Early Mucosal Healing With Infliximab Is Associated With Improved Longterm Clinical Outcomes in Ulcerative Colitis

Transcription

1 GASTROENTEROLOGY 2011;141: Early Mucosal Healing With Infliximab Is Associated With Improved Longterm Clinical Outcomes in Ulcerative Colitis JEAN FRÉDÉRIC COLOMBEL,* PAUL RUTGEERTS, WALTER REINISCH, DIRK ESSER, YANXIN WANG, YINGHUA LANG, ## COLLEEN W. MARANO, # RICHARD STRAUSS, # BJÖRN J. ODDENS, BRIAN G. FEAGAN,** STEPHEN B. HANAUER, GARY R. LICHTENSTEIN, DANIEL PRESENT, BRUCE E. SANDS, and WILLIAM J. SANDBORN *Department of Hepatogastroenterology, CHU Lille and INSERM-CIC9301, Universite Lille Nord de France; Department of Gastroenterology, University Hospital Gasthuisberg, Leuven, Belgium; Department of Gastroenterology and Hepatology, Universität Klinik für Innere Medizin III, AKH Wien, Vienna, Austria; Janssen Biologics, BV, Leiden, The Netherlands; Schering Corporation, Subsidiary of Merck & Co, Inc, Kenilworth, New Jersey; ## Department of Biostatistics, # Department of Immunology, Centocor Research & Development, a division of Johnson & Johnson Pharmaceutical Research & Development, LLC, Malvern, Pennsylvania; **Robarts Research Institute, University of Western Ontario, London, Ontario, Canada; University of Chicago, Chicago, Illinois; University of Pennsylvania, Philadelphia, Pennsylvania; Mount Sinai Medical Center, New York, New York; University of California San Diego, La Jolla, California BACKGROUND & AIMS: In the Active Ulcerative Colitis Trial (ACT)-1 and ACT-2, patients with ulcerative colitis treated with infliximab were more likely than those given placebo to have a clinical response, undergo remission, and have mucosal healing. We investigated the association between early improvement (based on endoscopy) and subsequent clinical outcome. METHODS: Patients underwent endoscopic evaluations at weeks 0, 8, 30, and 54 (ACT-1 only), and were categorized into 4 subgroups by week 8 (Mayo endoscopy subscore, 0 3). The association of week 8 endoscopy subscores, subsequent colectomy risk, symptoms and corticosteroid use outcomes were analyzed. Mucosal healing was defined as a Mayo endoscopy subscore of 0 (normal) or 1 (mild). RESULTS: Infliximab-treated patients with lower week 8 endoscopy subscores were less likely to progress to colectomy through 54 weeks of follow-up evaluation (P.0004). This trend was not observed among patients given placebo (P.47). Patients with lower endoscopy subscores achieved better symptomatic and corticosteroid use outcomes at weeks 30 and 54 (P.0001, infliximab; P.01, placebo). Among patients who achieved clinical response at week 8, trends in subsequent clinical outcomes by week 8 endoscopy subscores were generally consistent with that for the overall patient population; no trends were observed among patients who achieved clinical remission. CONCLUSIONS: The degree of mucosal healing after 8 weeks of infliximab was correlated with improved clinical outcomes including colectomy. Similar trends were observed for all outcomes except colectomy among the subgroup with clinical response at week 8. The degree of mucosal healing at week 8 among those in clinical remission did not predict subsequent disease course. Keywords: Endoscopic Improvement; Mucosal Healing; Colectomy; Symptomatic Outcomes. Ulcerative colitis is an inflammatory disorder characterized by rectal bleeding, persistent bloody diarrhea, and abdominal cramping. The clinical course of ulcerative colitis typically is characterized by periods of asymptomatic remission punctuated by unpredictable recurrent attacks of bloody diarrhea. Other symptoms include severe fecal urgency as a result of reduced rectal compliance, irritability, incontinence, weight loss, and general malaise. 1 Colonic mucosal involvement may present initially as vascular congestion, erythema, edema, and/or granularity. With increasing disease severity, mucosal friability, spontaneous bleeding, and small ulcers coalescing into large or linear ulcers may occur. 2 Until recently, the pharmacologic management of ulcerative colitis has relied mainly on 5-aminosalycilates (5-ASA), corticosteroids, and immunosuppressants including purine antimetabolites and cyclosporine. 3,4 The efficacy and safety of infliximab was shown in 728 patients with moderately to severely active ulcerative colitis who participated in the Active Ulcerative Colitis Trial (ACT)-1 and-2. In these trials, patients treated with infliximab at weeks 0, 2, and 6 and every 8 weeks thereafter were more likely to achieve clinical response, clinical remission, or have mucosal healing at weeks 8, 30, and 54 than were patients receiving placebo. 5 We recently showed that maintenance therapy with infliximab also reduced the risk of colectomy in patients with moderately to severely active ulcerative colitis. 6 Although endoscopic and histologic parameters tend to lag behind symptom improvement, 7 these measures are particularly useful in confirming clinical improvement in ulcerative colitis and may be important tools for predicting long-term outcome. 8 Importantly, residual acute inflammation after treatment in ulcerative colitis may suggest an increased risk of disease relapse. Mucosal healing has emerged as an important treatment goal in ulcerative colitis because evidence is accumulating that it can alter the course of the disease. 9,10 We now report on findings Abbreviations used in this paper: 5-ASA, 5-aminosalycilates; ACT, Active Ulcerative Colitis Trial; RESULTS-UC, Remicade Safety Under Long-term Study in Ulcerative Colitis by the AGA Institute /$36.00 doi: /j.gastro

2 October 2011 MUCOSAL HEALING BENEFIT WITH INFLIXIMAB IN UC 1195 from the ACT-1 and ACT-2 studies evaluating the association of long-term outcomes of colectomy, symptomatic remission, corticosteroid-free symptomatic remission, corticosteroid-free status, and sustained mucosal healing with endoscopy subscores achieved at week 8 after a course of induction therapy with infliximab. Materials and Methods Patients and Study Design These multicenter, randomized, double-blind, placebocontrolled studies were conducted globally at 62 (ACT-1) and 55 (ACT-2) sites between March 2002 and March 2005 (Clinical- Trials.gov numbers NCT and NCT ). The institutional review board or ethics committee at each site approved the protocol. All patients gave written informed consent. Patient enrollment criteria and study design were described previously. 5 Briefly, all eligible patients had an established diagnosis of ulcerative colitis, including endoscopic and histologic assessments. Eligible patients had moderately to severely active disease (total Mayo score, 6 12 points), with an endoscopy subscore of at least 2. Concurrent treatment was not required at enrollment for patients who, in the past, had inadequate response to or could not tolerate treatment with at least one of the following: corticosteroids and/or azathioprine or 6-mercaptopurine in ACT-1 and ACT-2 and/or 5-aminosalicylate-containing medications in ACT-2. Patients receiving concomitant 5-ASA and/or azathioprine/6-mercaptopurine were to keep their dosages stable throughout the trials except for corticosteroids, which were to be tapered at the discretion of the investigators after week 8. Patients previously exposed to infliximab or any other anti tumor necrosis factor agent were excluded. Eligible patients were randomized (in a 1:1:1 ratio) to receive intravenous infusions of infliximab (Remicade; Centocor Ortho Biotech, Inc., Malvern, PA) at a dose of 5 mg/kg or 10 mg/kg or placebo. Study medication was administered at weeks 0, 2, and 6 and then every 8 weeks through week 46 (ACT-1) or 22 (ACT-2). Patients then were followed up through week 54 (ACT-1) or 30 (ACT-2). Mayo Score The Mayo score consists of the following 4 subscores: stool frequency, rectal bleeding, endoscopy findings, and physician s global assessment. 5,11 Each subscore was rated on a scale from 0 to 3, indicating normal to severe disease activity; thus, the Mayo score ranged from 0 to 12, with higher scores indicating more severe disease. Total Mayo scores were determined at weeks 0, 8, 30, and Endoscopy Subscore Mucosal healing was assessed using the Mayo endoscopy subscore classification (0, normal or inactive disease; 1, mild disease with erythema, decreased vascular pattern, mild friability; 2, moderate disease with marked erythema, absent vascular pattern, friability, erosions; and 3, severe disease with spontaneous bleeding, ulceration). Mucosal healing was defined as an absolute endoscopy subscore of 0 or 1. Sustained mucosal healing was defined as mucosal healing at weeks 30 and 54. Colectomy Time to first colectomy was defined as the number of days elapsed from the date of the initial study agent administration to the date of first colectomy (partial or total) through 54 weeks after the first study drug administration. As reported previously, 6 colectomy and commercial infliximab use data through 54 weeks after the first infusion were obtained from ACT-1, ACT-2, and associated data sources (ACT-2 extension study, Remicade Safety Under Long-term Study in Ulcerative Colitis [RESULTS-UC, ClinicalTrials.gov number NCT ]; or a targeted post-study follow-up evaluation). For these analyses, all data were combined to ascertain time to colectomy. Symptomatic Remission, Corticosteroid-Free Symptomatic Remission, and Corticosteroid- Free Status Symptomatic remission was based on the patient-rated Mayo subscores of stool frequency (0, normal number for this patient; 1, 1 2 stools more than normal; 2, 3 4 stools more than normal; and 3, 5 stools more than normal) and rectal bleeding (0, no blood seen; 1, streaks of blood with stool less than half the time; 2, obvious blood with stool most of the time; and 3, blood alone passes). Symptomatic remission was defined as a Mayo stool frequency subscore of 0 or 1 and a Mayo rectal bleeding subscore of 0. Among patients who entered the study receiving corticosteroids, those who were not receiving corticosteroids for 6 days before the date preceding the visit day were considered to be corticosteroid-free at that visit. Patients who achieved symptomatic remission and were not receiving corticosteroids were in corticosteroid-free symptomatic remission. Statistical Methods The primary intent of the efficacy analyses was to assess how the mucosal endoscopy subscore at week 8 affected subsequent long-term clinical outcomes of cumulative incidence of colectomy through 54 weeks after the first dose of study medication, symptomatic remission, corticosteroid-free status, and corticosteroid-free symptomatic remission at weeks 30 and 54, mucosal healing at week 30 and sustained mucosal healing at both weeks 30 and 54. Patients were categorized into 4 subgroups based on their endoscopy subscore (0, 1, 2, or 3) at week 8. For analyses of the proportions of patients who achieved a clinical end point, patients who took prohibited medication for their ulcerative colitis symptoms, discontinued study medication for lack of efficacy, or underwent a colectomy or ostomy had the baseline value of their Mayo subscore or corticosteroid dose carried forward from the time of the event onward. In addition, patients with insufficient data had their last value carried forward. The incidence of colectomy over time was assessed by analyzing the time to first colectomy (partial or total) using the Kaplan Meier product limit method. Results are presented as Kaplan Meier curves. A log-rank test was performed to examine the trend in the incidence of colectomy through 54 weeks, with lower endoscopy subscores at week 8 using data from ACT-1 and ACT-2 combined. The total number of colectomies and the probability of being colectomy-free through 54 weeks were summarized by endoscopy subgroup. Patients who had a colectomy or discontinued and were lost to follow-up evaluation before week 8 were excluded from the analysis. In the colectomy study, a difference in the proportions of patients using commercial infliximab was noted in the placebo (27 of 244 patients; 11%) and combined infliximab groups (28 of 484; 6%). 6 This observation resulted in a post hoc sensitivity

3 1196 COLOMBEL ET AL GASTROENTEROLOGY Vol. 141, No. 4 analysis that took into consideration commercial infliximab use as rescue therapy after study discontinuation. In the sensitivity analysis, use of commercial infliximab was considered to be equivalent to a colectomy; therefore, an event occurred if a patient either received commercial infliximab or had a colectomy, whichever occurred first. The same sensitivity analysis was conducted to evaluate the potential impact of commercial infliximab use on colectomy by endoscopic subscore subgroup. Patients who received commercial infliximab or had a colectomy before week 8 were excluded from the sensitivity analysis. In both analyses, patients who did not have a colectomy (or did not receive commercial infliximab) or had incomplete colectomy follow-up evaluation through 54 weeks were censored at the date of last follow-up evaluation, and censoring was assumed to be noninformative. The proportions of patients who achieved symptomatic remission at week 30 (ACT-1 and ACT-2 combined) and week 54 (ACT-1) were summarized for each endoscopy subgroup. The proportion of patients who were corticosteroid-free and the proportion of patients who achieved corticosteroid-free symptomatic remission at week 30 (ACT-1 and ACT-2 combined) and week 54 (ACT-1) were summarized by endoscopy subgroup. The median doses of corticosteroids at week 30 (ACT-1 and ACT-2 combined) and week 54 (ACT-1) were provided for each endoscopy subgroup. All corticosteroid summaries excluded patients who were not receiving corticosteroids at baseline. The proportion of patients who achieved mucosal healing at week 30 (ACT-1 and ACT-2 combined) and sustained mucosal healing at both weeks 30 and 54 were summarized for each endoscopy subgroup. Cochran Armitage trend tests were performed to assess the trend in the proportions of patients with symptomatic remission, corticosteroid-free symptomatic remission, corticosteroid-free status, mucosal healing, and sustained mucosal healing with lower endoscopy subscores at week 8. Selected analyses (ie, colectomy, symptomatic remission, corticosteroid-free symptomatic remission, mucosal healing, and sustained mucosal healing) also were performed by week 8 endoscopy subscore for those patients who achieved clinical response (total Mayo score decrease from baseline of at least 3 points and at least 30% with an accompanying decrease in the rectal bleeding subscore of at least 1 point or an absolute rectal bleeding subscore of 0 or 1) or clinical remission (total Mayo score of 2 points or lower, with no individual subscore exceeding 1 point) at week 8. For all analyses, patients receiving infliximab 5 mg/kg or 10 mg/kg were pooled and analyzed separately from patients receiving placebo. Nominal P values are presented. Results Patient Disposition, Baseline Characteristics, and Baseline Concomitant Medications In the ACT-1 and ACT-2 trials, 728 patients were randomized to treatment: 244 to placebo and 242 each to infliximab 5 mg/kg and 10 mg/kg. Baseline demographics, disease characteristics, and concomitant medications were generally similar among treatment groups. 5 Mayo Endoscopy Subscores The distribution of baseline Mayo endoscopy subscores was similar among 484 infliximab-treated (0 [endoscopy subscore, 0], 2 [0.4%] [endoscopy subscore, 1], Figure 1. Distribution of Mayo endoscopy subscores at baseline, week 8, week 30 (ACT-1 and ACT-2 combined), and week 54 (ACT-1). 309 [64%] [endoscopy subscore, 2], 173 [36%] [endoscopy subscore, 3]) and 244 placebo-treated patients (0 [0], 1 [0.4%] [1], 146 [60%] [2], 97 [40%] [3]) (Figure 1). At week 8, a greater proportion of infliximab-treated patients (295 of 484; 61%) achieved endoscopy subscores of 0 or 1 compared with placebo-treated patients (80 of 244; 33%). Approximately 50% of infliximab-treated patients achieved endoscopy subscores of 0 or 1 at weeks 30 and 54 compared with 20% 30% of placebo-treated patients (Figure 1). Time to Colectomy Pooling all data sources (ACT-1, ACT-2, ACT-2 through extension study week E24, RESULTS-UC, and post-study follow-up evaluation), 466 infliximab-treated and 222 placebo-treated patients who did not have a colectomy by week 8 were analyzed for time to colectomy through 54 weeks after the first dose. Thirty-eight infliximab-treated patients across the 4 endoscopy subgroups (6 [0], 8 [1], 14 [2], and 10 [3]) had a colectomy through 54 weeks. Infliximab-treated patients with lower endoscopy subscores at week 8 after treatment were more likely to be colectomy-free through 54 weeks of follow-up evaluation (95% [0]; 95% [1]; 87% [2]; and 80% [3]; P.0004; Figure 2A). The Kaplan Meier curves separated as early as 8 weeks for infliximab-treated patients who achieved mucosal healing (endoscopy score of 0 or 1) at week 8 when compared with those who did not (P.001). Such separation was not observed between patients with an endoscopy subscore of 0 (ie, complete mucosal healing) and those with an endoscopy subscore of1(p.87). Twenty-seven placebo-treated patients (1 [0], 6 [1], 17 [2], and 3 [3]) had a colectomy through 54 weeks. A similar pattern could not be confirmed for placebo-treated patients (94% [0]; 89% [1]; 82% [2]; and 92% [3]; P.47; Figure 2B). Sensitivity analyses (459 infliximab-treated and 214 placebo-treated patients) further showed that achievement of lower endoscopy subscores at week 8 after infliximab therapy significantly reduced the incidence of colectomy or commercial infliximab use through 54 weeks (colectomy-free and commercial infliximab use-free through 54

4 October 2011 MUCOSAL HEALING BENEFIT WITH INFLIXIMAB IN UC 1197 likely to be in symptomatic remission at weeks 30 (71% [0]; 51% [1]; 23% [2]; and 9.7% [3]; P.0001) and 54 (73% [0]; 47% [1]; 24% [2]; and 10% [3]; P.0001). Furthermore, notably greater rates of symptomatic remission were displayed in infliximab-treated patients who achieved mucosal healing at week 8 compared with those who did not (Table 1). The highest rate of symptomatic remission occurred in patients with a week 8 endoscopy subscore of 0. Placebo-treated patients with lower endoscopy subscores also were more likely to be in symptomatic remission at weeks 30 and 54. However, the proportions of placebo-treated patients who achieved symptomatic remission in each endoscopy subgroup were lower than those for infliximab-treated patients (Table 1). Figure 2. (A) Combined infliximab group: Kaplan Meier estimate of the proportion of patients without a colectomy from weeks 8 through 54 by endoscopy subgroup. (B) Placebo group: Kaplan Meier estimate of the proportion of patients without a colectomy from weeks 8 through 54 by endoscopy subgroup. weeks: 92% [0]; 92% [1]; 84% [2]; and 69% [3]; P.0001; Supplementary Figure 1A). Kaplan Meier curves again separated as early as 8 weeks for infliximab-treated patients who achieved mucosal healing at week 8 when compared with those who did not. No trend for reduced incidence of colectomy or commercial infliximab use through 54 weeks was observed for placebo-treated patients (colectomy-free and commercial infliximab use-free through 54 weeks: 94% [0]; 79% [1]; 72% [2]; and 77% [3]; P.12; Supplementary Figure 1B). Symptomatic Remission at Weeks 30 and 54 Analysis of symptomatic remission included 484 infliximab-treated and 244 placebo-treated patients at week 30 and 243 infliximab-treated and 121 placebotreated patients at week 54. Infliximab-treated patients with lower endoscopy subscores at week 8 were more Corticosteroid-Free Symptomatic Remission, Corticosteroid-Free Status, and Corticosteroid Dose at Weeks 30 and 54 Analysis of corticosteroid-free status and corticosteroid-free symptomatic remission included 269 infliximab-treated and 139 placebo-treated patients at week 30 and 143 infliximab-treated and 79 placebo-treated patients at week 54 who were receiving corticosteroids at baseline. Infliximab-treated patients with lower endoscopy subscores at week 8 were more likely to be in corticosteroid-free symptomatic remission at week 30 (46% [0]; 34% [1]; 11% [2]; and 6.5% [3]; P.0001) and week 54 (47% [0]; 35% [1]; 5.3% [2]; and 5.3% [3]; P.0001). As was observed for symptomatic remission, the highest rate of corticosteroid-free symptomatic remission occurred in patients with a week 8 endoscopy subscore of 0. Furthermore, clear separation in the rates of corticosteroid-free symptomatic remission was displayed between infliximabtreated patients who achieved mucosal healing at week 8 compared with those who did not (Table 2). Placebo-treated patients with lower endoscopy subscores at week 8 also were more likely to be in corticosteroid-free symptomatic remission at weeks 30 and 54. However, the proportions of placebo-treated patients who achieved corticosteroid-free symptomatic remission in each endoscopy subgroup were lower than those for infliximab-treated patients (Table 2). In the subset of patients who received corticosteroids at baseline, infliximab-treated patients with lower endoscopy subscores at week 8 were more likely to have discontinued corticosteroids at weeks 30 (62% [0]; 46% [1]; 20% [2]; and 9.7% [3]; P.0001) and 54 (63% [0]; 46% [1]; 16% [2]; and 5.3% [3]; P.0001). Moreover, infliximab-treated patients who achieved mucosal healing at week 8 had greater corticosteroid discontinuation rates and lower corticosteroid doses compared with those who did not (Table 2). The proportion of patients discontinuing corticosteroids at weeks 30 and 54 was highest among those with an endoscopy subscore of 0 at week 8. Similar trends were observed for placebo-treated patients at weeks 30 and 54; however, the proportions of patients who were able to discontinue corticosteroids

5 1198 COLOMBEL ET AL GASTROENTEROLOGY Vol. 141, No. 4 Table 1. Association Between Week 8 Mayo Endoscopy Subscore and Symptomatic Remission at Week 30 (ACT-1 and ACT-2 Combined) and Week 54 (ACT-1) Week 8 Mayo endoscopy subscore a Symptomatic remission b Symptomatic remission b at week 30, n/n (%) P value c at week 54, n/n (%) P value c Infliximab 0 85/120 (71) /56 (73) /175 (51) 43/91 (47) 2 29/127 (23) 16/66 (24) 3 6/62 (9.7) 3/30 (10) Placebo 0 11/20 (55) /15 (67) /60 (38) 10/26 (39) 2 17/111 (15) 5/56 (8.9) 3 3/53 (5.7) 1/24 (4.2) n/n, number of patients who achieved endpoint/number of patients assessed. a Mayo endoscopy subscores were as follows: 0, normal or inactive disease; 1, mild disease; 2, moderate disease; and 3, severe disease. b Symptomatic remission was defined as a Mayo stool frequency subscore of 0 or 1 and a Mayo rectal bleeding subscore of 0. c Cochran Armitage trend test. were relatively lower and the median doses of corticosteroids were greater compared with infliximab-treated patients (Table 2). Mucosal Healing at Week 30 and Sustained Mucosal Healing at Weeks 30 and 54 Lower endoscopy subscores at week 8 were associated with increased rates of mucosal healing at week 30 (83% [0]; 67% [1]; 26% [2]; and 9.7% [3]; P.0001) and sustained mucosal healing at both weeks 30 and 54 (77% [0]; 54% [1]; 21% [2]; and 6.7% [3]; P.0001) in infliximab-treated patients (Table 3). Although more than 50% of infliximab-treated patients with mucosal healing at week 8 achieved mucosal healing at week 30 and sustained mucosal healing at weeks 30 and 54, those patients with a week 8 endoscopy subscore of 0 were most likely to achieve these outcomes. Similar trends were observed in placebo-treated patients; however, the rates of mucosal healing at week 30 and sustained mucosal healing at both weeks 30 and 54 were lower in placebo-treated patients than in infliximabtreated patients for each endoscopy subgroup (Table 3). Subgroup Analyses Among infliximab-treated patients who achieved clinical response at week 8, the trends in clinical outcomes by endoscopy subscore at week 8 generally were consistent with those in the overall patient population (Supplementary Table 1). There was no pattern displayed in the cumulative incidence of colectomy through 54 weeks across the endoscopy subgroups (Supplementary Figure 2A). Among infliximab-treated patients who achieved clinical remission at week 8, there was no difference in clinical outcomes (including colectomy) between infliximabtreated patients who had endoscopy subscores of 0 or 1 (Supplementary Table 2; Supplementary Figure 2B). Sample sizes of endoscopy subgroups were small in the subset of placebo-treated patients who achieved clinical response or clinical remission at week 8; therefore, caution should be used when interpreting these results. When analyzed by baseline endoscopy subscore, the magnitude of improvement in endoscopy scores at week 8 did not affect the clinical outcomes at weeks 30 and 54 (Supplementary Tables 3 5). However, among patients with an endoscopy subscore of 2 at week 8, those who improved from a baseline endoscopy subscore of 3 showed better clinical benefit than those who did not show a change from the baseline endoscopy subscore (ie, baseline endoscopy subscore of 2). Discussion Our study included 728 patients with moderately to severely active ulcerative colitis who participated in the placebo-controlled, double-blind ACT-1 and ACT-2 trials. As previously reported, the proportion of infliximabtreated patients who achieved mucosal healing at weeks 8, 30, or 54 was significantly greater than that for the placebo-treated patients (P.009 for all comparisons). 5 Infliximab-treated patients who achieved lower endoscopy subscores at week 8 (ie, endoscopic improvement) after treatment were less likely to progress to colectomy through 54 weeks of follow-up evaluation. Sensitivity analyses further showed that lower endoscopy subscores at week 8 after infliximab treatment significantly reduced the incidence of colectomy or commercial infliximab use through 54 weeks in infliximab-treated patients. Importantly, there was a dichotomy in colectomy rate between infliximab-treated patients who achieved mucosal healing (ie, endoscopy subscores of 0 or 1) at week 8 and those who did not; however, there was no further separation in colectomy outcome distinguishing between patients with an endoscopy subscore of 0 (ie, complete mucosal healing) vs 1. These data further illustrate the importance of achieving mucosal healing as a therapeutic goal in inflammatory bowel disease. Patients who have received conventional

6 October 2011 MUCOSAL HEALING BENEFIT WITH INFLIXIMAB IN UC 1199 Table 2. Association Between Week 8 Mayo Endoscopy Subscore and Corticosteroid Use, Corticosteroid-Free Status, and Corticosteroid-Free Symptomatic Remission at Week 30 (ACT-1 and ACT-2 Combined) and Week 54 (ACT-1) Week 30 Week 54 Corticosteroid-free symptomatic remission, n/n (%) P value c Median corticosteroid Corticosteroid-free, dose b n/n (%) P value c Corticosteroid-free symptomatic remission, n/n (%) P value c Median corticosteroid Corticosteroid-free, dose b n/n (%) P value c Week 8 Mayo endoscopy subscore a Infliximab /65 (62) /65 (46) /32 (63) /32 (47) /102 (46) 35/102 (34) /54 (46) 19/54 (35) /71 (20) 8/71 (11) /38 (16) 2/38 (5.3) /31 (9.7) 2/31 (6.5) /19 (5.3) 1/19 (5.3) Placebo 0 0 8/15 (53) /15 (33) /12 (42) /12 (33) /41 (37) 4/41 (9.8) 20 4/19 (21) 3/19 (16) /61 (8.2) 3/61 (4.9) 20 3/35 (8.6) 1/35 (2.9) /22 (4.6) 0/22 (0) 20 1/13 (7.7) 0/13 (0) n/n, number of patients who achieved endpoint/number of patients assessed. a Mayo endoscopy subscores were as follows: 0, normal or inactive disease; 1, mild disease; 2, moderate disease; and 3, severe disease. b For all placebo-treated and infliximab-treated patients by endoscopic subscore subgroup, the baseline median corticosteroid dose was 20 mg/day. c Cochran Armitage trend test. therapy to date and have achieved clinical response and/or remission may continue to experience subclinical inflammatory processes that leave them at risk of further disease relapses or the development of disease complications Mucosal healing has emerged as an important clinical outcome in the treatment of patients with Crohn s disease, 15,16 and similar findings appear to be emerging for patients with ulcerative colitis. In ulcerative colitis, a single referral center study looked for predictors of longterm outcome, including colectomy rates in outpatients treated with infliximab for refractory ulcerative colitis. 17 In this study, absence of short-term mucosal healing (Mayo endoscopic subscore of 0 or 1 at week 4 or 10 after the first infliximab infusion) was a predictor of subsequent colectomy. A population-based cohort study confirmed that mucosal healing had a significant impact on disease outcome in patients with ulcerative colitis. In patients with ulcerative colitis, 3 of 178 patients (2%) with mucosal healing at 1 year underwent colectomy by 5 years as compared with 13 of 176 patients (7%) without mucosal healing at 1 year (P.02). 9 Patients with healed mucosa also may have a lower risk of dysplasia and adenocarcinoma of the colon, 18 because active inflammation 19,20 is believed to increase the risk of colon cancer development in patients with active colitis. Decreasing colitis activity evidenced by endoscopic improvement at week 8 led to better subsequent clinical outcomes at weeks 30 and 54 as noted by differences in the rates of symptomatic remission, corticosteroid-free symptomatic remission, sustained mucosal healing, and corticosteroid-free status, along with the median corticosteroid dose across endoscopy subscore subgroups. The best clinical outcomes were observed in patients who achieved an endoscopy subscore of 0 (ie, complete mucosal healing). Moreover, consistent with the dichotomy in colectomy rates observed for infliximab-treated patients who did or did not achieve mucosal healing at week 8, notably greater rates of subsequent clinical outcomes were achieved at weeks 30 and 54 in patients who achieved mucosal healing at week 8. Although similar patterns were observed for placebo-treated patients, the overall proportion of patients in each endoscopy subgroup with clinical benefit was lower. We were unable to discern whether a trend exists in placebo-treated patients among endoscopic subscore subgroups at week 8 and cumulative incidence of colectomy through 54 weeks because of the imbalance in the sample sizes among endoscopy subgroups and the relatively small number of colectomy events. In the sensitivity analysis, which considered commercial infliximab use in patients equivalent to patients having a colectomy, placebo-treated patients with endoscopy subscores of 0 were less likely to undergo colectomy or use commercial infliximab compared with those in the other endoscopy subgroups. It is possible that the natural history of ulcerative colitis in those patients who achieved mucosal healing without infliximab therapy differs from that in patients who achieved mucosal healing with infliximab therapy.

7 1200 COLOMBEL ET AL GASTROENTEROLOGY Vol. 141, No. 4 Table 3. Association Between Week 8 Mayo Endoscopy Subscore and Mucosal Healing at Week 30 (ACT-1 and ACT-2 Combined) and Sustained Mucosal Healing at Weeks 30 and 54 (ACT-1) Week 8 Mayo endoscopy subscore a Mucosal healing at week 30, n/n (%) b P value c Sustained mucosal healing at weeks 30 and 54, n/n (%) b P value c Infliximab 0 100/120 (83) /56 (77) /175 (67) 49/91 (54) 2 33/127 (26) 14/66 (21) 3 6/62 (9.7) 2/30 (6.7) Placebo 0 16/20 (80) /15 (67) /60 (58) 7/26 (27) 2 19/111 (17) 4/56 (7.1) 3 1/53 (1.9) 0/24 (0.0) n/n, number of patients who achieved endpoint/number of patients assessed. a Mayo endoscopy subscores were as follows: 0, normal or inactive disease; 1, mild disease; 2, moderate disease; and 3, severe disease. b Mucosal healing was defined as an endoscopy subscore of 0 or 1. c Cochran Armitage trend test. The results of additional subgroup analyses indicated that trends by week 8 endoscopy subscore for subsequent clinical outcomes among infliximab-treated patients who were in clinical response at week 8 generally were consistent with those of the overall patient population. However, for the subset of infliximab-treated patients who achieved clinical remission at week 8, there was no difference in subsequent clinical outcomes between endoscopy subscores of 0 and 1 at week 8. By definition, patients who achieved clinical remission had no individual subscore greater than 1 and the total Mayo score could not exceed 2 points, thus subsequent clinical outcomes may reflect the clinical benefit already achieved at week 8. These results must be interpreted with the knowledge that the end point of symptomatic remission is not a validated end point and is not based on the total Mayo score, which was used to define the end point of clinical response and remission in the ACT-1 and ACT-2 studies. Lewis et al 21 recently showed that the partial Mayo score and a 6-point symptomatic subscore (composed of stool frequency and bleeding components) performed as well as the full Mayo score to identify patient-perceived clinical response. The present study only used Mayo subscores of patient-reported ulcerative colitis symptoms to define symptomatic remission and eliminate the confounding effect of the endoscopy assessment, which is a part of the total Mayo score and considered in the Physician s Global Assessment subscore. Also, the findings reported here may not apply to patients hospitalized with fulminant ulcerative colitis flares because these patients were excluded from study participation. This study showed the benefit of early mucosal healing for short-term and long-term outcomes of symptomatic and corticosteroid-free clinical remission, elimination of corticosteroid use, and sustained mucosal healing. Our results also suggest that incremental clinical benefit was achieved in patients who showed early mucosal healing in addition to clinical response. Although these findings are informative and mucosal healing ultimately may prove to be a treatment goal for patients with ulcerative colitis, caution should be exercised when extrapolating the correlation of symptomatic improvement with mucosal healing presented here to the correlation of clinical remission or clinical response with mucosal healing to clinical practice until formal studies are completed. Future clinical trials should be conducted in which patients are randomized prospectively to treatment algorithms tailored to achieve symptomatic or mucosal healing end points in the short term and then assessed for long-term outcomes including corticosteroid-free remission and risk of colectomy. In summary, our analyses indicated that achievement of early mucosal healing (ie, endoscopy subscores of 0 or 1 at week 8) after treatment with infliximab may lead to better long-term clinical outcomes for patients with moderately to severely active ulcerative colitis. Early mucosal healing was associated with a lower risk of colectomy through 54 weeks of follow-up evaluation and better long-term outcomes related to symptomatic remission, corticosteroidfree symptomatic remission and corticosteroid use, and subsequent mucosal healing at weeks 30 and 54. With the exception of colectomy outcomes, an endoscopic subscore of 0 (ie, complete healing) at week 8 was associated with a greater proportion of patients with long-term clinical benefit. Although no difference in the risk of colectomy by endoscopy subscore was observed in the week 8 clinical responders, lower endoscopy scores at week 8 were associated with symptomatic remission, corticosteroid-free symptomatic remission, and sustained mucosal healing. Among patients with clinical remission at week 8, an endoscopy score of 0 versus 1 did not predict subsequent colectomy, symptomatic remission, corticosteroid-free remission, or sustained mucosal healing. Overall, although similar patterns were observed for placebo-treated patients, the overall proportion of patients in each endoscopy subgroup with clinical benefit was lower.

8 October 2011 MUCOSAL HEALING BENEFIT WITH INFLIXIMAB IN UC 1201 Supplementary Material Note: To access the supplementary material accompanying this article, visit the online version of Gastroenterology at and at doi: /j.gastro References 1. Baumgart DC, Sandborn WJ. Inflammatory bowel disease: clinical aspects and established and evolving therapies. Lancet 2007; 369: Fefferman DS, Farrell RJ. Endoscopy in inflammatory bowel disease: indications, surveillance, and use in clinical practice. Clin Gastroenterol Hepatol 2005;3: Hanauer SB. Medical therapy for ulcerative colitis Gastroenterology 2004;126: Lichtiger S, Present DH, Kornbluth A, et al. Cyclosporine in severe ulcerative colitis refractory to steroid therapy. N Engl J Med 1994; 330: Rutgeerts P, Sandborn WJ, Feagan BG, et al. Infliximab for induction and maintenance therapy for ulcerative colitis. N Engl J Med 2005;353: Sandborn WJ, Rutgeerts P, Feagan BG, et al. Colectomy rate comparison after treatment of ulcerative colitis with placebo or infliximab. Gastroenterology 2009;137: Riley SA, Mani V, Goodman MJ, et al. Microscopic activity in ulcerative colitis: what does it mean? Gut 1991;32: Bitton A, Peppercorn MA, Antonioli DA, et al. Clinical, biological, and histologic parameters as predictors of relapse in ulcerative colitis. Gastroenterology 2001;120: Frøslie KF, Jahnsen J, Moum BA, et al. Mucosal healing in inflammatory bowel disease: results from a Norwegian population-based cohort. Gastroenterology 2007;133: Solberg IC, Lygren I, Jøhnsen J, et al. Mucosal healing after initial treatment may be a prognostic marker for long-term outcome in inflammatory bowel disease. Gut 2008;57(Suppl 2):A 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: Louis E, Collard A, Oger AF, et al. Behaviour of Crohn s disease according to the Vienna classification: changing pattern over the course of the disease. Gut 2001;49: Freeman HJ. Natural history and clinical behavior of Crohn s disease extending beyond two decades. J Clin Gastroenterol 2003; 37: Cosnes J, Cattan S, Blain A, et al. Long-term evolution of disease behavior of Crohn s disease. Inflamm Bowel Dis 2002;8: Vatn MH. Mucosal healing: impact on the natural course or therapeutic strategies. Dig Dis 2009;27: Rutgeerts P, Diamond RH, Bala M, et al. Scheduled maintenance treatment with infliximab is superior to episodic treatment for the healing of mucosal ulceration associated with Crohn s disease. Gastrointest Endosc 2006;63: Ferrante M, Vermeire S, Fidder H, et al. Long-term outcome after infliximab for refractory ulcerative colitis. J Crohns Colitis 2008; 2: Rutter MD, Saunders BP, Wilkinson KH, et al. Cancer surveillance in longstanding ulcerative colitis: endoscopic appearances help predict cancer risk. Gut 2004;53: Rutter M, Saunders B, Wilkinson K, et al. Severity of inflammation is a risk factor for colorectal neoplasia in ulcerative colitis. Gastroenterology 2004;126: Itzkowitz SH, Yio X. Inflammation and cancer IV. Colorectal cancer in inflammatory bowel disease: the role of inflammation. Am J Physiol Gastrointest Liver Physiol 2004;287:G7 G 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: Received December 22, Accepted June 17, Reprint requests Address requests for reprints to: Jean Frédéric Colombel, MD, Hopital Claude Huriez, Centre Hopitalier Universitaire de Lille, Rue Michel Polonovski, Lille, France jean-frederic. colombel@chru-lille.fr; fax: (33) Acknowledgments The corresponding author had full access to all of the data and takes full responsibility for the veracity of the data and data analysis. Some of the data displayed in this article were presented at the annual meeting of The American College of Gastroenterology 2010 (oral presentation; San Antonio, TX), United European Gastroenterology Week 2010 (poster; Barcelona, Spain), and 2010 Annual Advances in the Inflammatory Bowel Diseases, 2010 (oral presentation; Hollywood, FL). Conflicts of interest Paul Rutgeerts, William Sandborn, Jean Frédéric Colombel, Stephen Hanauer, Gary Lichtenstein, Walter Reinisch, Bruce Sands, and Brian Feagan have served as consultants for Centocor Research & Development, a division of Johnson & Johnson Pharmaceutical Research & Development, LLC and Schering Plough, a subsidiary of Merck & Co, Inc; Paul Rutgeerts, William Sandborn, Stephen Hanauer, Bruce Sands, Brian Feagan, Walter Reinisch, and Gary Lichtenstein have participated in continuing medical education events supported by unrestricted educational grants from Centocor Research & Development, a division of Johnson & Johnson Pharmaceutical Research & Development, LLC and Schering Corporation, a subsidiary of Merck & Co, Inc; Björn Oddens and Yanxin Wang are employees of Schering Corporation, a subsidiary of Merck & Co, Inc; Colleen Marano and Richard Strauss are employees of the Department of Immunology, Centocor Research & Development, a division of Johnson & Johnson Pharmaceutical Research & Development, LLC, and Yinghua Lang is an employee of the Department of Biostatistics, Centocor Research & Development, a division of Johnson & Johnson Pharmaceutical Research & Development, LLC; and Dirk Esser is a former employee of Janssen Biologics BV and is currently an employee of Boerhringer Ingelheim. Funding Supported by a research grant from Centocor Research & Development, a division of Johnson & Johnson Pharmaceutical Research & Development, LLC (Malvern, PA) and Schering Corporation, a subsidiary of Merck & Co, Inc (Kenilworth, NJ); this study was designed and conducted by the ACT-1 and ACT-2 Steering Committees, Centocor Research & Development, a division of Johnson & Johnson Pharmaceutical Research & Development, LLC and Schering Corporation (a subsidiary of Merck & Co, Inc) who jointly analyzed and interpreted the data, and contributed to the manuscript; Drs Colombel, Rutgeerts, and Sandborn prepared the first draft of the manuscript, and the ACT-1 and ACT-2 Steering Committee made the decision to publish; editorial and writing support was provided by James P. Barrett, an employee of the Medical Affairs Publications Group, Janssen Services, LLC.