Relapse rates and predictors for relapse in a real-life cohort of IBD patients after discontinuation of anti-TNF therapy

Objective: We investigated relapse rates after anti-tumor necrosis factor (anti-TNF) withdrawal in inflammatory bowel disease (IBD) patients, response to restart of anti-TNF treatment and predictors for relapse. Methods: IBD patients in remission receiving infliximab or adalimumab treatment for >= 1 year who discontinued treatment were included. Relapse rates and predictors for relapse were studied using survival and Cox regression analysis. Results: In total, 101 patients were included (77 CD, 24 UC). A total of 56 patients (55%) experienced a relapse (CD 38, UC 18) with a median time to relapse of 32 and 18 months in CD and UC, respectively. Of patients that were retreated with the same anti-TNF agent, 84% responded. A trough serum concentration >= 2 mu g/ml within 1 year prior to anti-TNF discontinuation was associated with a higher relapse rate in CD patients (HR 2.89; p = .018), which was more evident in patients requiring retreatment with biologicals, bowel-related surgery or experimental medication (HR: 4.18; p = .009). A young age (<17 years) at diagnosis was associated with a higher relapse rate (HR: 2.29; p = .040) and fecal calprotectin levels <25 mu g/g with a lower relapse rate in CD patients (HR: 0.34; p = .041). Relapse rates, requiring treatment with biologicals or experimental medication, was lower in UC patients who continued immunosuppressive treatment (HR: 0.26; p = .042). Conclusions: Approximately 55% of patients relapsed after anti-TNF withdrawal with a median time to relapse of 32 and 18 months in CD and UC, respectively. Retreatment with the same anti-TNF was successful in 84% of patients.