Discontinuation of short-term infliximab maintenance therapy in patients with Crohn's disease: outcomes and risk factors associated with relapse

Background: According to the current reimbursement policy and need for co-pay in China, most patients diagnosed with Crohn's disease are treated for around six scheduled infusions of infliximab. It is necessary and significant to investigate whether infliximab therapy can be safely discontinued among these patients. Methods: An observational retrospective study was conducted between February 2006 and October 2016. We included patients with Crohn's disease who were treated for around six infusions of infliximab with or without an antimetabolite and had been assessed in clinical remission and corticosteroid-free remission at the time of withdrawal. Patients were followed up for at least six months until October 31, 2016. We evaluated the risk of relapse and median time to relapse in patients after withdrawal of infliximab therapy using Kaplan–Meier method. Factors associated with time to relapse was identified using the multiple COX proportional hazards regression analysis. To determine a cutoff value for simplified endoscopic activity score for Crohn's disease (SESCD) score that predicted the relapse with the best sensitivity and specificity, a receiver-operating characteristic (ROC) curve was performed. Results: A total of 90 eligible patients were included in the study. After a median follow-up time of 15.5 months, 48 patients had a relapse. The median time to relapse was 23.2±2.5 months. The 6-month, 1-year, 2-year, 3-year and 5-year relapse-free survival were 91%, 80%, 49%, 38% and 19% respectively. According to the multivariate analysis, factors associated with time to relapse included age ≤20 years before infliximab initiation (HR=3.56, 95% CI: 1.76–7.18, p<0.001), disease duration ≥2 years before infliximab initiation (HR=1.97, 95% CI: 1.06–3.66, p=0.03), CDAI >220 before infliximab initiation (HR=2.28, 95% CI: 1.20–4.32, p=0.01) and CRP >3 ng/ml (HR=1.92, 95% CI: 1.03–3.55, p=0.01) at infliximab withdrawal. Among patients with six infusions of infliximab, we did not identify mucosal healing or deep remission at the time of drug withdrawal as predictive factors. Through ROC analysis, we found that an SES-CD score of more than 8 at infliximab initiation was the best cutoff for predicting relapse. Conclusions: Nearly 50% of patients with Crohn's disease who were treated for a short-course infliximab maintenance therapy had a relapse within 2 years after the infliximab cessation. By using combined clinical, biological and endoscopic markers, those with a higher risk of relapse can be identified.