On May 2, 2026, The Lancet published what is now the most rigorous clinical evidence to date on a question the recovery field has been watching for three years: can GLP-1 receptor agonists — the class of drugs that includes semaglutide, marketed as Ozempic and Wegovy — meaningfully reduce alcohol consumption in people with alcohol use disorder?
The short answer from this trial: yes, in this specific population, adjunctive to therapy, over 26 weeks.
The complications that follow that answer are where the science gets interesting.
The study: what they did and who was in it
The trial, led by Dr. Anders Fink-Jensen at Copenhagen University Hospital and conducted with international collaborators including NIH researchers, enrolled 108 adults who were both treatment-seeking for alcohol use disorder and had comorbid obesity. Eighty-eight completed the full 26-week protocol. Participants were randomized to receive either once-weekly subcutaneous semaglutide injections or placebo — both groups also received standard cognitive behavioral therapy sessions for AUD.
The design is significant. This wasn’t a post-hoc analysis of Ozempic prescriptions in electronic health records, the kind of study that has produced most of the prior GLP-1/AUD signal. It was a prospective, double-blind, placebo-controlled trial — the methodology that distinguishes “this drug might work” from “this drug works, under these conditions.”
The findings on the three primary measures were statistically significant: semaglutide reduced the frequency of heavy drinking days, reduced the number of drinks per drinking day, and reduced self-reported craving scores relative to placebo. All three moved in the right direction, all three with meaningful effect sizes, all three over the full 26-week duration.
What the biology suggests
GLP-1 receptors aren’t only in the gut. They’re expressed in the brain’s reward circuitry — specifically in the nucleus accumbens and ventral tegmental area, the same structures implicated in the reinforcement loop that drives compulsive alcohol use. The working hypothesis is that semaglutide dampens the dopamine response to alcohol consumption, reducing the rewarding signal that makes “just one” feel like a bad idea. This is mechanistically different from naltrexone, which blocks opioid receptors involved in alcohol’s euphoric effect, and from acamprosate, which stabilizes glutamate signaling disrupted during withdrawal.
If the effect is real and neurobiological — not just secondary to weight loss reducing alcohol’s bioavailability, which is a competing hypothesis the trial can’t fully rule out — it opens a potential third mechanism for AUD pharmacotherapy. The existing medications work for some people and not others; a drug that works via a distinct pathway could help patients who don’t respond to naltrexone or acamprosate. In a condition where fewer than 5% of people with AUD ever receive any medication-assisted treatment, a new option isn’t redundant.
The limitations that matter
The trial has three constraints that should inform how its results get applied clinically.
First, the population: every participant had both AUD and comorbid obesity, defined by BMI criteria. Semaglutide’s weight-loss mechanism means that people at healthy weight carry a higher risk of side effects, particularly muscle mass loss, and the drug isn’t indicated for them. The Lancet trial cannot be generalized to people with AUD who don’t have obesity. That’s a meaningful restriction — alcohol use disorder disproportionately affects people across the BMI spectrum, and the treatment gap for non-obese patients with AUD remains completely unaddressed by this finding.
Second, the duration: 26 weeks is promising but not sufficient to establish long-term efficacy or safety for a chronic condition like AUD. Alcohol use disorder is a relapsing illness with a trajectory measured in years, not months. The trial tells us semaglutide works over six months in this population; it says nothing about year two or year five, or about what happens when someone stops taking it.
Third, the concurrent therapy: all participants received CBT. The study cannot disentangle semaglutide’s effect from the CBT’s effect, or tell us whether semaglutide would work without structured psychosocial support. That matters enormously for how this drug gets used in practice, where most people with AUD don’t have access to a full course of cognitive behavioral therapy.
That matters enormously for how this drug gets used in practice, where most people with AUD don’t have access to a full course of cognitive behavioral therapy.
Dr. Fink-Jensen’s team noted these limitations explicitly in the published paper, which is how well-designed trials are supposed to work. The trial answers the question it was designed to answer, no more.
Where this sits in the AUD treatment landscape
Alcohol use disorder is the most undertreated serious medical condition in America by the ratio of people affected to people receiving care. Approximately 14.5 million Americans have AUD; fewer than 8% receive any treatment in a given year, and only about 2% receive an FDA-approved medication. Naltrexone, approved in 1994, is prescribed for AUD by fewer than 3% of primary care physicians who see patients with AUD. Acamprosate and disulfiram are prescribed even less frequently.
The May 2026 Lancet trial does not solve that problem. A new drug in the pipeline doesn’t address the fact that the existing drugs aren’t being used. But the GLP-1 story has done something the existing medication story hasn’t: it has generated mainstream cultural awareness that medication-based treatment for alcohol use disorder exists. The number of people who know what naltrexone is remains depressingly small. The number of people who know what Ozempic is — and who might ask their doctor about it for a drinking problem — is substantially larger.
That’s not a scientific argument. It’s a health communication argument. But in a disease where stigma and lack of awareness are primary barriers to treatment entry, cultural resonance matters.
The Lancet trial gives clinicians permission to consider semaglutide for patients with AUD and obesity who haven’t responded to or tolerated existing options. It does not give the field permission to declare GLP-1s the solution to alcohol use disorder. The solution to alcohol use disorder is a healthcare system that treats it like the chronic medical condition it is — which, for the majority of the 14.5 million people who have it, still doesn’t exist.
Related: The naltrexone prescribing gap — why fewer than 5% of people with alcohol use disorder ever receive medication · Explore alcohol use disorder resources
Sources Cited
- 01.A
- 02.A
- 03.A
- 04.C
Filed Under
sciencebiologytreatmentClinical Trial