Lancet trial: semaglutide cuts heavy drinking days in alcohol use disorder

A randomized, double-blind, placebo-controlled trial published April 30, 2026, in The Lancet found that adding once-weekly semaglutide to cognitive behavioral therapy reduced heavy drinking days in adults who had both alcohol use disorder and obesity ^[1][2]. The National Institutes of Health, whose scientists co-authored the study, called it the first randomized controlled trial evidence that a GLP-1 receptor agonist can lower heavy drinking in this group ^[2].

What the trial tested

The study was run at a single center, the Mental Health Center Copenhagen in Denmark, and enrolled 108 treatment-seeking adults aged 18 to 70 with a body mass index of 30 kg/m² or higher, at least six heavy drinking days, and an Alcohol Use Disorders Identification Test score above 15 ^[3]. Participants were assigned 1:1 to placebo or subcutaneous semaglutide 2.4 mg once weekly for 26 weeks, on top of standard cognitive behavioral therapy ^[2][3]. All participants received at least one dose and 88 completed the trial ^[3]. About 51% were men, and the average age was 52.3 years ^[3].

The primary endpoint was the change in heavy drinking between baseline and week 26 ^[3]. In the semaglutide group, the share of days that were heavy drinking days fell by 41.1 percentage points (95% CI, −48.7 to −33.5), compared with 26.4 percentage points (95% CI, −34.1 to −18.6) with placebo ^[3]. In practical terms, heavy drinking days dropped from roughly 17 per 30 days at baseline to about 5 with semaglutide, versus about 9 with placebo ^[1].

Other measures moved in the same direction. Total alcohol consumption fell by 1,550.2 grams per 30 days with semaglutide versus 1,025.9 grams with placebo, and mean drinks per day fell by 3.5 units versus 2.1 units ^[3]. Body weight dropped 11.2 kg with semaglutide versus 2.2 kg with placebo ^[3]. NIH said blood-alcohol biomarkers backed up the self-reported drinking data, and that reductions in blood pressure and other clinical measures were larger in the semaglutide group ^[2].

The two sources describe the size of the drinking benefit slightly differently. NIH characterized the semaglutide result as a 13.7% greater reduction than placebo ^[2], while the figures reported by AJMC imply a 14.7 percentage-point gap between the two arms ^[3]. NIH also reported a number needed to treat of 4.3 in this trial, compared with 7 or higher for currently approved alcohol use disorder medications ^[2].

Side effects and limits

Gastrointestinal problems were the most common adverse events and occurred more often with semaglutide, including nausea, loss of appetite, vomiting, abdominal pain, reflux and fatigue ^[3]. Five participants discontinued because of adverse effects, and there was one serious adverse event, abdominal pain, in the semaglutide group ^[3]. NIH described the side effects as transient and mild ^[2].

The authors listed clear limitations. No follow-up data were collected after treatment ended, so it is not known whether the drinking reductions last after semaglutide is stopped ^[3]. The study population was primarily White, which may limit how well the results apply elsewhere ^[3]. The trial was also small and conducted at one site ^[3].

Why it matters for patients

Only three medications are FDA-approved to treat alcohol use disorder, and NIAAA Director George Koob said existing options are "vastly underutilized" ^[2][3]. This trial does not change that. Semaglutide is approved in the US as Ozempic and Rybelsus for type 2 diabetes and as Wegovy for weight management; it is not approved for alcohol use disorder, and the study authors wrote that their finding supports "an expanded indication" — a regulatory step that has not happened ^[3].

It is also worth noting who was studied: adults with both alcohol use disorder and obesity, all receiving cognitive behavioral therapy. Everyone in the trial, including the placebo group, cut their heavy drinking substantially, which suggests the therapy itself did a lot of work ^[3]. An earlier randomized trial found no overall effect of a GLP-1 on heavy drinking across all participants, though a subgroup with obesity responded strongly ^[2].

What happens next

The researchers say they want to study GLP-1 effects over a longer period and in a larger, more diverse population to confirm the results ^[2]. NIDA Director Nora Volkow said the findings are "very encouraging" but that "questions remain" ^[2]. No timeline for follow-up trials, and no regulatory filing for an alcohol use disorder indication, has been announced in these sources.