Oneyear maintenance outcomes among ... - Wiley Online Library

Alimentary Pharmacology and Therapeutics

One-year maintenance outcomes among patients with moderatelyto-severely active ulcerative colitis who responded to induction therapy with adalimumab: subgroup analyses from ULTRA 2 W. J. Sandborn*, J.-F. Colombel†, G. D’Haens‡,§, G. Van Assche¶,**, D. Wolf††, M. Kron‡‡, A. Lazar‡‡, A. M. Robinson§§, M. Yang§§, J. D. Chao§§ & R. Thakkar§§

*University of California San Diego, La Jolla, CA, USA. † Centre Hospitalier Universitaire de Lille, Lille, France. ‡ Academic Medical Center, Amsterdam, the Netherlands. § Imelda GI Clinical Research Center, Bonheiden, Belgium. ¶ Univeristy of Toronto, Toronto, ON, Canada. **University of Leuven, Leuven, Belgium. †† Atlanta Gastroenterology Associates, Atlanta, GA, USA. ‡‡ Abbott GmbH & Co. KG, Ludwigshafen, Germany. §§ Abbott Laboratories, Abbott Park, IL, USA.

Correspondence to: Dr W. J. Sandborn, UCSD Inflammatory Bowel Disease Center, Division of Gastroenterology, UC San Diego Health System, 9500 Gilman Drive, La Jolla, CA 92093-0956, USA. E-mail: [email protected]

Publication data Submitted 26 September 2012 First decision 3 October 2012 Resubmitted 26 October 2012 Accepted 27 October 2012 EV Pub Online 23 November 2012

SUMMARY Background Patients with moderately-to-severely active ulcerative colitis (UC) are unlikely to continue anti-TNF therapy in the absence of early therapeutic response. Aim To assess week 52 efficacy, safety and benefit/risk balance of adalimumab treatment in patients with moderately-to-severely active UC failing conventional therapy who achieved clinical response at week 8 in the 52-week ULTRA 2 trial. Methods Patients randomised to adalimumab (160/80 mg, week 0/2; 40 mg, every other week thereafter) in ULTRA 2 who achieved clinical response at week 8 per partial Mayo score (Mayo score without endoscopy subscore) were assessed for week 52 clinical remission, clinical response, mucosal healing, steroid-free remission and steroid discontinuation rates, overall and by prior anti-TNF use. Benefit/risk balance for the overall ITT population (regardless of week 8 responder status) was assessed using ‘net efficacy adjusted for risk’ (NEAR) odds ratios. Safety was assessed using adverse event rates. Results Of 248 adalimumab-treated patients, 123 (49.6%) achieved clinical response at week 8. Of these, 30.9%, 49.6%, and 43.1% achieved clinical remission, clinical response, and mucosal healing, respectively, at week 52. Of the week 8 responders using corticosteroids at baseline (N = 90), 21.1% achieved steroid-free remission and 37.8% were steroid-free at week 52. NEAR odds ratios indicated a positive benefit/risk balance for achievement of week 8 and week 52 response or remission without serious adverse events or serious infections. No safety concerns were identified. Conclusions Adalimumab treatment was associated with a positive benefit/risk balance in the overall population of patients with moderately-to-severely active ulcerative colitis in ULTRA 2; early response was predictive of a positive outcome at 1 year (NCT00408629). Aliment Pharmacol Ther 2013; 37: 204–213

204

ª 2012 Blackwell Publishing Ltd doi:10.1111/apt.12145

Week 52 efficacy for responders to adalimumab induction in UC INTRODUCTION Ulcerative colitis (UC) is a chronic inflammatory disease affecting the colonic mucosa.1 The goal of medical therapy for this condition is to achieve remission, improve quality of life and prevent colectomy.2 Treatments for ulcerative colitis are limited and standard medical therapies, including aminosalicylates, corticosteroids and immunosuppressants, inadequately control the disease in a substantial proportion of patients. The approval of infliximab, an infused chimeric monoclonal antitumour necrosis factor (TNF) antibody, in 2005 provided an additional option for treatment of ulcerative colitis.3 The fully human anti-TNF antibody adalimumab has been shown in randomised, controlled clinical trials to induce4, 5 and maintain5 remission in patients with moderately-to-severely active ulcerative colitis. Despite the demonstrated efficacy of anti-TNF agents in treating patients with ulcerative colitis, as with any drug, not all patients will respond. In a real-world clinical situation, physicians are unlikely to continue antiTNF therapy in patients who do not demonstrate an early response, due to concerns about exposure to potential adverse events in the face of limited benefit and the cost of therapy. In clinical practice, response to treatment for ulcerative colitis can be assessed using the factors that are captured in the Mayo score. The Mayo score is a composite index of ulcerative colitis activity, which captures stool frequency, severity of rectal bleeding, patient status as evaluated by physician global assessment, and degree of mucosal ulcers as assessed by endoscopy.6 Partial Mayo score, which excludes the endoscopy subscore, has been reported to correlate as strongly as the full Mayo score with patient-perceived clinical status.7 A restricted analysis of the efficacy and safety outcomes among patients who demonstrate early response to anti-TNF agents in clinical trials (as measured by full or partial Mayo score) is an approach that can be used to model treatment outcomes in the real world. The assessment of a therapy’s benefit/risk profile is an important consideration in the selection of a treatment for an individual patient, but there is no standard approach to quantifying benefit/risk balance. One approach that has been advocated is the calculation of ‘number needed to treat’ (to characterise benefit) and ‘number needed to harm’ (to characterise risk)8; however, with this approach, benefit and risk are expressed separately. The Net Efficacy Adjusted for Risk, or NEAR, method is a recently suggested approach allowing a simple assessment of risk and benefit in a single measure.9 To date, NEAR methodology has not been widely adopted, but was used as part of a recent Cochrane systematic review of opioid use in rheumatoid arthritis.10, 11 The NEAR odds ratio (OR) provides Aliment Pharmacol Ther 2013; 37: 204-213 ª 2012 Blackwell Publishing Ltd

additional information over other benefit–risk analyses, in that it expresses the likelihood of a patient to achieve treatment success without experiencing a specific adverse event compared with a control. We performed a post hoc restricted analysis of data from the ULTRA 25 adalimumab induction and maintenance trial to describe efficacy and safety results at 1 year of treatment for patients with moderately-toseverely active ulcerative colitis unresponsive to standard therapy who responded to 8 weeks of induction therapy with adalimumab. In addition, we quantified the benefit/ risk of adalimumab treatment of ulcerative colitis, using the NEAR method, for induction and maintenance of remission and response in all patients in ULTRA 2.

METHODS ULTRA 2 (NCT00408629) was a multicentre, Phase 3, randomised, double-blind, placebo-controlled trial to evaluate the efficacy of adalimumab as induction and maintenance therapy for clinical remission in patients with moderately-to-severely active ulcerative colitis. The methods and results have been previously published.5 Briefly, patients with a Mayo score of 6–12 and endoscopy subscore  2 despite current or prior therapy with corticosteroids and/or immunosuppressants were included. Patients with prior anti-TNF exposure were eligible for inclusion in ULTRA 2. Patients were randomly assigned (1:1) to the addition of subcutaneous adalimumab (160 mg at week 0, 80 mg at week 2, then 40 mg every other week beginning at week 4) or placebo to current ulcerative colitis therapy through week 52. Randomisation was stratified by prior anti-TNF use. Concomitant medication doses remained stable, except corticosteroids, which could be tapered at the investigator’s discretion after week 8.5 Patients with inadequate response could move to open-label adalimumab (40 mg every other week) starting at week 12. Patients with continued inadequate response while on open-label adalimumab 40 mg every other week were permitted to move to open-label adalimumab 40 mg weekly. Full Mayo score was assessed at weeks 0, 8, 32 and 52 (or early discontinuation visit); partial Mayo score (Mayo score without the endoscopy subscore) was determined at all study visits (weeks 0, 2, 4, 8, 12, 16, 20, 26, 32, 38, 44 and 52/early discontinuation). Clinical remission per full Mayo score was defined as a Mayo score  2 with no individual subscore >1. Clinical response per full Mayo score was defined as a Mayo score decrease of  3 points and at least 30% from baseline, with a decrease in the rectal bleeding subscore of  1 or absolute rectal 205

W. J. Sandborn et al. bleeding subscore of 1 or 0. Clinical response per partial Mayo score was defined as a decrease in the partial Mayo score of  2 points and  30% from baseline, with the same rectal bleeding subscore criteria outlined above. Rectal bleeding and stool frequency subscores were determined conservatively, using the worst assessment from the 3 days prior to the study visit.5 The present analysis included patients from the intent-to-treat (ITT) population of ULTRA 2. Endpoints for induction included the proportion of randomised patients who achieved clinical response per partial Mayo score at weeks 2, 4, and 8, and per full Mayo score at week 8, and the proportion of adalimumab-treated patients using corticosteroids at baseline who responded at week 8 per full or partial Mayo score. Endpoints at week 52 included clinical remission (per full Mayo score), clinical response (per full Mayo score) and mucosal healing (endoscopy subscore of 1 or 0) for adalimumab-treated patients who achieved clinical response per full or partial Mayo score at week 8 (‘week 8 responders’), and clinical remission and clinical response in adalimumab-treated patients who did not achieve clinical response at week 8 (‘nonresponders’). Steroid-free clinical remission (per full Mayo score) at week 52 and steroid discontinuation by week 52 were assessed in the adalimumab-treated week 8 responders who used corticosteroids at baseline of ULTRA 2. Additional subgroup analyses by prior anti-TNF use among adalimumab responders for all endpoints at week 52 were also performed. Previously reported results from the full adalimumab-treated population5 are provided for reference.

Statistical methods For binary endpoints, nonresponder imputation (NRI) was used in which nonremission or nonresponse was assigned to patients with missing data or those receiving openlabel dosing at the time point evaluated. Additional analyses for week 52 endpoints for the adalimumab treatment group used modified NRI (mNRI), in which only missing data were imputed as nonresponse or nonremission (i.e. efficacy data for patients receiving open-label every other week or weekly therapy was included in the numerator). The Cochran-MantelHaenszel test was used for comparisons between adalimumab and placebo treatment groups (stratifying for prior anti-TNF use) for full and partial Mayo score response rates in the ITT set. Safety Adverse events were recorded to assess safety through week 8 for the full ITT population, and through week 52 206

for week 8 responders in the adalimumab-treated cohort and all patients in the placebo group. Safety in the full ITT population through week 52 has been reported.5

Benefit/risk assessment The benefit-to-risk balance of adalimumab treatment was measured using the NEAR method, as described by Boada et al.,9 for the likelihood of achieving efficacy free of serious adverse events or serious infections (which are a subset of serious adverse events) in the ITT population. NEAR odds ratios (OR, adalimumab/placebo) for efficacy free of serious adverse events or serious infections were calculated using actual (not estimated) data: one each for week 8 clinical response (per partial Mayo score) and clinical remission (per full Mayo score) after adalimumab induction, and one each for clinical response (per full Mayo score) and clinical remission (per full Mayo score) at week 52 during maintenance treatment. Partial Mayo response at week 8 was used because early patient evaluation after initiation of a therapeutic intervention is often made on the basis of symptom assessment. As required for the NEAR method, the observed numbers of patients experiencing four possible combinations of safety and efficacy outcomes were recorded for adalimumab- and placebo-treated patients for each adverse event of interest: (a) AE-free treatment success (best outcome), (b) no treatment success and no AE (‘neutral’ outcome), (c) treatment success with an AE (‘neutral’ outcome) and (d) no success with an AE (worst outcome).9 For the observed frequency calculations, NRI was used for efficacy determinations, whereby patients who had missing data or who moved to open-label therapy were considered not to have efficacy at the time point evaluated. AE determinations were based on occurrence of the AE being evaluated during the double-blind period of ULTRA 2. The odds of achieving AEfree efficacy (clinical remission or clinical response) were calculated for adalimumab- and placebo-treated patients by dividing the frequency of patients with AE-free treatment success (a) (the best outcome) by (b+c+d) (the sum of the frequencies for all other outcomes). NEAR odds ratios were calculated by dividing the odds of AE-free efficacy for adalimumab-treated patients by the odds for placebo-treated patients.9 RESULTS Patient demographics and clinical characteristics As previously reported, the baseline characteristics were similar in the overall population of adalimumab- and placebo-treated patients in ULTRA 25 (Table 1). Aliment Pharmacol Ther 2013; 37: 204-213 ª 2012 Blackwell Publishing Ltd

Week 52 efficacy for responders to adalimumab induction in UC Table 1 | Baseline demographics and clinical characteristics* (first 3 columns of the table are reprinted with permission from Elsevier)

Characteristics Male: n (%) Age, years: mean  s.d. Weight, kg: mean  s.d. Disease location: n (%) Pancolitis Descending colon Other Disease duration, years: mean  s.d. High sensitivity CRP, mg/L Mean  s.d. Median >ULN = 4.94mg/L†: n (%) Mayo score: mean  s.d. Partial Mayo score‡ Endoscopy subscore§ Rectal bleeding subscore‡ PGA subscore‡ Stool frequency subscore Medication at baseline: n (%) Corticosteroids Azathioprine/6-MP Aminosalicylates¶ Azathioprine/6-MP or steroids Azathioprine/6-MP and steroids Prior anti-TNF therapy

Placebo N = 246

Adalimumab N = 248

Adalimumab week 8 responders per partial Mayo score N = 123

152 (61.8) 41.3  13.22 77.1  17.31

142 (57.3) 39.6  12.47 75.3  17.71

72 (58.5) 38.8  11.91 75.0  17.99

120 (48.8) 96 (39.0) 30 (12.2) 8.5  7.37

120 (48.4) 96 (38.7) 32 (12.9) 8.1  7.09

66 (53.7) 40 (32.5) 17 (13.8) 8.5  7.40

13.1  36.78 4.2 116 (47.2) 8.9  1.75 6.5  1.55 2.5  0.50 1.7  0.94 2.2  0.57 2.6  0.66

14.5  32.07 4.1 113 (45.7) 8.9  1.50 6.5  1.39 2.5  0.50 1.7  0.85 2.2  0.55 2.5  0.71

9.0  18.88 2.8 44 (35.8) 8.7  1.54 6.3  1.43 2.4  0.49 1.7  0.84 2.3  0.56 2.3  0.78

140 (56.9) 80 (32.5) 155 (63.0) 175 (71.1) 45 (18.3) 101 (41.1)

150 (60.5) 93 (37.5) 146 (58.9) 193 (77.8) 50 (20.2) 98 (39.1)

90 (73.2) 47 (38.2) 78 (63.4) 104 (84.6) 33 (26.8) 35 (28.5)

ITT, intent-to-treat; PGA, Physician’s Global Assessment; MP, mercaptopurine; ULN, upper limit of normal. * 24 patient (14 placebo + 10 adalimumab) were excluded due to site noncompliance. † N = 246, placebo; N = 247, adalimumab; N = 493, total. ‡ N = 245, placebo; N = 247, adalimumab; N = 492, total. § N = 247, adalimumab; N = 493, total. ¶ Includes mesalazine (mesalamine), sulphasalazine, balsalazide, aminosalicylic acid, olsalazine.

Early clinical response Clinical response per partial Mayo score was statistically significantly greater in adalimumab-treated patients compared with placebo-treated patients as early as week 2. The proportion of patients achieving clinical response per partial Mayo score increased from week 2 to week 8 in both adalimumab- and placebo-treated patients (Figure 1); statistically significant differences between the adalimumab and placebo groups were observed at all time points. At week 8, of the 248 patients randomised to adalimumab treatment, 123 (49.6%) achieved clinical response per partial Mayo score, and, as previously reported, 125 (50.4%) achieved clinical response per full Mayo score.5 There was high concordance between the patients identified as having clinical response whether the full or partial Mayo score definitions were used; 117 Aliment Pharmacol Ther 2013; 37: 204-213 ª 2012 Blackwell Publishing Ltd

(93.6% and 95.1% respectively) of full or partial Mayo score responders achieved clinical response per both definitions. Most baseline characteristics for the week 8 responders per partial Mayo score (and full Mayo score; data not shown) were similar to those for the full adalimumab treatment group (Table 1), although the responder group had a lower median CRP concentration, a higher percentage of pancolitis, a higher rate of concomitant steroid use, and a lower rate of use of antiTNF agents in the past.

Clinical remission and clinical response at week 52 When assessed using NRI, 30.9% of adalimumab-treated week 8 responders (by partial Mayo score) achieved clinical remission per full Mayo score at week 52 (Figure 2a); the corresponding value for nonresponders was 207

W. J. Sandborn et al.

Figure 1 | Response rates through week 8 per partial Mayo score (PMS, weeks 2, 4 and 8) and full Mayo score (FMS, week 8 only).

Figure 2 | Clinical remission per full Mayo score at week 52 for adalimumab week 8 responders by partial Mayo score (NRI and mNRI) and all adalimumab-treated patients5 (NRI), (a) overall, and (b) by prior anti-TNF use. Clinical response per full Mayo score at week 52 for adalimumab week 8 responders by partial Mayo score (NRI and mNRI) and all adalimumab-treated patients5 (NRI), (c) overall, and (d) by prior anti-TNF use.

4.0% (Table S1). When remission occurring during open-label adalimumab treatment was included (mNRI), the 1-year remission rate was 37.4% for the week 8 responders (Figure 2a) and 8.0% for the nonresponders (Table S1). For comparison, 17.3% of adalimumab patients in the full ITT population achieved clinical remission (NRI).5 Patients na€ıve to anti-TNF agents tended to have higher remission rates than those with 208

prior anti-TNF exposure (Figure 2b). Similar patterns were observed for clinical response at week 52 (Figure 2c, d). Results for patients who responded at week 8 by full Mayo score were similar (Table S2).

Mucosal healing Over 40% of adalimumab-treated patients with clinical response at week 8 (by partial Mayo score) experienced Aliment Pharmacol Ther 2013; 37: 204-213 ª 2012 Blackwell Publishing Ltd

Week 52 efficacy for responders to adalimumab induction in UC mucosal healing at week 52 (Table 2). Week 8 responders who were na€ıve to prior anti-TNF treatment tended to have higher rates of mucosal healing than anti-TNFexposed patients (Table 2). Results for patients with week 8 response by full Mayo score were similar.

Steroid-free remission and discontinuation of steroid use at week 52 Of 150 adalimumab-treated patients using corticosteroids at baseline of ULTRA 2, 90 (60%) responded per full Mayo score, and 90 (60%) responded per partial Mayo score at week 8. In the placebo group, 140 patients used corticosteroids at baseline, and 51 (36.4%, P < 0.001) and 53 (37.9%, P < 0.001) responded at week 8 per full and partial Mayo scores respectively. The rates of steroid-free remission and steroid discontinuation at week 52 for adalimumab-treated week 8 responders (per partial Mayo score) are shown in Table 2. Similar results were observed for anti-TNF-na€ıve and -exposed patients. For reference, the rates of steroid discontinuation and steroid-free remission at week 52 for adalimumab-treated patients in the ITT population, as reported previously,5 are included in the Table. Results for week 8 responders by full Mayo score were similar. Safety No safety concerns were identified in the ITT population during the double-blind induction phase from baseline to week 8 (Table 3). In the double-blind maintenance phase,

adalimumab week 8 responders experienced fewer adverse events overall than the ITT placebo-treated patients, and fewer severe or serious adverse events, or events leading to discontinuation of the study drug. Adalimumab week 8 responders were more likely than the ITT placebo-treated patients to experience an opportunistic infection (all opportunistic infections in the adalimumab treatment group were nonserious local candidiasis), although the incidence of opportunistic infections was low overall (Table 4). There were no cases of demyelinating disease and no deaths were recorded in ULTRA 2.

Benefit/risk analysis In the ITT population, NEAR ORs indicated a significantly greater likelihood of achieving clinical remission or clinical response at week 8 (ORs: 1.8–2.0, Figure 3a) and clinical remission or clinical response at week 52 (ORs: 1.8–2.2, Figure 3b) free of serious adverse events or serious infections with adalimumab vs. placebo. DISCUSSION In this population of patients with moderately-toseverely active UC who had failed or were intolerant to immunosuppressants and/or corticosteroids, including patients with prior anti-TNF therapy, adalimumab treatment led to statistically significant improvements in disease activity, as assessed by clinical response, compared with placebo as early as week 2 (Figure 1). By week 8, approximately one-half of adalimumab-treated patients

Table 2 | Percentage of patients with mucosal healing at week 52 (ITT population and week 8 responders per partial Mayo score) and percentage of patients in steroid-free remission or with steroid discontinuation at week 52 (patients using corticosteroids at baseline, ITT population and week 8 responders per partial Mayo score), in the overall population and by prior anti-TNF exposure (NRI and mNRI) ITT

Mucosal healing, N Mucosal healing (NRI), n (%) Mucosal healing (mNRI), n (%) Baseline steroid users, N Steroid-free remission (NRI), n (%) Steroid-free remission (mNRI), n (%) Steroid discontinuation (NRI), n (%) Steroid discontinuation (mNRI), n (%)

Week 8 responders

ADA-treated

Anti-TNF-naïve

Anti-TNF-exposed

ADA-treated

Anti-TNF-naïve

Anti-TNF-exposed

248 62 (25.0)5

150 47 (31.3)5

98 15 (15.3)5

123 53 (43.1)

88 41 (46.6)

35 12 (34.3)

103 (41.5)

72 (48.0)

31 (31.6)

70 (56.9)

54 (61.4)

16 (45.7)

150

110

40

90

68

22

20 (13.3)5

15 (13.6)5

5 (12.5)5

19 (21.1)

14 (20.6)

5 (22.7)

25 (16.7)

19 (17.3)

6 (15.0)

24 (26.7)

18 (26.5)

6 (27.3)

36 (24.0)5

26 (23.6)

10 (25.0)

34 (37.8)

24 (33.3)

10 (45.5)

56 (37.3)

43 (39.1)

13 (32.5)

46 (51.1)

33 (48.5)

13 (59.1)

Aliment Pharmacol Ther 2013; 37: 204-213 ª 2012 Blackwell Publishing Ltd

209

W. J. Sandborn et al. Table 3 | Number of patients with treatment-emergent adverse events recorded in the double-blind induction phase (baseline to week 8); ITT

Any AE Any AE at least possibly drug-related Any severe AE Any serious AE Any AE leading to discontinuation of study drug Any infectious AE Any serious infection Any opportunistic infection† (excluding TB) Any allergic reaction Any injection site pain Any lupus-like syndrome Any malignancy Any nonmelanoma skin cancers Any haematology-related event Deaths

Placebo N = 246 n (%)

Adalimumab N = 247* n (%)

163 (66.3) 61 (24.8)

144 (58.3) 60 (24.3)

23 (9.3) 21 (8.5) 18 (7.3)

16 (6.5) 15 (6.1) 10 (4.0)

51 (20.7) 3 (1.2) 2 (0.8)

50 (20.2) 3 (1.2) 4 (1.6)

0 5 (2.0) 0 0 0 0 0

2 (0.8) 8 (3.2) 0 1 (0.4) 1 (0.4) 4 (1.6) 0

AE, adverse event; TB, tuberculosis. * One patient randomised to the adalimumab group did not receive adalimumab, so was not counted in the safety population, but was included for efficacy analyses. † All opportunistic infections were nonserious local candidiasis, except serious CMV colitis in a placebo patient.

achieved clinical response, measured by either full or partial Mayo score. Sustained efficacy was observed in the week 8 responders, even in patients who had previously failed anti-TNF therapy, and 1-year outcomes were superior in week 8 responders compared with the overall ITT population. Efficacy with adalimumab treatment was achieved in spite of reduction in concomitant therapy for ulcerative colitis; greater than one-third of the adalimumab-treated week 8 responders taking corticosteroid therapy at baseline had discontinued steroids at week 52. The overall adverse event profile was similar between adalimumab-treated patients and placebo-treated patients, and was consistent with the known profile of the drug across other indications. A positive benefit/risk balance was observed during induction and maintenance treatment in all patients. Patients with moderately-to-severely active UC who fail treatment with aminosalicylates have limited options for controlling their disease. Corticosteroids are effective for induction of remission in UC but not for maintenance therapy,2 and are associated with adverse effects 210

Table 4 | Treatment-emergent adverse event rates in the double-blind phase (baseline to end of double-blind treatment), adalimumab week 8 responders by partial Mayo score and all placebo patients

Any AE Any AE at least possibly drug-related Any severe AE Any serious AE Any AE leading to discontinuation Any infectious AE Any serious infection Any opportunistic infection* (excluding TB) Any allergic reaction Any injection site pain Any lupus-like syndrome Any malignancy Any nonmelanoma skin cancer Any haematology-related event Deaths

Placebo N = 246 PYs = 114.3 E (E/100PY)

Adalimumab, week 8 PMS responders N = 123 PYs = 93.7 E (E/100PY)

968 (846.9) 200 (175.0)

700 (747.1) 178 (190.0)

49 (42.9) 42 (36.7) 44 (38.5)

22 (23.5) 21 (22.4) 9 (9.6)

170 (148.7) 7 (6.1) 2 (1.7)

145 (154.8) 2 (2.1) 6 (6.4)

1 (0.9) 11 (9.6) 0 0 0

3 (3.2) 6 (6.4) 1 (1.1) 1 (1.1) 0

0

2 (2.1)

0

0

AE, adverse event; TB, tuberculosis. * All opportunistic infections were nonserious local candidiasis, except serious CMV colitis in a placebo patient.

involving multiple organ systems that limit their utility. Immunosuppressants are commonly used for maintenance therapy, as suggested by treatment guidelines,12 but are not approved for treating moderately-to-severely active UC, are limited by their slow onset of action (3– 4 months),2 and are associated with increased risk of neoplasms, infections and hepatotoxicity.13–15 Colectomy may be an option for controlling UC, but has associated risks and limitations, including post-operative mortality, frequent defecation (including at night), faecal incontinence, risk of pouchitis and female infertility.16–18 The majority of patients with clinical response as assessed by full Mayo score also met the criteria for partial Mayo score response. This finding is consistent with the report of a high correlation between full and partial Mayo score as reported by other investigators.7 Patients with week 8 response by either measure had similar results for all endpoints studied, suggesting that in clinical practice, early assessment of response based on Aliment Pharmacol Ther 2013; 37: 204-213 ª 2012 Blackwell Publishing Ltd

Week 52 efficacy for responders to adalimumab induction in UC

Figure 3 | NEAR ORs for clinical remission and clinical response free of SAE and serious infections (a) at week 8, and (b) at week 52. ITT population (NRI).

patient reports of stool frequency and rectal bleeding and the physician’s assessment of the patient’s status may be sufficient to identify patients with early response to anti-TNF therapy. The full ULTRA 2 results demonstrated a positive benefit/risk balance, whereby adalimumab-treated patients achieved greater efficacy than placebo-treated patients at week 8 and week 52, and safety similar to placebo for up to 1 year of treatment. A positive benefit/ risk balance was also documented in the ITT population through reduction in all-cause and UC-related hospitalisation19 in all patients, regardless of early response. AllAliment Pharmacol Ther 2013; 37: 204-213 ª 2012 Blackwell Publishing Ltd

cause hospitalisation is an objective indicator of both benefit and risk, as patients can be hospitalised for lack of efficacy due to their underlying disease, as well as safety issues arising from treatment. The present analysis further documents the positive benefit/risk balance of adalimumab treatment for UC in the overall population for up to 52 weeks, and supports initiating adalimumab treatment in patients who are candidates for anti-TNF therapy for UC for up to 8 weeks, then assessing response at week 8 to determine which patients should be considered for long-term therapy. No new safety signals were identified during the initial 8 weeks of adalimumab treatment compared with placebo among patients with ulcerative colitis who were receiving baseline treatment with standard UC therapy consisting of aminosalicylates, corticosteroids and thiopurines. The NEAR analysis provides additional insights into the benefit/risk profile of a treatment, in that it estimates the occurrence of AEs by efficacy status. A treatment would be less attractive if AEs occurred more frequently in the patients who experienced treatment success than those without efficacy. The NEAR results demonstrated that adalimumab-treated patients were more likely to achieve response free of serious AEs or serious infections at week 8 in the ITT population compared with placebo. NEAR ORs for maintenance adalimumab therapy in all patients were similar to those observed after induction therapy. Given the greater efficacy observed at week 52 in adalimumab week 8 responders, the ORs for maintenance therapy in this group of patients would probably be even greater, although the study design of ULTRA 2 (which did not re-randomise early responders to treatment continuation or placebo) does not allow for this type of analysis as there is no placebo group reflecting withdrawal of adalimumab therapy. Adalimumab and other anti-TNF therapies have been associated with adverse events including infections and, rarely, malignancies.20 The safety results in the overall adalimumab ulcerative colitis development program4, 5 demonstrated a similar pattern of adverse events as has been observed in clinical trials in other inflammatory disease states, including Crohn’s disease.21 The safety findings in ulcerative colitis should be considered in the context of the study design, in which adalimumab or placebo was added to baseline ulcerative colitis therapy, which included steroids and/or thiopurines in a majority of patients. Thus, ‘placebo’ does not represent ‘no therapy.’ The favourable adverse event profile observed for the adalimumab week 8 responders compared with that 211

W. J. Sandborn et al. observed for patients randomised to placebo may be in part explained by the greater rate of steroid discontinuation in the adalimumab group compared with the placebo group (which was 22.9% at week 52, as previously reported5). The fact that the study design did not allow re-randomisation of patients to continue or withdraw adalimumab according to week 8 response is a limitation. Nevertheless, an analysis of outcomes for patients who responded at week 8 provides interpretable results and is consistent with clinical practice and UC treatment guidelines. According to the European Crohn’s and Colitis Organisation guideline for UC, response to initial therapy should be assessed within ‘several weeks’, and maintenance therapy is recommended after ‘successful medical treatment of active disease.’12 In conclusion, our analyses of the results for patients with moderately-to-severely active UC demonstrate a positive benefit/risk balance for adalimumab treatment. Early response as assessed by either full or partial Mayo score is predictive of subsequent efficacy, and our results indicate that treating early responders might further enhance the benefit/risk profile of adalimumab.

AUTHORSHIP Guarantor of the article: William J. Sandborn. Author contributions: All authors contributed to the design of the study. WJS, JFC, GD, GVA and DW collected data. MK performed statistical analyses. All authors contributed to interpretation of the data and preparation of the manuscript. All authors and Abbott approved the final version of the manuscript; the authors maintained control over the final content. ACKNOWLEDGEMENTS Declaration of personal interests: WJS: Consultant: Abbott Laboratories, ActoGeniX NV, AGI Therapeutics, Inc., Alba Therapeutics Corporation, Albireo, Alfa Wasserman, Amgen, AM-Pharma BV, Anaphore, Astellas Pharma, Athersys, Inc., Atlantic Healthcare Limited, Axcan Pharma (now Aptalis), BioBalance Corporation, Boehringer-Ingelheim Inc, Bristol Meyers Squibb, Celgene, Celek Pharmaceuticals, Cellerix SL, Cerimon Pharmaceuticals, ChemoCentryx, CoMentis, Cosmo Technologies, Coronado Biosciences, Cytokine Pharmasciences, Eagle Pharmaceuticals, Eisai Medical Research Inc., Elan Pharmaceuticals, EnGene, Inc., Eli Lilly, Enteromedics, Exagen Diagnostics, Inc., Ferring Pharmaceuticals, Flexion Therapeutics, Inc., Funxional Therapeutics Limited, Genzyme Corporation, Genentech (now Roche), Gilead Sciences, Given Imaging, GlaxoSmithKline, Human Genome Sciences, Ironwood Pharmaceuticals (previously Microbia 212

Inc.), Janssen (previously Centocor), KaloBios Pharmaceuticals, Inc., Lexicon Pharmaceuticals, Lycera Corporation, Meda Pharmaceuticals (previously Alaven Pharmaceuticals), Merck Research Laboratories, MerckSerono, Millennium Pharmaceuticals (subsequently merged with Takeda), Nisshin Kyorin Pharmaceuticals Co., Ltd., Novo Nordisk A/S, NPS Pharmaceuticals, Optimer Pharmaceuticals, Orexigen Therapeutics, Inc., PDL Biopharma, Pfizer, Procter and Gamble, Prometheus Laboratories, ProtAb Limited, Purgenesis Technologies, Inc., Receptos, Relypsa, Inc., Salient Pharmaceuticals, Salix Pharmaceuticals, Inc., Santarus, Schering Plough Corporation (acquired by Merck), Shire Pharmaceuticals, Sigmoid Pharma Limited, Sirtris Pharmaceuticals, Inc. (a GSK company), S.L.A. Pharma (UK) Limited, Targacept, Teva Pharmaceuticals, Therakos, Tillotts Pharma AG (acquired by Zeria Pharmaceutical Co., Ltd), TxCell SA, UCB Pharma, Viamet Pharmaceuticals, Vascular Biogenics Limited (VBL), Warner Chilcott UK Limited, Wyeth (now Pfizer). Speakers fees: Abbott Laboratories, Bristol Meyers Squibb, and Janssen (previously Centocor). Financial support for research: Abbott Laboratories, Bristol Meyers Squibb, Genentech, Glaxo Smith Kline, Janssen (previously Centocor), Millennium Pharmaceuticals (now Takeda), Novartis, Pfizer, Procter and Gamble Pharmaceuticals, Shire Pharmaceuticals, and UCB Pharma. J-FC: Consultant: Abbott Laboratories, ActoGeniX NV, Albireo Pharma, Astra Zeneca, Bayer Schering Pharma, Biogen Idec, Boehringer-Ingelheim, BristolMyers Squibb, Cellerix, Centocor, Chemocentryx, Cosmo Technologies, Danone France, Elan, Genentech, Giuliani, Given Imaging, GlaxoSmithKline, Merck, Millennium, NeoVacs, Ocerra (previously named Renovia), Otsuka American, PDL Biopharma (previously named Protein Design Labs), Pfizer, Ribo Vacs Biotech, Schering-Plough, Shire, Synta, Teva and Petah Tikva, Therakos, UCB (previously named Celltech Therapeutics) and Wyeth. Speakers fees: Abbott Laboratories, Astra Zeneca, Centocor, Elan, Falk, Ferring, Given Imaging, Otsuka American, PDL, Schering-Plough, Shire and UCB. Financial support for research: Astra-Zeneca, Danisco, Danone, Dysphar, Ferring, Giuliani, Lesaffre, Mapi Naxis, Ocerra Therapeutics (previously named Renovia), Roquette, Schering-Plough and UCB. Shareholder: Intestinal Biotech Development. GD: Consultant: Consultant: Abbott Laboratories, Actogenix, Boehringer Ingelheim, Centocor, Cosmo Technologies, Elan, Engene, Ferring Pharmaceuticals, Giuliani, GlaxoSmithKline, Jansen Biologics, Merck, Millenium Pharmaceuticals, MSD, Neovacs, Novonordisk, Otsuka, PDL Biopharma, Pfizer, SetPoint, Shire, Takeda, Aliment Pharmacol Ther 2013; 37: 204-213 ª 2012 Blackwell Publishing Ltd

Week 52 efficacy for responders to adalimumab induction in UC Teva, UCB. Speakers fees: Abbott Laboratories, Tillotts, Tramedico, Ferring, MSD, UCB, Norgine, Shire. Financial support for research: Abbott Laboratories, Jansen Biologics, Given Imaging, MSD, DrFalk Pharma, Photopill. GVA: Consultant: Abbott, Biogen, BMS, MSD, Novartis. Speakers fees: Abbott, Ferring, MSD. Financial support for research: Abbott, Centocor, MSD, Pfizer. DW: Consultant: Abbott, GIVEN Imaging, Janssen Biotech Inc, Millennium Research Group, Prometheus Laboratories, Salix Pharmaceuticals, UCB Pharma. Speakers fees: Abbott, Janssen Biotech Inc, Prometheus Laboratories, UCB Pharma, Warner Chilcott. Financial support for research: Abbott, Bristol-Myers Squibb, GIVEN Imaging, Janssen Biotech Inc, Millennium Research Group, Prometheus Laboratories, UCB Pharma. MK, AL, AMR, MY, JDC, and RT are employees and shareholders of Abbott.

Declaration of funding interests: The study was funded by Abbott. Statistical support was provided by Qian Zhou, PhD, of Abbott. Writing support was provided by Laurinda Cooker, PhD, of Abbott.

SUPPORTING INFORMATION Additional Supporting Information may be found in the online version of this article: Table S1. Remission or response at week 52 for patients who did not respond to adalimumab at week 8 by either full or partial Mayo score (NRI and mNRI). Table S2. Adalimumab week 8 responders (per full Mayo score), overall and by prior anti-TNF exposure (NRI and mNRI): clinical remission and clinical response (per full Mayo score) and mucosal healing at week 52; steroid-free remission or steroid discontinuation at week 52 (patients with baseline steroid use).

REFERENCES 1. Ordas I, Eckmann L, Talamini M, Baumgart DC, Sandborn WJ. Ulcerative colitis. Lancet 2012; 380: 1606–19. 2. Kornbluth A, Sachar DB. Ulcerative colitis practice guidelines in adults: American College Of Gastroenterology. Practice Parameters Committee. Am J Gastroenterol 2010; 105: 501–23. 3. Remicade [package insert]. Horsham, PA USA: Janssen Biotech, Inc., 2011. 4. Reinisch W, Sandborn WJ, Hommes DW, et al. Adalimumab for induction of clinical remission in moderately to severely active ulcerative colitis: results of a randomised controlled trial. Gut 2011; 60: 780–7. 5. Sandborn WJ, van Assche G, Reinisch W, et al. Adalimumab induces and maintains clinical remission in patients with moderate-to-severe ulcerative colitis. Gastroenterology 2012; 142: 257–65. 6. Schroeder KW, Tremaine WJ, Ilstrup DM. Coated oral 5-aminosalicylic acid therapy for mildly to moderately active ulcerative colitis. A randomized study. N Engl J Med 1987; 317: 1625–9. 7. Lewis JD, Chuai S, Nessel L, et al. Use of the noninvasive components of the Mayo score to assess clinical response in ulcerative colitis. Inflamm Bowel Dis 2008; 14: 1660–6. 8. Altman DG, Schulz KF, Moher D, et al. The revised CONSORT statement for reporting randomized trials: explanation and elaboration. Ann Intern Med 2001; 134: 663–94. 9. Boada JN, Boada C, Garcia-Saiz M, et al. Net efficacy adjusted for risk Aliment Pharmacol Ther 2013; 37: 204-213 ª 2012 Blackwell Publishing Ltd

10.

11.

12.

13.

14.

15.

16.

(NEAR): a simple procedure for measuring risk:benefit balance. PLoS ONE 2008; 3: e3580. Whittle SL, Richards BL, Husni E, et al. Opioid therapy for treating rheumatoid arthritis pain. Cochrane Database Syst Rev 2011; 11: CD003113. Whittle SL, Richards BL, van der Heijde DM, et al. The efficacy and safety of opioids in inflammatory arthritis: a cochrane systematic review. J Rheumatol Suppl 2012; 90: 40–6. Travis SP, Stange EF, Lemann M, et al. European evidence-based Consensus on the management of ulcerative colitis: current management. J Crohns Colitis 2008; 2: 24–62. Bastida G, Nos P, Aguas M, et al. Incidence, risk factors and clinical course of thiopurine-induced liver injury in patients with inflammatory bowel disease. Aliment Pharmacol Ther 2005; 22: 775–82. Beaugerie L, Brousse N, Bouvier AM, et al. Lymphoproliferative disorders in patients receiving thiopurines for inflammatory bowel disease: a prospective observational cohort study. Lancet 2009; 374: 1617–25. Peyrin-Biroulet L, Khosrotehrani K, Carrat F, et al. Increased risk for nonmelanoma skin cancers in patients who receive thiopurines for inflammatory bowel disease. Gastroenterology 2011; 141: 1621–8. Hueting WE, Buskens E, van der Tweel I, et al. Results and complications after

17.

18.

19.

20.

21.

ileal pouch anal anastomosis: a meta-analysis of 43 observational studies comprising 9317 patients. Dig Surg 2005; 22: 69–79. Kaplan GG, McCarthy EP, Ayanian JZ, et al. Impact of hospital volume on postoperative morbidity and mortality following a colectomy for ulcerative colitis. Gastroenterology 2008; 134: 680–7. Navaneethan U, Venkatesh PGK, Shen B. Risks and benefits of ileal pouch– anal anastomosis for ulcerative colitis. Therapy 2010; 8: 83–99. Feagan B, Sandborn W, Yang M, et al. Adalimumab therapy reduces hospitalization and colectomy rates in patients with ulcerative colitis: Data from controlled clinical trials. Stockholm, Sweden: United European Gastroenterology Week, 2011. Burmester GR, Panaccione R, Gordon KB, et al. Adalimumab: long-term safety in 23 458 patients from global clinical trials in rheumatoid arthritis, juvenile idiopathic arthritis, ankylosing spondylitis, psoriatic arthritis, psoriasis and Crohn’s disease. Ann Rheum Dis 2012; doi: 10.1136/annrheumdis-2011-201244 [Epub ahead of print]. Colombel JF, Sandborn WJ, Panaccione R, et al. Adalimumab safety in global clinical trials of patients with Crohn’s disease. Inflamm Bowel Dis 2009; 15: 1308–19.

213