Wharton · Diabetes, obesity & metabolism 2022 · pooled analysis of randomized controlled trials · n=4182

Gastrointestinal tolerability of once-weekly semaglutide 2.4 mg in adults with overweight or obesity, and the relationship between gastrointestinal adverse events and weight loss.

Cited 179 times in the scientific literature.

Level 2 - randomized trial

Pooled post-hoc and mediation analysis of randomized controlled trials

PubMed 34514682 · doi:10.1111/dom.14551 · record verified 2026-08-27

What was done

This was a pooled post-hoc analysis of randomized controlled trial data from the STEP 1-3 trials evaluating adults with overweight or obesity randomized to 68 weeks of once-weekly semaglutide 2.4 mg (n = 2117) or placebo (n = 1262). Gastrointestinal (GI) adverse event (AE) incidence, severity, timing, and discontinuation rates were assessed. Mediation analysis estimated the proportion of weight loss attributable to versus independent of GI AEs. GI tolerability during maintenance was also evaluated using STEP 4 trial data in 803 participants who completed a 20-week semaglutide run-in.

What was found

GI AEs were more frequent with semaglutide than placebo: nausea (43.9% vs. 16.1%), diarrhoea (29.7% vs. 15.9%), vomiting (24.5% vs. 6.3%), and constipation (24.2% vs. 11.1%). For semaglutide, 98.1% of GI AEs were mild-to-moderate, 99.5% were non-serious, and they clustered during or shortly after dose escalation. Permanent treatment discontinuation due to GI AEs occurred in 4.3% of semaglutide-treated participants. Mean weight loss with semaglutide 2.4 mg was comparable in those without GI AEs (9.6%-17.1%) and with GI AEs (11.4%-17.7%). Mediation analysis indicated GI AEs accounted for <1 percentage point of the additional 7.6%-14.4% weight loss achieved with semaglutide compared to placebo.

Why it matters

These findings indicate that weight loss from semaglutide 2.4 mg is driven primarily by mechanism of action rather than as a secondary consequence of GI intolerance or nausea.

Limits

This is a post-hoc pooled analysis across clinical trials. STEP 4 maintenance data only included participants who had already tolerated a 20-week semaglutide run-in, creating survivor bias for maintenance tolerability estimates.

Cited by