Infliximab and Azathioprine for Crohn's Disease: A Super-Sonic Combination?

Abstract

Colombel JF, Sandborn WJ, Reinisch W, et al; SONIC Study Group. (Hôpital Claude Huriez and Centre d'Investigation Clinique, Centre Hospitalier Universitaire de Lille, Université Lille Nord de France, Lille, France). Infliximab, azathioprine, or combination therapy for Crohn's disease. N Engl J Med 2010;362:1383–1395. The current recommendation for the treatment of moderate to severe Crohn's disease (CD) is the so-called “step-up” approach (Am J Gastroenterol 2009;104:465–483). However, no trials had compared azathioprine alone to anti-tumor necrosis factor (TNF)-based therapeutic strategies in CD patients who are naïve to immunosuppressants and biological agents. Whether infliximab alone or in combination with azathioprine should be the preferred treatment strategy for CD also remained a subject of active debate. The authors of the Study of Biologic and Immunomodulator Naive Patients in Crohn's Disease (SONIC) trial investigated whether infliximab (Remicade; Centocor, Horsham, PA), azathioprine, or a combination of the 2 demonstrated greater efficacy for induction and maintenance of remission in patients with a history of corticosteroid use, but who were naïve to azathioprine, 6-mercaptopurine, methotrexate, or an anti-TNF biological agent. In this randomized, double-blind, multicenter trial, a total of 508 patients were recruited and randomized to either infliximab (5 mg/kg at weeks 0, 2, and 6, then every 8 weeks; n = 169), azathioprine (2.5 mg/kg daily; n = 170), or a combination of the 2 (n = 169). Each group had similar disease characteristics. Patients were followed for 30 weeks, with optional extension to 50 weeks, and received a follow-up phone call 4 weeks after completion or withdrawal from the trial. The primary end point of the trial was the rate of corticosteroid-free remission at week 26, at which point 318 patients remained in the trial. At week 26, 96 of the patients receiving combination therapy (56.8%), 75 of the patients receiving infliximab (44.4%), and 51 of the patients receiving azathioprine (30.0%) were in corticosteroid-free clinical remission. Importantly, the rates of remission with azathioprine were in line with those previously reported in larger scale trials that used the CD activity index as their measure of efficacy. The study also analyzed a series of secondary end points. The rate of corticosteroid remission at 34, 42, and 50 weeks showed the same trend as at 26 weeks. Mucosal healing was assessed in a subgroup of 309 patients in whom mucosal ulcerations were clearly detected at baseline and who completed the trial period. At week 26, mucosal healing was observed in 47 of 107 patients who were administered the combination therapy (43.9%), 28 of 93 patients administered infliximab alone (30.1%), and 18 of 109 patients administered azathioprine. Improvements were also observed in secondary end points, including mucosal healing and benefit at week 50; again, the benefit was greatest with combination therapy, then infliximab alone, and finally azathioprine alone, for each of these end points. In addition, when patients with baseline mucosal lesions were compared with patients in whom such lesions were not identified, healing was greater. C-reactive protein (CRP) was found to be predictive of a greater chance of treatment success in all 3 groups. The rate of corticosteroid-free clinical remission was higher in a subgroup of patients with high levels of CRP (defined as ≥0.8 mg/dL) who had been administered either the combination therapy or infliximab alone, compared with those patients from this subgroup who had been administered azathioprine. The same trend was also observed in patients with both high CRP and mucosal lesions at baseline. The safety of the 3 different treatment regimens was also compared, and found to be generally similar. Of note, infusion site reactions occurred in significantly more of the group administered infliximab alone (16.6%), compared with the group administered the combination therapy (5.0%) and the azathioprine group (5.6%). In a similar vein to the lower infusion site reactions in the patients administered combination therapy compared with infliximab alone, a lower rate of development of antibodies to infliximab was observed in the patients administered combination therapy (0.9%) compared with infliximab alone (14.6%). The rate of corticosteroid remission at 26 weeks was greatest in patients for whom the antibody test was inconclusive, which indicates the presence of infliximab in the serum, compared with patients for whom the test was positive or negative. The definitive results from the SONIC trial clearly establish the superiority of combination therapy or infliximab monotherapy over azathioprine alone to maintain corticosteroid-free remission. However, the study also raises several questions. First, should we still use azathioprine monotherapy in patients with CD? In a population-based study from northern France, the use of immunosuppressive therapy in pediatric patients was associated with a decreased risk for surgery (Gastroenterology 2008;135:1106–1113). However, in adults, experience from a referral center suggested that despite increasing use of azathioprine in patients with CD over time, surgical rates did not seem to change (Gut 2005;54:2137–2141). In addition, in a French prospective observational cohort study involving 19,486 patients with inflammatory bowel disease (IBD), the incidence of lymphoproliferative disorders was 0.90 per 1000 patient-years among patients receiving thiopurines compared with 0.26 per 1000 patient-years (0.10–0.57) in those who had never received thiopurines (P = .0054; Lancet 2009;374:1617–1625). The SONIC trials clearly show limited efficacy of azathioprine alone compared with the anti-TNF groups, with or without an immunomodulator. Therefore, to maximize clinical efficacy at 1 year, the results of the SONIC trial indicate that the combination of infliximab and azathioprine is preferable to azathioprine monotherapy. Second, should we continue combination therapy beyond 1 year? Even though efficacy of combination therapy was superior to infliximab alone and safety was identical throughout the 3 groups, safety issues should also be taken into account when initiating combination therapy. A meta-analysis suggested that immunomodulators may be associated with an increased risk of non-Hodgkin's lymphoma in adult CD patients (Clin Gastroenterol Hepatol 2009;7:874–881). Furthermore, 23 cases of hepatosplenic T-cell lymphoma were reported in patients with IBD receiving anti-TNF therapy alone or in combination with thiopurines (Inflamm Bowel Dis 2009;15:1281–1282). In addition, immunosuppressive medications, when used in combination with anti-TNF agents, may be associated with an increased risk of opportunistic infections in patients with IBD (Gastroenterology 2008;134:929–936). By contrast, data from both referral centers and randomized controlled trials showed that anti-TNF therapy alone was not associated with an increased risk of malignancies or infections. In Leuven, Belgium, after a median follow-up of 58 months, there was no difference in mortality, malignancies, and infection rate between 734 patients with IBD administered infliximab and control individuals (Gut 2009;58:501–508). Similarly, the Crohn's Therapy, Resource, Evaluation, and Assessment Tool registry, which includes 3179 CD patients who received infliximab, demonstrated that this biological agent was not an independent predictor of serious infections (Clin Gastroenterol Hepatol 2006;4:621–630). In a meta-analysis of 21 placebo-controlled trials enrolling 5356 individuals, anti-TNF therapy did not increase the risk of death, malignancy, or serious infection compared with control arms (Clin Gastroenterol Hepatol 2006;4:621–630). In a pooled analysis of 6 global clinical trials, evaluating the efficacy of adalimumab in 3160 CD patients, the frequency of infections, serious infections, malignancies, and deaths was similar to placebo (Inflamm Bowel Dis 2009;15:1308–1319). Overall, these findings underscore the need to assess the benefit/risk ratio of combination therapy beyond 1 year. Third, is it possible to improve the long-term safety profile of such potentially disease-modifying strategies while maintaining clinical benefit? A way to reduce the risk of lymphoma would be to stop azathioprine or infliximab in patients receiving long-term combination therapy. In a prospective multicenter trial called STORI from the Groupe d'Etudes Thérapeutiques des Affections Inflammatoires du tube Digestif, 115 patients with luminal CD who had been administered combined infliximab and immunosuppressant for ≥1 year and who were in stable remission without steroids for ≥6 months stopped their infliximab treatment and maintained immunomodulators at a stable dose. In multivariate analyses, 4 factors at the time of infliximab discontinuation were predictive of increased risk of relapse, namely CD Endoscopic Index of Severity ≥2, CRP ≥5 mg/L, hemoglobin ≤14.5 g/dL, and infliximab trough levels ≥2 μg/mL (Gastroenterology 2009;136:A-146). Recently, it was shown that withdrawal of azathioprine from patients with CD who had been in clinical remission for ≥6 months under azathioprine–infliximab combination therapy was associated with a high risk of relapse in patients who had either been taking the combination therapy for <27 months and/or in whom biological inflammation was present (Am J Gastroenterol 2010;105:1142–1149). These results therefore suggest that it may be possible to identify a subgroup of patients who are able to stop azathioprine treatment or anti-TNF therapy while maintaining sustained clinical benefit under azathioprine or infliximab monotherapy. Interestingly, similar efficacy has been found for methotrexate as for azathioprine to reduce formation of antibodies to infliximab and to improve its pharmacokinetics (Gut 2007;56:1226–1231). COMMIT was a randomized trial that compared the safety and efficacy of infliximab plus methotrexate to infliximab alone for long-term control of signs and symptoms of CD in patients requiring corticosteroids for disease control at baseline. Concomitant methotrexate added no benefit (Gastroenterology 2008;134[Suppl 1]:A-682c). However, because inclusion criteria in the COMMIT trial differed from those of the SONIC trial, whether methotrexate could be used in combination with infliximab to improve anti-TNF therapy efficacy while reducing Epstein–Barr virus-associated lymphoma risk will require additional investigations. Finally, can we extrapolate the results from the SONIC clinical trial to other anti-TNF agents and is immunogenicity a relevant issue when using other anti-TNF agents? In 2003, Baert et al were the first to demonstrate that the development of antibodies against infliximab correlated with an increased risk of infusion reactions and with a shorter duration of response owing to lower infliximab concentrations in CD (N Engl J Med 2003;348:601–608). In 2006, the European Crohn's and Colitis Organisation thus recommended the use of immunosuppressants in combination with infliximab (Gut 2006;55[Suppl 1]:i1-15). Recent safety concerns have led the experts to revise their recommendations on the systematic use of immunosuppressants in CD patients administered infliximab. Combinations involving immunosuppressive agents were associated with increased risks of opportunistic infections and lymphoma. Furthermore, Van Assche et al have shown that maintaining azathioprine after 6 months of combination therapy with infliximab did not provide additional clinical benefit after a 2-year follow-up (Gastroenterology 2008;134:1861–1868). In a pooled analysis of 4 randomized trials involving 1383 IBD patients, clinical efficacy and serum infliximab concentrations were broadly similar in patients who received infliximab maintenance therapy with concomitant immunomodulators and those receiving infliximab maintenance therapy alone (Aliment Pharmacol Ther 2009;30:210–226). In 2009, the American College of Gastroenterology recommended monotherapy with biologic agents in place of a combination therapy with immunosuppressive drugs, to maximize the benefit to risk ratio of CD therapy (Am J Gastroenterol 2009;104:465–483). It is now well established that all anti-TNF agents have the potential for immunogenicity, as has clearly been demonstrated for adalimumab (Gastroenterology 2009;137:1628–1640). We therefore hypothesize that broadly similar results would be observed with combination therapy including adalimumab or certolizumab pegol, even though this remains to be investigated in depth. Overall, what is the impact of SONIC in current CD management? The logical conclusion to draw from the SONIC trial is wider use of anti-TNF agents in CD patients. However, this runs the risk that physicians could advocate overtreatment as a consequence. Indeed, in a Norwegian population-based study, a large proportion of CD patients (44%) were in clinical remission during the last 5 years of follow-up (Clin Gastroenterol Hepatol 2007;5:1430–1438), indicating that the indiscriminate use of combination therapy may result in the overtreatment of most CD patients. The results from the SONIC trial nicely demonstrated higher efficacy rates in CD patients receiving anti-TNF–based strategies compared with those on azathioprine alone, an effect that was maximal in the group with high CRP levels and mucosal lesions. Therefore, if we combine these results with poor prognostic factors, such as complicated disease (stricture, fistula), perianal disease, rectal disease, extensive small bowel disease, upper gastrointestinal disease, high CRP levels, and/or severe endoscopic lesions, it seem to be clear that these patients may benefit from early intervention with combination therapy. Nevertheless, safety should be always kept in mind, especially when treating young males, who are at increased risk of developing lymphoma. Besides shedding new light on the management of patients with CD, the SONIC trial has some limitations. In addition to steroid-free remission and mucosal healing, information on hospitalization and surgery rates would allow us to determine whether combination therapy is disease modifying. Furthermore, whether or not combination therapy is cost effective has yet to be determined. Pending the results of long-term disease modification trials and considering efficacy, safety, and the cost of anti-TNF–based strategies, whether infliximab alone, or in combination with azathioprine, is the best therapeutic approach for CD patients without any poor prognostic factors remains a matter of debate.