- Top of page
- PATIENTS AND METHODS
Background: Our objective is to report the outcome of infliximab (IFX) in ulcerative colitis (UC) patients from a single center and to identify predictors of early clinical response.
Methods: The first 100 UC patients (45 female; median age, 37.9 years) who received IFX at a single center were included. Eighty-four patients received 5 mg/kg IFX, and 37 patients received a 3-dose IFX induction at weeks 0, 2, and 6. The Mayo endoscopic subscore, assessed by sigmoidoscopy before inclusion, was 1, 2, and 3 in 5%, 52%, and 43% of patients, respectively. Sixty percent had pancolitis, 63% were on concomitant immunosuppressive therapy, 9% were active smokers, 64% had C-reactive protein ≥5 mg/dL, and 44% were pANCA+/ASCA−. Five patients received IFX because of severe acute colitis refractory to intravenous corticosteroids.
Results: Early complete and partial clinical responses were observed in 41% and 24% of patients. Patients with early clinical response were significantly younger than nonresponders (median age, 35.7 versus 41.6 years, P = 0.041). Patients who were pANCA+/ASCA− had a significantly lower early clinical response (55% versus 76%; odds ratio [OR] = 0.40 (0.16–0.99), P = 0.049). Concomitant immunosuppressive therapy and the use of an IFX induction scheme did not influence early clinical response. Only 1 of 5 patients who received IFX for acute steroid-refractory colitis required colectomy within 2 months.
Conclusions: IFX is an efficient therapy in UC, as shown by 65% early clinical response. A pANCA+/ASCA− serotype and an older age at first IFX infusion are associated with a suboptimal early clinical response.
Ulcerative colitis (UC) is characterized by recurring episodes of inflammation limited to the mucosal layer of the colon, as well as rectal bleeding, diarrhea, and abdominal pain. Most patients with UC can be treated successfully with 5-aminosalicylates, corticosteroids, and/or immunomodulators, such as azathioprine and 6-mercaptopurine.1 Approximately 15% of UC patients will have a severe attack requiring hospitalization during their illness,2 and are treated primarily with high doses of corticosteroids intravenously.3–5 Despite intravenous corticosteroids, severe attacks have a high colectomy rate varying from 38% to 47%.4, 5 In addition, cyclosporine has been validated as an efficacious treatment in severe UC,6, 7 but is frequently associated with toxicity8 and only seems to postpone the inevitable colectomy.9, 10
In contrast to Crohn's disease (CD), UC has historically been considered a T-helper 2-driven disease, with a less prominent role for tumor necrosis factor (TNF). However, 2 lines of evidence suggest an important role of TNF in UC pathogenesis. First, several investigators found increased TNF levels in mucosal tissue, stools, colon perfusates, rectal dialysate, serum, and urine of UC patients with active colitis.11–14 Second, CD571, a monoclonal antibody against human TNF-α, significantly improved UC-like colitis in the cotton top tamarin.15
The initial open label and small randomized-controlled clinical trials with infliximab16, 17 resulted in conflicting data, based on the lack of a placebo-controlled arm, a lack of power, or the different population studied in each trial (steroid-refractory versus patients with moderate-to-severe UC).
The randomized, double-blind, multi-center placebo-controlled ACT1 and ACT2 trials, however, clearly showed a significant benefit of infliximab (IFX) in patients with moderate-to-severe active UC.18 Patients received an induction scheme with placebo or IFX 5 or 10 mg/kg body weight at weeks 0, 2, and 6 and maintenance therapy thereafter (placebo or IFX every 8 weeks), for a total of 22 weeks (ACT2) or 46 weeks (ACT1). A clinical response at 8 weeks was seen in 29%–37% for placebo versus 64%–69% for IFX 5 mg/kg versus 61%–69% for IFX 10 mg/kg (P < 0.001). At 30 and 54 weeks the clinical response rates remained significant better in the IFX-treated group. Furthermore, patients receiving IFX had significant higher clinical remission rates, mucosal healing and they were more able to stop corticosteroids.
The only randomized-controlled trial in patients with severe steroid-refractory UC so far, by Jarnerot et al,19 reported significantly lower colectomy rates 3 months after IFX (29% versus 67%, P = 0.017). Furthermore, the use of infliximab did not have any effect on postoperative complications.
Despite a clear benefit of IFX in UC patients in the ACT1 and ACT2 trials and the study by Jarnerot et al, questions about long-term safety, immunogenicity and cost-effectiveness remain of current interest. Because, as in CD, a subset of UC patients clearly do not respond to IFX, it is important to identify predictors of response to this rather expensive and potentially harmful drug, to allow careful selection of patients.
The aim of our study was to study clinical outcome in a consecutive series of 100 UC patients from a single referral center, and to identify possible predictors of short term clinical and endoscopic response to IFX.
- Top of page
- PATIENTS AND METHODS
As shown in the recent placebo-controlled multicenter ACT1 and ACT2 trials, IFX is an effective treatment in patients with moderate to severe UC.18 In the present work, we report similar results in a cohort of 100 UC patient from a single tertiary referral center. Short-term complete and partial clinical response was observed, respectively, in 41% and 24% of patients. Furthermore, a pANCA+/ASCA− serotype and an older age at first IFX infusion were associated with a less favorable outcome. The use of an IFX induction scheme (weeks 0, 2 and 6) had no influence on short-term clinical response.
Although IFX has been proved effective in both CD and UC, a subset of patients clearly do not respond. In the ACT1 and ACT2 trials, for example, 31 to 39% of patients treated with IFX did not show clinical response after 8 weeks.18 Identifying predictors of response to IFX is an important issue, as it might help in carefully selecting patients who will benefit this more expensive and potential harmful therapy. In CD, some clinical, serologic, and genetic predictors of early response to IFX have been identified,24 but up to now, this has not been the case for IFX treatment in UC.
We found an independent association between the UC-associated serotype pANCA+/ASCA− and a suboptimal early clinical response to IFX (55% versus 76%, OR = 0.40 (0.16–1.00), P = 0.049). Interestingly, similar finding have been suggested in CD. Taylor et al25 reported lower response to IFX in CD patients who were pANCA positive. Our own group also found lower response rates in CD patients with refractory intestinal disease carrying the pANCA+/ASCA− serotype. This association, however, lacked significance (P = 0.067), probably because only 10 of 183 CD patients carried this specific serotype.26 In contrast, Arnott et al27 did not find an association between serotype and IFX response in a smaller heterogeneous cohort of 74 IFX treated CD patients. We should note, however, that although the early clinical response rate was significantly lower in patients with the pANCA+/ASCA− serotype, it was still 55%. Indicating that also in these patients IFX could be administrated in an appropriate clinical setting.
Of interest, 6 UC patients in our cohort had a CD-like pANCA−/ASCA+ serology. Reanalysis of the clinical and histologic findings confirmed the diagnosis of UC.
Furthermore, in univariate analysis, we found a significant association between younger age at first IFX infusion and favorable early clinical response (median age (IQR) 35.7 (25.2–48.6) years versus 41.6 (30.6–57.3) years, Mann-Whitney U test, P = 0.041). This finding was independent of age at diagnosis and disease duration before first IFX infusion, but not from the serotype. In a previously reported paper, similar data were found in a Belgian cohort of CD patients who received IFX for refractory luminal (n = 137) or fistulizing (n = 103) disease.28 Two other papers from the United Kingdom and the United States, however, did not find such association, in a smaller cohort of CD patients of 74 and 100 CD patients, respectively.27, 29
Concomitant immunosuppressive therapy with azathioprine or methotrexate did not influence early clinical response in our cohort of UC patients. However, this does not exclude an influence on the long run. In CD, several studies showed a clear benefit of concomitant immunosuppressive therapy in patients with refractory luminal disease.28, 29
There was no different early clinical outcome after application of an induction scheme with IFX at weeks 0, 2, and 6 or a single infusion with IFX. Up to now, the randomized clinical trials with IFX in UC, only used the induction scheme in all their patients.18 This has become a common practice, mainly to reduce immunogenicity, and not to improve efficacy. In the ACCENT 1 trial, for example, the clinical benefit of 3 infusions over 1 single infusion was only 13%.30 Further studies, including long term responses, are necessary to conclude on these finding.
Because only one-half of the UC patients included in this study underwent an endoscopic evaluation, we were unable to conclude firmly on predictors of early mucosal healing.
During a median (IQR) follow-up of 2.7 (range, 0.4–6.4) years, we did not find predictors of colectomy, but only 11% of the 100 UC patients failed medical therapy. Longer follow-up is undertaken. Interestingly, from the 5 patients who received IFX for severe corticosteroid-refractory disease only 1 patient did not show response to IFX and needed a colectomy within 2 months.
In conclusion, IFX is an efficient therapy in UC. Patients with a pANCA+/ASCA− serotype had a significant lower clinical response 4 to 10 weeks after the IFX administration, but this was still 55%. Concomitant immunosuppressive therapy and the use of an induction scheme with IFX at weeks 0, 2 and 6, did not influence early clinical response.