Once-weekly semaglutide versus placebo in patients with alcohol use disorder and comorbid obesity: a randomised, double-blind, placebo-controlled trial
Klausen MK, Justesen SK, Pedersen JN, Rasmussen L, Jensen A, Jensen ME
The Lancet (2026) · n=108
Across 108 treatment-seeking adults with alcohol use disorder and comorbid obesity, 26 weeks of once-weekly semaglutide reduced heavy drinking days by −41.1 percentage points from baseline versus −26.4 in placebo — a 13.7-point between-group difference (p=0.0015), the first Lancet-tier RCT confirming a GLP-1RA effect on alcohol consumption.
This is the first large-scale randomised double-blind placebo-controlled trial of a GLP-1 receptor agonist for alcohol use disorder, published in The Lancet by the Fink-Jensen group at Copenhagen University Hospital with collaborating investigators from NIH NIAAA and NIDA. The trial enrolled 108 treatment-seeking patients with DSM-5 alcohol use disorder and comorbid obesity (53 women, 55 men) and randomised them 1:1 to once-weekly subcutaneous semaglutide or matched placebo for 26 weeks at a single Copenhagen centre. The primary endpoint was the reduction in the proportion of heavy drinking days from baseline.
Semaglutide produced a reduction of −41.1 percentage points (95% CI −48.7 to −33.5), compared with −26.4 percentage points (−34.1 to −18.6) in the placebo arm — a between-group treatment difference of −13.7 percentage points (−22.0 to −5.4; p=0.0015). 88 of 108 participants (81%) completed the full 26-week intervention. Adverse events were predominantly transient mild-to-moderate gastrointestinal symptoms occurring more frequently in the semaglutide arm, consistent with the known GLP-1RA tolerability profile.
The trial extends the preclinical and Phase 1 / open-label human signal of GLP-1RA effects on craving and consumption into a rigorous Lancet-tier RCT, and is the strongest existing efficacy data point for semaglutide in AUD.
Single-centre design at one Copenhagen hospital limits generalisability and introduces site-specific confounding. The enrolled population is restricted to AUD patients with comorbid obesity, so the trial cannot speak to AUD without an obesity component — clinically, the most-cited indication-area question is whether the alcohol effect tracks the weight effect or is independent. 26 weeks is short for an addictions trial; the literature standard is 12-month sustained-reduction outcomes, and the durability of the effect after semaglutide discontinuation is not addressed.
The primary endpoint (heavy drinking days) is responsive but biases toward subjects who reduce rather than abstain — total abstinence rates are not reported in the headline abstract. The treatment difference (−13.7 percentage points) is statistically robust but clinically modest given that the placebo arm itself dropped −26.4 percentage points, consistent with strong expectancy and trial-participation effects in AUD trials.
Funding includes the Novo Nordisk Foundation, and several co-authors have Novo Nordisk industrial relationships — the standard sponsor-affiliation conflict applies here. The trial was not powered for prespecified secondary cardiovascular or psychiatric outcomes.
07·Member discussion
No member discussion yet.
Member-only conversation lives here — cycle notes, practitioner commentary, pattern-matching. Be the first paying member to start the thread.