Outcome of elective withdrawal of anti-tumour necrosis factor-α therapy in patients with Crohn's disease in established remission

Background and Aims: Outcomes of cessation of anti-TNF therapy for Crohn's disease (CD) in clinical and/or endoscopic remission in routine clinical practice is uncertain. This study aimed to evaluate clinical outcomes and factors associated with relapse in CD patients following formal disease assessment and elective anti-TNF withdrawal. Methods: Prospective observational study of CD patients in whom anti-TNF therapy was stopped electively after >= 12months and follow-up of >= 6months. Investigations at assessment prior to cessation included >= 1 of clinical assessment, endoscopic and/or imaging. Relapse was defined as recurrent symptoms of CD requiring medical or surgical therapy. Results: Eighty-six patients received anti-TNF for a median duration of 23 (12-80) months for severe active luminal (70%), fistulating perianal (25.5%) and other fistulating disease (4.5%). Relapse rates at 90,180 and 365days were 4.7%, 18.6% and 36%, respectively. If anti-TNF dose escalation occurred 6months prior to withdrawal, 88% (7/8) relapsed. Based on multivariate analysis, risk factors for relapse include ileocolonic disease at diagnosis and previous anti-TNF therapy. An elevated faecal calprotectin (FC) is likely to predict relapse (p=0.02), with a PPV of 66.7% at >50 mu g/g. Of 36 patients who relapsed, 31 were retreated with anti-TNF, with an overall recapture rate of 93%. Conclusion: Relapse rates at 1year following elective withdrawal of anti-TNF are 36%, with high retreatment response rate. Predictors of relapse include ileocolonic involvement, previous anti-TNF therapy and raised FC. Endoscopic/radiologic assessment prior to cessation of therapy does not appear to predict those at lower risk of relapse.