Predictors of relapse following infliximab de‐escalation in patients with inflammatory bowel disease: the value of a strategy based on therapeutic drug monitoring

BACKGROUND: There are limited data concerning infliximab drug monitoring during de-escalation of the treatment of inflammatory bowel disease (IBD). AIM: To define the rate and the predictors of relapse following infliximab de-escalation in IBD patients in remission. METHODS: All IBD patients at a single referral centre in clinical and biological remission and in whom the dose of infliximab had been de-escalated were included. Patients in remission with a high trough level of infliximab (>7 mg/L) were considered to be trough level-based de-escalation patients. The data were retrieved from a prospective IBD database. Actuarial analysis was performed for statistical purposes. RESULTS: A total of 146 de-escalations were performed in 96 patients (Crohn's disease/ulcerative colitis: 68%/32%); 54 (37%) were based on clinical remission only, and 92 (63%) were based on clinical remission associated with a trough level above 7 mg/L. The cumulative probabilities of relapse following infliximab de-escalation were 16% and 47% at 1 and 2 years, respectively. Ulcerative colitis was associated with an increased risk of relapse (HR = 3.2, P = 0.005). Conversely, combination therapy at infliximab initiation (HR = 0.39, P = 0.0110) and trough level-based de-escalation were associated with decreased risk of relapse (HR = 0.45, P = 0.024). Trough levels before and after de-escalation were well correlated; a decrease by half was observed following a 2-week interval increase or a half-dose decrease. CONCLUSION: The use of trough levels to assess the feasibility of dose de-escalation seems to be a prerequisite for decreasing the risk of relapse.