Thrombin generation as objective parameter of ... - Wiley Online Library

Haemophilia (2014), 20, e7–e14

DOI: 10.1111/hae.12309

ORIGINAL ARTICLE Clinical haemophilia

Thrombin generation as objective parameter of treatment response in patients with severe haemophilia A and high-titre inhibitors  Z A R , * † C . P . B E L T R AN  -MIRANDA,‡ M. A. ESPARZA-FLORES,§ H . L U N A - Z AI  J . S O T O - P A D I L L A , § A . B E R G ES - G A R CIA , ¶ M . D . C . R O D RIG U E Z - Z E P E D A , * * M . T . P O M P A - G A R Z A † † and A . R . J A L O M A - C R U Z ‡ ‡ *CUCEI, Universidad de Guadalajara,, Guadalajara, Jalisco; †Doctorado en Genetica Humana, CUCS, Universidad de Guadalajara,, Guadalajara, Jalisco; ‡CUSUR, Universidad de Guadalajara,, Ciudad Guzm an, Jalisco; §Hospital de Pediatrıa, CMNO, IMSS, Guadalajara, Jalisco; ¶CMN La Raza, IMSS, Mexico, D.F.; **CMN Siglo XXI, IMSS, Mexico, D.F.; ††UMAE 25, Monterrey, Nuevo Le on; and ‡‡Divisi on de Genetica, Centro de Investigaci on Biomedica de Occidente, Guadalajara, Jalisco, Mexico

Summary. In Mexico, 15% of haemophilia A (HA) patients develop inhibitory alloantibodies in response to replacement therapy with factor VIII (FVIII), requiring bypass therapy such as activated prothrombin complex concentrate (APCC). Because bypass therapy has not been broadly available in Mexico even in recent years, this study aimed to evaluate the thrombin generation assay (TGA) in assessing the response to FVIII or APCC treatment in patients with severe HA positive to inhibitors. We studied 189 patients with severe HA. Clinical severity was verified by one-stage APTT-based clotting assay. Inhibitors to FVIII were investigated by the Nijmegen–Bethesda (N–B) method, and type of inhibition was assessed through serial plasma dilutions. Thrombin generation was measured with the calibrated automated thrombogram in inhibitor-positive plasmas previously spiked and incubated with FVIII or APCC. Data were analysed using ANOVA, Student or Fisher’s exact tests. We

detected 47 (24.9%) subjects with high-titre (5– 1700 N–B U mL 1) and 25 (13.2%) subjects with lowtitre inhibitor antibodies (0.6–4.7 N–B U mL 1). We found an association between kinetic behaviour and clinical response to FVIII (P = 0.0049) or vs. FVIII response evaluated with TGA (P = 0.0007). Global concordance between clinical and in vitro response was 70%. By evaluating the capacity of thrombin formation in a plasma sample, TGA predicts the response to FVIII or APCC therapy and allows individual optimization of resources in patients with severe HA and high-titre inhibitors. The inhibition pattern of the antibodies to FVIII:C correlated with the TGA parameters and showed an association with the clinical response to FVIII.

Introduction

patients with haemophilia receive on-demand therapy but the number of new patients who receive primary or secondary prophylaxis according to recent clinical guidelines is increasing in the major health care centres [1]. When patients develop inhibitors, they are treated with bypassing agents (APCC or rFVIIa) according to international criteria [2,3]. Bypassing products for inhibitor treatment are becoming more accessible [4] such as prophylaxis for patients with inhibitors [5]. In recent years, bypass therapy was not broadly available in Mexico and, alternatively, patients with inhibitors were treated with high doses of factor VIII (FVIII) in bolus or continuous infusion, sometimes

One of the most severe complications of replacement therapy in haemophilia is the development of inhibitors to the infused clotting factors, rendering ineffective control of bleeding episodes. In Mexico, most Correspondence: Ana Rebeca Jaloma Cruz, PhD, Centro de Investigaci on Biomedica de Occidente, IMSS, Sierra Mojada 800 Col. Independencia, C.P. 44340, A.P. 1-3838, Guadalajara, Jalisco, Mexico. Tel.: +52 33 3618 9410; fax: +52 33 3618 1756; e-mail: [email protected] Accepted after revision 7 October 2013 © 2013 John Wiley & Sons Ltd

Keywords: bypass therapy, inhibitor kinetics, inhibitor therapy, severe haemophilia A, thrombin generation assay, treatment response

e7

e8

 H. LUNA-Z AIZAR et al.

together with steroids [6,7]. Although it was not the treatment of choice, in some cases it was able to resolve the haemorrhage and, only when unsuccessful, bypassing agents were used [4]. According to the in vitro pattern of inactivation, two types of antibodies are recognized. Type I inactivates FVIII completely in a bimolecular reaction that follows second-order kinetics. Type II partially inhibits FVIII activity and follows a more complex kinetics [8,9]. It has been reported that patients with type II inhibitors, but not with type I inhibitors, usually respond well to infusion of high doses of FVIII [10]. We recently described an intermediate pattern of inactivation (type III) in a patient who did not respond to FVIII treatment [11]. Thrombin generation (TG) reflects the functional status of the blood coagulation mechanism as a result of the interplay between pro- and anticoagulant factors [12]. It can be easily measured with calibrated automated thrombography (CAT), a TG assay (TGA) that displays the course of the thrombin formation in a clotting plasma sample monitored by the splitting of a fluorogenic substrate [13]. TGA has been used as a global functional test to evaluate the hemostatic capacity in haemophilia patients or in various pathologies to assess bleeding or thrombotic risk and its modification with antithrombotic or hemostatic treatment [14–17], and has been proven to correlate with the clinical phenotype [18,19]. It has been used to identify severe haemophiliacs with a mild clinical phenotype using platelet-rich plasma [20]. With the development of a well-controlled laboratory/clinical protocol, it could also be used as a prognostic biomarker for therapeutic decisions [21]. The objective of this study was to investigate the usefulness of TGA to predict the response to FVIII and APCC treatment in patients with severe HA with high-titre inhibitors.

Materials and methods Patients One hundred and eighty-nine Mexican patients with severe HA suspected of – or known as – carrying an inhibitor to FVIII clotting activity (FVIII:C) were enrolled from health care institutions representing 20 states of Mexico. The study was approved by the corresponding local ethics committees according to a national institutional agreement. Written consent was obtained from all participants. A questionnaire (previously validated by haematologists who are ‘opinion leaders’ in Mexico) was provided for each patient and completed by their haematologist in regard to the treatment administered

Haemophilia (2014), 20, e7--e14

and the evolution of the last bleeding episode. A scale from 0 to 4 was applied to evaluate clinical response, considering positive response for values of 3–4 and negative response for values of 0–2 on the following bases: (i) reduction or disappearance of bleeding after administration of therapy and (ii) clinical evolution of injury (decreased pain, swelling and stiffness). Treatment generally administered was purified FVIII at normal or high doses, sometimes combined with steroids; therefore, clinical response was only considered for FVIII treatment.

Sample collection and plasma preparation A quantity of 9 mL of venous blood was collected into 4.5 mL Vacutainer polypropylene tubes (Becton Dickinson, Franklin Lakes, NJ, USA) containing 109 mmol L 1 trisodium citrate. Sampling was done in patients without recent treatment or in patients with at least 5 days since the last infusion of purified FVIII. TG analysis was done in platelet-poor plasma (PPP) obtained by centrifugation of the upper half volume of plasma supernatant and aliquoted into cryovials and frozen at 80°C until use. For coagulation and kinetics studies, PPP was obtained from the remaining plasma and aliquoted for immediate processing or frozen at 80°C. This procedure was performed within 30 min of venipuncture. Patients from other geographical states were assisted at the nearest hospital centre for collecting their blood samples under the supervision of any of the investigators who purposely travelled for sampling and appropriate preanalytical handling of the samples. Plasma aliquots were placed in dry ice at 40°C and transported to Guadalajara, Mexico where they were stored at 80°C until use. Pooled normal plasma (PNP) was obtained from blood samples of 35 healthy volunteers (20 men and 15 women) collected as previously described and separated into two portions: one remained without change as PNP for use in TGA; the other PNP portion was buffered with solid imidazole RA (Merck KGaA, Darmstadt, Germany) [Buffered Pooled Normal (BPN), 0.1 mol L 1, pH 7.4] for use in inhibitor studies according to the Nijmegen–Bethesda (N–B) method [22]. PNP and BPN were aliquoted and stored frozen at 80°C until use. All assays were measured at a single centre under similar conditions by a sole investigator.

Normal ranges Normal ranges for FVIII:C were obtained from the individual values of 35 healthy volunteers. Normal TGA parameters were established as the average value of 20 independent aliquots of PNP running in parallel with each set of tests.

© 2013 John Wiley & Sons Ltd

TREATMENT RESPONSE ASSESSED BY THROMBIN GENERATION

Factor VIII:C clotting activity, inhibitor and kinetics study in plasma Plasma FVIII:C was measured by one-stage APTTbased clotting assay. FVIII:C, inhibitor measurements and kinetic studies were carried out as described [11] using reagents [immunodepleted FVIII-deficient plasma as substrate, FSL-ACTINâ (rabbit brain thromboplastin + ellagic acid) and imidazole buffer 0.1 mol L 1] purchased from Dade-Behring (Newark, DE, USA) in a CA-540 coagulometer (Sysmex Corporation, Kobe, Japan) according to the manufacturer’s instructions. FVIII inhibitors were assessed by the N–B method [22]. Residual FVIII:C of the test mixtures was calculated as percentage of the normal control and converted to N–B units (N–B U mL 1); one N–B U is defined as the amount of inhibitor resulting in 50% of residual FVIII:C activity according to the classic Bethesda test [23] but indicating that the Nijmegen modification is used. Inhibition kinetics was studied at different antibody concentrations achieved by plasma dilutions as described by Ling et al. [24].

e9

and 150–200 U kg 1 body weight respectively). Plasmas were incubated with the referred therapeutic product for 60 min at 37°C. All experiments were carried out in duplicate. TG was triggered with 1 pM recombinant tissue factor and 4 lM of phospholipids: phosphatidylserine/phosphatidylcholine/phosphatidylethanolamine (PPP-Low Reagent). The microplate was inserted into the fluorometer and allowed to warm to 37°C (minimal time 5 min). The reaction was started by automatic dispensing of 20 lL of freshly prepared fluorogenic substrate–calcium working solution (Flu– Ca). All plasmas and reagents including Flu–Ca reached 37°C before use. The final volume in each well was 120 lL and the fluorescence signal was measured at 15–20-s intervals for 60 min. With each set of measurements, one aliquot of PNP was processed in parallel. During the measurement, TG curves (thrombograms) were generated using thrombinoscope software v.3.0.0.29 (Maastricht, The Netherlands) and lag time, thrombin peak, time to peak and endogenous thrombin potential (ETP) (area under the curve) were calculated. For the TG tests, we defined the positive response to the therapeutic agents with the ETP ≥60% of PNP value of each set of tests.

Thrombin generation measurement Thrombin generation was measured by CAT according to Hemker et al. [13] with some modifications using PPP of patients with inhibitors. Measurements were done in a 96-well plate Fluoroskan Ascent fluorometer (Thermolab Systems OY, Helsinki, Finland) equipped with a dispenser and a 390/460 filter set (excitation/emission) to detect fluorescence intensity. All TG reagents (PPP-Low reagent, a2M-Thrombin Calibrator, FluCa kit: Fluo-buffer and Fluo-substrate Z-Gly-Gly-Arg-amino-methyl-coumarin) were purchased from Thrombinoscope BV (Maastricht, The Netherlands), and 96-well round-bottomed microtiter plates were from Immulon 2HB (Thermo Labsystems, Woburn, MA, USA). Purified FVIII (HEMOFIL M 250 IU, Baxter HealthCare Corporation, Baxter Bioscience, Los Angeles, CA, USA) and APCC (FEIBA VH; Baxter Bioscience) were donated by BaxterMexico. Both products were reconstituted as recommended by the manufacturer, aliquoted and stored at 80°C prior to the spiking experiments. In a CAT experiment, thrombin concentrations were derived from two sets of readings: TG wells where TG takes place and CAL wells where a fixed amount of thrombin calibrator provides the fluorescent reference signal. Thrombin generation was measured in inhibitorpositive PPP under the following four conditions: (i) baseline, (ii) spiked with purified FVIII 2 U mL 1 of plasma (equivalent to therapeutic doses of 50– 100 U kg 1 body weight), (iii) and (iv) spiked with APCC 1 U mL 1 and 2 U mL 1 of plasma (equivalent to normal and high doses of APCC: 75–100 U kg 1 © 2013 John Wiley & Sons Ltd

Statistical analysis was applied to analyse differences between type of inhibitors and therapeutic products; Student t-test was made for comparison between mean values and Pearson’s regression test for correlation analysis between the baseline ETP and after spiking with both FVIII and APCC (Microsoft Office Excel 2007 and SPSS Statistics, v.10.0, Chicago, IL, USA). Fisher’s exact test was used for correlation studies with the program RXC for 2.2 cross-tab (Miller MP, 1997, Northern Arizona University). P < 0.05 was considered statistically significant.

ANOVA

Results FVIII clotting activity All participating patients showed FVIII:C 4% residual FVIII:C activity) and complex kinetics. (c) Type III: Complete inactivation but less steep slope than types I/II. (d) Differences among the three patterns of inactivation are more evident in log-linear scale. Grey area shows complete inactivation (≤4% of residual FVIII:C activity).

© 2013 John Wiley & Sons Ltd

TREATMENT RESPONSE ASSESSED BY THROMBIN GENERATION

e11

Table 1. Severe hemophilia A patients with type I inhibitors: inhibitor titres, clinical response to FVIII, ETP values and concordance between clinical and in vitro response to FVIII. Endogenous thrombin potential* Patient

Inhibitor N-B U mL

HA-25 HA-3 HA-58 HA-67 HA-59 HA-27 HA-14 HA-57 HA-55 HA-73 HA-47 HA-11 HA-19 HA-65 HA-68 HA-32 HA-2 HA-66 HA-7 HA-42 HA-72 HA-21 HA-60 HA-46 HA-69 HA-71 HA-8 HA-31 HA-1 Mean  SD

102 97 90 87 78 72 40 14 14 11 10 9.8 6.7 5.1 94 67 30 23 16 12 6 25 31 20 65 45 32 10 5.7 —

1

Clinical response to FVIII N N N N N N N N N N N N N N Y Y Y Y Y Y Y NA N N Y Y Y Y Y —

Baseline

VIII 2U mL 1

APCC 1U mL 1

APCC 2U mL 1

35 45 48 38 35 23 35 45 37 35 28 23 28 32 13 27 33 25 39 58 19 31 53 56 32 35 17 56 34 35.0  11.6

35 42 21 42 26 22 40 46 18 45 27 26 27 57 16 30 41 26 32 43 27 36 62 64 60 60 102 83 83 42.7  21.2

94 88 64 80 74 87 NA 78 52 64 61 63 59 52 41 96 103 NC 80 137 38 88 95 126 64 71 116 119 82 80.4  25.1

99 117 ES 125 ES 109 69 84 71 84 77 116 NC NC 65 ES ES 127 ES 144 87 ES ES ES 79 114 ES NC NC 97.9  24.1

Concordance clinical/in vitro response to FVIII N/N N/N N/N N/N N/N N/N N/N N/N N/N N/N N/N N/N N/N N/N Y/N Y/N Y/N Y/N Y/N Y/N Y/N NA/N N/Y N/Y Y/Y Y/Y Y/Y Y/Y Y/Y —

ETP, endogenous thrombin potential; FVIII, factor VIII; APCC, activated prothrombin complex concentrate; ES, ETP unavailable with the thrombinoscope software because of exhausted substrate due to consumption; NA, not available; N, no response; Y, responsive; NC, not carried out due to insufficient plasma. Mean values  1SD. *ETP expressed as percent of pooled normal plasma included in the corresponding assay.

Table 2. Severe hemophilia A patients with type II and III inhibitors: inhibitor titers, clinical response to FVIII, ETP values and concordance between clinical and in vitro response to FVIII. Endogenous thrombin potential* Patient

Inhibitor N-B U mL

HA-56 HA-40 HA-13 HA-30 HA-9 HA-36 HA-38 HA-20 HA-29 HA-6 HA-37 HA-5 HA-39 HA-49 † Mean  SD

91 60 7.3 7.2 6.0 5.0 5.0 5.0 38 21 9.5 6.4 6.3 6.2 —

1

Type

Clinical response to FVIII

II II II II II II II II III III III III III III —

Y Y Y Y Y Y Y NA NA Y N Y Y Y —

Baseline 49 54 47 36 23 49 53 56 50 32 56 25 43 50 44.5  11.1

VIII 2U mL

1

61 71 68 62 52 63 125 74 57 69 60 44 84 82 69.4  19.3

APCC 1U mL 83 101 106 101 51 64 72 91 55 63 83 67 48 102 77.6  20.3

1

APCC 2U mL ES 113 NC 146 69 93 88 94 66 94 113 98 76 NC 95.5  22.8

1

Concordance clinical/in vitro response to FVIII Y/Y Y/Y Y/Y Y/Y Y/N Y/Y Y/Y NA/Y NA/N Y/Y N/Y Y/N Y/Y Y/Y —

ETP, endogenous thrombin potential; FVIII: factor VIII; APCC, activated prothrombin complex concentrate; ES, ETP unavailable with the thrombinoscopesoftware because of exhausted substrate due to consumption; NA, not available; N, no response; Y, responsive; NC, not carried out due to insufficient plasma. Mean values  1 SD. *ETP expressed as percent of pooled normal plasma included in the corresponding assay. † Mean values  SD of type II and III inhibitors; there is no difference between them by ANOVA.

© 2013 John Wiley & Sons Ltd

Haemophilia (2014), 20, e7--e14

e12

 H. LUNA-Z AIZAR et al. (a)

(b)

(c)

Fig. 2. Thrombograms of patients’ plasmas after incubating with FVIII and APCC to evaluate in vitro response. (a) Patient HA-25 with 102 N– B U mL 1 of type I inhibitor: ETPFVIII 2U = 590 nM.min (35% of PNP). ETPAPCC 1U = 94% of PNP, ETPAPCC 2U = 99% of PNP. In vitro no response to FVIII agrees with the absence of clinical response to high doses of FVIII and in vitro response to APCC was excellent. (b) Patient HA-38 with 5.0 N–B U mL 1 of type II inhibitor: ETPFVIII 2U = 2300 nM.min (125% of PNP). ETPAPCC 1U = 1321 nM.min (72% of PNP), ETPAPCC 2U = 1630 nM.min of thrombin (88% of PNP). Clinically, the patient responded to high doses of FVIII. In this case, the optimum response was with FVIII. (c) Patient HA-56 with 91 N–B U mL 1 of type II inhibitor: ETPFVIII 2U = 1123 nM.min (61% of PNP); ETPAPCC 1U = 1529 nM.min (83% of PNP). ETPAPCC 2U = unavailable value, the thrombin curve did not return to baseline level because of substrate consumption; nevertheless, it is evident that response to APCC was >100%. The height of the peak was double that with 1 U mL 1. Clinically, the patient responded to high doses of FVIII but the response was optimal with 1 U mL 1 of APCC. The legend shows patient identification and conditions for the TGA: (………) PPP baseline; (_____) PPP incubated with FVIII 2 U mL 1; (- - - -) PPP incubated with APCC 1 U mL 1; ( )  PPP incubated with APCC 2 U mL 1. FVIII, factor VIII; APCC, activated prothrombin complex concentrate; ETP, endogenous thrombin potential; TGA, thrombin generation assay; PPP, platelet-poor plasma; PNP, pooled normal plasma.

clinical response to FVIII. This is the first study in Mexico that correlates objective parameters such as in vitro TG and type of inhibitor with clinical criteria of treatment response. Haemophilia (2014), 20, e7--e14

A recent study demonstrated the usefulness of in vitro and ex vivo TGA to predict the effectiveness of bypass therapies in patients undergoing elective surgery [21]. We use TGA to evaluate in vitro response to FVIII and APCC. The most useful parameter was the ETP reaching normalization in contrast to peak and rate (Fig. 3). Concentration of TF in the final mixture is a determinant factor of TG [18,25]; 1 pM of TF as used in this study improves test sensitivity; however, at this concentration of TF, the peak was not useful in determining TG capacity. This is in contrast to what was reported elsewhere [26,27], considering the normalization of thrombin peak as a positive response. In comparison, when ETP was ≥60%, we observed that the peak reached only ≥19% of the normal and TG continued over time with wide-flattened curves that did not reach the height of normal plasmas (see Fig. 2). This is explained because, in haemophilic plasmas, thrombin is generated at a lower rate than in normal plasmas. Correlation values between baseline ETP and ETP after spiking with both FVIII and APCC revealed a different behaviour of type I and type II/III inhibitors. Type I inhibitors had higher and significant correlation values, probably due to their homogeneous behaviour (refractory response to FVIII and good response to APCC) in contrast to the more heterogeneous pattern of type II/III inhibitors and without a sharp difference between baseline ETP and after FVIII (data not shown). In in vitro studies, Turecek et al. [26] obtained normalization of TGA parameters when using APCC concentration of 0.5–1 U mL 1 of plasma. We used APCC at 1–2 U mL 1 of plasma. In some cases, 2 U mL 1 APCC leads to a high TG. This causes substrate depletion (Fig. 2c), making the ETP calculation impossible as reported by others using a concentration >0.5 U mL 1 of APCC but made possible using 5 pM of TF [25]. For some plasmas, TG reached normal values with APCC 1 U mL 1 (equivalent to therapeutic doses of 65–100 IU kg 1 body weight) but caused substrate depletion with APCC at 2 U mL 1 of plasma (equivalent to high doses: 150–200 IU kg 1 body weight). This would indicate that APCC 2 U mL 1 was in excess to normalize TG. However, in most patients the TG was normalized at 2 U mL 1 of APCC (97%) (Tables 1 and 2), reinforcing the importance of individual therapeutic response evaluation by TGA to design individualized therapy in haemophilia patients without exposing them to thrombotic risk. It is known that lag time in patients with haemophilia is prolonged when contrasted with normal subjects, demonstrating a delay in TG initiation [18,25]. We observed that lag time was longer than baseline values in high-inhibitor plasmas incubated with FVIII, which did not occur in low-titre or negative-inhibitor © 2013 John Wiley & Sons Ltd

TREATMENT RESPONSE ASSESSED BY THROMBIN GENERATION

e13

Fig. 3. Mean values of thrombin generation assay parameters. Mean values of ETP, peak and rate of thrombin generation in inhibitor-positive PPP type I vs. II/III in baseline condition and incubated with FVIII and APCC. According to inhibitor type, significant differences were observed for response to FVIII in ETP (**P < 0.005) and peak (*P < 0.05) by Student t-test. Thrombin generation was normalized to 100% only for ETP but not for peak (60%) and rate (30%). ETP, endogenous thrombin potential; PPP, platelet-poor plasma; APCC, activated prothrombin complex concentrate.

Table 3. Response to FVIII evaluated by ETP and clinical criteria according to type of inhibitor. In vitro response*

Clinical response**

Inhibitor type

n

Yes

No

n

Yes

No

I II + III†

29 14

7 (24) 11 (79)

22 (76) 3 (21)

28 12

12 (43) 11 (92)

16 (57) 1 (8)

ETP, endogenous thrombin potential; FVIII, factor FVIII. Percentages by inhibitor type in parentheses. † No significant difference between type II and III inhibitors. Fisher’s exact test: *P = 0.0007; **P = 0.0049.

plasmas (data not shown). Longer lag time seems to be an inhibitory effect on FVIII:C. Otherwise, lag time was shortened in plasmas incubated with APCC according to its bypass effect and leading to normalization of TG parameters. This confirms the bypass agents as the first-line treatment in patients with inhibitors [21,28–30]. The inhibitory pattern on FVIII:C correlates with TG and the clinical response as previously observed (Table 3) [10,11]. Our results confirmed the usefulness of the kinetics inhibitor classification [10]. Most patients with type I inhibitor did not show clinical response to FVIII as replacement therapy in contrast to patients with type II inhibitors. Type III inhibitors showed a poor concordance between clinical and in vitro response, confirming their heterogeneity [31]. This study provides evidence of the strong correlation of both in vitro parameters to clinical response. The significant number of patients with severe HA with high-titre inhibitors that were studied strengthens the impact of this study. In addition, TG parameters are objective criteria to evaluate treatment response, which showed a significant prognostic value (70%). TG study may be limited because in vitro conditions do not reflect the physiological environment. However, in vitro TG has demonstrated to be a starting point for monitoring different clinical situations after © 2013 John Wiley & Sons Ltd

validation by ex vivo studies [21]. Standardized commercial reagents and dedicated software for automatic calculation of TG parameters decrease intra- and interassay variability, allowing a step towards the use of TGA in routine clinical practice [14–21,25–28,32].

Conclusion Thrombin generation assay predicts the response to FVIII or APCC therapy in patients with severe HA with high-titre inhibitors. Although bypass agents are the treatment of choice for inhibitors in haemophilia, the context of this study sustains TGA to evaluate the clinical response of Mexican patients to the available therapeutic resources. When quantitatively evaluating the capacity of thrombin formation in a plasma sample, TGA is a powerful tool that allows treatment individualization and optimization of therapeutic resources for bleeding events in inhibitor-positive patients and will improve overall cost-effectiveness in prophylactic treatments and surgery.

Acknowledgements Baxter Mexico S.A. de C.V. partially sponsored the project and provided support to Dr Luna-Zaizar and Dr Jaloma-Cruz for attending national and international scientific meetings to present the project. We thank the Federaci on de Hemofilia de la Rep ublica Mexicana, A.C. and the affiliated State Hemophilia Associations for their interest and valuable participation in the study. We are indebted to the haematologists from 25 institutions for the clinical evaluation of their patients: Aguilar G., Aguilar L.B., Berges A., Bernaldez R., Esparza M.A., Flores N., Garcıa J., Garibaldi R., Gonzalez M.C., Gonzalez M.G., Hernandez C., Hernandez M.P., Hernandez T., Lim on A., L opez C., Martınez M.C., Martınez M.A., Mejıa M., Miranda R., Novelo B., Olvera V., Perez J., Pompa M.T., Rangel B., Reyes U., Rodrıguez M.C., Soto J., Taboada C., Varela S., Velez M.A., Villegas H., Zaragoza V.M. and Zurita E. We thank Dr. Ana M. Contreras for valuable guide in the writing and review of the document. We appreciate the statistical advice of Ing. Rogelio Troyo-Sanroman. Sharon Morey, Executive Editor, Scientific Communications, assisted in the English editorial review. This study was supported by CONACyT SALUD-2003-C01-Sector 126 and BAXTER, Mexico.

Haemophilia (2014), 20, e7--e14

e14

 H. LUNA-Z AIZAR et al. designed the study, collected samples and clinical data, analysed data and wrote the paper.

Author contributions Hilda Luna designed the study, collected samples and clinical data, performed research, analysed data and wrote the paper. Claudia Beltran collected samples and clinical data. Marıa Amparo Esparza, Janet Soto, Adolfina Berges, Marıa del Carmen Rodrıguez and Marıa Teresa Pompa are paediatric haematologists, who collected clinical data. Ana Jaloma

References 1 Berges-Garcıa A, Velez-Marıa A, RıosOsuna MG et al. Estado actual y recomendaciones para la profilaxis de la hemofilia en Mexico. Rev Hematol Mex 2012; 13: 84–8. 2 Teitel J, Berntorp E, Collins P et al. A systematic approach to controlling problem bleeds in patients with severe congenital haemophilia A and high-titre inhibitors. Haemophilia 2007; 13: 256–63. 3 Gomperts ED, Astermark J, Gringeri A, Teitel J. From theory to practice: applying current clinical knowledge and treatment strategies to the care of hemophilia A patients with inhibitors. Blood Rev 2008; 22: S1–11. 4 Perez-Bianco R, Castro-Ozelo M, RibeiroVillacßa P et al. Diagnosis and treatment of congenital hemophilia with inhibitors. A Latin American perspective. Medicina (B Aires) 2008; 68: 227–42. 5 Martınez-Murillo C. Tratamiento profilactico en pacientes con hemofilia e inhibidores. Rev Hemo Trombo 2010; 3: 44–6. 6 Benıtez-Aranda H. Avances en el tratamiento de la hemofilia. Rev Med Inst Mex Seguro Soc 2005; 43: 135–8. 7 Esparza-Flores MA. Tratamiento preventivo y domiciliario de la hemofilia. Rev Med Inst Mex Seguro Soc 2005; 43: 139–41. 8 Biggs R, Austen DE, Denson KW, Borrett R, Rizza CR. The mode of action of antibodies which destroy factor VIII. Antibodies which give complex concentration graphs. Br J Haematol 1972; 23: 137–55. 9 Gawryl MS, Hoyer LW. Inactivation of factor VIII coagulant activity by two different types of human antibodies. Blood 1982; 60: 1103–9. 10 Gilles JG, Jacquemin MG, Saint-Remy JM. Factor VIII inhibitors. Thromb Haemost 1997; 78: 641–6. 11 Luna-Z aizar H, Esparza-Flores MA, L opez Guido B, Aguilar-L opez LB, Cortes-Alvarez CR, Jaloma-Cruz AR. Kinetics of factor VIII:C inhibitors and treatment response in

Haemophilia (2014), 20, e7--e14

12

13

14

15

16

17

18

19

20

21

Disclosures The authors declare that there are no conflicts of interest in relation to this study.

severe hemophilia A patients. Int J Lab Hematol 2009; 31: 673–82. Hemker HC, Giesen P, Al Dieri R et al. The calibrated automated thrombogram (CAT): a universal routine test for hyperand hypocoagulability. Pathophysiol Haemost Thromb 2002; 32: 249–53. Hemker HC, Giesen P, Al Dieri R et al. Calibrated automated thrombin generation measurement in clotting plasma. Pathophysiol Haemost Thromb 2003; 33: 4–15. Al Dieri R, Peyvandi F, Santagostino E et al. The thrombogram in rare inherited coagulation disorders: its relation to clinical bleeding. Thromb Haemost 2002; 88: 576– 82. Rugeri L, Beguin S, Hemker C et al. Thrombin-generating capacity in patients with von Willebrand’s disease. Haematologica 2007; 92: 1639–46. Tripodi A, Legnani C, Palareti G, Chantarangkul V, Mannucci PM. More on: high thrombin generation and the risk of recurrent venous thromboembolism. J Thromb Haemost 2009; 7: 906–7. Smid M, Dielis AW, Winkens M et al. Thrombin generation in patients with a first acute myocardial infarction. J Thromb Haemost 2011; 9: 450–6. Dargaud Y, Beguin S, Lienhart A et al. Evaluation of thrombin generating capacity in plasma from patients with haemophilia A and B. Thromb Haemost 2005; 93: 475– 80. Beltran-Miranda CP, Khan A, Jaloma-Cruz AR, Laffan M. Thrombin generation and phenotypic correlation in haemophilia A. Haemophilia 2005; 11: 326–34. Santagostino E, Mancuso ME, Tripodi A et al. Severe hemophilia with mild bleeding phenotype: molecular characterization and global coagulation profile. J Thromb Haemost 2010; 8: 737–43. Dargaud Y, Lienhart A, Negrier C. Prospective assessment of thrombin generation test for dose monitoring of bypassing therapy in hemophilia patients with inhibitors undergoing elective surgery. Blood 2010; 116: 5734–7.

22 Verbruggen B, Novakova I, Wessels H, Boezeman J, van den Berg M, Mauser-Bunschoten E. The Nijmegen modification of the Bethesda assay for factor VIII:C inhibitors: improved specificity and reliability. Thromb Haemost 1995; 73: 247–51. 23 Kasper CK, Aledort L, Aronson D et al. Proceedings: a more uniform measurement of factor VIII inhibitors. Thromb Diath Haemorrh 1975; 34: 869–72. 24 Ling M, Duncan EM, Rodgers SE et al. Classification of the kinetics of factor VIII inhibitors in haemophilia A: plasma dilution studies are more discriminatory than time-course studies. Br J Haematol 2001; 114: 861–7. 25 van Veen JJ, Gatt A, Bowyer AE, Cooper PC, Kitchen S, Makris M. The effect of tissue factor concentration on calibrated automated thrombography in the presence of inhibitor bypass agents. Int J Lab Hematol 2009; 31: 189–98. 26 Turecek PL, Varadi K, Keil B et al. Factor VIII inhibitor-bypassing agents act by inducing thrombin generation and can be monitored by a thrombin generation assay. Pathophysiol Haemost Thromb 2003; 33: 16–22. 27 Varadi H, Negrier C, Berntorp E et al. Monitoring the bioavailability of FEIBA with a thrombin generation assay. J Thromb Haemost 2003; 1: 2374–80. 28 Dargaud Y, Negrier C. Thrombin generation testing in haemophilia comprehensive care centres. Haemophilia 2010; 16: 223– 30. 29 Hoffman M, Dargaud Y. Mechanisms and monitoring of bypassing agent therapy. J Thromb Haemost 2012; 10: 1478–85. 30 Makris M, Hay CRM, Gringeri A, D’oiron R. How I treat inhibitors in haemophilia. Haemophilia 2012; 18: 48–53. 31 Lollar P. Pathogenic antibodies to coagulation factors. I: factor VIII and factor IX. J Thromb Haemost 2004; 2: 1082–95. 32 van Veen JJ, Gatt A, Makris M. Thrombin generation testing in routine clinical practice: are we there yet? Br J Haematol 2008; 142: 889–903.

© 2013 John Wiley & Sons Ltd