction to other studies showing a positive effect of exercise in patients who have received glucocorticoid injection prior to exercise therapy. Disclosure: K. Ellegaard, None; R. Christensen, None; S. R. Mortensen, None; C. Bartholby, None; S. Torp-Pedersen, None; T. Bandholm, None; B. DanneskioldSamsøe, None; H. Bliddal, None; M. Henriksen, Axellus A/S, 2, Mundipharma, 2, Norpharma, 9.

Rheumatology Research Foundation Special Session 2013 Rheumatology Research Foundation Edmond L. Dubois, MD Memorial Lectureship Tuesday, October 29, 2013, 4:30 PM–6:00 PM

2722 In Vivo Administration Of MiR-146a Protects C57BL/6 Mice From Pristane-Induced Pulmonary Hemorrhage Via Suppressing Type I Interferon Response. Dong Liang1, Shiyu Zhou2, Zheng Liu3, Zhengyuan Shan1, Philip Brohawn3, Yihong Yao3, John B. Harley1 and Nan Shen4. 1 Cincinnati Children’s Hospital Medical Center, Cincinnati, OH, 2Shanghai Institutes for Biological Sciences Chinese Academy of Sciences & Shanghai Jiao Tong University School of Medicine, Shanghai, China, 3MedImmune, LLC, Gaithersburg, MD, 4Shanghai Institute of Rheumatology, Renji Hospital, Shanghai Jiaotong University School of Medicine, Shanghai, China.

Background/Purpose: miR-146a as an endogenous regulator plays a critical role in resolving acute inflammation. The risk-associated genetic variant in miR-146a promoter was linked to reduced expression of miR146a in the peripheral blood leukocytes of lupus patients. Our previous studies also showed that overexpression of miR-146a in PBMCs from patients with SLE suppressed coordinate activation of type I interferon (IFN) pathway. Pristaneinduced lupus is a well established SLE murine model featuring abnormal activation of IFN pathway and unusual high penetrance of pulmonary capillaritis. This study explored a therapeutic potential of miR-146a in this induced murine model of lupus. Methods: The C57BL/6 (B6) mice were intravenously injected with PBS or miR-146a via lateral tail veins. The peripheral bloods from 3, 7 and 14 days post Pristane injection were evaluated by real time PCR or gene expression profile analysis. For comparison, the B6 mice were categorized into 3 groups, receiving respectively 3 consecutive injections of PBS (control group, n⫽9) or miR-146a either 3 days before (prevention group, n ⫽19) or 7 days after (treatment group, n⫽10) a single intraperiteneal injection of Pristane. The pulmonary hemorrhage was histologically investigated by HE staining 2 weeks post Pristane injection. Results: The injections of miR-146a mimics (agomirs) resulted in a dramatic increase in the expression of peripheral blood miR-146a in both the prevention and treatment groups as compared to the control group receiving PBS. HE staining showed that miR-146a administration rendered mice resistant to Pristane-induced hemorrhagic pulmonary capillaritis (Fig. 1). While in the control group, 56% mice injected with Pristane developed complete pulmonary hemorrhage, the prevalence was reduced to 25% in the treatment group. Of note, the complete hemorrhage was completely blocked in prevention group (Fig. 2). Furthermore, qPCR revealed that miR-146a was significantly lower in mice with pulmonary hemorrhage compared with mice developing no hemorrhage (p ⬍ 0.01), and inversely correlated with expression of Mx1 (p ⬍ 0.01), an IFN-inducible gene. Gene expression profile analysis showed that miR-146a injection substantially suppressed the IFN response, accompanied by reduced production of multiple pro-inflammatory cytokines and chemokines.

Conclusion: Our study provides evidence that miR-146a plays a suppressive role in the Pristane-induced pulmonary hemorrhage in B6 mice. It also highlights a potential pathogenic role of type I IFN pathway activation in the development of pulmonary capillaritis in Pristane-induced murine lupus. These findings suggest that SLE patients with pulmonary hemorrhage may benefit from therapeutic intervention to induce miR-146a expression. Disclosure: D. Liang, None; S. Zhou, None; Z. Liu, None; Z. Shan, None; P. Brohawn, None; Y. Yao, None; J. B. Harley, None; N. Shen, None.

2723 Role Of Interferon-Inducible RNA-Dependent Protein Kinase In The Pathogenesis Of Systemic Lupus Erythematosus. Aure´lie De Groof1, Benoıˆt Van den Eynde2, Fre´de´ric A. Houssiau1 and Bernard Lauwerys1. 1 Institut de Recherche Expe´rimentale et Clinique, Universite´ Catholique de Louvain, Brussels, Belgium, 2Institut de Duve, Universite´ Catholique de Louvain, Brussels, Belgium. Background/Purpose: PRKR (interferon-inducible RNA-dependent protein kinase) is an intracytoplasmic molecule that induces the production of interferon(IFN)-induced genes upon binding of double-stranded RNA viruses. PRKR is spontaneously over-expressed in PBMC from systemic lupus erythematosus (SLE) patients, and we previously demonstrated that in vitro exposure of SLE cells to 2-aminopurine (a pharmacological inhibitor of PRKR) results in a significant inhibition of IFN-induced genes and immunoglobulin production. In the present study, we wanted to further investigate the role of PRKR in the pathogenesis of SLE. Methods: Intraperitoneal 2-aminopurine versus PBS injections were administered to (NZB ⫻ NZW)F1 mice three times a week for a period of 2 months. In addition, PRKR-KO mice were backcrossed in B6.Sle1.Sle2.Sle3 tricongenic SLE-prone mice, resulting in the generation of PRKR-KO versus

S1162

Disclosure: A. De Groof, None; B. Van den Eynde, None; F. A. Houssiau, None; B. Lauwerys, None.

2724 cAMP Responsive Element Modulator (CREM)␣ Governs CD8 Expression and Contributes to the Generation of CD3ⴙCD4-CD8- T Cells in Lupus. Christian M. Hedrich1, Jose´ C. Crispıń1, Thomas Rauen1, Christina Ioannidis1, Tomohiro Koga2, Noe Rodriguez Rodriguez1, Sokratis A. Apostolidis2, Vasileios C. Kyttaris1 and George C. Tsokos1. 1BIDMC, Harvard Medical School, Boston, MA, 2Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, MA. Background/Purpose: CD3⫹CD4⫺CD8⫺ “double negative” T cells are expanded in the peripheral blood of patients with systemic lupus erythematosus (SLE) and lupus-prone mice. Double negative T cells infiltrate tissues, induce immunoglobulin production and secrete pro-inflammatory cytokines. Although double negative T cells have been claimed to derive from CD8⫹ T cells through down-regulation of CD8 surface co-receptors, the molecular mechanisms orchestrating this process remain unclear. The transcription factor cAMP responsive element modulator (CREM)␣, that is overexpressed in T cells from SLE patients, has been demonstrated to trans-regulate lupus-relevant genes and orchestrate epigenetic remodeling in mature T cells. Thus, CREM␣ is a promising candidate in the search for molecular mechanisms in T cell pathology in SLE. Methods: Primary human and murine T lymphocytes were isolated to assess epigenetic differences of the CD8 cluster in CD8⫹, CD4⫹, and double negative T cells using MeDIP and ChIP techniques. CD8 mRNA and protein expression was monitored in response to forced expression or knock-down of CREM␣. CREM␣-mediated effects on chromatin conformation were assessed using over-expression and knock-down techniques, followed by MeDIP or ChIP. Interactions between CREM␣ and epigenetic modifiers were established using Co-IPs and proximity ligation assays (PLA). Results: We link CREM␣ with transcriptional silencing of CD8A and CD8B in T cells from SLE patients and lupus prone MRL/lpr mice. CREM␣ trans-represses CD8 and mediates chromatin remodeling of the CD8 cluster through the recruitment of DNA methyltransferase (DNMT)3a and histone methyltransferase G9a.

Conclusion: We conclude that CREM␣ is essential for the expansion of double negative T cells in SLE and propose that CREM␣ may be utilized as disease biomarker and therapeutic target. An interaction between CREM␣ and G9a has been established applying proximity ligation assays. Ex vivo isolated CD8⫹ T cells exhibit interactions between CREM␣ and G9a that are enhanced after TCR-stimulation (120h). Nuclei are stained with DAPI, green signals represent CREM␣:G9a interactions. Disclosure: C. M. Hedrich, None; J. C. Crispıń, None; T. Rauen, None; C. Ioannidis, None; T. Koga, None; N. Rodriguez Rodriguez, None; S. A. Apostolidis, None; V. C. Kyttaris, None; G. C. Tsokos, None.

2725 Regulation Of Aberrant CD4 T Cell Activation By Suppressor Of Cytokine Signaling-1 (SOCS1) Mimetic Peptide, Has Relevance To Human Systemic Lupus Erythematosus. Tenisha Wilson, Cristina Armbruster, Simone Bedoya, Howard M. Johnson, Westley H. Reeves, Yi Li, Ying Yi Zheng, Laurence Morel and Joseph Larkin III. University of Florida, Gainesville, FL. Background/Purpose: Systemic lupus erythematosus (SLE) is an autoimmune disease, mediated by aberrantly activated CD4 T cells. Notably, suppressor of cytokine signaling-1 (SOCS1) regulates CD4 T cell activation. In addition, SOCS1⫹/⫺ (SOCS1 deficient) mice develop lupus-like disease. Although murine studies implicate a relationship between decreased SOCS1 expression and increased lupus pathology, the ability to translate these animal findings to human disease has not been established. Therefore, the goals of this study were to (a) establish clinical relevance for SOCS1 deficiency in regard to human SLE and (b) establish the mechanism by which SOCS1 deficiency can lead to lupus development. Methods: Human: qPCR was performed on PBMCs isolated from SLE patients (n⫽20) and healthy controls (HCs) (n⫽8) to assess SOCS1 expression, relative to GAPDH. SOCS1 expression was correlated with indicators of disease activity (monocyte CD64 MFI and C3 levels). Mouse: SOCS1⫹/⫺ mice were studied prior to the onset of lupus disease. C57BL/6 mice served as wild-type (WT) controls. Naı¨ve CD4⫹CD25⫺ T cells, isolated from spleen and LNs, were stimulated with anti-CD3 and anti-CD28 antibodies or only anti-CD3 antibody, with or without SOCS-1 KIR (mimetic peptide of SOCS1). Activation was assessed by flow cytometric analysis of CD25 and CD69. CFSE staining and 3H Thymidine incorporation were employed to measure proliferation. Akt activation was measured by western blotting. For statistical comparisons between SLE patients and HCs and between SOCS1⫹/⫺ and WT, the Mann-Whitney U test was performed. Spearman’s correlation was used for correlation data. Results: Similar to murine models of disease, SLE patients’ PBMCs had reduced expression of SOCS1 in comparison to HCs. Additionally, SOCS1 expression was negatively correlated with monocyte CD64 expression and positively correlated with C3 levels. Mechanistic data revealed that stimulating SOCS1⫹/⫺ naı¨ve CD4 T cells with anti-CD3 antibody alone resulted in an elevated population of CD69 and CD25 expressing cells. In parallel with the elevated population of activated cells under this suboptimal T cell activation condition, these SOCS1 deficient cells also displayed enhanced 3H thymidine incorporation and an increased CFSElow population - indicative of increased proliferation. The enhanced activation and proliferation was not due to an elevated population of memory T cells, but rather to a reduced requirement for Akt activation. Notably, SOCS1-KIR reduced the abnormal activation and proliferation. Conclusion: SOCS1 deficiency is clinically relevant to human SLE. We also determined that SOCS1 deficient CD4 T cells do not require CD28costimulation in order to undergo activation and proliferation. This has significant implications for lupus development, as SOCS1 deficient autoreactive T cells that have a reduced threshold for activation can activate

S1163

Tuesday, October 29

WT tricongenic animals. Exploratory studies on the role of PRKR in dendritic cell differentiation were performed using bone marrow cells from the original PRKR-KO mice versus Balb/c controls. The roles of 2-aminopurine and 6-mercaptopurine on IFN-induced gene expression were evaluated in TLR3transfected HEK293 cells. Results: While dsDNA antibody (Ab) titers and proteinuria significantly increase over time in PBS-treated (NZB ⫻ NZW)F1 mice, this is not the case in 2-aminopurine treated animals. 2-aminopurine therapy also results in a significant improvement of glomerulonephritis activity scores upon histological examination of the kidneys. Our first observations in PRKR-KO versus –WT SLE-prone mice indicate that dsDNA Ab titers display a significant increase from month 2 to month 10 in PRKR-WT B6.Sle1.Sle2.Sle3 mice, and this is not observed in their PRKR-KO littermates. IL-4 and GM-CSF exposure of bone-marrow cells results in the generation of myeloid-derived dendritic cells (moDC). We observed that the percentage of bone-marrow-derived CD11c-positive cells was significantly decreased in PRKR-KO mice, compared to Balb/c age- and gender-matched animals. Similarly, CD86-positive CD11c cells were significantly less numerous in PRKR-KO compared to Balb/c animals. Using 2-aminopurine and PRKR siRNA, we demonstrated that PRKR is involved in TLR3 signal transduction in TLR3-transfected HEK293 cells. Thus, poly I:C-induced expression of IFN␤ and IFN-induced genes is abrogated in these cells upon PRKR blockade. Intriguingly, we found that exposure of these cells to 6 mercaptopurine resulted in the same inhibitory effects on IFN␤ and IFN-induced gene expressio. Conclusion: Our data indicate that PRKR, an intracytoplasmic dangerrecognition molecule that has the ability to stimulate the expression of IFN-induced genes, is involved in the pathogenesis of SLE. The role of PRKR in the pathogenesis of the disease might be mediated by an increased differentiation and maturation of moDC, a hypothesis that we will verify in our PRKR-KO versus –WT SLE prone animals. Preliminary data indicate that the activity of PRKR is modulated by 6-mercaptopurine, the active metabolite of azathioprine.

autoreactive B cells, which can perpetuate disease. Significantly, SOCS1-KIR has the capacity to restore the proper T cell activation threshold. Based on this data, we propose that SOCS1-KIR should begin to be explored as a potential candidate for the prevention/treatment of SLE. Disclosure: T. Wilson, None; C. Armbruster, None; S. Bedoya, None; H. M. Johnson, None; W. H. Reeves, None; Y. Li, None; Y. Y. Zheng, None; L. Morel, None; J. Larkin III, None.

2726

Tuesday, October 29

A Polymorphism In The Mif Gene Is Associated With Cardiovascular Morbidity In Systemic Lupus Erythematosus – a Pilot Study. Eric F. Morand, Kathryn Connelly and Alberta Y. Hoi. Monash University, Melbourne, Australia. Background/Purpose: Chronic inflammation is believed to be responsible for accelerated atherosclerotic cardiovascular disease (CVD) in patients with SLE. Macrophage migration inhibitory factor (MIF) is a pro-inflammatory molecule implicated in the etiology of both SLE and atherosclerosis via amplification of macrophage recruitment and activation. A single nucleotide polymorphism (SNP) in the MIF promoter regulates MIF expression and is associated with SLE risk, and an anti-MIF therapy is currently in Phase I trials in SLE. The association of MIF polymorphisms with CVD risk in SLE is unknown. Methods: SLE patients (ACR criteria) attending a single centre were recruited fo a pilot study. Patients seen between 2007–2012 had disease activity (SLEDAI-2k) recorded at each visit, and organ damage (SLICCSDI) recorded at baseline and annually. CVD was defined using the cardiovascular domains of the SLICC-SDI criteria. Genomic DNA was isolated from whole blood, amplified by PCR, and MIF-173*C genotype determined by restriction fragment length polymorphism. Results: 153 SLE patients (80% female, median age 41y, disease duration 10y) were studied. The median (range) SDI and time-adjusted mean SLEDAI (AMS) were 1 (0–9) and 4 (0–22) respectively. The distribution of MIF-173 genotypes observed was G/G(69%) G/C(28%) and C/C(3%), thus a MIF-173C SNP was present in 31%. Organ damage (SDI) in the overall population correlated with disease duration (P⬍0.0001) and disease activity (AMS; P⬍0.0001). The major finding was that CVD was increased among patients carrying a MIF ⫺173*C allele. Among patients with SLE duration ⬎10y, CVD was 13-fold more frequent among patients carrying a MIF -173*C allele (odds ratio 12.75 (95%CI 2.2–75.2), p⫽0.0045). A significant association was also detected when stroke was included (OR⫽6.13, 95% CI 1.4–26.9, p⫽0.0204), and when AMI and/or stroke were analysed independently (OR⫽6.06, 95% CI 1.2–31.0, p⫽0.0375). There was no significant difference in traditional risk factors in relation to MIF -173*C allele, and no association detected with renal or CNS disease, corticosteroid use, or the frequency of episodes of persistently active disease. Conclusion: Although confirmation in larger cohorts is required, this pilot study suggests that carriage of the MIF-173C SNP is strongly associated with the risk of CVD in SLE. This suggests that anti-MIF therapy could alleviate SLE CVD risk in SLE. Disclosure: E. F. Morand, None; K. Connelly, None; A. Y. Hoi, None.

2727 Immunologic and Inflammatory Markers Of Impending Disease Flare In Systemic Lupus Erythematosus Patients Not Taking Immunosuppressive Medications In The Biomarkers Of Lupus Disease (BOLD) Study. Joel M. Guthridge1, Mikhail G. Dozmorov1, Melissa E. Munroe1, Krista M. Bean1, Sudhakar T. Sridharan2, Joan T. Merrill1 and Judith A. James3. 1Oklahoma Medical Research Foundation, Oklahoma City, OK, 2Pfizer Inc, Collegeville, PA, 3Oklahoma Medical Research Foundation and University of Oklahoma Health Sciences Center, Oklahoma City, OK. Background/Purpose: Systemic lupus erythematosus (SLE) is a clinically heterogeneous disorder characterized by a waxing and waning clinical course. Predictors of clinical disease flare have been difficult to identify, leading to cumulative damage, therapeutic toxicities and difficulties designing clinical trials. The goal of this study is to identify immunologic associations and pathogenic mechanisms of disease flares in SLE patients no longer taking immunosuppressive medications.

Methods: As part of the Biomarkers of Lupus Disease (BOLD) study, 41 SLE patients with moderately severe, but not organ-threatening disease activity were enrolled, background immunosuppressants (IS) stopped, intramuscular steroids given until disease suppression and then serially followed until clinical disease flare. SLE patients met ⬎ 4 ACR SLE classification criteria and had SLEDAI ⬎ 6 or BILAG ⬎/⫽ 2 B or 1 A scores at baseline. Peripheral blood specimens were collected at baseline, time of disease suppression and serially until they developed a significant enough flare to require new treatment (and minimum ⬎/⫽ 1 new BILAG B or SLEDAI increase of ⬎/⫽ 4 points). In addition to autoantibody levels and extensive immunophenotyping, 52 soluble inflammatory mediators, including cytokines, chemokines, and soluble receptors, using either xMAP multiplex technology or sandwich ELISA (BLyS and APRIL), were measured. Gene expression profiling, with globin depletion, was performed on a subset of baseline samples from SLE patients. Results: Forty of 41 SLE patients flared within 24 weeks, with 21 patients flaring within 60 days (early) and 13 flaring greater than 90 days (late) after stopping background IS. Patients who flared early were more likely to be of African-American descent, while Native American patients were more likely to flare late. Compared to late flaring patients, Caucasian SLE patients who flared early overexpressed 69 genes (including MMP9, CD11b (ITGAM), and MYH9) and showed decreased expression of 20 genes, including HLA-DRB5 and IFI6 at baseline. SLE patients who flared early also had higher baseline expression levels of CD11b on neutrophils (p⫽0.003) and monocytes (p⫽0.03) and higher CD86 expression on B cells (p⫽0.03) (all races combined). SLE patients who flared more than 90 days after stopping medications had higher baseline plasma levels of IL-1RA (p⫽0.03) and TNFRI (p⫽0.04), as well as higher levels of BLyS (p⫽0.01), IL-7 (p⫽0.03) and IFNg (p⫽0.04) at the flare visit. Conclusion: SLE patients who will flare earlier after withdrawal of ineffective immunosuppressants and transient steroid treatment have increased levels of activated B cells and increased expression of CD11b at baseline, while SLE patients who flared later have evidence of regulatory pathway engagement through secretion of IL1RA. Disclosure: J. M. Guthridge, None; M. G. Dozmorov, None; M. E. Munroe, None; K. M. Bean, None; S. T. Sridharan, Pfizer Inc, 3; J. T. Merrill, Pfizer Inc, 5; J. A. James, Pfizer Inc, 5.

ACR Concurrent Abstract Session Biology and Pathology of Bone and Joint II: Osteoclast Biology and Arthritis Tuesday, October 29, 2013, 4:30 PM–6:00 PM

2728 The Small Ubiquitin Related Modifier-1 (SUMO-1) Regulates Osteoclast Differentiation In Vitro and In Vivo. Svetlana Frank1, Daniel Umlauf2, Olli A. Ja¨nne3 and Thomas Pap4. 1University Hospital Mu¨nster, Muenster, Germany, 2University Hospital Muenster, Muenster, Germany, 3 Univ Helsinki, Inst Biomed Physiol, Biomedicum, Helsinki, Finland, 4 University Hospital Mu¨nster, Mu¨nster, Germany. Background/Purpose: Posttranslational modification of proteins by SUMO has been shown for a number of target molecules including transcription factors and is involved in a variety of cellular processes, including protein localization, transcriptional regulation, protein stability, cell survival and death. Previously, we have shown that the increased expression of SUMO-1 contributes to the inflammatory response in RA. Here, we investigated the role of SUMO-1 in osteoclastogenesis and studied the skeletal phenotype of SUMO-1⫺/⫺ mice. Methods: The skeletal phenotype of 8-week old SUMO-1⫺/⫺ and wild type mice was investigated by ␮CT-analysis of trabecular bone in the lumbar spine and femora. L5 vertebral bodies from these mice mice were embedded into methylmetacrylate and analyzed using van Kossa and tartrate-resistant acid phosphatase (TRAP) staining. For in vitro experiments, bone marrow macrophages (BMMs) were isolated from SUMO1⫺/⫺ mice and wild-type controls. The cells were differntiated into osteoclasts in the presence of macrophage colony-stimulating factor and

S1164

receptor activator of nuclear factor k-B ligand (NF-kB). Osteoclast differentiation was characterized by staining for TRAP. Using PCR, the expression levels of DC-STAMP, Cathepsin K and Integrin ␤3 were analyzed. Proliferation and cell viability of BMMs was determined using CyQuant proliferation assay and MMT test. Osteoclast resorption capacity was analyzed using a calcium phosphate bone resorption assay. Results: 8-weeks old SUMO-1⫺/⫺ mice had a 20% higher trabecular bone volume fraction compared with wt mice. Moreover, trabecular thickness was higher and trabecular separation was lower in SUMO-1⫺/⫺ mice. In addition, histological analyses revealed a significantly reduced number of osteoclasts in SUMO-1⫺/⫺ mice in vivo. The loss of SUMO-1 was associated with impaired osteoclast differentiation and with impaired bone resorption capacity in vitro. In PCR analysis, we found a decreased expression of DC-STAMP, Cathepsin K and Integrin ␤3 in osteoclasts differentiated from SUMO-1⫺/⫺ compared to wt mice. Proliferation and cell viability of BMMs were not affected by loss of SUMO-1. Using western blot analysis we found no differences in activation of MAPK p38 and p44/42 as well as in activation of NF-kB signaling in BMMs and osteoclasts in SUMO-1⫺/⫺ and wt mice. Conclusion: In our study, we found that SUMO-1⫺/⫺ mice have high bone mass owing to a decrease in number and function of osteoclasts. These finding are most likely due to decreased expression of osteoclast markers contributing to osteoclast fusion and to osteoclast resorption capacity. These data suggest that SUMO-1 is involved predominantly in the regulation of bone mass by osteoclast formation and activity, and therefore may be an interesting target for treating diseases associated with bone loss. Disclosure: S. Frank, None; D. Umlauf, None; O. A. Ja¨nne, None; T. Pap, UCB, Servier, Abbott, Bioiberica, MSD, Pfizer, Eli Lilly, 5, UCB, Servier, Abbott, Bioiberica, MSD, Pfizer, Eli Lilly, 8.

2729

Background/Purpose: In osteoarthritis (OA) it is evident that a subset of patients experiences chronic tissue inflammation and may benefit from anti-inflammatory treatment. The systemic inflammation marker C-reactive protein (CRP) has shown limited use in predicting disease severity, progression or response to anti-inflammatory treatment. The aim of this study was to investigate whether OA patients can be segregated into groups dependent on the present or absence of systemic and/or chronic tissue inflammation and whether such groups have different levels of MMP driven tissue destruction. Methods: A cross-sectional study of 342 patients: 281 patients had symptomatic knee OA and no planned total knee replacement (TKR) and 61 who underwent TKR. Serological biomarkers were measured by ELISA: high sensitive CRP (hsCRP), CRP degradation fragment (CRPM, chronic tissue inflammation), MMP-mediated degradation fragments of type I, II and III collagen; C1M (connective tissue), C2M (cartilage) and C3M (synovium). The associations between biomarkers and OA stage were investigated: Kellgren & Lawrence (KL) 0, Mild OA (n⫽12); KL1–2, Moderate OA (n⫽202); KL3–4, Severe OA (n⫽57); and KL3–4 with TKR (n⫽60). Cut-off values of CRPM and hsCRP were set as 12ng/mL and 5␮g/mL (mean⫹2SD of controls). Patients were divided in quartiles (Q, fig) based on the cut-off values. Reference values of the biomarkers were recorded for healthy controls. Data are shown as mean [95%-CI]. Results: hsCRP was only elevated with TKR (5.9 [3.6–8.2] ␮g/mL) compared to controls. In contrast, the mean levels of the CRPM were twice as high in the OA groups (10–14ng/mL) compared to controls (5ng/mL). C1M and C2M were significantly elevated in the TRKs compared to Moderate (p⬍0.001) and Severe (p⬍0.01). There was no difference between patient and controls in C3M. Patients in Q4 (fig) had significantly higher KL compared to patients in Q1 (p⬍0.0001), Q2 (P⫽0.017) and Q3 (p⬍0.0001). C1M, C2M and C3M were lower in Q1 compared to all other quartiles. Comparing Q2 with Q3 showed that C1M was higher (p⫽0.0005) in Q3, but C3M was lower (p⫽0.019). Thus, the populations identified by systemic and chronic tissue inflammation have different structural integrity.

Disclosure: A. S. Siebuhr, Nordic Bisocience, 3; K. K. Petersen, None; L. ArendtNielsen, None; L. Egsgaard, None; T. N. Eskehave, None; O. Simonsen, None; C. Christiansen, Nordic Bioscience Diagnostic, 1, Nordic Bioscience Diagnostic, 6; H. C. Hoeck, None; M. A. Karsdal, Nordic Biosciece, 1, Nordic Bioscience Diagnostic, 3; A. C. Bay-Jensen, Nordic Bioscience Diagnostic, 3.

2730 Netrin1 Is a Critical Autocrine Factor For Osteoclast Differentiation. Aranzazu Mediero1, Bhama Ramkhelawon1, Kathryn Moore1, P. Edward Purdue2, Steven R. Goldring2 and Bruce N. Cronstein3. 1NYU School of Medicine, New York, NY, 2Hospital for Special Surgery, New York, NY, 3 NYU School of Medicine, Division of Rheumatology, New York, NY. Background/Purpose: Netrins have been extensively studied for their role in axonal guidance during neural development. In addition, netrins are chemopulsants for a variety of non-neuronal cell types via binding to their receptors Unc5b and DCC. Although thought to suppress inflammation in several settings, netrin1, acting via Unc5b, inhibits macrophage migration directed by chemokines CCL2 and CCL19 to promote macrophage retention in and exacerbation of atherosclerotic plaque. We asked whether Netrin1 was expressed during osteoclast (OC) differentiation and whether it plays a role in OC differentiation. Methods: DEXAscan and MicroCT analysis were performed on Netrin1 deficient mice (radiation chimeras) and wildtype (WT, radiation chimeras) littermates. OC differentiation was studied as M-CSF/RANKL-stimulated differentiation of murine bone marrow precursors to TRAP⫹/multinucleated cells, in the presence/absence of recombinant Netrin1 and Unc5b antibody. Netrin1, Unc5b and DCC expression were studied by RT-PCR and Western Blot in primary bone marrow-derived osteoclasts. Netrin1 immunostaining was performed in human tissue obtained following primary prosthesis implantation or after prosthesis revision. Results: During OC differentiation cell-associated Netrin1 and Unc5b (but not DCC) protein expression increased by 30⫾2% and 98⫾4% respectively (p⬍0.001,n⫽4) and Netrin1 secretion increased by 66⫾2% (p⬍0.001, n⫽4). Consistently, RANKL stimulates an increase in Netrin1 and Unc5b mRNA expression during OC differentiation (25⫾4 and 3⫾0.5 fold change respectively p⬍0.001, n⫽4). Moreover, in Netrin1-deficient marrow precursors OC differentiation was diminished by 65⫾2% as compared to control (p⬍0.001, n⫽6), an effect reversed by addition of recombinant netrin1 to cultures (121⫾5% increased, p⬍0.5, n⫽4). An antibody to the netrin1 receptor Unc5b reduces OC formation by 57⫾6% (p⬍0.001, n⫽6) whereas an antibody to DCC had no effect on OC formation (5⫾4% reduction, p⫽NS

S1165

Tuesday, October 29

Identification Of Osteoarthritis Patients With Chronic Inflammation Driven Disease Progression. Anne Sofie Siebuhr1, Kristian Kjaer Petersen2, Lars Arendt-Nielsen2, Line Egsgaard2, Thomas Navndrup Eskehave3, Ole Simonsen4, Claus Christiansen1, Hans Christian Hoeck3, Morten Asser Karsdal1 and Anne C. Bay-Jensen1. 1Nordic Bioscience, Biomarkers and Research, Herlev, Denmark, 2Center for Sensory-Motor Interaction, Aalborg, Denmark, 3Center for Clinical and Basic Research and C4Pain, Aalborg, Denmark, 4Frederikshavn Hospital, Frederikshavn, Denmark.

Conclusion: All OA patients had surprisingly high levels of chronic tissue inflammation. OA patients could be divided into quartiles or 2 separate groups: i) those who may benefit from anti-inflammatory treatment (Q3, Q4) and ii) those eligible for a more tissue centric treatment (Q1, Q2). Patients with high CRPM (Q2 and Q4) had higher levels of the tissue degradation markers C1M, C2M and C3M suggesting that they had elevated tissue turnover. In alignment, those OA patients undergoing TKR had even higher levels of tissue turnover markers, suggesting a distinct TKR serological phenotype. Clearly different types of OA with different levels and types of tissue destruction could be identified by CRPM, but not CRP.

vs. control, n⫽6). Finally, DEXAscan and MicroCT analysis demonstrated an increase in bone mineral density (BMD), total volume (TV), bone volume (BV) and TV/BV in both cortical and trabecular bone in Netrin1 deficient mice when compared to WT (p⬍0.01 for all, n⫽5). Netrin1 immunostaining in human tissue biopsies reflect enhanced expression in tissue from implant revision when compared to primary implants. Conclusion: The chemorepulsant Netrin1 is required for osteoclast differentiation and stimulates OC differentiation by an autocrine mechanism. This finding suggests that Netrin1 may be a novel target to reduce OCmediated bone resorption and to prevent joint prosthesis loosening. Disclosure: A. Mediero, Filed a patent on use of adenosine A2AR agonists to prevent prosthesis loosening (pending)., 9; B. Ramkhelawon, None; K. Moore, None; P. E. Purdue, None; S. R. Goldring, Boehringer Ingelheim, 2, Pfizer Inc, Bone Therapeutics, Fidia Pharma, Inc, Abbott Laboratories, 5; B. N. Cronstein, Canfite Pharma, 1, NIH, Gilead, Takeda, AstraZeneca, 2, NYU School of Medicine, 3, Merck-SeronoBristolMyers Squibb, Novartis, CanFite Biopharmaceuticals, Cypress Laboratories, Regeneron (Westat, DSMB), Endocyte, Protalex, Allos, Inc., Savient, Gismo Therapeutics, Antares Pharmaceutical, Medivector, 5, Multiple patents on adenosine receptors and bone metabolism, pharmacology, 9.

2731

Tuesday, October 29

An Acetyl-Histone Mimetic Blocks Proinflammatory Activation Of Rheumatoid Arthritis Fibroblast-Like Synoviocytes. P. A. Kabala1, A.M. Grabiec1, C. Angiolilli1, Nicholas Smithers2, Jason Witherington2, Paul Peter Tak3, Rabinder Prinjha2 and Kris A. Reedquist1. 1Department of Clinical Immunology and Rheumatology Academic Medical Center, University of Amsterdam, Amsterdam, Netherlands, 2GlaxoSmithKline, Stevenage, United Kingdom, 3Academic Medical Center / University of Amsterdam, Department of Clinical Immunology and Rheumatology & GlaxoSmithKline, Amsterdam, Netherlands. Background/Purpose: Genetic backgrounds and environmental factors do not completely explain the etiology of disease and predisposition to rheumatoid arthritis (RA), and increasing attention has turned to the potential contributions of heritable epigenetic mechanisms such as DNA methylation, post-translational modifications of histones, and expression of non-coding microRNAs. Fibroblast-like synoviocytes (FLS) isolated from affected joints of RA patients maintain an aggressive phenotype in vitro, indicating potential alteration of epigenetic regulatory processes, and changes in global and gene-specific promoter DNA methylation and histone modification are observed in RA FLS. I-BET compounds, acetyl histone mimetics which interfere with reading of acetylated histones by BET family bromodomain proteins BRD2–4, prevent LPS-induced inflammatory gene expression in murine bone marrow-derived macrophages in vitro, as well as in vivo models of endotoxic shock and bacterial sepsis. This study was undertaken to assess the potential of I-BET to influence the inflammatory activation of RA FLS. Methods: RA fibroblast-like synoviocytes (FLS) were treated with IL-1␤ or TNF in the presence or absence of increasing concentrations of I-BET, and IL-6 and IL-8 production measured by ELISA. Cellular viability was assessed using MTT assay and cell death ELISA kits. Activation of intracellular signaling pathways was examined by immunoblotting. Total RNA was extracted and mRNA expression of genes regulated by IL-1␤ or TNF in RA FLS was analyzed using a low density quantitative PCR custom array. Results: BRD2–4 mRNA expression was readily detected in RA FLS. I-BET reduced RA FLS (n⫽6) production of IL-6 protein in response to IL-1␤ (1 ␮M, 70% reduction, P ⬍ 0.001) and TNF (50%, P ⬍ 0.001). IL-8 production in response to IL-1␤ (⬎75% reduction, P ⬍. 0.001) and TNF (⬎70%, P ⬍ 0.001) was also inhibited. Similar effects were observed on IL-6 and IL-8 mRNA expression. Inhibition of IL-6 and IL-8 production was maintained when I-BET treatment was delayed 1–4 hours post-stimulation. I-BET had no effect on cell viability or survival, and failed to modulate IL-1␤ or TNF–induced activation of MAPK or NF-␬B signaling pathways. Low density qPCR array analysis of 26 genes induced by IL-1␤ in RA FLS (n⫽3) demonstrated that I-BET reduced induction of 17 genes by ⬎50%, including TNF (⬎80% suppression), MMP-1 (⬎90%), MMP-3 (90%), SELE (⬎80%), VCAM-1 (⬎60%), CXCL-6 (⬎75%), CXCL-9 (⬎95%) and CXCL-11 (⬎90%). Conclusion: Our results demonstrate that disrupting the recruitment of BET family proteins to acetylated histones efficiently blocks RA FLS production of inflammatory mediators, including cytokines, chemokines and MMPs, in response to IL-1␤ and TNF. This study provides initial evidence that the development of synthethic compounds targeting interactions between epigenetic modifications, such as histone acetylation, and

the proteins which interpret these modifications may have therapeutic potential in the treatment of RA. Disclosure: P. A. Kabala, None; A. M. Grabiec, None; C. Angiolilli, None; N. Smithers, GlaxoSmithKline, 3; J. Witherington, GlaxoSmithKline, 3; P. P. Tak, GlaxoSmithKline, 3; R. Prinjha, GlaxoSmithKline, 3; K. A. Reedquist, GlaxoSmithKline, 2.

2732 Inhibition Of Autophagy Prevents Ovariectomy Induced Bone Loss. Neng-Yu Lin1, Joerg H. W. Distler1, Alfiya Distler1, Christian Beyer1, Georg Schett2 and Oliver Distler3. 1Department of Internal Medicine III and Institute for Clinical Immunology, University of Erlangen-Nuremberg, Erlangen, Germany, 2University of Erlangen-Nuremberg, Erlangen, Germany, 3 Department of Rheumatology, University Hospital Zurich, Zurich, Switzerland. Background/Purpose: Autophagy is an essential, homeostatic process by which cells digest unnecessary or damaged organelles. Atg7 is essentially required for formation of the autophagosome. The DMARD chloroquine (CQ) has been identified to arrest autophagy function by blocking the fusion between autophagosomes and lysosomes. Accumulating evidence demonstrates that autophagy is involved in the pathology of varies diseases including infections, cancer and neurodegeneration and recent reports also linked autophagy to osteoclastogenesis and rheumatoid arthritis. Methods: Systemic bone loss was induced by ovariectomy (OVX). Autophagy was modulated by genetic as well as pharmacologic approaches. For genetic inhibition of autophagy, we crossbred Atg7fl/fl mice with LysM Cre⫹ mice to generate Atg7fl/fl ⫻ LysM Cre⫹ mice with selective inactivation of autophagy function in the monocyte lineage. For pharmacological inhibition, eight-week-old mice underwent daily intraperitoneal injection of 20mg kg⫺1 CQ for a total of 50 days. We applied Microcomputed tomography to analyze bone density in vivo. TRAP staining and bone histomorphometry were used to confirm osteoclast number in tissue. To assess osteoclast formation and activity in vitro, Atg7fl/fl ⫻ LysM Cre⫹ BMCs and CQ treated BMCs were cultured on bone slices and analysed by the TRAP staining and Toluidine blue staining (n⫽6, for each group). To assess the autophagy effect on osteoblast activity in vitro, primary osteoblasts were treated with Cre Adenovirus and CQ to analyse ALP staining and measure osteoblast markers by RT-PCR. Results: To investigate the effect of CQ on OVX-induced systemic bone loss, we analyzed the bone density (bone volume/trabecular volume; Bv/Tv) and the trabecular number (Tb. N.) at the proximal tibias. Bv/Tv and Tb. N. were strongly reduced in sham-treated OVX mice. The decreases in Bv/Tv and in Tb. N. were reduced by 76 % and 60% in CQ treated OVX mice (n⫽6, p⬍0.05 for both). Inhibition of autophagy by selective knockdown of Atg7 in monocytic cells (Atg7fl/fl ⫻ LysM Cre-mice) also prevented OVX-induced osteoporosis and prevented OVX-induced decreases in Bv/Tv and in Tb. N.. Histomorphometric analyses demonstrated decreased osteoclast counts in CQ-treated mice and Atg7fl/fl ⫻ LysM Cre-mice, respectively, compared to controls. Consistent with these in vivo results, inhibition of autophagy, either by treatment with CQ or by knockdown of Atg7, also prevented osteoclast differentiation and osteoclast mediated bone resorption in vitro. In contrast to the effects on osteoclasts, inhibition of autophagy by genetic or pharmacologic approaches did not inhibit osteoblast activity. Conclusion: We demonstrate that inhibition of autophagy by genetic or by pharmacological approaches significantly reduced OVX-induced osteoporosis in mice. These findings identify the autophagy machinery as a potential target for the treatment of osteoporoses. Considering the potent anti-osteoporotic effects and the availability of approved inhibitors such as CQ, these findings may have direct translational implications. Disclosure: N. Y. Lin, None; J. H. W. Distler, None; A. Distler, None; C. Beyer, None; G. Schett, Celgene, 2, Abbott Laboratories, UCB, Roche, 5; O. Distler, Sanofi, Active Biotech, Pfizer, Actelion, and Novartis, 2, Actelion, Sanofi-Aventis, Roche, Genentech, Active Biotec, medac, Novartis, Sinoxa, 4D Science, 5.

2733 Spontaneous Osteoarthritis In Mice By Genetic Deletion Of Nfatc1 and Nfatc2. Susan Y. Ritter1, Matthew B. Greenblatt1, Kelly Tsang1, Dorothy Z. Hu2, John Wright1 and Antonios O. Aliprantis1. 1Brigham and Women’s Hospital, Boston, MA, 2Massachusetts General Hospital, Boston, MA. Background/Purpose: Osteoarthritis (OA) was once viewed as a mechanical disease of “wear and tear”, but recent advances suggest that OA results from active dysregulation of chondrocyte biology leading to catabo-

S1166

Disclosure: S. Y. Ritter, None; M. B. Greenblatt, None; K. Tsang, None; D. Z. Hu, None; J. Wright, None; A. O. Aliprantis, Nutech Medical, 5.

ACR Concurrent Abstract Session Genetics and Genomics of Rheumatic Disease II Tuesday, October 29, 2013, 4:30 PM–6:00 PM

2734 Epigenetic Control Of a Gene Regulatory Network Associated With Experimental Autoimmune Arthritis and Rheumatoid Arthritis. Andras Vida1, Janos Gal2, Gyorgyi Soos2, Attila Balog3, Laszlo G. Puskas4, Katalin Mikecz1, Tibor T. Glant1 and Tibor A. Rauch1. 1Rush University Medical Center, Chicago, IL, 2Bacs-Kiskun County Hospital, Kecskemet, Hungary, 3 Albert Szent-Gyorgyi University, Szeged, Hungary, 4Biological Research Center, Szeged, Hungary. Background/Purpose: Rheumatoid arthritis (RA) is an autoimmune inflammatory disorder that mainly affects the synovial joints, leading to destruction of articular cartilage and bone. There is accumulating evidence for the involvement of epigenetic events in the pathogenesis of RA. Recent genome-wide DNA methylation studies have reported disease-associated changes in DNA methylation profile. However, methylation status-dependent molecular mechanisms that can potentially drive or amplify inflammatory processes in arthritis have not been explored. Methods: We separated lymphocytes from mice with proteoglycaninduced arthritis (PGIA), an autoimmune model of RA, and surveyed arthritis-induced DNA methylation events at genome-wide level using a Methylated CpG Island Recovery Assay (MIRA)-chip method. In addition, we isolated RNA from lymphocytes of arthritic mice and from peripheral blood mononuclear cells (PBMC) of RA patients, and monitored the DNA

methylation-dependent changes in gene expression levels by quantitative RTPCR. Disease-associated DNA-protein interactions were analyzed using electrophoretic mobility shift assays. A cell culture-based in vitro transient expression system was employed to investigate the functional consequence of the binding of specific transcription factors to gene promoters. Results: Using the MIRA-chip method, we detected arthritis-associated epigenetic changes that preferentially occurred in B cells from mice with PGIA. Specifically, hypomethylation of the promoter of the gene encoding the “zinc finger and BTB domain containing 38” (ZBTB38) transcription factor resulted in a sharp upregulation of ZBTB38 expression in lymphocytes of arthritic mice relative to non-arthritic animals. Expression of the ZBTB38 gene was also elevated in PBMC of RA patients as compared to its expression in healthy subjects. Moreover, ZBTB38 could facilitate its own transcription, and stimulate the expression of a number of other genes including those encoding transcription factors, pro-inflammatory cytokines, as well as receptors for chemokines and Toll-like receptors. Intriguingly, many of the ZBTB38-targeted genes, identified in this study, have been implicated in RA pathogenesis. Conclusion: ZBTB38 is involved in an intricate gene regulatory network that seems to control inflammatory processes in RA. Exploration of epigenetic mechanisms involved in the regulation of a major RA-associated inflammatory pathway may lead to identification of new genes as novel biomarkers and potential drug targets for RA. Disclosure: A. Vida, None; J. Gal, None; G. Soos, None; A. Balog, None; L. G. Puskas, None; K. Mikecz, None; T. T. Glant, None; T. A. Rauch, None.

2735 Integrative Analysis Of Multiple Omics Technologies Reveals a Novel Therapeutic Target For Rheumatoid Arthritis (RA). David L. Boyle1, John Whitaker1, Beatrix Bartok1, Wei Wang2 and Gary S. Firestein1. 1UCSD School of Medicine, La Jolla, CA, 2UCSD, La Jolla, CA. Background/Purpose: Identifying novel therapeutic targets for immunemediated disease has become increasingly challenging. We used an unbiased method to integrate multiple “omics” methodologies and evaluate the overlap between RA-associated genes identified in genome-wide association studies (GWAS), genes that are differentially expressed or genes that are differentially methylated in RA fibroblast-like synoviocytes (FLS). This analysis narrowed the search to a limited number of genes and allowed us to focus on Engulfment and Cell Motility Protein-1 (ELMO1), a gene not previously considered as a therapeutic target. We show how this novel candidate gene plays a key role in the pathogenic behavior of RA FLS. Methods: Data were from three types of genome-wide RA assays were integrated: (i) for sequence variation we used the NCBI GWAS database and extracted all gene that were mapped to SNPs that had been implicated in RA susceptibility (www.genome.gov/gwasstudies/); (ii) for gene expression we used public microarray datasets of RA, OA, and normal (NL) FLS (Gene Expression Omnibus Database); (iii) for DNA methylation we used a set of differentially methylated genes that we previously identified in RA, OA, and NL FLS (Genome Medicine 5:40 2013). For functional studies, passage 4 to 6 RA FLS were cultured in medium or PDGF and migration was measured using a scratch assay. Cell invasion was determined by measuring cell invasion into Matrigel-coated transwell plates. Rac1 activation was determined by measuring the Rac1-GTP bound form of Rac1 by Western blot. Gene knockdown was performed using the siRNA and Amaxa technology. Results: The integrative analysis of the three unbiased genome-wide approaches identified 6 genes as different in RA compared to controls (OA and normal) in all datasets. Of the overlapping genes, we were particularly interested in the cytoplasmic engulfment protein, ELMO1, as it was differentially methylated in the promoter region, differentially expressed, and associated with RA due to an intron-6 polymorphism. This protein can bind to Rac1 in activated cells and potentially alter cell movement. We first confirmed that ELMO1 is expressed in RA FLS and synovium as determined by qPCR. siRNA knockdown was then performed, which decreased ELMO1 expression by ⬎90% compared to control siRNA. ELMO1 deficiency decreased PDFG-induced FLS migration by 40% (p⬍0.02). In an invasion assay, ELMO1 siRNA markedly decreased invasion of PDGF stimulated RA FLS into Matrigel matrix compared with control siRNA. The mechanism ELMO1 function was determined by determining the effect of ELMO1 deficiency on Rac1 activation. These experiments showed that ELMO1 siRNA decreased peak PDGF-induced Rac1 activation by 80%.

S1167

Tuesday, October 29

lism of the cartilage matrix. Little is understood regarding the transcription factors that regulate this catabolic program and ultimately the development of OA. Nuclear factor of activated T cells (NFATs) are a family of transcription factors with broad roles in vertebrate physiology including immune responses and tissue development. Since NFAT family members often display redundant functions and NFATc2-deficient mice were previously shown to develop OA at older ages, we tested the role of NFATc1 in OA. Methods: Mice bearing two copies of a floxed allele of Nfatc1 (Nfatc1fl/fl) were crossed to Col2-cre mice to delete the gene in cartilage (Nfatc1col2 mice). Nfatc1col2 mice were analyzed in the destabilization of the medial meniscus (DMM) model of OA. The Nfatc1col2 strain was bred onto the NFATc2-deficient background to generate Nfatc1col2Nfatc2⫺/⫺ mice. These mice were analyzed by microCT, immunohistochemistry and quantitative real-time PCR (qPCR) of mRNA isolated from dissected cartilage tissue. mRNA purified from human cartilage tissue from joint replacement surgery was also interrogated for NFATC1 and NFATC2 expression by qPCR. Results: NFATc1col2 mice were not more susceptible to OA in the DMM model than littermate controls, suggesting that in mice, NFATc1 deficiency is not sufficient to accelerate OA. In contrast, cartilage-specific ablation of Nfatc1 in Nfatc2⫺/⫺ mice led to a spontaneous, early onset, aggressive OA. The arthritis affects multiple joints including the knees, elbows and tarsal-metatarsal joints and is associated with subluxations. Histologically, joints from Nfatc1col2Nfatc2⫺/⫺ mice showed loss of proteoglycans, increases in collagen and aggrecan degradation products and eventual progression to cartilage effacement. MicroCT demonstrated increased osteophyte formation and changes to the subchondral bone in Nfatc1col2Nfatc2⫺/⫺ mice. Compared with littermate controls, cartilage dissected from the knee joints of Nfatc1col2Nfatc2⫺/⫺ mice showed increased expression of: Mmp13, Adamts5, Col2a1 and Htra1. Lastly, NFATC1 expression is downregulated in paired lesional versus macroscopically normal cartilage samples from OA patients (1.63 vs. 2.72, p⫽0.0065 by paired t-test). No significant change was observed in NFATC2 expression (2.44 vs. 2.06, p⫽0.214). Conclusion: NFATs are suppressors of OA and regulating NFATs or their transcriptional targets in chondrocytes may lead to novel disease modifying OA therapies. The Nfatc1col2Nfatc2⫺/⫺ mice have a highly penetrant, early onset and severe OA phenotype, which could be an attractive platform for the preclinical development of treatments to alter the course of this disease.

Conclusion: Integrative analysis of multiple unbiased genome-wide datasets is a novel method to identify potential therapeutic targets. One previously unanticipated target, ELMO1, emerged as a candidate gene in RA. ELMO1 plays a critical role as a regulator of FLS migration and matrix invasion by activating Rac1. These data show how one can use integrative studies to identify and validate novel therapeutic targets. Disclosure: D. L. Boyle, None; J. Whitaker, None; B. Bartok, None; W. Wang, None; G. S. Firestein, None.

2736

Tuesday, October 29

High Density Microarray Analysis Identifies Novel Differentially Expressed Long Noncoding RNAs In Rheumatoid Arthritis Synovial Fibroblasts. Mojca Frank Bertoncelj1, Michelle Trenkmann1, Christoph Kolling2, Beat A. Michel3, Renate E. Gay1 and Steffen Gay1. 1Center of Experimental Rheumatology, University Hospital Zurich and Zurich Center of Integrative Human Physiology (ZIHP), Zurich, Switzerland, 2Schultess Clinic, Zurich, Switzerland, 3Department of Rheumatology, University Hospital Zurich, Zurich, Switzerland. Background/Purpose: Long noncoding RNAs (lncRNAs) have emerged as key regulators of gene expression. Recently, several new susceptibility loci were identified for rheumatoid arthritis (RA), mapping to noncoding genomic regions (Eyre et al., Nat Gen 2012). Furthermore, we showed that the lncRNA HOTAIR is epigenetically repressed in RA synovial fibroblasts (SF), enhancing their activated phenotype. The aim of the present study was to determine the global lncRNA expression in RASF and to identify differentially expressed (DE) lncRNAs in RASF with a potential role in the pathogenesis of RA. Methods: Total RNA was isolated from cultured synovial fibroblasts (SF), passages 5–6, from 3 RA and 3 osteoarthritis (OA) age-matched female patients. Sample labeling and array hybridization were performed according to the Agilent One-Color Microarray-Based Gene Expression Analysis protocol. Fluorescently-labeled cRNA was hybridized to human LncRNA Array v3.0 (Arraystar) detecting 30,586 lncRNAs and 26,109 mRNAs. The arrays were scanned by the Agilent Scanner G2505C and analyzed by the Agilent Feature Extraction software. The Agilent GeneSpring GX v12.0 software was used for data processing. To identify DE transcripts the absolute fold change of normalized signal intensities between RASF and OASF was calculated (cut-off value 1.5, p⬍0.05). Gene ontology (GO) and pathway analysis were performed. Fourteen selected DE lncRNAs were confirmed by qPCR in a larger cohort of age- and gender-matched RASF and OASF (n⫽10 each) with normalization to GAPDH. Results: The microarray analysis identified 225 DE RNA transcripts in RASF compared to OASF. Among these, 36 lncRNAs (x-fold range: 1.5–4.1) and 87 mRNAs (1.5–10.7) were significantly down-regulated, while 64 lncRNAs (1.51–3.5) and 38 mRNAs (1.5–3.3) were significantly upregulated in RASF. Among identified lncRNAs, small nucleolar RNA host gene 1 (SNHG1) (dCt⫾SD OASF: 6.92⫾0.29; RASF: 6.53⫾0.45, p⫽0.038) and RP11–39708.4 (dCt⫾SD OASF: 12.61⫾0.26; RASF: 12.07⫾0.69, p⫽0.042) were confirmed by PCR to be upregulated in RASF. Although their function is not known, it is of interest that RP11–39708.4 is a natural antisense transcript in the locus of fibroblast growth factor 14 and SNHG1 is a host gene for several small nucleolar RNAs, including SNORD22 and SNORD25–31. Among DE mRNAs, several transcripts linked to the pathogenesis of RA were found, such as interleukin 8, peroxisome proliferator-activated receptor (PPAR) gamma and sirtuin 1. Additionally, several novel DE mRNAs were identified, such as tumor protein p73, cell division cycle-associated protein 3 and chemokine-like receptor 1. DE mRNAs clustered to p53, Wnt and PPAR signaling and cell cycle-regulating pathways with significant enrichment of GO terms including cell adhesion molecules, cadherin and beta-catenin binding as well as mitosis and nuclear division. Conclusion: In this study we identified novel DE lncRNAs in RASF, including SNHG1 and RP11–39708.4. This is the first study determining the global lncRNA profile in combination with differential expression of mRNAs in RASF, which will be a powerful tool to identify novel gene regulating pathways in RA. Disclosure: M. Frank Bertoncelj, IMI-BT Cure, IAR Epalinges, EURO-TEAM, 2; M. Trenkmann, EURO-TEAM, IMI BTCure, IAR Epalinges, KFSP USZ, 2; C. Kolling, None; B. A. Michel, None; R. E. Gay, EURO-TEAM, IMI BTCure, IAR Epalinges, 2; S. Gay, EURO-TEAM, IMI BTCure, IAR Epalinges, 2.

2737 In Vivo MiR-146a Administration Ameliorates Murine Lupus Nephritis. Dong Liang1, Shiyu Zhou2, Zheng Liu3, Zhengyuan Shan1, Philip Brohawn3, Yihong Yao3, Indu Raman4, Quan-Zhen Li4, John B. Harley1 and Nan Shen5. 1 Cincinnati Children’s Hospital Medical Center, Cincinnati, OH, 2Shanghai Institutes for Biological Sciences Chinese Academy of Sciences & Shanghai Jiao Tong University School of Medicine, Shanghai, China, 3MedImmune, LLC, Gaithersburg, MD, 4University of Texas Southwestern Medical Center, Dallas, TX, 5Shanghai Institute of Rheumatology, Renji Hospital, Shanghai Jiaotong University School of Medicine, Shanghai, China. Background/Purpose: New Zealand black and white F1 (NZBW/F1) is a classic mouse model of systemic lupus erythematosus (SLE). Type I interferon (IFN) infusion accelerates lupus proteinuria and nephritis. miR146a is a potent immune regulator to suppress inflammatory responses in T and dendritic cells. Potential involvement of this miRNA in lupus pathogenesis has been implicated in human SLE by both function and genetics. This study explored the novel functions and therapeutic potential of miR-146a in a mouse model of accelerated SLE. Methods: At 15 weeks, NZBW/F1 mice were infused with IFN␣ delivered by subcutaneously implanted osmotic pumps. The miR-146a (n ⫽ 8) or PBS (n ⫽ 7) administration began one week before the IFN␣ infusion and lasted for 9 weeks. Age matched NZBW/F1 mice were used as negative controls (n⫽10). Urine samples were tested for proteinuria by dipstick. Blood sera were used for autoantibody profiling. Gene expression profile data were collected from peripheral blood RNAs of both miR-146a and PBS treated groups. At the end of 9-week intervention, the kidney tissues were harvested and renal damage was histologically investigated by HE and IgG1 staining. Results: Administration of IFN␣ in NZBW/F1 mice resulted in an accelerated lupus. Proteinuria free survival rate differs between miR-146a treated and PBS-treated groups, with a significant delayed onset of proteinuria observed in the miR-146a-treated group (p ⬍ 0.01, Fig.1). IHC staining showed reduced IgG1 deposition in kidney tissues from mice treated with miR-146a, 9 weeks after miRNA treatment (Fig. 2). Gene expression profiling by microarray showed that miR-146a treatment decreased expression of immunoglobulin proteins (IgG1, IgG2b, IgM, and IgA). Consistent with these findings, autoantibody profiling of mouse sera showed that IFN␣ infusion triggered an induction of autoantibodies reactive with different glomerular or glomerular basement membrane antigens. However, the expression of some lupus relevant autoantibodies (anti-H3, anti-H4, anti-histone (total), anti-LC1, anti-MPO, anti-Ribophosphoprotein P1, anti-Sm-D1 and anti-U1-snRNP-BB’) was significantly suppressed following miR-146a treatment.

Conclusion: Overall, treatment with miR-146a attenuated lupus nephritis in the IFN␣ infused NZBW/F1 mice. The results are a proof of principal suggesting that interference with miR-146a may be a promising therapeutic strategy for treatment of lupus patients with renal involvement. Disclosure: D. Liang, None; S. Zhou, None; Z. Liu, None; Z. Shan, None; P. Brohawn, None; Y. Yao, None; I. Raman, None; Q. Z. Li, None; J. B. Harley, None; N. Shen, None.

2738 Microrna-Mediated Regulation Explains Allelic Risk Of TLR7 Variant Predisposing To Systemic Lupus Erythematosis. Yun Deng1, Jian Zhao1, Jennifer M. Grossman2 and Betty P. Tsao1. 1David Geffen School of Medicine University of California Los Angeles, Los Angeles, CA, 2UCLA David Geffen School of Medicine, Los Angeles, CA. Background/Purpose: We identified a genome-wide significant association of the TLR7 variant (rs3853839) with SLE susceptibility in multiple ancestries (Pmeta ⫽2.0⫻10⫺19, OR ⫽1.25), and the risk allele carriers exhibited elevated levels of TLR7 mRNA and protein and more pronounced

S1168

Canada, Europe and the UK for a combined population of 5,876 SSc cases and 9,517 controls. Results: As noted in the Table below, we identified and validated 4 novel SSc risk loci including DNASE1L3 at 3p14, SCHIP1 円 IL12A at 3q25, ATG5 at 6q21 and TREH/DDX6 at 11q23. Remarkably, the association of the rs35677470 missense variant in the DNASE1L3 locus with the ACA⫹ subset of patients is the most significant non-HLA association with SSc revealed to date (p ⫽ 2.70⫻10⫺32 OR⫽2.00). In addition, we further refined the area of association for the STAT4, IRF5/TNPO3 loci and related an observed peak of association in the PXK gene to the novel DNASE1L3 locus. Novel non-HLA loci associated with SSc and its subsets (p ⬍ 5⫻10⫺8) identified through Immunochip analysis. Locus DNASE1L3

SNP

Chr

Minor Allele

rs35677470

3p14

A

Comments Missense Arg⬎Cys

SCHIP112A

rs77583790

3q25

A

Intergenic

ATG5 TREH/DDX6

rs9373839 rs7130875

6q25 11q23

G G

Intronic Intergenic

Phenotype

MAF Cases/ CTRLs

p-value

OR

All SSc

0.088/0.062

1.20⫻10⫺15

1.43

lcSSc ACAⴙ All SSc lcSSc ACA⫹ All SSc All SSc

0.099/0.062 0.133/0.062 0.015/0.005 0.016/0.005 0.016/0.005 0.241/0.185 0.27/0.24

5.82⫻10⫺21 2.70ⴛ10ⴚ32 2.25⫻10⫺12 3.60ⴛ10ⴚ12 2.24⫻10⫺8 2.16⫻10⫺8 4.03⫻10⫺8

1.6 1.99 2.54 2.74 2.61 1.18 1.17

Conclusion: The DNASE1L3 association suggests that failure to clear apoptotic debris plays a role in SSc; that the IL12 pathway is key to SSc susceptibility; that autophagy (ATG5), previously unreported in SSc, may be an important mechanism; and that DDX6, which has been shown to regulate VEGF under hypoxic conditions may provide a clue to SSc vasculopathy. Disclosure: M. D. Mayes for the US Scleroderma GWAS Group, None; L. Bossini-Castillo for the Spanish Scleroderma Group, None; O. Gorlova, None; J. E. Martin, None; X. Zhou, None; W. Chen, None; S. Assassi, None; J. Ying, None; J. D. Reveille, None; P. K. Gregersen, None; A. T. Lee, None; M. Teruel, None; F. D. Carmona, None; B. P. C. Koeleman, None; M. A. Brown and the Immunochip Consortium, None; C. P. Denton, None; M. Baron for the Canadian Scleroderma Research Group, None; J. Broen, None; T. R. D. J. Radstake, None; J. Martin, None.

ACR Concurrent Abstract Session Innate Immunity and Rheumatic Disease Tuesday, October 29, 2013, 4:30 PM–6:00 PM

Disclosure: Y. Deng, None; J. Zhao, None; J. M. Grossman, None; B. P. Tsao, None.

2739 Immunochip Analysis Identifies New Susceptibility Loci For Systemic Sclerosis: Implications For Pathogenesis. Maureen D. Mayes for the US Scleroderma GWAS Group1, Lara Bossini-Castillo for the Spanish Scleroderma Group2, Olga Gorlova3, Jose Ezequiel Martin2, Xiaodong Zhou1, Wei Chen3, Shervin Assassi1, Jun Ying3, John D. Reveille1, Peter K. Gregersen4, Annette T. Lee4, Maria Teruel2, Francisco David Carmona5, Bobby P.C. Koeleman6, Matthew A. Brown and the Immunochip Consortium7, Christopher P. Denton8, Murray Baron for the Canadian Scleroderma Research Group9, Jasper Broen10, T.R.D.J. Radstake10 and Javier Martin11. 1University of Texas Health Science Center at Houston, Houston, TX, 2Instituto de Parasitologıá y Biomedicina Lo´pez-Neyra, IPBLN-CSIC, Granada, Spain, 3UT M.D. Anderson Cancer Center, Houston, TX, 4Feinstein Institute for Medical Research, Manhasset, NY, 5Instituto de Parasitologıá y Biomedicina Lo´pez-Neyra, IPBLN-CSIC, Armilla (Granada), Spain, 6Department of Medical Genetics, UMCU Utrecht, Utrecht, Netherlands, 7University of Queensland Diamantina Institute, Brisbane, Australia, 8Royal Free and University College Medical School, London, United Kingdom, 9Jewish General Hospital, Montreal, QC, 10University Medical Center Utrecht, Utrecht, Netherlands, 11Instituto de Parasitologia y Biomedicina Lo´pezNeyra, IPBLN-CSIC, Granada, Spain. Background/Purpose: The purpose of this study was to identify SSc risk loci shared with other autoimmune diseases on the Immunochip and to fine-map previously associated loci. Methods: We genotyped 1,959 SSc cases and 3,582 controls of European ancestry from the United States and Spain using the Immunochip custom array containing 196,524 SNP variants within 186 known autoimmune risk loci. Ten SNPs were then chosen for replication in 4,017 SSc cases and 5,935 controls from 6 additional populations of European ancestry from the US,

2740 High Efficacy Of Toll-Like Receptor 4 Targeting In Murine and Humanized Models Of Rheumatoid Arthritis In Comparison With IL-1 and TNF Inhibitors. Shahla Abdollahi-Roodsaz1, Marije I. Koenders2, Leo A. Joosten3, Fons A. van de Loo2 and Wim B. van den Berg1. 1Rheumatology Research and Advanced Therapeutics, Radboud University Nijmegen Medical Centre, Nijmegen, Netherlands, 2Rheumatology Research and Advanced Therapeutics, Radboud University Nijmegen Medical Center, Nijmegen, Netherlands, 3Department of Medicine, Radboud University Nijmegen Medical Center, Nijmegen, Netherlands. Background/Purpose: Increased expression of Toll-like Receptor (TLR) 4 and its endogenous agonists in rheumatoid joints suggest involvement in rheumatoid arthritis (RA). The aim of this study was to assess the therapeutic efficacy and downstream effects of TLR4 blockade in murine and humanized models compared with interleukin-1 (IL-1) and tumor necrosis factor (TNF) inhibitors. Methods: Mice with established collagen-induced arthritis (CIA) systemically received TNF, IL-1 and TLR4 inhibitors (Enbrel, Anakinra and purified B. quintana LPS, resp., each 2 mg/kg/day) using osmotic minipumps. Severe combined immunodeficient (SCID) mice engrafted with active RA synovial tissue, and ex vivo synovial cultures served to translate the findings into RA. Results: TLR4 blockade significantly suppressed clinical and histopathological manifestations of ongoing CIA to the same extent as IL-1 and TNF inhibitors. Targeting TLR4 substantially reduced serum IL-1␤ and IL-6, and was the only treatment capable of lowering serum TNFa. High-dose TLR4 inhibitor (8 mg/kg) was found significantly more

S1169

Tuesday, October 29

type I interferon (IFN) signature in SLE PBMCs. At or near the location of this variant are targeted sequences for predicted multiple microRNA (miRNAs) binding sites that may affect transcript degradation, leading us to explore whether differential miRNA binding could explain allelic risk of TLR7 for SLE. Methods: Each of the six bioinformatically predicted miRNAs was co-transfected with the TLR7 3’UTR reporter construct carrying either the risk or non-risk allele of rs3853839 into HEK 293 cells to assess which miRNA inhibits luciferase activity. To test effects of miRNA(s) on regulation of endogenous TLR7 expression and downstream IFN-␣ production, PBMCs from healthy donors were transfected with miRNAs or non-target control miRNAs for 24 hours, followed by stimulation with TLR7 agonist (CL097) for 24 hours to induce IFN-␣ production and measured in culture supernatants by ELISA. Cells were lysed for RNA extraction and mRNA levels of TLR7 and IFN scores (MX1, Ly6E, IFIT1 and IFIT3) were quantified by RT-PCR. Results: Transfection of each of six miRNAs into HEK-293 cells showed that (1) only miR-3148 and miR-2278 could reduce luciferase activity driven by the TLR7 3’UTR construct containing either risk or non-risk allele of rs3853839; (2) allelic differences in reduction of luciferase activity were observed by overexpression of miR-3148 at increasing concentrations [reduction in non-risk allele vs. risk-allele construct: 13.2% vs. 4.8%, P ⫽0.023 (6nM); 22.5% vs. 9.9%, P ⫽0.0012 (12nM); 21.4% vs. 8.5%, P ⫽0.0031 (48nM)], but only by overexpression of miR-2278 at the highest concentration [31.3% vs. 15.6%, P ⫽0.039 (48nM)]. Expression levels of miR-3148, but not miR-2278, were inversely correlated with TLR7 mRNA levels in PBMCs (R2 ⫽ 0.255, P ⫽0.001, and R2 ⫽0.086, P ⫽0.08, respectively). Compared to the non-target control miRNA, preliminary results of overexpression of miR-3148 in PBMCs (n⫽15) led to 15% reduction in TLR7 mRNA expression (P ⫽0.003), resulting in a trend of decreased downstream production of IFN-␣ (P ⫽0.07); whereas inhibition of miR-3148 led to 24% increase in TLR7 mRNA expression (P ⫽0.006) and a trend of increased production of IFN-␣ (P ⫽0.09). Overexpression or inhibition of miR-3148 in PBMCs showed no significant difference in IFN scores. Conclusion: Among the 6 tested miRNAs, miR-3148 and miR-2278 showed allelic difference in degradation of mRNA containing rs3853839, which results in increased expression of TLR7 mRNA in risk-allele carriers. The inverse correlation between miR-3148 and TLR7 mRNA levels and a trend of downregulation of IFN-␣ production in PBMCs ex vivo support miR-3148 regulates TLR7 expression and affects downstream IFN pathway. Our data suggests interactions between the genetic risk factor (rs3853839) and the epigenetic factor (miRNAs), implicating possibility of miRNA-based therapies to ameliorate SLE where excessive TLR7 activation exists.

Tuesday, October 29

effective than high-dose Enbrel (10 mg/kg; P ⬍ 0.05 using Bonferroni’s multiple comparison test). Importantly, TLR4 inhibition exceeded beneficial effects of TNF blocker by reducing serum IL-17 along with synovial gene expression of IL-23p19, IL-17 and the Th17-related transcription factor ROR␥t, while Th1 markers and type II collagen-directed T cell proliferation and antibody responses remained unaffected. In intact RA synovial biopsies, TLR4 stimulation (100 ng/ml E. coli LPS) potently induced TNF␣, IL-1␤, IL-6 and IL-8 production ex vivo, indicating its functional relevance. When transplanted into SCID mice, one single i.p. injection of TLR4 agonist significantly increased IL-6 production by RA synovium in vivo and sustained the otherwise declining IL-8 levels for up to 7 days. Importantly, TLR4 activation clearly reversed the therapeutic efficacy of anti-TNF treatment in the humanized RA synovium-SCID model, suggesting involvement in anti-TNF nonresponsiveness in subgroups of patients. Therapeutic value of TLR4 targeting in RA was revealed by inhibition of TLR4 in the RA synovium-SCID model. Blocking endogenous TLR4 activation in this model resulted in substantial reduction of spontaneous IL-6 and IL-8 release and synovial inflammation, thereby equaling anti-TNF. In RA synovial explant cultures ex vivo, TLR4 blockade by either B. quintana LPS or the small molecule inhibitor TAK242 suppressed several inflammatory cytokines. High-density (phospho)protein microarray of synovial protein lysates showed 27% reduction in NF␬Bp65 phospho-Ser536, but not NF␬Bp105/p50, levels by TLR4 blockade. Interestingly, TLR4 inhibition suppressed synovial expression of multiple other key signaling molecules including TAK1 phosphoThr187 and -Ser412 (26 and 23%, resp.), JAK1 phospho-Tyr1022 (22%), IRAK1 (22%), IRF3 (51%), Foxo3a (39%) and Btk (24%). Conclusion: Data in murine and humanized models position TLR4 upstream to a number of inflammatory and pathogenic pathways. The findings including impact on IL-17 production and anti-TNF responsiveness collectively impel future research on TLR4 as a potential therapeutic target in RA. Disclosure: S. Abdollahi-Roodsaz, None; M. I. Koenders, None; L. A. Joosten, None; F. A. van de Loo, None; W. B. van den Berg, None.

2741 Evidence That STAT3 controls NLRP3 inflammasome–dependent Release Of IL-1␤ and Pyronecrosis Through Regulation of mitochondrial activity. Jehad H. Edwan1, Jae Jin Chae2, Raphaela T. GoldbachMansky3 and Robert A. Colbert1. 1NIAMS NIH, Bethesda, MD, 2National Human Genome Research Institute, National Institutes of Health, Bethesda, MD, 3National Institute of Arthritis and Musculoskeletal and Skin Diseases, NIH, Bethesda, MD. Background/Purpose: Self-activating NLRP3 mutations are responsible for cryopyrin-associated periodic fever syndromes (CAPS), the most severe form of which is neonatal-onset multisystem inflammatory disease (NOMID). Spontaneous inflammatory responses initiated by NLRP3 mutations promote inflammasome-mediated IL-1beta processing and release, and can induce a rapid cell death in a process known as pyronecrosis that is dependent on cathepsin B activation. Emerging evidence suggests a role for mitochondria in activation of the inflammasome, and mitochondrial STAT3 has been implicated in regulating cellular respiration. The aim of this study was to determine whether STAT3 is involved in IL-1b release and pyronecrosis. Methods: We used whole blood cells from NOMID and healthy donors, THP-1 cells with STAT3 expression knocked down, THP-1 cells with NLRP3 expression knocked down, and monocytes derived from NRLP3 deficient mice. Cells were stimulated with LPS in the presence of inhibitors of STAT3, followed by ATP. Cell supernatants were collected and incubated with IL-1beta-capturing beads. Cells were fixed and permeabilized. Then beads were added back to cells, and the mixture of cells with beads was stained with anti-IL-1beta, CD14, CD16, and CD83 antibodies and then evaluated by flow cytometry. LPS stimulated cells were also evaluated using immunofluorescent, electron microscopy and western blot analysis. In addition, cell metabolism pathways were analyzed using an XF extracellular flux analyzer. Results: By flow analysis we found that a small population of monocytes, which undergoes pyronecrosis and is characterized by CD14hi/ CD16low and intracellular CD83 expression, is responsible for the majority of IL-1beta production and release. Using confocal microscopy to visualize pyronecrosis, we provided and evidence that this process is NLRP3 dependent. Blockade of STAT3 function with inhibitors leads to

complete inhibition of IL-1beta processing and release, as well as pyronecrosis in NOMID and ATP-stimulated healthy donor monocytes and in THP-1 cells. Similarly, STAT3 knockdown in THP-1 cells significantly inhibited IL-1beta release and pyronecrosis. Enhancement of the mitochondrial membrane potential in STAT3 knockdown cells bypasses the effect of STAT3 knockdown, and leads to increased IL-1beta processing and release, which was reversed with inhibitors of oxidative phosphorylation and glycolysis. Conclusion: Taken together, these data suggest a previously unrecognized role for mitochondrial STAT3 in mediating NLRP3 effects on pyronecrosis and IL-1beta release, and provide a novel therapeutic target for NOMID and other NRLP3-mediated inflammatory diseases. Disclosure: J. H. Edwan, None; J. J. Chae, None; R. T. Goldbach-Mansky, None; R. A. Colbert, None.

2742 Blau Syndrome-Associated NOD2 Mutations Limit Production Of IL-6 and KC/IL-8 In Knock-In Mice and In Patients Suggesting a Loss Of Function Disease Mechanism. Jae Dugan1, Eric Griffiths1, Paige Snow1, Holly L. Rosenzweig1, Carlos D. Rose2, Daniel Carr1, James T. Rosenbaum3 and Michael Davey1. 1Dept. of Veterans Affairs Medical Center, Portland, OR, 2Thomas Jefferson University/ AI duPont Hospital for Children, Wilmington, DE, 3Oregon Health and Science University, Portland, OR. Background/Purpose: Blau syndrome is an autosomal dominant disorder caused by mutations in nucleotide-binding oligomerization domain 2 (Nod2) and characterized by arthritis, dermatitis and uveitis. Nod2 binds muramyl dipeptide (MDP), a conserved structure from bacterial peptidoglycan, and activates NF-kB and MAPK signaling cascades. Prior in vitro studies using reporter assays indicated that Nod2 containing Blau mutations caused enhanced activation of NF-kB, suggesting a gain of function disease mechanism for Blau syndrome. However, studies with peripheral blood mononuclear cells (PBMCs) from patients with Blau syndrome stimulated with MDP show blunted cytokine responses. We tested the gain of function hypothesis in vivo by creating a knock-in mouse where a point mutation resulted in a change of arginine [R] to glutamine [Q] at position 314 (R314Q) of Nod2 (position 314 in mice corresponds to 334 in humans). Regulatory elements controlling Nod2 expression were not altered in this model. Methods: Knock in mice were studied for systemic cytokine responses. Bone marrow derived macrophages (BMDM) from knock in mice were studied by western blot analysis for Nod2 expression and intracellular signaling, and by ELISA assays. Macrophages derived from patients with Blau syndrome were studied for cytokine responses to MDP and phosphorylation of p38 mitogen activated protein kinase (MAPK). Results: R314Q heterozygous (⫹/m) and homozygous (m/m) mice did not spontaneously develop arthritis or dermatitis. BMDM from R314Q mice showed a reduction in full length (116 kD) Nod2 protein levels compared to wild type (WT) mice and they expressed an 80 kD protein that reacted with anti-Nod2 on western blotting. WT and mutant mice showed comparable amounts of Nod2 mRNA in BMDM and analysis of mRNA by PCR did not identify a splice variant. MDP treatment of BMDM showed reduced activation of NF-kB and p38 MAPK in ⫹/m and m/m compared to WT mice that correlated with the copy number of mutated Nod2, with the greatest reduction in m/m mice. In response to ip MDP, reduced levels of IL-6 and KC were detected in the serum of ⫹/m and m/m mice, also correlating with the copy number of the mutation. Macrophages derived from PBMCs of two patients with classic, familial Blau syndrome (a mother and son carrying an R334W mutation) showed negligible IL-6 and IL-8 production, and significantly reduced activation of p38 MAPK, in response to MDP compared to a healthy control. Conclusion: These data indicate that R314Q-Nod2 mice as well as patients with Blau syndrome studied here have lost the ability to respond to MDP, a pathway that is dependent on Nod2. Rather than a gain of function as previously thought, the mutations cause a deficiency of Nod2-dependent responses to MDP and raise the possibility that Blau syndrome may fall within the spectrum of an immunodeficiency disease. This observation may provide insights into why granulomas form in Blau syndrome and it might have important implications for treatment. Disclosure: J. Dugan, None; E. Griffiths, None; P. Snow, None; H. L. Rosenzweig, None; C. D. Rose, None; D. Carr, None; J. T. Rosenbaum, None; M. Davey, None.

S1170

2743 Targeted Activation Of The Metabolic Super-Regulator AMP-Activated Protein Kinase (AMPK) Blunts Gouty Inflammation. Ru L. Bryan1, Robert Terkeltaub2 and Yun Wang3. 1VA Medical Center/University of California San Diego, San Diego, CA, 2VA Medical Ctr/University of California San Diego, San Diego, CA, 3VA Medical Ctr, San Diego, CA.

Disclosure: R. L. Bryan, None; R. Terkeltaub, Takeda, Savient, ARDEA, BioCryst, Novartis, Regeneron, Sobi, Pfizer, Abbvie, 5; Y. Wang, None.

2744 Caspase 8 Modulates The Polarisome To Prevent Lung Fibrosis In a Murine Ssc-Like Disease Model. Carla M. Cuda1, Alexander Misharin2, Gokhan Mutlu2, Luisa Morales-Nebreda2, GR Scott Budinger2 and Harris R. Perlman2. 1Northwestern University Feinberg School of Medicine, Chicago, IL, 2Northwestern University, Chicago, IL. Background/Purpose: Pulmonary fibrosis has emerged as the leading cause of death in patients with Systemic Sclerosis (SSc). Currently available therapies are only marginally effective in treating this devastating complication and even patients who respond to therapy are left with significant respiratory morbidity. Caspase-8 is a cysteine-aspartic acid protease was originally identified as a key initiator of the apoptotic death receptor pathway and was later found to suppress necrotic programmed cell death (necroptosis) by inhibiting the receptor-interacting serine/threonine kinase 1/3 (RIPK1/3). We discovered a non-apoptotic role for caspase-8 in limiting the activation of macrophages and dendritic cells by toll-like receptor (TLR) agonists via a mechanism that involves RIPK. This novel function of caspase-8 is independent of its apoptotic function; unlike mice lacking Fas or overexpressing the general-baculoviral caspase inhibitor p35 in macrophages and DCs, the loss of caspase-8 does not result in changes in macrophage or DC numbers or affect their survival after bone marrow transplantation into lethally radiated mice.

Disclosure: C. M. Cuda, None; A. Misharin, None; G. Mutlu, None; L. MoralesNebreda, None; G. S. Budinger, None; H. R. Perlman, None.

2745 Reshaping Inflammatory Macrophage Development and Functions by a Monosaccharide Analogue In Rheumatoid Arthritis. Jun Li1, Hui-Chen Hsu2, PingAr Yang1, Qi Wu1, Bao Luo1, Amber L Rowse1, David M. Spalding1, James A Mobley1, S. Louis Bridges Jr.1 and John D. Mountz1. 1 University of Alabama at Birmingham, Birmingham, AL, 2Birmingham VA Medical Center, Birmingham, AL. Background/Purpose: Inflammatory macrophages (M⌽s) play key roles in pathogenesis of rheumatoid arthritis (RA). Fucosylation, comprising the transfer of a fucose (6-Deoxy-L-galactose) to proteins, is regulated by fucosyltransferases (FUTs) and involved in inflammation, oncogenesis, and cell differentiation. We have observed upregulated FUTs in synovial tissues from RA compared to osteoarthritis (OA) subjects. The purpose of the study is to determine: (i) the major cell types that produce FUTs; (ii) the roles of fucosylation; (iii) the efficacy of rebuilding the immune homeostasis by using a fucose analogue in RA. Methods: Twenty eight RA and OA subjects were recruited and the study is approved by UAB IRB. Q-PCR was performed to determine the expression of FUTs in synovial tissues and FACS sorted cells from RA synovial fluids. Inflammatory M⌽s were polarized by GM-CSF using monocytes from human PBMC, synovial fluid, and mouse bone marrow. Cells were treated with a fucose analog, 2-Deoxy-D-galactose (2-D-gal, inhibited the fucosylation mediated by FUT1/2) at different time points. Antigen presenting function was studied by using E␣-GFP peptide, DQ-Ova, and denatured bovine collagen II (CII)-FITC. Proteomics analysis was carried out by LCMS. Cytoskeleton images and video were collected using a Nikon spinning disk confocal microscope. Collagen-induced arthritis (CIA) was established in DBA/1J mice. 2-D-gal (200 mg/kg BW, every 2–3 days) was administered via I.P.. FACS and histopathology analyses were performed 6 weeks posterior primary CII immunization. Results: There is a highly positive correlation between TNF␣ with FUTs 1, 2, 3, 4, 5, 6, 7, 9, 10, 11, 12 (p⫽0.0001 for all), but not FUTs 8 (p⫽0.46) and 13 (p⫽0.47) in human RA synovia. In sorted cells from RA synovial fluid, FUTs 1, 3, 5, 7, 9 were highly expressed in M1 inflammatory M⌽, but not in M2 M⌽, synovial fibroblasts, and T cells (p⬍0.01), whereas FUTs 8 and 13 were predominately expressed in synovial fibroblasts. This highly indicated that subsets of FUTs might associate with the inflammatory M1 M⌽ characteristics. A fucose analog, 2-Deoxy-D-galactose (2-D-gal), precluded the differentiation of M1 M⌽. Phalloidin staining indicated 2-D-gal disrupted M⌽ actin-based cytoskeleton. Furthermore, LCMS analysis revealed that plectin-10, an actin regulatory protein, is the major target of 2-D-gal in M1 M⌽. These data suggested a potential role of fucosylation in antigen processing. Indeed, 2-D-gal treatment of fully differentiated M1 M⌽ for 2 days significantly reduced the uptake, processing and presentation of GFP-E␣ (from I-Ed␣), DQ-OVA, and FITC-Collagen II (CII) antigens (p⬍0.01). Additionally, 2-D-gal skewed the M1 M⌽ to M2 by increasing IL-10 secretion (p⬍0.01). In vivo, 2-D-gal treatment dramatically blocked bovine CII-induced arthritis (scores 9.5⫾1.7 vs 0.5⫾ 0.3, p⬍0.01) with reduced inflammatory M⌽ in draining LN (1.3⫾0.3% vs 0.5⫾0.1%, p⬍0.05), decreased TNF-␣ (130 vs 39 pg/ml, p⬍0.05), and anti-CII in the serum.

S1171

Tuesday, October 29

Background/Purpose: Gout is associated with both multiple dietary triggers and metabolic co-morbidities, including obesity, metabolic syndrome, and diabates. AMP-activated protein kinase (AMPK) is a master regulator of energy homeostasis, and tissue AMPK activity is decreased by high fat diet and in diabetes, and is conversely increased by clarotic restraint, exercise, and several drugs in the clinic for arthritis (sodium salicylate, high dose aspirin, methotrexate) and by metformin. AMPK is a heterotrimeric complex compoased of an alpha catalytic subunit and two regulatory beta and gamma subunits. Phosphorylation at Thr-172 within the catalytic domain of the a subunit is critical for AMPK activity. Activation of AMPK exerts anti-inflamamtory effects partly by suppressing NF-kB activation. Thus, we assessed the function of AMPK in monsodium urate (MSU) crystal-induced inflammation in vitro and in vivo. Methods: We added MSU crystals (0.2 mg/ml) for 18 h to bone marrow derived macrophages (BMDMs) from AMPKa1 knockout (KO) mice and C57BL/6 mice, and pre-treated with AMPK activators AICAR (1 mM) or highly AMPK selective A-769662 (0.25 mM). We analyzed conditioned media by ELISA for cytokines that drive gout arthritis, and cell lysates by Western blot for AMPK activity (phosphorylation of AMPKa). We also injected A-769662 (0.5 mM) one hour prior to injection of MSU crystals (3 mg) into mouse SQ air pouches in vivo, and collected samples 6 hrs later. Results: MSU crystals decreased AMPK activity (phosphorylation of AMPKa Thr172) in BMDMs, and both AICAR and A-769662 inhibited IL-1b and CXCL1 induction by 75% (p⫽0.0006) and 88% (p⫽0.0009), respectively. Conversely, MSU crystal-induced production of IL-1b and CXCL1 was enhanced 1.5 to 2-fold in AMPKa1 knockout BMDMs. Colchicine at 10 nM concentration achieved by standard, low dose therapy for acute gout, up-regulated AMPK activity in BMDMs. Last, A-769662 attenuated inflammatory responses to MSU crystals in vivo by 72% reduction of number of infiltrating leukocytes (p⫽0.04) and decrease in IL-1b and CXCL1 in pouch fluid 60% (p⫽0.003) and 50% (p⫽0.025), respectively. Conclusion: AMPKa1 deficiency enhances MSU crystal-induced macrophage IL-1b and CXCL1 release in vitro. Conversely, selective activation of AMPK by A-769662 markedly suppresses MSU crystal-induced inflammatory responses both in vitro and in vivo. We also established that nanomolar colchicine, in addition to certain other agents in the clinic for arthritis and diabetes, has the capacity to suppress gouty inflammation partly via AMPK activation. Our results indicate a novel, diet and drug targetable mechanism by which high fat diet, diabetes and the urate crystal by itself, are linked with heightened inflammatory potential of urate crystals in gout through decreased tissue activity of the metabolic super-regulator AMPK.

Methods: Mice lacking caspase 8 specifically in DCs or macrophages were generated (CreCD11cCasp8flox/flox CreLysMCasp8flox/flox) and examined using the bleomycin and adenoviral TGF-␤ models of lung fibrosis Flow cytometric analysis was used to characterize macrophage and DC. Luminexbased assays detected cytokine levels. Results: We show that mice deficient in caspase-8 specifically in macrophages and DCs (have markedly less lung fibrosis than their littermate controls following intratracheal treatment with either bleomycin or an adenovirus encoding an active form of TGF-␤. In wild-type animals, we found that the development of fibrosis was accompanied by a polarization of interstitial macrophages into an alternatively activated or M2 phenotype while interstitial macrophages from animals lacking caspase-8 remained polarized toward an M1 phenotype. M2 polarized macrophages have been associated with the release of profibrotic mediators involved in tissue fibrosis and M2 polarized macrophages are found in the lungs of patients with pulmonary fibrosis. Conclusion: These data suggest a new paradigm for fibrosis in which caspase-8 functions as a key component of a “polarisome” regulating the polarization of macrophages toward an M1 or M2 phenotype. Inhibiting the function of this polarisome in macrophages from patients with scleroderma may prevent the development of lung fibrosis.

Conclusion: Fucosylation, a hallmark of M1 M⌽, orientates M⌽ polarization and function. 2-D-gal, a fucose analog restores the deranged M1 M⌽ and leads to resolution of arthritis by inhibiting fucosylation of cytoskeleton molecules. Disclosure: J. Li, Arthritis Foundation, 2; H. C. Hsu, NIH (1R01AI083705-01A2); Lupus Research Institute, 2; P. Yang, None; Q. Wu, None; B. Luo, None; A. L. Rowse, None; D. M. Spalding, None; J. A. Mobley, None; S. L. Bridges Jr., None; J. D. Mountz, VA Merit Review Grant (1I01BX000600-01);NIH/NIAID (1AI 07111001A1);Rheumatology Research Foundation, 2.

ACR Concurrent Abstract Session Miscellaneous Rheumatic and Inflammatory Diseases I: Autoinflammatory Syndromes Tuesday, October 29, 2013, 4:30 PM–6:00 PM

2746

Tuesday, October 29

A Novel Autoinflammatory Disorder Characterized By Ectodermal Dysplasia, Metaphyseal Chondrodysplasia, Growth Failure and HyperIgD In a Single Family. Edward J. Oberle and James W. Verbsky. Children’s Hospital of Wisconsin, Milwaukee, WI. Background/Purpose: Autoinflammatory disorders are characterized by chronic inflammation, and a variety of systemic complaints. Defects in the NF-kb essential modifier, or NEMO, are associated with an immune deficiency characterized by ectodermal dysplasia. We present a novel autoinflammatory disorder following an autosomal dominant pattern characterized by chronic inflammation, metaphyseal chondrodysplasia, ectodermal dysplasia with hypohidrosis, hyper-IgD and growth failure not associated with immune deficiency in a family. Methods: Seven individuals from one family were identified. The proband, a male patient age 12 with known hypohydrosis, presented to our rheumatology clinic for chronic articular pain with bony hypertrophy of wrists, knees, and ankles. Standard labs were obtained including ESR, CRP, and CBC for all affected individuals. Plain radiographs and immunoglobulins were obtained from patient, father and sister. Films reviewed by two musculoskeletal radiologists. The proband underwent proximal tibial bone biopsy and marrow aspirate, as well as genetic testing for known autoinflammatory disorders. Results: The proband, his father age 30, two brothers age 7 and 9, and his 2-year-old sister were all found to have ectodermal dysplasia, short stature and failure to thrive (Figure 1). The youngest affected was evaluated for fevers and elevated inflammatory markers from birth. The patient’s father was admitted as a child for bone issues. All affected members exhibited chronically elevated ESR ranging between 64–118 mm/hr and elevated WBC. Radiographs of proband displayed symmetric lucent and sclerotic lesions of the metaphysis of distal radius/ulna, distal femora, and proximal and distal tibia/fibula (Figure 2). His father demonstrated widened metaphyses of distal ulna and tibia, but did not have lesions. Immunoglobulins were elevated in the patient and his father with IgD levels of 755 mg/L (⬍179 normal) and 600 mg/L, respectively. Bone biopsy of the proband was consistent with metaphyseal chondrodysplasia with marrow spaces consisting of loose fibrous stroma or cartilage, with scattered lymphocytes and plasmacytoid cells. Genetic testing was negative for mutations of CIAS1, LPIN2, MVK, IKBKG (NEMO), PSTP1P1, and TRAPS.

Figure 1. Pedigree.

Figure 2. Proband radiographs.

Conclusion: We propose a novel, autoinflammatory disorder resulting in osseous abnormalities, ectodermal dysplasia and Hyper IgD. Disclosure: E. J. Oberle, None; J. W. Verbsky, None.

2747 Canakinumab Treatment In Schnitzler’s Syndrome: A Multi-Center Randomized Placebo-Controlled 4-Month Study. Karoline Krause, Karsten Weller, Martin Metz and Marcus Maurer. Dept. of Dermatology and Allergy, Allergie-Centrum-Charite´, Charite´ – Universita¨tsmedizin Berlin, Germany, Berlin, Germany. Background/Purpose: Schnitzler’s syndrome (SchS) is an adultonset autoinflammatory disease characterized by urticarial exanthema and monoclonal gammopathy in combination with episodes of fever, arthralgia, fatigue, and bone and muscle pain. Anti-IL-1 targeting therapies in small patient numbers including an open-label study with canakinumab (CAN) showed to be effective in reducing the clinical symptoms of SchS. Methods: The current placebo-controlled study was designed to assess the effects of the selective anti-IL-1␤ humanized monoclonal antibody CAN on the clinical signs and symptoms of SchS in a larger patient cohort. A total of 20 patients with active disease enrolled in this multi-center trial. After a baseline period of up to 4 weeks, patients were randomized to receive a single CAN 150mg or placebo s.c. injection (day 0) and were evaluated for treatment response at day 7. This initial study period was followed by a 16-week open label phase with CAN injections upon confirmed relapse of clinical symptoms. Efficacy was determined by changes in the physician’s global assessment (PGA; range 0–20), a combined symptom score which includes 5 key symptoms of SchS (urticarial rash, fever, fatigue, arthralgia and myalgia), measurement of the inflammation markers C-reactive protein (CRP) and serum amyloid A (SAA) as well as changes in quality of life assessment (SF-36). Results: CAN was highly effective (P⫽0.001) in reducing median PGA total scores (14.0 to 2.0) within 7 days after first administration as compared to placebo treatment (15.0 to 13.0) in SchS patients. Also, significant (P⬍0.0001 – P⬍0.05) improvements were observed for each key symptom score. Median CRP reduced from 9.3mg/dL at baseline to 0.6mg/dL at day 7 in the CAN group vs increase from 3.0mg/dL to 5.0mg/dL for the placebo group. Similarly, median SAA levels reduced from 428mg/L to to 13mg/L for the CAN group vs increase from 160mg/L to 205mg/L for the placebo group. The median change from baseline between treatment groups for CRP (p⫽0.004) and SAA (p⫽0.002) was significant. Likewise, quality of life as measured by SF-36 significantly improved (P⫽0.001) for the CAN vs placebo groups at day 7. These improvements were maintained during the 16-week open label phase of the study. A total of 22 adverse events (AEs) were reported during the study including 3 serious AEs (2 hypertensive episodes in 1 patient and severe lumbago in another patient). Conclusion: In this 4-month study, CAN s.c. injections significantly improved the clinical signs and symptoms of SchS, reduced inflammation

S1172

markers, and enhanced quality of life. CAN treatment may be considered a promising therapeutic option in these patients. Disclosure: K. Krause, None; K. Weller, None; M. Metz, None; M. Maurer, Novartis Pharmaceutical Corporation, 2.

2748 Safety Results From The Beta Confident Registry In CanakinumabTreated Patients With Cryopyrin-Associated Periodic Syndrome. Hal Hoffman1, Jasmin B. Kuemmerle-Deschner2, Philip N. Hawkins3, T. van der Poll4, Ulrich A. Walker5, Michael Nebesky6, Ken Abrams7 and Hugh Tilson8. 1University of California at San Diego, San Diego, CA, 2University Hospital Tuebingen, Tuebingen, Germany, 3University College London Medical School, London, United Kingdom, 4Academic Medical Center, University of Amsterdam, Amsterdam, Netherlands, 5Universita¨ts-Poliklinik, Felix-Platter Spital, Basel, Switzerland, 6Novartis Pharma AG, Basel, Switzerland, 7Novartis Pharmaceuticals Corporation, East Hanover, NJ, 8The University of North Carolina, Gillings School of Global Public Health, Chapel Hill, NC.

References: 1. Arthritis Res Ther 2011, 13:R34, 2. Lachmann et al. J Exp Med 2009, 206: 1029–1036, 3. Arthritis Rheum 2008;58:2443–2452 Disclosure: H. Hoffman, Novartis, Regeneron, Sobi Biovitrum, 5; J. B. Kuemmerle-Deschner, Novartis,, 2, Novartis,, 5; P. N. Hawkins, None; T. van der Poll, Novartis,, 5; U. A. Walker, Novartis,, 5; M. Nebesky, Novartis,, 3; K. Abrams, Novartis,, 3, Novartis,, 1; H. Tilson, Bio Soteria, Bristol-Myers Squibb, Gilead, GlaxoSmithKline, HealthCore, Kendle, Merck, Novartis, 5, Glaxo SmithKline, Procter & Gamble, Other non-pharmaceutical holdings, 1.

Amyloidosis and Its Related Factors In Patients With Familial Mediterranean Fever: A Nationwide Multicenter Study. Timucin Kasifoglu1, Sule Yasar1, Ismail Sari2, Dilek Solmaz2, Soner Senel3, Hakan Emmungil4, Levent Kilic5, Sibel Yilmaz Oner6, Fatih Yildiz7, Sedat Yilmaz8, Muhammet Cinar8, Duygu Ersozlu Bakirli9, Muge Aydin Tufan7, Sema Yilmaz10, Veli Yazisiz11, Yavuz Pehlivan12, Cemal Bes13, Gozde Yildirim Cetin14, Sukran Erten15, Emel Gonullu1, Tuncer Temel1, Servet Akar2, Kenan Aksu4, Umut Kalyoncu5, Haner Direskeneli16, Eren Erken7, Bunyamin Kisacik12, Mehmet Sayarlioglu17 and Cengiz Korkmaz18. 1Eskisehir Osmangazi University School of Medicine, Eskisehir, Turkey, 2Dokuz Eylul University School of Medicine, Izmir, Turkey, 3Kayseri Erciyes University School of Medicine, Kayseri, Turkey, 4Ege University School of Medicine, Izmir, Turkey, 5 Hacettepe University School of Medicine, Ankara, Turkey, 6Marmara University School of Medicine, Istanbul, Turkey, 7Cukurova University School of Medicine, Adana, Turkey, 8Gulhane School of Medicine, Ankara, Turkey, 9Adana Numune Training and Research Hospital, Adana, Turkey, 10Selcuk University School of Medicine, Konya, Turkey, 11Akdeniz University School of Medicine, Antalya, Turkey, 12Gaziantep University School of Medicine, Gaziantep, Turkey, 13 Bolu Izzet Baysal School of Medicine, Bolu, Turkey, 14Kahramanmaras University School of Medicine, Kahramanmaras, Turkey, 15Ataturk Training and Research Hospital, Ankara, Turkey, 16Marmara University, School of Medicine, Istanbul, Turkey, 17Ondokuz Mayis University School of Medicine, Samsun, Turkey, 18Eskisehir Osmangazi University, Eskisehir, Turkey. Background/Purpose: Familial Mediterranean fever (FMF) is an autosomal recessive autoinflammatory disease that occurs worldwide and predominantly affects the populations arising from the Mediterranean origin. Secondary (AA) amyloidosis still remains the most devastating complication of FMF especially in untreated and noncompliant patients. However, pathogenesis and risk factors of amyloidosis still remains only partially understood in FMF. The primary aim of this study was to investigate the prevalence of amyloidosis and its related factors in a large number of FMF patients. Methods: Fifteen centers from the different geographical regions of Turkey were included in the study. Detailed demographic and medical data based on structured questionnaire and medical records were collected. The diagnosis of amyloidosis was based on histological proof of congophilic fibrillar deposits in tissue biopsies. Results: There were 2246 FMF patients. The male/female ratio was 0.87 (1049/1197). The mean ages of the patients were 33.6⫾0.25 years. Peritonitis was the most frequent clinical finding of the patients and it was present in 94.6% of the patients. The other clinical features were fever (91.9%), pleuritis (47.9%), arthritis (39.8%), erysipelas like erythema (ELE; 23.7%), myalgia (13%) and vasculitis (2.7%). Genetic testing was available in 1719 patients (76.5%). The most frequently observed genotype was homozygous M694V mutation which was present in 413 (24%) patients. Amyloidosis was present in 193 (8.6%) patients. Male sex, arthritis, M694V genotype, patients with end stage renal disease (ESRD), family history of amyloidosis and ESRD was significantly more prevalent in patients with amyloidosis compared with the amyloidosis negative subjects (Table 1). Patients with homozygous M694V mutations had significantly increased frequency of arthritis, ELE, amyloidosis, ESRD and family history of FMF and ESRD compared to the other genotypes. Table 1. Comparison of patients with or without amyloidosis

Male sex Peritonitis Pleuritis Arthritis Patients with ESRD M694V homozygosity Delay in the diagnosis (years) Family history of amyloidosis Family history of ESRD

S1173

Amyloidosis (ⴙ) nⴝ193 105, (54.4) 170, (88.1) 73, (37.8) 99, (51.3) 101, (52.3) 90, (45.6) 8 54, (28) 32, (16.6)

Amyloidosis (ⴚ) nⴝ2053 944, (46) 1956, (95.3) 1002, (48.8) 796, (38.8) 10, (0.5) 323, (15.7) 7 428, (20.8) 81, (3.9)

P 0.026 ⬍0.001 ⬍0.001 0.001 ⬍0.001 ⬍0.001 0.147 0.027 ⬍0.001

Tuesday, October 29

Background/Purpose: Cryopyrin-associated periodic syndrome (CAPS) comprises an extremely rare auto-inflammatory disorders, including familial cold auto-inflammatory syndrome (FCAS), MuckleWells syndrome (MWS), and neonatal-onset multisystem inflammatory disease/chronic infantile neurological cutaneous and articular syndrome (NOMID/CINCA).1 NLRP3 gene mutation results in overproduction of IL-1␤, leading to observed clinical symptoms of CAPS.2 Canakinumab (CAN) is a selective anti-IL-1␤ monoclonal antibody, approved for the treatment of CAPS.2,3 Here, we report the updated safety data for CAN in CAPS patients (pts) from the ongoing ␤-Confident Registry. The primary objective of the Registry is to monitor and further explore the overall safety of CAN, focusing primarily on serious infections, malignancies, hypersensitivity reactions, vertigo, and other selected adverse events in CAPS pts over a 5-year period (NCT01213641). Methods: The Registry includes pts with CAPS and other autoinflammatory disorders receiving CAN as part of regular medical care, after obtaining the written informed consent The study protocol does not mandate any visits or procedures, records all observed and reported adverse events (AEs) and serious adverse events (SAEs) or AEs potentially related to treatment with CAN. Cumulative safety data are reported, as incidence rate (number of events) per 100 patient-years (IR/100 pyr) from the date of first pt enrollment (November 19, 2009) until the data cut-off date (March 29, 2013). Additional safety data will be updated, as available, at the time of the conference presentation. Results: A total of 245 pts were enrolled in the Registry at the current cut-off date, of which 229 patients reported a median duration of 50 weeks (range: 0.9 – 137.3) of prior treatment with CAN at the time of Registry enrollment. Of the total, 100 pts reported 238 AEs with an IR of 75.9/100 pyr. Infections with an IR of 20.7/100 pyr, were the most commonly observed AEs and included nasopharyringitis, rhinitis and urinary tract infections (UTI) among others. Eight pts reported 13 events of vertigo as AEs, resulting in an IR of 4.2/100 pyr, of which 8 were suspected to be CAN related. With regards to SAEs, 37 events were reported by 27 pts with an IR of 11.8 /100 pyr that included one fatal case of malignancy (metastatic rectal adenocarcinoma in a 76 yr MWS patient) and 12 events of serious infections. No case of hypersensitivity to CAN was reported. A total of 14 pts permanently discontinued CAN: 4 each due to pts preference and AEs, 3 due to lack of therapeutic effects, and 3 due to unspecified reasons. Conclusion: No new or unexpected safety signals were reported to date in this ongoing Registry. Infections, as expected, were the most commonly occurring AEs and SAEs.

2749

Conclusion: In this nationwide study we revealed that 8.6% of our FMF patients had amyloidosis and homozygosity for M694V was the most common mutation in these patients. The latter finding confirms the association of homozygous M694V mutation with amyloidosis in Turkish FMF patients. Disclosure: T. Kasifoglu, None; S. Yasar, None; I. Sari, None; D. Solmaz, None; S. Senel, None; H. Emmungil, None; L. Kilic, None; S. Yilmaz Oner, None; F. Yildiz, None; S. Yilmaz, None; M. Cinar, None; D. Ersozlu Bakirli, None; M. Aydin Tufan, None; S. Yilmaz, None; V. Yazisiz, None; Y. Pehlivan, None; C. Bes, None; G. Yildirim Cetin, None; S. Erten, None; E. Gonullu, None; T. Temel, None; S. Akar, None; K. Aksu, None; U. Kalyoncu, None; H. Direskeneli, None; E. Erken, None; B. Kisacik, None; M. Sayarlioglu, None; C. Korkmaz, None.

Tuesday, October 29

2750 Long-Term Efficacy and Safety Of IL1 Receptor Antagonist In Schnitzler’s Syndrome: A French Multicenter Study. Antoine Neél1, Agathe Masseau1, Sebastien Barbarot1, Benoit Henry2, Pierre-Jean Weiller3, Olivier Decaux4, Xavier Kyndt5, Xavier Puechal6, Arnaud Hot7, Pierre Pottier1, Amar Smail8, David Launay9, Jean-Marie Berthelot1, Eric Hachulla9, Leonardo Astudillo10, Pierre-Yves Hatron9, Laurent Sailler10, Aurelien Lorleac’h11, Achille Aouba12, Be´range`re Cador4, Renato Fior13, Robin Dhote14, Fabrice Bonnet15, Jean-Dominique de Korwin16 and Mohamed Hamidou1. 1Nantes University Hospital, Nantes, France, 2Pitie´-Salpeˆtrie`re University Hospital, Paris, France, 3La Timone University Hospital, Marseille, France, 4Rennes University Hospital, Rennes, France, 5Hospital of Valenciennes,, Valenciennes, France, 6Hoˆpital Cochin, Paris, France, 7Edouard Herriot University Hospital, Lyon, France, 8Amiens University Hospital, Amiens, France, 9Claude Huriez University Hospital, Lille, France, 10 Toulouse University Hospital, Toulouse, France, 11Lorient Hospital, Lorient, France, 12Necker University Hospital, Paris, France, 13Antoine Bećle`re University Hospital, Clamart, France, 14Avicenne University Hospital, Bobigny, France, 15Bordeaux University Hospital, Bordeaux, France, 16Nancy University Hospital, Nancy, France. Background/Purpose: Schnitzler’s syndrome is a rare late onset autoinflammatory disease which associates a chronic/recurrent urticarial skin rash, a monoclonal gammopathy (mostly IgM kappa), and a variable combination of intermittent fever, osteoarticular pain, sclerotic bone lesions, lymphadenopathy and/or hepatosplenomegaly. This chronic disease can significantly alter patients’ quality of life through intermittent fever, rash, pain and sometimes profound weight loss or anaemia. Patients can also develop a hematologic malignancy (mainly Waldenstro¨m’s macroglobulinemia). In recent years several case reports have underscored the efficacy of Anakinra but long-term follow-up data are scare. Further, a few patients who did not respond to IL1 blockade have been reported recently. Methods: Retrospective analysis of a French multicenter cohort of 40 patients with Schnitzler’s syndrome. Results: In this cohort mean age at disease onset was 61y (53–74). Mean diagnostic delay was 3 years. Disease manifestations included urticarial rash (100%), intermittent fever (81%), weight loss (62%), bone and/or articular pain (86%) and anaemia (Hb⬍10g/dL; 46%). All patients had systemic inflammation and 80% had bone lesions. Monoclonal gammopathy was IgM kappa in 85% of cases. Mean follow-up was 9 years (2–35). Twenty eight patients (70%) received Anakinra. Thirteen of those were corticodependant (mean dose: 15mg/d) when Anakinra was introduced. All 28 patients experienced a dramatic improvement of all clinico-biological signs of disease activity. Mean hemoglobin level rose from 11,2 to 13,4 g/dL (p⬍0.001). At last follow-up, all patients remained on anti-IL1 therapy. A single patient was switched to Canakinumab, due to injection site reactions to anakinra. Eighty percent of patients were in complete remission under anti-IL1 monotherapy, 11% were in complete remission under anakinra plus low dose corticosteroids and 4% were in partial remission under anakinra monotherapy. Mean follow-up under anakinra was 40 months (2–73). No loss of effectiveness was observed. In case of treatment interruption most patients experienced disease flare within 24–48 h. Alternate day injections were sufficient to maintain remission in only 4 cases. Three cases of uncomplicated neutropenia were recorded. Six patients developed severe infections including pneumonia in 5 cases, with local predisposing factor in 4 (severe COPD in 3, enteral feeding in 1). Anakinra had no obvious effect on the monoclonal component. No lymphoproliferative disease occured. When last seen, all 12 patients without anakinra had an active disease with variable impact on quality of life. Their mean corticosteroids dosage was 7mg/d. Conclusion: Anakinra is dramatically effective in Schnitzler’s syndrome. Treatment failure should lead to reconsider the diagnosis. Long-term follow-up reveals no loss of effectiveness and a favourable safety and

tolerance profile. Alternate day injections are rarely sufficient to maintain remission. The effects regarding the risk of malignant transformation remain undertermined. Disclosure: A. Neél, None; A. Masseau, None; S. Barbarot, None; B. Henry, None; P. J. Weiller, None; O. Decaux, None; X. Kyndt, None; X. Puechal, None; A. Hot, None; P. Pottier, None; A. Smail, None; D. Launay, None; J. M. Berthelot, None; E. Hachulla, None; L. Astudillo, None; P. Y. Hatron, None; L. Sailler, None; A. Lorleac’h, None; A. Aouba, None; B. Cador, None; R. Fior, None; R. Dhote, None; F. Bonnet, None; J. D. de Korwin, None; M. Hamidou, None.

2751 Recurrent Pericarditis In Children and Adolescents: Etiology, Presentation, Therapies, and Outcomes In a Multicenter Retrospective Cohort Of 100 Patients. Antonio Brucato1, Massimo Imazio2, Marco Gattorno3, Antonella Insalaco4, Chiara Di Blasi Lo Cuccio1, Simona Marcora5, Rolando Cimaz6, Luca Cantarini7, Luciana Breda8, Manuela Marsili8, Fabrizia Corona9 and Alberto Martini10. 1USC Internal Medicine, Ospedale Papa Giovanni XXIII, Bergamo, Italy, 2Maria Vittoria Hospital, Torino, Italy, 3G. Gaslini Institute, Genova, Italy, 4Ospedale Pediatrico Bambino Gesu`, Roma, Italy, 5Ospedale Papa Giovanni XXIII, Bergamo, Italy, 6A. Meyer Children’s Hospital, Florence, Italy, 7Policlinico Le Scotte, University of Siena, Siena, Italy, 8University of Chieti G. D’Annunzio, Chieti, Italy, 9Paediatric Department University of Milano, Milano, Italy, 10Istituto G Gaslini, Pediatria II, Reumatologia, Genova, Italy. Background/Purpose: Acute pericarditis is defined as at least 2 of the following criteria: typical chest pain (sharp and pleuritic, improved by sitting up and leaning forward), pericardial friction rub, suggestive EKG changes (widespread ST elevation or PR depression), new or worsening pericardial effusion. Pericarditis may recur in 15–30% of cases, and these children are frequently followed by pediatric Rheumatologists. In up to 68% of pediatric patients and in more than 80% of adult cases a specific etiology cannot be detected and pericarditis is considered idiopathic. Suggested explanations of recurrences include: insufficient dose and/or duration of non-steroidal antiinflammatory drugs (NSAIDs), early corticosteroid treatment causing increased viral replication in pericardial tissue, too rapid tapering of corticosteroids, re-infection and exacerbation of the connective tissue disease. Methods: Multicenter, retrospective cohort study including all consecutive cases of recurrent pericarditis of patients aged ⬍18 years with at least 2 recurrences of pericarditis seen in referral centers in Italy. The study included 100 cases of recurrent pericarditis (median 13 years, range 1–17 years, 62 males). Results: Pericarditis was idiopathic or viral in 87.0% of cases, postpericardiotomy in 9.0% of cases; Familial Mediterranean Fever was diagnosed in 2.0% of cases, and a systemic inflammatory disease in 2.0% of cases. The majority of children had fever and CRP elevation at disease onset (96% fever, 98% CRP elevation), and pericardial effusion (57%), while pericardial rub (25%) and EKG changes (42%) were detected in a smaller percentage of patients. Corticosteroid-treated patients experienced more recurrences, side effects, and disease-related hospitalization (for all p⬍0.05). After a median follow up of 60 months (6–360 months), 470 recurrences were recorded (median 3, range 2–25). Duration of the active disease was unpredictable in the single patient and largely variable. Overall, 97% of cases had additional recurrences. ANA testing is ongoing. Additional adverse events during follow-up included: readmission in 74% of cases and cardiac tamponade in 13% of cases (in the first attacks). None of our patients diagnosed as having an idiopathic recurrent pericarditis developed a systemic autoimmune disease; 78% children were treated with NSAIDs, 62% with steroids and in 62% cases colchicine was added. Refractory cases (17%) were treated with immunosuppressant drugs (azathioprine, methotrexate, IVIG, Plaquenil), and 7 (7%) with Anakinra. Long term outcome was good, with no evolution in constrictive pericarditis, cardiomyopathy or systemic rheumatic diseases. Conclusion: this is the largest ever published case-series of pediatric recurrent pericarditis. NSAIDs and colchicines remain the mainstay of the therapy while corticosteroids should be used with extreme caution, particularly in pediatric patients. In cortico-dependent cases anti-IL1 drugs proved to be very effective. The long term outcome is good. Disclosure: A. Brucato, None; M. Imazio, None; M. Gattorno, None; A. Insalaco, None; C. Di Blasi Lo Cuccio, None; S. Marcora, None; R. Cimaz, None; L. Cantarini, None; L. Breda, None; M. Marsili, None; F. Corona, None; A. Martini, None.

S1174

ACR Concurrent Abstract Session Osteoarthritis II: Risk Factors and Natural History of Osteoarthritis Tuesday, October 29, 2013, 4:30 PM–6:00 PM

2752 Good, Moderate and Poor Outcome Trajectories Of Pain Severity In Early Symptomatic Knee Osteoarthritis; 5 Year Follow-Up Of Check study. Janet Wesseling1, Alex N. Bastick2, Sita M.A. Bierma-Zeinstra2 and Johannes W.J. Bijlsma1. 1University Medical Center Utrecht, Utrecht, Netherlands, 2Erasmus MC - University Medical Center, Rotterdam, Netherlands.

Disclosure: J. Wesseling, None; A. N. Bastick, None; S. M. A. Bierma-Zeinstra, None; J. W. J. Bijlsma, None.

2753 Relationship Of Pre-Radiographic MRI Lesions With Prevalent Frequent Knee Symptoms and Incident Persistent Knee Symptoms In Persons At Higher Risk For Knee Osteoarthritis. Leena Sharma1, Joan S. Chmiel1, Orit Almagor2, Dorothy D. Dunlop1, Marc C. Hochberg3, Charles Eaton4, C. Kent Kwoh5, Rebecca D. Jackson6, Joan M. Bathon7, Ali Guermazi8, Frank Roemer9, Michel Crema8, W. Jerry Mysiw6 and Michael C. Nevitt10. 1Northwestern University, Chicago, IL, 2Northwestern University Medical School, Chicago, IL, 3University of Maryland, Baltimore, MD, 4 Brown University, Providence, RI, 5University of Pittsburgh, Pittsburgh, PA, 6Ohio State University, Columbus, OH, 7Columbia University, New York, NY, 8Boston University, Boston, MA, 9Klinikum Augsburg, Augsburg, Germany, 10University of California, San Francisco, San Francisco, CA.

Table 1. Pre-radiographic MRI Lesions and Risk of Prevalent Frequent Knee Symptoms

Number of knees (1 knee per person)

Number of knees (row%) with frequent knee symptoms

OR (95% CI) Unadjusted

OR (95% CI) Adjusted for age, gender, BMI, previous knee injury, previous knee surgery

Cartilage damage (TF or PF)

642

178 (27.7%)

1.81 (1.21, 2.69)

1.75 (1.16, 2.63)*

Cartilage damage (TF only)

118

27 (22.9%)

0.86 (0.54, 1.36)

0.78 (0.49, 1.26)*

Cartilage damage (PF only)

257

64 (24.9%)

0.98 (0.70, 1.37)

0.95 (0.67, 1.35)*

Cartilage damage (TF and PF)

266

87 (32.7%)

1.73 (1.26, 2.40)

1.85 (1.32, 2.59)*

No cartilage damage (TF or PF)

208

36 (17.3%)

reference

reference

BML (TF or PF)

515

155 (30.1%)

2.00 (1.42, 2.80)

1.95 (1.38, 2.76)*

BML (TF only)

79

21 (26.6%)

1.08 (0.64, 1.82)

1.04 (0.60, 1.79)*

BML (PF only)

289

84 (29.1%)

1.34 (0.98, 1.87)

1.38 (0.99, 1.92)*

BML (TF and PF)

147

50 (34.0%)

1.68 (1.15, 2.47)

1.62 (1.09, 2.40)*

No BML (TF or PF)

335

59 (17.6%)

reference

reference

MT

180

51 (28.3%)

1.22 (0.84, 1.77)

1.22 (0.82, 1.81)*

No MT

670

163 (24.3%)

reference

reference

ME

117

39 (33.3%)

1.58 (1.04, 2.41)

1.67 (1.08, 2.59)*

No ME

733

175 (23.9%)

reference

reference

n ⫽ 850 persons K/L 0 in both knees [mean age 59.6 years (8.8, SD), BMI 26.7 kg/m2 (4.2), 475 (55.9%) women]. TF ⫽ tibiofemoral, PF ⫽ patellofemoral) * BMI also significant

Table 2. Pre-radiographic MRI Lesions and Risk of Incident Persistent Knee Symptoms

Number of knees (1 knee per person)

Number of knees (row%) with incident persistent knee symptoms

OR (95% CI) Unadjusted

OR (95% CI) Adjusted for age, gender, BMI, previous knee injury, previous knee surgery

Cartilage damage (TF or PF)

423

86 (20.3%)

1.94 (1.13, 3.35)

1.72 (0.98, 3.02)*

Cartilage damage (TF only)

84

13 (15.5%)

1.39 (0.65, 3.01)

1.17 (0.51, 2.73)*

Cartilage damage (PF only)

178

38 (21.3%)

2.07 (1.12, 3.80)

1.98 (1.04, 3.75)*

Cartilage damage (TF and PF)

161

35 (21.7%)

2.11 (1.14, 3.92)

2.00 (1.04, 3.84)

No cartilage damage (TF or PF)

155

18 (11.6%)

reference

reference

BML (TF or PF)

337

75 (22.3%)

2.09 (1.31, 3.33)

1.90 (1.18, 3.06)*

BML (TF only)

48

9 (18.8%)

1.69 (0.74, 3.84)

1.42 (0.58, 3.45)**

BML (PF only)

197

43 (21.8%)

2.04 (1.22, 3.42)

1.82 (1.07, 3.08)

BML (TF and PF)

92

23 (25.0%)

2.44 (1.32, 4.49)

2.31 (1.24, 4.31)

No BML (TF or PF)

241

29 (12.0%)

reference

reference

MT

109

30 (27.5%)

2.03 (1.24, 3.30)

1.97 (1.17, 3.33)*

No MT

469

74 (15.8%)

reference

reference

ME

61

18 (29.5%)

2.10 (1.16, 3.81)

1.69 (0.90, 3.18)*

No ME

517

86 (16.6%)

reference

reference

*BMI also significant **Previous surgery also significant n ⫽ 578 persons K/L 0 in both knees and without frequent knee symptoms at baseline [mean age 59.3 years (9.0), BMI 26.6 kg/m2 (4.3), 324 (56.1%) women].

Conclusion: In persons at higher risk but without any evidence of radiographic knee OA, cartilage damage, BML, ME, and BMI were associated with prevalent frequent knee symptoms, and cartilage damage, BML,

S1175

Tuesday, October 29

Background/Purpose: Knee pain is often the first sign of knee OA (osteoarthritis) and it is known that its course can be very different between patients over time. This study identifies distinct groups of patients with different trajectories of pain in symptomatic knee OA, and describes lifestyle and coping characteristics for each trajectory of pain. Lifestyle factors might be important elements for prevention, since they are modifiable in nature. Yet de role of lifestyle factors in different pain trajectories in early OA is unclear. Methods: Longitudinal data of five years follow-up of the CHECK (Cohort Hip and Cohort Knee) study was used. Participants had pain of knee, were aged 45–65 years, and had not yet consulted their physician for these symptoms or the consultation occurred within 6 months before inclusion. Pain severity was measured with numeric rating scale (0–10). Latent class growth analysis identified homogenous subgroups with distinct trajectories of pain. Multinomial regression analysis was used to examine different lifestyle and coping characteristics between the trajectories. Results: Longitudinal data of 5 years follow-up of 705 participants with symptomatic knee OA was analyzed. Three pain trajectories were identified based on their outcome: good, moderate and poor outcome. Participants with good outcome trajectory (n⫽222) had over time a slight decrease in pain severity and ended up with a low pain severity. Participants with moderate outcome trajectory (n⫽294) had a stable course of moderate pain over time. The poor outcome trajectory participants (n⫽189) had an increase of pain severity over time and ended up with severe pain. Compared to the good outcome group, participants in the moderate and poor outcome group were characterized by higher BMI (both OR’s 1.1; p⫽0.01), smoking (moderate outcome group OR⫽1.7, p⫽0.1; poor outcome group OR⫽ 2.3, p⫽0.02) by using passive coping strategies worrying (moderate outcome group OR⫽ 2.3, p⫽0.01; poor outcome group OR⫽3.7, p⬍0.001) and resting (moderate outcome group OR⫽ 1.5, p⫽0.1; poor outcome group OR⫽2.4, p⫽0.002). The passive coping strategy ‘retreating’ reduced the chance of belonging to the poor outcome group (moderate outcome group OR⫽ 0.6, p⫽0.05; poor outcome group OR⫽0.5, p⫽0.01). Baseline radiographic features of OA did not differ between the trajectories and did not have an association with the trajectories. However change from baseline to follow-up of radiological features (Kellgren & Lawrence grade and osteophyte area) differed between poor and good outcome. Conclusion: This study identified 3 trajectories of pain: good, moderate and poor outcome. Unhealthy lifestyle characteristics (high BMI and smoking) and passive coping strategies (worrying and resting) characterized the poor outcome group. The pain evolution of the participants in the poor outcome trajectory corresponds with an increase in radiological damage (Kellgren & Lawrence grade and osteophyte size). Distinguishing different trajectories could have implications for the treatment. Treatment for each pain trajectory in early OA might be tailored to lifestyle and coping characteristics.

Background/Purpose: The clinical significance of pre-radiographic MRI lesions in persons at risk for knee OA is unclear. Understanding whether such lesions are inconsequential or early disease will aid prevention and diseasemodifying strategy design. We hypothesized, in persons at risk but without radiographic OA, that cartilage damage, bone marrow lesions (BML), meniscal tear (MT), and meniscal extrusion (ME) are each associated with 1) prevalent frequent knee symptoms and 2) incident persistent knee symptoms. Methods: The Osteoarthritis Initiative is a cohort study of men and women age 45–79 years with or at increased risk to develop knee OA. 850 participants were bilateral K/L 0 at 12 months. In their 12-month right knee (left if right inadequate) MR images, we assessed cartilage, BML, MT, and ME using a modified MOAKS system, blinding readers to hypotheses and all other data. At baseline, of the 850, 578 were without frequent knee symptoms (knee symptoms or medication use for knee symptoms most days of a month in the past 12 months). Incident persistent knee symptoms were defined as frequent knee symptoms at 2 consecutive annual OAI visits. Multiple logistic regression models evaluated associations between each lesion and 1) prevalent frequent symptoms and 2) incident persistent symptoms by 60 months, adjusting for age, gender, BMI, knee injury, and knee surgery. Results: As shown in Table 1, cartilage damage, BML, and ME were each associated with prevalent frequent knee symptoms. As shown in Table 2, cartilage damage and BML (particularly patellofemoral), and MT were each associated with incident persistent symptoms.

MT, and BMI with the new development of persistent symptoms by 5 years. These findings suggest that these lesions, in persons at higher risk for knee OA, are clinically significant and may represent early disease. Disclosure: L. Sharma, None; J. S. Chmiel, None; O. Almagor, None; D. D. Dunlop, None; M. C. Hochberg, Abbvie, Inc, 5, Bioiberica SA, 5, Eli Lilly and Company, 5, Genentech, Inc, 5, Iroko Pharmaceuticals, LLC, 5, Merck, Inc, 5, Novartis Pharma AG, 5, Pfizer, Inc, 5, Savient Pharmaceuticals, 5; C. Eaton, None; C. K. Kwoh, None; R. D. Jackson, NIH, 2; J. M. Bathon, None; A. Guermazi, Boston Imaging Core Lab, 1, Merck Serono, 5, Sanofi-Aventis Pharmaceutical, 5, TissueGene, 5; F. Roemer, BICL, LLC, 1, BICL, LLC, 3, Merck-Serono, 5; M. Crema, BICL, LLC, 1, BICL, LLC, 3; W. J. Mysiw, None; M. C. Nevitt, None.

2754

Tuesday, October 29

Trajectory Of Cartilage Thickness Change During The Years Prior To Knee Replacement: Data From The Osteoarthritis Initiative. Felix Eckstein1, Robert M. Boudreau2, Zhijie Wang3, Michael J. Hannon3, Wolfgang Wirth1, Sebastian Cotofana1, Ali Guermazi4, Frank Roemer5, Michael C. Nevitt6, Markus John7, Christoph Ladel8, David J. Hunter9 and C. Kent Kwoh2. 1Paracelsus Medical University, Salzburg, Austria, 2University of Pittsburgh, Pittsburgh, PA, 3University of Pittsburgh School of Medicine, Pittsburgh, PA, 4Boston University, Boston, MA, 5Klinikum Augsburg, Augsburg, Germany, 6University of California, San Francisco, San Francisco, CA, 7Novartis Pharma AG, Basel, Switzerland, 8Merck KG, Darmstadt, Germany, 9University of Sydney, Sydney, Australia. Background/Purpose: Knee replacement (KR) represents a clinically important and cost-effective endpoint of knee osteoarthritis (OA). We have shown previously that, in a year prior to KR, medial cartilage loss (measured quantitatively with MRI) was greater in knees with KR than in controls, who displayed the same (baseline) radiographic disease stage. The best discrimination was observed in the central medial tibia (cMT). However, the trajectory of the more long-term cartilage thickness change prior to KR is unknown. The purpose of this study therefore was to examine rates of cartilage loss up to 4 years prior to KR. Methods: We studied knees from the Osteoarthritis Initiative (OAI) who received a KR between 12–60 month follow-up (12M–60M). Each knee with KR was matched with one control who did not receive a KR through 60M by sex, baseline KLG (0–1, 2, 3, 4 strata), and age (⫾5y). Sagittal 3T DESSwe MR images were used for segmenting the weight-bearing femoro-tibial cartilages (Chondrometrics GmbH), with blinding to acquisition order. The visit directly prior to KR (T0) and those 12, 24, 36, and 48 months prior to T0 (i.e. T-1 through T-4) were analyzed as available. Cartilage thickness in cMT was selected as the primary and that in the total medial compartment (MFTC) as the secondary endpoint. Comparisons included paired t-tests between case/control pairs, and conditional logistic regression model, with and without adjustment for pain and BMI at the beginning of the respective observation interval. Results: 220 knees of 190 OAI participants received a KR up to 60M (37@24M, 60@36M, 58@48M, and 65 @60M). Of these, 189 (58 baseline KLG 0–2, 69 KLG3, and 62 KLG4) had MRIs for at least one longitudinal interval, and a matched control (age 64⫾8.7; BMI 29⫾4.5; 85 male; 104 female). Analysis of slopes of annual change from all available time points revealed significantly greater rates of cMT cartilage loss in KRs than in controls (n⫽189: 94⫾137 vs. 55⫾104␮m; p⫽0.0017; odds ratio [OR] ⫽ 1.36 [95% CI: 1.11;1.66] without, and 1.34 [1.09;1.66] with adjustment for pain and BMI). Between T-1 and T0, cMT cartilage loss was 2.9⫻ that in controls (n⫽152: 114⫾209 vs. 39⫾159␮m; p⫽0.0007, OR 1.42 [1.14;1.75] without, and 1.48 [1.18;1.86] with adjustment). The difference was less in previous time intervals (ⱕ1.7x), but stronger for the T-2 to T0 period (n⫽127: 209⫾281 vs. 61⫾156␮m [3.4x]; p⬍0.0001, OR⫽1.61 [1.29;2.01] without, and 1.64 [1.30;2.06] with adjustment). However, no significant difference was observed during T-4 to T-2 (n⫽60: 119⫾255 vs. 125⫾175␮m; p⫽0.86, OR⫽0.97 [0.71;1.33] without, and 1.21 [0.81;1.81] with adjustment). Results for MFTC were similar to those for cMT. Conclusion: The rate of medial cartilage loss was substantially greater in knees with KRs than in those without during a two-year period prior to T0 (i.e. within 2 years proximate of the timing of the KR), even after matching for baseline radiographic disease status. During earlier time intervals (i.e. 4 to 2 years prior to T0) the rates of medial cartilage loss were not different between KRs and controls. This trajectory of cartilage thickness change resembles that observed for change in symptoms prior to KR. Disclosure: F. Eckstein, Chondrometrics, 3, Chondrometrics, 4, Merck Serono, 5, Abbvie, 5, Novartis Pharmaceutical Corporation, 2; R. M. Boudreau, None; Z. Wang, None; M. J. Hannon, None; W. Wirth, Chondrometrics, 4, Chondrometrics, 3; S.

Cotofana, Chondrometrics GmbH, 3; A. Guermazi, Boston Imaging Core Lab, 4, Merck Serono, 5, Sanofi-Aventis Pharmaceutical, 5, TissueGene, 5, Boston Core Imaging Lab BICL), 3; F. Roemer, Boston Core Imaging Lab BICL), 4, Boston Core Imaging Lab BICL), 3, Merck Serono, 5; M. C. Nevitt, None; M. John, Novartis, 1, Novartis, 3; C. Ladel, Merck Pharmaceuticals, 3; D. J. Hunter, None; C. K. Kwoh, Novartis Pharmaceutical Corporation, 5.

2755 The Clinical Significance and Natural History Of Knee Bone Marrow Lesions Over 8 Years. Yi Chao Foong1, Hussain Ijaz Khan1, Dawn Aitken1, Changhai Ding1, Flavia Cicuttini2 and Graeme Jones1. 1Menzies Research Institute Tasmania, University of Tasmania, Hobart, 7000, Australia, 2Department of Epidemiology and Preventive Medicine, Monash University, Melbourne, 3004, Australia. Background/Purpose: There is increasing evidence to suggest that bone marrow lesions (BMLs) play a key role in the pathogenesis of OA. Whilst many studies have focused on short and medium term changes in BMLs, no study has explored BMLs over longer timeframes. The aim of this study was to describe the natural history of knee BMLs and their association with knee pain over eight years. Methods: 199 subjects which consisted of 109 adult offspring of subjects who had a knee replacement and 90 controls who were initially matched by age and sex were studied. BMLs were measured at the two and ten year follow-ups using T2 weighted fat saturation MRI measuring the maximum area of the lesion at the medial and lateral tibial/femoral and patella sites. Knee pain was assessed by the Western Ontario and McMaster Universities Arthritis Index (WOMAC). Multiple linear and logistic regression models were performed to investigate potential associations between BMLs and pain. Results: At the two year follow-up, 64% of participants (n⫽128/199) had one or more BMLs, with a total of 231 BMLs present. Over eight years, of the 231 BMLs, 71% (164/231) increased in size, 8% (18/231) remained stable and 21% (49/231) decreased in size or resolved completely (Figure 1). Of those participants with no BMLs at baseline (71/199), 52% (37/71) developed one or more incident BMLs. In the whole sample, 31% of participants (62/199) developed a new BML at 1 site, 22% (44/199) at 2 sites, 9% (18/199) at 3 sites and 1% (2/199) at 4 sites or more. BML natural history did not vary between offspring and controls. After adjusting for age, sex and BMI, the development of a new BML was associated with developing pain in those without pain at baseline (␤⫽3.71, 95% CI 1.02–6.41). Those with a large BML at two years also had greater odds of having pain at ten years (OR⫽1.95, 95% CI 1.05–3.62). Eight year change in total BML size predicted an increase in total WOMAC pain in offspring but not controls (Table 1). The magnitude of the association in offspring was stronger in males as compared to females. Table 1. Change in WOMAC pain score per change in BML size (linear regression) Univariable ␤ (95% CI) Total Controls Offspring

0.46 (⫺0.17, 1.08) ⫺0.09 (⫺0.71, 0.53) 2.68 (1.22, 4.13)

Multivariable* ␤ (95% CI) 0.47 (⫺0.16, 1.11) ⫺0.04 (⫺0.69, 0.60) 2.65 (1.20, 4.10)

Females* ␤ (95% CI) 0.14 (⫺0.58, 0.87) ⫺0.14 (⫺0.93, 0.66) 2.22 (0.10, 4.34)

Males* ␤ (95% CI) 2.49 (0.86, 4.12) 1.56 (⫺1.10, 4.23) 3.17 (1.04, 5.31)

*Adjusted for age, sex, BMI. ␤coefficients are the change in pain score per unit change in BML size (cm2).

Figure 1. Natural history of BMLs

Conclusion: Over eight years, the proportion of BMLs increasing in size was more than triple those decreasing in size and stable BMLs were rare.

S1176

Incident BMLs were common, with slightly over half of our participants developing new BMLs, whilst one-fifth of BMLs resolved. Change in BML size was associated with pain only in those with a strong family history of OA, suggesting that knee pain associated with BMLs may have a genetic component. Disclosure: Y. C. Foong, None; H. I. Khan, None; D. Aitken, None; C. Ding, None; F. Cicuttini, None; G. Jones, None.

2756 Serum Periostin Is Associated With Prevalent Knee Osteoarthritis and Predicts Disease Progression In Women: The Ofely Study. Jean-Charles Rousseau1, Elisabeth Sornay-Rendu1, Cindy Bertholon1, Patrick Garnero2 and Roland Chapurlat3. 1INSERM UMR 1033, Lyon, France, 2INSERM, UMR 1033, Lyon and Cisbio Bioassays, Bagnols/ Ce`ze, France, 3INSERM UMR 1033 and Universite´ de Lyon, Hoˆpital Edouard Herriot, Lyon, France.

Disclosure: P. Saxena, None; R. F. Loeser, None; W. Pathmasiri, None; S. Sumner, None; D. Beavers, None; D. J. Hunter, None; S. P. Messier, None.

ACR Concurrent Abstract Session Rheumatoid Arthritis - Clinical Aspects V: Observational Studies in Rheumatoid Arthritis Tuesday, October 29, 2013, 4:30 PM–6:00 PM

Disclosure: J. C. Rousseau, None; E. Sornay-Rendu, None; C. Bertholon, None; P. Garnero, None; R. Chapurlat, None.

2757

2758

Metabolomics For The Identification Of Urinary Biomarkers In Knee Osteoarthritis Progression. Puja Saxena1, Richard F. Loeser2, Wimal Pathmasiri3, Susan Sumner3, Daniel Beavers1, David J. Hunter4 and Stephen P. Messier5. 1Wake Forest University School of Medicine, Winston-Salem, NC, 2Wake Forest School of Medicine, Winston-Salem, NC, 3NIH Eastern Regional Comprehensive Metabolomics Resource Core, Research Triangle Park, NC, 4University of Sydney, Sydney, Australia, 5Wake Forest University, Winston-Salem, NC.

Quantifying The Risk Of Incident Type II Diabetes Following Oral Glucocorticoid Therapy In Patients With Rheumatoid Arthritis: Association With Dose and Duration Of Use. William G. Dixon1, Mohammad Movahedi1, Marie-Eve Beauchamp2, David W. Ray3 and Michal Abrahamowicz2. 1The University of Manchester, Manchester, United Kingdom, 2McGill University, Montreal, QC, 3Institute of Human Development, The University of Manchester, Manchester, United Kingdom.

Background/Purpose: Obesity is an important risk factor for knee OA, not only due to effects of excess adipose tissue on joint loading, but also due to potential metabolic effects. It is not known if metabolic differences in people with OA might contribute to OA progression. The purpose of this study was to use a metabolomics approach to determine if a metabolic signature could distinguish overweight and obese adults with knee OA who exhibited radiographic progression during an 18 month exercise and/or weight loss intervention from those who did not progress. Methods: Urine samples collected from 24 overweight or obese participants in the Intensive Diet and Exercise for Arthritis (IDEA) trial at baseline and at 18 months of follow-up were selected from two subgroups

Background/Purpose: Glucocorticoid (GC) therapy is widely used in patients with rheumatoid arthritis (RA). Its association with incident type II diabetes mellitus (DM) is accepted, but the extent of the risk and its relationship with GC dose and treatment duration in RA are not known. Methods: Adult patients with RA were identified from a large UK primary care research database using a validated algorithm during the study period 1992–2009. Patients with prevalent DM at the time of their first code for RA were excluded. GC exposure was identified from oral GC prescriptions and was analysed using multivariable time-dependent Cox models with conventional exposure measures (eg on/off treatment) and a novel weighted cumulative dose (WCD) model that accounted for

S1177

Tuesday, October 29

Background/Purpose: Periostin (POSTN) is a secreted vitamin K-dependent (Gla-containing) protein produced by osteoblasts and chondrocytes. The aim of this study was to investigate the relationships between serum POSTN and both prevalent and incident knee osteoarthritis in women. Methods: We investigated 753 women (mean age: 62.7 ⫾ 11.2 yr) from the OFELY (Os des Femmes de Lyon) cohort, 19.3 % being premenopausal. Knee radiographs were performed and scored according to Kellgren & Lawrence at the ninth annual follow-up (baseline for this analysis) and 4 years later. Progression of knee OA was defined as an increase of the KL score ⱖ 1 during the 4 years follow-up. Evaluation of the spine (X-rays), hip (self-reported) and hand OA (clinical exam) have been performed at the same visit. Serum POSTN was measured at baseline by ELISA (USCNK, China). Results: Serum POSTN was significantly lower in the 110 women with a KL score ⱖ 2 at baseline, compared to the 481 controls with a KL score ⬍ 2 (1118.9 ⫾ 307 ng/ml vs 1182.3 ⫾ 292 ng/ml, p ⫽ 0.018 after adjustment for age). During the 4 year follow-up, 181 women had a radiological progression in their knee OA. Baseline POSTN levels were significantly lower in progressors than in non progressors (1140 ⫾ 296 vs 1182 ⫾ 294 ng/ml, p ⫽ 0.039 after adjustment for age and for the presence of prevalent knee, spine, hip and hand OA). For each increase of one POSTN quartile, the risk of progression decreased by 0.82 (IC: 0.69–0.98, p ⫽ 0.034) after adjustment for age and for OA at the other anatomical sites. Conclusion: We showed for the first time that serum POSTN is associated with prevalence and progression of knee OA in women. This biomarker may be of interest to study OA pathophysiology.

(n⫽12 each): a group that exhibited radiographic progression of knee OA and an age, race, sex, and BMI matched group who did not progress. Progression was defined as a decrease in medial joint space width of ⱖ 0.7mm from baseline to 18 months (knees flexed at a 15° angle using a positioning device and the x-ray beam centered on the joint space). Non-progressors were defined as individuals who had a decrease in joint space width of ⱕ0.35mm. Morning second void urine samples were collected at baseline and 18 months. After centrifuging and addition of internal standards, 1H NMR spectra were acquired using a 950 MHz NMR spectrometer for each study sample and for a quality control sample prepared by pooling individual samples NMR data was processed by automated integration over the spectral window, and bins were normalized to the total intensity for each spectrum. Principal component analysis (PCA) and orthogonal partial least square discriminate analysis (OPLS-DA) were used to reduce the dimensionality and enable visualization of the separation of progressors and non-progressors. NMR bins that distinguished the study phenotypes were determined based on inspection of loadings plots and variable importance plots. Bins that distinguished the study groups were mapped to corresponding metabolites. Results: The median age of this IDEA subgroup was 67 yrs. (range 60–78) and median BMI was 31.2 kg/m2 (range 27.0–40.4). The mean minimum JSW change was ⫺1.03 (SD⫽0.29, range ⫺1.7 to ⫺0.76) mm for progressors and 0.05 (SD⫽0.28 range ⫺0.26 to ⫹0.68)mm for non-progressors. After applying PCA, two outliers (non-progressors) with high levels of urine glucose were excluded. Using OPLS-DA, five urinary metabolites were determined that separated the progressors from non-progressors. At baseline, relatively higher levels of hippurate and glycolate and lower levels of lysine were identified in progressors. At 18 month follow-up, relatively higher levels of hippurate, glycolate, and sarcosine and lower levels of lysine and choline characterized the progressor group. Conclusion: Metabolic differences were found for the first time to correlate with radiographic progression in overweight and obese adults with knee OA. These results merit further investigation in a larger sample set. Given that hippurate is a gut-flora derived metabolite, differences in the gut microbiome could be contributing to OA progression.

Tuesday, October 29

past doses and treatment duration and timing. Incident DM was defined as a READ code for type II DM, at least two oral anti-diabetic prescriptions or abnormal blood results (blood sugar, HbA1C or glucose tolerance test). Hazard ratios (HR) associated with different GC patterns were adjusted for gender, age, family history of DM, hypertension, prior cumulative dose of oral GC, current DMARDs and ever NSAID use. Results: 22,535 adult RA patients were included. 70% were female with a median age of 60 years (IQR 49–71). Median follow-up time per patient was 5.4 years. 2,324 patients were diagnosed with type II DM during follow-up: incidence 13.7 events/1000 person years (pyrs) in unexposed patients versus 21.8 events/1000pyrs after GC exposure. Ignoring dose and duration, the conventional HR was 1.42 (1.30–1.56) in ever GC users compared with non-users. This equates to one additional case of DM per year for every 172 patients who received GCs at some time in the past. The WCD model showed risk increased with increased dose and duration of recent use, especially in the past 8 months. 5mg prednisolone equivalent (PEQ) for past 1, 3 and 6 months were associated with HRs of 1.14, 1.36 and 1.46, respectively, compared to non-use. Doses more than eight months ago added little to the current risk of DM, e.g. increasing the duration of 5mg PEQ exposure from the past 6 to 12 months increased the HR to only 1.49 compared to non-use. Risk increased significantly with dose: 30mg PEQ for 1, 3 or 6 months generated HRs of 2.2, 6.3 and 9.8, respectively, compared to non-use. 5mg or 30mg PEQ for 6 months equates to one additional case for every 140 or 7 patients treated, respectively. Conclusion: Oral GC therapy is a clinically important and quantifiable risk factor for incident Type II DM in patients with RA, influenced by dose and duration of exposure within the last eight months. Disclosure: W. G. Dixon, None; M. Movahedi, None; M. E. Beauchamp, None; D. W. Ray, None; M. Abrahamowicz, None.

Results: The primary cohorts included 57,251 RA patients (232,985 person-years [PY]) - mean (SD) baseline age 58.2 (13.8) years, 24.5% men, and RA duration 8.2 (11.7) years (Table 1). Some important demographic differences appeared; for example, smoking was most common in NOAR, and SRR had the highest proportion with early RA. Mean baseline HAQ and DAS28 were 0.88 (0.67) and 4.1 (1.5). Specific outcome event rates were determined in each registry (Table 2). CVD rates varied the most with the lowest MI rate (0.09 per 100 PY) seen in IORRA and the highest (0.39) in SRR. Rates of hospitalized infections were quite consistent across registries. Sensitivity analysis returned higher mortality and MACE rates along with wider confidence intervals when standardized also for RCT HAQ levels. Additional sub-cohort analyses showed small and mostly consistent changes among registries. Table 1. Baseline characteristics in 5 RA registries’ primary cohorts. Baseline characteristic Region N Age, ⬎60 (%) Gender, female (%) Smoking, ever (%) RA duration, ⬎5 yrs (%) SJC, mean (SD) TJC, mean (SD) CRP, ⬍10 mg/L (%) DAS28, ⬎3.2 (%) HAQ, ⬎1.1 (%) RF⫹ (%) MTX from baseline (%) Other nb-DMARDs from baseline (%)

CORRONA USA 24,176 47.3 76.0 38.1 56.8 4.3 (5.5) 4.2 (5.8) 72.4 58.3 35.0 72.4 64.6 30.1

SRR NOAR West Europe 18,527 1,564 54.1 51.0 70.3 69.9 NA 66.9 29.6 44.9 6.3 (5.7) 5.8 (5.1) 5.8 (5.8) 7.7 (8.3) 47.3 37.8 75.9 75.0 39.2 54.5 71.7 89.6 71.5 33.8 22.4 24.2

CORRONA Int. IORRA Other 2,727 10,255 34.7 41.0 84.7 82.2 26.5 37.7 61.2 51.5 2.9 (4.3) 2.8 (3.4) 5.7 (6.7) 2.8 (3.8) 64.9 62.8 67.4 70.5 49.3 30.3 75.3 78.8 68.9 74.7 52.8 76.2

Table 2. Standardized incidence rates (with 95% confidence intervals)* per 100 person-years of selected outcome events in 5 RA registries’ primary cohorts.

2759 Can RA Registries Provide Contextual Safety Data For Modern RCTs? Kaleb Michaud1, Johan Askling2, Hisashi Yamanaka3, Deborah Symmons4, Marie Holmqvist2, Thomas Frisell2, George Reed5, Dimitrios A. Pappas6, Eiichi Tanaka3, Eisuke Inoue3, Suzanne M.M. Verstappen4, Christopher Garwood4, Laura Horne7, Kathy Lampl7, Niklas Berglind8, Stefan Franzen8, Fredrik Nyberg9, Trung Tran10, Meilien Ho11 and Jeffrey D. Greenberg12. 1National Data Bank for Rheumatic Diseases & University of Nebraska Medical Center, Omaha, NE, 2Karolinska Institutet, Stockholm, Sweden, 3Tokyo Women’s Medical University, Tokyo, Japan, 4 Arthritis Research UK Epidemiology Unit, The University of Manchester, Manchester, United Kingdom, 5University of Massachusetts Medical School, Worcester, MA, 6Columbia University, New York, NY, 7 AstraZeneca R&D Wilmington, Wilmington, DE, 8AstraZeneca R&D Mo¨lndal, Mo¨lndal, Sweden, 9AstraZeneca R&D, Mo¨lndal, Sweden, 10 MedImmune LLC, Gaithersburg, MD, 11AstraZeneca R&D Alderley Park, Macclesfield, United Kingdom, 12NYU Hospital for Joint Diseases, New York, NY. Background/Purpose: For ethical reasons, modern phase-III RCTs in rheumatoid arthritis (RA) have limited placebo exposure time, resulting in uncertainties when interpreting low frequency adverse events (AEs). We developed and implemented a novel method for providing contextual AE rates through coordinated analyses of 5 RA registries. Methods: Participating observational registries included CORRONA (USA), CORRONA International (East Europe, India, Latin America), IORRA (Japan), NOAR (UK), and SRR (Sweden). By harmonizing AE definitions in a typical RA clinical trial program and corresponding available outcomes from the registries we defined specific outcomes in 4 areas: mortality, cardiovascular disease (CVD), infection, and malignancy. Each registry defined a primary cohort (all patients included from January 2000) for optimal sample size. To address comparability and potential bias several sub-cohorts were also defined for sensitivity analyses, based on disease activity, treatment, calendar time or duration of follow-up, and relevant RCT exclusions. Rates were standardized for the potential confounders age and sex, and in a sensitivity analysis also HAQ score to the distribution in a typical RA program.

CORRONA

SRR

NOAR

CORRONAINT

IORRA

All-cause mortality

0.42 (0.38, 0.46)

0.67 (0.63, 073)

0.80 (0.66, 0.98)

NA

0.74 (0.67, 0.81)

Major adverse cardiovascular events (MACE)

0.45 (0.40, 0.50)

0.77 (0.71, 0.83)

0.50 (0.39, 0.66)

0.36 (0.15, 0.81)

0.31 (0.26, 0.37)

Acute myocardial infarction (MI)

0.21 (0.18, 0.25)

0.39 (0.35, 0.43)

0.21 (0.14, 0.33)

NA

0.09 (0.07, 0.13)

Stroke

0.20 (0.17, 0.24)

0.31 (0.27, 0.35)

0.16 (0.10, 0.27)

NA

0.12 (0.09, 0.16)

Hospitalized congestive heart failure (CHF)

0.20 (0.11, 0.38)

0.16 (0.14, 0.19)

0.13 (0.07, 0.24)

0.42 (0.16, 0.92)

NA

Hospitalized infection

1.30 (1.18, 1.42)

1.62 (1.52, 1.72)

1.56 (1.30, 1.88)

1.51 (0.99, 2.24)

1.14 (1.05, 1.25)

Malignancies excluding non-melanoma skin cancer (NMSC)

0.64 (0.58, 0.71)

0.87 (0.80, 0.94)

0.77 (0.60, 0.99)

0.56 (0.27, 1.07)

0.65 (0.57, 0.75)

Event

Solid malignancies

0.53 (0.47, 0.58)

0.86 (0.80, 0.94)

0.88 (0.70, 1.10)

0.37 (0.16, 0.81)

0.58 (0.49, 0.67)

Lymphomas

0.06 (0.04, 0.08)

0.06 (0.04, 0.08)

0.09 (0.03, 0.21)

NA

0.06 (0.04, 0.10)

* Standardized according to the age and sex distribution in a typical RA clinical trial program. ⫽ Not Available. Too few events - incidence rates not calculated

Conclusion: While potential concern for bias may remain, we have successfully determined background safety rates that provide context for RCTs, through a coordinated approach between observational RA registries. This new approach may have utility to satisfy future regulatory requirements and support safety assessments. Disclosure: K. Michaud, University of Nebraska Medical Center, 3, National Data Bank for Rheumatic Diseases, 3; J. Askling, AstraZeneca, 2, AstraZeneca, 5; H. Yamanaka, AstraZeneca, Abbott, AbbVie, Chugai, Takeda, Pfizer, Daiichi Sankyo, Mitsubishi Tanabe, Teijin, Nippon Kayaku, Taishotoyama, BristolMyers Squibb, Astellas, Eisai, MSD, Santen, GlaxoSmithKline, Asahikasei, Janssen, 2, AstraZeneca, Abbott, AbbVie, Chugai, Takeda, Pfizer, Daiichi Sankyo, Mitsubishi Tanabe, Teijin, Nippon Kayaku, Bristol-Myers Squibb, Astellas, Eisai, 5, Abbott, AbbVie, Chugai, Takeda, Pfizer, Mitsubishi Tanabe, Teijin, Bristol-Myers Squibb, Astellas, Eisai, 8; D. Symmons, AstraZeneca, 2, AstraZeneca, 5; M. Holmqvist, None; T. Frisell, None; G. Reed, CORRONA, Inc., 3; D. A. Pappas, CORRONA Inc, 3, Novartis Pharmaceutical Corporation, 5, Columbia University, 6; E. Tanaka, None; E. Inoue, None; S. M. M. Verstappen, None; C. Garwood, None; L. Horne, AstraZeneca, 3, AstraZeneca, 1; K. Lampl, AstraZeneca, 3, AstraZeneca, 1; N. Berglind, AstraZeneca, Bristol-Myers Squibb, 1, AstraZeneca, 3; S. Franzen, AstraZeneca, 1, AstraZeneca, 3; F. Nyberg, AstraZeneca, 1, AstraZeneca, 3; T. Tran, MedImmune LLC, 3; M. Ho, AstraZeneca, 3; J. D. Greenberg, AstraZeneca, 5, Corrona, 5, Pfizer Inc, 5, Corrona, 1.

S1178

2760 Incidence and Severity Of Myocardial Infarction In Subjects Receiving Anti Tumour Necrosis Factor Drugs For Rheumatoid Arthritis: Results From Linking the British Society For Rheumatology Biologics Register For Rheumatoid Arthritis and Myocardial Ischaemia National Audit Project. Audrey SL Low1, Kimme L. Hyrich2, Mark Lunt3, Louise K. Mercer1, Christopher Gale4, Kath Watson1, British Society for Rheumatology Biologics Registers (BSRBR) Control Centre Consortium1, William G. Dixon5, Deborah P. Symmons3 and On behalf of the BSRBR6. 1Arthritis Research UK Epidemiology Unit, The University of Manchester, Manchester, United Kingdom, 2Arthritis Research UK Epidemiology Unit, Manchester Academic Health Sciences Centre, Institute of Inflammation and Repair, The University of Manchester, Manchester, United Kingdom, 3Arthritis Research UK Epidemiology Unit, University of Manchester, Manchester, United Kingdom, 4 Department of Cardiology, York Teaching Hospital NHS Foundation Trust, York, United Kingdom, 5The University of Manchester, Manchester, United Kingdom, 6British Society for Rheumatology, London, United Kingdom.

Table.

Years of follow-up per subject, median (IQR) Person-years of exposure, pyrs Number of verified MIs Crude incidence rate of verified MIs from BSRBR & MINAP per 10,000 personyears (95% CI) Risk of MI between nbDMARD and TNFitreated subjects: Unadjusted HR (95%CI) Risk of MI between nbDMARD and TNFitreated subjects: PD-adjusted HR (95% CI) Number of verified MIs with additional MINAP data Phenotype of MI: n (%) of STEMI Median peak CK, IU/L (IQR) Cardiac arrest, n (%) Median length of hospital stay, days (IQR)

nbDMARD (nⴝ3058) 3.5 (1.8, 4.9)

TNFi (nⴝ11200) 5.3 (3.6, 6.4)

p-value ⫺

10337 58 56 (43, 73)

55636 194 35 (30, 40)

⫺ – ⫺

Referent

0.78 (0.58, 1.05)

⫺

Referent

0.61 (0.41, 0.89)

–

35

108

–

16 (46) 290 (172, 1598) 3 (8.6) 6 (5, 9)

53 (49) 691 (150, 1293) 5 (4.7) 6 (4, 8)

0.32 0.19 0.48 0.46

Variables in PD: age, gender, disease duration, DAS28, HAQ score, steroid use, number of previous nbDMARDs, entry year to study, hypertension, diabetes, smoking, chronic lung disease, aspirin, statin, NSAID/COX2-inhibitor use, all at baseline.

Disclosure: A. S. Low, None; K. L. Hyrich, None; M. Lunt, None; L. K. Mercer, None; C. Gale, None; K. Watson, None; B. S. F. R. B. R. Control Centre Consortium, None; W. G. Dixon, None; D. P. Symmons, None; O. B. O. T. BSRBR, Abbvie, 2, Amgen, 2, Swedish Orphan Biovitrum SOBI), 2, Merck Pharmaceuticals, 2, Pfizer Inc, 2, Roche Pharmaceuticals, 2, UCB Pharma Ltd, 2.

2761 Comparative Risks Of Herpes Zoster Among RA Patients Switching Biologics In The U.S. Medicare Program. Huifeng Yun1, Fenglong Xie1, Elizabeth S. Delzell1, Lang Chen1, Emily Levitan1, James Lewis2, Kenneth G. Saag1, Timothy Beukelman1, Kevin L. Winthrop3, John Baddley1 and Jeffrey R. Curtis1. 1University of Alabama at Birmingham, Birmingham, AL, 2University of Pennsylvania, Philadelphia, PA, 3Oregon Health & Science University, Portland, OR. Background/Purpose: Several newer biologics have been approved for treatment of rheumatoid arthritis (RA) in the United States. However, their comparative risks of herpes zoster infection are not well understood. The objective of the present study is to evaluate whether the risks of herpes zoster infections associated with various biologics differ among Medicare RA population. Methods: Using Medicare data from 2006–2011 for patients with RA, we identified new users of abatacept, adalimumab, certolizumab, etanercept, golimumab, infliximab, rituximab and tocilizumab. New users were defined specific to each drug as no use of that therapy in the prior 12 month ‘baseline’ period. To increase homogeneity of patients characteristics to allow comparison with biologics typically not used as first line agents, patients were required to have used another biologic during baseline (i.e. they were ‘switchers’). Eligible subjects must have been continuously enrolled in Medicare Parts A, B and D in baseline and throughout follow up and could not have had zoster medications during baseline. Follow up started from the drug initiation date and ended at the earliest date of: zoster infection, a 30 day gap in current exposure, death, loss of Medicare coverage or Dec 31, 2011. We identified herpes zoster infection using physician diagnosis code and use of antiviral medication within 7 days. Confounding was controlled through a person-specific infection risk score that was separately derived among biologic-naı¨ve new users of anti-TNF and non-biologic DMARDs. We calculated the incidence rate of herpes zoster infection for each biologic and compared their herpes zoster infection risks during follow-up using Cox regression adjusting for infection risk score decile, disability status, glucocorticoids use during baseline, methotrexate use during baseline, most recent biologic during baseline and Medicaid eligibility. Results: Of 25,274 biologic switchers, 11.3% used etanercept, 15.9% adalimumab, 6.1% certolizumab, 4.4% golimumab, 12.6% infliximab, 29.0% abatacept, 14.5% rituximab and 6.3% tocilizumab. Among 24,237 patients who had no history of herpes zoster infection, we identified 336 herpes zoster infections yielding similar crude incidence rates across different biologics. After adjustment for potential cofounders and compared to infliximab users, the adjusted hazard ratios of all types of biologics were not significantly different from infliximab users. Among 1,037 patients with a history of herpes zoster infection during baseline, we identified 14 recurrent herpes zoster infections

S1179

Tuesday, October 29

Background/Purpose: Subjects with rheumatoid arthritis (RA) are at increased risk of myocardial infarction (MI) compared to subjects without RA, with the increased risk potentially driven by inflammation. Anti- tumour necrosis factor (TNFi) drugs may modulate the risk and severity of the MI. The aim of this analysis was to compare the risk and severity of MI in subjects treated with TNFi to non-biologic drugs (nbDMARDs) using linked data from BSRBR-RA (an ongoing prospective study of the safety of biologic therapy in RA) and the Myocardial Ischaemia National Audit Project (MINAP), a national database of hospitalisations with MI in England and Wales. Methods: This analysis included subjects with RA starting TNFi (etanercept, infliximab, adalimumab) and a biologic-naı¨ve comparator cohort receiving nbDMARDs recruited between 2001–2008; all followed via physician and patient questionnaires and flagged with the national death register. All subjects were also linked to MINAP using deterministic matching which provided both additional events and further event data regarding MI severity. Events from both datasets were verified using the American Heart Association/European Society of Cardiology criteria for MI with additional criteria of thrombolysis/ angioplasty and MI listed as the underlying cause of death on death certificates, coded using International Classification of Diseases 10. When estimating incidence of MI, subjects were censored at first MI, death, 90 days following discontinuation of TNFi, last physician follow-up or 04/20/2010, whichever came first. The risk of MI was compared between subjects receiving nbDMARDs and TNFi using a Cox regression model, adjusted on deciles of propensity scores (PD) (Table). In a subset of MIs with additional event data from MINAP; MI phenotype and severity (using surrogates: cardiac arrest, peak creatine kinase (CK) levels and length of hospital stay) were compared between treatment groups using descriptive statistics.

Results: Using the linked data, a total of 252 verified first MIs were analysed: 58 in 3058 nbDMARD subjects and 194 in 11200 TNFi-treated subjects. The PD-adjusted hazard ratio (HR) of MI in TNFi referent to nbDMARD was 0.61 (95%CI 0.41, 0.89). The estimate was unchanged when all follow-up time following first dose of TNFi was included. There were 143 MIs (nbDMARD: 35, TNFi: 108) with additional MINAP data. No statistical differences were observed in MI severity when compared between nbDMARD and TNFi-treated subjects, although there was a trend towards higher peak CK levels recorded in TNFi-treated subjects (Table). Conclusion: Subjects receiving TNFi drugs had a decreased risk of MI compared to subjects receiving nbDMARDs over the medium term which could relate to attributes of the drug itself or better overall disease control. The severity of MIs at presentation appears to be similar between both treatment groups.

yielding an incidence rate for recurrence of 2.1 per 100 person years. We did not have enough data to look at drug-specific associations with recurrent infection. Table. Events, absolute incidence rate and adjusted hazard rate of herpes zoster infection by different types of biologics Biologic Exposures Abatacept Adalimumab Certolizumab Etanercept Golimumab Rituximab Tocilizumab Infliximab

Events 118 40 14 34 ⬍11 60 15 45

Crude incidence rate per 100 py 1.81 1.69 2.03 1.86 1.74 1.99 2.11 1.69

Adjusted Hazard Ratio* (95% CI) 0.88 (0.61–1.28) 0.96 (0.62–1.49) 1.14 (0.64–2.04) 1.01 (0.65–1.58) 0.97 (0.49–1.94) 0.94 (0.63–1.41) 0.92 (0.50–1.69) 1.00 (Ref)

*Adjusted for age, gender, infection risk score decile, steroid use during baseline, recent biologic use during baseline, original reason for Medicare coverage and Medicaid eligibility.

Conclusion: Among patients with RA, the risk for herpes zoster was similar across biologics, include those with non-TNF mechanisms of action.

Results: 1502 RA patients, 359 PsA patients and 7887 NRC were included. Overall, the mean age of the cohort was 55 years; 70% were female, and 98% were Caucasian. RA patients on average were older, more likely to be female, taking steroids and immunosuppressive drugs, and had higher ESR and lower median BMI than PsA patients. Statin use was not different across all groups. During observation, there were 1195 incident diabetes cases: 199 in the RA, 47 in the PSA and 859 in the NRC, with IR 23.2, 23.0 and 17.7 respectively, with an incidence rate ratio (IRR) of 1.32 (1.13–1.54) and 1.30 (0.97–1.75) for RA and PsA respectively compared to NRC. The adjusted HR (95% CI) for DM was 1.14 (0.94–1.38) and 1.28 (0.94–1.75) for RA and PsA respectively compared to NRC. Defining DM as HgA1c ⱖ 6.5%, there were 708 incident cases: 122 in the RA, 25 in the PsA and 561 in the NRC, with IR 13.8, 11.8 and 11.2 respectively and IRR 1.05 (0.70–1.57) and 1.22 (1.01–1.49) for RA and PsA respectively compared to NRC. The HR (95% CI) was 1.06 (0.87–1.30) and 1.12 (0.82–1.52) for RA and PsA respectively compared to NRC. Conclusion: In this cohort, RA and PsA were not associated with increased risk of incident DM, defined by both clinical and stricter laboratory criteria. Our results suggest that an excessive burden of DM does not account for the high risk of CVD in these patients. Disclosure: S. Mathew, None; X. Tang, None; H. L. Kirchner, None; M. C. M. Wasko, None; A. Bili, None.

Tuesday, October 29

2763 Disclosure: H. Yun, None; F. Xie, None; E. S. Delzell, Amgen, 2; L. Chen, None; E. Levitan, Amgen, 2; J. Lewis, Pfizer, Prometheus, Lilly, Shire, Nestle, Janssen, AstraZeneca, Amgen, 5, Centocor, Shire, Takeda, 2; K. G. Saag, Ardea; Regeneron; Savient; Takeda, 5, Ardea; Regeneron; Savient: Takeda, 2; T. Beukelman, Genentech and Biogen IDEC Inc., 5, Novartis Pharmaceutical Corporation, 5, Pfizer Inc, 2; K. L. Winthrop, Pfizer Inc, 2, Genentech Inc., Pfizer, UCB, Regeneron, 5; J. Baddley, BMS, 2, Pfizer Inc, 2; J. R. Curtis, Roche/Genentech, UCB, janssen, CORRONA, Amgen, Pfizer, BMS, Crescendo, AbbVie, 2, Roche/Genentech, UCB, janssen, CORRONA, Amgen, Pfizer, BMS, Crescendo, AbbVie, 5.

2762 Rheumatoid and Psoriatic Arthritis Are Not Associated With Higher Risk Of Incident Diabetes Mellitus. Susan Mathew1, Xiaoqin Tang2, H. Lester Kirchner3, Mary Chester M. Wasko4 and Androniki Bili1. 1 Geisinger Medical Center, Danville, PA, 2Geisinger Center for Health Research, Danville, PA, 3Geisinger Health System, Danville, PA, 4Temple University School of Medicine, Pittsburgh, PA. Background/Purpose: Rheumatoid arthritis (RA) and psoriatic arthritis (PsA) are chronic inflammatory conditions that are associated with increased risk of insulin resistance and cardiovascular disease (CVD). However, the evidence for the association of these conditions with incident diabetes mellitus (DM) has been inconsistent. The purpose of this study was to evaluate this association in a cohort of patients with physician entered diagnosis and laboratory values. Methods: We studied a retrospective cohort of patients with RA or PsA from 1/1/2001 to 12/31/2012 with age- (⫾ 2.5 years) and gendermatched non-rheumatic controls (NRC) in a tertiary health center. Data were extracted from electronic health records. RA and PsA were defined using ICD-9 codes 714.0 and 696.0, respectively, entered twice by a rheumatologist; with a primary care physician within the health system were included. Patients with prevalent DM were excluded. Patients were followed over time for development of DM, defined by ICD-9 250, random blood glucose ⬎200 mg/dl, HgA1c ⱖ6.5% or use of anti-diabetic medications; a more stringent DM diagnosis, with DM defined as HgA1c ⱖ6.5%, was also examined. A Poisson regression model was used to estimate the incidence rate (IR) of DM per 1000 person-years for the RA and PsA groups separately compared to the NRC. Cox proportional hazard regression (HR) models, accounting for the correlation induced by matching, were used to estimate the association on developing incident DM after adjusting for age, gender, race, body mass index ( BMI), hypertension, hyperlipidemia, ESR, number of office visits, number of drug classes, glucocorticoid, immunosuppressive agent and statin use in the year prior to cohort entry. For the primary outcome, our study had 80% power to detect a 25% difference in the risk for incident diabetes between the RA/PsA and NRC groups.

Carotid Atherosclerosis As a Predictor Of Mortality In Rheumatoid Arthritis. Inmaculada del Rincon1, Roy W. Haas1, Jose Felix Restrepo1, Daniel F. Battafarano2, Daniel H. O’Leary3, Emily Molina4 and Agustin Escalante1. 1University of Texas Health Science Center, San Antonio, TX, 2Brooke Army Medical Ctr, San Antonio, TX, 3Tufts UniversityBoston Campus, Boston, MA, 4University of Texas Health Science Center at San Antonio, San Antonio, TX. Background/Purpose: Patients with rheumatoid arthritis (RA) have higher mortality than do persons of the same age and sex without RA. This is due in part to an increased risk of atherosclerosis and cardiovascular (CV) disease. The extent of atherosclerosis can be estimated noninvasively using high-resolution carotid ultrasound. In this study we examined the association of carotid atherosclerosis, as measured by carotid ultrasound, with mortality in a cohort of RA patients. Methods: We recruited RA patients during a visit to their rheumatologist, and invited them to participate in a comprehensive clinical assessment that included a high-resolution carotid ultrasound. We also ascertained age, sex, RA duration, the CV risk factors and erythrocyte sedimentation rate (ESR), among other variables, using a predetermined protocol. After the assessment, we followed patients prospectively per protocol. We obtained a death certificate for all patients who died, from which we classified the causes of death using ICD9 codes. Deaths were classified as CV if the death certificate listed a CV condition, corresponding to ICD9 codes 390–459. We used standard time-to-event techniques to examine the association between the carotid ultrasound findings and all-cause and CV-mortality. Mortality rates are shown per 100 personyears. Rates and hazard ratios (HR) are shown with 95% confidence intervals. Results: We recruited 1,328 RA patients, of whom 1,197 had a carotid ultrasound. We followed them prospectively over 6,500 person-years, during which 206 deaths occurred, for a mortality rate of 3.1 (2.7, 3.6). Death was attributed to a CV cause in 105 cases, for a CV mortality rate of 1.7 (1.4, 2.1). The presence of carotid plaque resulted in an increased all-cause mortality rate, 4.6 (3.9, 5.47), compared to 1.6 (1.2, 2.1) with absent plaque. The Figure shows a Kaplan Meier curve for all-cause mortality in patients with and without plaque (P ⬍ 0.0001, log rank test). Carotid plaque was also associated with an increased CV mortality rate, 2.7 (2.1, 3.4) vs. 0.8 (0.6, 1.2). The carotid intimamedia thickness (IMT) was associated with increased all-cause mortality, with a HR of 2.28 per mm of IMT (2.00, 2.59), as well as CV mortality, with a HR of 2.26 per mm IMT (1.91, 2.68). The associations of both carotid plaque and the IMT with all-cause mortality were independent of age, sex, CV risk factors and the ESR. Their associations with CV mortality were independent of the all same variables, with the exception of the ESR.

S1180

more patients in the T plus I group (42%) than in the T group (25%) (p⬍0.05). Radiographic non-progression at 52 weeks was achieved by more patients in the T plus I group (41%) than in the T group (18%) (p⬍0.05). Especially, swollen and tender joints counts at 52 weeks are significantly improved T plus I group(SJC: 3.6⫾1.3, TJC:4.2⫾2.1) than T group.(SJC: 8.5⫾4.2, TJC: 6.2⫾3.5)(p p⬍0.05). However C reactive protein at 52 is not significantly difference each group (T plus I group: 0.21⫾0.12 mg/L, T group: 0.26⫾0.22 mg/L) (p⫽0.67). Conclusion: Results of this reveal that combination of intra-articular steroid injection and tocilizumab can achieve a high clinical and radiological remission rate in early RRP RA.

Conclusion: Carotid atherosclerosis is significantly associated with mortality in RA, both all-cause and due to CV causes. The precise role of carotid ultrasound in mortality risk stratification and identification of candidates for intervention to reduce mortality in RA is an interesting area in need of study. Disclosure: I. del Rincon, None; R. W. Haas, None; J. F. Restrepo, None; D. F. Battafarano, None; D. H. O’Leary, None; E. Molina, None; A. Escalante, None.


2764 Combination of Intra-Articular Steroid Injection and Tocilizumab More Effective Than Tocilizumab in Rapid Radiographic Progression Patients With Rheumatoid Arthritis. A Randomized, Open Label, x Ray Reader Blinded Study. Kensuke Kume1, Kanzo Amano1, Susumu Yamada1, Toshikatsu Kanazawa2, Hiroshi Komori2, Hiroyuki Ohta3, Noriko Kuwaba4 and Kazuhiko Hatta5. 1Hiroshima Clinic, Hiroshima, Japan, 2Hiroshima Clinic, Hiroshima, Japan, 3Hiroshima Clinic, Hiroshima, Japan, 4Sanki Clinical Link, Hiroshima, Japan, 5Hatta Clinic, Kure, Japan. Background/Purpose: Treatment of rheumatoid arthritis (RA) should aim at full remission. However, recent publications described rapid radiographic progression (RRP) existed despite initial biologics and methotrexate combination therapy in early RA. In RRP, initial biologics and methotrexate might be inadequate. We reported that infliximab plus intra-articular steroid injection is more effective than infliximab in RRP patients in RA. How about other biologics? To compare remission and radiographic non-progression in RRP patients treated with tocilizumab or with tocilizumab plus intra-articular steroid injection. Methods: We designed a single-blind (X ray reader and assessment physician), randomized controlled trial. We screened 32 RRP (CRP⬎ 35mg/L, RF ⫹, and ACPA⫹) early (disease duration⬍6 months) RA patients for inclusion. 28 were randomly allocated tocilizumab group (T group) or tocilizumab plus intra-articular steroid injection group (T plus I group). All patients were taking methotrexate (from 12 to 22mg a week). For T plus I group, palpate examinations of both MP and PIP joints, wrists, elbows, shoulders, and knees were performed every 4 weeks. If swollen joints were existed, intra-articular steroid injections were intensified in each swollen joints by clinician’s decision. Co- primary endpoints were proportion of patients showing clinical remission (SDAI ⬍3.3) and radiographic non-progression (⌬ modified total Sharp score ⱕ0.5) at 52 weeks. Analysis was by intention-to-treat with last observation carried forward to missing data. Results: The characteristics of each group at baseline were not significantly different. Clinical remission at 52weeks was achieved by

Disclosure: K. Kume, None; K. Amano, None; S. Yamada, None; T. Kanazawa, None; H. Komori, None; H. Ohta, None; N. Kuwaba, None; K. Hatta, None.

2765 A Double-Blind Randomized Placebo-Controlled Trial Of Lovastatin in Patients with Rheumatoid Arthritis. Cynthia Aranow1, John J. Cush2, Marcy B. Bolster3, Christopher C. Striebich4, Maria Dall’Era5, Meggan Mackay1, Ewa Olech6, Tracy M. Frech7, J. Box8, Richard M. Keating9, Mary Chester M. Wasko10, E. William St Clair11, Alan Kivitz12, Betty Diamond1, Anne Davidson1, Meagan Spychala13, Ellen A. Goldmuntz14 and Autoimmunity Centers of Excellence15. 1Feinstein Institute for Medical Research, Manhasset, NY, 2Baylor Research Institute and Baylor University Medical Center, Dallas, TX, 3Medical University of South Carolina, Charleston, SC, 4University of Colorado Denver, Aurora, CO, 5University of California, San Francisco, San Francisco, CA, 6Oklahoma Medical Research Foundation, Oklahoma City, OK, 7 University of Utah School of Medicine, SLC, UT, 8Carolina Bone and Joint, Charlotte, NC, 9The University of Chicago, Chicago, IL, 10Temple University School of Medicine, Pittsburgh, PA, 11Duke Unversity Medical Center, Durham, NC, 12Altoona Center for Clinical Research, Duncansville, PA, 13Rho, Inc., Chapel Hill, NC, 14NIAID/NIH, Bethesda, MD, 15National Institutes of Health, Bethesda, MD. Background/Purpose: HMG-CoA reductase inhibitors (statins) are standard treatment for hyperlipidemia. In addition to lipid lowering abilities, statins exhibit multiple anti-inflammatory effects. The objectives of this study were to determine if treatment of patients with Rheumatoid Arthritis (RA) with lovastatin decreased CRP or reduced disease activity. Methods: We conducted a randomized double blind placebo controlled 12 week trial. 64 patients with mildly active RA (defined as 2–8 tender joints and 1–6 swollen joints) and an elevated CRP (⬎ 5mg/L) were randomized (1:1) to receive lovastatin 80mg or placebo. Patients could be on stable prednisone ⱕ 10 mg, DMARDs and/or biologic therapy. The primary efficacy endpoint was the reduction in mean log CRP. Secondary endpoints included disease activity, RF and anti-CCP antibody titers. Safety was a co-primary endpoint; hepatic and muscle toxicities were of particular interest. Results: Baseline features of the treatment groups were similar. The mean baseline CRP was 12.2 mg/L in the lovastatin arm and 12.6 mg/L in the placebo arm; mean baseline DAS-28 CRP was 3.5 and 3.6 in the lovastatin and placebo arms, respectively. No significant differences in mean log CRP reduction or % change in CRP from baseline between the two treatment arms were observed. No significant difference between the lovastatin and placebo arms was observed in a longitudinal model of the estimated mean log CRP. Disease activity assessed by DAS28 did not change from baseline in either lovastatin or placebo treated groups (⫺0.42 and ⫺0.58, respectively, ns). At week 12, clinical responses were comparable in subjects receiving lovastatin or placebo: ACR 20 (29% vs.

S1181

Tuesday, October 29

Rheumatoid Arthritis Treatment - Small Molecules, Biologics and Gene Therapy: Efficacy of Approved Biologics I Tuesday, October 29, 2013, 4:30 PM–6:00 PM

References: 1) Effectiveness of initial treatment allocation based on expert opinion for prevention of rapid radiographic progression in daily practice of an early RA cohort. Durnez A, et al. Ann Rheum Dis. 2011 Apr;70(4):634–7. Epub 2010 Dec 21. 2) A matrix risk model for the prediction of rapid radiographic progression in patients with rheumatoid arthritis receiving different dynamic treatment strategies: post hoc analyses from the BeSt study. Visser K, et al. Ann Rheum Dis. 2010 Jul;69(7):1333–7. Epub 2010 May 24. 3) A pilot risk model for the prediction of rapid radiographic progression in rheumatoid arthritis. Vastesaeger N, et al. Rheumatology (Oxford). 2009 Sep; 48(9):1114–21. Epub 2009 Jul 9. Review.

40%) and Good/Moderate EULAR response (42% vs. 44%). Autoantibody titers were stable during the course of the study with no group differences. Although not statistically different, the frequency of subjects receiving biologic therapy was greater in the lovastatin treated group than placebo (59% vs. 38%). A post-hoc analysis of subjects not using biologic therapy (n⫽32) demonstrated a significantly greater proportion achieving ⱖ15% reduction in CRP from baseline in the lovastatin treated group (83%) compared to placebo (33%; p ⫽ 0.019). No analogous difference was observed in subjects receiving biologics (47% vs. 60%). The mean change from baseline CRP in subjects not using biologics was also numerically greater in subjects taking lovastatin (⫺4.75) than placebo (⫺2.00). Lovastatin was well tolerated with no serious safety concerns. Several subjects experienced transient reversible elevations of transaminases. Clinical myositis was not observed. Conclusion: Statins were well tolerated in our patient population. This study showed no anti-inflammatory or clinical effects after 12 weeks of treatment with lovastatin. However, we observed a potential modest effect of lovastatin in subjects not using biologics, suggesting statins may be antiinflammatory in selected patients. Sponsored by NIAID Autoimmunity Centers of Excellence: U19 AI056362, U19AI056363; NCT00710021 Disclosure: C. Aranow, None; J. J. Cush, None; M. B. Bolster, None; C. C. Striebich, None; M. Dall’Era, None; M. Mackay, None; E. Olech, None; T. M. Frech, None; J. Box, None; R. M. Keating, None; M. C. M. Wasko, None; E. W. St Clair, None; A. Kivitz, None; B. Diamond, None; A. Davidson, None; M. Spychala, None; E. A. Goldmuntz, None; A. Centers of Excellence, None.

Tuesday, October 29

2766 Improvement In Insulin Resistance Greater With The Use Of Infliximab Following Intensive Treatment Of Early Rheumatoid Arthritis. LesleyAnne Bissell1, Elizabeth Hensor2, Sarah L. Mackie2, Agata Burska2, Jackie L. Nam2, Lukasz Kozera2, Helen I. Keen2, Edith Villeneuve2, Heike Eberl3, Helena Donica4, Philip G. Conaghan5, Jacqueline Andrews2, Paul Emery6 and Ann W. Morgan2. 1Leeds Institute of Rheumatic and Musculoskeletal Medicine, University of Leeds and NIHR Leeds Musculoskeletal Biomedical Research Unit, Leeds Teaching Hospitals NHS Trust, Leeds, United Kingdom, 2NIHR-Leeds Musculoskeletal Biomedical Research Unit and Leeds Institute of Rheumatic and Musculoskeletal Medicine, University of Leeds, Leeds, United Kingdom, 3Roche Molecular Diagnostics, Burgess Hill, United Kingdom, 4Department of Biochemical Diagnostics, Medical University of Lublin, Lublin, Poland, 5NIHR Leeds Musculoskeletal Biomedical Research Unit, Leeds, United Kingdom, 6Leeds Musculoskeletal Biomedical Research Unit, Leeds Teaching Hospitals NHS Trust, Leeds, United Kingdom. Background/Purpose: Insulin resistance (IR), N-terminal pro-brain natriuretic peptide (NT-proBNP) and total cholesterol/high density lipoprotein cholesterol ratio (TC/HDL-C) profiles have been proposed as surrogate measures of CVD in RA, with some improving with suppression of disease activity. In early RA, few randomised controlled trials (RCTs) have compared the effects of DMARDs with TNFi on these biomarkers. A recent cohort study has linked TNFi to a reduced risk of diabetes. We aimed to determine whether TNFi had greater influence over DMARDs on homeostasis model assessment of IR (HOMA-IR), NT-proBNP and TC/HDL-C. Methods: The Infliximab as Induction therapy for Early rheumatoid Arthritis (IDEA) multicentre double-blind RCT recruited 112 DMARD-naı¨ve RA patients (pts) (1987 ACR criteria; 3–12 months duration). Pts were randomised 1:1 to IFX⫹MTX or MTX with single-dose 250mg IV methylprednisolone (MP) as induction therapy. Treatment was blinded to week 26 then guided using a treat to target (T2T) approach. 120mg IM MP was given in both groups at weeks 6, 14 and 22 if DAS⬎2.4 at that visit. A single centre conducted the CV sub-study. Fasting glucose, lipids, insulin and NT-proBNP were measured at baseline, week 26 (w26) and 78 (w78). Multiple imputation by chained equations, using predictive mean matching, created 20 datasets; linear regression analyses adjusted for baseline values. HOMA-IR (glucose*insulin/405) and NT-proBNP values were ln-transformed prior to analysis. Results: CV biomarker data were available for 86 pts; 7 had known CVD and were excluded from the subsequent analysis. Of the 79 pts included (age 51.6 [range 19–75], 71% female, 57% RF⫹ve, 72% ACPA⫹ve), 38 received IFX and 41 MP. Baseline clinical characteristics were similar between the two groups, including CV risk factors. DAS44 remission rates did not differ

between TNFi and DMARD groups at w26 (32% vs. 37%, p⫽0.956) and w78 (43% vs. 54%, p⫽0.593). In both groups the three surrogate CVD markers improved on average at w26 and w78 (see Table 1). TC/HDL-C or NT-proBNP did not differ, but the TNFi group showed greater improvement in HOMA-IR at w78. Adjusting for IA/IM steroid injection dose did not alter the result. Table 1. Changes in surrogate measures of CVD over 78 weeks Baseline

Change at week 26

MTXⴙIV MTXⴙIFX Steroid MTXⴙIV nⴝ38 nⴝ41 MTXⴙIFX Steroid TC/HDL-C (mean (SD))

⫺0.67

⫺0.96

Adjusted difference (95% CI), p-value 0.11 (⫺0.45, 0.66), p⫽0.709

Change at week 78

MTXⴙIFX

MTXⴙIV Steroid

⫺0.86

⫺1.01

Adjusted difference (95% CI), p-value ⫺0.08 (⫺0.89, 0.73), p⫽0.838

5.14 (1.60)

5.71 (2.26)

HOMA-IR (geometric mean)*

2.17

2.54

FU/BL 0.68 FU/BL 0.69

IFX/IVS 0.87 (0.65, 1.16) p⫽0.333

FU/BL 0.51 FU/BL 0.87

IFX/IVS 0.51 (0.35, 0.76) p⫽0.001

NT-proBNP (geometric mean)*

82.47

62.92

FU/BL 0.80 FU/BL 0.82

IFX/IVS 1.06 (0.77, 1.46) p⫽0.708

FU/BL 0.85 FU/BL 0.88

IFX/IVS 1.06 (0.73, 1.52) p⫽0.768

*it was not possible to calculate SD in original units for log-transformed variables

Conclusion: Treatment of early RA was associated with improvement in TC/HDL-C, IR and NT-proBNP. To our knowledge this is the first double-blinded RCT to compare the change in HOMA-IR with TNFi versus non-TNFi in early DMARD-naive RA. We determined a greater long-term improvement in HOMA-IR in those treated with TNFi; on average at w78 HOMA-IR values were around half (0.51 times) as high as those treated with MTX/IV steroid. When implementing a T2T approach, there appears to be an advantage with the use of TNFi. Longer term follow up is underway to determine if these findings translate into reduced overt CVD. Disclosure: L. A. Bissell, None; E. Hensor, None; S. L. Mackie, None; A. Burska, None; J. L. Nam, None; L. Kozera, None; H. I. Keen, UCB, 5; E. Villeneuve, Janssen, Bristol-Myers-Squibb, Abbott, Amgen, UCB, 5; H. Eberl, Roche Pharmaceuticals, 3; H. Donica, None; P. G. Conaghan, Pfizer Inc, Janssen Pharmaceutica Product, L.P., 5, Bristol-Myers Squibb, Pfizer Inc, 8; J. Andrews, None; P. Emery, Abbvie, MSD, UCB, Pfizer, Roche, BMS, 5, Abbvie, MSD, UCB, Pfizer, Roche, BMS, 8; A. W. Morgan, Merck Pharmaceuticals, 2.

2767 Tocilizumab In Combination Therapy and Monotherapy Versus Methotrexate In Methotrexate-Naive Patients With Early Rheumatoid Arthritis: Clinical and Radiographic Outcomes From a Randomized, PlaceboControlled Trial. Gerd R. Burmester1, William Rigby2, Ronald F. van Vollenhoven3, Jonathan Kay4, Andrea Rubbert-Roth5, Ariella Kelman6, Sophie Dimonaco7 and Nina Mitchell7. 1Charite´ - Universita¨tsmedizin Berlin, Berlin, Germany, 2Dartmouth-Hitchcock Medical Center and Dartmouth Medical School, Lebanon, NH, 3Karolinska Institute, Stockholm, Sweden, 4 UMass Memorial Medical Center and University of Massachusetts Medical School, Worcester, MA, 5University of Cologne, Cologne, Germany, 6Genentech, South San Francisco, CA, 7Roche, Welwyn Garden City, United Kingdom. Background/Purpose: Recent recommendations support intensive treatment of patients (pts) with early rheumatoid arthritis (RA) to achieve remission or low disease activity.1–3 Tocilizumab (TCZ) was not previously studied exclusively in an early RA population. The purpose of this study was to assess the efficacy and safety of TCZ ⫾ methotrexate (MTX) vs MTX in MTX-naive pts with early RA (defined as ⱕ2 y since diagnosis). Methods: Pts were randomized 1:1:1:1 (double-dummy, doubleblind) to receive TCZ 8 mg/kg (TCZ8) ⫹ MTX (primary intervention), TCZ8 monotherapy (TCZ8 MONO), TCZ 4 mg/kg (TCZ4) ⫹ MTX, or MTX for 104 wks. Pts received IV TCZ q4w, and MTX starting at 7.5 mg qw and escalating to 20 mg qw by wk 8. Inclusion criteria included RA duration ⱕ2 y, DAS28 ⬎3.2, MTX-naive, elevated ESR or CRP, and presence of RF or anti-CCP antibodies or radiographic erosion(s). Primary endpoint was proportion of pts achieving DAS28 remission (DAS28 ⬍2.6) at wk 24. Key secondary endpoints included mean changes from baseline (BL) to wk 52 in van der Heijde–modified Total Sharp Score (mTSS) and improvement in physical function (using HAQ-DI). A hierarchy of statistical testing was implemented to control the type 1 error rate for multiplicity. This trial is ongoing; 52-wk data are reported here.

S1182

Results: The intent-to-treat population consisted of 1157 pts. BL characteristics were similar in all treatment groups: mean RA duration, 0.4–0.5 y; mean DAS28, 6.6–6.7; mTSS, 5.66–7.72. Statistically significantly greater proportions of TCZ8 ⫹ MTX vs MTX pts achieved DAS28 remission and ACR20/50/70 responses at wks 24 and 52 (p ⬍ 0.05); statistically significant improvements in mean mTSS and HAQ-DI were also observed at wk 52 (p ⬍ 0.05; Table). TCZ8 MONO was also statistically significant for the primary endpoint of percentage of DAS28 remission responders in comparison to MTX at wk 24 (p ⬍ 0.05). Both TCZ8 MONO and TCZ4 ⫹ MTX exhibited numerically greater improvements vs MTX across key secondary endpoints and were more efficacious than MTX in preventing structural joint damage (Table). Adverse events (AEs) observed with TCZ were consistent with its known safety profile. Incidences of AEs and serious AEs were similar across groups, while serious infections were highest with combination therapy (Table). Overall, 9 deaths were observed across all groups; causes of death were variable. Table. Primary and Selected Secondary Efficacy Endpoints (ITT Population) and Safety Data TCZ8ⴙMTX TCZ8 MONO TCZ4ⴙMTX MTX n ⴝ 290 n ⴝ 292 n ⴝ 288 n ⴝ 287 44.8*** 4.77 74.5* 56.9**

38.7*** 3.70 70.2 47.6

31.9††† 2.72 73.6† 47.9

15.0

38.6** 0.08** ⫺0.81* 18.4b

30.1 0.26† ⫺0.67 14.2

34.7† 0.42† ⫺0.75 16.7b

25.4 1.14 ⫺0.64 10.0

28.5b

22.6

22.6

16.4

b

b

b

65.2 43.2

24.5

20.5

22.2

13.2

256 (88.3) 31 (10.7) 10 (3.4)

250 (85.6) 25 (8.6) 8 (2.7)

256 (88.6) 29 (10.0) 11 (3.8)

235 (83.3) 24 (8.5) 6 (2.1)

2 (0.7)

1 (0.3)

4 (1.4)

2 (0.7)

ITT, intent-to-treat; OR, odds ratio. efficacy comparisons versus MTX (MTX ⫹ placebo). ***p ⬍ 0.0001; **p ⬍ 0.001: *p ⬍ 0.05. †††p ⬍ 0.0001 and †p ⬍ 0.05; these comparisons occurred after break in the hierarchical testing sequence. a Analyses were based on safety population to 52 weeks (TCZ8 ⫹ MTX, n ⫽ 290; TCZ8 MONO, n ⫽ 292; TCZ4 ⫹ MTX, n ⫽ 289; MTX, n ⫽ 282); multiple occurrences of the same AE in a patient were counted only once. b p ⬍ 0.05; ACR/EULAR Boolean remission, ACR/EULAR index (SDAI) remission, and CDAI remission were performed as post hoc exploratory analyses; therefore, p values were not adjusted for multiplicity.

Conclusion: TCZ is effective as combination therapy and monotherapy in MTX-naive pts with early active RA. TCZ resulted in greater improvements from BL in signs, symptoms, and physical function and in inhibition of structural joint damage in all treatment groups vs MTX alone. Of the 3 TCZ treatment groups, efficacy outcomes for TCZ vs MTX were consistently greatest in the TCZ8 ⫹ MTX group. The overall safety of TCZ was consistent with its known profile.

Intravenous Golimumab Inhibits Radiographic Progression and Maintains Clinical Efficacy and Safety In Patients With Active Rheumatoid Arthritis Despite Methotrexate Therapy: 2-Year Results Of a Phase 3 Trial Of Intravenous Golimumab. Michael E. Weinblatt1, Clifton O. Bingham III2, Alan M. Mendelsohn3, Lilianne Kim3, Kim Hung Lo3, Lenore Noonan3, Daniel Baker3 and Rene Westhovens4. 1Brigham and Women’s Hospital, Boston, MA, 2Johns Hopkins University, Baltimore, MD, 3Janssen Research & Development, LLC., Spring House, PA, 4University Hospital KU Leuven, Leuven, Belgium. Background/Purpose: To evaluate long-term clinical/radiographic efficacy of IV GLM 2mg/kg⫹MTX in active RA despite MTX through wk112. Methods: 592 pts with active RA (ⱖ6/66 SJC, ⱖ6/68 TJC, CRPⱖ1.0mg/ dL, RF and/or anti-CCP positive) despite ⱖ3 months of MTX (15–25mg/wk) participated in this multicenter, randomized, double-blind, placebo (PBO)controlled study. Pts were randomized to IV GLM 2mg/kg or PBO at wks0&4 and q8wks; all pts continued stable MTX. PBO pts with ⬍10% improvement in SJC⫹TJC at wk16 could early escape to IV GLM 2mg/kg (wks16&20, q8wks). All PBO pts received IV GLM 2mg/kg starting at wk24. Primary endpoint was wk14 ACR20. Radiographs of hands and feet at wks 0, 24(wk16, early escape), 52, and 100 were scored by 2 independent readers and adjudicator (as needed) using the vdH-S score. Reading Session 2 included wk0, wk52, and wk100 scoresIn general, analyses at wks 24, 52, and 100 were performed using ITT methodology, with imputation for missing data. Results: 82% of pts (486/592) continued through wk112; 106 pts d/c, mostly due to AEs(44pts) and few due to lack of efficacy (12pts). At wk14, significantly (p⬍0.001) larger proportions of GLM⫹MTX vs PBO⫹MTX pts had ACR20/50/70, DAS28-CRP good/moderate responses, and greater improvements in HAQ. Clinical improvements were maintained through wk100 (final efficacy visit), when ACR20/50/70 responses among all GLM⫹MTX-treated pts were 68.1%, 43.8%, and 23.5%, resp; DAS28-CRP moderate/good response was 81.9%; and median improvement from wk0 in HAQ was 0.5; 67.1% of GLM⫹MTX pts had improvement in HAQ ⱖ0.25 from baseline. GLM⫹MTX-treated pts continued to have significantly less radiographic progression based on vdH-S total and subscores vs PBO⫹MTX at wk24, and PBO⫹MTXa`GLM ⫹ MTX at wk52 and wk100. Pts randomized to PBO⫹MTX who began GLM at wk16/24 demonstrated marked slowing of radiographic progression from wk24–52 and from wk52–100. Through wk112 (final safety visit), the mean follow-up for all GLM-treated pts was 96wks. AEs and serious AEs occurred in 79% and 18%, resp, of GLM-treated pts (vs 49% and 2% at wk24). 3 cases of TB and 2 serious opportunistic infections (cryptococcal pneumonia, intervertebral discitis) were reported through wk112. 6pts (1.0%) died: 1 PBO⫹MTX and 5 (0.8%) GLM⫹MTX (pneumonia/MI, dehydration, abdominal TB, unknown ⫻ 2). Through wk112, the proportion of infusions with infusion reactions was 0.4% and the proportion of pts with infusion reactions was 3.9% (vs. 1.1% and 3.5%, respectively, at wk24). RADIOGRAPHIC PROGRESSION (mean ⴞ standard deviation) Baseline Total vdH-S Total vdH-S change from baseline at wk24 Total vdH-S change from baseline at wk52 Total vdH-S change from baseline at wk100 Total vdH-S change from wk24-wk 52 Total vdH-S change from wk52wk100

Placebo 3 GLM 2 mg/kg ⴙ MTX (nⴝ197) 50.26 ⫾ 59.85 1.09 ⫾ 3.19

GLM 2mg/kg ⴙ MTX (nⴝ395) 47.59 ⫾ 54.63 0.03 ⫾ 1.90 ***

All GLM 2 mg/kg ⴙ MTX (nⴝ592) – – 0.49 ⫾ 3.46

1.22 ⫾ 3.98

0.13 ⫾ 3.11**

2.10 ⫾ 7.42

0.74 ⫾ 6.32**

1.19⫾ 6.73

0.12 ⫾ 2.44

0.15 ⫾ 1.83

0.14 ⫾ 2.06

0.80 ⫾ 3.03

0.58 ⫾ 3.07

0.64 ⫾ 3.06

**, ***p-value vs. placebo ⫹ MTX ⬍0.01, 0.001, respectively. Pts with missing total vdH-S score at wk52 excluded.

References: 1. Arthritis Care Res 2012;64:625; 2. Ann Rheum Dis 2010;69:631; 3.Ann Rheum Dis 2010;69:964 Disclosure: G. R. Burmester, Roche, Abbott, Pfzier, UCB, Merck Sharp and Dohme and Bristol-Myers Squibb, 2, Roche, Chugai, Pfizer, UCB and BristolMyers Squibb, 5, Roche, Pfizer, Merck Sharp and Dohme, Abbott and BristolMyers Squibb, 8; W. Rigby, Roche Pharmaceuticals, 5; R. F. van Vollenhoven, Abbott Immunology Pharmaceuticals, 2, BMS, 2, GSK, 2, MSD, 2, Pfizer Inc, 2, Roche Pharmaceuticals, 2, UCB, 2, Abbott Immunology Pharmaceuticals, 5, BMS, 5, GSK, 5, MSD, 5, Pfizer Inc, 5, Roche Pharmaceuticals, 5, UCB, 5; J. Kay, Ardea Biosciences, Eli Lilly, Fidia Farmacutici, SpA, Pfizer, Roche, Sanofi-Aventis, 2, Amgen, Baxter Healthcare Corporation, BMS, Celgene, fourteen22 Inc., Genentech, Hospira, Inc., Horizon Pharma, Inc., Janssen, Medac Pharma Inc., PanGenetics, B.V., Pfizer, Roche, Savient Pharmaceuticals, Inc., Sun Pharmaceutical Industries Ltd., UCB,, 5; A. Rubbert-Roth, Roche, Chugai, Pfizer, 2, Roche, Chugai, Pfizer, UCB, MSD, 5, Roche, UCB, MSD, 8; A. Kelman, Genentech and Biogen IDEC Inc., 3; S. Dimonaco, Roche Pharmaceuticals, 3; N. Mitchell, Roche Pharmaceuticals, 1, Roche Pharmaceuticals, 3.

Conclusion: IV GLM⫹MTX significantly inhibited radiographic progression (vdH-S scores) at wks 24,52 and 100. Among PBO-treated pts who began GLM at wk16/24, marked slowing of radiographic progression, to rates similar to pts randomized to GLM, was observed from wk24–52 and from wk52–100. IV GLM⫹MTX also significantly improved and maintained RA signs/symptoms in pts with active RA despite ongoing MTX and continued to demonstrate an acceptable safety profile through wk112. Disclosure: M. E. Weinblatt, Janssen Research & Development, LLC., 5; C. O. Bingham III, Janssen Pharmaceutica Product, L.P., 2, Janssen Pharmaceutica Product, L.P., 5, UCB, 2, UCB, 5, Pfizer, 2, Pfizer Inc, 5, AbbVie/Abbott, 5, Amgen, 5, BMS, 2, BMS, 5, Celgene, 5, Corrona, 2, Genetech/Roche, 2, Genetech/rRoche, 5, Novartis Pharmaceutical Corporation, 5, Mesoblast, 2; A. M. Mendelsohn, Janssen Research & Development, LLC., 3; L. Kim, Janssen Research & Development, LLC., 3; K. H. Lo, Janssen Research & Development, LLC., 3; L. Noonan, Janssen Research & Development, LLC., 3; D. Baker, Janssen Research & Development, LLC., 3; R. Westhovens, BMS, 8, Janssen; Galapagos, 9, Roche Pharmaceuticals, 2.

S1183

Tuesday, October 29

Efficacy DAS28-ESR remission (⬍2.6), 24 wks (%, OR) ACR20 ACR50 24 wks (%) ACR70 Mean ⌬mTSS, BL to 52 wks Mean ⌬HAQ-DI, BL to 52 wks ACR/EULAR Boolean remission, 24 wks (%) ACR/EULAR index (SDAI) remission, 24 wks (%) CDAI remission, 24 wks (%) Safetya Patients with AEs, n (%) Patients with serious AEs, n (%) Patients with serious infections, n (%) Deaths, n(%)

2768

2769

Tuesday, October 29

Clinical Remission After 52 Weeks Of Treatment Is a Predictor Of Adalimumab-Free Disease Control In Patients With Early Rheumatoid Arthritis: Hopeful 2 Study. Yoshiya Tanaka1, Hisashi Yamanaka2, Naoki Ishiguro3, Nobuyuki Miyasaka4, Katsuyoshi Kawana5, Tadamichi Kubo5, Aki Kuroki5 and Tsutomu Takeuchi6. 1University of Occupational and Environmental Health, Japan, Kitakyushu, Japan, 2Tokyo Women’s Medical University, Tokyo, Japan, 3Nagoya Univeristy Graduate School of Medicine, Nagoya, Japan, 4Tokyo Medical and Dental University, Tokyo, Japan, 5 AbbVie GK, Tokyo, Japan, 6Keio University School of Medicine, Tokyo, Japan. Background/Purpose: Although it is possible to achieve remission or low disease activity (LDA) with the combination of methotrexate (MTX) and biologics for many patients with early rheumatoid arthritis (RA), it is uncertain whether disease is controlled following withdrawal of biologics. The purpose of this observational study was to assess the effect of withdrawal of adalimumab (ADA) on disease activity following treatment with MTX plus ADA, and to identify predictors of biologic-free disease control in patients with early RA. Methods: Patients with early RA received blinded ADA 40 mg every other week (EOW) plus MTX 6–8 mg every week (EW) or MTX 6–8 mg EW alone for 26 weeks. Thereafter, all patients received open-label ADA 40 mg EOW plus MTX 6–8 mg EW for 26 weeks in the HOPEFUL 1 study. At week 52, patients could be enrolled in the 52-week observational, follow-up, HOPEFUL 2 study, during which time they received ADA plus MTX treatment (ADA-continued group), or MTX alone (ADA-withdrawal group) at the investigator’s discretion. The primary outcomes of this study were disease activity score based on 28 joints count based on erythrocyte sedimentation rate (DAS28-ESR), Health Assessment QuestionnaireDisability Index (HAQ-DI), and modified total Sharp score (mTSS) at week 104. The factors correlated with the outcomes were also analyzed. Results: Among 278 patients completing the 52-week HOPEFUL 1 study, 220 were enrolled in the HOPEFUL 2 study. At week 52, baseline characteristics including DAS28-ESR, HAQ-DI and mTSS were comparable between the ADA-continued (N⫽106) and ADA-withdrawal (N⫽114) groups. 70% and 50% from each group achieved low disease activity (LDA, DAS28-ESR⬍3.2) and clinical remission (DAS28-ESR⬍2.6), respectively. Table shows disease activity scores at week 104. Mean DAS28-ESR score at week 104 was significantly higher and the percentage of the patients who achieved LDA or clinical remission was significantly lower in the ADAwithdrawal group. There was no significant difference in HAQ-DI score and ⌬mTSS in the two groups at week 104. In the ADA-withdrawal group, lower baseline CRP scores and lower DAS28-ESR scores at week 52 predicted LDA at week 104 in multivariate analysis. The cut-off point to sustain LDA through week 104 without using ADA was DAS28-ESR⬍ 2.6 at week 52. Table. Disease activity scores at week 104 in ADA-continued group and ADA-withdrawal group

DAS28-ESR, Mean⫾SD % of DAS28-ESR⬍3.2 % of DAS28-ESR⬍2.6 HAQ-DI, Mean⫾SD ⌬mTSS from week 52, Mean⫾SD

ADAcontinued group

ADAwithdrawal group

P value

2.70⫾1.08 72.5 53.8 0.20⫾0.29 0.8⫾3.9

3.20⫾1.24 55.8 36.8 0.26⫾0.39 0.6⫾1.8

0.006 0.021 0.026 0.609 0.335

Conclusion: More patients who continued ADA therapy sustained LDA and clinical remission after 1 year. Nevertheless, half of ADA-withdrawal RA patients sustained LDA for 1 year. Achieving DAS28-ESR-remission after 52 weeks of treatment using ADA was the key determinant for biologic-free disease control in early RA patients. Disclosure: Y. Tanaka, BMS, MSD, Chugai, Mitsubishi-Tanabe, Astellas, Abbvie, Daiichi-Sankyo, 2, Mitsubishi-Tanabe, Abbott, Abbvie, Eisai, Chugai, Janssen, Pfizer, Takeda, Astellas, Daiichi-Sankyo, GSK, AstraZeneca, Eli Lilly, Quintiles, MSD, Asahi Kasei, 8, Mitsubishi-Tanabe, Abbott, Abbvie, Eisai, Chugai, Janssen, Pfizer, Takeda, Astellas, Daiichi-Sankyo, GSK, AstraZeneca, Eli Lilly, Quintiles, MSD, Asahi Kasei, 5; H. Yamanaka, AbbVie GK, 2, Bristol-Myers Squibb, 2, Chugai Pharmaceutical Co, 2, Eisai Co, 2, Janssen Pharmaceutical K.K., 2, Mitsubishi Tanabe Pharma Corporation, 2, Otsuka Pharmaceutical Co, 2, Pfizer Japan Inc, 2, Takeda Industrial Pharmaceutical Co, 2, UCB Japan Co, 2, Abbvie GK, 8, Bristol-Myers Squibb, 8, Chugai Pharmaceutical Co, 8, Eisai Co, 8, Janssen Pharmaceutical K.K, 8, Mitsubishi Tanabe Pharma Corporation, 8, Otsuka Pharmaceutical Co, 8, Pfizer Japan Inc, 8, Takeda Pharmaceu-

tical Co, 8, UCB Japan Co, 8, AbbVie GK, 5, Bristol-Myers Squibb, 5, Chugai Pharmaceutical Co, 5, Eisai Co, 5, Janssen Pharmaceutical K.K., 5, Mitsubishi Tanabe Pharma Corporation, 5, Otsuka Pharmaceutical Co, 5, Pfizer Japan Inc, 5, Takeda Pharmaceutical Co, 5, UCB Japan Co, 5; N. Ishiguro, Abbott Japan, 2, Astellas Pharmaceutical, 2, Bristol-Myers Squibb, 2, Chugai Pharmaceutical Co, 2, Eisai Co, 2, Janssen Pharma, 2, Mitsubishi Tanabe Pharmaceutical Co, 2, Pfizer Japan, 2, Takeda Pharma, 2; N. Miyasaka, AbbVie GK, 2, Astellas Pharmaceutical, 2, Banyu Pharmaceutical, 2, Chugai Pharmaceutical Co, 2, Daiichi Sankyo Pharmaceutical Co, 2, Eisai Co, 2, Janssen Pharmaceuticals, 2, Mitsubishi Tanabe Pharma Corporation, 2, Takeda Pharmaceutical Co, 2, Teijin Limited, 2; K. Kawana, AbbVie GK, 3; T. Kubo, AbbVie GK, 3; A. Kuroki, AbbVie GK, 3; T. Takeuchi, Astra Zeneca K.K., Eli Lilly Japan K.K., Novartis Pharma K.K., Mitsubishi Tanabe Pharma Co., and Asahi Kasei Medical K.K, 5, AbbVie GK., Bristol–Myers K.K., Chugai Pharmaceutical Co., Ltd., Eisai Co., Ltd., Janssen Pharmaceutical K.K., Mitsubishi Tanabe Pharma Co., Pfizer Japan Inc., and Takeda Pharmaceutical Co., Ltd., 8, AbbVie GK., Astellas Pharma, Bristol–Myers K.K., Chugai Pharmaceutical Co, Ltd., Daiichi Sankyo Co., Ltd., Eisai Co., Ltd., Janssen Pharmaceutical K.K., Mitsubishi Tanabe Pharma Co., Nippon Shinyaku Co., Ltd., Pfizer Japan Inc., Sanofi–Aventis K.K., 2, Santen Pharmaceutical Co., Ltd., Takeda Pharmaceutical Co., Ltd., and Teijin Pharma Ltd., 2.

ACR Concurrent Abstract Session Sjo¨gren’s Syndrome: Basic Science Tuesday, October 29, 2013, 4:30 PM–6:00

PM

2770 Complex Functional Effects Within The HLA Contribute To Sjo¨gren’s Syndrome Pathogenesis and May Influence Both Transcriptional Regulation and Peptide Binding. Christopher J. Lessard1, He Li2, Indra Adrianto2, John A. Ice2, Mikhail G. Dozmorov2, Roland Jonsson3, Maureen Rischmueller4, Gunnel Nordmark5, Xavier Mariette6, Corinne MiceliRichard7, Marie Wahren-Herlenius8, Torsten Witte9, Michael T. Brennan10, Roald Omdal11, Lars Ro¨nnblom5, Patrick M. Gaffney2, Wan-Fai Ng12, Nelson L. Rhodus13, Barbara M. Segal14, Jonathan D. Wren2, R. Hal Scofield15, Juan-Manuel Anaya16, John B. Harley17, Courtney G. Montgomery2 and Kathy L. Sivils1. 1University of Oklahoma Health Sciences Center, Oklahoma City, OK, 2Oklahoma Medical Research Foundation, Oklahoma City, OK, 3University of Bergen, Bergen, Norway, 4Queen Elizabeth Hospital, Adelaide, Australia, 5Uppsala University, Uppsala, Sweden, 6Biceˆtre University Hospital, Le Kremlin Bicetre, France, 7Universite´ Paris Sud, Le Kremlin Biceˆtre, France, 8Karolinska Institutet, Stockholm, Sweden, 9Medical University Hannover, Hanover, Germany, 10Carolinas Medical Center, Charlotte, NC, 11Stavanger University Hospital, Stavanger, Norway, 12Newcastle University, Newcastle upon Tyne, United Kingdom, 13University of Minnesota, Minneapolis, MN, 14Hennepin County Medical Center, Minneapolis, MN, 15US Department of Veterans Affairs Medical Center, Oklahoma City, OK, 16School of Medicine and Health Sciences, Universidad del Rosario. Center for Autoimmune Diseases Research (CREA), Bogota´, Colombia, 17Cincinnati Children’s Hospital Medical Center, Cincinnati, OH. Background/Purpose: Primary Sjo¨gren’s syndrome (pSS) is a common, heterogeneous exocrinopathy. Etiology involves complex environmental, genetic and genomic influences driving innate and adaptive autoimmune responses. This study sought to integrate genome-wide association study (GWAS) and classical HLA allele associations in SS, and to explore this genetically complex region for insight into disease mechanisms through bioinformatic approaches. Methods: Genotype data was collected for 1638 pSS cases and 6754 controls on Illumina Omni1-Quad or ImmunoChip arrays. HLA classical allele imputation was performed using HiBag with a certainty threshold of ⬎80% to define relationships with single nucleotide polymorphisms (SNPs). A set of 9122 pSS associated SNPs (P⬍5⫻10E-5) were tested for enrichment using GenomeRunner to determine if the set of SNPs co-localize with various functional genomic elements such as specific transcription factor binding sites. A subset of SS associated SNPs that were significantly enriched in genomic locations near functional elements related to the transcriptional regulator, RFX5, were tested by expression quantitative trait locus (eQTL) analysis using the MATRIXeQTL package in R using transcript levels measured by Illumina WG-6 arrays in 133 subjects. Results: After adjusting for the most significant GWAS variant (rs115575857 in HLA Class II; P⫽1.65⫻10E-114), we found a second independent effect peaking at rs116232857 (P⫽1.33⫻10E-96). Imputation and regression analyses identified the previously reported ancestral haplotype including DQB1*0201 (P⫽1.38⫻10E-95), DQA1*0501 (P⫽8.50⫻10E-94), and DRB1*0301 (P⫽2.19⫻10E-84) and indicated rs115575857 was synonymous with this haplotype. The association at rs116232857 persisted and

S1184

accounted for DQB1*0501, but DQB1*0501 could not account for rs116232857, suggesting rs116232857 either tags multiple functional effects or, alternatively, DQB1*0501 is a false positive. Statistically significant co-localization (P⫽1.53⫻10E-14) was observed showing that 160 HLA region variants are located within or near (100 bp) ChIP-seq peaks for RFX5, a key transcriptional regulator of HLA Class I and II loci. Further analyses of RFX5 related variants identified multiple eQTLs (P-values between 8.31⫻10E-11 and 4.54⫻10E-33) in both the Class I and Class II HLA loci and include HLA-DRB6, HLA-C, HLA-DPB1, HLA-DQA1, and HLA-A. Risk alleles for these eQTLs reside on both pSS associated haplotypes. Conclusion: We conclude that the ancestral haplotype of DQB1*0201, DQA1*0501, and DRB1*0301 corresponds to our top GWAS effect, rs115575857 and the second independent effect at rs116232857 is novel. RFX5 was identified as a novel potential functional candidate as the RFX5 binding elements were enriched for pSS associated variants affecting HLA Class I and II expression levels. This work suggests that complex functional elements in the HLA region involved in both transcriptional regulation and peptide binding impact SS. Disclosure: C. J. Lessard, None; H. Li, None; I. Adrianto, None; J. A. Ice, None; M. G. Dozmorov, None; R. Jonsson, None; M. Rischmueller, None; G. Nordmark, None; X. Mariette, None; C. Miceli-Richard, None; M. Wahren-Herlenius, None; T. Witte, None; M. T. Brennan, None; R. Omdal, None; L. Ro¨nnblom, None; P. M. Gaffney, None; W. F. Ng, None; N. L. Rhodus, None; B. M. Segal, None; J. D. Wren, None; R. H. Scofield, None; J. M. Anaya, None; J. B. Harley, None; C. G. Montgomery, None; K. L. Sivils, None.

2771

Background/Purpose: NK cells are an important subset of cells involved in innate immunity. Their possible role has never been studies in pSS pathogeny. We aimed to assess the involvement of NCR3/NKp30, a NKspecific activating receptor regulating the cross-talk between NK and dendritic cells and type II IFN secretion, and its receptor named B7H6 in pSS pathogenesis. Methods: First, a cohort of 584 pSS patients (ASSESS ⫹ KB cohort) and 451 controls of Caucasian ancestry, addressed by 48 AIMs, was used for exploratory genetic study. Nine single nucleotide polymorphisms (SNPs) within the 6p21.3 NCR3 locus and 3 additional SNP proxies for HLA-DR2,HLA-DR3 and TNF-308 were genotyped. Two NCR3 SNPs (rs11575837, rs2736191) and the SNP proxy for HLA-DR3 (rs2187668) were genotyped in the replication study that included 436 pSS Scandinavian patients and 441 healthy controls. Then, NKp30 mRNA levels were investigated in 102 pSS patients from the French ASSESS cohort according to their genotype. Second, we performed phenotypic characterization of NK cells in 38 pSS patients compared to 30 age-matched controls. The functional relevance of expression levels of NKp30 on NK cells was assessed by a cross-linking assay to analyze degranulation and IFN-␥ secretion. Third, we assessed the presence of NK cells by immunohistochemistry (IHC) and transcriptional level of B7H6 within salivary glands. Last we investigated the NKp30-dependent cross-talk between NK cells and epithelial cells within salivary glands. Results: Our case-control study of genetic polymorphisms of the NCR3/ NKp30 gene demonstrated that the rare allele of the rs11575837 (G⬎A) residing in the promoter was protective for pSS and was associated with

Disclosure: G. Nocturne, None; S. Rusakiewicz, None; D. Sene, None; G. Nordmark, None; M. L. Eloranta, None; P. Eriksson, None; E. Theander, None; H. Forsblad-d’Elia, None; R. Omdal, None; M. Wahren-Herlenius, None; R. Jonsson, None; L. Ro¨nnblom, None; J. Nititham, None; K. E. Taylor, None; C. J. Lessard, None; K. L. Moser, None; J. E. Gottenberg, None; L. A. Criswell, None; C. Miceli-Richard, None; L. Zitvogel, None; X. Mariette, None.

2772 Identification Of a Sjo¨gren’s Syndrome-Associated Variant That Influences OAS1 Isoform Switching. He Li1, John A. Ice2, Jennifer A. Kelly2, Indra Adrianto2, Stuart B. Glenn2, Kimberly S. Hefner3, Evan G. Vista4, Donald U. Stone1, Raj Gopalakrishnan5, Glen D. Houston1, David M. Lewis1, Michael Rohrer5, Pamela Hughes5, John B. Harley6, Courtney G. Montgomery2, James Chodosh7, James A. Lessard8, Juan-Manuel Anaya9, Barbara M. Segal10, Nelson L. Rhodus5, Lida Radfar1, R. Hal Scofield2, Christopher J. Lessard1 and Kathy L. Sivils2. 1University of Oklahoma Health Sciences Center, Oklahoma City, OK, 2Oklahoma Medical Research Foundation, Oklahoma City, OK, 3Hefner Eye Care and Optical Center, Oklahoma City, OK, 4University of Santo Tomas, Taguig City, Philippines, 5University of Minnesota, Minneapolis, MN, 6US Department of Veterans Affairs Medical Center, Cincinnati, OH, 7Massachusetts Eye and Ear Infirmary, Harvard Medical School, Boston, MA, 8Valley Bone and Joint Clinic, Grand Forks, ND, 9School of Medicine and Health Sciences, Universidad del Rosario, Bogota´, Colombia, 10Hennepin County Medical Center, Minneapolis, MN. Background/Purpose: Sjo¨gren’s syndrome (SS) is a common, progressive autoimmune exocrinopathy characterized by symptoms of dry eyes and mouth present in 0.7–1% of the European population. Our previous gene expression profiling (GEP) study has demonstrated overexpression of transcripts induced by interferons (IFN) in SS patients. Here, we sought to identify and characterize underlying genetic contributions to dysregulation of IFN pathways in SS. Methods: IFN signature genes of interest were selected from GEP studies performed in 180 SS cases and 73 controls using Illumina Human WG-6 v3.0 microarray data and evaluated for cis-expression quantitative trait loci (eQTL) in 222 subjects by integration with genome-wide association study (GWAS) data. Gene splicing patterns were evaluated using microarray data and supplemented with RNA-sequencing (RNA-seq) performed in 57 SS cases and 27 controls on the Illumina platform. Transcripts measured by RNA-seq were reconstructed using Cufflinks and the relative abundance of isoforms was compared across samples according to genotypes for loci of interest. Results: GEP showed that OAS1, an IFN-inducible gene involved in inhibition of virus replication, was significantly overexpressed in SS patients. Multiple cis-eQTL were identified in OAS1 with the most significant peaking at rs10774671, strengthening prior evidence of this variant for disease association (P⫽6⫻10⫺3) obtained in our large GWAS dataset consisting of 395 cases and 1975 controls. We further replicated this genetic association in an independent set of 648 cases and 2927 controls followed by meta-analysis using a weighted Z score (Pmeta⫽9⫻10⫺6; OR⫽0.79). The rs10774671 A allele conferred risk, is a splice site variant located at the intersection between intron-5 and exon-6, and thus may switch the primary normal isoform, p46, to various alternatives. To characterize functional impact of this variant, we evaluated alternative splicing events using both microarray and RNA-seq data. Variation in splicing was detectable by a microarray probe that specifically recognizes a truncated form of OAS1 (p42). Both microarray and RNA-seq showed that the risk allele A, which demolishes the splicing consensus sequence, was correlated with higher expression of p42 (Pmicro⫽2⫻10⫺16 and Pseq⫽1⫻10⫺15). RNA-seq results also showed correlation of the A risk allele with higher proportions of p48 and p44 isoforms (P⫽9⫻10⫺8 and P⫽4⫻10⫺4, respectively), but a lower expression of the functionally normal isoform, p46 (P⫽4⫻10⫺30).

S1185

Tuesday, October 29

The NKp30/B7H6 Axis Contributes To Pathogenesis In Primary Sjo¨gren’s Syndrome. Gaetane Nocturne1, Sylvie Rusakiewicz2, Damien Sene3, Gunnel Nordmark4, Maija-Leena Eloranta5, Per Eriksson6, Elke Theander7, Helena Forsblad-d’Elia8, Roald Omdal9, Marie Wahren-Herlenius10, Roland Jonsson11, Lars Ro¨nnblom5, Joanne Nititham12, Kimberly E. Taylor13, Christopher J. Lessard14, Kathy L. Moser14, Jacques-Eric Gottenberg15, Lindsey A. Criswell12, Corinne Miceli-Richard1, Laurence Zitvogel16 and Xavier Mariette17. 1Universite´ Paris Sud, Le Kremlin Biceˆtre, France, 2IGR INSERM U1015, Villejuif, France, 3Pitie-Salpetriere Hospital, Paris, France, 4 Uppsala University, Uppsala, Sweden, 5Section of Rheumatology, Uppsala University, Uppsala, Sweden, 6Rheumatology/AIR, Linko¨ping University, Linko¨ping, Sweden, Linko¨ping, Sweden, 7Lund University, Malmo¨, Sweden, 8 University of Gothenburg, Gothenburg, Sweden, 9Stavanger University Hospital, Stavanger, Norway, 10Karolinska Institutet, Stockholm, Sweden, 11 University of Bergen, Bergen, Norway, 12University of California, San Francisco, Rosalind Russell Medical Research Center for Arthritis, San Francisco, CA, 13University of California, San Francisco, San Francisco, CA, 14 Oklahoma Medical Research Foundation, Oklahoma City, OK, 15Strasbourg University Hospital, Strasbourg, France, 16IGR INSERM U1015, Villejuif, France, 17Biceˆtre University Hospital, Le Kremlin Bicetre, France.

reduced gene transcription and function. We also demonstrated that circulating levels of NCR3/NKp30 were markedly increased among pSS patients compared with controls and correlated with higher NCR3/NKp30 IFN-␥ secretion by NK cells. Excess accumulation of NK cells in minor salivary glands correlated with the severity of the exocrinopathy. B7-H6, the ligand of NKp30, was expressed by salivary epithelial cells and regulated by TNF-␣ triggering NKp30 mediated-effector functions. Conclusion: These findings suggest that NK cells are involved in pSS pathogeny. Different levels of evidence (genetics, mRNA expression, function in blood, presence in the target organ as well as the ligand) demonstrate an NKp30-dependent inflammatory state in salivary glands. Blockade of the B7H6/NKp30 axis could be clinically relevant in pSS.

Conclusion: We identified OAS1 as a novel candidate SS locus that confers risk through a functional eQTL at rs10774671. This splice site variant switches the primary p46 isoform to multiple alternatives with decreased OAS1 enzyme activity potentially contributing to reduced ability to inhibit viral replication. These results indicate the risk allele may cause vulnerability to viral infection that contributes to SS susceptibility.

2774

Disclosure: H. Li, None; J. A. Ice, None; J. A. Kelly, None; I. Adrianto, None; S. B. Glenn, None; K. S. Hefner, None; E. G. Vista, None; D. U. Stone, None; R. Gopalakrishnan, None; G. D. Houston, None; D. M. Lewis, None; M. Rohrer, None; P. Hughes, None; J. B. Harley, None; C. G. Montgomery, None; J. Chodosh, None; J. A. Lessard, None; J. M. Anaya, None; B. M. Segal, None; N. L. Rhodus, None; L. Radfar, None; R. H. Scofield, None; C. J. Lessard, None; K. L. Sivils, None.

Background/Purpose: Sjo¨gren’s Syndrome (SS) is characterized by the destruction of the lacrimal and salivary grands due to autoreactive lymphocyte infiltration in the early phase and subsequent intralobular fibrosis and acinar atrophy at late phase of the disease. Involvement of plasmacytoid dendritic cells (pDCs) and mast cells (MCs) has been suggested, although their detailed role in pathology is poorly understood. The aim of this study was to investigate the role pDCs and MCs in the disease course of SS. Methods: Lip biopsy specimen from 178 SS patients and 7 Sicca syndrome (Sicca) patients were analyzed. NanoZoomer Digital Pathology (NDP), a system to convert a histology slide into high-resolution digital slides and enables sequential pathological analysis of multiple sections and measurement of accurate specimen area was utilized for analysis. Density of pDCs and MCs (/mm2) was measured with NDP and histological scoring was performed on lymphocytes infiltration, acinar atrophy and intralobular fibrosis. pDC, CXCL13 and CD68 were detected by immunohistochemistry and MCs were detected by toluidine blue staining. Histological stage was based on ‘Greenspan grade’. Results: Significant lymphocytes infiltration, acinar atrophy and intralobular fibrosis were observed in SS compared to Sicca patients with salivary gland destruction. Both, pDCs and MCs were significantly increased in SS compared to Sicca (12.6⫾10.2 vs 3.4⫾3.7, 43.7⫾21.8 vs 21.5⫾6.9). Positive correlation between density of pDCs with lymphocytes infiltration score (pee0.02) and density of MCs with intralobular fibrosis score (pee0.03) was observed in primary SS but not in secondary SS. Moreover, in the primary SS, density of CXCL13 positive infiltrating cells positively correlated with the degree of lymphocyte infiltration score and density of pDCs (p⬍0.001); by double immunostaining of antiCXCL13 and anti-CD68, we found that most of CXCL13 positive infiltrating cells are from Macrophage(57.8⫾22.1%) and the localization of PDC and CXCL13 positive infiltrating cells are almost same around duct. On the other hand, no correlation was observed with secondary SS patients. Conclusion: Our data suggests the specific role of pDCs in lymphocyte infiltration in primary SS. Correlation of CXCL13 expression, a chemoattractant for lymphocyte recruitement, with lymphocyte infiltration and density of pDCs further suggests the important role of pDCs. CXCL13 is induced by type 1 IFN, whereas pDCs are the major source of type 1 IFN. Therefore, it likely that type 1 IFN produced by pDCs in the early stage of the disease induces CXCL13 production of infiltrating cells, especially Macrophage, contributing to lymphocyte infiltration constructing germinal center. In addition, MCs seems to be involved in intralobular fibrosis which occurs in the mid to late stage of SS.

2773

Tuesday, October 29

Genome-Wide DNA Methylation Patterns In naı¨ve CD4ⴙ T Cells From Patients With Primary Sjo¨gren’s Syndrome. Nezam I. Altorok1, Patrick S. Coit1, Travis Hughes1, Kristi A. Koelsch2, R. Hal Scofield3, Kathy L. Sivils3, A. Darise Farris4 and Amr H. Sawalha1. 1University of Michigan, Ann Arbor, MI, 2Oklahoma Medical Research Foundation, Okalahoma City, OK, 3 Oklahoma Medical Research Foundation, Oklahoma City, OK, 4Oklahoma Medical Research Foun, Oklahoma City, OK. Background/Purpose: Primary Sjo¨gren’s syndrome (PSS) is a systemic autoimmune disease characterized by inflammation of the lacrimal and salivary glands and dryness of the eyes and mouth. There is evidence to suggest that PSS shares common pathogenic factors with other autoimmune diseases such as lupus, including common genetic susceptibility loci. Recent evidence strongly supports an epigenetic contribution to the pathogenesis of lupus, however, very little is known about the epigenetics of PSS. Methods: We performed a genome-wide DNA methylation study in naı¨ve CD4⫹ T cells in eleven PSS patients before receiving any treatment compared to age-, sex-, and ethnicity-matched healthy controls. Naı¨ve CD4⫹ T cells were isolated from PBMCs by FACS or indirect labeling and magnetic bead separation. Cell purity was ⬎ 95%. DNA was isolated, and treated with sodium bisulfite. Cytosine methylation was quantified in more than 485,000 CpG sites that cover 99% of RefSeq genes, with an average of 17 CpG sites per gene region using the Illumina Infinium HumanMethylation450 BeadChip array. Differentially methylated CpG sites between PSS patients and controls with a fold difference ⱖ 1.2 were identified. A false discovery rate (FDR) of 5% was applied to correct for multiple testing and differential methylation was considered statistically significant if the FDR corrected P value was ⱕ 0.01. Results: We identified 553 hypomethylated and 200 hypermethylated CpG sites in naı¨ve CD4⫹ T cells from PSS patients compared to healthy matched controls, representing 311 hypomethylated and 115 hypermethylated gene regions. Hypomethylated genes in PSS include LTA, coding for Lymphotoxin ␣, which is involved in LT␤ receptor signaling pathway, activation of follicular dendritic cells, and expression of interferon ␣. Other relevant genes such as GSTM1, CD247, TNFSF25, PTPRC and PDCD1 were also hypomethylated. The interferon signature pathway was represented by hypomethylation of STAT1, IFI44L, and IFITM1. A group of genes encoding for members of the solute carrier proteins, which are membrane transport proteins that are important for maintenance of cell function were hypomethylated (SLC11A1, SLC11A2, SLC22A23, SLC25A25, ALC25A3, SLC25A33, SLC6A20), whereas, SLC9A1, which is important for the maintenance of PH homeostasis was hypermethylated in PSS patients compared to controls. In addition, the transcription factor RUNX1 was hypermethylated in patients, suggesting an impact on the differentiation of hematopoietic stem cells, and possible connection to lymphoma predisposition. Gene ontology (GO) analysis of hypomethylated genes demonstrated enrichment of genes involved in lymphocyte activation (P⫽ 1.10E-04), and immune response (P⫽ 2.20E-03). GO terms for hypermethylated genes included antigen processing and presentation (P⫽ 8.00E-06), and positive regulation of RNA metabolic process (P⫽ 2.10E-02). Conclusion: This is the first epigenome-wide DNA methylation study in PSS. Our data suggest wide-spread DNA methylation changes in naı¨ve CD4⫹ T cells in PSS, and highlight several genes and pathways that may be involved in the pathogenesis of this disease and that will be the subject for our future investigation. Disclosure: N. I. Altorok, None; P. S. Coit, None; T. Hughes, None; K. A. Koelsch, None; R. H. Scofield, None; K. L. Sivils, None; A. D. Farris, None; A. H. Sawalha, None.

Distinct Role Of Plasmacytoid Dendritic Cells and Mast Cells In The Pathogenesis Of Sjo¨gren’s Syndrome. Jidong Zhao, Kunihiro Yamaoka, Satoshi Kubo, Shingo Nakayamada and Yoshiya Tanaka. University of Occupational and Environmental Health, Japan, Kitakyushu, Japan.

Disclosure: J. Zhao, None; K. Yamaoka, None; S. Kubo, None; S. Nakayamada, None; Y. Tanaka, None.

2775 Adenosine A2b Receptor Agonist Bay60–6583 Restores Salivary Gland Function In a Mouse Model For Sjo¨gren’s Syndrome. Barbara Szczerba, Paulina Rybakowska, Paromita Dey, Harini Bagavant and Umesh Deshmukh. University of Virginia, Charlottesville, VA. Background/Purpose: Primary Sjo¨gren’s syndrome (pSS) is a chronic autoimmune disorder mainly affecting the exocrine glands. However, it is now clear that both immune and non-immune mechanisms are responsible salivary gland dysfunction in this disorder. Thus, drugs capable of inducing immunosuppression as well as directly influencing salivary gland function would be highly desirable as therapeutics for SS. Adenosine is a major anti-inflammatory metabolite generated during tissue injury and it signals through four distinct G protein-coupled receptors, termed A1AR, A2aAR, A2bAR and A3AR. In addition it can have direct effects on different aspects of cellular metabolism. Thus, in this study we have tested the efficacy of adenosine A2b receptor agonist Bay60–6583 for restoration of salivary gland function in a mouse model for SS-like disorder.

S1186

Methods: The B6.Aec1Aec2 mouse model for SS-like disorder was used for this study. Gene expression of adenosine receptors on human and mouse submandibular gland cell lines, as well as mouse salivary glands was determined by real time PCR. ERK phosphorylation in cell lines treated with the drug and presence of serum autoantibodies were analyzed by western blotting. Mice with established glandular dysfunction were treated with the drug, 8 times, every third day by intraperitoneal route. Saliva production was measured following pilocarpine injection. Cytokine levels in sera were measured using Bioplex assay system. Formalin fixed salivary gland sections were stained with H and E and presence of inflammatory cell infiltrates quantified by Stereology. Results: All adenosine receptors were expressed in the mouse salivary glands and in the mouse and human cell lines. Bay60–6583 readily induced ERK phosphorylation in the human and mouse cell lines. Mice treated with Bay60–6583 showed significant improvement in their saliva production than control (vehicle treated) mice. Salivary glands from drug treated mice showed significantly lower inflammatory cell infiltrates. Surprisingly, the inflammatory cytokine and autoantibody levels between the vehicle and drug treated mice were not different. Conclusion: Our study has established for the first time the presence of functional adenosine receptors on salivary gland cells. The restoration of glandular function in drug treated mice demonstrates the potential of targeting adenosine receptors for therapeutic intervention in SS. Our study also demonstrates that in this mouse model, circulating inflammatory cytokine and autoantibody levels were dissociated from glandular dysfunction. This finding suggests that in some SS patients, biomarkers other than circulating cytokines and autoantibodies might be more relevant for following disease progression. Disclosure: B. Szczerba, None; P. Rybakowska, None; P. Dey, None; H. Bagavant, None; U. Deshmukh, None.

Conclusion: This large nationwide primary care cohort demonstrates an increased risk of CVD among AS patients which is evident from the time of diagnosis. Clinicians should be aware of the excess CVD risk in this patient group which may help inform their overall management, with the view to optimize long-term quality of life. Disclosure: L. E. Dean, None; G. J. Macfarlane, Abbott Laboratories, 2, Pfizer Inc, 2; A. G. MacDonald, None; G. T. Jones, Abbott Laboratories, 2, Pfizer Inc, 2.

2776 Ankylosing Spondylitis Is Associated With An Increased Likelihood Of Cardiovascular Disease: A Nationwide Matched Cohort Study In Primary Care. Linda E. Dean1, Gary J. Macfarlane1, Alan G. MacDonald2 and Gareth T. Jones1. 1University of Aberdeen, Aberdeen, United Kingdom, 2 Aberdeen Royal Infirmary, Aberdeen, United Kingdom. Background/Purpose: Cardiovascular disease (CVD) risk is well established in rheumatoid arthritis but less so in ankylosing spondylitis (AS). Due to a similar inflammatory profile, AS patients may be expected to be at increased risk of CVD, particularly ischemic heart disease, hyperlipidaemia and hypertension. Thus, the aim of this study was to determine whether there is an increased risk of CVD among persons with AS. Methods: Design: matched cohort study. The Scottish Primary Care Clinical Informatics Unit collects demographic, diagnostic and clinical information from a nationally representative sample of general practices in Scotland, UK. As almost all UK residents are registered with a general practitioner and referrals to specialist services are made via this route, the database provides a source of nationwide health data. In April 2007, the database consisted of 3,256,108 patients, approximately 60% of the Scottish population. The ‘exposed’ cohort comprised all patients with a prior clinical diagnosis of AS (indicated by disease-specific codes). These patients were matched at a ratio of 4:1 to non-AS patients by age (⫹/⫺ 1yr), gender and area-level deprivation. AS patients were followed up from date of diagnosis until either a CVD diagnosis or the end of follow-up period. Non-AS patients were assigned the same index date as their matched exposed patient and followed to the same end point. All those with a CVD diagnosis prior to the index date were excluded. Data was extracted on the following CVD outcomes: ischemic heart disease; congestive heart failure; cerebrovascular disease; cardiomyopathy; pericarditis; valvular disease; conduction disturbances; hyperlipidaemia; hypertension; atherosclerosis; and heart valve replacement. Differences in

2777 Ankylosing Spondylitis and Non-Radiographic Axial Spondyloarthritis: Part of a Common Spectrum Or Distinct Diseases? Analysis of a Longitudinal Prospective Cohort. Dinny Wallis1, Nigil Haroon1, Renise Ayearst2, Adele Carty1 and Robert D. Inman1. 1Toronto Western Hospital, University of Toronto, Toronto, ON, 2University of Toronto, Toronto Western Hospital, Toronto, ON. Background/Purpose: Historically ankylosing spondylitis (AS) has been defined by the modified New York classification criteria which require the presence of radiographic sacroiliitis. More recently the use of MRI has led to the identification of patients with features of axial spondyloarthritis (SpA) who do not fulfill the modified New York criteria (non-radiographic axial SpA, nr-axSpA). The natural history of axial SpA is not completely understood and current knowledge is drawn largely from European data. We aimed to compare the features of AS and nr-axSpA in a North American cohort. Methods: Data were analyzed for all patients enrolled in a longitudinal spondyloarthritis cohort between January 2003 and December 2012 meeting modified New York criteria for AS or the ASAS classification criteria for nr-axSpA. Categorical variables were compared using Fisher’s exact test with two-tailed p values. For continuous variables, a mean patient value was calculated based on all visits for that patient. Variables were compared using t-tests. Results: 639 patients with AS and 73 patients with nr-axSpA were included. Of the nr-axSpA patients, 40 demonstrated inflammation on MRI, and 33 were classified according to ASAS clinical criteria for axial SpA (of whom 15 had MRI showing no inflammation and 18 did not have MRI). The proportion of male patients was higher in AS than in nr-axSpA (76.2% vs. 47.9%, p⬍0.0001). CRP and ESR levels were higher in AS than nr-axSpA (CRP 11.4 vs. 5.2, p⬍0.0001; ESR 13.7 vs. 9.9, p⫽0.018). Disease duration at last clinic visit was shorter in nr-axSpA than AS

S1187

Tuesday, October 29

ACR Concurrent Abstract Session Spondylarthropathies and Psoriatic Arthritis: Clinical Aspects and Treatment: Clinical Features of Spondyloarthritis Tuesday, October 29, 2013, 4:30 PM–6:00 PM

the occurrence of CVD between cohorts were examined using (a) Kaplan-Meier survival curves, and (b) Cox proportional hazards regression. Results were adjusted for smoking status and summarised as hazard ratios (HR). Results: 1,964 AS patients, plus 7,856 matched non-AS patients, were identified (mean age 62yrs, 77% male). Median follow-up time was 16yrs (IQR 8–26). The risk of CVD, over time, is shown in Figure 1. Persons with AS were significantly more likely to receive a diagnosis of CVD than individuals without AS (HR 1.15; 95%CI 1.05–1.26) and, in particular, experienced a 20% increase in the risk of hypertension (1.20; 1.07–1.34). No significant individual associations were found with any of the other CVD diagnoses under investigation, including ischemic heart disease, heart failure or hyperlipidemia.

Tuesday, October 29

(12.1y versus 17.7y, p⫽0.0002), providing some indirect support to the notion that nr-axSpA may represent a subset of axial SpA seen earlier in the course of the disease than AS. The proportions of patients receiving biologic therapy, non-steroidal anti-inflammatory drugs, disease modifying anti-rheumatic drugs and glucocorticoids were similar in AS and nr-axSpA. Further investigation of gender, inflammation and radiographic severity in the combined axial SpA cohort revealed that 31.9% of patients with a normal CRP were female, compared to only 14.6% of patients with an elevated CRP (p⬍0.0001). 96.6% patients with an elevated CRP had radiographic AS, compared to 86.8% of those with a normal CRP (p⬍0.0001). When data were analyzed for only the female patients, the difference in acute phase response between AS and nr-axSpA lost significance (CRP 9.4 versus 5.2, p⫽0.09; ESR 16.3 versus 11.5, p⫽0.07).

Conclusion: Analysis of this North American SpA cohort has identified some key differences between subsets of axial SpA, and highlights the influence of gender on inflammation and radiographic severity, thereby influencing the clinical classification of axial SpA patients. Disclosure: D. Wallis, Janssen Pharmaceutica Product, L.P., 2; N. Haroon, Janssen Pharmaceutica Product, L.P., 5, Pfizer Inc, 5, Amgen, 5, Abbott Laboratories, 5; R. Ayearst, None; A. Carty, None; R. D. Inman, Abbvie, Janssen, Pfizer, UCB, 5.

2778 Mortality In United States Veterans With The HLA-B27 Gene. Jessica Walsh1, Brian C. Sauer1, Daniel O. Clegg1, Grant W. Cannon2, Xi Zhou1 and Chiachen Teng1. 1George E. Wahlen Veteran Affairs Medical Center, Salt Lake City, UT, 2Salt Lake City VA and University of Utah, Salt Lake City, UT. Background/Purpose: HLA-B27 (B27) is carried by in 6–8% of Americans and is strongly associated with spondyloarthritis (SpA). SpA has been associated with premature mortality. However, increased mortality was not suspected in B27 positive (B27⫹) individuals without SpA, until recently published data from the 2009 National Health and Nutrition Examination Survey reported a lower prevalence of B27 in randomly selected people older than 50, compared to younger people (7.3% vs. 3.6%, OR 0.4, CI 0.3–0.8). The purpose of this study was to better define the relationship between the B27 gene and mortality, by comparing mortality in B27⫹ and B27- veterans with clinically available B27 test results. Methods: The Corporate Data Warehouse was used to identify veterans with available B27 test results. Mortality differences between B27⫹ and B27- veterans were compared with Cox proportional hazard ratios. Logistic regression was used to evaluate the relationships between death and covariates including age at B27 testing, sex, race, and SpA diagnoses codes from rheumatology encounters. Results: Among 17,209,732 veterans, 32,327 had both available vital status data and B27 test results (Figure 1). The mean age at testing was 50.5 for B27⫹ and 50.1 for B27- veterans. Male gender was recorded in 92.9% of B27⫹ veterans and 88.3% of B27- veterans (Table 1). The hazard ratio comparing mortality in B27⫹ and B27- veterans was 1.16 (CI 1.04 – 1.30), after adjustment for age at B27 testing, sex, race, and SpA codes (Table 2).

Table 1. Demographics, B27 testing, & spondylorthritis

Age at B27 testing (mean) Male gender Race White Black Other* Unknown/declined to answer Years between B27 testing & study end or death (mean) SpA ICD-9 code from a rheumatology encounter

B27ⴚ n ⴝ 27,398 ⴞ SD or (%)

B27 ⴙ n ⴝ 4929 ⴞ SD or (%)

50.1 ⫾ 14.6 24,201 (88.3)

50.5 ⫾ 14.3 4578 (92.9)

19,321 (70.5) 5146 (18.8) 445 (1.6) 2486 (9.1) 5.1 ⫾ 3.3

3823 (77.6) 433 (8.8) 92 (1.9) 581 (11.8) 5.3 ⫾ 3.3

2381 (8.7)

1717 (34.8)

*Other races included American Indian or Alaska Native, Asian, Native Hawaiian or Other Pacific Islander

Table 2. Mortality in B27⫺ and B27⫹ veterans

B27⫺ B27⫹

Person time (Years after B27 testing)

Deceased (%)

Adjusted HR* (95% CI)

P

140,120 26,060

1981 (7.2) 411 (8.3)

Ref. 1.16 (1.04–1.30)

0.01

*Adjusted for age at B27 testing, sex, race, & presence of SpA diagnoses codes

Conclusion: Mortality was higher in B27⫹ veterans than B27- veterans, and the difference was not explained by age, sex, race, or SpA diagnoses codes. The mortality difference was likely conservative because the B27group may have been disproportionately enriched with conditions mimicking SpA that are associated with increased mortality, such as chronic pain or rheumatoid arthritis. Additional research is required to estimate survival in unselected individuals from the general population and to identify specific disease processes that contribute to premature mortality. Recognizing these diseases may provide mechanistic insights into B27 functions and may lead to interventions that improve survival in B27⫹ individuals. Disclosure: J. Walsh, None; B. C. Sauer, None; D. O. Clegg, None; G. W. Cannon, None; X. Zhou, None; C. Teng, None.

S1188

2780

A Novel Evidence-Based Detection Of Undiagnosed Spondyloarthritis In Patients Presenting With Acute Anterior Uveitis: The DUET (Dublin Uveitis Evaluation Tool) Algorithm. Muhammad Haroon1, Michael Anthony O’Rourke2, Pathma Ramasamy2, Conor Murphy2 and Oliver FitzGerald1. 1Dublin Academic Medical Centre, St. Vincent’s University Hospital, Dublin, Ireland, 2Royal Victoria Eye and Ear Hospital, Dublin, Ireland.

Cumulative Exposure To Elevated Inflammatory Markers Is Associated With Increased Burden Of Atherosclerosis In Psoriatic Arthritis Patients: A Cohort Study. Lihi Eder1, Arane Thavaneswaran1, Vinod Chandran1, Richard J. Cook2 and Dafna D. Gladman1. 1University of Toronto, Toronto Western Hospital, Toronto, ON, 2University of Waterloo, Waterloo, ON.

Background/Purpose: The prevalence of mostly undiagnosed underlying Spondyloarthropathy (SpA) is more than 60% among patients presenting with acute anterior uveitis (AAU). To date, there are no formal guidelines or referral pathways for AAU patients developed or endorsed by any international or national societies. The objectives of our study were: (1) to investigate the prevalence of undiagnosed SpA in patients presenting with idiopathic AAU; (2) to develop and validate an assessment algorithm for referral from Ophthalmologists of appropriate AAU patients to Rheumatology that will aid the early diagnosis of the SpA Methods: All consecutive patients attending emergency department of local ophthalmology hospital with AAU, but who did not have a known diagnosis of SpA, were eligible to partake in this study. Patients with any other known cause of AAU were excluded. DEVELOPMENT COHORT: The rheumatologic referral was made as per a test algorithm (Figure-1), and those who did not require a referral, remained part of the study as a control group; these patients also underwent detailed rheumatologic evaluation. VALIDATION COHORT: To confirm the findings from the development cohort, we recruited a validation cohort with similar entry and exclusion criteria. Algorithm version-1 (revised form of test algorithm which is now named the DUET algorithm, figure-1) was used in this cohort to identify patients requiring rheumatologic referral. Results: DEVELOPMENT COHORT: 104 consecutive patients from September 2011 through to June 2012 were recruited. However, 3 of these patients were lost to follow up prior to rheumatologic evaluation. After rheumatologic evaluation of the entire cohort (n⫽101), 41.6% (n⫽42) had undiagnosed SpA as per ASAS classification criteria. Our test algorithm was noted to have: sensitivity 100%, specificity 53.5%, PPV 61% and NPV 100%. Further regression analysis resulted in the development of the DUET algorithm which made the following improvements in the assessment of the development cohort: sensitivity 95%, specificity 98%, PPV 97.5%, NPV 96.6, positive LR 56.19, and negative LR 0.04. VALIDATION COHORT: To obtain a 95% confidence interval of at least as narrow as ⫾10% or ⫾8% for all the statistics, a sample size of 44 or 69 participants, respectively, was deemed sufficient. Hence, consecutive 74 idiopathic AAU patients were recruited from November 2012 through to April 2013, but 2 of these patients were lost to follow-up. After rheumatologic evaluation of the cohort (n⫽72), 40% (n⫽29) were diagnosed with SpA, with the following performance of DUET algorithm - sensitivity 96%, specificity 97%, PPV 96.5, NPV 97.6, positive likelihood ratio 41.5 and negative likelihood ratio of 0.03

Background/Purpose: Cardiovascular morbidity is increased in patients with psoriatic arthritis (PsA). Traditional cardiovascular risk factors do not fully explain this excessive risk. It is unclear whether the cumulative burden of inflammation in the skin and the joints contribute to the development of atherosclerotic plaques. We aimed to investigate whether higher burden of arthritis and psoriasis over time is associated with the development of atherosclerotic plaques among patients with PsA. Methods: A retrospective cohort analysis was conducted in patients attending a large PsA clinic. Patients were assessed at 6–12 month intervals according to a standard protocol. Information about demographics, medical history, musculoskeletal and skin examination, patient-reported outcomes and laboratory tests was collected at each visit. The cumulative effect of inflammation was measured by a time-adjusted arithmetic mean (AM) of all available measurements from the first visit to the clinic. The following variables were considered as predictors: Psoriasis Activity and Severity Index (PASI), Erythrocyte Sedimentation Rate (ESR), total leukocyte counts (TLC), Tender and Swollen joint count (TJC, SJC), C - reactive protein (CRP), Psoriatic Arthritis Disease Activity Score (PASDAS) and Disease Activity for PsA (DAPSA). Ultrasound assessment of the carotid arteries was performed and Total Plaque Area (TPA) was measured. This measure represented the extent of atherosclerosis and was considered the outcome of interest. TPA was stratified into 4 categories: 1) TPA⫽0 (no plaques), 2) 0⬍TPAⱕ0.1 cm2, 3) 0.1⬍TPAⱕ0.4 cm2, 4) TPA⬎0.4 cm2. The association between the various AM variables and TPA categories was assessed using logistic ordinal regression model adjusted age and gender. Results: A total of 235 patients with PsA were analyzed. Their mean age was 55.4⫾11.4 years and the duration of psoriasis and PsA were 27⫾13.5 and 16.2⫾11.7 years, respectively. 56.6% of the participants were males. Patients in higher TPA categories were older at the time of assessment (p⬍0.001), were more likely to be smokers (p⫽0.008), hypertensive (p⫽0.002), diabetics (p⬍0.001) and were older at the onset of psoriasis (p⬍0.001) and PsA (p⬍0.001). In a multivariable regression model adjusted for age and sex AM-ESR was associated with higher TPA categories (category 3 vs. 1 Odds Ratio (OR) 1.04, 95% Confidence Interval (CI) 1, 1.08, p⫽0.04, category 4 vs. 1 OR 1.05, 95% CI 1.01, 1.1, p⫽0.02). The following variables were also associated with higher probability of having severe atherosclerosis (being in TPA category 4 vs. 1): AM-TLC (OR 1.45, 95% CI 1.09, 1.92, p⫽0.01) and AM-DAPSA (OR 1.08, 95% CI 1.02, 1.14, p⫽0.009). A trend for an association was observed between AM-PASDAS and being in the highest TPA category (OR 1.73, 95% CI 0.97, 3.09, p⫽0.06). No significant association was found between AM-PASI, AM-CRP, AMTJC, AM-SJC and damaged joint count and TPA category in multivariate analysis. Conclusion: An exposure to increased cumulative inflammation is associated with development of atherosclerotic plaques among patients with PsA. Disclosure: L. Eder, None; A. Thavaneswaran, None; V. Chandran, None; R. J. Cook, None; D. D. Gladman, None.

2781 The Disease Characteristics and Predictors Of Minimal Disease Activity On TNF Blockers- Results From A Longitudainal Observational Cohort. Amir Haddad1, Arane Thavaneswaran1, Ioana Ruiz Arruza1, Vinod Chandran1, Richard J. Cook2 and Dafna Gladman1. 1University of Toronto, Toronto Western Hospital, Toronto, ON, 2University of Waterloo, Waterloo, ON.

Figure 1.

Conclusion: Approximately 40% of patients presenting with idiopathic AAU have undiagnosed SpA. A simple to apply algorithm is described with excellent sensitivity and specificity. Disclosure: M. Haroon, None; M. A. O’Rourke, None; P. Ramasamy, None; C. Murphy, None; O. FitzGerald, UCB, PFIZER, ABBOTT, ROCHE, MSD, BMS, 2, Janssen Pharmaceutica Product, PFIZER, ABBOTT, ROCHE, BMS, MSD, 5, Pfizer, ABBOTT, UCB, ROCHE, JANSSEN, 8.

Background/Purpose: A state of Minimal Disease Activity (MDA) has been defined and validated as a target for treatment in PsA. The purpose of the study is to identify disease characteristics and predictors of MDA in PsA patients treated with TNF blockers. Methods: Patients fulfilled the CASPAR criteria and were followed regularly every 3–6 months and completed radiographic evaluation at 2-year intervals. TNF␣ blockers were prescribed when patients failed standard of care. Patients were considered in MDA when met at least 5/7 of the criteria defined by Coates et al[i].Sustained MDA was defined as MDA lasting for

S1189

Tuesday, October 29

2779

Tuesday, October 29

ⱖ 12 months. Patients achieving MDA were compared to patients who did not achieve MDA. A proportional odds discrete time survival analysis model was applied adjusting for gender, age at visit, duration of PsA, presence of abnormal ESR and damaged joint count at each visit to identify predictors for MDA. Results: Of the 306 patients treated with anti-TNF agents 23 were in MDA at baseline and 57 were prescribed anti-TNF agents prior to enrolment. 226 were in non-MDA state and constituted the study population. 145/226 patients achieved MDA after an average of 1.30 (1.68) years and for a mean duration of 3.46 (2.25) years. Patients who achieved MDA were more likely males (71.9% vs. 55.2%), younger at diagnosis (34.4 vs. 39.0 years), had a lower actively inflamed joint count (7.3 vs. 15.1), dactylitis (9.1% vs 13.9%), enthesitis (10.8% vs. 17.4% and tenosynovitis (9.2% vs. 21.7%), but more likely to have clinical damage (72.7% vs. 52.2%) and higher modified Steinbrocker score (25.5 vs. 11.5) as well as axial involvement (46.2% vs. 25%) at baseline compared to patients that didn’t achieve MDA. Non-MDA patients had higher BMI (31.3 vs. 28.8), more were classified with functional class III/IV (33.3% vs. 10.1%) and had a lower SF-36 physical (28.1 vs. 43.9) and mental (41.9 vs. 49.9) component summary scores at baseline compared to the MDA group. HLA B*27 was found in 24.8% patients with MDA compared to 9.2% patients who didn’t achieve MDA (P⫽0.007). The majority of patients in both groups were also treated with DMARDs and NSAIDs. No significant difference was found in disease duration, alcohol use, smoking, ESR, CRP or PASI score between the two groups. Patients who achieved MDA had less radiographic progression over time compared to patients without MDA. The survival analysis showed that after adjusting for characteristics at each visit male gender increased the odds of achieving MDA (OR⫽1.68 CI 1.11, 2.53, P⫽0.01), whereas an abnormal ESR lowered the odds of achieving MDA (OR⫽0.46 CI 0.27, 0.80, P⫽0.06). As for sustained MDA, only normal ESR was found to a predictor (OR 0.06 CI 0.03–0.15, P⬍0.0001). Conclusion: 63% of the study group achieved MDA after an average duration of 1.3 years and those patients had less active disease at baseline but more damage with less radiographic progression over time. A normal ESR at each visit and male gender were predictors of MDA in patients treated with TNF blockers, the presence of HLA B*27 has a prognostic value also in identifying patients who achieve MDA. [i] Ann Rheum Dis. 2010 Jan;69(1):48–53) Disclosure: A. Haddad, None; A. Thavaneswaran, None; I. Ruiz Arruza, None; V. Chandran, None; R. J. Cook, None; D. Gladman, Abbott, Bristol-Myers Squibb, Celgene, Johnson & Johnson, MSD, Novartis, Pfizer, UCB Pharma, 2, Abbott, Bristol-Myers Squibb, Celgene, Johnson & Johnson, MSD, Novartis, Pfizer, UCB Pharma, 5.

for remission induction. Patients who suffered a severe disease flare (BVAS/ WG ⬎ 3 or one major BVAS/WG item) between 6 and 18 months were eligible for open label RTX (OLR) (375mg/m2 once weekly times four). Results: 17 patients received two courses of RTX. Baseline characteristics are presented in the table. Characteristics of patients treated with open label RTX N ⴝ 17 Originally assigned to RTX PR3-ANCA positive GPA Newly diagnosed disease Mean time to OLR (range, days) BVAS/WG at OLR (range) Mean prednisone dose at OLR (range, mg) Detectable B-cells at flare Rising ANCA at flare

16 (94%) 14 (82%) 15 (88%) 6 (35%) 367 (225–556) 4.8 (3–11) 8.4 (0–40) 16 (94%) 14 (82%)

After retreatment, patients were followed for an average of 301 days (range 35–427). Treatment with OLR led to remission (BVAS/WG⫽0) in 15 of 17 patients (88%) by an average of 55 days (range 1–181). One patient with diffuse alveolar hemorrhage did not improve and died 7 weeks after the initial flare. Another patient reached a BVAS/WG of 1 before suffering a limited flare at 12 months after OLR. Six months after OLR, 15 patients (88%) were in remission (BVAS/ WG ⫽ 0), 8 (47%) had achieved complete response (BVAS/WG ⫽ 0 and prednisone ⱕ 10mg/day) and 6 patients (35%) were in complete remission (BVAS/WG ⫽ 0 and prednisone ⫽ 0). After 12 months, 13 patients (76%) had achieved complete responses and 8 (47%) had reached complete remission. There were 4 limited and no severe flares among the 17 patients (BVAS/WG 2.5) over one year of follow up after OLR. There were a total of 3 severe (grade ⱖ 3) adverse events after OLR, including one death (described above), metastatic colon cancer, and severe sinusitis. Conclusion: This prospective study indicates that retreatment of GPA or MPA flares with rituximab is effective in re-inducing remission. Disclosure: E. Miloslavsky, Genentech Inc, 9; U. Specks, None; P. A. Merkel, None; P. Seo, None; R. F. Spiera, Roche Pharmaceuticals, g, 2; C. A. Langford, BristolMyers Squibb, 9, Genentech and Biogen IDEC Inc., 9; G. S. Hoffman, None; C. G. M. Kallenberg, Roche, 8; E. W. St Clair, None; N. Tchao, None; L. Ding, None; D. Ikle, Rho, 3; B. Jepson, Rho, 3; P. Brunetta, Genentech Inc, 3; J. H. Stone, Genentech and Biogen IDEC Inc., 2, Genentech and Biogen IDEC Inc., 5, Roche Pharmaceuticals, 2.

2783 ACR Concurrent Abstract Session Vasculitis III Tuesday, October 29, 2013, 4:30

PM–6:00 PM

2782 Retreatment With Rituximab In The Rituximab In ANCA-Associated Vasculitis (RAVE) Trial. Eli Miloslavsky1, Ulrich Specks2, Peter A Merkel3, Philip Seo4, Robert F. Spiera5, Carol A. Langford6, Gary S. Hoffman7, Cees G.M. Kallenberg8, E. William St Clair9, Nadia Tchao10, Linna Ding11, David Ikle12, Brett Jepson12, Paul Brunetta13 and John H. Stone14. 1Massachusetts General Hopsital, Boston, MA, 2Mayo Clinic, Rochester, MN, 3University of Pennsylvania and VA Medical Center, Philadelphia, PA, 4Johns Hopkins Vasculitis Center, Baltimore, MD, 5Hospital for Special Surgery, New York, NY, 6Cleveland Clinic, Cleveland, OH, 7Cleveland Clinic Foundation, Cleveland, OH, 8University Medical Center Groningen, Groningen, Netherlands, 9Duke Unversity Medical Center, Durham, NC, 10Immune Tolerance Network, Bethesda, MD, 11NIAID, Bethesda, MD, 12 Rho, Chapel Hill, NC, 13Genentech, So San Francisco, CA, 14Massachusetts General Hospital, Boston, MA. Background/Purpose: Retrospective studies have demonstrated that repeat rituximab treatment may be effective in re-inducing remission in relapsing ANCA-associated vasculitis. We analyzed data from the Rituximab in ANCA-associated vasculitis (RAVE) trial in order to determine the safety and efficacy of a second course of rituximab for relapsing granulomatosis with polyangiitis (GPA) or microscopic polyangiitis (MPA). Methods: Randomized controlled trial comparing rituximab (RTX, n⫽99) to cyclophosphamide (CYC) followed by azathioprine (AZA, n⫽98)

Rituximab Versus Azathioprine For Maintenance In Antineutrophil Cytoplasmic Antibodies-Associated Vasculitis: Follow Up At 39 Months. Benjamin Terrier1, Christian Pagnoux2, Alexandre Karras3, Chahera Khouatra4, Olivier Aumaıˆtre5, Pascal Cohen1, Francois Maurier6, Olivier Decaux7, He´le`ne Desmurs-Clavel8, Pierre Gobert9, Thomas Quemeneur10, Claire Blanchard-Delaunay11, Pascal Godmer12, Xavier Puećhal13, Luc Mouthon14 and Loic Guillevin15. 1Cochin University Hospital, Paris, France, 2Division of Rheumatology, Mount Sinai Hospital, Toronto, ON, 3Hoˆpital Europeén Georges Pompidou, APHP, Paris, France, 4CHU Lyon, Lyon, France, 5 Division of internal Medicine, Centre Hospitalier Universitaire, Hoˆpital Gabriel Montpied, Clermont–Ferrand, Clermont–Ferrand, France, 6Division of internal Medicine, CHR Metz, Metz, Metz, France, 7Rennes University Hospital, Rennes, France, 8Hospices Civils de Lyon, Hoˆpital Louis Pradel, Lyon, France, 9Centre Hospitalier d’Avignon, Avignon, France, 10CHR de Valenciennes, Valenciennes, France, 11Hoˆpital de Niort, Niort, France, 12 Department of Internal Medicine, Centre Hospitalier Bretagne Atlantique de Vannes, Vannes, France, 13Hoˆpital Cochin, AP-HP, Universite´ Paris Descartes, Sorbonne Paris Cite´, Paris, France, 14Department of Internal Medicine, Referral Center for Rare Autoimmune and Systemic Diseases, Hoˆpital Cochin, AP–HP, Universite´ Paris Descartes, Paris, France, Paris, France, 15Division of Internal Medicine, Hoˆpital Cochin, University Paris Descartes, Paris, France. Background/Purpose: Rituximab was shown to be as effective as cyclophosphamide to induce remission in patients with ANCA-associated vasculitis (AAV). The prospective, randomized, controlled MAINRITSAN trial compared rituximab (RTX) to azathioprine (AZA) to maintain ANCAassociated vasculitis (AAV) remission. Once remission was obtained with a conventional regimen using corticosteroids and cyclophosphamide, patients were randomly assigned to receive a 500-mg RTX infusion on D1, D15, 5.5

S1190

months later, then every 6 months for a total of 5 infusions over 18 months, or AZA for 22 months at the initial dose of 2 mg/kg/d. The primary endpoint was the major relapse rate (EULAR/ACR criteria) at 28 months. This study demonstrated that 500 mg of rituximab (RTX) every 6 months was superior to azathioprine (AZA) to maintain ANCA-associated vasculitis (AAV) remission during the 28-month follow-up, with a similar profile of tolerance. This study describes the extended follow-up of patients included in the MAINRITSAN trial. Methods: Extended follow-up was ascertained from the patients included in the trial. Data on relapse and survival were collected from physician records. All patients were analyzed according to the group to which they were randomized, except for 8 patients who were excluded from the extended follow-up because of severe violation to protocol. Results: One hundred and nine patients were analyzed for the extension of follow-up. Median duration of follow-up was 38.6 months (IQR, 33.5–45.2 months). Seven out of 55 (12.7%) patients in the RTX arm and 26/54 (48.1%) patients in the AZA arm had at least one major relapse. Overall, the risk of major relapse remained significantly lower in the RTX arm compared to the AZA arm (hazard ratio 0.20, 95% CI 0.11 to 0.45, P⬍0.0001). During follow-up, 3 patients died in the AZA arm but none in the RTX arm. Causes of death were infection, cancer and mesenteric ischemia. Overall survival rate was better in the RTX arm compared to the AZA arm (P⫽0.07).

initially received corticosteroids (CS) alone. Because some patients required add-on therapies (AT) during follow-up, baseline characteristics associated with their use were sought. Methods: Patients’ data were updated in 2012. Chapel Hill definitions classified EGPA, PAN and MPA. Analyzed AT were all cytotoxic agents, biologics (except omalizumab), intravenous immunoglobulins (IVIg) (⬎2 g/kg) and plasma exchange. Univariate and multivariate analyses were performed. Results: The study included 193 patients (75 EGPA, 61 MPA and 57 PAN) initially treated with CS alone. Mean⫾SD overall follow-up was 97.6⫾39.6 months, with no difference among entities. During follow-up, 86/193 (24 PAN, 32 MPA and 30 EGPA) patients required AT (mean follow-up since CS onset: 23.3⫾34.1 months) because CS failed (37%), relapse (52%) or CS dependency (10%): 49 received IV cyclophosphamide (CYC), 13 oral CYC, 56 azathioprine, 15 methotrexate, 9 mycophenolate mofetil, 7 IVIg, 6 plasma exchange, 1 infliximab and 1 cyclosporine. The significant association of mononeuritis multiplex (MNM) with AT use (univariate analysis, Fig 1; P⫽0.008) was confirmed by multivariate analysis, with MM being the only factor independently associated with requiring AT (hazard ratio⫽1.81 [95% CI: 1.12–2.93]; P⫽0.02). AT prescription rates were comparable for the 3 entities. At last visit, 165/193 (85%) were alive, with 94 (57%) and 28 (17%), respectively, still taking CS and/or cytotoxic agent or biotherapy. Overall survival reached 90% at 7 years and was comparable for patients who had taken ⱖ1 AT vs those treated only with CS during follow-up (P⫽0.564). However, patients given ⱖ1 vs 0 AT had significantly higher Vasculitis Damage Indexes (VDI): 2.93⫾2.09 vs 1.96⫾1.40 (P⬍0.001), reflecting more frequent osteoporosis (33 vs 18%, P⫽0.013) or peripheral neuropathy (60 vs 38%, P⫽0.004).

Tuesday, October 29

Conclusion: Despite relapses in the RTX arm after the end of the study period, RTX remains associated with a lower risk of relapse than AZA. RTX tends to be associated with a better overall survival compared to AZA. Disclosure: B. Terrier, None; C. Pagnoux, None; A. Karras, None; C. Khouatra, None; O. Aumaıˆtre, None; P. Cohen, None; F. Maurier, None; O. Decaux, None; H. Desmurs-Clavel, None; P. Gobert, None; T. Quemeneur, None; C. BlanchardDelaunay, None; P. Godmer, None; X. Puećhal, None; L. Mouthon, None; L. Guillevin, None.

2784 Mononeuritis Multiplex Predicts The Need For Immunosuppressive Or Immunomodulatory Drugs For Eosinophilic Granulomatosis With Polyangiitis, Polyarteritis Nodosa and Microscopic Polyangiitis Patients Without Poor-Prognosis Factors. Maxime Samson1, Xavier Puećhal2, Herve´ Devilliers3, Camillo Ribi4, Pascal Cohen5, Boris Bienvenu6, Christian Pagnoux7, Luc Mouthon2, Loic Guillevin8 and French Vasculitis Study Group (FVSG)2. 1Department of Internal Medicine, Referral Center for Rare Autoimmune and Systemic Diseases, Hoˆpital Cochin, AP–HP, Universite´ Paris Descartes, Paris, France; Internal Medicine and Clinical Immunology, Hoˆpital du Bocage, Dijon, France, Paris, France, 2Department of Internal Medicine, Referral Center for Rare Autoimmune and Systemic Diseases, Hoˆpital Cochin, AP–HP, Universite´ Paris Descartes, Paris, France, Paris, France, 3Internal medicine, Hoˆpital Geńe´ral, Dijon, France, Dijon, France, 4 Immunology and Allergology, Internal Medicine, University Hospitals of Geneva, Geneva, Switzerland, Geneva, Switzerland, 5Cochin University Hospital, Paris, France, 6Division of Internal Medicine, Centre Hospitalier Re´gional Universitaire de Caen, Coˆte de Nacre, Caen, France, Caen, France, 7 Rheumatology, Mount Sinai Hospital, Toronto, Canada, Toronto, ON, 8 Referral Center for Rare Autoimmune and Systemic Diseases, Hoˆpital Cochin, AP–HP, Universite´ Paris Descartes, Paris, France. Background/Purpose: Patients with eosinophilic granulomatosis with polyangiitis (EGPA), polyarteritis nodosa (PAN) or microscopic polyangiitis (MPA) without poor-prognosis factors, as defined by their 1996 Five-Factor Scores (FFS)⫽0, were included in 2 prospective CHUSPAN trials and

Figure 1. Probability of survival without prescription of new add-on therapies of 193 patients with EGPA, PAN or MPA without baseline poor-prognosis factors (FFS⫽0). P determined with log-rank test.

Conclusion: Despite the good and comparable overall survival of baseline-FFS⫽0 EGPA, PAN or MPA patients, 45% required AT, mostly those with MNM, and their VDI were significantly higher, indicating more sequelae than those of the other FFS⫽0 patients. Hence, this MNM subpopulation might be more likely to fail on CS alone, thereby supporting prospective evaluation of their initial cytotoxic agent use. Disclosure: M. Samson, None; X. Puećhal, None; H. Devilliers, None; C. Ribi, None; P. Cohen, None; B. Bienvenu, None; C. Pagnoux, None; L. Mouthon, None; L. Guillevin, None; F. V. Study Group (FVSG), None.

2785 Urinary Inflammatory Cells Strongly Reflect the Disease Activity and Renal Function in Antineutrophil Cytoplasmic Autoantibody-Associated Vasculitis. Yoko Wada1, Minoru Sakatsume1, Masaaki Nakano2 and Ichiei Narita1. 1Niigata University Graduate School of Medical and Dental Sciences, Niigata, Japan, 2School of Health Sciences, Faculty of Medicine, Niigata University, Niigata, Japan. Background/Purpose: The antineutrophil cytoplasmic autoantibody (ANCA)- associated vasculitides (AAVs) include microscopic polyangiitis (MPA), granulomatosis with polyangitis (GPA; formerly Wegener’s), and eosinophilic granulomatosis with polyangitis (EGPA). These small-vessel vasculitides are characterized by necrotizing inflammation of the vessel wall, particularly affecting small arteries, arterioles, and capillaries in systemic organs, and the kidney is one of the most frequently involved organs. We have already reported that T cells and macrophages appear in the urine of patients with glomerulonephritis, accompanied by active cellular infiltration

S1191

Tuesday, October 29

such as cellular crescent formation and diffuse interstitial cell infiltration, but not in the urine of patients with glomerulonephritis without the active inflammatory lesions. In this study, we examined the correlation between the numbers of urinary inflammatory cells and disease activity in AAVs, and assessed the utility of urinary immune cell analysis. Methods: Thirty patients with AAVs (MPA; n⫽26, GPA; n⫽3, EGPA; n⫽1), who had been referred to Niigata University Hospital between 2004 and 2012, were recruited for this study. The patients were divided into two groups according to clinical renal function (renal involvement (RI) group, n⫽24); non-renal involvement (non-RI) group, n⫽6)). Flow-cytometric analysis of urinary inflammatory cells was performed for each subject. Numbers of urinary T cells or macrophages were determined by multiplying the number of viable cells in the gated mononuclear cell region in each sample by the percentage of urinary CD3-positive or CD14-positive cells in the population, respectively. The numbers of urinary CD3-positive cells and CD14-positive cells, laboratory data including the titers of serum ANCA, serum creatinine and C-reactive protein, and the Birmingham vasculitis activity score (BVAS), were examined in each subject, and compared between the RI and non-RI groups. We then examined the correlations between the numbers of CD3-positive or CD-14-positive cells and laboratory data and BVAS by Spearman’s rank correlation coefficient in all subjects. Results: The total numbers of urinary CD3-positive cells and CD14positive cells were significantly elevated (⬎120/ml urine) in all patients in the RI group and 5 of 6 patients in the non-RI group. The number of urinary CD3-positive cells was positively correlated with the titer of serum ANCA (r⫽0.41, p⫽0.02) and BVAS (r⫽0.38, p⫽0.038), while the number of urinary CD14-positive cells was positively correlated with serum Cr (r⫽0.38, p⫽0.038) and negatively correlated with eGFR (r⫽⫺0.38, p⫽0.03). Conclusion: These results indicate the usefulness of urinary immune cell analysis for assessment of both kidney function and disease activity in patients with AAVs. Disclosure: Y. Wada, None; M. Sakatsume, None; M. Nakano, None; I. Narita, None.

2786 Review Of The Expert Panel Methodology In The Diagnostic and Classification Criteria For Vasculitis Study: A Pilot Study. Cristina Ponte1, Anthea Craven2, Joanna Robson2, Peter C. Grayson3, Ravi Suppiah4, Richard A. Watts5, Peter A. Merkel6 and Raashid A. Luqmani2. 1Rheumatology and Metabolic Bone Diseases Department, Hospital de Santa Maria, CHLN and Rheumatology Research Unit, Instituto de Medicina Molecular, Faculdade de Medicina da Universidade de Lisboa, Lisbon, Portugal, 2Nuffield Department of Orthopaedics Rheumatology and Musculoskeletal Sciences, Botnar Research Centre, University of Oxford, Oxford, United Kingdom, 3Section of Rheumatology & the Clinical Epidemiology Unit, Boston University School of Medicine, Vasculitis Center, Boston, MA, 4 Auckland District Health Board, Auckland, New Zealand, 5Rheumatology Department Ipswich Hospital and University of East Anglia, Ipswich, United Kingdom, 6University of Pennsylvania, Philadelphia, PA. Background/Purpose: The Diagnostic and Classification Criteria for Vasculitis (DCVAS) Study is a multinational observational study to develop diagnostic criteria and to update classification criteria for the primary systemic vasculitides. By 2015 the database will include clinical, laboratory and radiology data from over 2000 patients with vasculitis and 1500 comparator patients who present with features similar to vasculitis. To avoid the inherent circularity of using the submitting physician diagnosis as the gold standard, a reference diagnosis for each patient will be established using a combination of expert panel opinion and data-driven methods (e.g. machine learning algorithms). The aim of this analysis was to evaluate the methodology by which the expert panel will assess individual patient data to establish the reference diagnosis. Methods: By November 2012, 1662 patients had been recruited; 391 had complete 6 month follow-up data. Forty cases were randomly extracted and developed into clinical vignettes (CVs). The CVs were assessed for diagnoses by 6 independent experts using an online platform. Ten patients were assessed by all experts and the other 30 were each assessed by 2 of the 6 experts, randomly chosen. The experts first chose between primary vasculitis, secondary vasculitis, or other illness; then the respective major class (small-, medium-, or large-vessel vasculitis, or no predominant size vasculitis); and then the subtype or the specific disease for each category. For each answer a level of certainty (unlikely, possible, probable, definitive, or unknown) was

provided. The diagnoses of the expert panel and the submitting physician were compared. Results: The 40 clinical vignettes represented 26 women and 14 men, with a mean age of 62.5 ⫾ 20.3 years (range 20–86 years). Data from all 120 CV reviews were available for analysis. Treating clinicians submitted a diagnosis of primary vasculitis in 32 patients (17 small-vessel, 1 mediumvessel, 13 large-vessel and 1 with no predominant size); secondary vasculitis in 2, and other illness in 6. The expert panel agreed with the submitted diagnosis of primary vasculitis in 97% of the cases (definite 54%, probable 35% and possible 8%). However, only 78% of the submitted patients with primary vasculitis were classified as having the same sub-type of vasculitis when compared with the expert panel diagnosis (9% could not be sub-typed within the correct major class of vasculitis, 9% classified the cases with another sub-type of the same major class, 3% chose another major class or diagnosis and 2% selected the unknown option). There was an intraclass correlation coefficient of 0.82 (confidence interval 0.57–0.95) in the 10 CVs assessed by the 6 experts, indicating low variability between evaluators. Conclusion: An expert panel agreed with the individual submitting physician regarding a diagnosis of some form of primary vasculitis in nearly all cases, but disagreement about the exact form of vasculitis occurred in 22% of cases. Physician-based opinion may be more reliable for defining general categories of vasculitis than for defining specific subtypes. This exercise highlights the potential for diagnostic bias when using physician-opinion to define the gold standard diagnosis. Disclosure: C. Ponte, None; A. Craven, None; J. Robson, None; P. C. Grayson, None; R. Suppiah, None; R. A. Watts, None; P. A. Merkel, None; R. A. Luqmani, None.

2787 The Effect Of Smoking On The Clinical Expression Of ANCAAssociated Vasculitis. Neil Basu1, Aladdin Mohammad2, Richard A. Watts3, Paul Gatenby4, Luis F. Flores-Suarez5 and Alfred Mahr6. 1University of Aberdeen, Aberdeen, United Kingdom, 2Skåne University Hospital, Lund, Sweden, 3Rheumatology Department Ipswich Hospital and University of East Anglia, Ipswich, United Kingdom, 4ANU Medical School, The Canberra Hospital, Canberra, Australia, 5Instituto Nacional de Enfermedades Respiratorias, Mexico City, Mexico, 6Department of Internal Medicine, Hospital Saint-Louis, Paris, France. Background/Purpose: It is well recognized that smoking has immunomodulatory effects in several chronic inflammatory disorders. For ANCAassociated vasculitis (AAV), which includes granulomatosis with polyangiitis (GPA) and microscopic polyangiitis (MPA), the influence of smoking on disease risk and expression is not well understood. A small retrospective study suggested a protective effect of smoking on ENT manifestations. We investigated the relationship between smoking and AAV characteristics within two large, prospectively collected cohorts. Methods: The analyses used data from newly-diagnosed patients with AAV who were enrolled in 4 international randomized controlled trials of the Europe Vasculitis Society (EUVAS). Disease characteristics at diagnosis (age, sex, diagnosis, ANCA subtype, system involvement, BVAS, creatinine, CRP) were compared according to smoking status (current versus noncurrent; ever versus never). The analyses were repeated for the cumulative disease characteristics of consecutive patients with AAV from the Vasculitis Quality of Life (VASQoL) study, a multi-center cross-sectional study involving rheumatology and renal clinics across the United Kingdom. Univariate analyses used Chi2 and Mann-Whitney tests for categorical and continuous variables, respectively; logistic regression was used for ageadjusted analyses. Results: Of the 535 EUVAS patients, smoking data was available for 342 (GPA/MPA ratio: 186/156; males: 57%; median age: 60 years) among which 36 (11%) were current smokers and 158 (46%) ever smokers. Of the 360 VASQoL patients (GPA/MPA ratio: 265/95; males: 49%; median age: 64 years), all with complete smoking data, 26 (7%) were current smokers and 191 (53%) were ever smokers. Compared with non-current smokers, current smokers were significantly more likely to have vasculitis-related gastrointestinal (GI) manifestations in the EUVAS and in the VASQoL cohorts (17% vs. 6% [p⫽0.026] and 12% vs. 4% [p⫽0.044], respectively). In addition, VASQoL current smokers were significantly more likely to experience cutaneous (48% vs. 28%, p⫽0.035) and less likely to experience ENT (44% vs. 64%, p⫽0.046) manifestations. No other differences in AAV characteristics, including anti PR3/MPO status, were observed in the EUVAS and VASQoL cohorts although current smokers were younger than noncurrent smokers (p⫽0.047 and p⫽0.021, respectively). Age-adjustment did

S1192

not alter the interpretation of the identified associations of current smoking with disease characteristics. Both EUVAS and VASQoL ever-smokers were more likely to be males (both p⬍0.0001) and older (both p⫽0.02) but ever-smoking was not associated with clinical AAV characteristics in either cohort. Conclusion: These analyses of 2 independent AAV cohorts indicate that current smoking is positively associated with GI involvement, a finding which is in keeping with the established association between smoking and risk for Crohn’s disease. The VASQoL data also replicated a previous report supporting that smokers less often present with ENT manifestations. The lack of phenotype-modifying effects of ever-smoking could suggest an acute immune-modulatory effect of smoking in AAV. Disclosure: N. Basu, None; A. Mohammad, None; R. A. Watts, None; P. Gatenby, None; L. F. Flores-Suarez, None; A. Mahr, None.

ACR/ARHP Combined Session ACR/ARHP Combined Pediatrics Abstract Session Tuesday, October 29, 2013, 4:30 PM–6:00 PM

2788 Reliability and Responsivenss Of The Standardized Universal Pain Evaluations For Rheumatology Providers For Children and Youth (SUPER-KIDZ). Nadia Luca1, Jennifer N. Stinson1, Susanne M. Benseler1, Brian M. Feldman1, Dorcas Beaton2 and Ahmed Bayoumi3. 1Hospital for Sick Children, Toronto, ON, 2Scientist, Institute for Work & Health, Toronto, ON, 3Keenan Research Centre of the Li Ka Shing Knowledge Institute of St. Michael’s Hospital, Toronto, ON.

Disclosure: N. Luca, None; J. N. Stinson, None; S. M. Benseler, None; B. M. Feldman, None; D. Beaton, None; A. Bayoumi, None.

Enhancing Uveitis Screening Compliance In Juvenile Idiopathic Arthritis Patients. Melissa S Oliver1, Jennifer E. Weiss2, Suzanne C. Li2, Kathleen A. Haines2, Ginger L. Janow2, Esi Morgan DeWitt3 and Yukiko Kimura2. 1 UMDNJ New Jersey Medical School, Newark, NJ, 2Joseph M Sanzari Children’s Hospital, Hackensack University Medical Center, Hackensack, NJ, 3Cincinnati Children’s Hospital Medical Center, Cincinnati, OH. Background/Purpose: Chronic uveitis is one of the more severe morbidities associated with juvenile idiopathic arthritis (JIA). Risk of developing uveitis, and the frequency of uveitis screening visits, differs between the different JIA subtypes. Through participation in the Pediatric Rheumatology Care & Outcomes Improvement Network (PR-COIN), a quality improvement collaborative, we developed a uveitis screening tool for patients to bring to their ophthalmologist to improve communication between providers. This tool includes education for the family on uveitis. Our objective was to determine if use of this tool improved adherence to uveitis screening recommendations and documentation of uveitis screening. Methods: A retrospective chart review was conducted of 243 JIA patients without uveitis identified from our Childhood Arthritis & Rheumatology Research Alliance Registry and PR-COIN databases in the 13 months before and after beginning our participation in PR-COIN. Collected data included demographics, documentation of the uveitis assessment, and documentation of patient education. Effectiveness of implementation of the screening tool was assessed by comparing compliance with uveitis screening in the two time periods. Continuous variables were summarized as median (interquartile range). Categorical variables were summarized as frequency (percent). Paired binary outcomes of screening compliance were examined using McNemar’s test. Results: Data from 184 patients were included (table 1). Patients were excluded if they did not follow up during the study period or if charts were unavailable for review. Prior to our participation in PR-COIN, 132 (71.7%) patients reported compliance with screening but only 50 (27%) patients had documentation of their uveitis screening visits. Following introduction of our uveitis screening tool, documentation of screening increased significantly to 45.7% (table2). Of the 184 patients, 52 (28%) were non-compliant by self-report prior to participation in PR-COIN. Of those 52 patients, 32 (61.5%) increased their compliance following implementation of our uveitis screening tool. Table 1. Demographics and characteristics of JIA patients (n ⫽ 184) Age at diagnosis (years) Median (Interquartile range) Range: minimum - maximum Female n (%) Duration of JIA disease (years) Median (Interquartile range) Range: minimum - maximum JIA sub-type n (%) Polyarticular rheumatoid factor (⫺) Oligoarticular, persistent Enthesitis related arthritis Systemic arthritis Psoriatic arthritis Oligoarticular, extended Polyarticular rheumatoid factor(⫹) ANA Positive n (%) Yes

9.3 (4.5–13.2) 0.3–17.9 127 (69.0) 4.4 (3.0–7.5) 0.7–18.0 65 (35.3) 39 (21.2) 20 (10.9) 19 (10.3) 17 (9.2) 13 (7.1) 11 (6.0) 50 (27.2)

Table 2. Summary of patient and ophthalmologist compliance before and after implementation of our uveitis screening tool (n ⫽ 184)

Patient self-report of compliance with uveitis screening schedule n (%) Ophthalmologist documentation of eye exam on chart

13 months prior to joining PRCOIN n(%)

13 months after joining PRCOIN n(%)

P-Valuea

132 (71.7)

143 (77.7)

0.1308

50 (27.2)

84 (45.7)

0.0001

a, P-value is based on the McNemar’s test performed on repeated binary outcomes

Conclusion: Since joining PR-COIN and implementing a uveitis education and communication tool, there was improvement in patient compliance with uveitis screening. There was significant improvement in documentation

S1193

Tuesday, October 29

Background/Purpose: Pain is the most common symptom in children and youth with juvenile idiopathic arthritis (JIA), however, currently there is no comprehensive validated pain measure for this population. The Standardized Universal Pain Evaluations for Rheumatology Providers for Children and Youth (SUPER-KIDZ) is a new multi-dimensional online pain tool, developed to fill this gap in clinical care. The objective of this study was to determine the test-retest reliability and responsiveness of the computerized 20-item version of the SUPER-KIDZ pain tool in children with JIA. Methods: A single center prospective cohort study of JIA patients aged 8–18 years was performed. The SUPER-KIDZ questionnaire was administered to children expected to have stable pain for test-retest reliability analysis of each item using intra-class correlation coefficients (ICC) and weighted Cohen’s kappa. Responsiveness of each SUPER-KIDZ item to change in pain was evaluated in patients undergoing intra-articular steroid injection(s) who are expected to have improvement in pain. Measures of responsiveness included standardized response mean (SRM), Wilcoxon signed rank test, linear mixed model regression, and receiver operating characteristic (ROC) curve analysis. Internal consistency of the three SUPER-KIDZ subscales (sensory, interference, emotional) was measured using ordinal reliability alpha and item-total correlation. Results: Fifty-one children were included, of which 40 (78%) were female, and had a median of 3 active joints (1–5) and median physician global assessment of 2.5 cm (1.5–4) on 10 cm visual analog scale. Internal consistency was acceptable (ordinal ␣⫽0.73–0.92) for the sensory, interference and emotional SUPER-KIDZ subscales. Good test-retest reliability (ICC or weighted kappa ⱖ0.80) was found for 15 SUPER-KIDZ items in at least one analysis. Reliability was strongest for the items on pain intensity, pain frequency, pain duration and physical function, and weakest for questions related to sleep, having fun, catastrophizing, and feeling angry. At 2 weeks post-injection, 16 items were responsive to change in pain (SRM⫽0.66–0.82, significant Wilcoxon signed rank and/or linear mixed model regression). ROC curve analysis of 9 items gave an area under the curve of ⱖ0.70, adequately distinguishing between improved and unimproved subjects. The questions less responsive to change in pain were those related to fatigue frequency and emotional function (feeling angry, cheerful, worried). Conclusion: The majority of items of the new online SUPER-KIDZ tool have excellent test-retest reliability and responsiveness properties. The questions regarding fatigue and emotional function are less responsive to change after a joint injection procedure and could be tested after a cognitive intervention. If validity is demonstrated, this measure could be implemented as a standardized comprehensive pain tool for JIA patients, thereby fulfilling a longstanding gap in the care of patients with JIA.

2789

by ophthalmology of the patient’s visit. Simple tools such as a uveitis screening form can facilitate communication with the ophthalmologist and better educate JIA patients on the importance of uveitis screening, thus improving their care. Disclosure: M. S. Oliver, None; J. E. Weiss, None; S. C. Li, None; K. A. Haines, None; G. L. Janow, None; E. Morgan DeWitt, None; Y. Kimura, None.

2790

Tuesday, October 29

The Research In Arthritis In Canadian Children Emphasizing Outcomes (ReACCh Out) Cohort: Are We Achieving Clinically Important Outcomes? Deborah M. Levy1, Shirley ML Tse1, Elizabeth Stringer2, Jaime Guzman3, Roberta A. Berard4, Karen Watanabe Duffy5, Dax Rumsey1, Mercedes O. Chan3, Rosie Scuccimarri6, Adam M. Huber2, Lori B. Tucker7, Rae SM Yeung1, Ciaran M. Duffy5, Kiem Oen8 and The ReACCh Out Investigators1. 1The Hospital for Sick Children and University of Toronto, Toronto, ON, 2IWK Health Centre and Dalhousie University, Halifax, NS, 3 BC Children’s Hospital and University of British Columbia, Vancouver, BC, 4 Children’s Hospital of Western Ontario, London, ON, 5Children’s Hospital of Eastern Ontario and University of Ottawa, Ottawa, ON, 6Montreal Children’s Hospital and McGill University, Montreál, QC, 7University of British Columbia, Vancouver, BC, 8Children’s Hospital of Winnipeg and University of Manitoba, Winnipeg, MB. Background/Purpose: Recent data suggests that achievement of inactive disease and early disease remission may result in improved outcomes for patients with juvenile idiopathic arthritis (JIA). Objectives of our study were to examine the probabilities of achieving i) active joint count (AJC)⫽ 0; ii) inactive disease; and iii) disease remission for patients treated with contemporary treatments in a large prospective longitudinal inception cohort of patients with JIA. Methods: ReACCh Out recruited consecutive patients diagnosed with JIA at 16 sites across Canada (Jan 2005 – Dec 2010), with prospective data collection every 6 months for the first 2 years, then yearly. Clinical information included the six ACR core outcome measures and medications. Inactive disease was defined as active joint count (AJC) ⫽0, absence of systemic symptoms, enthesitis or uveitis and a physician global activity (PGA) of ⬍ 1 cm on a 10 cm VAS. Remission was defined as ⱖ 12 months with inactive disease with no anti-rheumatic or anti-uveitis medications. Descriptive statistics (median and interquartile range (IQR)) and KaplanMeier Survival analyses were examined; patients were censored at their last study visit or study end date. Results: 1104 patients with newly diagnosed (ⱕ 6 months) active JIA with ⱖ1 follow-up visit were analyzed. Patients were predominantly female (63%), age at diagnosis was 9.3 (3. 9, 13.0) years. Time from diagnosis to enrollment was 0.3 (0, 1.6) months. Follow-up to last visit or study end was 34.2 (21.5, 48) months. Patients were classified into ILAR subtypes at the 6 month visit: oligoarticular (416, 38%), polyarticular RF negative (235, 21%), polyarticular RF positive (46, 4%), psoriatic (64, 6%), enthesitis related arthritis (ERA) (157, 14%), systemic (sJIA)(76, 7%) and undifferentiated (110, 10%). Treatment received included intraarticular steroid injections in 46%, DMARDs in 55%, and biologics in 12%. Almost all patients (92%) achieved AJC⫽0 during the study period at a median of 7.0 mos (IQR 3.5, 13.3) from diagnosis. Patients with RF positive polyarthritis were last to reach AJC ⫽0. Inactive disease was achieved by 81% of subjects at a median of 13.0 mos (IQR 7.7, 21.7) from diagnosis. Time to first episode of inactive disease was shortest in the oligoarthritis group, and longest in the RF positive polyarthritis group. Table 1 shows the time to achieve AJC ⫽0 and inactive disease by JIA subtype. By survival analysis, the probability of disease remission by 4 years of disease was oligoarthritis (41%), polyarticular RF negative (8%), polyarticular RF positive (0%), psoriatic (47%), ERA (28%), sJIA (29%) and undifferentiated (30%). Table 1. Disease duration (in months) to achieve outcomes N Oligoarthritis Polyarticular RF negative Polyarticular RF positive Psoriatic Enthesitis Related Arthritis Systemic Undifferentiated

412 235 46 62 157 75 110

AJC ⴝ 0 Median (IQR) 5.9 (3.3, 10.0) 10.2 (5.0, 16.5) 10.0 (6.7, 20.2) 7.2 (4.7, 14.9) 6.8 (3.0, 14.3) 3.6 (2.0, 11.3) 6.4 (3.2, 15.9)

N 410 232 46 63 153 74 108

Inactive Disease Median (IQR) 9.6 (6.8, 14.9) 14.9 (10.8, 27.3) 24.2 (13.5, 39.5) 12.6 (7.7, 18.1) 16.0 (10.7, 24.5) 13.1 (6.7, 36.4) 14.4 (9.4, 22.7)

Conclusion: Almost all patients achieve an AJC ⫽0, and most attain inactive disease; however, the probability of remission remains low for the polyarticular subtypes. Disclosure: D. M. Levy, None; S. M. Tse, None; E. Stringer, None; J. Guzman, None; R. A. Berard, None; K. Watanabe Duffy, None; D. Rumsey, None; M. O. Chan, None; R. Scuccimarri, None; A. M. Huber, None; L. B. Tucker, None; R. S. Yeung, None; C. M. Duffy, None; K. Oen, None; T. ReACCh Out Investigators, None.

2791 Telephone Consultation Usage In a Pediatric Rheumatology Clinic: Considerations In Optimizing Nursing Resources. Julie Lemieux1, Audrey Tran2, Vincent Brienza3 and Roman Jurencak3. 1Children’s Hospital of Eastern Ontario, Ottawa, ON, 2University of Western Ontario, London, ON, 3 University of Ottawa, Ottawa, ON. Background/Purpose: Telephone consultation is essential to the delivery of patient care in the ambulatory care clinics at the Children’s Hospital of Eastern Ontario (CHEO). Our objective was to quantify and analyze all telephone calls received by the pediatric rheumatology nurses at CHEO. Methods: As per CHEO policy and procedure, all telephone calls nurses receive are documented on standard forms. Calls documented by the nursing staff in the Division of Pediatric Rheumatology over a six-month period from Jan 1, 2012 until June 30, 2012 were retrospectively analyzed for selected characteristics. An incoming call addressing a new concern relating to a specific patient was considered to be the index call. Index calls and all subsequent documented communications generated by the index call were then examined. Calls that only confirmed appointments and calls confirming that a message was received were excluded from analysis. Results: 321 index calls were received during the study period (0.6 calls per clinic hour), generating a total of 780 follow-up calls. The most frequent patient diagnosis was juvenile idiopathic arthritis (57%) and most patients were 12–18 years old (52%). 79% of calls were placed by the patient’s mother, while only 1% of callers were the patient themselves. 44% of calls lasted 1–5 minutes, 29% lasted 5–10 minutes and 27% lasted more than 10 minutes. The most common reasons for call were concerns relating to rheumatologic condition and medications taken (50%). Large number of calls was related to administrative issues such as requests for appointment change, prescription refills, etc (34%). Pain management was discussed in 29% of all calls. Nurses managed independently 40% of calls; when other health care providers were consulted, the physician was approached in 91% of cases. Only 8% of index calls resulted in an advanced clinic appointment. Conclusion: This study provides a descriptive analysis of calls nurses receive in a tertiary care pediatric rheumatology clinic. While nursing telephone consultation is beneficial in providing inter-disciplinary patient care and in minimizing clinic visits, the associated workload is significant and utilization of this service needs to be optimized, including re-distribution of administrative calls. Teenage patients rarely call themselves and their independence needs to be encouraged. Disclosure: J. Lemieux, None; A. Tran, None; V. Brienza, None; R. Jurencak, None.

2792 Pediatric Rheumatology Care and Outcomes Improvement Network Demonstrates Performance Improvement On Juvenile Idiopathic Arthritis Quality Measures. Julia G. Harris1, Esi Morgan DeWitt2, Ronald M. Laxer3, Stacy P. Ardoin4, Beth S. Gottlieb5, Judyann C. Olson1, Murray H. Passo6, Jennifer E. Weiss7, Daniel J. Lovell8, Tzielan C. Lee9, Sheetal S. Vora10, Nancy Griffin2, Jason A. Stock2, Lynn M. Darbie2 and Catherine A. Bingham11. 1Children’s Hospital of Wisconsin, Milwaukee, WI, 2Cincinnati Children’s Hospital Medical Center, Cincinnati, OH, 3University of Toronto, Toronto, ON, 4Ohio State University College of Medicine, Columbus, OH, 5 Cohen Children’s Medical Center of New York, New Hyde Park, NY, 6 Medical University of South Carolina, Charleston, SC, 7Hackensack University Medical Center, Hackensack, NJ, 8Cincinnati Children’s Hospital Medical Center, Cincinnati, OH, 9Stanford University School of Medicine, Stanford, CA, 10University of North Carolina Chapel Hill, Chapel Hill, NC, 11 Penn State Hershey Children’s Hospital, Hershey, PA. Background/Purpose: Pediatric Rheumatology Care and Outcomes Improvement Network (PR-COIN) is a multi-site learning network designed to improve outcomes of juvenile idiopathic arthritis (JIA) care. Teams collect point of care data on measures of process of care and outcomes of care for the purposes of analysis to guide improvement activities. Eleven North American

S1194

pediatric rheumatology centers participate. This report illustrates our improvement in several JIA process quality measures (QMs). Methods: Process of care QMs targeted for improvement include measurement of: arthritis-related pain, physician global assessment, joint count, health-related quality of life, physical function, as well as screening for uveitis, medication toxicity, and tuberculosis per guidelines. Outcome measures for JIA include clinical inactive disease, no or mild pain level, and optimal physical functioning. Network goals were determined for each process and outcome measure. Data are collected with IRB approval and informed consent, and the shared registry for data entry is the ACR’s Rheumatology Clinical Registry. Site-specific and aggregate data are analyzed and displayed monthly via statistical process control charts allowing PR-COIN to track performance over time. Individual centers use established quality improvement methodology to reach and exceed pre-determined goals. Results: Data from 3231 encounters for 905 JIA patients have been collected since April 2011. QMs with performance meeting or exceeding initial goals include: documentation of complete joint count every 180 days, measurement of arthritis-related pain at every visit, functional assessment every 180 days, and documentation of baseline toxicity labs. For PR-COIN network as a collective unit, QMs improved in five processes– measurement of functional ability, completion of ongoing medication toxicity labs, documentation of complete joint count, documentation of annual behavioral counseling, and measurement of health-related quality of life. All of these measures had a sustained shift above the baseline mean, demonstrating special cause. In addition, five sites have demonstrated individual improvement in at least one process QM. Conclusion: PR-COIN sites are collectively and individually demonstrating significant improvements in JIA process of care QMs. Quality improvement efforts in PR-COIN are ongoing with the goal of improving the outcome for patients with JIA.

2793 Towards Developing a Rheumatology-Specific Transition Of Care Program. Rina Mina1, Janalee Taylor2, Pamela A. Heydt3, Terry M. Moore4, Julie V. Ranz2, Mary Beth Burns2, Paula G. Melson5, Abigail Nye2, Jill Segerman2, Yolanda Farhey6, John Houk6, Avis Ware6, Jennifer L. Huggins2 and Lisa Vaughn2. 1Cincinnati Children’s Hospital Medical Center/ University of Cincinnati School of Medicine, Cincinnati, OH, 2Cincinnati Children’s Hospital Medical Center, Cincinnati, OH, 3Childrens Hospital Medical Center, Cincinnati, OH, 4Cincinnati Chiildren’s Hospital Medical Center, Cincinnati, OH, 5Children’s Hosp Medical Center, Cincinnati, OH, 6 University of Cincinnati, Cincinnati, OH.

Disclosure: R. Mina, None; J. Taylor, None; P. A. Heydt, None; T. M. Moore, None; J. V. Ranz, None; M. B. Burns, None; P. G. Melson, None; A. Nye, None; J. Segerman, None; Y. Farhey, None; J. Houk, None; A. Ware, None; J. L. Huggins, None; L. Vaughn, None.

S1195

Tuesday, October 29

Disclosure: J. G. Harris, None; E. Morgan DeWitt, None; R. M. Laxer, None; S. P. Ardoin, None; B. S. Gottlieb, None; J. C. Olson, None; M. H. Passo, Pfizer Inc, 5; J. E. Weiss, None; D. J. Lovell, None; T. C. Lee, None; S. S. Vora, None; N. Griffin, None; J. A. Stock, None; L. M. Darbie, None; C. A. Bingham, None.

Background/Purpose: Because of their considerable medical challenges, adolescents and young adults with pediatric-onset rheumatic diseases are often reliant on the health care system, and any interruption in their care could have serious consequences. Their successful transition of care from pediatric to adult health care systems is crucial for maintaining their health and health-related quality of life. As such, an organized transition of care program is highly important for these patients. Our objective was to determine components of a transition program from pediatric to adult rheumatology care using survey method. Methods: A checklist oriented towards developmentally appropriate goals for transition of care was formulated after review of existing literature and other transition checklists in use for other diseases. An online survey to gather approval for the checklist and resolve transition of care related questions was sent to members of the pediatric rheumatology multidisciplinary staff at Cincinnati Children’s Hospital Medical Center and to its adult rheumatology counterpart at the University of Cincinnati. The participants included attending pediatric and adult rheumatologists on staff, fellows-in-training, nursing staff, physical and occupational therapists, and social worker. All have direct patient contact. Results: Response rate was 84% (27 of 32). The respondents unanimously accepted the self-management transition checklist in terms of content and format. The checklist contained transition goals that reflected developmental skills of the patients. All respondents want the checklist to be incorporated into the patient’s electronic chart. From the pediatric rheumatology side, there was no agreement as to (1) the frequency of transition readiness assessment, and (2) at what age should the preparation phase of transition should commence. In addition, majority agreed on the (1) usefulness of scheduled independent visits for adolescent patients to (67%) and (2) necessity of actual doctor visit with an adult rheumatologist prior to the transfer-of-care. The adult rheumatology staff and fellows unanimously thought that patients should be able to communicate independently to care providers and be informed about their disease. Majority (60%) of the adult rheumatology providers felt that they need more information and education about caring for patients with pediatric onset rheumatic diseases. Conclusion: Agreement was reached on several relevant aspects of a transition of care checklist and program for patients with pediatric onset rheumatic disease. Validation of this checklist is ongoing as are interventional support to meet gaps in the transition process.

ACR Concurrent Abstract Session B cells in Systemic Lupus Erythematosus Wednesday, October 30, 2013, 9:00 AM–10:30 AM

2794

Wednesday, October 30

Characterisation Of Antigens Driving In Situ Autoantibody Production In Human Lupus Tubulointerstitial Nephritis (TIN). Andrew Kinloch1, Scott Henderson1, Natalya Kaverina1, Anthony Chang1, D. James Haddon2, Justin Jarrel2, Carole Henry Dunand1, Patrick C. Wilson1, Paul Utz2 and Marcus R. Clark1. 1University of Chicago, Chicago, IL, 2Stanford University School of Medicine, Stanford, CA. Background/Purpose: We have demonstrated that the degree of tubulointerstitial nephritis (TIN) on diagnostic biopsy predicts progression to renal failure. TIN is associated with tertiary lymphoid structures, in situ B cell oligoclonal expansion and ongoing antigen-driven somatic hypermutation. These phenomena support the hypothesis that local antigen is driving in situ lymphocyte selection and activation. Therefore, we cloned and characterized the antigenic specificities of in situ selected antibodies in human TIN. Methods: Human kidney biopsies with lupus TIN were stained for the proliferation marker ki-67 (6/8), or the post GC marker CD38 (1/8). Positive single cells were isolated by laser capture microscopy. An eighth sample was single cell sorted for CD138⫹ plasmablasts. mRNA was purified and cDNA reverse transcribed. Matched IgG heavy and light chain variable regions were PCR amplified, cloned into human IgG1 heavy or light chain expression vectors and expressed as mAbs in HEK-293 cells. Antigen screening methods included clinical ELISAs, crithidia IIF (for DNA), human protein arrays (Invitrogen), confocal microscopy of HEp-2 cells. Antigens targeted in HEp-2 cells were sought by immunoprecipitation followed by mass spectrometry. Candidate antigen identity was confirmed by multi-color confocal microscopy, and purified antigen ELISAs and protein arrays. In situ expression pattern of target antigen was confirmed by immunofluorescence using kidney from lupus TIN and normal controls. Sera from lupus patients, with (n⫽40) and without nephritis (n⫽20), were tested for IgG reactivity with candidate antigens. Results: 27 mAbs were generated. By HEp-2 analysis, 3/27 yielded nuclear speckled-patterns, 1/27 nucleolar-, 7/27 cytoskeletal-, 3/27 nuclear and cytoplasmic-, and 1/27 reacted with the golgi apparatus. None were DNA, Sm or RNP positive. Vimentin was confirmed as the dominant targeted antigen (40% of mAbs, from 7/8 patients), immunoprecipitated from HEp-2 cells, bound by TIN mAbs on arrays and ELISA, and yielding a co-staining pattern with an anti-vimentin antibody (V9, DAKO). Vimentin reactive TIN mAbs reacted strongly with both glomeruli and inflamed TI, but little with normal TI. Approximately 70% of serum samples reacted with vimentin by protein array and this tended to be more common in nephritis patients. Antigens targeted more strongly in nephritic serum included myosin and SSB (median % false discovery rate ⬍.001). Conclusion: Vimentin is an autoantigen upregulated during inflammation capable of driving in situ antibody production. These data support a model inwhich a positive feed-back loop of inflammation contributes to renal failure. Disclosure: A. Kinloch, None; S. Henderson, None; N. Kaverina, None; A. Chang, None; D. J. Haddon, None; J. Jarrel, None; C. Henry Dunand, None; P. C. Wilson, None; P. Utz, None; M. R. Clark, None.

2795 B Cell Derived Cytokines Induce Glomerular Injury in Mice. Alfred Kim1, Shreeram Akilesh2, Jeffrey Miner3 and Andrey Shaw3. 1Washington Univ School of Med, St. Louis, MO, 2University of Washington, Seattle, WA, 3Washington University School of Medicine, Saint Louis, MO. Background/Purpose: Renal involvment remains the leading cause of mortality for SLE patients, and is associated with proteinuria and foot process effacement. In subsets of LN patients, B cell depletion therapies have been efficacious in lowering disease activity including glomerulopathy. The contributions of B cells to proteinuria and foot process effacement remain unknown. The development of a murine model of B-cell induced proteinuria and identification of pathogenic factors would enhance our understanding of immune-based glomerular diseases. Methods: The B cell model antigen model hen egg lysozyme (HEL) was biotinylated, complexed to avidin and injected into mice. Naı¨ve HEL-specific B cells were then adoptively transferred and proteinuria assessed. Kidneys were processed for immunofluorescence and scanning electron microscopy

(SEM). Cultured podocyte membrane ruffling was assessed with DIC videomicroscopy. IL-4 expression in mice was achieved by hydrodynamically injecting murine IL-4 in the piggyBac vector system. Results: HEL embedded within the glomerular basement membrane (GBM) following IV injection. Proteinuria occurred after the transfer of naive HEL-specific B cells and associated with focal foot process effacement. No antibody or complement deposition was observed in proteinuric glomeruli. Intravital two-photon microscopy demonstrated that HEL-specific B cells arrested trafficking within glomeruli only in the presence of glomerularlocalized HEL. This demonstrated that B cells were capable of inducing glomerular injury and proteinuria. These data suggested cytokines secreted by activated B cells may be responsible for podocyte injury. Since foot process effacement is the histologic correlate of actin cytoskeletal rearrangement, we hypothesized that cytokines mediate podocyte injury through alterations in the actin cytoskeleton. We found that IL-4 induced unstable actin cytoskeletal changes leading to membrane ruffling. In addition, IL-4 generated foot process retractions on ex vivo fragments of renal cortex. Hydrodynamic DNA injection of wild-type mice with plasmid encoding IL-4 lead to proteinuria, which was reversed by JAK1/3 inhibition. Conclusion: We developed a novel model of B cell-induced proteinuria with focal foot process effacement. B cell derived cytokines such as IL-4 induced alterations in foot process morphology, leading to proteinuria. Transgenic IL-4 mice are associated with glomerulosclerosis independent of autoantibody production. We believe that IL-4 plays a direct role in podocyte injury through disruptions in the actin cytoskeleton. This has important implications in developing therapies to preserve podocyte function, limiting glomerular injury. Disclosure: A. Kim, ACR/RRF, 2; S. Akilesh, None; J. Miner, None; A. Shaw, None.

2796 Effects Of BAFF Inhibition On B Cell Selection In Murine SLE. Alexis Boneparth1, Ramalingam Bethunaickan2, Weiqing Huang2 and Anne Davidson2. 1Feinstein Institute for medical Research, Manhasset, NY, 2Feinstein Institute for Medical Research, Manhasset, NY. Background/Purpose: BAFF inhibition is a new B cell targeted therapy approved for the treatment of moderately active SLE. Although BAFF regulates selection of naı¨ve autoreactive B cells, belimumab only modest decreases autoantibody titers. To determine whether BAFF inhibition alters the quality of the autoantibody response by affecting selection of autoreactive B cells we generated NZW/BXSB.yaa (W/B) mice bearing the 3H9 site directed Ig heavy chain transgene. An extra copy of TLR7 conferred by the Yaa locus accelerates disease in the males. 3H9 associates with diverse light chains to generate anti-DNA and anti-cardiolipin specificity. We have previously shown that the autoreactive B cell response in 3H9.W/B mice uses predominantly Vk5-43 and Vk5-48 light chains that confer autoreactivity in their germline configuration and that such 3H9/Vk5-expressing B cells are rare in the naı¨ve repertoire but are expanded in germinal centers (GCs). Male W/B mice have defects in tolerance both at the GC entry checkpoint and in negative selection of B cells that acquire autoreactivity as a consequence of somatic mutation. We now wish to know whether BAFF inhibition influences either of these two checkpoints. Methods: To create a physiologic setting in which autoreactive B cells compete for survival with non-autoreactive B cells, we generated 50% 3H9/50% wt W/B bone marrow chimeras in which transferred 3H9. TLR7⫹/⫹ or TLR7⫹/⫺cells are GFP⫹ and can be easily identified. To determine when negative selection of 3H9⫹ cells occurs, we enumerated the % 3H9⫹/GFP⫹ cells in immature bone marrow, transitional, marginal zone (MZ), follicular (Fo), GC and plasma B cells using flow cytometry. To investigate the effect of TLR7 dose and BAFF inhibitors on selection of the autoreactive B cell repertoire, we performed single cell PCR and sequencing to determine the repertoire of light chains associating with the 3H9 heavy chain in Fo, GC, and plasma cells. Results: When competition from wt B cells is provided, significant deletion of 3H9 B cells occurs at the immature and transitional stage in the bone marrow regardless of TLR7 status or BAFF inhibition. Relative enrichment of 3H9 B cells occurs in the GC TLR7⫹/⫹ compared with TLR7⫹/⫺ chimeras. BAFF-R-Ig treatment significantly decreases B cell percentage and total numbers but does not decrease the percentage of immature or mature 3H9 B cells or mediate deletion of anergic cells. Data from single cell sorts of GFP⫹ B cells indicates comparable naı¨ve B cell repertoires but a decrease in selection of high affinity anti-chromatin Vk5-48/

S1196

Jk4 light chains among TLR7⫹/⫺ and BAFF-R-Ig treated mice compared with TLR7⫹/⫹ untreated controls. Conclusion: Preliminary data indicates that when competition is provided, negative selection of naı¨ve 3H9⫹ cells appears to be TLR7 and BAFF independent. In contrast, although autoreactive B cells still entered the germinal centers in mice treated with BAFF inhibition, modest changes were observed in germinal center selection. Future studies will explore the specific effects of BAFF inhibition on B cells that acquire autoreactivity in the GC as a consequence of somatic mutation. Disclosure: A. Boneparth, None; R. Bethunaickan, None; W. Huang, None; A. Davidson, GSK, 8, Eisai, 5.

2797 B-Lymphocyte Stimulator and A Proliferation Inducing Ligand In Lupus Nephritis: Low Serum Levels Of BLyS Predict Treatment Response. Ioannis Parodis1, Agneta Zickert1, Elisabet Svenungsson1, Vivianne Malmstro¨m1 and Iva Gunnarsson2. 1Rheumatology Unit, Department of Medicine, Karolinska Institutet, Stockholm, Sweden, 2Karolinska Institutet, Stockholm, Sweden.

Disclosure: I. Parodis, None; A. Zickert, None; E. Svenungsson, None; V. Malmstro¨m, None; I. Gunnarsson, None.

Bone Marrow In Systemic Lupus Erythematosus Patients Contain a Highly Elevated Proportion Of Somatically Mutated, Activated Naive Cells Comprised Of Substantial Clonal Expansions. Jennifer Hom1, Christopher Tipton1, Christopher Fucile2, Bridget Neary1, Chungwen Wei1, F. Eun-Hyung Lee1, Alex Rosenberg2 and Inaki Sanz1. 1Emory University, Atlanta, GA, 2University of Rochester, Rochester, NY. Background/Purpose: Systemic lupus erythematosus (SLE) is a disease where autoreactive antibodies are produced as a result of abnormal B cell activation and broken self-tolerance. This activation is perhaps most prominently observed in elevated levels of circulating activated cells within the naı¨ve compartment of SLE patients. The origin, diversity, and direct developmental consequences of this elevated activation are largely unclear though. In this study, we use a variety of multi-color flow cytometry panels to isolate and sort specific bone marrow (BM) and circulating B cell populations for immunoglobulin heavy chain (IGH) deep sequencing analysis. By developing a novel program to trace their inter-population relationships through mutational analysis, we aim to further characterize these cells and their contribution to the pathogenesis of SLE. Methods: Peripheral blood and BM aspirates from healthy donors and volunteers with SLE were collected and sorted using several multi-color flow cytometry panels. The IGH gene was amplified from these cells using multiple sets of VH family-specific and isotype-specific primers. Deep sequencing was conducted using Illumina MiSeq technology, and results were analyzed using an internally developed analysis program that performs quality filtering, detailed mutational analysis, and identification and matching of clonal lineages. Results: We found that a population of circulating mitotracker green (MTG) positive, CD23- activated naı¨ve cells were highly elevated in flaring SLE patients, slightly elevated in non-flaring patients, and largely absent in healthy controls. This population was also identified in BM of SLE patients, and it was found to be highly elevated in all SLE samples. Deep sequencing analysis showed that unlike most other populations in the circulation and BM, this population was highly oligoclonal, with some clones accounting for greater than 5% of the total population. Sequences from these activated naı¨ve cells were somatically mutated and typically found to be of lower VH mutation rates than memory and plasmablasts (PB), but higher than resting naı¨ve cells. We found clonal relations between several circulating and BM B cell subsets and tracked the lineage maturation using mutation analysis. Surprisingly, the activated naı¨ve fraction was highly related to PB and through mutational tree analysis, they were found to inhabit the more proximal portion of the same lineage branches as the more distal PB. Conclusion: Our data suggest that in SLE, there is an expansion of activated MTG⫹, CD23- naive cells which may be an early implication of abnormal B cell activation as a result of damaged self-tolerance. We show that these cells are highly related to circulating and BM antibody-producing PB, which are typically also expanded in SLE. Disclosure: J. Hom, None; C. Tipton, None; C. Fucile, None; B. Neary, None; C. Wei, Biogen Idec, 2; F. E. H. Lee, None; A. Rosenberg, None; I. Sanz, Pfizer Inc, 5, Biogen Idec, 9.

2799 Identification Of IL-10 Producing Plasma Cells In Human and Its Deficiency In Systemic Lupus Erythematosus Patients. Xiaoqian Wang1, James Roger2 and Ignacio Sanz3. 1Emory University, Allergy, Immunology and Rheumatology, Atlanta, GA, 2University of Rochester, Rochester, NY, 3 Emory University School of Medicine, Atlanta, GA. Background/Purpose: Regulatory B cells are active participants in down-regulating inflammation and autoimmunity, through the production of IL-10. Despite intensive studies in mice, there is only a paucity of data concerning the regulation of IL-10 production in human. The purpose of this study is to understand IL-10 production by human B cells and to explore the development of regulatory B cells either in healthy individuals or in patients with SLE. Methods: Healthy controls (HC) (n⫽15) and SLE patients (n⫽20) fulfilling the American College of Rheumatology revised classification criteria were included in this study. Total B cells were purified with MACS-negative selection and different B cell subsets were obtained by sorting. IL-10 producing B cells were identified with intracellular staining, Elispot, and dual color Flurospot assay. Elisa was used to determine the level of IL-10 in B cell cultures.

S1197


Background/Purpose: B-lymphocytes have a pivotal role in the pathogenesis of Systemic Lupus Erythematosus (SLE). B-Lymphocyte Stimulator (BLyS) has an important role in the activation, differentiation and maintenance of activated B-cells. A PRoliferation Inducing Ligand (APRIL) is involved in the induction and maintenance of B- and T-cell responses. The aim of our study was to assess serum levels of BLyS and APRIL in patients with lupus nephritis (LN) and investigate how these levels are affected by immunosuppression. Methods: Sixty-four patients with biopsy-proven LN (52 proliferative nephritis, PN, and 12 membranous nephritis, MN) and 64 gender- and age-matched controls were included in our study (mean age 34 years). The patients were treated with corticosteroids combined with cyclophosphamide (CYC, n⫽45), mycophenolate mofetil (n⫽11), rituximab (n⫽7) or azathioprine (n⫽1) and underwent a second renal biopsy after completed induction therapy. Serum was collected before and after induction treatment (mean time 8 months) for the patients and at recruitment for the controls. The mean proteinuria at baseline was 2.2 g/d. BLyS and APRIL levels were estimated by ELISA (R&D Systems and eBioscience, respectively). The renal biopsies were assessed according to the ISN/RPS classification system, as well as Activity Index (AI) and Chronicity Index (CI). We defined clinical response (CR) as ⱖ50% reduction in proteinuria, normal or improved renal function (ⱖ25% increase of GFR) and inactive urinary sediment. For complete response the proteinuria at follow-up should be ⬍0.2 g/d and for partial response between 0.2 and 2 g/d. We defined histopathological response (HR) as ⱖ50% improvement in AI for partial response and additionally demanded a lack of signs for active inflammation (ISN/RPS class I, II, III C or IV C) in the follow-up biopsy for complete response. Results: BLyS levels were significantly higher in patients with LN than in controls (p⬍0.001), but remained unchanged after induction treatment (p⫽0.99). APRIL levels were significantly higher in patients compared to controls at baseline (p⫽0.005), but not at follow-up (0.14). This is consistent with a significant reduction of APRIL levels at follow-up (p⬍0.001). This decrease was more prominent in patients who received CYC (p⫽0.006). APRIL levels decreased significantly in both responding and non-responding patients. In the patient group with PN, the decrease of APRIL in nonresponding patients was not statistically significant (p⫽0.13). Interestingly, the ROC-curve for BLyS by treatment response revealed a potential predictive power. Further analysis showed that low baseline levels of BLyS had high positive predictive value for both CR and HR. Conclusion: We observed an overall decrease in serum levels of APRIL after immunosuppression. No decrease was found in non-responding patients with PN, implying that APRIL could be of importance in this subgroup. BLyS levels remained unchanged after immunosuppressive treatment, suggesting that conventional treatments do not affect BLyS-producing cells. However, low baseline levels of BLyS were found to predict both CR and HR, suggesting that BLyS should be evaluated as a potential biomarker of response in LN.

2798

Results: In vitro and ex vivo analysis determined for the first time, that human plasma cells (PC: CD19⫹CD27⫹⫹CD38⫹⫹CD138⫹) and plasmablasts (PB: CD19⫹CD27⫹⫹CD38⫹⫹CD138⫺) are major sources of IL-10. This cytokine was secreted from PB and PC generated in vitro from cultured memory cells using Elispot assay and intracellular flow cytometry. Both assay identified 1–2% of IL-10 positive PB/PC. Similarly, ex vivo analysis of unstimulated B cell subsets sorted 7 days after flu vaccination of HC identified PB and PC as the only source of IL-10 and up to 1% of PC and PB are able to produce IL-10. In contrast, circulating PB and PC spontaneously expanded in the circulation of active SLE patients failed to produce IL-10 ex vivo. In vitro stimulation of purified B cell subsets with CpGDNA and IL-2 coupled with CFSE labeling identified the IgD(⫹)CD27(⫹) nonswitched memory B-cell subpopulation as the main IL-10-producing precursor B cell in a division-linked fashion. Of interest, IL-10 production from stimulated B cells was significantly lower in SLE patient (380.1⫾60.94 pg/ml) than in HC (857.6⫾97.06 pg/ml) (p⬍0.001), in direct correlation with the reduction in non-switched memory cells typical of this disease. In contrast, SLE patients with normal distribution of B cell subsets had normal IL-10 secretion. Conclusion: We provide the first description of spontaneous IL-10 production by human PB and PC in HC and its deficiency in SLE. These findings suggest that IL-10 could negatively regulate antibody secreting cells in an autocrine fashion that would be deficient in SLE thereby enhancing autoantibody production. Our data also identify non-switched memory B cells as an important source of potent regulatory B cells and indicate that the decrease of this population typically observed in SLE may be a major contributor to the deficiency of regulatory IL-10 in this autoimmune disease.

Table 1. Baseline demographics and characteristics of Japanese patients with RA.

Characteristic Female Age (years) Disease duration (years) Rheumatoid factor positive MTX dose (mg) DAS28-CRP

Denosumab 60 mg Q3M Q2M N ⴝ 82 N ⴝ 85

Placebo N ⴝ 88

Q6M N ⴝ 85

76 (86.4) 57.0 ⫾ 10.6 2.3 ⫾ 1.3 60 (68.2)

65 (76.5) 54.4 ⫾ 10.6 2.2 ⫾ 1.3 59 (69.4)

59 (72.0) 52.0 ⫾ 11.7 2.3 ⫾ 1.3 56 (68.3)

66 (77.6) 54.6 ⫾ 10.5 2.3 ⫾ 1.4 57 (67.1)

7.6 ⫾ 1.8 3.5 ⫾ 0.9

8.0 ⫾ 2.0 3.2 ⫾ 0.9

8.4 ⫾ 2.2 3.3 ⫾ 0.9

8.3 ⫾ 2.0 3.3 ⫾ 0.8

Data presented are Mean ⫾ SD or n (%).

Changes in the bone erosion score (deltaERO) at Month 12 were 0.99 ⫾ 2.69 in placebo, 0.27 ⫾ 0.98 in Q6M (Compared with placebo, p⫽0.0082), 0.14 ⫾ 0.53 in Q3M (p⫽0.0036), and 0.09 ⫾ 1.52 in Q2M (p⬍0.0001) (Figure 1). Percentages of patients without increase in ERO (ⱕ0) at Month 12 were 62.5% in placebo, 78.8% in Q6M (p⫽0.0173), 80.5% in Q3M (p⫽0.0099), and 83.5% in Q2M (p⫽0.0019).

Disclosure: X. Wang, None; J. Roger, None; I. Sanz, Pfizer Inc , 5, Biogen Idec, 9.

ACR Concurrent Abstract Session Imaging in Rheumatoid Arthritis Wednesday, October 30, 2013, 9:00 AM–10:30

AM

Figure 1. Mean (⫾ SD) Bone Erosion Score Change From Baseline (Full Analysis Set)


2800 Dose-Response Effects Of Denosumab, a Novel Subcutaneous RANKL Inhibitor, On The Progression Of Bone Erosion In Japanese Patients With Rheumatoid Arthritis Treated With Methotrexate: Results Of Phase II DRIVE Study—A Twelve Month Placebo Controlled, Randomized, Double Blind Study. Tsutomu Takeuchi1, Yoshiya Tanaka2, Naoki Ishiguro3, Hisashi Yamanaka4, Toshiyuki Yoneda5, Harry K. Genant6 and De´sireé van der Heijde7. 1Keio University School of Medicine, Tokyo, Japan, 2 University of Occupational and Environmental Health, Kitakyushu, Japan, 3 Nagoya University Graduate School of Medicine, Nagoya, Japan, 4Institute of Rheumatology Tokyo Women’s Medical University, Tokyo, Japan, 5 Indiana University School of Medicine, Indianapolis, IN, 6University of California, San Francisco, CCBR-Synarc, Newark, Tiburon, CA, 7Leiden University Medical Center, Leiden, Netherlands. Background/Purpose: Denosumab is a fully human monoclonal antibody (IgG2 subclass) that binds specifically to RANKL. It inhibits osteoclastinduced bone resorption by preventing the binding of RANKL to RANK, thereby inhibiting progression of marginal bone erosion at the joints where inflammatory cytokines are highly expressed, stopping structural joint damage. In this study, we evaluated the dose-frequency of denosumab for safety and efficacy in Japanese patients with RA. Methods: DRIVE Study (AMG162-D-J201) was a placebo-controlled, 4-arm randomized (1:1:1:1), double-blind, parallel-group study in patients receiving methotrexate treatment. Denosumab 60 mg (every 6 months (Q6M), 3 months (Q3M), or 2 months (Q2M)) or placebo was administered subcutaneously for 12 months. The primary endpoint was change in the bone erosion score assessed by the modified Sharp-van der Heijde method from baseline to 12 months. One-tailed Shirley-Williams’s test were performed and differences were considered significant when p⬍0.025. Results: Among 350 patients enrolled, 346 (placebo, n⫽88; Q6M, n⫽86; Q3M, n⫽85; Q2M, n⫽87) were treated with the study drugs (Table 1), and 35 (placebo n⫽6; Q6M, n⫽8; Q3M, n⫽13; Q2M, n⫽8) were discontinued.

No significant differences in the types and the overall incidence of adverse events (AEs) were observed among denosumab-treated groups and the placebo group. No hypocalcemia, osteonecrosis of the jaw, or atypical femoral fracture were observed. Conclusion: Greater suppression of the bone erosion from baseline to 12 months and the less erosive progression were observed after denosumab treatment in all doses compared with placebo, with lower progression rates in groups with shorter dosing interval. Denosumab was well tolerated in Japanese patients with RA and shortening the dosage interval did not increase AE incidence. Disclosure: T. Takeuchi, Abott Japan, Astellas Pharma, Bristol-Myers, Chugai Pharmaceutical, Daiichi Sankyo, Eisai, Mitsubishi Tanabe Pharma, Pfizer, Sanofiaventis, Santen Pharmaceutical., Takeda Pharmaceutical, Teijin Pharma, AbbVie, Asahi Kasei Pharma, Taisho Toyama Pharma, 2, Abott Japan, Bristol-Myers, Chugai Pharmaceutical, Eisai, Janssen Pharmaceutical, Mitsubishi Tanabe Pharma, Pfizer Japan, Takeda Pharmaceutical, Astellas Pharma, Daiichi Sankyo, Astra Zeneca, Eli-Lilly Japan, Novartis Pharma, Asahi Kasei Pharma, AbbVie, 5; Y. Tanaka, Bristol-Myers, Mitsubishi Tanabe Pharma, AbbVie, MSD, Chugai Pharmaceutical, Astellas Pharma, Daiichi Sankyo, 2, Mitsubishi Tanabe Pharma, Eisai, Chugai Pharmaceutical, Abott Japan, Astellas Pharma, Daiichi Sankyo, AbbVie, Janssen Pharmaceutical, Pfizer Japan, Takeda Pharmaceutical, Astra Zeneca, Eli Lilly Japan, GlaxoSmithKline, Quintiles, MSD, Asahi Kasei Pharma, 5; N. Ishiguro, Takeda Pharmaceutical, Mitsubishi Tanabe Pharma, Astellas Pharma, Chugai Pharmaceutical, Abott Japan, Bristol-Myers, Eisai, Janssen Pharmaceutical, Kaken Pharmaceutical, Pfizer Japan, Taisho Toyama Pharmaceutical, Otsuka Pharmaceutical, Daiichi Sankyo, 5; H. Yamanaka, Abott Japan, Bristol-Myers, Chugai Pharmaceutical, Eisai, Mitsubishi Tanabe Pharma, Takeda Pharmaceutical, Pfizer Japan, 5; T. Yoneda, Daiichi Sankyo, 5; H. K. Genant, Amgen, Bristol-Myers Squibb, Eli Lilly & Co., Genentech, GlaxoSmithKline, Merck & Co., Novartis Pharmaceuticals, Pfizer, Roche Pharmaceutical, ONO Pharma USA, Servier, and Synarc, 5; D. van der Heijde, AbbVie, Amgen, AstraZeneca, Bristol-Myers Squibb, Celgene, Centocor Biotech, Chugai Pharmaceutical, Eli Lilly & Co., GlaxoSmithKline, Imaging Rheumatology, Merck & Co., Novartis Pharmaceuticals, Pfizer, Roche Pharmaceutical, Sanofi-Aventis, Schering-Ploug, 5.

S1198

2801 Comparative Longitudinal Analysis Of Periarticular Bone Structure In Patients Treated With Methotrexate In Combination Of Either TNF Blockers Or Tocilizumab. Sebastian Kraus, Matthias Englbrecht, Juergen Rech, Roland Kocijan, Georg A. Schett and Stephanie Finzel. University of Erlangen-Nuremberg, Erlangen, Germany.

Background/Purpose: The aim was to investigate if a treat-to-target strategy with methotrexate and intra-articular glucocorticoid suppressed synovitis and osteitis, and halted structural damage progression in early rheumatoid arthritis (ERA), and if added adalimumab provided an additional effect, as judged by magnetic resonance imaging (MRI). Methods: In a double-blinded placebo-controlled investigator-initiated trial, 180 DMARD naı¨ve ERA patients were randomized 1:1 to methotrexate, intra-articular glucocorticoid injections and placebo/adalimumab (1). Eightyfive patients (placebo/adalimumab: 43/42) had contrast-enhanced MRI of the right wrist and the 2nd–5thmetacarpohalangeal joints at months 0, 6 and 12. Synovitis, osteitis, tenosynovitis, bone erosion and joint space narrowing (JSN) were scored with validated methods (2–4). Results: Synovitis, osteitis and tenosynovitis scores decreased highly significantly (p⬍0.0001) during the 12-months follow-up, with change scores of mean ⫺3.7 (median ⫺3.0[range ⫺13;11]), ⫺2.2 (⫺1 [⫺31;32]) and ⫺3.3 (⫺1[⫺31; 32]), respectively. No overall change in MRI erosion and JSN scores were observed, with change scores of 0.1 (0 [-16;13]) and 0.2 (0 [-3;4]), respectively. See table for status scores. Clinical disease activity scores and patient-related measures decreased highly significantly during follow-up (table). Among MRI, clinical and biochemical outcome measures, the tenosynovitis score at month 6 was the only measure that differed significantly between the treatment groups; placebo group: 3.9 (2 [0;18])/the adalimumab group: 1.3 (0 [0;11]), MannWhitney: p⬍0.035. Furthermore, the osteitis score decreased significantly from month 0 to month 6 and 12 in the adalimumab group, but not in the placebo group, Wilcoxon; pⱕ0.002 and pⱖ0.062, respectively. Table. MRI and clinical status values for patients in the placebo and the Adalimumab treatment groups at 0, 6 and 12 months, mean (median [range]) 0 months

6 months

12 months

Disclosure: S. Kraus, None; M. Englbrecht, None; J. Rech, None; R. Kocijan, None; G. A. Schett, None; S. Finzel, None.

Synovitis (0–21) Placebo group Adalimumab group BME (0–69) Placebo group Adalimumab group Tenosynovitis (0–30) Placebo group Adalimumab group Erosions (0–30) Placebo group Adalimumab group JSN (0–84) Placebo group Adalimumab group DAS28 Placebo group Adalimumab group

2802

Values are presented as mean (median [range]). The numbers in bold writing indicate differences between treatment groups. Between baseline and follow-up visit differences were tested using Mann-Whitney’s test: .NS: Not significant; *ⱕ0.005; **ⱕ0.001; ***⬍0.0001.

A Treat-To-Target Strategy With Methotrexate and Intra-Articular Triamcinolone With Or Without Added Adalimumab Reduces Synovitis, Osteitis and Tenosynovitis and Halts Structural Damage Progression In Early Rheumatoid Arthritis: The Opera Magnetic Resonance Imaging Sub-Study. Mette Bjørndal Axelsen1, Iris Eshed2, Kim HørslevPetersen3, Kristian Steengaard-Petersen4, Merete L. Hetland5, Jakob M. Møller6, Peter Junker7, Jan Pødenphant8, Torkell Ellingsen9, Palle Ahlquist10, Hanne M. Lindegaard11, Asta Linauskas12, Annette Schlemmer13, Mette Yde Dam14, Ib Hansen15, Hans Chr Horn16, Christian G. Ammitzbøll17, Anette Jørgensen17, Sophine B. Krintel18, Johnny Raun19, Julia S. Johansen20, Niels Steen Krogh21 and Mikkel Østergaard5. 1Copenhagen University, Copenhagen, Denmark, 2Sheba Medical Center, Tel Hashomer, Israel, 3South Jutland Hospital, Graasteen, Denmark, 4Aarhus University Hospital, Aarhus, Denmark, 5Copenhagen University Hospital Glostrup, Copenhagen, Denmark, 6 Copenhagen University Hospital at Herlev, Copenhagen, Denmark, 7Odense University Hospital, Odense C, Denmark, 8Copenhagen University at Gentofte, Hellerup, Denmark, 9Diagnostic Centre Region Hospital Silkeborg Denmark, 8600 Silkeborg, Denmark, 10Vejle Regional Hospital, Vejle,

8.2 (8 [0–19]) 7.6 (7 [0–21])

4.9 (4 [0–13])*** 5.0 (5 [0–13])**

4.7 (4 [0–15])*** 4.8 (4 [0–15])*

6.2 (1 [3–35]) 4.3 (1 [0–34])

3.9 (0 [0–35])NS 1.4 (0 [0–11])**

3.5 (0 [0–36])NS 2.1 (0 [0–31])*

7.3 (5 [0–26]) 5.2 (3 [0–23])

3.9 (2 [0–18])* 1.3 (0 [0–11])***

2.0 (0 [0–20])** 1.6 (0 [0–30])*

12.5 (6 [0–64]) 10.9 (8 [0–43])

11.6 (5 [0–49])NS 9.9 (8 [0–30])NS

13.7 (6 [0–61])NS 10.3 (8 [0–30])NS

(0 [0–8]) 0.6 (0 [0–7])

(0 [0–9])NS 0.6 (0 [0–7])NS

1.4 (0 [0–12])NS 0.9 (0 [0–7])NS

5.3 (5.2 [3.5–8.1]) 5.4 (5.4 [3.3–7.5])

2.7 (2.5 [1.7–5.0])*** 2.6 (2.4 [1.7–6.0])***

2.6 (2.3 [1.7–4.6])*** 2.4 (2.0 [1.7–4.7])***

Conclusion: A treat-to-target strategy with methotrexate and intraarticular glucocorticoid in ERA patients effectively decreased synovitis, osteitis and tenosynovitis and halted structural damage progression judged by MRI. The addition of Adalimumab provided further suppression of osteitis and tenosynovitis. References: 1. Hørslev-Petersen K et al. Ann Rheum Dis. Online First 7 mar 2013; 2. Østergaard et al. J Rheumatol. 2003;30:1385–6; 3. Haavardsholm et al. Ann Rheum Dis. 2007;66:1216–20; 4. Østergaard et al. J Rheumatol. 2011;38:2045–50 Disclosure: M. B. Axelsen, Abbott Laboratories, 2; I. Eshed, None; K. Hørslev-Petersen, Abbott Laboratories, 2, UCB Nordic, 5; K. Steengaard-Petersen, Abbott Laboratories, 5, Wyeth Pharmaceuticals, 5, Pfizer Inc, 5, Danish Rheumatism Association, 2; M. L. Hetland, None; J. M. Møller, None; P. Junker, The Danish Rheumatism Association, 2; J. Pødenphant, None; T. Ellingsen, None; P. Ahlquist, None; H. M. Lindegaard, Lilly, MSD, Nordpharma, Roche Pharmaceuticals, 5; A. Linauskas, None; A. Schlemmer,: MerckSharpDohme, 5, Roche Pharmaceuticals, 5, Wyeth/Pizer, 5; M. Yde Dam, None; I. Hansen, None; H. C. Horn, Abbott Laboratories, 5; C. G. Ammitzbøll, None; A. Jørgensen, None; S. B. Krintel, None; J. Raun, None; J. S. Johansen, None; N. S. Krogh, None; M. Østergaard, Abbott, Pfizer, Centocor, 2, Abbott Pfizer, Merck, Roche, UCB, 5, Abbott, Pfizer, Merck, BMS, UCB, Mundipharma, 8.

S1199


Background/Purpose: Inhibition of tumor necrosis factor alpha (TNFi) and interleukin-6 receptor (IL6Ri, tocilzumab) are among the most potent therapeutic strategies in patients with rheumatoid arthritis (RA) and have shown to retard structural bone damage. Comparative longitudinal studies on the impact of TNFi or IL-6i on periarticular bone damage in RA, however, have not been performed. Methods: Observational study on 54 patients with RA receiving TNFi (N⫽25) or IL6Ri (N⫽29) in combinaton with methotrexate over one year. All patients received baseline and one-year follow-up examination of the 2nd– 4th metacarpophalangeal and wrist joints for presence of erosions and osteophytes using high-resolution peripheral quantitative computed tomography (HR-pQCT). Erosions and osteophytes were quantified both numerically and metrically at baseline and one year follow-up. Demographic and disease specific characteristics such as age, gender, disease duration and -activity were recorded. Results: Both cohorts were comparable for age, gender, disease duration and –activity, autoantibody status as well as baseline bone damage. At baseline 185 erosions (MCP:N⫽136; wrist:N⫽49) and 340 osteophytes (MCP:N⫽248; wrist:N⫽92) were detected in the 29 patients of the IL-6i group and 123 erosions (MCP:N⫽84; wrist;N⫽39) and 270 osteophytes (MCP: N⫽176; wrist: N⫽94) were detected in the 25 patients of the TNFi group. Erosion numbers significantly increased in the MCPs (⫹0.321mm3, t(83)⫽⫺2.92, p⫽0.004) and tended to increase in the wrist joints: (⫹0.13mm3, t(48)⫽⫺0.52, p⫽0.096) in the TNFi group, whereas they remained constant in IL-6i treated patients both in wrist and MCP joints. No significant volumetrical change of erosive changes of erosions was observed in either group. Osteophyte numbers (TNFi:⫹0.27mm, t(175)⫽⫺6.99, p⬍0.001; IL6i:⫹ 0.1mm, t(245)⫽⫺3.49, p⫽0.001) and size (TNFi: ⫹0.19mm, t(174)⫽⫺7.01, p⬍0.0001; IL-6i:⫹0.08mm, t(247)⫽⫺3, p⫽0.003) in the MCP joints significantly increased in both groups. In the wrists, osteophytes increased both in size (⫹0.9mm, t(96)⫽2.49, p⫽0.015) and numbers (⫹0.19mm, t(93)⫽⫺3.14, p⫽0.002) in the TNFi group while remaining constant in size and increasing in numbers (⫹0.185mm, t(91)⫽⫺3, p⫽0.004) in the IL-6i group. Conclusion: This study suggests that IL-6Ri allows an even better control of structural bone damage than TNFi in patients with erosive RA. Furthermore, the observation that HR-pQCT allows detecting structural progression in TNFi treated RA patients is important and indicated that high-level imaging technology allows very detailed analysis of changes of bone composition in RA patients, which could be valuable to improve the monitoring of strucrural effects of anti-rheumatic drugs.

Denmark, 11Odense University Hospital, Odense, Denmark, 12Vendsyssel Hospital, Hjørring, Denmark, 13Aalborg University Hospital, Aalborg, Denmark, 14Silkeborg Regional Hospital, Silkeborg, Denmark, 15Viborg Hospital, Viborg, Denmark, 16Vejle Hospital, Vejle, Denmark, 17Arhus University Hospital, Aarhus, Denmark, 18Copenhagen University Hospital at Glostrup, Glostrup, Denmark, 19University of Southern Denmark, Graasten, Denmark, 20 Herlev Hospital, Herlev, Denmark, 21ZiteLab ApS, Copenhagen, Denmark.

2803

2804

Can Ultrasonographic Findings Predict Response To Tumor Necrosis Factor-␣ Inhibitor Treatment In Rheumatoid Arthritis? Nevsun Inanc1, Gu¨lsen Ozen2 and Haner Direskeneli1. 1Marmara University, School of Medicine, Istanbul, Turkey, 2Marmara University School of Medicine, Istanbul, Turkey.

Radiological Outcome Of Joints With MRI Detected Subclinical Inflammation In Early Arthritis Patients. A. Krabben, W. Stomp, J. A. B. van Nies, T.W.J. Huizinga, D. van der Heijde, J.L Bloem, M. Reijnierse and A. H. M. van der Helm-van Mil. Leiden University Medical Center, Leiden, Netherlands.

Background/Purpose: Tumor necrosis factor-␣ inhibitors (TNFi) are highly effective in patients with rheumatoid arthritis (RA), while not effective in all, with predictors of response being necessary. Although genetic, inflammatory and serologic biomarkers are under major investigation for this purpose, little is known about the predictive value of ultrasonographic parameters in RA. We aimed to investigate the ability of ultrasonographic parameters to predict which patients with RA will benefit from the treatment with TNFi in terms of EULAR response Methods: Biologic naive RA patients starting treatment with TNFi were examined longitudinally by ultrasonography (US) (both Gray-Scale [GS] and Power Doppler [PD]) of 28 joints according to standard scans of EULAR guideline and clinically (tender/swollen joint counts, DAS28 and HAQ scores) at baseline and after 3 and 6 months. US examinations were performed by an experienced sonographer (NI) using a MyLab 70 US machine (Esaote, Italy) equipped with 6–18 and 4–13 MHz broad band multi-frequency linear transducer. US synovitis GS and PD signals were semiquantitatively graded from 0 to 3. Total PD and GS synovitis scores of all sites are recorded as sum scores of PD and GS, respectively. The clinical response was evaluated according to the EULAR response criteria at 3rd month. Potential ultrasonographic predictors of response were identified using multivariate binary logistic regression models. Results: The study cohort consisted of 42 RA patients (F/M⫽33/9, mean age 49.0⫾10.7 years) with a mean disease duration of 9.1⫾7.5 years and mean baseline DAS28 score of 5.5⫾1.0. Rheumatoid factor (RF) and anti-cyclic citrullinated peptide antibody (CCP) positivity were 76.2% and 64.3%, respectively. Baseline characteristics of TNFi responders (30/42) and non-responders (12/42) are shown in Table 1. Swollen joint count (p⫽ 0.05), sum scores of baseline PD (p⫽ 0.048), GS (p⫽ 0.048) and PD⫹GS (p⫽ 0.046) differed significantly between responders and non-responders. Baseline PD sum score was the only ultrasonographic parameter in the multivariate analysis predicting which patients achieve good/moderate EULAR response with TNFi at 3rd month (p⫽ 0.004). The mean PD⫹GS, PD and GS sum scores decreased significantly from baseline to 3 months (p⬍0.001 for all parameters) whereas decrease between 3rd and 6th months was nonsignificant for all US parameters (p⫽0.128, p⫽0.266, p⫽ 0.105, respectively).

Background/Purpose: Extremity-MRI is becoming important in Rheumatoid Arthritis (RA) research as it is a sensitive tool to assess inflammation. Inflammation is measured by three features: bone marrow edema (inflammation of bone), synovitis and tenosynovitis. Part of the inflammatory lesions detected by MRI are subclinical, indicating that these were not observed by physical examination. The relevance of subclinical inflammation detected by MRI in a population of early arthritis patients for to the disease course is not known. In this study we investigated the relevance of subclinical inflammation on MRI in early arthritis patients for radiological joint damage. Methods: 179 early arthritis patients (median symptom duration of 15 weeks) included in the Leiden Early Arthritis Clinic underwent a 68-tender and 66-swollen joint count (including MCP, wrist and MTP joints) and a 1.5T MRI of the MCP (2–4), wrist and MTP (1–5) joints at the most painful side at baseline. Synovitis and bone marrow edema were scored according to the RAMRIS method; tenosynovitis at the wrists and MCP joints was also determined. Scoring was performed by two readers and the average scores were studied. Radiographs of hands and feet were made at baseline and after one year of follow-up and scored according to the Sharp-van-der-Heijde scoring method (SHS). Results: 1,790 small joints of 179 patients were studied. Of these joints 1,525 (85%) were not swollen at physical examination. Of all clinically non-swollen joints, any subclinical inflammation was present on MRI in 25%; 15% had bone marrow edema, 15% synovitis and 18% tenosynovitis. When evaluating these clinically non-swollen joints in relation to the radiographs at year-1, 7% of the joints had SHS ⱖ1 and 2% SHS progression ⱖ1. The subclinical inflamed joints were evaluated with progression of radiological joint damage over the first year (SHS progression ⱖ1) as outcome. When comparing the non-swollen joints with and without any inflammation on MRI, 4% versus 1% had SHS progression ⱖ1, (RR 3.5 (1.3–9.6)). For bone marrow edema this was 6.5% versus 1% (RR 5.3 (2.0–14.0), for synovitis this was 5% versus 1% (RR 3.4 (1.2–9.3)) and for tenosynovitis this was 4% versus 1% (RR 3.0 (0.7–12.7)). Hence, although radiological progression was infrequent, it was significantly increased in joints with subclinical synovitis or bone marrow edema. For comparison we studied the outcome of the swollen joints. 8% of all swollen joints had SHS progression ⱖ1 over the first year, indicating that the frequency of progression in subclinical inflamed joints almost equaled the frequency of progression in clinically swollen joints. Conclusion: Joints with subclinical inflammation detected by MRI have an increased risk of progression in structural damage in early arthritis. This points to the relevance of subclinical inflammation for the disease outcome in RA.


Table 1. Baseline characteristics of TNFi responders and nonresponders Responders (nⴝ30) Age (years) Disease duration (years) Treatment delay (months) RF titer (IU/mL) Anti-CCP titer (U/mL) DAS28 Pain VAS (0-100 mm) ESR (mm/h) CRP (mg/L) Tender joint count (0–28) Swollen joint count (0–28) HAQ score Prednisolone dose, mg/day Sum score of PD (0–84) Sum score of GS (0–84) Sum score of PDⴙGS (0–168)

48.8 ⫾ 10.6 8.0 ⫾ 7.2 32.2 ⫾ 51.0 166.4 ⫾ 233.9 78.7 ⫾ 147.0 5.5 ⫾ 0.9 66.0 ⫾ 21.7 43.7 ⫾ 20.7 27.2 ⫾ 34.1 8.0 ⫾ 6.8 5.7 ⫾ 5.0 1.1 ⫾ 0.6 5.7 ⫾ 3.1 12.5 ⫾ 9.8 16.1 ⫾ 11.7 29.0 ⫾ 21.4

Non-responders (nⴝ12)

p Value

ⴞ ⴞ ⴞ ⴞ ⴞ ⴞ ⴞ ⴞ ⴞ ⴞ ⴞ ⴞ ⴞ ⴞ ⴞ ⴞ

0.81 0.13 0.85 0.31 0.10 0.99 0.067 0.61 0.90 0.12 0.05 0.22 0.93 0.048 0.048 0.046

49.6 11.9 35.2 295.1 232.2 5.5 79.2 40.0 28.6 12.0 9.8 1.3 5.8 20.1 25.7 45.6

11.5 7.8 38.2 586.8 457.5 1.3 16.7 22.5 28.4 9.5 7.8 0.4 2.2 13.2 15.6 28.5

The values were presented as mean⫾SD

Conclusion: Our data underline that baseline PD scores, despite similar clinical features, can predict which patients will respond to TNFi therapy. Ultrasonographic response to TNFi treatment can be achieved substantially in the first 3 months. Beyond 3rd month changes in US scores are mostly nonsignificant. Disclosure: N. Inanc, None; G. Ozen, None; H. Direskeneli, None.

Disclosure: A. Krabben, None; W. Stomp, None; J. A. B. van Nies, None; T. W. J. Huizinga, None; D. van der Heijde, None; J. L. Bloem, None; M. Reijnierse, None; A. H. M. van der Helm-van Mil, None.

2805 Synovial Explant Inflammatory Mediator Production Is Associated With Synovitis While Not With Bone Marrow Edema In Rheumatoid Arthritis: A Cross Sectional Study. Martin Andersen1, Mikael Boesen2, Karen Ellegaard1, Robin Christensen1, Kalle So¨derstro¨m3, Søren Torp-Pedersen1, Bente Danneskiold-Samsøe1, Else Marie Bartels1, Nina Vendel4, Niels H. Søe5, Pieter Spee3, Ulrik GW Mørch6, Lars Karlsson3 and Henning Bliddal1. 1 The Parker Institute, Department of Rheumatology, Copenhagen University Hospital, Bispebjerg and Frederiksberg, Denmark, Frederiksberg, Denmark, 2 Department of Radiology, Copenhagen University Hospital, Bispebjerg and Frederiksberg, Denmark, Frederiskberg, Denmark, 3Translational Immunology, Biopharmaceutical Research Unit, Måløv, Novo Nordisk, Denmark, Måløv, Denmark, 4Department of Anaesthesiology, Intensive Care and Operations, Gentofte University Hospital, Denmark, Hellerup, Denmark, 5 Department of Orthopedics, Section of Hand Surgery, Gentofte University Hospital, Denmark, Hellerup, Denmark, 6Biomarkers, Søborg, Novo Nordisk, Denmark, Søborg, Denmark. Background/Purpose: Synovitis and bone damage may represent two distinct but overlapping pathological processes in rheumatoid arthritis (RA).

S1200

Whereas the bulk of synovial cells contribute to general joint inflammation, the interplay between synovial cells, neighboring cartilage and bony tissue at the margins of the joint contribute to bone degradation(1). The aim of this study was investigate if the levels of inflammatory cytokines interleukine 6 (IL-6), interleukine 8 (IL-8), macrophage chemo-attractant protein 1 (MCP-1) and macrophage inflammatory protein 1 beta (MIP-1b), which all play key roles in RA pathology, correlate with disease intensity and/or can discriminate between key pathological processes in RA. Methods: Fifty-seven synovial sites from the hand joints of twenty-five RA patients were evaluated by color Doppler ultrasound (CDUS, 52 synovial sites) and/or high field magnetic resonance imaging (MRI, 39 synovial sites) and subsequently synovectomized by a needle-arthroscopic procedure. Synovial tissue was cultured for 72h, and the concentration of Il-6, Il-8, MCP-1 and MIP-1b in the culture supernatants were measured by Multiplex immunoassays. Ultrasound Doppler was determined as the color fraction. MRI bone marrow edema, synovitis, and erosion score were estimated by the rheumatoid arthritis magnetic resonance score [RAMRIS]. The concentrations of inflammatory mediators were compared to imaging data from the individual sites using repeated samples from the same patients. Since data was clustered within patients, a mixed linear model was applied for the statistical analysis. Parsimony in the statistical model was achieved omitting covariates with (P⬎0.1) from the model. Results: Ultrasound color fraction and the RAMRIS synovitis score were statistically associated with the release of MCP-1 (P⫽0.004 and P⫽0.04, respectively), IL-8 (P⫽0.04 and P⫽0.04, respectively), and MIP-1b (P⫽0.01 and P⫽0.05, respectively), but not with IL-6 (P⫽0.29 and P⫽0.21, respectively). IL-6 was the only supernatant mediator, which was associated to the RAMRIS erosion component (P⫽0.002). There were no statistically significant associations between mediator release and the RAMRIS bone marrow edema component: MCP-1(P⫽0.07), IL-6 (P⫽0.93), IL-8 (P⫽0.40) and MIP-1b (P⫽0.96). Conclusion: Levels of IL-8, MCP-1, MIP-1b were significantly associated with clinical assessment of synovitis, but not bone erosion, indicating that these molecules play key roles in general inflammation in the joint, and underlining their potential as biomarkers to measure levels of synovitis. In contrast, levels of IL-6 significantly associted with clinical assessment of bone erosion, indicating that this cytokine, which is known to have direct activating effects on osteoclasts, has potential as a biomarker for measuring bone destructive activity in RA patients. Acknowledgments: Supported by unrestricted grants from Novo Nordisk, The Danish Agency for Science Technology and Innovation and the Oak Foundation.

Disclosure: V. Varisco, None; M. Vigano’, None; A. Batticciotto, None; P. Lampertico, None; A. Marchesoni, None; P. Gibertini, None; R. Pellerito, None; G. Rovera, None; R. Caporali, None; M. Todoerti, None; M. Covelli, None; A. Notarnicola, None; P. Sarzi-Puttini, None.

2807

Disclosure: M. Andersen, Novo Nordisk, 3; M. Boesen, None; K. Ellegaard, None; R. Christensen, Abbott, Axellus A/S, Bristol-Myers Squibb, Cambridge Weight Plan, Norpharma, Pfizer, and Roche, 5, Axellus A/S, Cambridge Weight Plan, Mundipharma, and Roche, 2, Abbott, Axellus, Bayer HealthCare Pharmaceuticals, Biogen Idec, Bristol-Myers Squibb, Cambridge Weight Plan, Ipsen, Laboratoires Expanscience, MSD, Mundipharma, Norpharma, Pfizer, Roche, and Wyeth, 8; K. So¨derstro¨m, Novo Nordisk, 1, Novo Nordisk, 3; S. Torp-Pedersen, None; B. Danneskiold-Samsøe, None; E. M. Bartels, None; N. Vendel, None; N. H. Søe, None; P. Spee, Novo Nordisk, 1, Novo Nordisk, 3; U. G. Mørch, Novo Nordisk, 1, Novo Nordisk, 3; L. Karlsson, Novo Nordisk, 1, Novo Nordisk, 3; H. Bliddal, None.

ACR Concurrent Abstract Session Infection Related Rheumatic Diseases Wednesday, October 30, 2013, 9:00 AM–10:30

Background/Purpose: B cell-targeted therapy using rituximab (RTX), anti-CD20 monoclonal antibody, is an effective treatment for Rheumatoid arthritis (RA). However, the safety of such therapy in hepatitis B surface antigen (HBsAg) negative/anti-hepatitis B core antigen (anti-HBc) positive recipients is unknown. The aim of this study is to evaluate the risk of HBsAg seroreversion in this kind of patients. Methods: We retrospectively reviewed 306 patients of our multicentre database affected by RA and treated with RTX from August 2006 to December 2011 in 5 italian outpatient rheumatologic Clinics. Complete serological screening for HBV status before the first administration of RTX and adequate post-treatment follow-up were available in 35 HBsAg negative/ anti-HBc positive patients who did not undergo antiviral prophylaxis with Lamivudine. All patients (75% female, median age 60 years, median disease duration 8 yrs, 100% serum HBV DNA negative by sensitive PCR assay, 87% anti-HBs positive) has been treated with one or more disease-modifying anti-rheumatic drugs (83% MTX, 26% CYS, 80% PDN, 8% LEF, 6% AZA) and were eligible for RTX therapy according to the international guidelines. RTX was administered for a median of 3 cycles (range: 1–8) and was ongoing in 76% of cases. Clinical and laboratory examinations, including serum HBsAg and serum HBV DNA were assessed every 6 months or in case of alanine aminotransferase (ALT) elevation and at the end of the follow-up period. Results: During a median follow-up of 45 months (range: 12–80) 27% of patients had anti-HBs titer reduction (2 patient with a complete lost of anti-HBs levels). All patients remained viremic free except one (3%) who had an increase of serum HBV DNA (from undetectable to 24 and 44 IU/mL, 1 week apart) not associated to either HBsAg seroreversion or ALT increase. The mild virological breakthrough occurred 5 months after the first RTX administration and required Lamivudine treatment which successfully suppressed viral replication. Another patient (3%) had an ALT flare which was not related to HBV reactivation. Conclusion: A retrospectively review of our multicentre experience suggest that in RA patients HBsAg negative/anti-HBc positive treated with RTX, Lamivudine prophylaxis should be avoid using an adequate monitoring of HBsAg or HBV DNA levels.

AM

2806 Low Risk Of Hepatitis B Virus Surface Antigen Seroreversion In HBsAg Negative/Anti-Hbc Positive Carriers Undergoing Rituximab For Rheumatoid Arthritis. Valentina Varisco1, Mauro Vigano’2, Alberto Batticciotto1, Pietro Lampertico3, Antonio Marchesoni4, Patrizia Gibertini4, Raffaele Pellerito5, Guido Rovera5, Roberto Caporali6, Monica Todoerti6, Michele

Management Of Vertebral Osteomyelitis: A Randomized Clinical Trial Comparing 6 Versus 12 Weeks Of Antibiotic Treatment. Louis Bernard1, Aure´lien Dinh2, Idir Ghout3, Vale´rie Zeller4, Bertrand Issartel5, Nadia Belmatoug6, Michel Dupon7 and Denis Mulleman8. 1University hospital of Tours, Tours, France, 2University hospital of Paris, Garches, France, 3University Hospital of Paris, Boulogne, France, 4La Croix Saint Simon Hospital, Paris, France, 5Clinique du Tonkin, Lyon, France, 6University hospital of Paris, Clichy, France, 7University Hospital of Bordeaux, Bordeaux, France, 8 CHU Trousseau Tours, Tours, France. Background/Purpose: As for most of bone and joint infection, optimal treatment duration for vertebral osteomyelitis (VO) is unknown. In an era of increasing bacterial resistance, we compare the effectiveness of 6 and 12 weeks antibiotic treatment. Methods: We carried out anational, open blind, randomized controlled non-inferiority multicentre trial in 2 parallel groups comparing 6 versus 12 weeks of antibiotic treatment duration. Adult patients with clinical and radiologic diagnosis of VO (MRI or CT) and a reliable microbiological identification (positive and significative blood culture or needle biopsy) were recruited since 2007 to 2011. At one year, after the end of antibiotic therapy, cure was define as the absence of clinical and biological signs related to VO without the need for additional or alternative antibiotic therapy. Results: 359 patients were randomized, 175 in the group 12 weeks treatment group and 176 in the group 6 weeks; 240 (69%) were male with mean age 61 years old. Eight patients were secondarily excluded. Median

S1201


References: (1) Lafeber FP, Van der Laan WH. Progression of joint damage despite control of inflammation in rheumatoid arthritis: a role for cartilage damage driven synovial fibroblast activity. Ann Rheum Dis 2012; 71(6):793–5.

Covelli7, Antonella Notarnicola7 and Piercarlo Sarzi-Puttini1. 1Rheumatology Unit, L. Sacco University Hospital, Milan, Italy, 2San Giuseppe hospital, University of Milan, Milan, Italy, 3Fondazione IRCCS Ca` Granda Ospedale Maggiore Policlinico, University of Milan, Milan, Italy, 4Istituto Ortopedico Gaetano Pini, Milano, Italy, 5Ospedale Mauriziano, Turin, Italy, 6Rheumatology, IRCCS Policlinico San Matteo Foundation and University of Pavia, Pavia, Italy, 7D.I.M.I.M.P, Rheumatology Unit—University of Bari, Bari, Italy.

duration between first symptoms and diagnosis was 49 (1–80) days; 182 (52%) patients were febrile at diagnosis and 237 (68%) had positive blood culture. Infective endocarditis was present in 70 (20%) cases. Main bacteria involved were Staphylococcus aureus (n⫽143, 41%), Coagulase negativeStaphylococci (n⫽52, 15%) and Streptococcus non enterococcus (n⫽63, 18%). At baseline, characteristics were similar in both group of treatment duration. During follow up, difference was non significative in both groups regarding median duration of parenteral antibiotictherapy, median length of hospitalization stay, most frequent antibiotics for oral treatment (rifadin and fluoroquinolon). Adverse events occurred in 100 cases of whom 26 deaths. In the intention to treat analysis, cure rates in the 6 weeks treatment group and in the 12 weeks treatment group were respectively 91% (159/171) and 91% (160/176)with a difference non significative between the two groups. Conclusion: We have demonstrated in this large multicentre randomized clinical trial that 6 weeks of antibiotic treatment duration is as effective as 12 for VO. Disclosure: L. Bernard, None; A. Dinh, None; I. Ghout, None; V. Zeller, None; B. Issartel, None; N. Belmatoug, None; M. Dupon, None; D. Mulleman, None.

2808


Favorable Outcome Of Hepatitis E Virus Infection In Patients With Inflammatory Arthritides Treated With Immunosupressants. He´le`ne Bauer1, Cećile Luxembourger2, Sophie Fournier2, Alain G. Cantagrel3, Jean Marie Peron2, Anne Marie Roque Afonso4, Pascal Claudepierre5, S. Fabre6, Christophe Hudry7, Guillaume Lefevre8, Antoine Martin9, Laurent Messer10, Beátrice Pallot Prades11, Christian Roux12, Christelle Sordet13, Claire Veissier14, Daniel Wendling15, Jacques-Eric Gottenberg13 and Jean Sibilia16. 1 CHU de Hautepierre, Strasbourg, France, 2CHU de Toulouse, Toulouse, France, 3Centre Hospitalier Universitaire de Toulouse, Toulouse, France, 4 virologie hoˆpital Paul Brousse, Villejuif, France, 5Henri Mondor Teaching Hospital, AP-HP, Cre´teil, France, 6CHU MONTPELLIER, Montpellier, France, 7Hopital Cochin, Paris, France, 8Lille University Hospital, Internal Medicine Department, Lille, France, 9Centre Hospitalier De St Brieuc, St Brieuc, France, 10Hopitaux Civils de Colmar, Colmar, France, 11CHU St Etienne, St Etienne, France, 12CHU L’ Archet University Nice, Nice, France, 13 Strasbourg University Hospital, Strasbourg, France, 14CHU de Bordeaux, Bordeaux, France, 15University Hospital, Besancon, France, 16CHU Hautepierre, Strasbourg, France. Background/Purpose: In inflammatory rheumatic diseases, the prevalence, clinical presentation and outcome of hepatitis E, an ubiquitous viral infection, remain unknown. We therefore addressed these issues in a large retrospective study. The main objective of this study was to investigate the severity of acute hepatitis E and the risk of chronic viral replication in patients suffering from inflammatory rheumatisms and treated with immunosuppressive drugs. Methods: All French rheumatology and internal medicine practitioners registered on the Club Rhumatisme et Inflammation (nearly 2,000 physicians),were repeatedly sent newsletters asking to report observations of acute hepatitis E virus (HEV) infection in patients with inflammatory arthritides. They were subsequently sent a standardized and detailed questionnaire on baseline characteristics of the patients and the course of HEV infection. Results: Nineteen observations of hepatitis E were collected. They occurred in patients with rheumatoid arthritis (n ⫽ 9), axial spondyloarthritis (n ⫽ 4), psoriatic arthritis (n ⫽ 3), juvenile idiopathic arthritis (n ⫽1), Jaccoud arthropathy (n ⫽ 1) and undifferentiated arthritis (n ⫽ 1), treated with methotrexate (n ⫽ 13), anti-TNF␣ therapy (n ⫽ 6), rituximab (n ⫽ 4), abatacept (n ⫽ 3) or tocilizumab (n ⫽ 1). Eight patients were treated with corticosteroids with a median dose of 4.5 mg/d. Most of the patients had few symptoms, except asthenia. A woman, suffering from psoriatic arthritis treated with cyclosporine developed a bilateral Parsonage Turner syndrome, which was considered as an extrahepatic HEV-related manifestation. All of the patients had acute elevation of aspartate and alanine aminotransferase levels, and 7 patients also had moderate cholestasis. Two patients had an acute hepatic failure with decreased prothrombin time. The hepatitis E diagnosis either relied on positive PCR detection for HEV RNA (n ⫽ 13 patients), or on positive IgM result without PCR assessment (n ⫽ 2) or with negative PCR for HEV (n ⫽ 4). No other aetiology of the hepatitis could be found, except in 1 patient who had a HEV and HAV coinfection. Mean current follow-up is 13.4 months (standard deviation: 8.8 months). Treatment of HEV infection included the discontinuation of immunosuppressants in 18 out of 19 patients (adalimumab was maintained in 1 patient) and ribavirine in 5 patients. Liver enzymes normalized within 1 month in 7 patients, 3 months in 6 patients, and 4 months in 2 patients. Ten previously positive PCR tests were repeated and

all became negative within 3 months after HEV infection.Among the 15 patients with a follow-up of at least 3 months, all are considered to be cured by their clinicians and an immunosuppressant (the same as before HEV infection in 14 patients, etanercept in 1 patient) could be reinitiated in all of them with no viral reactivation. Conclusion: The diagnosis of acute hepatitis E should be considered in patients with inflammatory arthritides treated with immunosuppressants and elevated liver enzymes. The outcome of HEV infection is usually favourable, after transient discontinuation of immunosuppressants associated or not with ribavirin, with no reported evolution towards chronicity. Disclosure: H. Bauer, None; C. Luxembourger, None; S. Fournier, None; A. G. Cantagrel, None; J. M. Peron, None; A. M. Roque Afonso, None; P. Claudepierre, None; S. Fabre, None; C. Hudry, None; G. Lefevre, None; A. Martin, None; L. Messer, None; B. Pallot Prades, None; C. Roux, None; C. Sordet, None; C. Veissier, None; D. Wendling, None; J. E. Gottenberg, None; J. Sibilia, None.

2809 Prevalence and Clinical Presentation Of Lyme Arthritis In a Large Cohort Of Patients With Recent-Onset Arthritis. Jeska K. De VriesBouwstra, Nathalie D. van Burgel, Tom Vreeswijk, Aloys C. M. Kroes, Tom W. J. Huizinga and Annette H. M. van der Helm-van Mil. Leiden University Medical Center, Leiden, Netherlands. Background/Purpose: Lyme arthritis is a relatively uncommon form of arthritis that is relevant to identify because it requires specific therapy. The presents study aimed to describe 1. the seroprevalence of Lyme antibodies, 2. the prevalence of Lyme arthritis, and 3. the diagnostics accuracy of serologic testing for Borreliae among patients with recent-onset arthritis in The Netherlands. Methods: The present study included all patients who presented to the Leiden Early Arthritis Clinic during the period of February 1993 to April 1997 (cohort 1) and the period July 2003 to June 2008 (cohort 2). In sera from cohort 1, antibodies against Borreliae were detected by the IgG and IgM flagellin-Enzyme Immune Assay (EIA). All available positive samples were retested with the IgG/IgM C6 EIA. In cohort 2, the C6 Lyme EIA Kit was used. All positive EIA results from both cohorts were confirmed by the Borrelia Europe LINE blot and a second EIA, the Liaison® Borrelia burgdorferi IgG (VlsE) and IgM (VlsE⫹OspC). The clinical records of the patients with positive or equivocal serology were thoroughly studied; Lyme disease was not diagnosed in case of a clear alternative diagnosis. The proportion of patients with Lyme arthritis was determined, as well as the diagnostic accuracy of Borrelia serology. Results: Of 1180 patients with recent-onset arthritis, 53 patients had positive serology, indicating a seroprevalence of 4,5% (Figure 1 and 2). Eight cases of definite Lyme arthritis (0.7%) were identified These patients were characterized by younger age, and typically presented with monoarthritis or oligoarthritis including a knee. Retrospectively, possible Lyme arthritis was diagnosed in seven (0,6%) patients. In the remaining patients with positive or equivocal serology different clinical diagnoses were made (rheumatoid arthritis n⫽ 14; spondylarthritis n⫽ 11; systemic inflammatory disease n⫽ 3; osteoarthritis n⫽ 2; other n⫽8). Based on the low prevalence of Lyme arthritis, the positive predictive value of serology for Lyme disease was low (10–28%). By selecting patients with a higher prior chance of Lyme disease, eg in patients presenting with an oligo- or monoarthritis of large joints, the positive predictive value increased to 67%.

S1202

Table 1. Patients characteristics and interferon-gamma release assay (IGRA) conversion rate. Characteristics

Conclusion: The prevalence of Lyme arthritis among patients with recent-onset arthritis in The Netherlands is low, 0.7%. Although sensitivity and specificity of Lyme serology are reported to be good, serologic testing for Lyme arthritis contributes poorly to clinical decision making in patient with recent-onset arthritis due to the low a priori chance on Lyme arthritis.

Diagnosis AS RA Age, yrs, mean (SD) ⬍40 ⱖ 40 Gender Male Female TNF␣ antagonists Etanercept Adalimumab Infliximab Duration of TNF␣ inhibitor therapy ⬍ 2 yrs ⱖ 2 yrs Tuberculosis prophylaxis No Yes

All (Nⴝ117)

Conversion Adjusted (Nⴝ12) OR

92 (78.63) 25 (21.37)

7 (7.61) 5 (20.00)

0.42 1.00

72 (61.54) 45 (38.46)

1 (1.39) 11 (24.44)

76 (64.96) 41 (35.04)

95% CI

p-value

(0.06–3.07)

0.394

1.00 21.19

(2.12–211.57)

0.009*

6 (7.89) 6 (14.63)

1.00 0.34

(0.05–2.30)

0.271

43 (36.75) 37 (31.62) 37 (31.62)

7 (16.28) 1 (2.70) 4 (10.81)

1.00 0.19 0.53

(0.02–1.96) (0.11–2.47)

0.164 0.418

74 (63.25) 43 (36.75)

5 (6.76) 7 (16.28)

1.00 2.31

(0.46–11.70)

0.313

102 (87.18) 15 (12.82)

8 (7.84) 4 (26.67)

1.00 1.70

(0.25–11.54)

0.590

*p ⬍ 0.05

Disclosure: J. K. De Vries-Bouwstra, None; N. D. van Burgel, None; T. Vreeswijk, None; A. C. M. Kroes, None; T. W. J. Huizinga, None; A. H. M. van der Helm-van Mil, None.

2810 Follow-Up Testing Of Interferon-Gamma Release Assays For The Diagnosis Of Hidden Tuberculosis Infection In Patients Receiving Tumor Necrosis Factor Alpha Antagonists. Chan-Nam Son, Tae-Hwan Kim, Il-Hoon Sung, Jae-Bum Jun and Dae-Hyun Yoo. Hanyang University Hospital for Rheumatic Diseases, Seoul, South Korea.

Disclosure: C. N. Son, None; T. H. Kim, None; I. H. Sung, None; J. B. Jun, None; D. H. Yoo, None.

2811 Disease Characteristics and Treatment Patterns In US Veterans With Rheumatoid Arthritis and Concomitant Hepatitis C Infection. Ruchika Patel and Joshua Baker. University of Pennsylvania, Philadelphia, PA. Background/Purpose: The prevalence of concurrent rheumatoid arthritis (RA) and hepatitis C (HCV) is estimated at 0.02%, affecting around 40,000 Americans. To our knowledge, no existing studies evaluated disease characteristics and treatment patterns in RA patients with and without HCV infection. Our aim was to evaluate disease characteristics, component and composite measures of disease activity, and treatment patterns in HCVpositive subjects with RA compared to HCV-negative subjects with RA in a large national registry Methods: We utilized the Veterans Affairs Rheumatoid Arthritis Registry (VARA) to identify subjects with RA and concomitant HCV. Subjects who were identified by the physician to have comorbid HCV infection at the time of enrollment were considered HCV-positive. The registry includes baseline and longitudinal measures that are updated and entered into the VARA database by the treating physicians at routine clinical visits. Differences in disease characteristics, disease activity measures, and treatment patterns were assessed over the first 5 clinical visits. Chi2 tests and T-tests or ranksum tests of significance were utilized for group comparisons. Linear and logistic regression analyses assessed group differences after adjustment for potential confounders (age, race and days from enrollment). Results: Of 1870 registry participants, 53 were identified with concomitant HCV (2.8%). At enrollment, HCV-positive subjects were younger [60.7 ⫾5.9 vs. 67.1 ⫾11.1 (p⬍0.0001)], more likely to be African American (62% vs. 38%, p⫽0.001) and to smoke (p⫽0.001). Other disease characteristics (anti-citrullinated protein antibodies, erosions and disease duration) were similar between groups. At enrollment, HCV-positive subjects had greater DAS28 [4.7 ⫾1.85 vs. 4.0 ⫾1.61 (p⫽0.05)], tender (p⫽0.001) and swollen joint counts (p⫽0.006). By visit 2, there were no observed group differences in disease activity. During the five-visit period, HCV-positive patients had more frequent follow up [12.1 ⫾ 5.3 vs. 17.4 ⫾ 11.8 (p⫽0.009)]. No differences were observed in patient global score, global evaluator score,

S1203


Background/Purpose: Anti-tumor necrosis factor alpha (Anti-TNF␣) therapy is often used in patients with rheumatic diseases who do not respond to conventional treatment. Risk of tuberculosis infection is high in patients receiving anti-TNF␣ treatments. Therefore, tuberculosis infection prophylaxes are recommended prior to anti-TNF␣ therapy if a tuberculin skin test or interferon-gamma release assay (IGRA) is positive. However, little data is available on the conversion of IGRAs in patients with rheumatic diseases who received anti-TNF treatment. We evaluated the utility of follow-up IGRAs for the diagnosis of latent tuberculosis infection and newly developing tuberculosis in patients receiving TNF␣ antagonists. Methods: The study participants (n⫽117) were enrolled from September 2008 to August 2012, among ankylosing spondylitis (AS) and rheumatoid arthritis (RA) patients registered in tertiary care center. These patients had a negative IGRA screening before receiving antiTNF␣ therapy and were evaluated by follow-up IGRA for detection of latent and newly developing tuberculosis until May 2013. We retrospectively examined data of the subjects according to age, gender, tuberculosis prophylaxis, IGRA conversion and TNF␣ blockers, including type and treatment duration. The QuantiFERON-TB Gold In-Tube test was used for the initial and follow-up screenings to detect tuberculosis infection. Results: The median interval between initial and follow-up IGRAs in patients was 20.5 months, and the median age was 38.9 years. Of the 117 patients (92 AS and 25 RA), 105 patients (89.7%) showed consistently negative results, and IGRA conversion was found in 12 patients (10.3%). Among the 92 AS patients, IGRA conversion was found in 7 patients (7.6 %). Among the 25 RA patients, 5 patients with positive conversion (20%) were found. One RA patient developed extrapulmonary tuberculosis. IGRA conversion rate was higher in older patients (ⱖ 40 years) (24.4%) than it was in younger patients (⬍ 40 years) (1.4%) (p ⬍ 0.05) (Table 1). AS patients were younger and used anti-TNF␣ inhibitors for extended time period. IGRA conversion rate was not significantly different between AS (7.6%) and RA (20%) (p ⫽0.13).

Conclusion: In patients with rheumatic diseases receiving TNF␣ blocker, IGRA conversions were observed. Patients with old age had higher IGRA conversion rate than patients with young age. These data suggest that follow-up IGRAs may identify false negative results of screening test and detect latent tuberculosis reactivation and new developing tuberculosis in patients receiving TNF␣ blockers in Korea. Larger-scaled studies may be needed for further evaluation about follow-up IGRAs.

Multidimensional Health Assessment Questionnaire (MD-HAQ), ESR, or CRP. At baseline, methotrexate use was less common among HCV-positive subjects (Table 1). By visit 2, subjects with HCV were more likely to use prednisone and, by visit 4, were more likely to use hydroxychloroquine. These differences in medications use persisted through the fifth visit (Table 1). Table 1. Medication use among HCV-positive and HCV-negative subjects with Rheumatoid Arthritis at enrollment and over the initial 5-visits in VARA. Odds of use is adjusted for group differences in age, race, and days from enrollment for each visit. Visit # Hydroxychloroquine 1 2 3 4 5 Sulphasalazine 1 2 3 4 5 Methotrexate 1 2 3 4 5 Prednisone 1 2 3 4 5 Anti-TNFs 1 2 3 4 5

HCVⴚ (%)

HCVⴙ (%)

Odds Ratio (95% CI)

28 28 26 25 23

34 36 42 43 42

1.26 (0.71–2.27) 1.23 (0.68–2.25) 1.59 (0.87–2.93) 1.91 (1.00–3.67)* 2.29 (1.12–4.68)*

12 13 13 12 11

21 21 23 19 21

1.68 (0.84–3.34) 1.43 (0.71–2.89) 1.58 (0.79–3.16) 1.16 (0.55–2.46) 1.69 (0.78–3.59)

45 47 43 40 36

21 21 19 15 15

0.32 (0.16–0.62)* 0.24 (0.12–0.47)* 0.20 (0.01–0.40)* 0.15 (0.07–0.34)* 0.16 (0.07–0.37)*

37 38 34 31 28

40 55 49 53 48

1.14 (0.65–2.01) 1.85 (1.02–3.38)* 1.44 (0.78–2.66) 2.32 (1.16–4.67)* 2.41 (1.13–5.14)*

19 19 19 17 17

26 30 40 30 26

1.38 (0.73–2.60) 1.67 (0.89–3.13) 2.32 (1.25–4.29)* 1.71 (0.88–3.33) 1.46 (0.72–2.98)


* P value ⬍ 0.05

Background/Purpose: For 8 years, the New York City Rheumatology Objective Self Assessment Clinical Exam (NYC-ROSCE) has been used to assess trainee competencies such as patient care and communication. Most recently, we have refocused the NYC-ROSCE to assess professionalism, using a validated professionalism questionnaire. This year we asked which individual skills were most likely to correlate with overall achievement in the area of professionalism. Methods: The 2012 NYC-ROSCE included 5 patient-centered stations focusing on rheumatic disease. For the first time, stations were designed to challenge fellows with patient psychosocial dilemmas, in order to assess specific aspects of the patient-doctor relationship including physician professionalism. Rheumatology trainees (n⫽28) and faculty MD-evaluators (n⫽32) from 6 NY rheumatology training programs participated. Professional actors (n⫽30) were trained to role-play patients. Quantitative assessments of the trainees were made at each station (9-point Likert scale) by patient- and MD-evaluators in the areas of: maintaining composure, partnering with the patient, being open and honest, professionalism, empathy and accountability. Qualitative free text comments were solicited after each station, regarding the strengths and weaknesses of each fellow’s performance during the encounter. In addition, trainees rated their own performance after each encounter. Immediate oral feedback on professionalism was given to each trainee by both patient-actors and MD-evaluators. Results: When assessing professionalism, MD-evaluators tended to rate trainees lower (6.74⫹/⫺0.58) than patient-actors (7.12⫹/⫺0.78), suggesting that physicians and patients may apply different criteria when assessing professionalism during observed encounters. Trainee self-evaluations for professionalism were in the range of the MD-evaluators rather than the patient-actors (6.70⫹/⫺1.16). MD-evaluators also tended to give lower ratings for accountability (6.32⫹/⫺0.82) than patient-actors (6.68⫹/⫺0.85) or the trainees themselves (6.53⫹/⫺1.04). In contrast, ratings of empathy were found to be closely in agreement for trainees (6.62⫹/⫺1.04), MDevaluators (6.58⫹/⫺0.69) and patient-actors (6.55⫹/⫺1.05. MD-evaluators most often commented on the trainees’ word choice and body language, ability to formulate a clear plan, ability to recognize and address patient’s fears, and the need to actively partner with the patient. For all evaluators, excessive use of medical jargon was most often cited as the area that needed improvement. The use of jargon correlated inversely with the ratings of professionalism for MD-evaluators and patient-actors. Conclusion: The introduction this year of new and challenging patient encounters allowed us to assess which patient-centered traits most contributed to perceptions of achievement in the area of professionalism. We found that our evaluators considered the ability to minimize medical jargon to be one of the most important markers of the competency of professionalism. This finding suggests that patients view communication as correlating most strongly with their doctor’s professionalism.

Conclusion: At enrollment, HCV-positive subjects were younger, more likely to be African American, smoke, had higher joint counts, and were less likely to be prescribed methotrexate. Our data suggest more frequent follow up and use of prednisone and hydroxychloroquine over time among subjects with RA and concomitant HCV, may result in comparable early disease control compared to RA subjects without HCV.

Disclosure: J. Berman, None; J. Aizer, None; A. R. Bass, None; A. Davidson, None; E. Dwyer, None; T. R. Fields, Takeda Pharmaceuticals, 8, Takeda Pharmaceuticals , 9, Savient Pharmaceuticals, 8, Pfizer Pharmaceuticals, 8; J. Kang, None; L. Kerr, None; S. Krasnokutsky-Samuels, None; D. M. Lazaro, None; S. A. Paget, None; J. S. Schwartzman-Morris, None; M. H. Pillinger, None.

Disclosure: R. Patel, None; J. Baker, None.

2813

2812

Expert Panel Consensus On Content Of a Rheumatology Objective Structured Clinical Examination. Lisa G. Criscione-Schreiber1, Marcy B. Bolster2, Beth L. Jonas3, Richard Sloane1, Jeffrey Hawley4 and Kenneth S. O’Rourke5. 1Duke University Health System, Durham, NC, 2Massachussetts General Hospital, Boston, MA, 3University of North Carolina at Chapel Hill, Chapel Hill, NC, 4Duke University School of Nursing, Durham, NC, 5Wake Forest School of Medicine, Winston-Salem, NC.

A Rheumatology Objective Structured Clinical Examination Using Challenging Patient Scenarios Shows Trainee Use Of Medical Jargon Correlates Inversely With Patient Perceptions Of Professionalism. Jessica Berman1, Juliet Aizer2, Anne R. Bass2, Anne Davidson3, Edward Dwyer4, Theodore R. Fields2, Jane Kang4, Leslie Kerr5, Svetlana KrasnokutskySamuels6, Deana M. Lazaro7, Stephen A. Paget2, Julie S. SchwartzmanMorris8 and Michael H. Pillinger9. 1Hospital for Special Surgery, Weill Cornell Medical College, New York, NY, 2Hospital for Special Surgery Weill Cornell Medical College, New York, NY, 3Feinstein Institute for Medical Research, Manhasset, NY, 4Columbia School of Medicine, New York, NY, 5Mount Sinai School of Medicine, New York, NY, 6NYU School of Medicine, New York, NY, 7Brooklyn VA, Brooklyn, NY, 8Albert Einstein College of Medicine, Bronx, NY, 9NYU School of Medicine, Division of Rheumatology, New York, NY.

Background/Purpose: Rheumatology objective structured clinical examinations (ROSCEs) are assessment tools to evaluate learner performance in simulated clinical experiences. ROSCEs allow standardization of learner assessment in competencies that are challenging to measure in other ways. While several regional groups in the United States conduct ROSCEs, none have been validated for use across programs. We aimed to establish agreement among subspecialty experts regarding the important benchmarks for rheumatology fellows to achieve in 11 theoretical ROSCE stations. Methods: First, each author independently created a grid listing important attributes to assess in 11 possible ROSCE stations. The grids were combined (by LCS) into a 173-item checklist survey to assess the importance of each item for future use in ROSCEs. IRB exemption was granted to send the survey by email to 37 expert rheumatologists, defined as individuals who have helped create and administer regional ROSCEs. Participants in the expert

ACR Concurrent Abstract Session Medical Education Wednesday, October 30, 2013, 9:00

AM–10:30 AM

S1204

panel survey rated the importance of each item using a 5-point Likert scale (1⫽not important to 5⫽very important). Minimum and maximum scores, the mean, standard deviation, and lower 95% confidence level for the mean were calculated for each survey item. Consensus for high importance was predefined as a lower bound of the 95% CI ⱖ 4.0. Results: 11 individuals completed the expert panel survey. 123 of the 173 items (71%) met statistical cutoff for consensus to retain, distributed among the 11 proposed stations. Several items were rejected that had population means of ⱖ 4.0 but did not meet the pre-determined definition for consensus. The percentage of retained items for individual stations ranged from 23.5 to 100% (Table). All items were retained for the core interpersonal and professionalism items included as part of general patient counseling and for x-ray interpretation tasks. Only 23% of items were retained for a station involving rehabilitation medicine, and 46% of items for a microscope use/synovial fluid analysis station. Table. Item retention summary. Items were retained if statistical analysis revealed a lower bound of the 95% CI ⱖ 4.0. Station Phone call from primary care provider X-rays Non-formulary request letter to insurance company Osteoporosis/DXA interpretation Rehabilitation General counseling Counseling: antiRo⫹ woman pre-conception Counseling: new scleroderma Counseling: new biologic for RA Knee arthrocentesis Microscope synovial fluid Total

Agreement score range

Percentage of items retained

Number of items assessed

Number of items retained

18

10

1.98–5

55.5%

19 16

19 12

4.19–5 3.03–5

100% 75%

17

14

2.98–4.74

82.4%

17 13 13

4 13 11

2.51–4.6 4.08–5 3.13–5

23.5% 100% 84.6%

11

8

3.13–4.6

72.7%

14

9

3.34–5

64.3%

20 15

16 7

3.18–5 2.70–5

80% 46.7%

173

123

1.98–5

71.1%

Disclosure: L. G. Criscione-Schreiber, None; M. B. Bolster, American College of Rheumatology, 6, Rheumatology Research Foundation, 6, American College of Rheumatology, 6; B. L. Jonas, None; R. Sloane, None; J. Hawley, None; K. S. O’Rourke, None.

2814 The Resident-Fellow Interaction: Limiting Barriers and Maximizing Learning. Eli Miloslavsky1, Amy Sullivan2, Jeremy Richards2, Jakob I. McSparron3, David Roberts2 and Alberto Puig4. 1Massachusetts General Hopsital, Boston, MA, 2Beth Israel Deaconess Medical Center, Boston, MA, 3 Brigham and Women’s Hospital, Boston, MA, 4Massachusetts General Hospital, Boston, MA. Background/Purpose: Subspecialty fellows may play an important role in Internal Medicine (IM) residents’ education and career specialty choice (Horn 2008). This may be particularly important in rheumatology where a projected workforce shortage may require the expansion of the current training capacity (Deal 2007), and over 75% of fellows decide to pursue a career in rheumatology during residency (Kolasinski 2007). In this context, fellows may be underutilized as clinical teachers and role models, possibly

Disclosure: E. Miloslavsky, Genentech Inc, 9; A. Sullivan, None; J. Richards, None; J. I. McSparron, None; D. Roberts, None; A. Puig, None.

2815 Preliminary Entrustable Professional Activities in Muscloskeletal Ultrasound For Rheumatology Fellowship Training. Amy C. Cannella1, Eugene Y. Kissin2, Gurjit S. Kaeley3, Jay B. Higgs4, Kelly J. Caverzagie5 and Karina D. Torralba6. 1Omaha Veterans Affairs Hospital and University of Nebraska Medical Center, Omaha, NE, 2Boston University, Boston, MA, 3 University of Florida, Jacksonville, FL, 4San Antonio Military Medical Center, Fort Sam Houston, TX, 5University of Nebraska Medical Center, Omaha, NE, 6Loma Linda University, Loma Linda, CA. Background/Purpose: As part of the Next Accreditation System (NAS), the Accreditation Council for Graduate Medical Education (ACGME) has mandated a change in the evaluation of postgraduate trainees by including the development of Entrustable Professional Activities (EPAs) for each specialty and subspecialty. EPAs are the routine professional activities which define a profession. They also provide a means of framing the 6 ACGME competencies in a way that allows them to be meaningful to learners and faculty. The American College of Rheumatology (ACR) is developing EPAs for Rheumatology training. Although not currently an ACGME requirement, many fellowship programs have incorporated Musculoskeletal Ultrasound (MSUS) in their curricula. We seek to define EPAs for MSUS in rheumatology as an initial step towards the development of standardized curriculum and assessment tools. Methods: A group of educators utilized a modified nominal group technique (NGT) to develop EPAs for MSUS. A literature search was completed on MSUS education, certification, best practices and curriculum development to generate a list of potential EPAs specific for MSUS in rheumatology, framed within the ACGME competency and milestone vernacular.

S1205


Conclusion: In this single round expert panel survey, we established national consensus on 123 items to assess on 11 proposed ROSCE stations. This study represents the first use of such rigorous methods to establish checklist content agreement for an OSCE in any medical field. Next steps include 1) defining behavioral anchors (milestones) for performance on individual stations and 2) finalizing checklists and station manuals that can be shared among rheumatology programs nationally to improve the validity of regional OSCEs and initiate use in programs without regional collaborators. Use of these checklists by regional rheumatology groups will create the framework for collection of summary data and perhaps eventual creation of a high-stakes rheumatology OSCE.

due to the existence of barriers to an effective resident-fellow interaction (Miloslavsky 2010). To our knowledge, the working relationship between IM subspecialty fellows and IM residents has not been examined. We conducted a study to determine the barriers and facilitating factors to the IM resident-fellow interaction on the wards. Methods: We conducted 4 focus groups: IM residents at the Massachusetts General Hospital (MGH) and the Brigham and Women’s Hospital (BWH), IM fellows at BWH/MGH, and IM fellows at MGH who received teaching awards. There were 32 participants from all residency classes and 7 IM subspecialties including rheumatology. Four investigators analyzed focus group transcripts via a theory driven immersion-crystallization process using activity theory (Engestrom 2011). Results: Five themes of barriers were identified: - Intrinsic: residents’ and fellows’ expectations of each other; absence of familiarity/personal relationship between residents and fellows; interest and availability (fellows - willingness to see the consult, residents - being available/interested in discussing the consult); knowledge (residents - patient history, fellows - subject matter). - Logistical: resident work hours (e.g. frequent patient handoffs); primary team structure (e.g. team-based care model, de-regionalized teams). - Attending-related: primary team attending (role in formulating the consult question, communication); consult attending (role modeling). - Workload: resident and fellow patient census and the duration/ complexity of the consult. - Teaching: time available to teach and fellows’ teaching skills. Focus group participants cited increasing familiarity/personal relationships between residents and fellows as a critical variable in improving the interaction. Other means of overcoming barriers included limiting fellow pushback on residents’ consult requests, standardizing fellows’ and residents’ expectations about the consult interaction, and improving fellows’ teaching skills. Positive and negative feedback loops appear to be important in the resident-fellow interaction, with positive interactions strengthening future ones and negative interactions creating additional barriers. There was broad agreement between fellows from different subspecialties and residents with respect to both barrier and facilitating factors. Conclusion: The resident-fellow interaction faces multiple barriers from both systems and personal domains, however many of these barriers may be modifiable. Future efforts should focus on implementing strategies to overcome these barriers. Such efforts may enhance IM residents’ rheumatologic clinical skills and potentially influence their career choice.

Each item was rated for inclusion as an EPA using a Likert scale, where 1 represented the least agreement and 9 represented the most agreement. A median score of 7–9 was considered to have highest agreement, 4–6 was intermediate agreement, and ⬍3 was rejected. An initial postal round was followed by a teleconference call, where each group member had the opportunity to discuss, defend or dispute the item. A second round was then completed to develop a final list of EPAs. Results: 35 initial EPAs were developed for scoring in each round. In the postal round, 24 and 11 EPAs met criteria for highest and intermediate agreement, respectively. In the final round, 19 EPAs met criteria for highest agreement and included concepts such as indications, limitations and proper performance of MSUS. Eight EPAs met criteria for intermediate agreement, and included concepts such as communication with referring providers and patients regarding MSUS findings. The remaining 8 EPAs were rejected for reasons such as redundancy with predicted sub-specialty EPAs, and a lack of specificity for MSUS. As a result of conferencing, several EPAs were merged into single EPAs to enhance cohesiveness and included concepts such as combining patient positioning with other strategies to obtain optimal images, and approaches to self-reflection and self-directed learning. Out of the initial 35 items, a final list of 16 EPAs was developed for MSUS in Rheumatology. Conclusion: A preliminary set of EPAs specific for the practice of MSUS in rheumatology was developed. These EPAs are the initial step in the development of assessment tools for determination of MSUS competency for a rheumatology fellow in training. Further studies will include validation of these EPAs by a larger group of educators and practitioners of MSUS. Disclosure: A. C. Cannella, None; E. Y. Kissin, None; G. S. Kaeley, None; J. B. Higgs, None; K. J. Caverzagie, None; K. D. Torralba, None.

2816


Pilot Phase Outcomes From The ACR/Carra Inter-Institutional Mentoring Program In Pediatric Rheumatology. Peter A. Nigrovic1, Eyal Muscal2, Lakshmi N. Moorthy3, Sampath Prahalad4, Marisa S. Klein-Gitelman5, Meredith P. Riebschleger6, B. Anne Eberhard7 and Rayfel Schneider8. 1 Boston Children’s Hospital, Boston, MA, 2Baylor College of Medicine, Houston, TX, 3Robert Wood Johnson Medical School-UMDNJ, New Brunswick, NJ, 4Emory Children’s Center, Atlanta, GA, 5Children’s Memorial Hospital, Chicago, IL, 6University of Michigan Health System, Ann Arbor, MI, 7Cohen Children’s Hospital Medical Center, New Hyde Park, NY, 8 Pediatric Rheumatology Collaborative Study Group (PRCSG), Cincinnati, OH. Background/Purpose: In pediatric rheumatology, the small size of many academic programs translates into limited mentoring options for early career physicians. To address this “mentorship gap,” the American College of Rheumatology (ACR) and the Childhood Arthritis and Rheumatology Research Alliance (CARRA) joined together to develop AMIGO, the ACR/ CARRA Mentoring Interest Group, that now enrolls more than 80 pediatric rheumatology fellows and junior faculty in the US and Canada. We report program evaluation data from the AMIGO pilot project, launched in November 2011 with 20 mentor-mentee dyads. Methods: Mentees and mentors participating in the AMIGO pilot phase were surveyed via online questionnaire 17 months after initial match to determine dynamics of contact and perceived benefit in domains relevant to professional success and work-life balance. Results: Of 20 initial matches, 19 were still functioning while one had dissolved due to departure of the mentee from North America. Ninety-five percent of the participants reported at least two substantial interactions. Total mentoring time was estimated at under 60 minutes by approximately half of respondents, though 20% reported more than 3 hours of interaction. Thirty-one percent of participants rated the quality of the match as good and 51% as excellent (defined as ‘excellent overlap of mentee needs and mentor experience’). All mentees considered the mentoring helpful in providing career guidance, with substantial benefit also in scholarship, and job satisfaction. Benefits reported by mentors included improvement of their mentoring skills and development of their academic portfolios. Both mentees and mentors reported improved connectedness to the wider pediatric rheumatology community. All participants favored continuation of the program (Table 1).

Table 1. Pilot survey data at 17 months after initiation of a mentoring program Mentees (n, %) Mentors (n, %) Overall (n, %) Responses (%) Benefit to mentee Career path/job search Research/scholarship Job satisfaction Work-life balance Benefit to mentors Mentoring skills Professional portfolio Enhanced connection to community Should AMIGO be continued?

18/19 (95)

19/19 (100)

37/38 (97)

18 (100) 12 (67) 13 (72) 8 (44)

13 (68) 10 (53) 12/17 (71) 12/18 (67)

31 (84) 22 (59) 25/35 (71) 20/36 (56)

NA NA 14/17 (82)

12/18 (67) 8/17 (47) 14 (75)

NA NA 28/36 (78)

18 (100)

19 (100)

37 (100)

Conclusion: The success of the pilot AMIGO program indicates that a North American inter-institutional mentoring program in pediatric rheumatology is feasible. Participants identified benefits to both mentees and mentors in multiple domains, most prominently in career guidance, a core goal of the program. Although longer-term evaluation of the full AMIGO program is needed, the success of the pilot program suggests that AMIGO could serve as a model for mentoring more broadly within the ACR and potentially in other medical subspecialties. Disclosure: P. A. Nigrovic, Baxter Healthcare, 2, Novartis Pharmaceutical Corporation, 5; E. Muscal, None; L. N. Moorthy, None; S. Prahalad, None; M. S. Klein-Gitelman, None; M. P. Riebschleger, None; B. A. Eberhard, None; R. Schneider, None.

2817 Objective Assessment Of Musculoskeletal Physical Examination Skills During Continuing Education Programs For Primary Care Providers Adds Significant Information Not Obtained Through Self-Assessment. Andrea M. Barker1, Michael J. Battistone1, J Peter Beck1, Jorie Butler2, Marissa Grotzke1, Timothy A. Huhtala1, Amy C. Cannella3, David I. Daikh4, Meika A Fang5, Antonio A. Lazzari6, Pedro Roldan7, Joan Marie Von Feldt8 and Grant W. Cannon1. 1Salt Lake City VA and University of Utah, Salt Lake City, UT, 2Salt Lake City VA, Salt Lake City, UT, 3Omaha Veterans Affairs Hospital and University of Nebraska Medical Center, Omaha, NE, 4 University of California, San Francisco, San Francisco, CA, 5VA Greater Los Angeles Healthcare System, Los Angeles, CA, 6Boston VA Medical Center, Boston, MA, 7Leesburg VA CBOC, Leesburg, FL, 8Univ of Pennsylvania/Philadelphia VAMC, Philadelphia, PA. Background/Purpose: Self-assessment is the most frequently used method in the evaluation of continuing medical education (CME) programs. This multi-institutional project was designed to examine convergent validity of learners’ self-assessment competency ratings with their scores on a 2-station objective structured clinical examination (OSCE) during a musculoskeletal (MSK) CME program. Methods: The VA Salt Lake City Health Care System developed a 3-day MSK Mini-Residency program in collaboration with six other VA facilities. This program for primary providers emphasizes physical examination and clinical management of common shoulder and knee disorders. Curriculum was introduced through focused didactics followed by small group hands-on practice sessions involving simulated patients. At the conclusion of the program, participants completed a self-reported competency assessment utilizing a 5-point Likert scale and an OSCE for both the shoulder and knee with a rater unaware of the self-reported competency. Self-assessment ratings were classified as low (⬍4.0), medium (4.0–4.5), and high (⬎4.5). OSCE ratings were classified with the same numerical boundaries based on a weighted scoring system for each component of the physical exam and then percentage of exam completed correctly (low ⬍80%, medium 80%-90%, high ⬎90%). Results: A total of 107 and 109 participants completed all components of the shoulder and knee assessments, respectively. Over half the participants were rated high on both self-assessment and OSCE for the shoulder, though correlation was weak (Pearson coefficient ⫽ 0.28). For the knee, the OSCE ratings were more variable with approximately one-third of participants in each category while the majority of the participants assessed themselves as being highly competent. Correlation was again weak (Pearson coefficient ⫽ 0.05).

S1206

Table 1. Self-assessment ratings compared to OSCE ratings for shoulder. SHOULDER Low Self-Assessment Rating

Low Med High

OSCE Rating Med High 0 5 (4.7%) 9 (8.4%) 14 (13.1%)

1 (0.9%) 8 (7.5%) 18 (16.8%) 27 (25.2%)

0 10 (9.3%) 56 (52.3%) 66 (61.7%)

1 (0.9%) 23 (21.5%) 83 (77.6%) 107

Results: Treatment with Cit-ME ameliorated the clinical score of the diseased arthritic rats. Rats treated with the Cit-ME had significantly less arthritic symptoms compared to untreated rats at day 21 (Figure 1). Amelioration of disease manifestations was associated with increased apoptosis rate of T cells.

Table 2. Self-assessment ratings compared to OSCE ratings for knee. KNEE Low Self-Assessment Rating

Low Med High

OSCE Rating Med High 0 13 (11.9%) 26 (23.9%) 39 (35.8%)

0 7 (6.4%) 28 (25.7%) 35 (32.1%)

1 (0.9%) 8 (7.3%) 26 (23.9%) 35 (32.1%)

1 (0.9%) 28 (25.7%) 80 (73.4%) 109

Conclusion: Following an intense MSK CME program, the majority of participants rated themselves as highly competent for the shoulder and knee. However, correlation of self-reported competency and OSCE rating was weak, particularly for the knee. Greater variability in the knee OSCE rating and correlation coefficient may be due to the exam maneuvers being more technically difficult to perform as they require more “hands-on” skills of the examiner. These data suggest that an OSCE should be included in evaluation of participants in CME programs to assess competency in the clinical skills being taught. Disclosure: A. M. Barker, None; M. J. Battistone, None; J. P. Beck, None; J. Butler, None; M. Grotzke, None; T. A. Huhtala, None; A. C. Cannella, None; D. I. Daikh, None; M. A. Fang, None; A. A. Lazzari, None; P. Roldan, None; J. M. Von Feldt, None; G. W. Cannon, None.

ACR Concurrent Abstract Session Rheumatoid Arthritis - Animal Models II Wednesday, October 30, 2013, 9:00 AM–10:30 AM

2818

WITHDRAWN

Figure 1. Treatment with Cit-ME suppressed Adjuvant induced arthritis (AIA) clinical score in Lewis rats

2819

Background/Purpose: Antigen-induced peripheral tolerance is a potentially efficient and specific therapeutic approach to attenuate autoimmunity. Citrullinated peptides are major targets of disease-specific autoantibodies in Rheumatoid Arthritis (RA). Currently, citrullinated peptides serve primarily as biomarkers for the diagnosis of RA by measuring titers of anti-citrullinated protein antibodies. In an attempt to develop a citrullinated peptides-based specific immunotherapy for RA, we previously showed the potential of two citrullinated peptides (citrullinated-filaggrin and citrullinated-␤-fibrinogen) to up-regulate TGF-␤ mRNA expression and concomitant expansion of regulatory T cell population. In addition, our citrullinated peptides reduced the expression of inflammatory cytokines (INF-␥, TNF-␣ and IL-17), reduced the percentage of pathogenic IL-17⫹ cells and increased the apoptosis rate of T cells following incubation with RA-derived peripheral blood mononuclear cells (PBMC). In view of the multiplicity of citrullinated target autoantigens in RA we tailored a multi-epitope citrullinated peptide (Cit-ME) derived from major prevalent citrullinated autoantigens (citrullinated filaggrin, fibrinogen, vimentin and collagen type II). The later was tested for treatment of adjuvant induced arthritis (AIA) via immune tolerance induction by attenuating the disease manifestations. Methods: Seven days following induction of AIA in Lewis female rats we administrated Cit-ME (300mg/rat) by 8 subcutaneous injections given on alternate days. Clinical scoring was performed once a week during the experiment. At the end of the experiment rats spleens were analyzed for apoptosis of CD4⫹ T cells by flow cytomety.

Conclusion: We demonstrated that citrullinated peptides induced immune tolerance in an experimental model of AIA. Disclosure: H. Amital, None; S. Gertel, None; Y. Shoenfeld, None.

2820 Disease-Regulated Local Interleukin-10 Gene Therapy Diminishes Synovitis and Articular Cartilage Damage In Experimental Arthritis. Eline A. Vermeij1, Mathijs G.A. Broeren1, Miranda B. Bennink1, Onno J. Arntz2, Inger Gjertsson3, Wim B. van den Berg1 and Fons A.J. van de Loo1. 1 Rheumatology Research and Advanced Therapeutics, Department of Rheumatology, Radboud University Nijmegen Medical Centre, Nijmegen, Netherlands, 2Radboud University Nijmegen Medical Centre, Nijmegen, Netherlands, 3Department of Rheumatology and Inflammation Research, Institute of Medicine, University of Gothenburg, Gothenburg, Sweden, Gothenburg, Sweden. Background/Purpose: Rheumatoid arthritis (RA) is a chronic destructive autoimmune disease and in most patients the disease follows an intermittent course with periods of exacerbation and remission. Today’s treatment is based on continuous immunosuppression, irrespective of the patient’s inflammatory status. An attractive alternative treatment would provide a disease-regulated therapy that offers flexible drug delivery, high during flares and low during remission. To explore this concept we expressed the anti-inflammatory cytokine interleukin (IL)-10 gene under the control of an inflammation dependent promoter in a mouse model of RA, the streptococcal cell wall arthritis (SCW). Methods: C57Bl6/N mice were injected intra-articularly in the knee joint with lentiviral vectors expressing a luciferase reporter or the therapeutic protein IL-10. The disease-regulated proximal promoters of S100a8, Cxcl1, Mmp13, Saa3, IL-1b, and TSG6 were selected from endogenous genes differentially regulated in the inflamed synovium of arthritic mice. The constitutive PGK promoter was used as a positive control. Arthritis was

S1207


Administration Of a Multi-Epitope Citrullinated Peptide Attuneuates adjuvant Induced Arthritis In Rats Via Induction Of Immune Tolerance. Howard Amital1, Smadar Gertel2 and Yehuda Shoenfeld3. 1Sheba Medical Center,Tel-Hashomer, Tel-Aviv University, Tel-hashomer, Israel, 2Sheba Medical Center, Tel-Hashomer, Israel, 3Zabludowicz Center for Autoimmune Diseases, Sheba Medical Center, Affiliated to Sackler Faculty of Medicine, Tel-Aviv University,Tel-Aviv, Tel-Aviv, Israel.

induced by injection of 25␮g SCW into the knee joint cavity 4 days after lentivirus injection. At 1, 4, and 7 days after arthritis induction, in-vivo bioluminescent imaging was performed or mice were sacrificed and knee joints were dissected for either histological analysis, or RNA isolation for qPCR analysis. Results: The 6 disease-regulated promoters all showed a different activation profile during the course of the disease. Two promoters were selected for IL-10 overexpression in the SCW model; the Saa3 promoter which showed an immediate and high upregulation at day 1 after arthritis induction, and the MMP13 promoter which showed a delayed response and peaked at day 4. For both regulated promoters, overexpression of IL-10 showed significant less synovitis at day 4 after arthritis induction (decrease from 2.2⫾0.2 to 1.6⫾0.2) and significant less cartilage proteoglycan (PG) depletion at day 4 (decrease from 2.6⫾0.2 to 1.3⫾0.2) and day 7 (decrease from 2.2⫾0.3 to 0.7⫾0.3) after arthritis induction. At day 4, IL1-Ra and SOCS3 genes were upregulated by Saa3-IL10, whereas at day 7 both Saa3-IL10 and MMP13-IL10 caused an upregulation of IL1Ra and SOCS3 gene expression. IL-1Ra is known to counteract the detrimental effects of IL-1 on cartilage damage and SOCS3 can inhibit the JAK/STAT pathway and subsequent inflammation. Therefore, these IL-10 induced changes in gene expressions can explain the diminished synovitis and PG depletion. Probably because IL-10 is expressed in the synovial tissue at day 1 of SCW arthritis without any treatment, a major therapeutic difference between the MMP13 and SAA3 promoter was not evident in this study. Conclusion: Local inflammation-dependent IL-10 gene therapy suppresses experimental arthritis and is a promising strategy in the development of novel treatments for RA. Disclosure: E. A. Vermeij, None; M. G. A. Broeren, None; M. B. Bennink, None; O. J. Arntz, None; I. Gjertsson, None; W. B. van den Berg, None; F. A. J. van de Loo, None.

2821


Evaluation Of Selective Manipulation Of The CD28 Co-Stimulation Pathway In The Rhesus Monkey Model Of Collagen-Induced Arthritis. Michel P.M. Vierboom1, Elia Breedveld1, Bert ’t Hart1, Flora Coulon2 and Bernard Vanhove3. 1Biomedical Primate Research Centre, Rijswijk, Netherlands, 2Institut National de la Sante´ et de la Recherche Me´dicale, Nantes, France, 3INSERM UMR-S 1064, Nantes, France. Background/Purpose: T-cells are important in the pathogenesis of rheumatoid arthritis (RA). T-cell activation depends on at least two signals. Next to the first signal that is provided by the binding of the T-cell receptor (TCR) to a peptide bound by an MHC molecule it requires a second signal provided by one or more co-stimulatory pathways. If only the first signal is present without co-stimulation T cells become anergic. One of the most important pathways, which is of therapeutic interest is that between CD80/86 on antigen-presenting cells and CD28 on naı¨ve T-cells. This interaction is necessary for T-cell activation. Intervention in this pathway has resulted in clinical success in the treatment of RA as demonstrated by treatment with the CD80/86 antagonist CTLA4-Ig (Abatacept). Here we present the beneficial result of treatment with a novel CD28 antagonist, which blocks CD28 instead of CD80/86, in a nonhuman primate model of inflammatory arthritis. Methods: FR104, is a monovalent and pegylated humanized Fab’ antibody fragment directed against CD28. FR104 functions as a CD28 antagonist that prevents the interaction between CD28-CD80/86 without inhibiting the interaction with CTLA-4 and PDL-1, thereby promoting immune regulation. FR104 is a primate specific antibody and safety and efficacy was tested in a collagen-induced arthritis (CIA) model in the rhesus monkey. The rhesus CIA model is an autoimmune-mediated model of polyarthritis with inflammation and erosion of joints that shares several important cellular and histopathological features with (RA). Treatment with either Vehicle (N⫽5) or FR104 (N⫽7) started at the day of induction and continued weekly until day 42 (7 administrations) Results: Treatment with FR104 prevented the development of clinical arthritis leading to a significant disease free survival compared to placebo treated animals. This was supported by unchanged production of cartilage breakdown products and minimal histological changes. This was associated with a suppressed production of CRP and IL-6. Treatment with FR104 also resulted in a robust suppression of collagen type II-specific proliferation and antibody (IgM/IgG) production and prevented the development of joint swelling and the subsequent destruction of cartilage and bone. Conclusion: FR104 was well tolerated after the multiple dosing in this nonhuman primate model of arthritis. Treatment resulted in a robust suppression of collagen specific immune responses and inhibited the development

clinical arthritis. In addition it prevented the development of neutralizing antibodies, a phenomenon that is regularly observed in the model with treatments targeting inflammatory cytokines like IL-6. The current study demonstrated the strong potential of CD28 as a therapeutic target of antagonist antibodies in a range of inflammatory disorders, including RA. Disclosure: M. P. M. Vierboom, None; E. Breedveld, None; B. ’t Hart, None; F. Coulon, None; B. Vanhove, Effimmune, 9.

2822 IKK␧ Deficiency Prolongs Neutrophil Survival Paradoxically Prolonging Inflammation In The Absence Of a Type I Interferon Response. Maripat Corr1, Christopher Chung2, Seong-Kyu Kim3, D. L Boyle2 and G. S. Firestein2. 1Univ of California-San Diego, La Jolla, CA, 2UCSD School of Medicine, La Jolla, CA, 3Catholic University of Daegu School of Medicine, Daegu, South Korea. Background/Purpose: Deficiency of the IKK-related kinases IKK␧ synergizes with the anti-inflammatory effects of IFN␤ in a murine model of arthritis, and this effect is mediated by IL-1Ra. To further explore the relationship between IFN signaling and IKK␧, IKK␧ null mice were intercrossed with IL1Ra or type I interferon receptor deficient mice and evaluated in the passive K/BxN model of arthritis. Methods: IKK␧ null, Il1rn (which encodes IL-1Ra) null, IL1Ra/IKK␧ null, IFNAR1 null, IFNAR1/IKK␧ null and wild type mice were injected with 150 ul of pooled K/BxN sera on day 0. Clinical response and histologic scores were assessed. Gene expression was measured by quantitative PCR. Mice were injected intraperitoneally with thioglycollate. The peritoneal lavage was evaluated for cell counts. The number and viability of neutrophils was determined by flow cytometry. Gene expression and protein production were tested by qPCR and western blot respectively. Results: IKK␧ null mice had a modestly attenuated course of arthritis (area under the curve [AUC] for ankle diameter⫽5.1 vs. 5.9 for wild type controls, P⬍0.05). In contrast IFNAR1 null mice (AUC 6.3 vs. 5.9, P⬍0.05) and Il1rn⫺/⫺ mice (AUC 11.4 vs. 5.9 P⬍0.01) had more severe serum transfer arthritis than wild type mice. The IL1Ra/IKK␧ null had arthritis similar (AUC 12.1) to the Il1rn⫺/⫺ mice (AUC 11.4). Unexpectedly, IFNAR1/IKK␧ null mice (AUC 7.9, P⬍0.05) had even greater disease severity than the IFNAR1 null (AUC 6.3) mice. The ankle histology scores in each strain correlated well with the clinical course. Of interest, quantitative PCR of arthritis paws showed a significant increase in the neutrophil chemotactic protein GCP2 in the IFNAR1 (98⫹10) and IFNAR1/IKK␧ (81⫹11) null compared to the WT (19⫹1) mice (p⬍0.05). There was no difference in IL-6, IL-1␤, IL-1a or IFNb in the IL1Ra/IKK␧ null or IFNAR1/IKK␧ null mice compared to the individual knockout controls. Concordant with higher GCP2 expression mice deficient in IFN signaling recruited higher numbers of neutrophils into the peritoneum after thioglycollate challenge: 4.6⫹1 (WT), 6.1⫹2 (IKK␧), 15.8⫹4.8 (IFNAR1) and 13.5⫹2.8 (IFNAR1/IKK␧) ⫻106 neutrophils (p⬍0.05). The neutrophils recovered from the IFNAR1/IKK␧ mice, however, expressed higher mRNA and protein levels of the anti-apoptotic protein Mcl1 than those of the individual knockout controls, suggesting that the deficient neutrophils had defective apoptosis. Conclusion: IKK␧ deficiency did not offset increased severity in IL-1Ra deficiency supporting a direct role of IL-1Ra in reducing arthritis. Surprisingly, IKK␧ deficiency increased disease severity in the IFNAR null mice, most likely due to increased neutrophil recruitment and prolonged survival. Disclosure: M. Corr, NIAMS-NIH, 2, UCSD, 3; C. Chung, None; S. K. Kim, None; D. L. Boyle, NIH, 2; G. S. Firestein, NIH, 2.

2823 FLIP Deficiency In Dendritic Cells Promotes Spontaneous Inflammatory and Erosive Arthritis. Qi Quan Huang1, Harris R. Perlman1, Robert Birkett2, Renee E. Koessler1, Syamal K. Datta3 and Richard M. Pope4. 1 Northwestern University, Chicago, IL, 2Northwestern University Medical School, Chicago, IL, 3Northwestern University FSM, Chicago, IL, 4Northwestern Univ Med School, Chicago, IL. Background/Purpose: FLIP is an anti-apoptotic protein induced by chronic inflammation. In this study a mouse line with FLIP deleted in CD11c⫹ dendritic cells (DCs), that spontaneously develops arthritis, was employed to define the potential role of tolerance induction and adaptive immune.

S1208

Methods: Mice with FLIP deficient in DC (Flip f/f, CD11c cre/⫹ or KO) were generated by crossing Flip f/f with CD11ccre transgenic mice. Immune cells in lymphoid organs were analyzed by Flow-cytometry employing multifluorochrome-conjugated antibodies. Arthritis was evaluated by joint swelling and/or deformity. Pathology was analyzed by histologic HE staining. The levels of IgG and cytokines were determined by ELISA and serum rheumatoid factors (RF) by rabbit IgG-ELISA. In vivo antigen presentation was determined by T cell proliferation to OVA, monitored by CFSE labeled CD4⫹ T cells from OTII mice. All data are analyzed comparing with gender matched littermate controls. Results: The KO mice spontaneously developed progressive, erosive arthritis, starting at 6 weeks of age and reaching ⱖ 80% incidence at ⱖ 4 months. For mice at ⱖ 5 months, histological examination of knees, ankles, and paws demonstrated increased joint inflammation with neutrophils, macrophages and lymphocytes, bone erosion, pannus formation and cartilage destruction (p⬍ 0.05-0.001). Further, increased TNF␣, IL-1␤, IL-6, IL-10, CXCL5, INF-g and RANKL, but reduced OPG, were present in ankle homogenates (p⬍ 0.001). Old KO mice expressed increased circulating INF-␥, IgG1 and RF (p⬍ 0.05-0.001). All 5 month KO mice developed lymphadenopathy with increased B cells and plasmablasts. Examination of 4 week old (young) KO mice, prior to the onset of arthritis, demonstrated reduction in the size and number of cells in the thymus. The subset of DCs expressing high CD11c and MHCII and were CD8⫹ were significantly reduced in the thymus, as well as the spleen, and lymph nodes (LN) young mice (p⬍ 0.05-0.001). Further, single positive CD4⫹ or CD8⫹ T cells, the ratio of CD4:CD8 cells and CD4⫹/CD25⫹/Foxp3⫹ Treg cells were reduced (p⬍ 0.05-0.001) in the thymus of young KO mice. CD3⫹,CD4⫹ and CD3⫹,CD8⫹ cells were reduced and B cells were increased in the LNs of young mice. In the LNs of old KO mice CD11c DC and CD3⫹,CD8⫹ cells normalized, while CD3⫹,CD4⫹ and B cells were increased. Finally, the young Flip f/f, CD11c cre/⫹ mice demonstrated reduced in vivo antigen presentation properties, identified by reduced CD4⫹ T proliferation in response to OVA. Conclusion: These observations suggest that reduction of FLIP in DC results in reduced positive and negative selection, reduced Tregs and increased B cells, IFN-g, and autoantibodies. DCs expressing high levels of CD11c and MHCII are necessary for normal tolerance development and restraining the occurrence of inflammatory arthritis. Disclosure: Q. Q. Huang, None; H. R. Perlman, None; R. Birkett, None; R. E. Koessler, None; S. K. Datta, None; R. M. Pope, None.

(CRP), erythrocyte sedimentation rate (ESR), interleukin-6 (IL-6), tumor necrosis factor-a (TNF-a), and cystatin-C, were measured in RA. Subclinical atherosclerosis was assessed using electron beam computer tomography, and arterial stiffness using AIX. We compared traditional risk factors, UACR and measures of subclinical CVD between RA and controls using Wilcoxon rank sum and Fisher’s exact test. The association between MA and CAC and AIX was adjusted for age, sex and race using linear or logistic multivariate analyses as appropriate. Results: Patients with RA had a higher UACR [median (IQR): 7.6mg/g (4.0–15.5) than controls (5.6 (3.3–9.0), p⫽0.02). The presence of MA was significantly associated with HTN in RA (p⫽0.01). No significant association was observed with age, gender, smoking, HOMA-IR, or metabolic syndrome. In RA, but not controls, there was a significant association between AIX and log transformed UACR, ␤ coefficient of 1.9 (95% CI 0.4–3.4), p⫽0.01. The association remained significant after adjusting for age, sex, and race. CAC was not significantly associated with UACR in RA or controls [Table]. UACR was significantly associated with higher levels of VCAM-1 (␳⫽0.2, p⫽0.01) and lower levels of IL-10 (␳⫽⫺0.2, p⫽0.02) in RA. Table. Association between UACR and subclinical CVD in RA and controls: UACR

AIX

RA Controls

CAC

RA Control

Unadjusted b coef. (95% CI) P value

Adjusted for age, sex, and race b coef. (95% CI) P value

1.9 (0.4–3.4) ⫺1.3 (⫺3.6–1.0) R (95% CI) 0.9 (0.7–1.2) 0.6 (0.4–1.1)

1.5 (0.1–2.8) ⫺0.9 (⫺2.8–1.0) OR (95% CI) 0.7 (0.5–1.1) 0.7 (0.4–1.3)

0.01 0.26 P value 0.64 0.11

0.03 0.34 P value 0.13 0.31

Conclusion: Urinary microalbumin is increased in patients with RA compared to controls and is associated with increased arterial stiffness as measured by AIX in RA, independent of other CVD risk factors. MA is not associated with CAC. In RA, urinary microalbumin is associated with higher levels of VCAM-1, which mediates the adhesion of inflammatory cells to vascular endothelium, and with lower levels of IL-10, an anti-inflammatory cytokine. Disclosure: K. Becetti, None; A. Oeser, None; J. F. Solus, None; P. Raggi, None; C. M. Stein, NIH, 2; C. P. Chung, Vanderbilt Physician Development Award, 2.

2825


The Influence Of Early Menopause On Cardiovascular Risk In Women With Rheumatoid Arthritis. Emily Pfeifer, Cynthia S. Crowson, Shreyasee Amin, Sherine E. Gabriel and Eric L. Matteson. Mayo Clinic, Rochester, MN.

2824 Urinary Microalbumin Is Associated With Arterial Stiffness In Patients With Rheumatoid Arthritis. Karima Becetti1, Annette Oeser1, Joseph F. Solus2, Paolo Raggi3, C. Michael Stein2 and Cecilia P. Chung1. 1Vanderbilt University, Nashville, TN, 2Vanderbilt Medical Center, Nashville, TN, 3 University of Alberta, Edmonton, AB. Background/Purpose: Rheumatoid arthritis (RA) is associated with an increased risk for cardiovascular disease (CVD). Urinary microalbumin is a risk factor for CVD in the general population, but its association with CVD in RA is less well defined. Thus, we examined the association between urinary microalbumin and CVD, using coronary artery calcium (CAC) and augmentation index (AIX), measures of coronary atherosclerosis and vascular stiffness, respectively. Methods: In a cross-sectional study, we evaluated 136 patients with RA and 79 controls with no diabetes or clinical history of CVD. MA was defined as urine albumin to creatinine ratio (UACR) ⬎ 30mg/g in a spot urine. Traditional CVD risk factors such as age, gender, smoking, family history of CVD, body mass index (BMI), hypertension, and lipids were recorded, and insulin resistance (defined using the homeostasis model assessment), metabolic syndrome and Framingham risk scores were calculated. Disease duration, DAS 28 score and inflammatory markers, including vascular cell adhesion molecule-1 (VCAM-1), interleukin-10 (IL-10), C-reactive protein

Background/Purpose: Lifetime exposure to female sex hormones may play a role in the development and severity of rheumatoid arthritis (RA). These same hormones have also been found to play a role in the development of cardiovascular disease (CVD) in women. Since RA is associated with an increased risk of CVD, the purpose of this study was to determine if early menopause, representing a surrogate for lower lifetime exposure to female sex hormones, affects the risk of developing CVD in women with RA. Methods: A population-based inception cohort of 600 women with RA who first fulfilled 1987 ACR criteria for RA between 1955 and 2007 was assembled and followed until death, migration or 12/31/2008. Age at menarche, gravidity, parity, age at menopause, and duration, if any, of hormone replacement therapy (HRT), along with occurrence of CVD (including coronary artery disease, heart failure, cerebrovascular disease and peripheral vascular disease) was ascertained by review of medical records. Cox proportional hazard models were used to compare women within this cohort who underwent early menopause, defined as natural or artificial menopause at age ⱕ 45 years, to those within the cohort who did not undergo early menopause. Results: This study included 600 women with RA age ⱖ 45 years at diagnosis, of whom 199 experienced early menopause (mean age ⫾ SD: 64.3 ⫾ 12.0 years; 67.x % rheumatoid factor (RF) positive) and 401 did not (mean age: 62.8 ⫾ 11.4 years; 64.9% RF positive). The mean age at menopause in those who experienced early menopause was 40.9 ⫾ 5.0 years,

S1209


Rheumatoid Arthritis - Clinical Aspects VI: Cardiovascular Disease in Rheumatoid Arthritis Wednesday, October 30, 2013, 9:00 AM–10:30 AM

while in those who did not the mean age at menopause was 50.7 ⫾ 2.8 years. Among women without prior CVD, 73 with early menopause and 132 without early menopause developed CVD during a mean follow-up of 11.8 years. Women who underwent early menopause were at significantly higher risk for developing CVD when compared to women who did not experience early menopause (hazard ratio (HR): 1.41; 95% CI: 1.05–1.91, adjusted for age, calendar year of RA diagnosis and cardiovascular risk factors including smoking status, body mass index at RA diagnosis, diabetes mellitus and hypertension). This difference remained significant when age of menopause was defined as the end of HRT for women with artificial menopause who started HRT therapy at the time of artificial menopause (HR: 1.40; 95% CI: 1.03–1.88). CVD risk was increased in women with higher gravidity and parity (HR 1.06 per pregnancy increase; 95% CI 1.004–1.12 and HR 1.07 per 1 birth increase, 95% CI 1.01–1.14 respectively). There was a strong non-linear relationship between gravidity and parity and cardiovascular outcomes, whereby the increase risk of CVD was detected only at very high values (⬎7) of gravidity and parity. No associations were found between the development of CVD and cause of early menopause (natural vs. artificial), the use of HRT or the length of time on HRT. Conclusion: The risk of CVD in women with RA was significantly higher in those who experience early menopause, and like other known risk factors should increase clinician concern for development of CVD in these patients. Further investigation is needed to determine the role that female sex hormones play in the development of CVD in women with RA. Disclosure: E. Pfeifer, None; C. S. Crowson, None; S. Amin, None; S. E. Gabriel, None; E. L. Matteson, None.

2826


Performance Of Five Current Risk Algorithms In Predicting Cardiovascular Events In Rheumatoid Arthritis Patients. Elke.E.A. Arts1, Calin Popa1, Alfons A. den Broeder2, Anne G. Semb3, Tracey Toms4, George Kitas4, Piet L.C.M. van Riel1 and Jaap Fransen1. 1Radboud University Nijmegen Medical Centre, Nijmegen, Netherlands, 2Sint Maartenskliniek, Nijmegen, Netherlands, 3Diakonhjemmet Hospital, Oslo, Norway, 4Dudley Group of Hospitals NHS Trust, Russells Hall Hospital, Dudley, United Kingdom. Background/Purpose: Cardiovascular risk in rheumatoid arthritis (RA) is increased. The cardiovascular (CV) risk algorithms used in the general population may underestimate the risk of cardiovascular disease (CVD) in the RA population1. This study was undertaken to assess the predictive ability of 5 established CV risk models for the 10-year risk of fatal and non-fatal CVD in European patients with Rheumatoid Arthritis Methods: Included were the Framingham risk score (FRS), the Systematic Coronary Risk Evaluation score (SCORE), the modified SCORE according to the EULAR recommendations for CV risk2 (M-SCORE), the Reynolds risk score (RRS) and the QRisk II risk score. Prospectively collected data from the Nijmegen early RA inception cohort were used. Patients with CVD prior to enrollment were excluded. Discriminatory ability for CV risk prediction was estimated by the area under the receiver operating characteristic (ROC) curve. Calibration was assessed by comparing the observed versus expected number of events using Hosmer-Lemeshov tests and calibration plots. Sensitivity and specificity were calculated for the cut-off values of 10% and 20% predicted risk Results: In total, n⫽1050 patients were included. During follow-up, 145 patients developed a CV event. The mean⫾SD age was 54⫾13.8 years, 66% were female and 74% were rheumatoid factor positive. Areas under the ROC curve were 0.75 (95% CI; 0.71–0.80), 0.75 (95% CI;0.71–0.80), 0.76 (95% CI; 0.73–0.80), 0.75 (95% CI; 0.71–0.79) and 0.77(95% CI; 0.73–0.80) for the SCORE, M-SCORE, FRS, Reynolds and QRisk II respectively, indicating moderate to good discrimination between patients with and without a CV event. All five models underestimated CV risk at low and middle observed risk levels, and overestimated CV risk at high observed risk levels (Fig. 1). For the 10% and 20% cut-off values used as indicators for CV preventive treatment, sensitivity ranged from 66–87% and 43–67% respectively and specificity ranged from 44–57% and 67–81% respectively. Depending on the model, 13% to 35% of observed CV events occurred in RA patients who were classified as low risk (⬍10%) for CVD.

Figure 2. Observed (closed bars) versus predicted (open bars) CV event rate (%) in deciles of predicted risk, for the SCORE (A), M-SCORE (B), FRS (C), Reynolds (D) and QRisk II (E) risk algorithms.

Conclusion: Established risk models generally underestimate CV risk in RA patients. There is an unmet need for development of a RA-specific CV risk model. References: 1. Crowson CS, Matteson EL, Roger VL, et al. Usefulness of Risk Scores to Estimate the Risk of Cardiovascular Disease in Patients With Rheumatoid Arthritis. Am J Cardiol. 2012;10:420–24. 2. Peters MJ, Symmons DP, McCarey D, et al. EULAR evidence-based recommendations for cardiovascular risk management in patients with rheumatoid arthritis and other forms of inflammatory arthritis. Ann Rheum Dis. 2010;69:325–31. Disclosure: E. E. A. Arts, None; C. Popa, None; A. A. den Broeder, None; A. G. Semb, Merck/Schering-Plough, Abbott, BMS, Pfizer/Wyeth, Genentech and Roche, 5; T. Toms, None; G. Kitas, None; P. L. C. M. van Riel, None; J. Fransen, None.

2827 Effect Of Biologic Agents On Lipids and Cardiovascular Risk In Rheumatoid Arthritis Patients. Dimitrios A. Pappas1, Ani John2, Jeffrey R. Curtis3, George W. Reed4, Jeffrey D. Greenberg5, Ashwini Shewade2, Daniel H. Solomon6, Joel M. Kremer7 and Tanya Sommers4. 1Columbia University, New York, NY, 2Genentech Inc., South San Francisco, CA, 3University of Alabama at Birmingham, Division of Clinical Immunology and Rheumatology, Birmingham, AL, 4CORRONA, Inc., Southborough, MA, 5New York Hospital for Joint Diseases, New York, NY, 6Harvard Medical School, Brigham and Women’s Hospital, Division of Rheumatology, Division of Pharmacoepidemiology, Boston, MA, 7Albany Medical College and The Center for Rheumatology, Albany, NY. Background/Purpose: The risk for cardiovascular disease (CVD) is increased in patients (pts) with RA. The interplay between traditional CV risk factors and inflammatory burden may be responsible for this increased CVD risk. Therefore, CVD risk scores incorporating measures of inflammation such as the Reynolds risk score (RRS) may be more appropriate than the Framingham risk score or total cholesterol/HDL ratio to predict CVD risk in pts with rheumatoid arthritis (RA), though not yet validated for RA pts

S1210

(Ridker, 2007, 2008). This study investigated the effects of biologics on lipids, CRP and CVD risk using RRS. Methods: Pts with moderate disease activity (CDAI ⬎10) initiating a biologic DMARD participated in a comparative effectiveness prospective study (CERTAIN) nested within CORRONA. Characteristics, including LDL, HDL, total cholesterol (TC) (desirable ⬍ 200) and triglycerides (TG), high sensitivity C-reactive protein (hsCRP) and RRS in pts initiating a TNF-␣ inhibitor (TNFi) or non-TNFi (rituximab [RTX], abatacept [ABT] or tocilizumab [TCZ]), were measured at baseline and at 3 months. Linear, mixed effect regression models were fit to compare the effect of biologics on lipids, CRP and RRS outcomes after 3 months of therapy. In all models adjustment took place for the baseline levels of the outcome, biologically plausible covariates and other characteristics that were imbalanced among the four groups of medications. Results: 779 initiations of a biologic were analyzed: 435 (55.8%) TNFi, 45 (5.8%) RTX, 161 (20.7%) ABA and 138 (17.7%) TCZ. Overall baseline characteristics: 75.9% women, 86.3% Caucasian, 65.5% seropositive. Mean ⫾ SD age was 55.9 ⫾ 13.3 years, RA disease duration 8.8 ⫾ 9.5 years; CDAI 28.6 ⫾ 12.7. 35.9% of pts were biologic naı¨ve and 24.6% received anti-hyperlipidemic therapy. History of prior CVD was present in 8.0% of pts; 7.8% had diabetes mellitus and 41% were obese (BMI ⬎ 30). The adjusted effect of biologic agents on lipid levels, CRP and RRS after 3 months of treatment are summarized in Figure 1. ABT and RTX had a similar effect on lipid levels compared to TNFi after 3 months while TCZ increased TC, LDL and TG levels compared with TNFi after 3 months. Significantly different changes in CRP were observed across different biologics at 3 months. For CVD risk, as measured by RRS, ABT, RTX and TCZ had a similar effect compared with TNFi.

Background/Purpose: Cardiovascular (CV) risk in rheumatoid arthritis (RA) is increased. The CV risk algorithms for the general population may underestimate the risk of cardiovascular disease (CVD) in RA patients1. The objective of this study was to test the performance of the original and updated SCORE risk algorithm for the prediction of the 10-year risk of CVD in RA patients. Methods: Data from the Nijmegen early RA inception cohort (n⫽1017) were used. The systematic coronary risk evaluation algorithm (SCORE) recalibrated for the Dutch population was used as a basis for the new model. This algorithm was recalibrated in this cohort (SCORE recalibrated) by first adjusting the weights of only the risk factors that were included in the original SCORE (age, gender, smoking, systolic blood pressure and total cholesterol: HDL ratio) and in the following this step the original SCORE was updated (SCORE updated) by including other risk factors as assessed in univariate and multivariate Cox proportional hazard regression analysis (significant at p-value ⬍0.2). Predictive performance was assessed by the area under the receiver operating characteristic (ROC) curve and by comparing the observed versus expected number of CV events using Hosmer-Lemeshov tests and calibration plots. Results: During follow-up, 144 patients had a CV event. In addition to the risk factors of the original SCORE, the updated model included BMI, diabetes, hypertension and high disease activity (DAS28⬎5.1) at baseline. Areas under the ROC curve were 0.72 (95% CI; 0.68–0.77, 0.71 (0.67–0.75), 0.75 (0.71–0.79), for the original, the recalibrated and the updated SCORE respectively (figure 1A), indicating moderate to good discrimination. The agreement between the observed and the predicted CV events was poor for the original score which underestimated CV risk at low and middle observed risk levels, which was confirmed in the Hosmer and Lemeshow (H-L) test that indicated poor model fit (p⬍ 0.001). The updated SCORE still showed some underestimation (figure 1B) in the low-middle risk groups (ⱕ 20% CV risk) and overestimation in the highest risk groups, but overall model fit and predictive accuracy improved (H-L test p⫽1.0).

Figure 1. Panel A: ROC-curves for the different SCORE algorithms. Panel B: Calibration plot of the updated SCORE. Depicted are the observed events versus predicted events (observed versus predicted probabilities).

Ridker P, et al. JAMA 2007;297:611; Ridker P, et al. Circulation 2008;118;22 Disclosure: D. A. Pappas, CORRONA, 3, Novartis, 5; A. John, Genentech, 3; J. R. Curtis, Roche/Genentech, UCB, Janssen, CORRONA, Amgen, Pfizer, BMS, Crescendo, AbbVie, 2, Roche/Genentech, UCB, Janssen, CORRONA, Amgen, Pfizer, BMS, Crescendo, AbbVie, 5; G. W. Reed, CORRONA, 3; J. D. Greenberg, CORRONA, 1, Astra Zeneca, Corrona, Novartis, Pfizer, 5; A. Shewade, Genentech, 3; D. H. Solomon, Amgen, Lilly, CORRONA, 2, Pfizer, Novartis, Lilly, BMS, 9, UpToDate, 7; J. M. Kremer, CORRONA, 1, CORRONA, 3; T. Sommers, CORRONA, 3.

References: 1. Crowson CS, Matteson EL, Roger VL, Therneau TM, Gabriel SE. Usefulness of Risk Scores to Estimate the Risk of Cardiovascular Disease in Patients With Rheumatoid Arthritis. Am J Cardiol. 2012.3 Disclosure: E. E. A. Arts, None; C. Popa, None; A. A. den Broeder, None; A. G. Semb, Merck/Schering-Plough, Abbott, BMS, Pfizer/Wyeth, Genentech and Roche, 5; T. Toms, None; G. Kitas, None; P. L. C. M. van Riel, None; J. Fransen, None.

2829

2828

Statin Adherence and Risk Of Mortality In Patients With Rheumatoid Arthritis: A Population-Based Study. Mary De Vera1, Michal Abrahamowicz2 and Diane Lacaille3. 1Arthritis Research Centre of Canada, Richmond, BC, 2McGill University, Montreal, QC, 3Arthritis Research Centre of Canada, University of British Columbia, Richmond, BC.

The Performance Of The Original and An Updated Cardiovascular Risk Algorithm (SCORE) In Patients With Rheumatoid Arthritis. Elke.E.A. Arts1, Calin Popa1, Alfons A. den Broeder2, Anne Grete Semb3, Tracey Toms4, George Kitas4, Piet L.C.M. van Riel1 and Jaap Fransen1. 1Radboud University Nijmegen Medical Centre, Nijmegen, Netherlands, 2Sint Maartenskliniek, Nijmegen, Netherlands, 3Diakonhjemmet Hospital, Oslo, Norway, 4 Dudley Group of Hospitals NHS Trust, Russells Hall Hospital, Dudley, United Kingdom.

Background/Purpose: Poor adherence with statin therapy is associated with increased mortality in the general population, but no corresponding data are available among patients with rheumatoid arthritis (RA). Since cardiovascular diseases (CVD) are the primary cause of excess mortality in RA, RA-specific data are highly relevant to RA clinical care. We evaluated the impact of statin adherence on risk of mortality in individuals with RA. Methods: We conducted a cohort study of all incident statin users in a population-based cohort of RA identified from physician billing data, fol-

S1211


Conclusion: Our study showed that there was a differential effect on lipid levels and CRP when compared amongst biologics with different mechanisms of action. However, these changes did not translate to an increased CVD risk as measured by RRS. In order to better understand CV risk in RA, long term outcome studies are needed to evaluate the interplay among lipids, inflammation and other CVD risk factors.

Conclusion: CV risk predictions according to the updated SCORE risk algorithm were more accurate compared to the original SCORE risk algorithm, particularly in the low and intermediate CV risk groups. The next step will be to externally validate these results in a different cohort.

lowed from May 1996 to March 2010 using administrative health data. Each individual’s observational time was divided into 90-day periods from the date of the first statin prescription until date of death or end of follow-up. For each period, we calculated the proportion days covered (PDC) of statin use and created three adherence categories: high adherence (PDC ⱖ 0.80), poor adherence (0 ⬍ PDC ⬍ 0.80), and non-use (PDC ⫽ 0). Outcomes evaluated include deaths due to all causes and due to any CVD. We used Cox’s proportional hazards analyses, modeling risk of death, with statin adherence as a time-dependent variable representing adherence in the current 90 day interval. We considered the following covariates: age, gender, comorbidities (e.g. acute myocardial infarction, cerebrovascular accidents, infections), use of diabetes, hypertension, and congestive heart failure medications, and use of medications known to influence cardiac risk (e.g. hormone replacement therapy), measured at baseline. Markers of RA severity and RA medication use that could influence mortality risk (DMARDs, glucocorticosteroids, methotrexate and rate of RA-related medical visits) were also included as time-dependent covariates. Covariates were included in the final model if they were significant predictors of death or if they had a confounding effect on the association between statin adherence and death. Results: The cohort of RA patients with incident statin use included 6,525 individuals with 36,097 person-years of follow-up (60% females; mean age 67 years). We documented 1,193 deaths overall with 386 due to CVD. Adjusted HRs for statin adherence status in the current period and mortality risk are summarized (Table). Statin Adherence Status High adherence (PDCⱖ0.80) ref Poor adherence (0⬍PDC⬍0.80) Non-use (PDC⫽0)

CVD Mortality aHR (95% CI)

All Cause Mortality aHR (95% CI)

1.00 1.70 (1.27, 2.28) 1.54 (1.21, 1.95)

1.00 1.69 (1.42, 1.99) 1.83 (1.60, 2.08)

block which progressed 2 days later to 3rd degree. The neonate was not paced until 4.5 months of age. Another pregnancy resulted in a secondary outcome of a prolonged fetal PR interval ⬎ 150 msec at 26 weeks for which the mother was treated with 3 days of 4mg dexamethasone. The PR interval normalized and the EKG at birth was normal. In the one ongoing pregnancy, all echos through 37 weeks were normal (notable since the most vulnerable period for complete block is 18–24 weeks). Overall, the rate of recurrence of cardiac NL was 5.3%. Figure 1 reveals that therapeutic HCQ levels can be achieved by the mid second trimester in mothers who were not previously on HCQ prior to conception. Figure 2 substantiates cord blood levels of HCQ.

Conclusion: These population-based data indicate that RA patients who are poorly adherent with statins have a higher risk of death from CVD and from all causes. Findings emphasize the importance of discussing adherence with statin therapy during health care professional encounters with RA patients. Disclosure: M. De Vera, None; M. Abrahamowicz, None; D. Lacaille, None.

ACR Concurrent Abstract Session Systemic Lupus Erythematosus Clinical Aspects: Pregnancy Wednesday, October 30, 2013, 9:00 AM–10:30 AM


2830 First Stage Of a Simon’s Two-Stage Optimal Approach Supports Placental Transfer Of Hydroxychloroquine and a Reduced Recurrence Rate Of The Cardiac Manifestations Of Neonatal Lupus. Peter M. Izmirly1, Nathalie Costedoat-Chalumeau2, Amit Saxena1, Amanda Zink1, Zoey Smith1, Deborah Friedman3 and Jill P. Buyon1. 1New York University School of Medicine, New York, NY, 2Hopital Cochin, Paris, France, 3New York Medical College, Valhalla, NY. Background/Purpose: A previous case control study suggested that hydroxychloroquine (HCQ) might prevent the development of cardiac Neonatal Lupus (cardiac NL) in anti-SSA/Ro antibody (ab) exposed fetuses of SLE mothers. In a subsequent study the use of HCQ reduced the nearly 10 fold increased recurrence rate of cardiac NL in an international cohort of anti-SSA/Ro ab women. Based on these encouraging data, an open label prospective study was initiated to evaluate whether HCQ reduces the recurrence rate of cardiac NL Methods: A Phase II trial using a Simon’s two-stage optimal approach was employed to allow for early stopping due to absence of treatment efficacy. In this first stage of the study, 19 anti-SSA/Ro ab pregnant patients with a previous child with cardiac NL (1st degree block excluded) were enrolled and if 3 or more children with cardiac NL were born, the study would be terminated for inefficacy. The protocol called for initiation of HCQ by 10 weeks gestation. Serial echocardiograms (echos) to evaluate for the development of the primary (2nd or 3rd degree block) and secondary (1st degree block) outcomes were performed. Maternal and cord blood HCQ levels were evaluated to assess treatment adherence, pathobiology and efficacy. Results: Eighteen pregnancies in 17 women have been completed. Seventeen babies (including a set of twins) had normal serial fetal echos and normal EKGs at birth. One pregnancy resulted in a primary outcome of 3rd degree block. In this pregnancy, the echo at 20 weeks revealed 2nd degree

Conclusion: These prospective data affirm that HCQ may prevent the recurrence of cardiac NL. The feasibility of the strategy using the Simon’s two-stage design was substantiated. Moreover, the study unequivocally demonstrates that measurable HCQ levels can be reached prior to the vulnerable period and are present in the cord blood, reflective of placental transfer. Disclosure: P. M. Izmirly, None; N. Costedoat-Chalumeau, None; A. Saxena, None; A. Zink, None; Z. Smith, None; D. Friedman, None; J. P. Buyon, NIH, 2.

2831 Increased Risk Of Autism Spectrum Disorders In Children Born To Women With SLE: Preliminary Data From The OSLER Cohort. Evelyne Vinet1, Susan Scott2, Christian A. Pineau1, Lawrence Joseph3, Ann E. Clarke1, Eric Fombonne4, Robert W. Platt3 and Sasha Bernatsky1. 1 McGill University Health Center, Montreal, QC, 2McGill University Health Centre, Montreal, QC, 3McGill University, Montreal, QC, 4Oregon Health and Sciences University, Portland, OR. Background/Purpose: Recent experimental data suggest in utero exposure to maternal antibodies and cytokines as important risk factors for autism spectrum disorders (ASD). Interestingly, women with SLE display autoantibodies (e.g. anti-N-methyl-D-aspartate receptor and antiphospholipid antibodies) and cytokines (e.g. interleukin-6), which, in animal models, alter fetal

S1212

brain development and induce behavioural anomalies in offspring. To date, no one has specifically assessed the risk of ASD in children of SLE mothers. Using the “Offspring of Systemic Lupus Erythematosus mothers Registry (OSLER)”, we aimed to determine if children born to SLE mothers have an increased risk of ASD compared to children born to mothers without SLE. Methods: OSLER is a large population-based cohort, which includes all women who had ⱖ1 hospitalization for delivery after SLE diagnosis, identified through Quebec’s universal healthcare databases (1989–2009). OSLER also includes a randomly selected control group of women, matched at least 4:1 for age and year of delivery, who did not have a diagnosis of SLE prior to or at the time of delivery. We identified children born live to SLE mothers and their matched controls, and ascertained ASD based on ⱖ1 hospitalization or physician visit with a relevant diagnostic code, through to end of database follow-up. We performed multivariate analyses to adjust for maternal demographics, sex and birth order of child, and obstetrical complications. In a subsample analysis of children with maternal public drug coverage throughout pregnancy, we further assessed relevant in utero medication exposures. Results: 509 women with SLE had 719 children, while 5824 matched controls had 8493 children. Mean maternal age and follow-up were respectively 30.3 [standard deviation (SD) 5.0] and 9.1 (SD 5.8) years. Children born to women with SLE had more records of ASD diagnoses compared to controls [1.4% (95% CI 0.8, 2.5) vs 0.6% ( 95%CI 0.5, 0.8)]. Mean age at ASD diagnosis was slightly younger in offspring of SLE mothers (3.8 years, 95%CI 1.8, 5.8) as opposed to controls (5.7 years, 95%CI 4.9, 6.5). In multivariate analyses (Table 1), children born to women with SLE had substantially increased risk of ASD versus controls (HR 2.31, 95%CI 1.03, 5.16). In the subsample of children with maternal drug coverage (n⫽1925), in utero medication exposures were rare in the 18 ASD cases: none were exposed to antimalarials, antidepressants, or immunosuppressants, while only one case born to a SLE mother and another born to a control mother were respectively exposed to corticosteroids and anticonvulsants.

reduces the risk of cardiac NLE through disruption of toll-like receptors signalling. The effect of HCQ on the risk of extra-cardiac NLE has not been specifically studied. The aim of this study was to assess if maternal intake of HCQ or chloroquine throughout pregnancy reduces the risk of NLE in the offspring. We hypothesized that these drugs would confer protection against NLE. Methods: A case-control study was performed on a large single-center cohort of NLE and unaffected children, on whom prospective data has been collected since 1984. Inclusion criteria were: (1) first child born from a mother positive for anti-Ro and/or anti-La antibodies with a diagnosis of either cutaneous lupus, systemic lupus erythematosus, Sjogren’s syndrome, dermatomyositis or rheumatoid arthritis and, (2) assessment of the child in the NLE clinic at least once in his first 6 months of life. Descriptive statistics and logistic regressions were performed. Results: The study population consisted of 220 children, of whom 62 were exposed to HCQ or chloroquine throughout gestation (Table). NLE was diagnosed in 98 patients; 12 had cardiac NLE (11 congenital heart block and 1 cardiomyopathy). Neutropenia (n⫽41) was the most frequent extra-cardiac NLE feature followed by hepatitis (n⫽36), skin involvement (n⫽32), thrombocytopenia (n⫽4) and extraventricular obstructive hydrocephalus (n⫽4). One hundred and twelve children did not develop NLE. Ten children were classified as having no cardiac NLE involvement but could not be diagnosed as true unaffected children as one or more blood test components were missing. No statistically significant protective effect of HCQ or chloroquine exposure was found on the risk of NLE (OR 0.79; p⫽0.45). Similar results were found when extra-cardiac NLE cases were analyzed separately (OR 0.85; p⫽0.60). Only 1 of 62 children exposed to HCQ or chloroquine developed cardiac-NLE compared to 11 of 158 unexposed children (OR 0.22; p⫽0.19). On multivariable logistic regression, anti-La titer ⱖ100 U/mL was the only significant predictor of NLE (OR 2.23; p⫽0.03). The mother’s age, diagnosis, intake of azathioprine, anti-Ro titers ⱖ50 U/mL and the child’s gender did not significantly impact on the risk of NLE.

Table 1. Multivariate analyses of the risk of autism spectrum disorders (ASD) in SLE mothers versus controls (n⫽9212)

Table. Characteristics of mothers of NLE cases and unaffected children

Covariates Maternal SLE No Yes Education level High school or less College or more Race/ethnicity Other Caucasian Preterm birth No Yes Small for gestational age No Yes Gestational diabetes No Yes

Multivariate HR for ASD (95% CI)

Unaffected children (nⴝ112)

Mothers Age at birth of child Diagnosis SLE or cutaneous lupus Sjogren’s syndrome Others Anti-Ro ⱖ50 U/mL* Anti-La ⱖ100 U/mL§ On HCQ or chloroquine On azathioprine On prednisone Older child with CHB

Reference 2.31 (1.03, 5.16)

31.7 ⫾ 4.8

32.2 ⫾ 4.6

Reference 0.87 (0.51, 1.49)

71 (72) 20 (21) 7 (7) 62 (67) 26 (28) 26 (27) 5 (5) 28 (29) 1 (1)

93 (83) 12 (11) 7 (6) 63 (66) 14 (14) 35 (31) 13 (12) 40 (36) 2 (2)

Reference 1.22 (0.66, 2.23) Reference 1.44 (0.59, 3.54)

p value 0.44 0.13

0.96 0.01 0.45 0.09 0.32 1.00

Mean⫾SD; n (%) *n⫽93 NLE and 95 unaffected children; §n⫽94 NLE and 103 unaffected children

Reference 1.75 (0.82, 3.72)

Conclusion: In the largest single-center case-control study of children born to anti-Ro and/or anti-La antibody positive women diagnosed with a connective tissue disease, HCQ or chloroquine exposure throughout gestation did not result in a significantly lower risk of NLE. High titers of anti-La antibodies in the mother were associated with an increased risk of NLE.

Reference 2.38 (0.85, 6.66)

a Matching for maternal age, calendar year, and accounting for birth order and sex; hazard ratio (HR); confidence interval (CI); systemic lupus erythematosus (SLE)

Conclusion: Compared to children from the general population, children born to women with SLE have a substantially increased risk of ASD. Our findings are supported by previous experimental data and should prompt further research on the potential role of SLE-related autoantibodies, such as N-methyl-D-aspartate receptor antibodies, in ASD. Disclosure: E. Vinet, None; S. Scott, None; C. A. Pineau, None; L. Joseph, None; A. E. Clarke, None; E. Fombonne, None; R. W. Platt, None; S. Bernatsky, None.

2832 The Effect Of Maternal Antimalarial Intake During Pregnancy On The Risk Of Neonatal Lupus. Julie Barsalou, Edgar Jaeggi and Earl D. Silverman. The Hospital for Sick Children, University of Toronto, Toronto, ON. Background/Purpose: Neonatal Lupus (NLE) results from passive transfer of anti-Ro and/or anti-La antibodies to the fetus during gestation. It has been suggested that prenatal exposure to hydroxychloroquine (HCQ)

Disclosure: J. Barsalou, GlaxoSmithKline, 9; E. Jaeggi, None; E. D. Silverman, None.

2833 Increased Cardiac Septal Defects In Children Born To Women With Systemic Lupus Erythematosus: Results From The OSLER Study. Evelyne Vinet1, Susan Scott2, Christian A. Pineau2, Ann E. Clarke3, Robert W. Platt3 and Sasha Bernatsky1. 1McGill University Health Center, Montreal, QC, 2McGill University Health Centre, Montreal, QC, 3McGill University, Montreal, QC. Background/Purpose: Cardiac septal defects, which include atrial septal defects (ASD) and ventricular septal defects (VSD), are the most frequent congenital heart anomalies in the general population. Uncontrolled small studies of children born to women with SLE suggest a potentially increased prevalence of structural heart defects, such as ASD and VSD. In a large population-based study, we aimed to determine if offspring born to mothers

S1213


NLE (nⴝ98)

a

with SLE have an increased risk of ASD and VSD compared to offspring born to mothers without SLE. Methods: The “Offspring of SLE mothers Registry (OSLER)” includes all women who had ⱖ1 hospitalization for delivery after SLE diagnosis, identified through Quebec’s universal healthcare databases (1989–2009). OSLER also includes a randomly selected control group of women, matched at least 4:1 for age and year of delivery, who did not have a diagnosis of SLE prior to or at the time of delivery. We identified children born live to SLE mothers and their matched controls, and ascertained ASD and VSD based on ⱖ1 hospitalization or physician visit with a relevant diagnostic code, within the first 12 months of life. We performed multivariate analyses to adjust for maternal demographics, sex and birth order of child, and maternal co-morbidities. In a subsample analysis of children with maternal public drug coverage throughout pregnancy, we further adjusted for relevant maternal medications Results: 509 women with SLE had 719 children, while 5824 matched controls had 8493 children. Compared to controls, children born to women with SLE experienced more ASD [2.9% (95%CI 1.9, 4.4) vs 0.8% (95%CI 0.6, 1.0) and VSD [1.7% (95%CI 1.0, 2.9) vs 0.7% (95%CI 0.5, 0.9)]. Among children with cardiac septal defects, those born to SLE mothers had more repair procedures compared to controls [9.7% (95%CI 3.4, 24.9) vs 4.4% (95%CI 1.9, 9.9)]. In multivariate analyses (n⫽9212), children born to women with SLE had a substantially increased risk of both ASD (OR 3.36, 95%CI 1.99, 5.70) and VSD (OR 2.48, 95%CI 1.20, 4.76) compared to controls. In addition, offspring of SLE mothers had a substantially increased risk of having a cardiac septal defect repair compared to controls (OR 4.90, 95%CI 1.11, 21.73). When accounting for the possibility of detection bias by excluding children with ⱖ 1 fetal echocardiography (n⫽331), adjusted effect estimates were similar to the primary multivariate analysis results for both ASD (OR 2.48, 95%CI 1.26, 4.87) and VSD (OR 2.09, 95%CI 0.96, 4.53). In the subsample analysis further controlling for relevant maternal medications (n⫽1925), though a trend remained for increased risk of ASD and VSD (combined) for offspring of SLE mothers versus controls, due to reduced sample size the 95% CI was wide and included the null value (both ASD and VSD combined; OR 1.83, 95%CI 0.59, 5.69). Conclusion: Compared to children from the general population, children born to women with SLE have an increased risk of ASD and VSD, as well as an increased risk of having a cardiac septal defect repair. The effect of SLE on cardiac septal defects does not seem to be explained by detection bias and might be independent of medication exposures. Disclosure: E. Vinet, None; S. Scott, None; C. A. Pineau, None; A. E. Clarke, None; R. W. Platt, None; S. Bernatsky, None.


2834 Maternal data Analysis Of The French Registry Of 205 Cases Of Immune Congenital Heart Block (neonatal lupus). Kateri Levesque1, Nathalie Morel2, Gae¨lle Guettrot-Imbert3, Mohamed Hamidou4, Jean Loup Pennaforte5, Pauline Orquevaux5, Jean-Charles Piette6, Christophe Deligny7, Zahir Amoura6, Olivier Meyer8, Olivier Fain9, Agathe Masseau4, Holly Bezanahary10, Pascal Cathebras11, Elizabeth Diot12, Yves Dulac13, Loic Guillevin14, Eric Hachulla15, Jean-Louis Pasquali16, Anne BesanconBergelin17, Bernard Bonnotte18, Je´rome Lebidois19, Alice Maltret20, Elisabeth Villain20 and Nathalie Costedoat-Chalumeau1. 1Hopital Cochin, Paris, France, 2COCHIN, Paris, France, 3Centre Hospitalier de Clermont-Ferrand, Clermont-Ferrand, France, 4Nantes University Hospital, Nantes, France, 5 CHU Reims, Reims, France, 6CHU Pitie´-Salpeˆtrie`re, Paris, France, 7Centre hospitalier Universitaire de Fort de France, Fort de France, Martinique, 8 Bichat University Hospital, Paris, France, 9Internal Medicine, Jean Verdier Hospital, Bondy, France, 10University Hospital of Limoges, Limoges, France, 11 University Hospital St Etienne, St Etienne, France, 12Department of Internal Medicine, Hoˆpital Bretonneau, Centre Hospitalier Re´gional Universitaire de Tours, Tours, France, Tours, France, 13University Hospital Toulouse, Paris, France, 14Division of Internal Medicine, Hoˆpital Cochin, University Paris Descartes, Paris, France, 15University Hospital Lilles, Lille CEDEX, France, 16 Hoˆpitaux Universitaires de Strasbourg, Hoˆpital civil, Service de me´decine interne et immunologie clinique, Strasbourg, France, 17Internal Medicine, Le Mans Hospital, Le Mans, France, 18Centre Hospitalier de Dijon, Dijon, France, 19Cardiac Institute, Paris, France, 20Groupe Hospitalier Necker Enfants Malades, Paris, France. Background/Purpose: Congenital heart block (CHB) occurs in 1 to 2 % of pregnancies exposed to anti-SSA antibodies. Few data are available

regarding the risk of the mothers of fetuses/children with CHB to develop an autoimmune disease. Methods: The inclusion criteria in the French registry of neonatal lupus are the positivity of anti-SSA and/or anti-SSB antibodies and a manifestation of neonatal lupus. In this retrospective study, we analyzed the data of the mothers who had a foetus/child with CHB. Results: 184 mothers of 205 fetuses/children with CHB were included: 99.5% (183) had anti-SSA antibodies and 69.6% (128) had anti-SSB antibodies. At the time of their first diagnosis of CHB, 136 mothers (73.9%) were asymptomatic. The 48 mothers (26.1%) with an auto-immune disease had a systemic lupus erythematosus (SLE, n⫽20, associated with antiphospholipid syndrome with venous thrombosis in 2 cases), a sjogren syndrome (n⫽12), an undifferentiated connective tissue disease (UCTD, n⫽10), a rheumatoid arthritis (n⫽3), a idiopathic thrombocytopenic purpura (n⫽1), an autoimmune hepatitis (n⫽1) and a systemic scleroderma (n⫽1). Clinical manifestations of SLE mainly included cutaneous and articular manifestations and only 8 mothers had more severe manifestations (renal, neurological, pericarditis, hematological). Only one had received cyclophosphamide. After a median follow up period of 8.8 years [1 day-36.5 years], 75 mothers (40.8%) remained asymptomatic. The 109 mothers (59.2%) with an autoimmune disease had a Sjogren syndrome (n⫽42), an SLE (n⫽37, including 2 associated with an antiphospholipid syndrome), an UCTD (n⫽20), a rheumatoid arthritis (n⫽3), an SLE associated with a Sjogren syndrome (n⫽2), a idiopathic thrombocytopenic purpura (n⫽2), an autoimmune hepatitis (n⫽1), a systemic scleroderma (n⫽1) and a primary obstetrical antiphospholipid syndrome (n⫽1). Conclusion: At the time of the CHB diagnosis in their offspring, a third of the mothers had a diagnosis of autoimmune disease. During the follow up period, another third developed an autoimmune disease. The remaining third was still asymptomatic after a median follow up period of 8.8 years [1 day-36.5 years]. Disclosure: K. Levesque, None; N. Morel, None; G. Guettrot-Imbert, None; M. Hamidou, None; J. L. Pennaforte, None; P. Orquevaux, None; J. C. Piette, None; C. Deligny, None; Z. Amoura, None; O. Meyer, None; O. Fain, None; A. Masseau, None; H. Bezanahary, None; P. Cathebras, None; E. Diot, None; Y. Dulac, None; L. Guillevin, None; E. Hachulla, None; J. L. Pasquali, None; A. Besancon-Bergelin, None; B. Bonnotte, None; J. Lebidois, None; A. Maltret, None; E. Villain, None; N. Costedoat-Chalumeau, None.

2835 Safety Of Gardasil® Vaccine In Systemic Lupus Erythematosus. J. Patricia Dhar, Lynnette Essenmacher, Renee Dhar, Ardella Magee, Joel Ager, Malini Venkatram, Harpreet Sagar and Robert Sokol. Wayne State University, Detroit, MI. Background/Purpose: Cervical neoplasia is increased in women with SLE presumably due to cervical infection with oncogenic Human Papillomavirus (HPV) types which persist in an immunosuppressed host. Vaccinating women with SLE against HPV is thus an important part of health prevention in this population. Gardasil® immunizes against the HPV types that cause the majority of cervical cancer (types 16 and 18) and genital warts (types 6 and 11), and has been shown to be protective against these HPV-related diseases. Methods: For this ongoing trial, twenty women ages 18–50 years with a history of mild to moderate SLE and minimally active or inactive SLE disease received Gardasil® at the standard dosing schedule (0, 2 months, 6 months). This study was approved by the Human Investigation Committee at Wayne State University and the U.S. Food and Drug Administration. Patients were excluded if they had active disease (SELENA-SLEDAI ⬎2), a history of severe disease, deep venous thrombosis, were on ⬎400 mg/day of hydroxychloroquine, were on ⬎15 mg/day of prednisone, or had active infections. To date, 20 patients have completed the second vaccine shot. Patients were monitored for adverse events (AE), SLE flare, and generation of thrombogenic antibodies and thrombosis. Results: The women in the study were predominantly African-American (75%), mean age 39 years with mean age at diagnosis of SLE at 29.8 years. All patients met American College of Rheumatology (ACR) criteria for SLE; 25% had a history of smoking, 95% had 4 or more sexual partners, 35% had a history of sexually transmitted diseases, and only 21% used condoms on a regular basis. History of abnormal pap smears occurred in 40%, ranging from ASCUS (atypical glandular cells of undetermined significance) to CIN 3 (cervical intraepithelial neoplasia grade 3). Most of our patients had multiple comorbidities in addition to SLE. Vaccine site reactions (VSRs) occurred in 50%, 90% being mild, with the most common reaction being pain. This

S1214

compares favorably to data from the current prescribing label showing frequency of VSRs in normal women to be 83.9% for Gardasil® vs. 75.4% for controls. For the non-vaccine site adverse events (nvAE), 90% of our cohort experienced at least one nvAE; there were 105 nvAEs reported from 18 patients and 90% of these nvAEs were mild. There was one serious AE related to underlying osteoarthritis and cervical disc disease. The most common nvAEs reported were musculoskeletal (n⫽28) followed by cardiopulmonary (n⫽25), headache (n⫽21), and dermatologic (n⫽19). None of the nvAEs were related to vaccine or SLE. There was no flare of SLE, thrombosis, or generation of thrombogenic antibodies in any patient. Conclusion: Preliminary data from our study shows that Gardasil® vaccine is safe to use in women with SLE. Vaccine site reactions are not increased in SLE patients. Other than vaccine site reactions, there were no related short term adverse events. Gardasil® vaccine administration did not result in any lupus flare or thrombosis. Women with SLE should be immunized with Gardasil® vaccine as part of their health prevention program, particularly since this population is at increased risk for HPV-related cervical dysplasia. Disclosure: J. P. Dhar, Merck, Inc, 2, GSK, Inc, 2; L. Essenmacher, Merck, Inc, 3; R. Dhar, None; A. Magee, Merck, Inc, 3; J. Ager, None; M. Venkatram, None; H. Sagar, None; R. Sokol, None.

ACR Concurrent Abstract Session Spondylarthropathies and Psoriatic Arthritis: Clinical Aspects and Treatment: Clinical and Imaging Aspects of Axial Spondyloarthritis Wednesday, October 30, 2013, 9:00 AM–10:30 AM

Results: Phase 1: The item pool contained 251 items in 44 categories. Phase 2 was performed based on data collected along an international cross sectional study among 1915 AS patients (mean age 51.2⫾3.6, 53% male, BASDAI 5.5⫾2.4) in 4 continents. For 82 items of the functioning part a unidimensional scale, fit to the Rasch model and absence of Differential Item Function (DIF) could be confirmed. Phase 3: An expert committee selected 50 functioning items using predefined selection criteria. Phase 4: An international cross sectional study with 628 AS patients (mean age 48.5⫾14.2, 51.6% male, BASDAI 5.6⫾2.3) was conducted in 4 continents. Misfit was identified in 4 items and DIF in 15 items. More than 50% of the items showed a residual correlation between each other above a value of 0.2 in the initial round. Phase 5: Based on results of the analyses in step 4, the consensus members agreed on 17 final items. In the 17 items fit to the model, no residual correlation and absence of constant DIF could be confirmed with a PersonSeparation Index of 0.82. The item location has been shown to be 0.00 ⫾ 1.84 with a fit residual of 0.06 ⫾ 1.24 and the person location has been shown to be 0.01 ⫾ 1.80 with a fit residual of ⫺0.30 ⫾ 0.67. The chi-square probability was 0.73. Conclusion: The ASAS HI measure contains 17 dichotomous items addressing categories of pain, emotional functions, sleep, sexual functions, mobility, self-care, community life and employment. In covering much of the ICF Core Set for AS, these items represent a whole range of abilities of patients with AS. The questionnaire will be translated and further tested for sensitivity to change. ASAS HI can be used in clinical trials and clinical practice as a new composite index that captures relevant information on the health status of the patients. Disclosure: U. Kiltz, None; D. van der Heijde, None; A. Boonen, None; A. Cieza, None; G. Stucki, None; M. A. Khan, None; W. P. Maksymowych, None; H. Marzo-Ortega, None; J. D. Reveille, None; S. Stebbings, None; C. Bostan, None; J. Braun, None.

2836

Background/Purpose: The impact of ankylosing spondylitis (AS) on a patient’s life can be considerable. Patients suffer from pain, stiffness and fatigue, are limited in their activities, and restricted in social participation. The International Classification of Functioning, Disability and Health (ICF) is a model to systematically classify and describe functioning, disability, and health in human beings. However no ICF-based patient-reported outcome measure has been developed. the objective is to develop a measure to assess health in patients with AS, the ASAS (Assessments of SpondyloArthritis international Society (ASAS)) Health Index, based on the Comprehensive ICF Core Set for AS. Methods: Development has been performed in five phases. Table 1. Phases of development for the ASAS Health Index Phase

Objectives

Methods Linkage of various assessment tools for functioning and health to ICF categories

1st postal patient survey

Development of a pool of items representing the categories of the Comprehensive ICF Core Set for AS Patient preference and weighting of the items per ICF category Item reduction (within and across ICF categories)

Expert consultation 2nd postal patient survey Consensus Meeting

Agreement on item reduction Validation of the draft version and further item reduction Agreement on a final version

Ia

Preparatory

Ib

Patient meeting

II III IV V

Relative weight to each item, patients distributed 100 points per ICF category. Factor Analysis, Rasch Analysis, Spearman rank correlation coefficient Nominal Consensus Process Testing psychometric properties Rasch Analysis Nominal Consensus Process

2837 Higher Disease Activity Leads To More Damage In The Early Phases Of Ankylosing Spondylitis: 12-Year Data From The OASIS Cohort. Sofia Ramiro1, A.M. van Tubergen2, De´sireé van der Heijde3, Carmen Stolwijk2, Maxime Dougados4, Filip Van den Bosch5 and Robert Landewe´6. 1Academic Medical Center, University of Amsterdam, Amsterdam, Netherlands, 2 Maastricht University Medical Center, Maastricht, Netherlands, 3Leiden University Medical Center, Leiden, Netherlands, 4Rheumatology B Department, Paris-Descartes University, Cochin Hospital, Paris, France, 5Gent University Hospital, Gent, Belgium, 6Academic Medical Center Amsterdam & Atrium Medical Center, Heerlen, Netherlands. Background/Purpose: For years, it was unclear if inflammation and radiographic progression were related in ankylosing spondylitis (AS), but studies were only of short follow-up and not analysed optimally. We here present a true longitudinal analysis on the long-term relationship between disease activity and radiographic damage. Methods: Patients from the Outcome in AS International Study (OASIS) were followed-up for 12 years, with biannual clinical and radiographic assessments. Two readers independently scored the x-rays according to the modified Stoke Ankylosing Spondylitis Spine Score (mSASSS) and scores were averaged. Disease activity measures include the BASDAI, ASDASCRP, CRP, ESR, patient’s global assessment and spinal pain. The relationship between disease activity measures and radiographic damage was investigated using generalized estimating equations. Auto-regressive models with 2-year time-lags were used and analyses were adjusted for potential confounders. Different models were constructed: model 1 – with ASDAS as a continuous measure; model 2 – with ASDAS disease activity states; model 3 – with BASDAI continuous and CRP; model 4 – with BASDAI categories and CRP; model 5 – with patient global and CRP; model 6 – with spinal pain and CRP. Models were repeated replacing CRP by ESR. Model fit was compared using the quasi-likelihood information criterion (QIC). Interactions were tested for gender and disease duration. Results: A total of 185 patients were included (70% males, mean (SD) age: 43(12) years, mean symptom duration: 20(12) years and 83% HLA-B27 positive). All disease activity measures, except ESR, were significantly longitudinally associated with radiographic progression (Table). Neither medication (NSAIDs, NSAID score or biologicals), nor the presence of extra-articular manifestations (uveitis, psoriasis or IBD) were confounding this relationship. The models with ASDAS had the best fit (i.e. lowest QIC): An increase in one ASDAS-unit led to an increase in 0.72 mSASSS-units and a ‘very high disease activity state’ (i.e. ASDAS⬎3.5) compared to inactive disease (i.e. ASDAS⬍1.3) represented an additional progression of 2.31

S1215


Development of a Health Index in Patients With Ankylosing Spondylitis (ASAS HI)—Final Result of a Global Initiative Based On the International Classification of Functioning, Disability and Health (ICF) Guided By Assessment of Spondyloarthritis International Society (ASAS). Uta Kiltz1, De´sireé´ van der Heijde2, Annelies Boonen3, Alarcos Cieza4, Gerold Stucki5, Muhammad Asim Khan6, Walter P. Maksymowych7, Helena MarzoOrtega8, John D. Reveille9, Simon Stebbings10, Cristina Bostan11 and Juergen Braun1. 1Rheumazentrum Ruhrgebiet, Herne, Germany, 2Leiden University Medical Center, Leiden, Netherlands, 3Maastricht University Medical Center, Maastricht, Netherlands, 4University Southampton, Southampton, United Kingdom, 5University of Lucerne, Lucerne, Switzerland, 6Case Western Reserve University Hospital, Cleveland, OH, 7University of Alberta, Edmonton, AB, 8Leeds Musculoskeletal Biomedical Research Unit and University of Leeds, Leeds, United Kingdom, 9Univ of Texas Health Science Center at Houston, Houston, TX, 10University of Otago, Otago, New Zealand, 11Paraplegic Research Unit, Nottwil, Switzerland.

mSASSS-units per 2 years (Table). Results were consistent across all disease activity measures. The effect of ASDAS on mSASSS was higher in males vs females (␤ 0.98 vs ⫺0.06) and in patients with shorter vs longer (⬍18 vs ⱖ18 years) disease duration (0.84 vs 0.16). Table. Longitudinal relationship between disease activity measures and radiographic damage* Variable Previous mSASSS (0–72)

Univariable regression ␤ (95% CI) (N ⴝ 174 - 185) –

Multivariable regression 1 ␤ (95% CI) (N ⴝ 183)


1.03 (1.01; 1.05) 1.03 (1.01; 1.05)



1.03 (1.01; 1.05) 1.03 (1.01; 1.05)



1.03 (1.01; 1.05)

1.03 (1.01; 1.05)

BASDAI (0–10)

0.24 (0.09; 0.40)

§

§

0.21 (0.06; 0.37)

§

§

§

ASDAS

0.72 (0.41; 1.04)

0.72 (0.41; 1.04)

§

§

§

§

§

Results: New fatty lesions did not develop at all in sites (quadrants or VUs) where there was no osteitis at baseline. New fatty lesions at year 2 (Y2) or Y3 primarily developed in SI-joint quadrants in which osteitis disappeared between baseline (BL) and Y2, this was the case in 28.6% quadrants (10 out of 35) (see Figure 1, section A2). In the spine only 1 VU developed a new fatty lesion in which osteitis disappeared between BL and Y2/Y3.

ASDAS disease activity states - Moderate vs inactive (ⱖ1.3 and (⬍2.1 vs ⬍1.3)

0.57 (⫺0.56; 1.69)

§

0.57 (⫺0.56; 1.69)

§

§

§

§

- High vs inactive (ⱖ2.1 and ⬍3.5 vs ⬍1.3)

0.91 (⫺0.17; 1.99)

§

0.91 (⫺0.17; 1.99)

§

§

§

§

- Very high vs inactive (ⱖ3.5 vs ⬍1.3)

2.31 (1.11; 3.51)

§

2.31 (1.11; 3.51)

§

§

§

§

BASDAI disease activity states - Moderate vs ⫺0.19 (⫺0.98; 0.60) inactive (ⱖ2 and ⬍4 vs ⬍2)

§

§

§

⫺0.29 (⫺1.09; 0.52)

§

§

- High vs inactive (ⱖ4 and ⬍6 vs ⬍2)

1.13 (0.30; 1.97)

§

§

§

1.06 (0.21; 1.90)

§

§

- Very high vs inactive (ⱖ6 vs ⬍2)

1.11 (0.12; 2.11)

§

§

§

0.82 (⫺0.19; 1.84)

§

§

CRP (mg/l)

0.02 (0.01; 0.04)

§

§

0.02 (0.00; 0.04)

0.02 (0.00; 0.04)

Patient’s global assessment (0–10)

0.20 (0.08; 0.32)

§

§

§

§

0.17 (0.05; 0.29)

§

Pain (0–10)

0.25 (0.12; 0.38)

§

§

§

§

§

0.22 (0.00; 0.04)

5349

5458

5665

5579

5664

5473

QIC of the model

0.02 (0.00; 0.04) 0.02 (0.01; 0.04)

*All models are time-lagged (2 years) and auto-regressive § Not included in the model

Conclusion: In AS disease activity is unequivocally longitudinally associated with radiographic progression. ASDAS is the best measure to reflect this relationship. The effect of disease activity on radiographic damage is more pronounced in men and in the earlier phases of the disease. These findings may give support to use ASDAS as a treat-to-target. Disclosure: S. Ramiro, None; A. M. van Tubergen, None; D. van der Heijde, None; C. Stolwijk, None; M. Dougados, None; F. Van den Bosch, None; R. Landewe´, None.


2838 Relationship Between Active Inflammatory Lesions In The Spine and Sacroiliac Joints and Development Of Fatty Lesions On Whole-Body MRI In Patients With Early Axial Spondyloarthritis—Long-Term Data Of The Esther Trial. In-Ho Song1, Kay-Geert A. Hermann2, Hildrun Haibel3, Christian Althoff4, Denis Poddubnyy3, Joachim Listing5, Anja Weiss6, Ekkehard Lange7, Bruce Freundlich8, Martin Rudwaleit9 and Joachim Sieper10. 1Charite´ Universita¨tsmedizin Berlin, Campus Benjamin Franklin, Berlin, Germany, 2Charite´ Universita¨tsmedizin Berlin, Campus Mitte, Berlin, Germany, 3Charite´ Universita¨tsmedizin Berlin, Berlin, Germany, 4Charite Universita¨tsmedizin Berlin, Campus Mitte, Berlin, Germany, 5 German Rheumatism Research Center, Berlin, Germany, 6German Rheumatism Research Centre, Berlin, Germany, 7Pfizer Pharma AG, Berlin, Germany, 8University of Pennsylvania, Philadelphia, PA, 9Endokrinologikum Berlin, Berlin, Germany, 10Charite´ Universita¨tsmedizin Berlin, Campus Benjamin-Franklin, Berlin, Germany. Background/Purpose: In patients with early axial spondyloarthritis (SpA) with evidence of active inflammation on magnetic resonance imaging (MRI) in the spine and/or sacroiliac (SI-)joints at baseline [1] we previously reported about the relationship between inflammation and the development of fatty lesions in the subchondral bone marrow [2]. We now assessed this relationship in the spine and SI-joints on MRI in these patients with 3 years of continuous treatment with etanercept (ETN). Methods: Wb-MRIs of those 40 patients who reached the end of year 4 were scored for active inflammation (osteitis) and fatty lesions in the SI-joint quadrants and spine vertebral units (VUs). For this analysis we pooled the data of the patients who were continously treated with ETN for 3 consecutive years. Scoring was performed by two radiologists, blinded for treatment arm and MRI time point. Fatty lesions were scored for absence or presence according to our recently published score [2]. For this analysis, the presence or absence of osteitis and fatty lesions were only counted if both scorers agreed.

Figure 1. Development of new fatty lesions depending on the presence of inflammation on magnetic resonance imaging related to sacroiliac joint quadrants and spine vertebral units (VUs) with no fatty lesions at baseline (BL). Y⫽ year; n⫽ number of sites (SI-joint quadrants or spine vertebral units)

Persistent osteitis at Y2 or new development of osteitis (occurred rarely) was associated with a low rate of new development of fatty lesions, 16% (3/19) and 33% (1/3) for SI-joint and spine, respectively (A1, B1). New onset of osteitis at Y2 was not associated with new fatty lesions. At baseline 74 SI-joint quadrants and 26 spine VUs showed fatty lesions. Until Y2 and Y3 only in 3 SI-joint quadrants and 0 VUs fatty lesions disappeared. The mean spine fatty lesion scores were 1.1 (2.1) at BL, 1.4 (2.5) at Y2 and 1.3 (2.3) at Y3. The mean SI-joint fatty lesion scores were 4.6 (6.3), 5.2 (6.4) and 4.7 (6.3), respectively. Conclusion: Our data indicate a strong relationship between presence of inflammation and new development of fatty lesions. Furthermore there was no increase of fatty lesions over continous treatment of axial SpA patients with etanercept over 3 years.. Whether this is predictive of stopping radiographic progression needs to be further investigated. References: [1] Song I.-H. et al. Ann Rheum Dis. 2011 Apr;70(4):590–6. [2] Song I.-H. et al. Ann Rheum Dis. 2011 Jul;70(7):1257–63. Disclosure: I. H. Song, Pfizer, Merck Sharp Dohme/Schering Plough, AbbVie , 5; K. G. A. Hermann, None; H. Haibel, AbbVie,MSD, Chugai, 8, AbbVie, MSD, 5; C. Althoff, None; D. Poddubnyy, Merck Sharp Dohme/Schering Plough, AbbVie, 5; J. Listing, None; A. Weiss, None; E. Lange, Pfizer Inc, 3; B. Freundlich, Pfizer Inc, 9; M. Rudwaleit, Pfizer, Merck Sharp Dohme/Schering Plough, AbbVie, UCB, 5; J. Sieper, Pfizer, Merck Sharp Dohme/Schering Plough, AbbVie, UCB, 5.

2839 Effects Of Treatment On Spinal Fat Lesions As Assessed By MRI With The Fat Spondyloarthritis Spine Score. Susanne Juhl Pedersen1, Zheng Zhao2, Stephanie Wichuk3, Robert GW Lambert3 and Walter P. Maksymowych3. 1Copenhagen University Hospital at Glostrup, Copenhagen, Denmark, 2Department of Rheumatology, University of Alberta and PLA General Hospital, Beijing, PR China, Beijing, China, 3University of Alberta, Edmonton, AB. Background/Purpose: Fat lesions develop after resolution of inflammation in the spine (1), and have also been shown to predict development of new syndesmophytes (2). Consequently, spinal fat lesions may constitute an

S1216

important measure of treatment efficacy and a surrogate for structural damage progression. The aim of this study was to investigate the effects of different treatments on development of fat lesions on spinal MRI. Methods: A total of 135 patients with axial spondyloarthritis (SpA) had MRIs performed with a mean (SD) follow-up of 1.7 (0.8) years. Patients either received no anti-inflammatory treatment (n⫽12), or were treated with non-steroid anti-inflammatory drugs (NSAIDs) only (n⫽55), tumor-necrosis-factor-alpha (TNF␣) inhibitor only (n⫽15) or NSAIDs and TNF␣ inhibitor in combination (n⫽53). Spinal fat lesions were assessed by the Fat Spondyloarthritis Spine Score (FASSS), where fat lesions are scored based on anatomical location and recorded dichotomously (present/absent) at each vertebral endplate (3). Scoring range per disco-vertebral unit (DVU) for the cervical spine is 0–8, and for the thoracic and lumbar spine: 0–24, resulting in a total score range of 0–456 for all 23 DVUs. Two rheumatologists evaluated pairs of scans blinded to clinical, biochemical or imaging data. Inter-class correlation coefficients were 0.96 (CI 95% 0.94–0.97) for baseline and 0.86 (0.80–0.90) for change scores. Results: Patients treated only with NSAIDs had significantly shorter disease duration, lower CRP and structural damage on X-rays as assessed by mSASSS as compared to patients treated with TNF␣ inhibitor alone or in combination with NSAIDs (results not shown, Mann-Whitney test). Patients treated only with NSAIDs had lower FASSS at baseline (p⫽0.03) and had less change (p⫽0.01) as compared to patients treated only with TNF␣ inhibitors (Table 1). Furthermore, significantly more patients treated only with NSAIDs decreased or had no change in FASSS score during follow-up as compared to patients receiving no anti-inflammatory treatment (p⫽0.009) or TNF␣ inhibitor alone (p⫽0.03) (Table 1) (Chi2 test). Table 1. FASSS baseline and change scores, and number and frequency of patients with decrease/no change or increase in FASSS scores stratified according to treatment. FASSS scores Change

Change in FASSS score Decreased/ no change Increased

Treatment group

Baseline

1: No anti-inflammatory treatment (n⫽12) 2: NSAIDs only (n⫽55) 3: TNF␣ inhibitor only (n⫽15) 4: NSAIDs and TNF␣ inhibitor (n⫽53)

10.3 (0; 187)

2.0 (⫺1; 9.5)

2 (16.6)

10 (83.3)

11.5 (0; 115) 18.5 (1; 118)

0 (⫺9.5; 50) 8.5 (⫺8; 51)

32 (58.2) 4 (26.7)

23 (41.8) 11 (73.3)

10.0 (0; 93)

1.5 (⫺35.5; 34)

21 (39.6)

32 (60.4)

Results are provided a median (range) or N (%).

References: 1. Chiowchanwisawakit et al. ARD 2010; 2. Chiowchanwisawakit et al. AR 2011; 3. Østergaard et al. J Rheum 2009 Disclosure: S. J. Pedersen, None; Z. Zhao, None; S. Wichuk, None; R. G. Lambert, None; W. P. Maksymowych, None.

2840 Predictors Of Spinal Mobility Progression In A Multi-ethnic Cohort Of Patients With Ankylosing Spondylitis. Roozbeh Sharif1, Trisha M. Parekh1, Lianne S. Gensler2, MinJae Lee3, Mohammad Rahbar4, Laura A. Diekman4, Michael H. Weisman5, Michael M. Ward6, Shervin Assassi3 and John D. Reveille3. 1University of Texas Medical Branch, Galveston, TX, 2University of California, San Francisco, San Francisco, CA, 3University of Texas Health Science Center at Houston, Houston, TX, 4The University of Texas Health Science Center at Houston, Houston, TX, 5Cedars-Sinai Medical Center, Los Angeles, CA, 6NIAMS/ NIH, Bethesda, MD. Background/Purpose: Clinicians utilize several different clinical measurements (metrology) in patients with Ankylosing Spondylitis (AS) to assess the spinal mobility. We aimed to examine the pattern of change in metrology over time and to evaluate the factors associated with these changes.

Table 1. Univariable longitudinal analysis of demographic and clinical factors associated with deterioration of mobility over time among patients with ankylosing spondylitis Hip Hip Lateral Cervical Cervical Chest external internal lumbar Occiput Schober Predictor factor flexion rotation expansion rotation rotation spine flexion to wall test Age ⱖ40 years

ⴚ24.4 ⴚ29.6 (

2013 Annual Meeting Abstract Supplement

2013 Annual Meeting Abstract Supplement

Suggest Documents

2015 Annual Meeting Abstract Supplement - Society of Toxicology

Meeting Registration Rates • 2013 ACCP Annual Meeting

Meeting Registration Form • 2013 ACCP Annual Meeting

Kidney Week 2013 - JASN Abstract Supplement

Kidney Week 2013 - JASN Abstract Supplement

Kidney Week 2013 - JASN Abstract Supplement

TMS 2013 Annual Meeting Publications

2013 Annual General Meeting - Norbord

abstract supplement

Abstract Supplement

2014 Joint Annual Meeting ABSTRACT BOOK

2014 Joint Annual Meeting ABSTRACT BOOK

2014 Joint Annual Meeting ABSTRACT BOOK

2014 Joint Annual Meeting ABSTRACT BOOK

2014 Joint Annual Meeting ABSTRACT BOOK

SICB - 2013 meeting - Abstract Details

TMS 2013 Annual Meeting Publications

29th Annual LAMA/ATA Meeting Program (2013)

AAAS Annual Meeting, Boston, February 2013

DCPA 2013 Annual Meeting report - Google Sites

TMS 2013 Annual Meeting Publications

abstract supplement

Abstract Supplement - MAFIADOC.COM

abstract supplement