Efficacy of biologic drugs in short-duration versus long-duration inflammatory bowel disease: A systematic review and an individual-patient data meta-analysis of randomized controlled trials.

BACKGROUND:

Starting biologic treatment early in the course of inflammatory bowel disease (IBD) may associate with higher efficacy, especially in Crohn's disease (CD).

METHODS:

A systematic review and individual-patient-data meta-analysis of all placebo-controlled trials of biologics approved for IBD at study inception (Oct 2015), using Vivli data-sharing platform. The primary outcome was the proportional biologic/placebo treatment effect on induction-of-remission in patients with short-duration (≤18months) versus long-duration disease (>18months) analyzed separately for CD and ulcerative colitis (UC). We used meta-regression to examine the impact of patients' characteristics on the primary outcome. Study PROSPERO registration: CRD42018041961.

RESULTS:

We included 25 trials, testing infliximab, adalimumab, certolizumab, golimumab, natalizumab or vedolizumab (6,168 CD, 3,227 UC patients). In CD, induction-of-remission rates were higher in pooled placebo and active arms' patients with short-disease duration≤18 months (41.4%, 244/589) compared with disease-duration>18months (29.8%, 852/2857, meta-analytically estimated OR=1.33, 95%CI:1.09-1.64). The primary outcome, proportional biologic/placebo treatment effect on induction-of-remission, was not different in short-duration disease ≤18 months (n= 589, OR 1.47, 95%CI:1.01-2.15) compared with longer disease-duration (n=2857, OR 1.43, 95%CI:1.19-1.72). In UC trials, both the proportional biologic/placebo remission-induction effect and the pooled biologic-placebo effect were stable regardless of disease duration. Primary outcome results remained unchanged when tested using alternative temporal cut-offs and when modelled for individual-patients' co-variates, including prior anti-TNFs exposure.

CONCLUSION:

There are higher rates of induction-of-remission with biologics and with placebo in early CD, resulting in a treatment-to-placebo effect ratio which is similar across disease durations. No such relationships between disease-duration and outcomes is found in UC.