Objective: Infliximab is an effective treatment for refractory or fistulizing Crohn's disease (CD). However, about 30% of patients do not respond to infliximab for unknown reasons. Identifying predictive factors of response is important for optimizing clinical management and for better understanding infliximab's mechanisms of action. The aim of this study was to assess whether demographic or clinical parameters influence short-term response to infliximab.
Methods: The first 240 CD patients of the Belgian Infliximab Expanded Access Program were studied for response to infliximab treatment and assessed at 4 (refractory luminal CD) or 10 wk (fistulizing CD) after the first infusion. Detailed demographic and clinical information on age, sex, type of disease (fistulizing or refractory), Crohn's Disease Activity Index score, C-reactive protein (CRP), smoking habits, disease duration, localization of disease, concomitant medication, and previous surgery were obtained from all patients. Logistic regression and decision tree analysis were performed.
Results: There were 73.5% responders and 26.5% nonresponders to treatment. Stepwise logistic regression identified age (OR = 0.971, 95% CI = 0.947-0.995, p = 0.018), isolated ileitis (OR = 0.359, 95% CI = 0.177-0.728, p = 0.004), and previous surgery (OR = 0.429, 95% CI = 0.233-0.787, p = 0.006) as inversely correlated with response, whereas isolated colitis (OR = 1.905, 95% CI = 1.010-3.597, p = 0.046) and concomitant immunosuppressive treatment (OR = 2.670, 95% CI = 1.430-5.016, p = 0.0022) were positively correlated with response to infliximab. Surprisingly, smoking habits were not retained as predictors for response. Decision tree analysis provided a working algorithm based on age and immunosuppressive treatment that warrants further exploration.
Conclusions: In this large cohort of infliximab-treated CD patients, young age, Crohn's colitis, and concomitant immunosuppressive treatment were identified as independent variables favoring short-term response to infliximab.