- Email: [email protected]
Postoperative Therapy With Infliximab Prevents Long-term Crohn’s Disease Recurrence
Clinical Gastroenterology and Hepatology 2014;-:-–-
Postoperative Therapy With Infliximab Prevents Long-term Crohn’s Disease Recurrence Miguel Regueiro,* Kevin E. Kip,‡ Leonard Baidoo,* Jason M. Swoger,* and Wolfgang Schraut§ *Inflammatory Bowel Disease Center and Division of Gastroenterology, Hepatology and Nutrition, and §Division of Gastrointestinal Surgery, Department of Surgery, University of Pittsburgh Medical Center, Pittsburgh, Pennsylvania; and ‡ Research Center, College of Nursing, University of South Florida, Tampa, Florida BACKGROUND & AIMS:
A previous randomized, placebo-controlled study showed that infliximab maintenance therapy prevented recurrence of Crohn’s disease 1 year after an ileocolonic resection. We evaluated recurrence of Crohn’s disease, on the basis of endoscopic examination and/or the need for additional surgical resection, beyond the first postoperative year.
METHODS:
In a prospective, open-label, long-term follow-up study, 24 patients previously randomly assigned to receive infliximab for 1 year after an ileocolonic resection were given the option to continue, stop, or start infliximab therapy. The primary end point was the time to recurrence of Crohn’s disease, on the basis of endoscopic evidence (endoscopic recurrence), from the initial assignment to postoperative infliximab or placebo. Secondary end points were rate of endoscopic recurrence, time to reoperation, and rate of surgical recurrence in relation to the total time on infliximab.
RESULTS:
All patients were followed for at least 5 years after surgery. Patients assigned to the infliximab group in the first year after surgery had a longer mean time to first endoscopic recurrence (1231 – 747 days) than patients originally assigned to the placebo group (460 – 121 days, P [ .003). Colonoscopies identified Crohn’s disease recurrence in 22.2% of patients who received long-term infliximab and in 93.9% of those not on infliximab (P < .0001). Compared with no infliximab, the adjusted rate ratio for being in endoscopic remission while on infliximab was 13.47 (95% confidence interval, 3.52–61.53; P [ .0001). Patients originally assigned to the infliximab group had a mean longer time to surgery (1798 – 359 days) than patients originally assigned to the placebo group (1058 – 529 days, P [ .04). The rate of surgical recurrence (required additional surgical resection) was significantly lower among patients who received infliximab for most of the followup period than patients who received it for shorter periods (20.0% vs 64.3%, P [ .047).
CONCLUSIONS:
Postoperative infliximab maintenance beyond 1 year prevents recurrence of Crohn’s disease.
Keywords: IBD; Anti-TNF; Clinical Trial; Tumor Necrosis Factor; Treatment; Drug.
espite the increased use of immunomodulators and anti–tumor necrosis factor alpha (anti-TNF) medications for the treatment of inflammatory bowel disease, approximately two-thirds of Crohn’s disease patients will require an intestinal resection, and more than 50% will require at least 1 additional surgery in their lifetime.1–3 Cigarette smoking, penetrating disease, recurrent surgery, and postoperative endoscopic recurrence are well-established risk factors that predict subsequent intestinal resection.4,5 Nitroimidazole antibiotics and thiopurine analogs may be effective treatment for prevention of postoperative recurrence.6–9 Recently, anti-TNF agents have shown efficacy in the prevention of Crohn’s disease after intestinal resection.10–12 Most of the studies on the postoperative prevention of Crohn’s disease with anti-TNF agents have focused on
D
1-year clinical and endoscopic outcomes.12–14 Long-term postoperative anti-TNF studies have been limited to approximately 3 years of follow-up and primarily consist of retrospective reviews or case series.10,15–18 There is only 1 open-label infliximab postoperative study that followed patients for 3 years.19 There remain important unanswered questions about long-term postoperative Crohn’s disease management with anti-TNF therapy, such as (1) does initiating infliximab shortly after
Abbreviations used in this paper: AZA, 6-mercaptopurine; TNF, tumor necrosis factor.
azathioprine;
© 2014 by the AGA Institute 1542-3565/$36.00 http://dx.doi.org/10.1016/j.cgh.2013.12.035
6MP,
2
Regueiro et al
surgery prevent long-term recurrence and additional surgery?, (2) does starting infliximab in response to endoscopic recurrence induce remission and avoid future surgery?, and (3) does stopping infliximab after the first postoperative year of remission lead to recurrence or sustain anti-TNF–free remission? This is a prospective, open-label study from our previously published randomized controlled trial of infliximab vs placebo for postoperative Crohn’s disease.12 We evaluated long-term endoscopic and surgical recurrence in patients continuing, newly starting, or stopping infliximab after the first postoperative year.
Methods Study Design The study design consisted of a prospective, openlabel long-term follow-up study (>5 years) of patients from our previously published randomized controlled trial of infliximab vs placebo for postoperative Crohn’s disease.12 The study was conducted at the Inflammatory Bowel Disease Center at the University of Pittsburgh Medical Center, with collection of data beyond the initial 1-year randomized controlled study approved by the Institutional Review Board at the University of Pittsburgh Medical Center. Patients were followed through December 2012 to obtain at least 5 years of data on all patients.
Patients Between 2004 and 2007, 24 adult patients underwent an ileocolonic resection for Crohn’s disease. All patients had been enrolled within 4 weeks of surgery and were randomly assigned in a blocked 1:1 manner to infliximab 5 mg/kg or identical-appearing placebo infusions at 0, 2, 6, and every 8 weeks for 1 year. At the end of 1 year, an ileocolonoscopy was performed, and an ileal Crohn’s disease activity score was recorded on the basis of the methods of Rutgeerts et al.5 On completion of the 1-year postoperative randomized controlled trial, patients were followed for at least 5 years from the date of surgery, during which time they received standard of care treatment by their gastroenterologist. All patients were treated at University of Pittsburgh Medical Center, with clinical information recorded in the electronic medical record system. The data captured in the longterm postoperative follow-up period included colonoscopy with ileal endoscopic scores, additional surgeries, medications, indication for stopping infliximab, and disease behavior.
Treatment At the 1-year postoperative ileocolonoscopy, patients and physicians were blinded to the study treatment
Clinical Gastroenterology and Hepatology Vol.
-,
No.
-
(placebo or infliximab). At that time, the physician informed the patient of their endoscopic findings and discussed the options of open-label infliximab every 8 weeks or other treatment. For those who received infliximab after the postoperative year 1 colonoscopy, this meant initiation of infliximab for the placebo patients and uninterrupted maintenance for those who had been randomized to infliximab. For those who did not receive infliximab after the first postoperative year, this meant continuation without infliximab for the placebo patients and stopping infliximab for those who had been randomized to it. Beyond the 1-year randomized study, physicians had the option to start or stop any therapy as they deemed appropriate.
Study Outcomes The primary study outcome was the time to endoscopic Crohn’s disease recurrence by initial postoperative assignment to infliximab or placebo. Beyond the first year of the randomized controlled study, ileocolonoscopy was performed at the discretion of the treating gastroenterologist. Although the optimal time interval for colonoscopic inspection beyond the first postoperative year has not been defined, it is our practice to complete a second ileocolonoscopy within 2 years of ileocolonic resection. The Rutgeerts endoscopic score was recorded.5 Although this scoring system has never been validated, it is the most widely accepted endoscopic metric in postoperative Crohn’s disease studies. The scores were as follows: i0, no lesions; i1, 5 or fewer aphthous lesions; i2, more than 5 aphthous lesions with normal mucosa between the lesions or skip areas of larger lesions or lesions confined to the ileocolonic anastomosis; i3, diffuse aphthous ileitis with diffusely inflamed mucosa; and i4, diffuse inflammation with large ulcers, nodules, and/or narrowing. Endoscopic recurrence was defined by an endoscopic score of i2, i3, or i4. Endoscopic remission was defined by a score of i0 or i1. Secondary study outcomes included (1) rate of endoscopic recurrence stratified by use vs non-use of infliximab and 6-mercaptopurine/azathioprine (6MP/ AZA), (2) time to reoperation by the initial assignment to infliximab or placebo, and (3) rate of endoscopic and surgical recurrence by total time on infliximab. Surgical recurrence was defined as any intestinal resection for Crohn’s disease or a complication of Crohn’s disease.
Statistical Methods Baseline demographic and clinical characteristics of the study sample are presented as means and standard deviations for continuous variables and percentages for categorical variables. In patient-level analyses, the time on study, percentage of time on infliximab, and number of colonoscopies performed were compared by initial
-
2014
random assignment by use of Student t tests. Rates of endoscopic disease recurrence and recurrent surgery at any time during follow-up were compared by random assignment by use of Fisher exact test. Among patients who experienced endoscopic disease recurrence, the mean number of days to recurrence was compared by random assignment by use of Student t tests. Because of small numbers, corresponding days to recurrent surgery were compared by use of the Wilcoxon rank sum test. Kaplan–Meier estimates of surgery-free survival were calculated by random assignment, stratified by approximate median time on infliximab therapy during the full study period (<60% of the time vs 60% of the time). The cutoff of >60% of time was assigned to identify a group of patients who had been on infliximab for the majority of the study period. The log-rank statistic was used to compare the 4 respective survival curves. In analyses in which colonoscopy was the unit of analysis (n ¼ 84), generalized estimating equations specifying the binomial distribution, log link, and a compound symmetry correlation structure were used to estimate rate ratios of being in remission in relation to use of biologic medications at the time of endoscopic evaluation. Univariate estimates and estimates adjusted for concomitant use of 6MP/AZA, and number of days between successive colonoscopies, are presented. A 2-sided P value <.05 was used to define statistical significance.
Results Study Population The 24 participants in the original study (infliximab, n ¼ 11; placebo, n ¼ 13) were enrolled between October 2004 and April 2007 and completed their participation by December 2012. The mean length of study participation was 6.5 0.6 years, with a minimum of 5.4 and maximum of 7.7 years. Mean years of participation were similar between patients assigned to infliximab (6.4 0.8) or to placebo (6.5 0.5, P ¼ .88). As previously reported, baseline characteristics of the 2 groups were similar for age, gender, duration of Crohn’s disease, disease behavior, disease location, prior infliximab exposure, and prior surgical resection (Supplementary Table 1).12 For the full cohort, the mean age at entry was 37.8 13.8 years, 33.3% were female, 25.0% were current smokers and continued to smoke throughout the longterm follow-up, 50.0% had Crohn’s disease for more than 10 years, and 91.7% had surgery for a penetrating/ perforating complication. For 66.7% of patients, this was their first intestinal resection for Crohn’s disease, for 25% this was their second surgery, and for 8.3% this was their third surgery. There were 45.8% of patients taking 6MP/AZA, and 33.3% had received prior treatment with infliximab. In the long-term follow-up period, none of the patients were prescribed 5-aminosalicylates, antibiotics, or methotrexate for their Crohn’s disease.
Infliximab Prevents Postoperative Crohn’s Disease
3
Follow-up Colonoscopies, Surgeries, and Use of Infliximab After the index (trial entry) surgery, the mean number of colonoscopies performed during follow-up ranged between 2 and 5 and was similar between infliximab-assigned (3.4 1.0) and placebo-assigned (3.5 1.1) patients (P ¼ .69). Despite initial random assignment to infliximab, the mean percentage of time during the entire course of the study that these patients were on infliximab therapy (50.3 40.4) was similar to the placebo-assigned (53.0 29.6) group (P ¼ .85). Of the 13 placebo-assigned patients, 1 never started infliximab in long-term follow-up, and 12 started infliximab therapy in the year that followed their 1-year initial placebo regimen (1 patient started within the first year after undergoing recurrent surgery). Overall, 23 of the 24 patients (95.8%) initiated infliximab therapy at some time during the study, and of these patients, 14 (60.9%) stopped therapy for the following reasons: infusion reaction (n ¼ 2), lupus-like reaction (n ¼ 1), breast cancer (n ¼ 1), patient decision (n ¼ 8), and loss of insurance (n ¼ 2). None of the patients stopped infliximab because of a loss of response. Percentage use of infliximab during follow-up was similar between nonsmokers and smokers (42.5 36.5 vs 54.2 35.0, P ¼ .50).
Primary Outcome Time to endoscopic recurrence by assignment to infliximab or placebo (endoscopy, patient-level analysis). Eight of the 11 infliximab patients elected to stop
infliximab after the 1-year randomized controlled trial (Figure 1). All 8 patients had endoscopic recurrence after stopping infliximab; 5 have had another surgery, and 3 have not. One infliximab patient had an endoscopic recurrence (i3) at the end of the 1-year study but elected to stay on infliximab for control of arthritis symptoms. Despite the infliximab, this patient has had a persistent i3 endoscopic score through 6 years of follow-up but has not had Crohn’s disease symptoms or required another resection. The 2 patients originally assigned to infliximab who continued infliximab for the duration of long-term follow-up have remained in remission (i0, i1). Of the 8 infliximab patients who stopped infliximab at some time, the time to recurrence was relatively abrupt, with a range of 8.0–30.0 months and mean of 18.2 7.2 months. Twelve of the 13 patients (92.3%) in the placebo arm showed endoscopic evidence of disease recurrence at some point. Of these 13 patients, 12 elected to initiate infliximab after the 1-year randomized controlled trial. The one who did not elect to start infliximab had an ileal score of 1 at 1 year but ultimately progressed to i4 and had another resection (Figure 1). Seven of the 12 (2 were i3, 5 were i2) who initiated infliximab responded with
4
Regueiro et al
Clinical Gastroenterology and Hepatology Vol.
-,
No.
-
Figure 1. Long-term (>5 years) follow-up of patients initially randomized to infliximab or placebo. Recurrence is defined as endoscopic recurrence (i2, i3, i4). Remission is defined as endoscopic score of i0 or i1. Surgery is defined as intestinal resection for recurrent Crohn’s disease or complication of Crohn’s disease. All surgery patients had 1 intercurrent surgery during follow-up, except 1 of the infliximab patients who stopped infliximab had 2 surgeries and the 1 placebo patient who did not receive anti-TNF treatment had 2 surgeries. IFX, infliximab; PBO, placebo; RCT, randomized controlled trial.
endoscopic remission and have avoided additional surgery. Five of the 12 (4 were i4 and 1 was i3) who initiated infliximab did not achieve endoscopic remission and required another surgical resection. Overall, patients
originally assigned to infliximab had a longer time to first endoscopic recurrence compared with patients originally assigned to placebo (1231 747 days vs 460 121 days, P ¼ .003) (Figure 2).
Figure 2. Plot of days to endoscopic disease recurrence by initial random assignment to infliximab or placebo. The black-filled portion of each horizontal bar depicts number of days on infliximab therapy; the gray portion of each horizontal bar depicts number of days not on infliximab therapy.
-
2014
Infliximab Prevents Postoperative Crohn’s Disease
Secondary Outcomes Rate of endoscopic recurrence by overall use of infliximab and/or immunomodulators (endoscopy, colonoscopylevel analysis). For the 24 patients, a total of 84
colonoscopies were performed after entry into the trial and during long-term follow-up. The median time between successive follow-up colonoscopies was 451 days, with interquartile range of 406–734 days and minimum and maximum of 65 and 1515 days, respectively. The median time between colonoscopies did not differ between those originally assigned to placebo or to infliximab. Twenty-four colonoscopies occurred at the completion of the1-year trial period, and the remaining colonoscopies were distributed as second (n ¼ 24), third (n ¼ 22), fourth (n ¼ 8), and fifth (n ¼ 6) year thereafter. Endoscopic ileal disease activity scores at the time of colonoscopy were i0 (n ¼ 30, 35.7%), i1 (n ¼ 10, 11.9%), i2 (n ¼ 15, 17.9%), i3 (n ¼ 16, 19.0%), and i4 (n ¼ 13, 15.5%). This corresponded to endoscopic recurrence in 44 of 84 colonoscopies (52.4%). Among the colonoscopic evaluations, 12 (14.3%) were performed after a second intercurrent surgical resection. Biologic medication status at the time of colonoscopy, defined as medication being given during the majority (>60%) of time since the previous evaluation, was distributed as none (n ¼ 33, 39.3%), infliximab (n ¼ 45, 53.6%), and adalimubab or certolizumab (n ¼ 6, 7.1%) (Table 1). Twenty-five colonoscopies (29.8%) were done on patients on 6MP/AZA (ie, irrespective of biologic therapy). As shown in Figure 3, 77.8% of colonoscopies done on patients receiving infliximab disclosed disease in remission compared with just 6.1% of colonoscopies done on patients not receiving biologic therapy (P < .0001). This analysis reflects the intermittent nature of use of infliximab and strong temporal relationship between exposure to infliximab and the probability of being in disease remission. To illustrate, compared with those not
5
on biologic therapy, the adjusted rate ratio of patients being in remission while on infliximab was 13.47 (95% confidence interval, 3.52–61.53; P ¼ .0001). A similar protective effect (50% remission, P ¼ .02) was observed for other biologic therapy (adalimubab or certolizumab), although only 6 colonoscopies were evaluated in this circumstance. Results were not appreciably different after statistical adjustment for age, gender, duration of Crohn’s disease, infliximab use before trial entry, surgical resection before trial entry, smoking status, or concomitant use of 6MP/AZA. Being a smoker was not associated with endoscopic recurrence (rate ratio ¼ 0.87; 95% confidence interval, 0.52–1.47; P ¼ .61). In subgroup analyses (Figure 4) the highly protective effect of infliximab on endoscopic recurrence was evident. Only 22.2% of colonoscopies in which the patient was taking infliximab had endoscopic recurrence, compared with 93.9% recurrence in those not on infliximab (P ¼ .0002). The role of 6MP/AZA in combination with or as primary treatment of postoperative Crohn’s disease was also evaluated. Of the 26 colonoscopies in which use of both infliximab and 6MP/AZA was absent, all 26 (100%) showed endoscopic evidence of disease recurrence. This compared with just 18.5% of the 27 colonoscopies in which infliximab was being received in the absence of 6MP/AZA (P < .0001). The rates of endoscopic recurrence were not significantly different between those on monotherapy infliximab compared with those on combination infliximab with 6MP/AZA (18.5% vs 27.8%, P ¼ .49). Finally, 5 of the 7 colonoscopies (71.4%) in which 6MP/AZA was administered without infliximab showed evidence of disease recurrence, which was not statistically different from those on no medications (100%, P ¼ .08). Time to surgical recurrence by assignment to infliximab or placebo (surgery, patient-level analysis). Five of
11 infliximab-assigned patients (45.5%) and 6 of 13 placebo-assigned patients (46.2%) underwent recurrent
Table 1. Rate Ratios of Being in Remission (i0, i1) During Follow-up Colonoscopies and in Relation to Use of Biologic Medications n
No. in remission
% in remission
Unadjusted RR
Adjusted RRa
95% CI
P value
33 45 6 33 51
2 35 3 2 38
6.1 77.8 50.0 6.1 74.5
1.0 12.98 8.21 1.0 12.34
1.0 13.47 9.09 1.0 12.95
— 3.52–61.53 1.75–47.26 — 3.35–50.10
— .0001 .009 — .0002
Excluding colonoscopies performed after intercurrent surgery None 30 2 6.7 Infliximab 40 30 75.0 Other biologic 2 0 0.0 None 30 2 6.7 Any biologic 42 30 71.4
1.0 12.01 — 1.0 10.74
1.0 12.44 — 1.0 11.18
— 3.19–48.47 — — 2.84–44.03
— .0003 — — .0006
Biologic use None Infliximab Other biologic None Any biologic
CI, confidence interval; RR, rate ratio. a Adjusted for concomitant use of 6MP/AZA and number of days between successive colonoscopies. Within-subject correlation of multiple colonoscopies modeled by using a compound symmetry correlation structure.
6
Regueiro et al
Clinical Gastroenterology and Hepatology Vol.
-,
No.
-
assigned patients without reoperation received infliximab for most of the post 1-year trial follow-up period. Rate of recurrence by total time on infliximab. Among patients on infliximab therapy for at least 60% of the full study period and irrespective of initial random assignment, the rate of surgical re-resection was significantly lower compared with those with less frequent use of infliximab (20.0% vs 64.3%, P ¼ .047, Figure 6).
Discussion
Figure 3. Percentage of patients in remission (ileal score, i0 or i1) of Crohn’s disease at 5–8 years of endoscopic evaluation by use vs non-use of biologic therapy in the period that preceded endoscopic evaluation.
surgery during long-term follow-up (Figure 5). Of the 11 infliximab-assigned patients, reoperations during the years of follow-up were distributed as follows: 0 (n ¼ 6, 54.5%), 1 (n ¼ 4, 36.4%), and 2 (n ¼ 1, 9.1%). This was similar (P ¼ .98) to the distribution among the 13 placebo-assigned patients: 0 (n ¼ 7, 53.8%), 1 (n ¼ 5, 38.5%), and 2 (n ¼ 1, 7.7%). Despite this similar rate of re-resectional surgery, patients originally assigned to infliximab had a longer time interval to surgery compared with patients originally assigned to placebo (1798 359 days vs 1058 529 days, P ¼ .04) (Supplementary Figure 1). As shown in Figure 5, four of the 5 infliximab-treated patients who underwent reoperation during long-term follow-up did so after ceasing therapy at the end of the 1-year trial period. Four of the 6 infliximab-assigned patients without reoperation remained on infliximab for all or nearly all of the follow-up period. Five of the 7 placebo-
This is the longest postoperative Crohn’s disease study to date and provides insight into the natural course of disease after surgical resection as well as the impact of postoperative treatment. Patients originally assigned to infliximab had a significantly longer time to first endoscopic recurrence and the need for re-resection compared with placebo patients. When considering each colonoscopy as a separate event, more than three-fourths of the colonoscopies done on patients on infliximab revealed endoscopic remission, compared with less than 10% of those on no biologic therapy. The endoscopic recurrence rates were similar for patients receiving infliximab monotherapy or in combination with 6MP/AZA, and the thiopurines alone did not protect against recurrence. Patients were stratified into those on infliximab for the majority of time (>60%) and those for less than that. Irrespective of initial assignment to placebo or infliximab, among patients on infliximab therapy for the majority of the entire follow-up period, there was a higher rate of endoscopic remission and lower rate of surgical recurrence compared with those not on infliximab. Does initiating infliximab shortly after surgery prevent long-term recurrence and additional surgery? All patients who initiated infliximab after surgery and continued long-term have avoided surgery for nearly 8
Figure 4. Colonoscopylevel analysis of rate of endoscopic disease recurrence (i2, i3, i4) stratified by use vs non-use of infliximab and by 6MP/ AZA status overall and in absence of use of infliximab. The 6 colonoscopies with adalimumab or certolizumab were excluded. IFX, infliximab.
-
2014
Infliximab Prevents Postoperative Crohn’s Disease
7
Figure 5. Plot of days to recurrent surgery by initial random assignment to infliximab or placebo. The black-filled portion of each horizontal bar depicts number of days on infliximab therapy; the gray portion of each horizontal bar depicts number of days not on infliximab therapy.
years. These findings are consistent with other published studies of shorter duration of follow-up. The studies by Sorrentino et al11 and Papamichael et al17 reported 2year endoscopic recurrence of 0% and 25% with infliximab and adalimumab, respectively. Similarly, Yoshida et al19 found a 3-year clinical remission rate of 93% and 1-year endoscopic recurrence of 21% in patients started on infliximab within 4 weeks of surgery. Savarino et al16 reported a 3-year endoscopic recurrence of 0% in patients initiating adalimumab 2 weeks after an
Figure 6. Crude rates of surgery during entire study period stratified by initial random assignment and cumulative time on infliximab therapy (<60% vs 60%). IFX, infliximab; pts, patients.
intestinal resection. In the study by Araki et al,15 the 3year surgical recurrence rate was only 3% in patients administered infliximab within 8 weeks of surgery compared with 34% not treated with infliximab. Although none of these studies are randomized and all have small numbers, the low rates of endoscopic and surgical recurrence are consistent with our results. Does starting infliximab in response to endoscopic recurrence induce remission and avoid future surgery? It is not known whether initiating anti-TNF treatment immediately after surgery prevents future Crohn’s disease complications and surgery more effectively than waiting to treat endoscopic recurrence. The patients who received infliximab within 4 weeks of surgery had a longer time to endoscopic and surgical recurrence compared with the placebo patients who started infliximab 1 year after surgery in response to endoscopic recurrence. All of the patients who initiated infliximab within 4 weeks of surgery and continued long-term treatment have avoided additional surgery. In contrast, 58% of the placebo patients who started infliximab in response to 1-year endoscopic recurrence have had improvement in endoscopic inflammation, but nearly one-half have required additional surgery. It is possible that performing a colonoscopy sooner than 1 year after surgery would have led to earlier initiation of treatment and better outcomes. There have been several postoperative Crohn’s disease studies that have initiated anti-TNF treatment in response to early endoscopic recurrence. Yamamoto et al20 showed that 75% of patients starting infliximab 6 months after
8
Regueiro et al
surgery for endoscopic recurrence improved endoscopic inflammation compared with 38% of AZA and 0% of mesalamine patients. Complete mucosal healing was achieved in 38% of patients on infliximab when compared with 13% of those receiving AZA and 0% of mesalamine-treated patients. Sorrentino et al21 reported similar results, with 69% of patients having endoscopic improvement and 54% with complete mucosal healing after receiving infliximab for 6-month postoperative Crohn’s disease recurrence. In a study by Papamichael et al,17 adalimumab administered to patients with endoscopic recurrence 6 months after surgery was able to improve endoscopic inflammation in 60%. All 3 of these studies initiated anti-TNF treatment in response to endoscopic recurrence at 6 months. The overall endoscopic improvement rates (60%–70%) are similar to 58% we found in placebo patients initiating infliximab 1 year after surgery. Whether earlier endoscopic surveillance would improve these rates further and whether this strategy would approach the consistently high endoscopic remission rate (80%–100%) seen in the studies initiating anti-TNF treatment within 1 month of surgery are not known. The postoperative model provides a glimpse into the timing of medical therapy vis à vis mucosal healing and thus prevention or postponement of recurrent Crohn’s disease. Patients who undergo a resection that removes the entirety of the diseased intestine may be considered in the “deepest” remission. Starting “top down” biologic treatment immediately after surgery rather than waiting for disease recurrence provides the best rates of longterm prevention.10–19 This parallels the experience with previously published medical treatment trials that showed greater benefit in patients with short Crohn’s disease duration compared with a less robust response in those with longer disease duration.22–25 The reason that Crohn’s disease patients with longer disease duration have an attenuated medication response is currently in the domain of speculation but may reflect irreversible structural bowel damage, vascular changes, or possibly an altered microbiome and cytokine profile. We did find that patients with less severe postoperative endoscopic recurrence responded better to the initiation of infliximab. Patients with an ileal score of 2 were able to induce endoscopic remission, whereas none of the i4 patients and only 2 of the i3 patients had endoscopic improvement. All of the patients with i4 scores and half of the i3 patients required another surgery despite initiation of infliximab. Lacking in the management of postoperative Crohn’s disease is a noninvasive measure to diagnose recurrent mucosal inflammation that would allow initiation of medication before the development of irreversible intestinal changes. Such a diagnostic tool would avoid overtreating postoperative patients who may never develop recurrence and target patients for treatment at the earliest sign of Crohn’s disease activity. Does stopping infliximab after the first postoperative year of remission lead to recurrence or sustain anti-
Clinical Gastroenterology and Hepatology Vol.
-,
No.
-
TNF–free remission? The optimal duration of anti-TNF treatment for patients with clinical and endoscopic remission, ie, deep remission, is not known. Recent studies have suggested that factors such as absence of mucosal inflammation noted by endoscopy or serologic markers and lack of recent corticosteroids or anti-TNF dose intensification may predict those who can stop anti-TNF treatment and avoid relapse.26–28 Most of the patients in these studies were on concomitant immunomodulators at the time of anti-TNF discontinuation. In all 3 studies, however, approximately one-half of the patients relapsed within 1 year of stopping the anti-TNF treatment. In our postoperative study, nearly threefourths of the infliximab patients stopped treatment at 1 year after resection while in complete endoscopic remission. All of these patients subsequently had endoscopic recurrence, and most required another surgery. Sorrentino et al18 reported 83% of their postoperative patients in endoscopic remission for 3 years developed endoscopic recurrence after stopping infliximab. The high postoperative recurrence rates after stopping infliximab may reflect a more aggressive Crohn’s disease course. There are several limitations in our study, mainly the small sample size and the open-label design. Another limitation is that our study was conducted at a tertiary care center, and the patients were at high risk for recurrence; the majority had penetrating disease, onethird had more than 1 prior surgery, and more than half of the patients had failed immunomodulators. The lack of control was also a limitation in that patients could stop and start treatment at the discretion of their physician, and there were no restrictions on concomitant medications. Approximately one-half of the patients were on a thiopurine analog for the duration of the study, and approximately one-fourth were unable to quit cigarette smoking after surgery. Although the sample size was too small to analyze, concomitant immunomodulators and smoking did not seem to impact on Crohn’s disease recurrence. Interestingly, nearly two-thirds of our patients stopped their anti-TNF treatment, with most because of personal preference rather than an adverse event. Despite these limitations, we had complete medication, endoscopic, and surgical data on all patients for at least 5 years after surgery. In conclusion, infliximab administered within 4 weeks of an ileocolonic resection and continued for at least 1 year prevents endoscopic and surgical recurrence. Stopping infliximab after 1 year of postoperative remission results in endoscopic and surgical recurrence. Initiation of infliximab in response to postoperative endoscopic recurrence reduces mucosal inflammation and prevents re-resection in those patients with less advanced disease (i2) but not severe disease (i4). Patients at a high risk for postoperative Crohn’s disease recurrence may benefit from long-term anti-TNF maintenance. Results from postoperative anti-TNF studies with larger sample size and randomization to immediate “top down” treatment vs waiting for endoscopic recurrence are anticipated.
-
2014
Supplementary Material Note: To access the supplementary material accompanying this article, visit the online version of Clinical Gastroenterology and Hepatology at www.cghjournal.org, and at http://dx.doi.org/10.1016/j.cgh.2013.12.035.
References 1. Cosnes J, Nion-Larmurier I, Beaugerie L, et al. Impact of increasing use of immunosuppressants in Crohn’s disease on the need for intestinal surgery. Gut 2005;54:237–241. 2. Lazarev M, Ullman T, Schraut WH, et al. Small bowel resection rates in Crohn’s disease and the indication for surgery over time: experience from a large tertiary care center. Inflamm Bowel Dis 2010;16:830–835. 3. Terdiman JP. Prevention of postoperative recurrence in Crohn’s disease. Clin Gastroenterol Hepatol 2008;6:616–620. 4. Bennell O, Lapidus A, Hellers G. Risk factors for surgery and postoperative recurrence in Crohn’s disease. Ann Surg 2000; 231:38–45. 5. Rutgeerts P, Geboes K, Vantrappen G, et al. Predictability of the postoperative course of Crohn’s disease. Gastroenterology 1990;99:956–963. 6. Rutgeerts P, Hield M, Beboes K, et al. Controlled trial of metronidazole treatment for prevention of Crohn’s disease recurrence after ileal resection. Gastroenterology 1995;108:1617–1621. 7. Rutgeerts P, Van Assche G, Vermeire S, et al. Ornidazole for prophylaxis of postoperative Crohn’s disease recurrence: a randomized, double-blind, placebo-controlled trial. Gastroenterology 2005;128:856–861. 8. Hanauer SB, Korelitz BI, Rutgeerts P, et al. Postoperative maintenance of Crohn’s disease remission with 6mercaptopurine, mesalamine, or placebo: a 2-year trial. Gastroenterology 2004;127:723–729. 9. Ardizzone S, Maconi G, Sampietro GM, et al. Azathioprine and mesalamine for the prevention of relapse after conservative surgery for Crohn’s disease. Gastroenterology 2005;127:730–740. 10. Sorrentino D, Terrosu G, Avellini C, et al. Prevention of postoperative recurrence of Crohn’s disease by infliximab. Eur J Gastroenterol Hepatol 2006;18:457–459. 11. Sorrentino D, Terrosu G, Avellini C, et al. Infliximab with low-dose methotrexate for prevention of postsurgical recurrence of ileocolonic Crohn’s disease. Arch Intern Med 2007; 167:1804–1807. 12. Regueiro M, Schraut W, Baidoo L, et al. Infliximab prevents Crohn’s disease recurrence after ileal resection. Gastroenterology 2009;136:441–450. 13. Aguas M, Bastida G, Cerrillo E, et al. Adalimumab in prevention of postoperative recurrence of Crohn’s disease in high-risk patients. World J Gastroenterol 2012;18:4391–4398. 14. De Cruz P, Kamm MA, Hamilton AL, et al. Adalimumab prevents postoperative Crohn’s disease recurrence and is superior to thiopurines: early results from the POCER Study. Gastroenterology 2012;142:S212.
Infliximab Prevents Postoperative Crohn’s Disease
9
17. Papamichael K, Archavlis E, Lariou C, et al. Adalimumab for prevention and/or treatment of post-operative recurrence of Crohn’s disease: a prospective, two-year, single center, pilot study. J Crohn’s Colitis 2012;6:924–931. 18. Sorrentino D, Paviotti A, Terrosu G, et al. Low-dose maintenance therapy with infliximab prevents postsurgical recurrence of Crohn’s disease. Clin Gastroenterol Hepatol 2010;8:591–599. 19. Yoshida K, Ken F, Hiroki I, et al. Scheduled infliximab monotherapy to prevent recurrence of Crohn’s disease following ileocolic or ileal resection: a 3-year prospective randomized open trial. Inflamm Bowel Dis 2012;18:1617–1623. 20. Yamamoto T, Umegae S, Matsumoto K. Impact of infliximab after early endoscopic recurrence following ileocolonic resection of Crohn’s disease: a prospective pilot study. Inflamm Bowel Dis 2009;15:1460–1466. 21. Sorrentino D, Terrosu G, Paviotti A, et al. Early diagnosis and treatment of postoperative endoscopic recurrence of Crohn’s disease: partial benefit by infliximab—a pilot study. Dig Dis Sci 2012;57:1341–1348. 22. Hyams J, Crandall W, Kugathasan S, et al. Induction and maintenance infliximab therapy for the treatment of moderateto-severe Crohn’s disease in children. Gastroenterology 2007; 132:863–873. 23. D’Haens G, Baert F, Assche G, et al. Early combined immunosuppression or conventional management in patients with newly diagnosed Crohn’s disease: an open randomized trial. Lancet 2008;371:660–667. 24. Schreiber S, Colombel JF, Bloomfield R, et al. Increased response and remission rates in short-duration Crohn’s disease with subcutaneous certolizumab pegol: an analysis of PRECiSE 2 randomized maintenance trial data. Am J Gastroenterol 2010; 105:1574–1582. 25. Colombel JF, Sandborn WJ, Reinisch W, et al. Infliximab, azathioprine, or combination therapy for Crohn’s disease. N Engl J Med 2010;362:1383–1395. 26. Waugh AW, Garg S, Matic K, et al. Maintenance of clinical benefit in Crohn’s disease patients after discontinuation of infliximab: long-term follow-up of a single centre cohort. Aliment Pharmacol Ther 2010;9:1129–1134. 27. Louis E, Mary JY, Vernier-Massouille G, et al. Maintenance of remission among patients with Crohn’s disease on antimetabolite therapy after infliximab therapy is stopped. Gastroenterology 2012;142:63–70. 28. Molnar T, Lakatos PL, Farkas K, et al. Predictors of relapse in patients with Crohn’s disease in remission after 1 year of biological therapy. Aliment Pharmacol Ther 2013;37: 225–233.
Reprint requests Address requests for reprints to: Miguel Regueiro, MD, University of Pittsburgh School of Medicine, Inflammatory Bowel Disease Center, Division of Gastroenterology, Hepatology and Nutrition, University of Pittsburgh Medical Center, 200 Lothrop Street, PUH- C Wing Mezzanine Level, Pittsburgh, Pennsylvania 15213. e-mail: [email protected]; fax: (412) 383-8913.
15. Araki T, Uchida K, Okita Y, et al. The impact of postoperative infliximab maintenance therapy on preventing the surgical recurrence of Crohn’s disease: a single-center paired casecontrol study. Surg Today 2014;44:291–296.
Acknowledgments The authors thank the Sigal Family Research & Education Inflammatory Bowel Disease Fund for their support and Marilyn Pesci, RN, for her assistance with study coordination and data collection.
16. Savarino E, Dulbecco P, Bodini G, et al. Prevention of postoperative recurrence of Crohn’s disease by adalimumab: a case series. Eur J Gastroenterol Hepatol 2012;24:468–470.
Conflicts of interest This author discloses the following: Miguel Regueiro has served as a consultant for Abbvie, Janssen, Shire, Takeda, and UCB. The remaining authors disclose no conflicts.
9.e1
Regueiro et al
Clinical Gastroenterology and Hepatology Vol.
-,
No.
-
Supplementary Figure 1. Kaplan–Meier plot of surgery-free survival stratified by initial random assignment and cumulative time on infliximab therapy (<60% vs 60%). The highly divergent survival curves did not achieve statistical significance (P ¼ .23), likely because of low statistical power. IFX, infliximab.
Supplementary Table 1. Baseline Demographics at Entry Into Randomized Controlled Trial12 Infliximab (n ¼ 11)
Placebo (n ¼ 13)
Baseline demographics
n
%
n
%
P valuea
Female Age 40 y Active smoker Duration of Crohn’s disease >10 y Disease location at surgery Ileum only Ileum and colon Phenotype B2 (stricture) B3 (fistula) Prior infliximab Surgical resectionsb 1 2 3 Concomitant immunomodulator Mesalamine agent
5 6 5 7
45.5 54.5 45.5 63.6
3 5 1 5
23.1 38.5 7.7 38.5
.39 .68 .06 .41 .78
2 9
18.2 81.8
3 10
23.1 76.9
0 11 3
0.0 100.0 30.0
2 11 5
15.4 84.6 38.5
7 3 1 4 1 Median 43 13
63.6 27.3 9.1 36.4 9.1 25%, 75% 28, 49 1, 19
9 3 1 7 4 Median 32 9
69.2 23.1 7.7 53.8 30.8 25%, 75% 26, 45 2, 12
Age (y) Duration of Crohn’s disease (y)c
.48
NOTE. Reprinted with permission from Regueiro et al.12 a P values were calculated by exact methods, c2 tests for categorical variables, and Wilcoxon tests for continuous variables. b Includes surgical resection related to study enrollment. c Duration of less than 1 year was coded as 0.5 years.
1.0 1.0
.44 .33 .34 .35
Postoperative Therapy With Infliximab Prevents Long-term Crohn’s Disease Recurrence Miguel Regueiro,* Kevin E. Kip,‡ Leonard Baidoo,* Jason M. Swoger,* and Wolfgang Schraut§ *Inflammatory Bowel Disease Center and Division of Gastroenterology, Hepatology and Nutrition, and §Division of Gastrointestinal Surgery, Department of Surgery, University of Pittsburgh Medical Center, Pittsburgh, Pennsylvania; and ‡ Research Center, College of Nursing, University of South Florida, Tampa, Florida BACKGROUND & AIMS:
A previous randomized, placebo-controlled study showed that infliximab maintenance therapy prevented recurrence of Crohn’s disease 1 year after an ileocolonic resection. We evaluated recurrence of Crohn’s disease, on the basis of endoscopic examination and/or the need for additional surgical resection, beyond the first postoperative year.
METHODS:
In a prospective, open-label, long-term follow-up study, 24 patients previously randomly assigned to receive infliximab for 1 year after an ileocolonic resection were given the option to continue, stop, or start infliximab therapy. The primary end point was the time to recurrence of Crohn’s disease, on the basis of endoscopic evidence (endoscopic recurrence), from the initial assignment to postoperative infliximab or placebo. Secondary end points were rate of endoscopic recurrence, time to reoperation, and rate of surgical recurrence in relation to the total time on infliximab.
RESULTS:
All patients were followed for at least 5 years after surgery. Patients assigned to the infliximab group in the first year after surgery had a longer mean time to first endoscopic recurrence (1231 – 747 days) than patients originally assigned to the placebo group (460 – 121 days, P [ .003). Colonoscopies identified Crohn’s disease recurrence in 22.2% of patients who received long-term infliximab and in 93.9% of those not on infliximab (P < .0001). Compared with no infliximab, the adjusted rate ratio for being in endoscopic remission while on infliximab was 13.47 (95% confidence interval, 3.52–61.53; P [ .0001). Patients originally assigned to the infliximab group had a mean longer time to surgery (1798 – 359 days) than patients originally assigned to the placebo group (1058 – 529 days, P [ .04). The rate of surgical recurrence (required additional surgical resection) was significantly lower among patients who received infliximab for most of the followup period than patients who received it for shorter periods (20.0% vs 64.3%, P [ .047).
CONCLUSIONS:
Postoperative infliximab maintenance beyond 1 year prevents recurrence of Crohn’s disease.
Keywords: IBD; Anti-TNF; Clinical Trial; Tumor Necrosis Factor; Treatment; Drug.
espite the increased use of immunomodulators and anti–tumor necrosis factor alpha (anti-TNF) medications for the treatment of inflammatory bowel disease, approximately two-thirds of Crohn’s disease patients will require an intestinal resection, and more than 50% will require at least 1 additional surgery in their lifetime.1–3 Cigarette smoking, penetrating disease, recurrent surgery, and postoperative endoscopic recurrence are well-established risk factors that predict subsequent intestinal resection.4,5 Nitroimidazole antibiotics and thiopurine analogs may be effective treatment for prevention of postoperative recurrence.6–9 Recently, anti-TNF agents have shown efficacy in the prevention of Crohn’s disease after intestinal resection.10–12 Most of the studies on the postoperative prevention of Crohn’s disease with anti-TNF agents have focused on
D
1-year clinical and endoscopic outcomes.12–14 Long-term postoperative anti-TNF studies have been limited to approximately 3 years of follow-up and primarily consist of retrospective reviews or case series.10,15–18 There is only 1 open-label infliximab postoperative study that followed patients for 3 years.19 There remain important unanswered questions about long-term postoperative Crohn’s disease management with anti-TNF therapy, such as (1) does initiating infliximab shortly after
Abbreviations used in this paper: AZA, 6-mercaptopurine; TNF, tumor necrosis factor.
azathioprine;
© 2014 by the AGA Institute 1542-3565/$36.00 http://dx.doi.org/10.1016/j.cgh.2013.12.035
6MP,
2
Regueiro et al
surgery prevent long-term recurrence and additional surgery?, (2) does starting infliximab in response to endoscopic recurrence induce remission and avoid future surgery?, and (3) does stopping infliximab after the first postoperative year of remission lead to recurrence or sustain anti-TNF–free remission? This is a prospective, open-label study from our previously published randomized controlled trial of infliximab vs placebo for postoperative Crohn’s disease.12 We evaluated long-term endoscopic and surgical recurrence in patients continuing, newly starting, or stopping infliximab after the first postoperative year.
Methods Study Design The study design consisted of a prospective, openlabel long-term follow-up study (>5 years) of patients from our previously published randomized controlled trial of infliximab vs placebo for postoperative Crohn’s disease.12 The study was conducted at the Inflammatory Bowel Disease Center at the University of Pittsburgh Medical Center, with collection of data beyond the initial 1-year randomized controlled study approved by the Institutional Review Board at the University of Pittsburgh Medical Center. Patients were followed through December 2012 to obtain at least 5 years of data on all patients.
Patients Between 2004 and 2007, 24 adult patients underwent an ileocolonic resection for Crohn’s disease. All patients had been enrolled within 4 weeks of surgery and were randomly assigned in a blocked 1:1 manner to infliximab 5 mg/kg or identical-appearing placebo infusions at 0, 2, 6, and every 8 weeks for 1 year. At the end of 1 year, an ileocolonoscopy was performed, and an ileal Crohn’s disease activity score was recorded on the basis of the methods of Rutgeerts et al.5 On completion of the 1-year postoperative randomized controlled trial, patients were followed for at least 5 years from the date of surgery, during which time they received standard of care treatment by their gastroenterologist. All patients were treated at University of Pittsburgh Medical Center, with clinical information recorded in the electronic medical record system. The data captured in the longterm postoperative follow-up period included colonoscopy with ileal endoscopic scores, additional surgeries, medications, indication for stopping infliximab, and disease behavior.
Treatment At the 1-year postoperative ileocolonoscopy, patients and physicians were blinded to the study treatment
Clinical Gastroenterology and Hepatology Vol.
-,
No.
-
(placebo or infliximab). At that time, the physician informed the patient of their endoscopic findings and discussed the options of open-label infliximab every 8 weeks or other treatment. For those who received infliximab after the postoperative year 1 colonoscopy, this meant initiation of infliximab for the placebo patients and uninterrupted maintenance for those who had been randomized to infliximab. For those who did not receive infliximab after the first postoperative year, this meant continuation without infliximab for the placebo patients and stopping infliximab for those who had been randomized to it. Beyond the 1-year randomized study, physicians had the option to start or stop any therapy as they deemed appropriate.
Study Outcomes The primary study outcome was the time to endoscopic Crohn’s disease recurrence by initial postoperative assignment to infliximab or placebo. Beyond the first year of the randomized controlled study, ileocolonoscopy was performed at the discretion of the treating gastroenterologist. Although the optimal time interval for colonoscopic inspection beyond the first postoperative year has not been defined, it is our practice to complete a second ileocolonoscopy within 2 years of ileocolonic resection. The Rutgeerts endoscopic score was recorded.5 Although this scoring system has never been validated, it is the most widely accepted endoscopic metric in postoperative Crohn’s disease studies. The scores were as follows: i0, no lesions; i1, 5 or fewer aphthous lesions; i2, more than 5 aphthous lesions with normal mucosa between the lesions or skip areas of larger lesions or lesions confined to the ileocolonic anastomosis; i3, diffuse aphthous ileitis with diffusely inflamed mucosa; and i4, diffuse inflammation with large ulcers, nodules, and/or narrowing. Endoscopic recurrence was defined by an endoscopic score of i2, i3, or i4. Endoscopic remission was defined by a score of i0 or i1. Secondary study outcomes included (1) rate of endoscopic recurrence stratified by use vs non-use of infliximab and 6-mercaptopurine/azathioprine (6MP/ AZA), (2) time to reoperation by the initial assignment to infliximab or placebo, and (3) rate of endoscopic and surgical recurrence by total time on infliximab. Surgical recurrence was defined as any intestinal resection for Crohn’s disease or a complication of Crohn’s disease.
Statistical Methods Baseline demographic and clinical characteristics of the study sample are presented as means and standard deviations for continuous variables and percentages for categorical variables. In patient-level analyses, the time on study, percentage of time on infliximab, and number of colonoscopies performed were compared by initial
-
2014
random assignment by use of Student t tests. Rates of endoscopic disease recurrence and recurrent surgery at any time during follow-up were compared by random assignment by use of Fisher exact test. Among patients who experienced endoscopic disease recurrence, the mean number of days to recurrence was compared by random assignment by use of Student t tests. Because of small numbers, corresponding days to recurrent surgery were compared by use of the Wilcoxon rank sum test. Kaplan–Meier estimates of surgery-free survival were calculated by random assignment, stratified by approximate median time on infliximab therapy during the full study period (<60% of the time vs 60% of the time). The cutoff of >60% of time was assigned to identify a group of patients who had been on infliximab for the majority of the study period. The log-rank statistic was used to compare the 4 respective survival curves. In analyses in which colonoscopy was the unit of analysis (n ¼ 84), generalized estimating equations specifying the binomial distribution, log link, and a compound symmetry correlation structure were used to estimate rate ratios of being in remission in relation to use of biologic medications at the time of endoscopic evaluation. Univariate estimates and estimates adjusted for concomitant use of 6MP/AZA, and number of days between successive colonoscopies, are presented. A 2-sided P value <.05 was used to define statistical significance.
Results Study Population The 24 participants in the original study (infliximab, n ¼ 11; placebo, n ¼ 13) were enrolled between October 2004 and April 2007 and completed their participation by December 2012. The mean length of study participation was 6.5 0.6 years, with a minimum of 5.4 and maximum of 7.7 years. Mean years of participation were similar between patients assigned to infliximab (6.4 0.8) or to placebo (6.5 0.5, P ¼ .88). As previously reported, baseline characteristics of the 2 groups were similar for age, gender, duration of Crohn’s disease, disease behavior, disease location, prior infliximab exposure, and prior surgical resection (Supplementary Table 1).12 For the full cohort, the mean age at entry was 37.8 13.8 years, 33.3% were female, 25.0% were current smokers and continued to smoke throughout the longterm follow-up, 50.0% had Crohn’s disease for more than 10 years, and 91.7% had surgery for a penetrating/ perforating complication. For 66.7% of patients, this was their first intestinal resection for Crohn’s disease, for 25% this was their second surgery, and for 8.3% this was their third surgery. There were 45.8% of patients taking 6MP/AZA, and 33.3% had received prior treatment with infliximab. In the long-term follow-up period, none of the patients were prescribed 5-aminosalicylates, antibiotics, or methotrexate for their Crohn’s disease.
Infliximab Prevents Postoperative Crohn’s Disease
3
Follow-up Colonoscopies, Surgeries, and Use of Infliximab After the index (trial entry) surgery, the mean number of colonoscopies performed during follow-up ranged between 2 and 5 and was similar between infliximab-assigned (3.4 1.0) and placebo-assigned (3.5 1.1) patients (P ¼ .69). Despite initial random assignment to infliximab, the mean percentage of time during the entire course of the study that these patients were on infliximab therapy (50.3 40.4) was similar to the placebo-assigned (53.0 29.6) group (P ¼ .85). Of the 13 placebo-assigned patients, 1 never started infliximab in long-term follow-up, and 12 started infliximab therapy in the year that followed their 1-year initial placebo regimen (1 patient started within the first year after undergoing recurrent surgery). Overall, 23 of the 24 patients (95.8%) initiated infliximab therapy at some time during the study, and of these patients, 14 (60.9%) stopped therapy for the following reasons: infusion reaction (n ¼ 2), lupus-like reaction (n ¼ 1), breast cancer (n ¼ 1), patient decision (n ¼ 8), and loss of insurance (n ¼ 2). None of the patients stopped infliximab because of a loss of response. Percentage use of infliximab during follow-up was similar between nonsmokers and smokers (42.5 36.5 vs 54.2 35.0, P ¼ .50).
Primary Outcome Time to endoscopic recurrence by assignment to infliximab or placebo (endoscopy, patient-level analysis). Eight of the 11 infliximab patients elected to stop
infliximab after the 1-year randomized controlled trial (Figure 1). All 8 patients had endoscopic recurrence after stopping infliximab; 5 have had another surgery, and 3 have not. One infliximab patient had an endoscopic recurrence (i3) at the end of the 1-year study but elected to stay on infliximab for control of arthritis symptoms. Despite the infliximab, this patient has had a persistent i3 endoscopic score through 6 years of follow-up but has not had Crohn’s disease symptoms or required another resection. The 2 patients originally assigned to infliximab who continued infliximab for the duration of long-term follow-up have remained in remission (i0, i1). Of the 8 infliximab patients who stopped infliximab at some time, the time to recurrence was relatively abrupt, with a range of 8.0–30.0 months and mean of 18.2 7.2 months. Twelve of the 13 patients (92.3%) in the placebo arm showed endoscopic evidence of disease recurrence at some point. Of these 13 patients, 12 elected to initiate infliximab after the 1-year randomized controlled trial. The one who did not elect to start infliximab had an ileal score of 1 at 1 year but ultimately progressed to i4 and had another resection (Figure 1). Seven of the 12 (2 were i3, 5 were i2) who initiated infliximab responded with
4
Regueiro et al
Clinical Gastroenterology and Hepatology Vol.
-,
No.
-
Figure 1. Long-term (>5 years) follow-up of patients initially randomized to infliximab or placebo. Recurrence is defined as endoscopic recurrence (i2, i3, i4). Remission is defined as endoscopic score of i0 or i1. Surgery is defined as intestinal resection for recurrent Crohn’s disease or complication of Crohn’s disease. All surgery patients had 1 intercurrent surgery during follow-up, except 1 of the infliximab patients who stopped infliximab had 2 surgeries and the 1 placebo patient who did not receive anti-TNF treatment had 2 surgeries. IFX, infliximab; PBO, placebo; RCT, randomized controlled trial.
endoscopic remission and have avoided additional surgery. Five of the 12 (4 were i4 and 1 was i3) who initiated infliximab did not achieve endoscopic remission and required another surgical resection. Overall, patients
originally assigned to infliximab had a longer time to first endoscopic recurrence compared with patients originally assigned to placebo (1231 747 days vs 460 121 days, P ¼ .003) (Figure 2).
Figure 2. Plot of days to endoscopic disease recurrence by initial random assignment to infliximab or placebo. The black-filled portion of each horizontal bar depicts number of days on infliximab therapy; the gray portion of each horizontal bar depicts number of days not on infliximab therapy.
-
2014
Infliximab Prevents Postoperative Crohn’s Disease
Secondary Outcomes Rate of endoscopic recurrence by overall use of infliximab and/or immunomodulators (endoscopy, colonoscopylevel analysis). For the 24 patients, a total of 84
colonoscopies were performed after entry into the trial and during long-term follow-up. The median time between successive follow-up colonoscopies was 451 days, with interquartile range of 406–734 days and minimum and maximum of 65 and 1515 days, respectively. The median time between colonoscopies did not differ between those originally assigned to placebo or to infliximab. Twenty-four colonoscopies occurred at the completion of the1-year trial period, and the remaining colonoscopies were distributed as second (n ¼ 24), third (n ¼ 22), fourth (n ¼ 8), and fifth (n ¼ 6) year thereafter. Endoscopic ileal disease activity scores at the time of colonoscopy were i0 (n ¼ 30, 35.7%), i1 (n ¼ 10, 11.9%), i2 (n ¼ 15, 17.9%), i3 (n ¼ 16, 19.0%), and i4 (n ¼ 13, 15.5%). This corresponded to endoscopic recurrence in 44 of 84 colonoscopies (52.4%). Among the colonoscopic evaluations, 12 (14.3%) were performed after a second intercurrent surgical resection. Biologic medication status at the time of colonoscopy, defined as medication being given during the majority (>60%) of time since the previous evaluation, was distributed as none (n ¼ 33, 39.3%), infliximab (n ¼ 45, 53.6%), and adalimubab or certolizumab (n ¼ 6, 7.1%) (Table 1). Twenty-five colonoscopies (29.8%) were done on patients on 6MP/AZA (ie, irrespective of biologic therapy). As shown in Figure 3, 77.8% of colonoscopies done on patients receiving infliximab disclosed disease in remission compared with just 6.1% of colonoscopies done on patients not receiving biologic therapy (P < .0001). This analysis reflects the intermittent nature of use of infliximab and strong temporal relationship between exposure to infliximab and the probability of being in disease remission. To illustrate, compared with those not
5
on biologic therapy, the adjusted rate ratio of patients being in remission while on infliximab was 13.47 (95% confidence interval, 3.52–61.53; P ¼ .0001). A similar protective effect (50% remission, P ¼ .02) was observed for other biologic therapy (adalimubab or certolizumab), although only 6 colonoscopies were evaluated in this circumstance. Results were not appreciably different after statistical adjustment for age, gender, duration of Crohn’s disease, infliximab use before trial entry, surgical resection before trial entry, smoking status, or concomitant use of 6MP/AZA. Being a smoker was not associated with endoscopic recurrence (rate ratio ¼ 0.87; 95% confidence interval, 0.52–1.47; P ¼ .61). In subgroup analyses (Figure 4) the highly protective effect of infliximab on endoscopic recurrence was evident. Only 22.2% of colonoscopies in which the patient was taking infliximab had endoscopic recurrence, compared with 93.9% recurrence in those not on infliximab (P ¼ .0002). The role of 6MP/AZA in combination with or as primary treatment of postoperative Crohn’s disease was also evaluated. Of the 26 colonoscopies in which use of both infliximab and 6MP/AZA was absent, all 26 (100%) showed endoscopic evidence of disease recurrence. This compared with just 18.5% of the 27 colonoscopies in which infliximab was being received in the absence of 6MP/AZA (P < .0001). The rates of endoscopic recurrence were not significantly different between those on monotherapy infliximab compared with those on combination infliximab with 6MP/AZA (18.5% vs 27.8%, P ¼ .49). Finally, 5 of the 7 colonoscopies (71.4%) in which 6MP/AZA was administered without infliximab showed evidence of disease recurrence, which was not statistically different from those on no medications (100%, P ¼ .08). Time to surgical recurrence by assignment to infliximab or placebo (surgery, patient-level analysis). Five of
11 infliximab-assigned patients (45.5%) and 6 of 13 placebo-assigned patients (46.2%) underwent recurrent
Table 1. Rate Ratios of Being in Remission (i0, i1) During Follow-up Colonoscopies and in Relation to Use of Biologic Medications n
No. in remission
% in remission
Unadjusted RR
Adjusted RRa
95% CI
P value
33 45 6 33 51
2 35 3 2 38
6.1 77.8 50.0 6.1 74.5
1.0 12.98 8.21 1.0 12.34
1.0 13.47 9.09 1.0 12.95
— 3.52–61.53 1.75–47.26 — 3.35–50.10
— .0001 .009 — .0002
Excluding colonoscopies performed after intercurrent surgery None 30 2 6.7 Infliximab 40 30 75.0 Other biologic 2 0 0.0 None 30 2 6.7 Any biologic 42 30 71.4
1.0 12.01 — 1.0 10.74
1.0 12.44 — 1.0 11.18
— 3.19–48.47 — — 2.84–44.03
— .0003 — — .0006
Biologic use None Infliximab Other biologic None Any biologic
CI, confidence interval; RR, rate ratio. a Adjusted for concomitant use of 6MP/AZA and number of days between successive colonoscopies. Within-subject correlation of multiple colonoscopies modeled by using a compound symmetry correlation structure.
6
Regueiro et al
Clinical Gastroenterology and Hepatology Vol.
-,
No.
-
assigned patients without reoperation received infliximab for most of the post 1-year trial follow-up period. Rate of recurrence by total time on infliximab. Among patients on infliximab therapy for at least 60% of the full study period and irrespective of initial random assignment, the rate of surgical re-resection was significantly lower compared with those with less frequent use of infliximab (20.0% vs 64.3%, P ¼ .047, Figure 6).
Discussion
Figure 3. Percentage of patients in remission (ileal score, i0 or i1) of Crohn’s disease at 5–8 years of endoscopic evaluation by use vs non-use of biologic therapy in the period that preceded endoscopic evaluation.
surgery during long-term follow-up (Figure 5). Of the 11 infliximab-assigned patients, reoperations during the years of follow-up were distributed as follows: 0 (n ¼ 6, 54.5%), 1 (n ¼ 4, 36.4%), and 2 (n ¼ 1, 9.1%). This was similar (P ¼ .98) to the distribution among the 13 placebo-assigned patients: 0 (n ¼ 7, 53.8%), 1 (n ¼ 5, 38.5%), and 2 (n ¼ 1, 7.7%). Despite this similar rate of re-resectional surgery, patients originally assigned to infliximab had a longer time interval to surgery compared with patients originally assigned to placebo (1798 359 days vs 1058 529 days, P ¼ .04) (Supplementary Figure 1). As shown in Figure 5, four of the 5 infliximab-treated patients who underwent reoperation during long-term follow-up did so after ceasing therapy at the end of the 1-year trial period. Four of the 6 infliximab-assigned patients without reoperation remained on infliximab for all or nearly all of the follow-up period. Five of the 7 placebo-
This is the longest postoperative Crohn’s disease study to date and provides insight into the natural course of disease after surgical resection as well as the impact of postoperative treatment. Patients originally assigned to infliximab had a significantly longer time to first endoscopic recurrence and the need for re-resection compared with placebo patients. When considering each colonoscopy as a separate event, more than three-fourths of the colonoscopies done on patients on infliximab revealed endoscopic remission, compared with less than 10% of those on no biologic therapy. The endoscopic recurrence rates were similar for patients receiving infliximab monotherapy or in combination with 6MP/AZA, and the thiopurines alone did not protect against recurrence. Patients were stratified into those on infliximab for the majority of time (>60%) and those for less than that. Irrespective of initial assignment to placebo or infliximab, among patients on infliximab therapy for the majority of the entire follow-up period, there was a higher rate of endoscopic remission and lower rate of surgical recurrence compared with those not on infliximab. Does initiating infliximab shortly after surgery prevent long-term recurrence and additional surgery? All patients who initiated infliximab after surgery and continued long-term have avoided surgery for nearly 8
Figure 4. Colonoscopylevel analysis of rate of endoscopic disease recurrence (i2, i3, i4) stratified by use vs non-use of infliximab and by 6MP/ AZA status overall and in absence of use of infliximab. The 6 colonoscopies with adalimumab or certolizumab were excluded. IFX, infliximab.
-
2014
Infliximab Prevents Postoperative Crohn’s Disease
7
Figure 5. Plot of days to recurrent surgery by initial random assignment to infliximab or placebo. The black-filled portion of each horizontal bar depicts number of days on infliximab therapy; the gray portion of each horizontal bar depicts number of days not on infliximab therapy.
years. These findings are consistent with other published studies of shorter duration of follow-up. The studies by Sorrentino et al11 and Papamichael et al17 reported 2year endoscopic recurrence of 0% and 25% with infliximab and adalimumab, respectively. Similarly, Yoshida et al19 found a 3-year clinical remission rate of 93% and 1-year endoscopic recurrence of 21% in patients started on infliximab within 4 weeks of surgery. Savarino et al16 reported a 3-year endoscopic recurrence of 0% in patients initiating adalimumab 2 weeks after an
Figure 6. Crude rates of surgery during entire study period stratified by initial random assignment and cumulative time on infliximab therapy (<60% vs 60%). IFX, infliximab; pts, patients.
intestinal resection. In the study by Araki et al,15 the 3year surgical recurrence rate was only 3% in patients administered infliximab within 8 weeks of surgery compared with 34% not treated with infliximab. Although none of these studies are randomized and all have small numbers, the low rates of endoscopic and surgical recurrence are consistent with our results. Does starting infliximab in response to endoscopic recurrence induce remission and avoid future surgery? It is not known whether initiating anti-TNF treatment immediately after surgery prevents future Crohn’s disease complications and surgery more effectively than waiting to treat endoscopic recurrence. The patients who received infliximab within 4 weeks of surgery had a longer time to endoscopic and surgical recurrence compared with the placebo patients who started infliximab 1 year after surgery in response to endoscopic recurrence. All of the patients who initiated infliximab within 4 weeks of surgery and continued long-term treatment have avoided additional surgery. In contrast, 58% of the placebo patients who started infliximab in response to 1-year endoscopic recurrence have had improvement in endoscopic inflammation, but nearly one-half have required additional surgery. It is possible that performing a colonoscopy sooner than 1 year after surgery would have led to earlier initiation of treatment and better outcomes. There have been several postoperative Crohn’s disease studies that have initiated anti-TNF treatment in response to early endoscopic recurrence. Yamamoto et al20 showed that 75% of patients starting infliximab 6 months after
8
Regueiro et al
surgery for endoscopic recurrence improved endoscopic inflammation compared with 38% of AZA and 0% of mesalamine patients. Complete mucosal healing was achieved in 38% of patients on infliximab when compared with 13% of those receiving AZA and 0% of mesalamine-treated patients. Sorrentino et al21 reported similar results, with 69% of patients having endoscopic improvement and 54% with complete mucosal healing after receiving infliximab for 6-month postoperative Crohn’s disease recurrence. In a study by Papamichael et al,17 adalimumab administered to patients with endoscopic recurrence 6 months after surgery was able to improve endoscopic inflammation in 60%. All 3 of these studies initiated anti-TNF treatment in response to endoscopic recurrence at 6 months. The overall endoscopic improvement rates (60%–70%) are similar to 58% we found in placebo patients initiating infliximab 1 year after surgery. Whether earlier endoscopic surveillance would improve these rates further and whether this strategy would approach the consistently high endoscopic remission rate (80%–100%) seen in the studies initiating anti-TNF treatment within 1 month of surgery are not known. The postoperative model provides a glimpse into the timing of medical therapy vis à vis mucosal healing and thus prevention or postponement of recurrent Crohn’s disease. Patients who undergo a resection that removes the entirety of the diseased intestine may be considered in the “deepest” remission. Starting “top down” biologic treatment immediately after surgery rather than waiting for disease recurrence provides the best rates of longterm prevention.10–19 This parallels the experience with previously published medical treatment trials that showed greater benefit in patients with short Crohn’s disease duration compared with a less robust response in those with longer disease duration.22–25 The reason that Crohn’s disease patients with longer disease duration have an attenuated medication response is currently in the domain of speculation but may reflect irreversible structural bowel damage, vascular changes, or possibly an altered microbiome and cytokine profile. We did find that patients with less severe postoperative endoscopic recurrence responded better to the initiation of infliximab. Patients with an ileal score of 2 were able to induce endoscopic remission, whereas none of the i4 patients and only 2 of the i3 patients had endoscopic improvement. All of the patients with i4 scores and half of the i3 patients required another surgery despite initiation of infliximab. Lacking in the management of postoperative Crohn’s disease is a noninvasive measure to diagnose recurrent mucosal inflammation that would allow initiation of medication before the development of irreversible intestinal changes. Such a diagnostic tool would avoid overtreating postoperative patients who may never develop recurrence and target patients for treatment at the earliest sign of Crohn’s disease activity. Does stopping infliximab after the first postoperative year of remission lead to recurrence or sustain anti-
Clinical Gastroenterology and Hepatology Vol.
-,
No.
-
TNF–free remission? The optimal duration of anti-TNF treatment for patients with clinical and endoscopic remission, ie, deep remission, is not known. Recent studies have suggested that factors such as absence of mucosal inflammation noted by endoscopy or serologic markers and lack of recent corticosteroids or anti-TNF dose intensification may predict those who can stop anti-TNF treatment and avoid relapse.26–28 Most of the patients in these studies were on concomitant immunomodulators at the time of anti-TNF discontinuation. In all 3 studies, however, approximately one-half of the patients relapsed within 1 year of stopping the anti-TNF treatment. In our postoperative study, nearly threefourths of the infliximab patients stopped treatment at 1 year after resection while in complete endoscopic remission. All of these patients subsequently had endoscopic recurrence, and most required another surgery. Sorrentino et al18 reported 83% of their postoperative patients in endoscopic remission for 3 years developed endoscopic recurrence after stopping infliximab. The high postoperative recurrence rates after stopping infliximab may reflect a more aggressive Crohn’s disease course. There are several limitations in our study, mainly the small sample size and the open-label design. Another limitation is that our study was conducted at a tertiary care center, and the patients were at high risk for recurrence; the majority had penetrating disease, onethird had more than 1 prior surgery, and more than half of the patients had failed immunomodulators. The lack of control was also a limitation in that patients could stop and start treatment at the discretion of their physician, and there were no restrictions on concomitant medications. Approximately one-half of the patients were on a thiopurine analog for the duration of the study, and approximately one-fourth were unable to quit cigarette smoking after surgery. Although the sample size was too small to analyze, concomitant immunomodulators and smoking did not seem to impact on Crohn’s disease recurrence. Interestingly, nearly two-thirds of our patients stopped their anti-TNF treatment, with most because of personal preference rather than an adverse event. Despite these limitations, we had complete medication, endoscopic, and surgical data on all patients for at least 5 years after surgery. In conclusion, infliximab administered within 4 weeks of an ileocolonic resection and continued for at least 1 year prevents endoscopic and surgical recurrence. Stopping infliximab after 1 year of postoperative remission results in endoscopic and surgical recurrence. Initiation of infliximab in response to postoperative endoscopic recurrence reduces mucosal inflammation and prevents re-resection in those patients with less advanced disease (i2) but not severe disease (i4). Patients at a high risk for postoperative Crohn’s disease recurrence may benefit from long-term anti-TNF maintenance. Results from postoperative anti-TNF studies with larger sample size and randomization to immediate “top down” treatment vs waiting for endoscopic recurrence are anticipated.
-
2014
Supplementary Material Note: To access the supplementary material accompanying this article, visit the online version of Clinical Gastroenterology and Hepatology at www.cghjournal.org, and at http://dx.doi.org/10.1016/j.cgh.2013.12.035.
References 1. Cosnes J, Nion-Larmurier I, Beaugerie L, et al. Impact of increasing use of immunosuppressants in Crohn’s disease on the need for intestinal surgery. Gut 2005;54:237–241. 2. Lazarev M, Ullman T, Schraut WH, et al. Small bowel resection rates in Crohn’s disease and the indication for surgery over time: experience from a large tertiary care center. Inflamm Bowel Dis 2010;16:830–835. 3. Terdiman JP. Prevention of postoperative recurrence in Crohn’s disease. Clin Gastroenterol Hepatol 2008;6:616–620. 4. Bennell O, Lapidus A, Hellers G. Risk factors for surgery and postoperative recurrence in Crohn’s disease. Ann Surg 2000; 231:38–45. 5. Rutgeerts P, Geboes K, Vantrappen G, et al. Predictability of the postoperative course of Crohn’s disease. Gastroenterology 1990;99:956–963. 6. Rutgeerts P, Hield M, Beboes K, et al. Controlled trial of metronidazole treatment for prevention of Crohn’s disease recurrence after ileal resection. Gastroenterology 1995;108:1617–1621. 7. Rutgeerts P, Van Assche G, Vermeire S, et al. Ornidazole for prophylaxis of postoperative Crohn’s disease recurrence: a randomized, double-blind, placebo-controlled trial. Gastroenterology 2005;128:856–861. 8. Hanauer SB, Korelitz BI, Rutgeerts P, et al. Postoperative maintenance of Crohn’s disease remission with 6mercaptopurine, mesalamine, or placebo: a 2-year trial. Gastroenterology 2004;127:723–729. 9. Ardizzone S, Maconi G, Sampietro GM, et al. Azathioprine and mesalamine for the prevention of relapse after conservative surgery for Crohn’s disease. Gastroenterology 2005;127:730–740. 10. Sorrentino D, Terrosu G, Avellini C, et al. Prevention of postoperative recurrence of Crohn’s disease by infliximab. Eur J Gastroenterol Hepatol 2006;18:457–459. 11. Sorrentino D, Terrosu G, Avellini C, et al. Infliximab with low-dose methotrexate for prevention of postsurgical recurrence of ileocolonic Crohn’s disease. Arch Intern Med 2007; 167:1804–1807. 12. Regueiro M, Schraut W, Baidoo L, et al. Infliximab prevents Crohn’s disease recurrence after ileal resection. Gastroenterology 2009;136:441–450. 13. Aguas M, Bastida G, Cerrillo E, et al. Adalimumab in prevention of postoperative recurrence of Crohn’s disease in high-risk patients. World J Gastroenterol 2012;18:4391–4398. 14. De Cruz P, Kamm MA, Hamilton AL, et al. Adalimumab prevents postoperative Crohn’s disease recurrence and is superior to thiopurines: early results from the POCER Study. Gastroenterology 2012;142:S212.
Infliximab Prevents Postoperative Crohn’s Disease
9
17. Papamichael K, Archavlis E, Lariou C, et al. Adalimumab for prevention and/or treatment of post-operative recurrence of Crohn’s disease: a prospective, two-year, single center, pilot study. J Crohn’s Colitis 2012;6:924–931. 18. Sorrentino D, Paviotti A, Terrosu G, et al. Low-dose maintenance therapy with infliximab prevents postsurgical recurrence of Crohn’s disease. Clin Gastroenterol Hepatol 2010;8:591–599. 19. Yoshida K, Ken F, Hiroki I, et al. Scheduled infliximab monotherapy to prevent recurrence of Crohn’s disease following ileocolic or ileal resection: a 3-year prospective randomized open trial. Inflamm Bowel Dis 2012;18:1617–1623. 20. Yamamoto T, Umegae S, Matsumoto K. Impact of infliximab after early endoscopic recurrence following ileocolonic resection of Crohn’s disease: a prospective pilot study. Inflamm Bowel Dis 2009;15:1460–1466. 21. Sorrentino D, Terrosu G, Paviotti A, et al. Early diagnosis and treatment of postoperative endoscopic recurrence of Crohn’s disease: partial benefit by infliximab—a pilot study. Dig Dis Sci 2012;57:1341–1348. 22. Hyams J, Crandall W, Kugathasan S, et al. Induction and maintenance infliximab therapy for the treatment of moderateto-severe Crohn’s disease in children. Gastroenterology 2007; 132:863–873. 23. D’Haens G, Baert F, Assche G, et al. Early combined immunosuppression or conventional management in patients with newly diagnosed Crohn’s disease: an open randomized trial. Lancet 2008;371:660–667. 24. Schreiber S, Colombel JF, Bloomfield R, et al. Increased response and remission rates in short-duration Crohn’s disease with subcutaneous certolizumab pegol: an analysis of PRECiSE 2 randomized maintenance trial data. Am J Gastroenterol 2010; 105:1574–1582. 25. Colombel JF, Sandborn WJ, Reinisch W, et al. Infliximab, azathioprine, or combination therapy for Crohn’s disease. N Engl J Med 2010;362:1383–1395. 26. Waugh AW, Garg S, Matic K, et al. Maintenance of clinical benefit in Crohn’s disease patients after discontinuation of infliximab: long-term follow-up of a single centre cohort. Aliment Pharmacol Ther 2010;9:1129–1134. 27. Louis E, Mary JY, Vernier-Massouille G, et al. Maintenance of remission among patients with Crohn’s disease on antimetabolite therapy after infliximab therapy is stopped. Gastroenterology 2012;142:63–70. 28. Molnar T, Lakatos PL, Farkas K, et al. Predictors of relapse in patients with Crohn’s disease in remission after 1 year of biological therapy. Aliment Pharmacol Ther 2013;37: 225–233.
Reprint requests Address requests for reprints to: Miguel Regueiro, MD, University of Pittsburgh School of Medicine, Inflammatory Bowel Disease Center, Division of Gastroenterology, Hepatology and Nutrition, University of Pittsburgh Medical Center, 200 Lothrop Street, PUH- C Wing Mezzanine Level, Pittsburgh, Pennsylvania 15213. e-mail: [email protected]; fax: (412) 383-8913.
15. Araki T, Uchida K, Okita Y, et al. The impact of postoperative infliximab maintenance therapy on preventing the surgical recurrence of Crohn’s disease: a single-center paired casecontrol study. Surg Today 2014;44:291–296.
Acknowledgments The authors thank the Sigal Family Research & Education Inflammatory Bowel Disease Fund for their support and Marilyn Pesci, RN, for her assistance with study coordination and data collection.
16. Savarino E, Dulbecco P, Bodini G, et al. Prevention of postoperative recurrence of Crohn’s disease by adalimumab: a case series. Eur J Gastroenterol Hepatol 2012;24:468–470.
Conflicts of interest This author discloses the following: Miguel Regueiro has served as a consultant for Abbvie, Janssen, Shire, Takeda, and UCB. The remaining authors disclose no conflicts.
9.e1
Regueiro et al
Clinical Gastroenterology and Hepatology Vol.
-,
No.
-
Supplementary Figure 1. Kaplan–Meier plot of surgery-free survival stratified by initial random assignment and cumulative time on infliximab therapy (<60% vs 60%). The highly divergent survival curves did not achieve statistical significance (P ¼ .23), likely because of low statistical power. IFX, infliximab.
Supplementary Table 1. Baseline Demographics at Entry Into Randomized Controlled Trial12 Infliximab (n ¼ 11)
Placebo (n ¼ 13)
Baseline demographics
n
%
n
%
P valuea
Female Age 40 y Active smoker Duration of Crohn’s disease >10 y Disease location at surgery Ileum only Ileum and colon Phenotype B2 (stricture) B3 (fistula) Prior infliximab Surgical resectionsb 1 2 3 Concomitant immunomodulator Mesalamine agent
5 6 5 7
45.5 54.5 45.5 63.6
3 5 1 5
23.1 38.5 7.7 38.5
.39 .68 .06 .41 .78
2 9
18.2 81.8
3 10
23.1 76.9
0 11 3
0.0 100.0 30.0
2 11 5
15.4 84.6 38.5
7 3 1 4 1 Median 43 13
63.6 27.3 9.1 36.4 9.1 25%, 75% 28, 49 1, 19
9 3 1 7 4 Median 32 9
69.2 23.1 7.7 53.8 30.8 25%, 75% 26, 45 2, 12
Age (y) Duration of Crohn’s disease (y)c
.48
NOTE. Reprinted with permission from Regueiro et al.12 a P values were calculated by exact methods, c2 tests for categorical variables, and Wilcoxon tests for continuous variables. b Includes surgical resection related to study enrollment. c Duration of less than 1 year was coded as 0.5 years.
1.0 1.0
.44 .33 .34 .35