Harms with placebo in trials of biological therapies and small molecules as maintenance therapy in inflammatory bowel disease: a systematic review and meta-analysis.

Background: Randomised placebo-controlled trials for the induction of inflammatory bowel disease (IBD) remission involve potential harms to those receiving placebo. Whether these harms are also apparent with placebo during maintenance of remission trials in IBD is unclear. We aimed to examine the potential harms associated with receiving placebo in trials of licensed biologics and small molecules for maintenance of remission of ulcerative colitis and luminal Crohn's disease in a meta-analysis.

Methods: We performed a systematic review and meta-analysis. We searched several medical literature databases including MEDLINE (from Jan 1, 1946, to May 31, 2024), Embase and Embase Classic (Jan 1, 1947, to May 31, 2024), and the Cochrane Central Register of Controlled Trials from database inception to May 31, 2024, for randomised placebo-controlled trials of licensed biologics and small molecules for maintenance of remission in adults with IBD reporting data on adverse events over a period of 20 weeks or more. There were no language restrictions or prespecified exclusion criteria. We extracted summary data and pooled data using a random-effects model for any treatment-emergent adverse event, drug-related adverse event, infection, worsening of IBD activity, withdrawal due to adverse events, serious adverse events, serious infection, serious worsening of IBD activity, or venous thromboembolic events, reporting relative risks (RRs) for placebo versus active drug with 95% CIs for each outcomes. The protocol for this meta-analysis was registered with PROSPERO (CRD42024542624).

Findings: Our search identified 10 826 citations, of which 45 trials including 16 562 patients (10 319 [62·3%] receiving active drug and 6243 [37·7%] placebo) were eligible. The risks of any treatment-emergent adverse event (7297/9546 [76·4%] patients on active drug vs 4415/5850 [75·5%] on placebo; RR 1·01, 95% CI 0·99-1·04; I² =47%), serious infection (260/10 242 [2·5%] vs 155/6149 [2·5%]; 0·97, 0·79-1·19; I² =0%), or venous thromboembolic event (12/4729 [0·3%] vs 9/2691 [0·3%]; 0·72, 0·31-1·66; I² =0%) were not significantly lower with active drug than placebo. The risks of any infection (3208/8038 [39·9%] vs 1713/4809 [35·6%]; 1·14, 1·05-1·23; I² =60%) or any drug-related adverse event (1094/2997 [36·5%] vs 609/1950 [31·2%]; 1·24, 1·02-1·50; I² =75%) were higher with active drug than placebo. However, the risks of any worsening of IBD activity (1038/8090 [12·8%] vs 1181/5191 [22·8%]; 0·58, 0·52-0·64; I² =40%), any withdrawal due to adverse events (610/10 282 [5·9%] vs 561/6207 [9·0%]; 0·71, 0·60-0·84; I² =43%), any serious adverse events (1066/10 292 [10·4%] vs 742/6198 [12·0%]; 0·85, 0·77-0·94; I² =17%), or any serious worsening of IBD activity (101/5707 [1·8%] vs 143/3640 [3·9%]; 0·55, 0·42-0·71; I² =0%) were lower with active drug than placebo. 21 randomised controlled trials were judged as low risk of bias across all domains.

Interpretation: In maintenance of remission trials in IBD, placebo was associated with some clinically significant potential harms. Patients should be counselled about these before participating in clinical trials and consideration given to alternative designs to test novel drugs in IBD.

Funding: None.