GLP-1 Agonists and Alcohol Reduction: Emerging Research

In February 2025, a team led by Christian Hendershot at the University of North Carolina published the first randomized clinical trial of semaglutide for alcohol use disorder. The results, published in JAMA Psychiatry, showed that a low dose of semaglutide reduced alcohol craving and consumption in a controlled laboratory setting.

The paper landed on years of preclinical animal data, growing observational evidence, and thousands of patient self-reports on social media and in clinical practice. The collective picture is one of the most intriguing developments in addiction medicine in decades: a drug designed for diabetes and obesity may also help people drink less.

But this story is early. Very early. Here is what we actually know, what we do not, and why the excitement -- though warranted -- needs some guardrails.

The Hendershot Trial: First Controlled Human Evidence

The Hendershot et al. trial was a Phase 2, double-blind, randomized, parallel-arm study conducted at an academic medical center in the US. It enrolled 48 non-treatment-seeking adults with alcohol use disorder between September 2022 and February 2024.

Key design details:

Participants received either subcutaneous semaglutide (titrated up to a relatively low dose) or placebo for 9 weeks
The primary outcome was alcohol consumption during a posttreatment laboratory self-administration task -- participants were given access to their preferred alcoholic beverage in a controlled setting
Secondary outcomes included daily alcohol use, craving, and smoking behavior

What the Trial Found

In the lab setting:

Semaglutide significantly reduced grams of alcohol consumed (effect size: -0.48; 95% CI: -0.85 to -0.11; P = .01)
Semaglutide significantly reduced peak breath alcohol concentration (effect size: -0.46; 95% CI: -0.87 to -0.06; P = .03)

In daily life (measured throughout the 9-week treatment):

Semaglutide did not significantly affect average drinks per calendar day
Semaglutide did not significantly reduce total number of drinking days
Semaglutide did significantly reduce drinks per drinking day (P = .04) -- meaning participants drank less when they did drink
Weekly alcohol craving was significantly reduced (P = .01)
A treatment-by-time interaction showed semaglutide predicted greater reductions in heavy drinking over time relative to placebo

Bonus finding: In a subsample of participants who smoked cigarettes, semaglutide predicted greater reductions in cigarettes per day -- consistent with the broader addiction research on GLP-1 drugs.

How Big Are These Effects?

The effect sizes in the lab task were medium to large (-0.46 to -0.48), which compares favorably to existing FDA-approved medications for alcohol use disorder. The researchers noted that "effect sizes of semaglutide on alcohol consumption were larger than those associated with naltrexone or acamprosate," though they emphasized that head-to-head trials would be needed to confirm this comparison.

Limitations to Keep in Mind

Small sample (n = 48). This was designed to detect a signal, not to definitively prove efficacy.
Non-treatment-seeking participants. These were people with AUD who were not actively trying to quit. Their baseline drinking may have been lower than in treatment-seeking populations.
Short duration (9 weeks). We do not know whether the effects persist with longer treatment.
Low dose. The semaglutide dose was lower than what is used for weight management (2.4 mg). Whether higher doses would produce larger effects -- or different side effects -- is unknown.
No serious adverse events. This is encouraging, but 48 people over 9 weeks provides limited safety data, especially for potential interactions between semaglutide and alcohol.

The Preclinical Foundation: 15 Years of Animal Data

The Hendershot trial did not emerge from nowhere. It was built on roughly 15 years of animal research showing consistent effects of GLP-1 receptor agonists on alcohol-related behaviors.

Exenatide (Exendin-4)

Exenatide was the first GLP-1 RA tested for alcohol effects. Studies in mice and rats showed that systemic administration:

Blocked alcohol-induced dopamine release in the nucleus accumbens -- the brain region most associated with reward and "wanting"
Reduced alcohol-induced locomotor stimulation (a marker of alcohol's rewarding effects in rodents)
Reduced alcohol intake and alcohol-seeking behavior in rats after 8 months of alcohol exposure
Prevented relapse drinking in the alcohol deprivation model (where animals are given a period of forced abstinence and then re-exposed to alcohol)

Liraglutide

Liraglutide produced similar findings across species:

Decreased alcohol intake and preference in male rats, male mice, and vervet monkeys (nonhuman primates)
Attenuated alcohol-induced dopamine release in the nucleus accumbens
Decreased alcohol-induced conditioned place preference (a measure of whether the animal "likes" being in a location associated with alcohol)
Prevented relapse drinking after a period of deprivation
Suppressed the motivation to consume alcohol in alcohol-preferring rat strains

Semaglutide

Semaglutide-specific preclinical data is more limited but consistent with the other GLP-1 RAs. In mice, semaglutide suppressed relapse drinking in the alcohol deprivation model.

Dulaglutide

Weekly dulaglutide treatment for 5-9 weeks reduced both alcohol consumption and preference in male and female rats throughout the entire treatment period.

The Preclinical Consensus

A 2025 review in Endocrinology summarized the animal data: "Preclinical studies revealed that GLP-1R agonists lowered alcohol intake, the motivation to consume alcohol, alcohol-seeking, relapse-drinking, and abstinence symptoms. These effects are observed by all tested GLP-1R agonists and across different strains, species, and sexes."

The consistency across different drugs, different species, and different experimental paradigms is what made the field confident enough to move into human trials.

How It Works: The Dopamine Connection

The mechanism connecting GLP-1 drugs to alcohol reduction centers on the brain's reward circuitry -- the same system implicated in emotional eating and food noise.

GLP-1 Receptors in Reward Regions

GLP-1 receptors exist throughout the brain's mesolimbic and mesocortical circuits, including:

Ventral tegmental area (VTA): The origin of dopamine neurons that project to reward regions. The VTA is where the "wanting" signal starts.
Nucleus accumbens (NAc): The primary target of VTA dopamine neurons. This is where reward is registered and craving is generated.
Prefrontal cortex (PFC): Involved in impulse control, decision-making, and the ability to weigh long-term consequences against short-term urges.

When you drink alcohol, dopamine floods the NAc. This is what makes alcohol feel good and what makes you want more. GLP-1 receptor agonists appear to dampen this dopamine surge. The alcohol still enters your system, but the brain's reward response is muted.

The Virginia Tech Study: Peripheral Mechanisms

A 2025 study from Virginia Tech explored an additional mechanism. Participants on GLP-1 drugs "consistently reported feeling less intoxicated." This suggests the drugs may also act through peripheral pathways -- affecting how alcohol is absorbed or metabolized in the gut -- not just through central brain effects.

If confirmed, this dual mechanism (central reward reduction plus peripheral intoxication reduction) could explain why GLP-1 drugs seem to reduce drinking more effectively than the modest changes in any single pathway would predict.

A Comprehensive Mechanism Model

A 2025 review in Clinical and Medical Sciences proposed that GLP-1 RAs reduce substance-seeking behavior through multiple parallel mechanisms:

Modulation of dopaminergic signaling (reducing reward)
Modulation of glutamatergic signaling (reducing craving)
Modulation of GABAergic signaling (reducing impulsivity)
Improved satiety and reduced general reward sensitivity
Better impulse control through prefrontal cortex effects

Real-World Evidence: Large Observational Studies

While the randomized trial evidence is limited to the Hendershot trial and a couple of smaller studies, the observational data is more extensive.

The Finnish Registry Study

A major Finnish cohort study published in JAMA Psychiatry (2024) analyzed nationwide registry data with a median follow-up of more than 8 years. Among patients with AUD and comorbid obesity or type 2 diabetes:

Individuals were at "markedly lower risk" of alcohol-related hospitalizations when using GLP-1 agonists compared to periods when they were not using them
Semaglutide and liraglutide showed particularly strong associations with decreased AUD hospitalization risk
The reduction in hospitalization risk was lower than that seen with officially approved AUD medications -- meaning GLP-1 agonists appeared to outperform existing treatments (though this was not a head-to-head comparison and confounding is possible)

Other Observational Data

Multiple observational studies have found that semaglutide reduces AUD risk by 50-56% within 12 months compared to other anti-obesity medications. These are large-effect signals, though observational data cannot prove causation.

Systematic Review and Meta-Analysis (2025)

A systematic review published in eClinicalMedicine pooled three randomized trials comparing GLP-1 RAs to placebo. The meta-analysis showed a non-significant overall association (SMD -0.24, 95% CI: -0.70 to 0.23). However, the Hendershot and Probst trials individually found statistically significant reductions in alcohol outcomes. The non-significant pooled result was likely driven by the third trial (Klausen et al.), which tested exenatide in combination with cognitive behavioral therapy -- and the CBT may have attenuated the drug's observable effect by helping both groups equally.

A pooled analysis of AUDIT score changes across trials demonstrated a significant reduction (mean difference -7.81 points; 95% CI: -9.02 to -6.60), suggesting meaningful clinical improvement in overall alcohol use patterns.

The Ongoing Clinical Trials

The research pipeline is robust. As of early 2026, multiple Phase 3 trials are underway or recently completed:

Four trials of oral or injectable semaglutide for AUD, two of which have completed enrollment but not yet reported results
Two trials of tirzepatide (the dual GIP/GLP-1 agonist), which is particularly interesting because observational data suggests tirzepatide may have even larger effects on alcohol outcomes than semaglutide. Given tirzepatide's greater weight loss efficacy, this raises the question of whether the dual-agonist mechanism provides additional benefit for reward-driven behaviors.
Additional mechanistic studies using neuroimaging to map exactly how GLP-1 drugs alter the brain's response to alcohol cues

Important Caveats and Open Questions

Who Would Benefit Most?

The early evidence suggests GLP-1 drugs may work best for mild to moderate AUD rather than severe dependence. The Hendershot trial enrolled non-treatment-seeking participants, and the Finnish registry data showed the strongest effects in people with comorbid obesity.

Exenatide studies found a striking pattern: the drug significantly reduced alcohol intake in overweight AUD patients but did not alter consumption in normal-weight AUD patients. This suggests metabolic status may modify the treatment response -- which makes sense if part of the mechanism involves peripheral metabolic pathways.

Sex Differences

Animal studies have found sex-dependent differences in GLP-1 RA effects on alcohol behavior. The mechanisms behind these differences are not fully understood but may involve sex hormones. Human trials so far have been too small to detect sex-specific effects, but future studies will need to address this.

The Cost Problem

Semaglutide costs roughly $1,000-1,300 per month without insurance coverage. Currently approved AUD medications (naltrexone, acamprosate, disulfiram) cost a fraction of that -- generic naltrexone can be as low as $30-50 per month. Even if GLP-1 drugs prove effective for AUD, widespread adoption will require either insurance coverage for this indication or significant cost reductions.

The Administration Barrier

Weekly subcutaneous injections are a potential barrier to treatment adherence, especially in populations with AUD who may have inconsistent healthcare engagement. Oral semaglutide exists but is being tested at different doses than the injectable form, and its efficacy for AUD specifically has not been established.

This Is Not an FDA-Approved Treatment for AUD

As of this writing, no GLP-1 receptor agonist is approved for the treatment of alcohol use disorder. Prescribing semaglutide "off-label" for AUD is a clinical decision that should be made carefully, weighing the promising but limited evidence against the costs, potential side effects, and availability.

The Bigger Picture: From Metabolic Drug to Reward Drug

The alcohol findings are part of a broader pattern. GLP-1 receptor agonists appear to reduce reward-driven behavior across multiple domains:

Food: Reduced food noise, decreased craving for high-calorie foods
Alcohol: Reduced craving and consumption (this article)
Nicotine: Reduced cigarette consumption in the Hendershot trial; potential protective effect in the Oxford cohort study
Other substances: Preclinical data suggests effects on cocaine and opioid-seeking behavior

This pattern suggests GLP-1 drugs are not specifically "anti-alcohol" or "anti-food." They appear to reduce the brain's general reward-seeking intensity. The clinical implications are enormous: one drug class could potentially address multiple reward-driven conditions simultaneously.

But this generalization of reward dampening also raises questions about quality of life. If a drug reduces the pleasure from food, alcohol, and nicotine, does it also reduce pleasure from exercise, sex, music, or social connection? The emotional "flatness" reported by some GLP-1 users hints that this question deserves serious attention.

What This Means for Patients

If You Take a GLP-1 Drug and Notice Reduced Alcohol Interest

You are not imagining it. There is a biological basis for what you are experiencing. The drug is likely modulating the same reward circuits that drive both appetite and alcohol craving.

If you want to use this effect to reduce your drinking, that is reasonable -- but tell your prescriber. They can help you monitor for mood changes, track your progress, and connect you with behavioral support if needed.

If you are taking a GLP-1 drug for weight loss and your relationship with alcohol changes in ways that concern you -- drinking more to "feel something," for instance, or replacing food-based reward-seeking with alcohol -- that also warrants a conversation with your provider.

If You Have AUD and Are Wondering About GLP-1 Drugs

The research is promising but not yet definitive. FDA-approved medications for AUD (naltrexone, acamprosate) have decades of evidence behind them and are much more affordable. Behavioral treatments -- cognitive behavioral therapy, motivational interviewing, 12-step facilitation -- also have strong evidence.

If you have AUD and also have obesity or type 2 diabetes, the case for discussing a GLP-1 drug with your doctor is stronger, because the metabolic benefits are established and the alcohol reduction may be an added benefit.

The Bottom Line

The alcohol reduction effects of GLP-1 receptor agonists are one of the most exciting developments in addiction research in years. The Hendershot trial provides the first controlled human evidence that semaglutide can reduce alcohol craving and consumption. The preclinical data is extensive and consistent. The observational evidence suggests real-world effects.

But we are at the very beginning. A single Phase 2 trial with 48 participants, however well-designed, is not enough to change clinical practice. The Phase 3 trials currently underway will determine whether GLP-1 drugs become legitimate tools in the AUD treatment arsenal -- or whether the early promise fades under more rigorous testing.

For now, what the data supports: GLP-1 receptor agonists reduce reward-driven behavior across multiple domains, including alcohol. The effect is biologically plausible, consistently demonstrated in animals, and supported by early human evidence. If the upcoming trials confirm these findings, we may be looking at a new class of addiction treatment built on a mechanism no one predicted when these drugs were first designed for diabetes.

This article is part of the GLP-1 and Mental Health series. For related topics, see our coverage of semaglutide and depression, GLP-1 drugs and suicidal ideation, emotional eating and food noise, and body image after weight loss.