Semaglutide and alcohol craving, two 2026 trials

Two trials, two different questions

The oral trial was published 29 July 2026. It randomized 50 treatment-seeking adults with moderate or severe alcohol use disorder. Everyone had a body mass index of at least 25. Oral semaglutide was 3 mg a day for 4 weeks, then 7 mg a day for 4 weeks, or placebo, for 8 weeks. The drug was bought commercially and overencapsulated. The protocol ran under an FDA investigational new drug exemption. The preregistered main test was a single laboratory craving score at week 6, after people held and smelled a drink they were not allowed to consume.

The weekly trial enrolled 108 treatment-seeking adults with alcohol use disorder and a body mass index of at least 30. ClinicalTrials.gov lists it as completed phase 2, with Wegovy once a week titrated to a maximum of 2.4 mg. The Lancet paper says enrollment ran from 10 June 2023 to 4 February 2025, that the pens came from a hospital pharmacy, and that everyone was offered up to 10 sessions of cognitive behavioral therapy. The main test was the change in heavy drinking days at 26 weeks.

The main laboratory craving test missed

At week 6, the single craving question, rated 0 to 20 after the drink cue, did not differ significantly. The paper gives b of -0.614, a 95% interval from -3.563 to 2.335, and p of 0.68. That analysis had 26 people on placebo and 21 on semaglutide. A longer laboratory craving questionnaire at the same visit did differ, with p of 0.027. The other single-item laboratory questions did not.

Day-to-day craving, measured with the Penn Alcohol Craving Scale outside the laboratory, was lower on semaglutide, with p of 0.030. The paper's own discussion says the single laboratory item missed and that other craving measures, drinking, and alcohol-related consequences still moved. Those are not the same result as the test the trial named first.

A missed primary test is the result for that test. It does not erase the secondary measures, and the secondary measures do not convert the primary test into a hit.

Heavy drinking days, and one result that changed after a fill-in

In the last 4 weeks, heavy drinking days were lower on oral semaglutide than on placebo, with p of 0.008. Drinks per day were not significant, with p of 0.057. Drinks per drinking day were lower, with p of 0.041. The number of drinking days was not, with p of 0.58. The groups did not differ significantly in weight loss. Baseline body mass index did not significantly change those drinking results.

Fifty people were randomized and 47 finished 8 weeks. All 3 who left were in the semaglutide group. One left after a rash the investigators judged probably related to the study drug. On the drinking-risk scale, 17 of 21 people still in the semaglutide group at week 8 had dropped at least one World Health Organization risk level, versus 14 of 26 on placebo, with p of 0.045. A drop of two or more levels was not significant. When the three who left were filled in as no change, neither comparison stayed significant. The paper reports that second analysis in the same paragraph.

Weekly semaglutide 2.4 mg, with therapy in both groups

In the weekly trial, heavy drinking days fell by 41.1 percentage points on semaglutide and by 26.4 percentage points on placebo. The estimated difference was 13.7 percentage points, with a 95% interval from 22.0 to 5.4 in the paper's negative direction and p of 0.0015. All 108 people who received a first injection were in the analysis. Eighty-eight people, 81%, finished 26 weeks. The registry's primary outcome matches the paper: change in the percentage of heavy drinking days.

Secondary results in the paper include 1.5 fewer drinks per drinking day than placebo, p of 0.0051, and a Penn Alcohol Craving Scale difference of 3.1 points, p of 0.0020. The scale runs from 0 to 30. Three people stopped at 0.5 mg, two at 1.0 mg, and five at 1.7 mg. The rest of the semaglutide group reached 2.4 mg. Heavy drinking days and weight loss were correlated in the semaglutide group and not in the placebo group. The authors say only three people on semaglutide lost less than 2.5 kg, so the paper cannot tell whether the drinking result is separate from weight loss.

Both groups received talk therapy. The 13.7 point figure is the gap between groups, not the full drop from each person's own baseline.

What the papers say about side effects

In the oral trial, adverse events were common in both groups and mostly mild. The groups did not differ significantly in the share of people with at least one event, or in any single event. Events judged definitely related to the drug were more common on semaglutide. One person on semaglutide stopped for a maculopapular rash. One serious event, a myocardial infarction, was in the placebo group after that person had finished the study drug. More people on placebo than on semaglutide correctly guessed their assignment. The discussion treats functional unblinding as a limit of this drug class.

In the weekly trial, nausea was reported by 31 of 54 people on semaglutide and 4 of 54 on placebo. Loss of appetite was 19 versus 8. The paper calls the gastrointestinal effects mostly mild to moderate and transient. Five people stopped because of side effects, four of them on semaglutide. One person on semaglutide was admitted for abdominal pain at the 0.25 mg dose and stayed in the trial. The authors say safety data in alcohol use disorder are still scarce, and that a setting without weekly visits could look different.

Wegovy's listed uses do not include alcohol use disorder

NovoCare's Wegovy page, opened 27 October 2026, says the injection and the tablets are used with a reduced-calorie diet and increased physical activity to reduce the risk of major cardiovascular events in adults with known heart disease and obesity or overweight, and to help adults with obesity, or some adults with overweight and a weight-related medical problem, lose weight and keep it off. Alcohol use disorder is not in that list. The same page says Wegovy should not be used with other semaglutide products or other GLP-1 receptor agonists.

The oral trial's doses were 3 mg and then 7 mg a day. NovoCare's tablet prices on that page are $149 a month at 1.5 mg, $199 at 4 mg, and $299 at 9 mg and 25 mg. One month is a bottle of 30 tablets. Those strengths are not the trial doses. The weekly trial used the Wegovy pen, toward 2.4 mg, supplied by a hospital pharmacy, not a telehealth cash offer. The American Journal of Psychiatry paper says three medications are FDA-approved for alcohol use disorder and does not name them in the text opened here. It also says semaglutide is approved for weight loss and for glucose control in type 2 diabetes, which is a different sentence from an alcohol-use approval.

An investigational-new-drug trial is a research use. It is not the use printed on the Wegovy page.

The commission pages sell weight management

Sesame's weight-loss page, MEDGm's GLP-1 page, and Shed's Wegovy page were opened the same day. The fetched text of each one does not say alcohol use disorder. Sesame describes the Wegovy pill as an oral GLP-1 medication for weight loss. MEDGm's craving lines are about food and food preoccupation. Shed describes Wegovy as approved for chronic weight management. The prices on those pages are in the pill-cost note. They are not a fee for the trials above.

FDA's unapproved-GLP-1 page, opened the same day, says compounded drugs are not FDA approved and are not reviewed for safety, effectiveness, or quality before marketing. The oral trial used commercially purchased semaglutide. The weekly trial used Wegovy pens. A compounded semaglutide price in this catalog is neither of those products. The price module under the first section is the catalog of commission-earning Wegovy offers, with each row's own check date. It is a weight-management price list.