Once-weekly semaglutide was associated with greater reductions in heavy drinking days compared with placebo over 26 weeks among treatment-seeking patients with alcohol use disorder and obesity, according to a randomized clinical trial published in The Lancet.
The single-center, double-blind trial in Denmark included 108 patients aged 18 to 70 years with moderate to severe alcohol use disorder and body mass index of 30 kg/m² or higher. Notably, 85% of patients met criteria for severe alcohol use disorder, with a mean of 17 heavy drinking days per month and mean alcohol consumption of about 2,200 g over 30 days.
Patients were randomly assigned to receive semaglutide 2.4 mg once weekly or placebo. Both groups were offered up to 10 sessions of cognitive behavioral therapy and attended weekly study visits.
Alcohol-Related Outcomes
The primary endpoint was change in heavy drinking days from baseline to week 26, assessed with Timeline Followback. Heavy drinking was defined as at least 60 g of alcohol per day for men and at least 48 g per day for women.
Heavy drinking days decreased by 41 percentage points from baseline with semaglutide and 26 percentage points with placebo, for an estimated between-group difference of 14 percentage points.
Total alcohol consumption declined by 1,550 g per 30 days with semaglutide compared with 1,026 g with placebo. Drinks per drinking day decreased by 3.5 units compared with 2.1 units.
Alcohol craving, Alcohol Use Disorders Identification Test scores, Alcohol Use Disorders Identification Test consumption scores, and World Health Organization risk drinking levels also improved more with semaglutide. The number needed to treat for a two-level decrease in World Health Organization risk drinking level was 4.3. The authors noted that this compares favorably with estimates reported for approved medications, which are typically 7 or higher for the same endpoint.
The study also reported moderate effect sizes (Cohen’s d) across key alcohol outcomes, supporting the clinical relevance of the observed reductions.
“These data, when added to the growing evidence, demonstrate the potential of [glucagon-like peptide-1 receptor agonists] as a novel treatment for alcohol use disorder,” wrote lead study author Mette Kruse Klausen, MD, of Mental Health Centre Copenhagen, Copenhagen University Hospital–Bispebjerg and Frederiksberg, and colleagues.
Biomarkers and Metabolic Effects
Biomarker findings were consistent with lower alcohol exposure. Phosphatidyl ethanol decreased with semaglutide and did not change with placebo, and gamma-glutamyl transferase declined to a greater extent with semaglutide.
Patients receiving semaglutide also had greater reductions in bodyweight, waist circumference, body mass index, and glycated hemoglobin. Bodyweight declined by 11.2 kg with semaglutide compared with 2.2 kg with placebo.
Among patients receiving semaglutide, greater weight loss was associated with larger reductions in heavy drinking days, whereas no such association was observed in the placebo group. The investigators noted that caloric intake was not assessed, limiting interpretation of whether reduced alcohol consumption was independent of weight loss.
Subgroups and Placebo Response
Prespecified subgroup analyses showed reductions in heavy drinking days among patients with severe alcohol use disorder and among those with 12 to 17 heavy drinking days at baseline. However, the moderate alcohol use disorder subgroup included only 16 patients, and subgroup analyses were not powered to detect differences between groups.
The placebo group also showed a 26 percentage point reduction in heavy drinking days. Investigators noted that weekly study visits and cognitive behavioral therapy may have contributed to this response, which could influence how the findings translate to routine clinical settings.
Safety
Adverse events were mainly gastrointestinal and occurred more often with semaglutide. Nausea was reported in 57% of patients receiving semaglutide and 7% receiving placebo. Five patients discontinued the trial because of side effects, including four in the semaglutide group.
One patient receiving semaglutide was hospitalized for abdominal pain and continued in the trial. The study also reported small, asymptomatic increases in amylase levels in the semaglutide group, without cases of pancreatitis.
Adverse events were assessed through patient self-report and routine clinical inquiry rather than structured instruments, which may affect estimates of frequency and severity.
Interpretation and Limitations
The researchers noted several limitations, including enrollment restricted to patients with obesity, a predominantly White study population, and the absence of alcohol-use follow-up following week 26. The trial was conducted at a single center, which may further limit generalizability.
In addition, all participants received cognitive behavioral therapy and frequent clinical contact, which may not reflect usual care. The durability of treatment effects beyond the study period remains unknown.
In an accompanying Comment, Christian S. Hendershot, PhD, of the Keck School of Medicine of the University of Southern California, and Klara R. Klein, MD, PhD, of the University of North Carolina School of Medicine, wrote that the findings support cautious optimism but highlighted ongoing uncertainties, including generalizability, variability in response, and long-term effects. They also noted that glucagon-like peptide-1 receptor agonists are already being prescribed off-label for alcohol use disorder despite limited evidence.
Disclosures
The trial was funded by the Research Foundation, Mental Health Services, Capital Region of Denmark; Novo Nordisk Foundation; Novavi Foundation; Hartmann Foundation; and Augustinus Foundation. The manufacturer had no role in study design, data collection, data analysis, or manuscript preparation.
Several researchers reported consulting, advisory, or research relationships with pharmaceutical and biotechnology companies, including Novo Nordisk. Full disclosures are available in the study.
Source: The Lancet
