Can Oral Semaglutide Reduce Heavy Drinking in Alcohol Use Disorder?

Oral semaglutide significantly reduced heavy drinking days, drinking intensity, craving, and alcohol-related consequences among treatment-seeking adults with moderate to severe alcohol use disorder (AUD) in a phase 2 randomized, double-blind, placebo-controlled trial. However, the drug missed its primary end point, a lab-based measure of cue-elicited craving, and did not significantly reduce drinks per calendar day—defined as the average daily consumption across all 28 days of the measurement window, including non-drinking days—1 of its 2 key secondary end points.¹

The findings, published in the American Journal of Psychiatry, extend earlier evidence on injectable semaglutide to the oral formulation and to a treatment-seeking population with greater illness severity than prior study samples.

50 Adults With Moderate to Severe AUD Randomized 1:1

Investigators at the University of Colorado Anschutz Medical Campus randomized 50 adults with DSM-5–defined moderate or severe AUD to oral semaglutide (3 mg/day for 4 weeks, then 7 mg/day for 4 weeks) or placebo for 8 weeks. The trial used the National Institute on Alcohol Abuse and Alcoholism's Human Laboratory Program paradigm, pairing an in-clinic alcohol cue-reactivity assessment with naturalistic drinking measures collected via the Timeline Follow-Back interview. Forty-seven of 50 participants (94%) completed the 8-week treatment period, and medication adherence, verified through urinary riboflavin, video-monitored dosing, and plasma drug concentrations, was high in both arms.

Heavy Drinking Days Drop Despite Missed Primary End Point

Semaglutide did not significantly separate from placebo on the trial's primary end point, a single-item visual analog scale (VAS) measure of cue-elicited craving at week 6. It also did not significantly reduce drinks per calendar day, one of 2 preregistered key secondary end points. However, it did significantly reduce heavy drinking days per week relative to placebo.

Where semaglutide showed benefit was across a range of secondary and exploratory outcomes during the last 4 weeks of treatment:

• Drinks per drinking day (consumption counted only on days participants actually drank) decreased significantly
• Naturalistic craving, measured with the Penn Alcohol Craving Scale, decreased significantly
• Alcohol-related negative consequences, measured with a PROMIS alcohol measure, decreased at a significantly greater rate than with placebo
• Among the subset of participants who used cannabis at baseline, semaglutide significantly reduced cannabis use days over the treatment period

Semaglutide-treated participants were also significantly more likely than those on placebo to drop their World Health Organization risk drinking level by a full category or more over the treatment period.

Effect sizes for heavy drinking days and drinks per drinking day were medium to large, which the study authors noted may exceed the effects reported for currently approved AUD medications. Lead author Joseph Schacht, PhD, associate professor at the CU Anschutz School of Medicine, said in a statement that the results "suggest oral semaglutide may help reduce heavy and harmful drinking, even in people who are not trying to quit alcohol entirely," and could offer an option for patients who have not responded to existing therapies.²

Adverse Events Mostly Mild, Similar Across Groups

Adverse events, most of them mild, occurred at similar rates in both groups, though events judged "definitely" related to study medication were more common with semaglutide.¹ One participant discontinued semaglutide because of a maculopapular rash; a myocardial infarction occurred in the placebo group after the participant had completed study medication. The 2 treatment groups did not differ significantly in weight change during the 8-week period.

Small, Single-Site Sample Limits Findings

The authors note several constraints on interpretation. The trial was conducted at a single site with a modest sample (N = 50), and the authors describe the results as hypothesis-generating rather than confirmatory; both the study team and CU Anschutz indicate that larger, longer-term trials are planned.^1,2 The sample was 94% White and, by enrollment design, entirely overweight or obese, which the authors state limits generalizability to non-White, Hispanic, lower-weight, or less cue-reactive patients. The 8-week treatment period was also relatively brief, and no biological marker of alcohol consumption, such as phosphatidylethanol, was used to corroborate self-reported drinking.

Placebo-group participants were significantly more likely than those on semaglutide to correctly guess their group assignment, raising the possibility of functional unblinding that the study's sample size did not permit the authors to fully account for statistically. The authors attribute this partly to the high public profile of adverse events and weight-loss effects associated with glucagon-like peptide-1 receptor agonists.

However, the findings “could represent a new treatment option for AUD, particularly for those who have not benefited from existing medications, and may reduce alcohol-related health and social harms,” Schacht continued.² “Importantly, even reducing heavy drinking can lead to meaningful improvements for patients and families.”

References

1. Schacht JP, Sakai JT, Raymond K, Shelton R. Oral semaglutide for alcohol use disorder: a randomized clinical trial. Am J Psychiatry. Published online July 29, 2026. doi:10.1176/appi.ajp.20260003
2. Can oral semaglutide reduce alcohol use? CU Anschutz clinical trial shows fewer heavy drinking days. News release. University of Colorado Anschutz Medical Campus; July 29, 2026. Accessed July 29, 2026. https://www.eurekalert.org/news-releases/1137949