GLP-1 Agonists and Alcohol: The Drinking Effect
GLP-1 Psychology
36% fewer drinks
In the first randomized trial, semaglutide significantly reduced drinks per drinking day and weekly alcohol craving compared to placebo in adults with alcohol use disorder.
Hendershot et al., JAMA Psychiatry, 2025
Hendershot et al., JAMA Psychiatry, 2025
View as imagePeople taking semaglutide for weight loss or diabetes started reporting something unexpected: they no longer wanted to drink. Anecdotal reports accumulated through 2022 and 2023, and researchers who had been studying GLP-1 receptors in brain reward circuits were not surprised. The GLP-1 receptor is expressed in the nucleus accumbens, ventral tegmental area, and other addiction-relevant brain regions, and preclinical evidence that GLP-1 agonists reduce alcohol consumption in rodents had been building for years. In February 2025, the first randomized clinical trial confirmed what the animal data and patient reports had suggested: semaglutide reduces alcohol craving and drinking in humans with alcohol use disorder[1]. This article is part of the GLP-1 psychology cluster, examining how these drugs affect behavior and mental health beyond metabolic endpoints.
Key Takeaways
- In the first RCT (48 adults with AUD), semaglutide significantly reduced drinks per drinking day, heavy drinking days, and weekly alcohol craving versus placebo over 9 weeks (Hendershot et al., JAMA Psychiatry, 2025)
- Semaglutide reduces alcohol drinking in rodents by enhancing GABA signaling in the central amygdala and infralimbic cortex, brain regions central to addiction (Chuong et al., JCI Insight, 2023)
- A case series of 6 patients on semaglutide for weight loss showed a mean 9.5-point reduction in AUDIT scores, with 100% of patients improving (Richards et al., 2023)
- In a Swedish cohort of over 5 million, semaglutide users had 36% lower risk of alcohol use disorder hospitalization (adjusted HR 0.64) compared to non-users
- Semaglutide attenuates alcohol-induced dopamine release in the nucleus accumbens without blocking normal dopamine function, selectively dampening reward-driven signaling
- Phase 3 trials are underway; no GLP-1 agonist is approved for alcohol use disorder, and only 3 FDA-approved medications currently exist for AUD
How GLP-1 receptors ended up in addiction neuroscience
GLP-1 is best known as a gut hormone that stimulates insulin secretion and suppresses appetite. But GLP-1 receptors are also densely expressed in brain regions that process reward, motivation, and habit formation: the ventral tegmental area (VTA), nucleus accumbens (NAc), central amygdala, and prefrontal cortex. These are the same circuits hijacked by addictive substances.
The connection between GLP-1 and addictive behavior was first demonstrated in rodent studies showing that GLP-1 receptor activation in the VTA and NAc reduced the rewarding properties of cocaine, nicotine, and alcohol. The mechanism is not broad dopamine suppression. GLP-1 receptor agonists selectively attenuate drug-induced dopamine release in the nucleus accumbens while preserving baseline dopaminergic function. This context-dependent modulation, dampening reward-driven hyperactivity without flattening normal motivation, distinguishes GLP-1 agonists from traditional dopamine-blocking approaches to addiction. The finding also helps explain why patients on GLP-1 drugs do not typically report anhedonia or loss of pleasure from non-drug activities, a problem that plagues other pharmacological interventions targeting reward circuits. GLP-1 agonists appear to recalibrate rather than suppress the reward system.
The animal evidence: semaglutide reduces alcohol intake in rats
Aranas et al. (2023): both sexes, relapse prevention
Aranas and colleagues at the University of Gothenburg demonstrated that semaglutide reduced alcohol intake and relapse-like drinking in both male and female rats[2]. The relapse finding is particularly significant because preventing return to heavy drinking after a period of abstinence is one of the hardest challenges in alcohol use disorder treatment. The effect was consistent across sexes, which is relevant because sex differences in alcohol pharmacology and addiction neurobiology are well-documented.
Chuong et al. (2023): the GABA mechanism
An NIH-led team including George Koob (then director of the National Institute on Alcohol Abuse and Alcoholism) published a mechanistic study in JCI Insight[3]. Semaglutide dose-dependently reduced binge-like alcohol drinking in mice and reduced both binge and dependence-induced drinking in rats.
The mechanism: semaglutide increased spontaneous inhibitory postsynaptic current (sIPSC) frequency in two brain regions central to addiction, the central amygdala (CeA) and infralimbic cortex (ILC). This means semaglutide enhanced GABAergic (inhibitory) signaling in regions that regulate compulsive alcohol seeking. Enhanced GABA release in the central amygdala reduces the anxiety and negative emotional states that drive compulsive drinking in dependent individuals.
One critical nuance: the GABA-enhancing effect was present in alcohol-naive rats but absent in alcohol-dependent rats. This suggests chronic alcohol exposure may alter how GLP-1 receptors interact with GABA circuits, which has implications for whether semaglutide would work differently in early-stage versus severe, long-term alcohol dependence.
Aranas et al. (2024): combining GLP-1 with amylin
The Gothenburg group followed up by testing whether combining a GLP-1 agonist (dulaglutide) with an amylin receptor agonist (salmon calcitonin) would enhance alcohol reduction[4]. Adding the amylin agonist to ongoing GLP-1 treatment reduced alcohol intake in both sexes without tolerance development. When both drugs were started simultaneously, tolerance developed over time. This sequencing effect suggests that established GLP-1 receptor activation may create a foundation for additional peptide-based interventions against alcohol use.
Human evidence: from case reports to randomized trials
Case series: semaglutide patients who stopped wanting to drink
Richards et al. published a case series of six patients prescribed semaglutide for weight loss who had co-existing alcohol use disorder[5]. All six (100%) showed significant improvement in AUD symptoms. The mean Alcohol Use Disorder Identification Test (AUDIT) score dropped by 9.5 points, a clinically meaningful change. Patients described reduced interest in alcohol rather than forced abstinence. They simply wanted to drink less.
Quddos et al. (2023): semaglutide and tirzepatide in obesity patients
Quddos and colleagues analyzed alcohol consumption patterns in individuals with obesity taking semaglutide or tirzepatide. Both drugs were associated with reduced alcohol consumption[6]. This observational data added to the signal from case reports but could not establish causality because patients taking GLP-1 agonists were also losing weight, changing their diets, and potentially modifying other behaviors simultaneously.
Swedish cohort: population-level signal
A large Swedish cohort study involving over 5 million individuals with more than 8 years of follow-up found that semaglutide users had the lowest risk of alcohol use disorder hospitalization among GLP-1 agonist users (adjusted hazard ratio 0.64, representing a 36% risk reduction). The risk of any substance use disorder hospitalization was also lower (aHR 0.68). A separate analysis found semaglutide use was associated with 50% reduced risk of recurrent AUD diagnoses and 75% lower risk compared to naltrexone or topiramate, the current standard pharmacotherapies.
Hendershot et al. (2025): the first randomized trial
The study that moved this from hypothesis to evidence was published in JAMA Psychiatry in February 2025[1]. This phase 2, double-blind, randomized trial enrolled 48 non-treatment-seeking adults with alcohol use disorder. Participants received low-dose semaglutide (titrated from 0.25 mg to 1.0 mg weekly over 9 weeks) or placebo.
Results: semaglutide significantly reduced drinks per drinking day and heavy drinking days compared to placebo. Weekly alcohol craving was significantly lower in the semaglutide group. In a post-treatment laboratory alcohol self-administration test, semaglutide-treated participants consumed less alcohol. Among participants who also smoked, semaglutide reduced cigarettes per day, suggesting a broader reward-modulating effect.
Limitations are real: the trial was small (48 participants), short (9 weeks), and used low doses. The participants were not seeking treatment for AUD, which means results may not generalize to treatment-seeking populations with more severe dependence. Mean weight loss in the semaglutide group was 5%, raising the question of whether some behavioral changes were mediated by weight loss rather than direct brain effects.
The ghrelin connection
Ghrelin and GLP-1 have opposing effects on alcohol reward. Ghrelin increases alcohol craving through the same mesolimbic dopamine circuits that GLP-1 agonists suppress. Jerlhag published a comprehensive review in 2024 arguing that the ghrelin and GLP-1 systems represent complementary therapeutic targets for alcohol use disorder[7]. Blocking ghrelin signaling reduces the "wanting" component of alcohol reward, while activating GLP-1 signaling reduces the "liking" and compulsive consumption components. A 2025 review by the same group formally positioned GLP-1 receptor agonists as promising therapeutic targets for AUD based on the converging preclinical and clinical evidence[8].
This peptide interplay matters because it suggests that ghrelin receptor antagonists and GLP-1 agonists might be combined for more complete suppression of alcohol reward signaling than either approach alone. No such combination has been tested clinically.
How this compares to existing AUD medications
Only three medications are FDA-approved for alcohol use disorder: naltrexone (an opioid receptor antagonist), acamprosate (which modulates glutamate signaling), and disulfiram (which causes unpleasant reactions to alcohol). All three have modest efficacy. Naltrexone, the most widely used, reduces heavy drinking days by about 17% compared to placebo in meta-analyses.
Semaglutide operates through an entirely different mechanism. Where naltrexone blocks opioid receptors involved in alcohol's hedonic effects, semaglutide modulates dopaminergic and GABAergic signaling through GLP-1 receptors that naltrexone does not target. This mechanistic distinction is why the Swedish cohort data showing semaglutide outperforming naltrexone (75% lower risk of recurrent AUD with semaglutide versus naltrexone or topiramate) is particularly striking, though it comes from observational data with significant confounding. Patients prescribed semaglutide differ from those prescribed naltrexone in ways that could explain the difference without invoking superior drug efficacy.
The potential for combining GLP-1 drugs with existing AUD treatments has not been tested but is mechanistically appealing. GLP-1 and opioid pathways converge on reward processing through different receptor systems, suggesting additive or synergistic effects are plausible.
What we do not know
The evidence for GLP-1 agonists reducing alcohol consumption is building rapidly, but significant gaps remain.
Dose-response: the Hendershot trial used low-dose semaglutide. Whether higher doses (2.4 mg, as used for weight loss) produce stronger effects is unknown. Higher doses cause more gastrointestinal side effects, which could independently reduce alcohol intake through nausea and reduced appetite rather than central reward circuit modulation.
Dependence severity: the GABA-enhancing mechanism observed by Chuong et al. was absent in alcohol-dependent rodents. If chronic alcohol exposure alters GLP-1-GABA interactions, semaglutide might be more effective for early-stage problem drinking than for severe, long-term alcohol dependence.
Duration of effect: all studies to date are short. Whether semaglutide's alcohol-reducing effects persist with months or years of treatment, or whether tolerance develops, is unknown. The broader question of whether GLP-1 drugs can treat addiction hinges partly on durability.
Mechanism specificity: semaglutide also reduced intake of non-alcoholic caloric solutions in the Chuong study. The effect may not be alcohol-specific but rather a general reduction in reward sensitivity, which could have implications (both positive and negative) for motivation, pleasure, and quality of life.
Weight loss confounding: patients taking GLP-1 agonists lose weight, change their eating patterns, and often report altered relationships with food and social eating. Disentangling the direct neurological effects on alcohol reward from the indirect behavioral changes accompanying weight loss is methodologically challenging.
The Bottom Line
GLP-1 receptor agonists, particularly semaglutide, reduce alcohol craving and consumption through selective modulation of brain reward circuits involving dopamine and GABA signaling. The first randomized trial (Hendershot et al., JAMA Psychiatry, 2025) confirmed preclinical findings, showing reduced drinking and craving in adults with AUD. Population-level data from Sweden show a 36% lower risk of AUD hospitalization among semaglutide users. Phase 3 trials are underway, but no GLP-1 drug is approved for alcohol use disorder. Whether effects persist long-term and whether they work equally in severe dependence remain open questions.