For decades, the search for a pill that quiets addictive cravings has come up mostly empty. Now, a weight-loss injection designed to mimic a gut hormone is doing something researchers did not anticipate: in a rigorous trial of 108 people with alcohol use disorder, it cut heavy drinking days by 41% from baseline, more than a third better than placebo, on top of standard therapy [1]. The drug is semaglutide, better known as Ozempic or Wegovy, and the result, published in The Lancet in May 2026, is forcing a rethink of how addiction works in the brain.
The Trial That Started It All
The Copenhagen trial, run by Mette Kruse Klausen, Anders Fink-Jensen and colleagues at the University of Copenhagen, ran from June 2023 to February 2025 and enrolled 108 treatment-seeking adults (53 women and 55 men) with moderate-to-severe alcohol use disorder and a body mass index of 30 or higher [1]. Half received a weekly 2.4-milligram semaglutide injection, the weight-loss dose, while the other half received saline. Both groups also received 10 sessions of cognitive behavioural therapy, the kind of gold-standard counselling that has long been a backbone of addiction treatment.
By the end of 26 weeks, the picture was striking. Participants on semaglutide had reduced the share of their heavy drinking days (defined as four or more standard drinks for women and five or more for men) by 41.1 percentage points from their own baseline, while the placebo group dropped 26.4 percentage points. That 13.7-percentage-point gap between arms was statistically robust (95% CI −22.0 to −5.4; p=0.0015) [1]. Eighty-one percent of participants, 88 of 108, stuck with the full intervention. The side effects, mostly nausea and other transient gastrointestinal complaints, were the same mild-to-moderate issues already familiar to anyone who has taken a GLP-1 drug for diabetes or weight loss.
So what does a 13.7-point edge actually mean in a clinic? Even in a strong placebo-plus-therapy arm, the kind of response AUD trials routinely see, a 13.7-percentage-point jump on top of that is the kind of separation that rarely shows up in phase 2 addiction work. The co-authors include George Koob, the architect of modern addiction neuroscience, and Nora Volkow, the former head of the National Institute on Drug Abuse, lending the result unusual credibility in a field littered with failed phase 2 trials [1]. When a study of this design lands that big a gap, addiction researchers pay attention.
Rewiring the Reward Circuit
To understand why a diabetes drug might curb problem drinking, follow the receptors. GLP-1 receptors are scattered through the brain's reward circuitry, including the ventral tegmental area, the nucleus accumbens, the laterodorsal tegmental area, the hypothalamus, the amygdala, the striatum and the hippocampus [2]. The brain itself makes GLP-1 in a small brainstem region called the nucleus tractus solitarius, then ships it forward as a neurotransmitter, modulating dopamine release in the mesolimbic pathway that turns alcohol (and food, and drugs) into a felt reward [3].
In animal studies, GLP-1 receptor agonists blunt the dopamine spike that alcohol, cocaine, nicotine and amphetamines normally produce in the nucleus accumbens, without flattening normal reward function [3]. Rodents given GLP-1 drugs drink less alcohol, seek it less vigorously and relapse less. Vervet monkeys bred to drink heavily reduce their intake on liraglutide and exenatide without vomiting [3]. The drugs shrink the reward gap between using and abstaining, which is what makes sobriety feel less like deprivation.
Anna Lembke, the Stanford psychiatrist who wrote Dopamine Nation, frames the mechanism in plain language: "Consuming drugs or alcohol, and participating in any activity that is pleasurable or rewarding will temporarily increase dopamine firing above baseline in the reward pathway, making this area of the brain a good target for medications to treat addiction" [4]. Patients on GLP-1 drugs, she says, often describe the experience as the medication "turning down the volume of food noise" in the brain. So what does that mean for someone trying to cut back on drinking? The same volume knob appears to be working on alcohol cravings too, which is why the most surprising clinical signal has come from patients who never asked for a weight-loss drug in the first place.
The Rest of the Evidence
The Copenhagen trial did not arrive in a vacuum. A smaller American study, published in JAMA Psychiatry in February 2025 by Christian Hendershot and colleagues, gave low-dose semaglutide to adults with alcohol use disorder for nine weeks. The drug significantly reduced alcohol intake during a laboratory self-administration task and lowered weekly craving scores compared with placebo [5]. The newer trial simply used a higher dose, ran longer and grew the effect size.
Out in the real world, a 2024 analysis by Fares Qeadan and colleagues tracked thousands of patients with opioid or alcohol use disorders who were already taking GLP-1 drugs for diabetes or weight loss. Among those with alcohol use disorder, semaglutide prescriptions were associated with a 36% lower rate of alcohol-related hospitalisations; liraglutide was associated with a 28% lower rate [6]. The patterns held even after the researchers adjusted for the usual confounders. "Pathways implicated in addiction also contribute to pathological overeating and obesity," noted a 2025 review led by Lorenzo Leggio of NIDA and NIAAA, which is one reason the obesity-plus-AUD overlap is becoming a research sweet spot [7]. Taken together, the animal work, the small American pilot, the Copenhagen phase 2b and the real-world data all point in the same direction.
Why Doctors Are Already Prescribing It
The numbers are stark. Alcohol use disorder accounts for about 5% of deaths worldwide each year, and in the United States, fewer than one in four people with an alcohol or drug problem received any treatment in 2023 [1, 7]. The FDA has approved fewer than 10 medications for addiction, and only three of those (naltrexone, acamprosate and disulfiram) target alcohol, despite more than 48 million Americans being affected [4].
So clinicians are voting with their prescription pads. A 2026 commentary in The Lancet warned that "optimism about incretin therapies has led to off-label prescribing of GLP-1 therapies for alcohol use disorder," and Psychiatric News documented the same trend, noting that the official label for semaglutide still does not include alcohol use disorder [8, 9]. Lembke sees the move as pragmatic rather than reckless: "In order to adapt to this environment, we need all the tools we can find. It's the reality of the world we live in now" [4]. Her broader framing is that GLP-1 drugs demonstrate "exciting early promise in stemming the rising tide of addictive disorders", even as longer studies are still needed. For a primary care doctor with a patient whose drinking will not budge, that calculus often favours prescribing.
The Caveats That Matter
The Klausen trial was a single-centre phase 2b study, not a phase 3 confirmatory mega-trial. Every participant had a BMI of 30 or higher, which means the result may not generalise to non-obese patients with alcohol use disorder. Earlier work with a different GLP-1 drug, exenatide, found no significant alcohol reduction in a non-obese population, with a hint of benefit only in the obese subgroup, which is the design clue the Copenhagen team followed [1, 7]. Both arms of the new trial also received cognitive behavioural therapy, so the published effect is for semaglutide on top of counselling, not the injection alone.
The 41% figure, the one bouncing around the press, is a within-arm change from each participant's own baseline, not the active-versus-placebo contrast. The cleaner head-to-head number is 13.7 percentage points [1]. Adverse events were mostly the expected gastrointestinal grumbling; nothing alarming, but no one yet knows what five or ten years of continuous GLP-1 use does to a non-diabetic brain. The phase 3 data, and an eventual FDA review, will determine whether semaglutide ever earns a label that mentions a drink.
For now, the takeaway is both more modest and more interesting than the headlines suggest. A drug designed for diabetes and now famous for weight loss appears to nudge the brain's reward thermostat down a few notches, enough to help some people with alcohol use disorder drink less. The researchers behind the trial believe the effect is real. The clinicians already writing the prescriptions believe it is worth acting on. The rest of medicine is, cautiously, taking notes.
Medical Disclaimer
This article discusses off-label use of semaglutide and other GLP-1 receptor agonists for alcohol use disorder. The U.S. Food and Drug Administration, the European Medicines Agency, Australia's Therapeutic Goods Administration and other major regulators have not approved these medications for the treatment of alcohol use disorder. Off-label prescribing is a legal and common practice in which a licensed clinician prescribes a drug for a use not specifically listed on its approved label, but it carries different risk and liability profiles than on-label use. Anyone considering semaglutide or a related medication, on- or off-label, should consult a qualified health professional. Alcohol use disorder is a serious medical condition; help and information are available from a primary care provider, an addiction specialist, or national services such as the U.S. SAMHSA National Helpline (1-800-662-4357) and Australia's National Alcohol and Other Drug Hotline (1800 250 015).