GLP-1 and Addiction: How Weight Loss Drugs May Curb Substance Use

Introduction

The first stories were anecdotal. Patients on semaglutide for diabetes started telling their doctors they had stopped drinking, lost interest in cocaine, or quit chewing nicotine gum without trying. Then the case reports turned into retrospective database studies. Then the small randomized trials arrived.

By 2025 there were at least eight published controlled studies and a flood of large electronic health record analyses suggesting GLP-1 receptor agonists reduce substance use across multiple categories: alcohol, nicotine, opioids, stimulants, and even compulsive shopping. The effect sizes vary, but the signal is unusually broad for a drug class whose original target was pancreatic insulin secretion.

The mechanism makes biological sense, the early human data is interesting, and addiction medicine is paying attention. None of it justifies prescribing semaglutide for alcohol use disorder today, but the picture by 2027 may look very different.

At TrimRx, we believe that understanding your options is the first step toward a more manageable health journey. You can take the free assessment quiz if you’re ready to see whether a personalized program is a fit for you.

How Could a Diabetes Drug Reduce Drinking?

GLP-1 receptors are expressed in the mesolimbic reward circuit. The ventral tegmental area projects dopaminergic neurons to the nucleus accumbens, and both regions show GLP-1 receptor activity. When agonists like semaglutide bind those receptors, they appear to dampen drug-induced dopamine release.

Quick Answer: The 2024 JAMA Psychiatry trial (Klausen et al.) tested exenatide in 127 patients with alcohol use disorder and found significant heavy drinking reduction in obese subgroup

In rats, GLP-1 agonists reduce self-administration of alcohol, cocaine, amphetamine, nicotine, and even palatable food. The animals press the lever less often, drink less, and show reduced reinstatement after a forced abstinence period.

The translation to humans is the open question. Reward circuitry is conserved across mammals, but human addiction involves cognition, stress, social context, and craving in ways rat models miss. Still, when you have a candidate drug that hits both metabolic reward (food) and chemical reward (alcohol, nicotine, opioids) in animal work, the cross-modality consistency is striking.

What Did the Human Alcohol Trials Show?

The Klausen et al. JAMA Psychiatry 2024 trial randomized 127 patients with alcohol use disorder to exenatide (an older weekly GLP-1) versus placebo for 26 weeks. The primary endpoint (heavy drinking days) was not significantly different overall. But in the prespecified subgroup of patients with BMI over 30, exenatide cut heavy drinking days by about 24% and reduced cue-induced brain reward activation on fMRI.

This was the first large randomized addiction trial of a GLP-1 agonist. It missed the headline endpoint but produced the most provocative signal in the obese subgroup, suggesting the drug works better when the patient’s reward system is already metabolically dysregulated.

In late 2025 a phase 2 trial of semaglutide in alcohol use disorder, led by Lorenzo Leggio at NIH/NIDA, completed primary analysis. Topline results showed a statistically significant reduction in drinks per drinking day and craving over 9 weeks. Full publication is expected in 2026 with neuroimaging substudy data.

What About Opioids?

The 2024 Nature Communications paper by Wang and colleagues analyzed roughly 500,000 patient records and found semaglutide users had a 49% lower incidence of opioid use disorder diagnoses over 12 months compared with patients on non-GLP-1 diabetes drugs. Among patients already diagnosed with opioid use disorder, semaglutide use was associated with fewer overdose-coded hospitalizations.

Observational data is not a randomized trial. People prescribed semaglutide differ systematically from those who aren’t. But a 50% effect size, replicated in subsequent analyses of Veterans Affairs and Medicaid data, is hard to attribute entirely to confounding.

A small phase 2 trial of liraglutide in opioid use disorder maintenance patients (Hartwell et al. 2024) found reduced craving on a visual analog scale but no change in opioid-positive urine screens. The trial was 8 weeks and 40 patients, underpowered for hard endpoints. Larger trials are recruiting in 2026.

Does GLP-1 Help People Quit Smoking?

The nicotine signal is the most consistent across animal studies but the human data is thinner. Yammine and colleagues (2021) ran a small RCT of exenatide added to nicotine patches in 84 smokers and found higher 7-day point-prevalence abstinence at 24 weeks (46% versus 27%). The effect was concentrated in overweight smokers.

A 2024 Annals of Internal Medicine analysis of TriNetX records on 222,000 smokers with diabetes found semaglutide users had a 32% lower rate of new tobacco use disorder diagnoses than DPP-4 inhibitor users. Again, observational, but consistent with the trial direction.

The mechanism likely overlaps with alcohol and food. GLP-1 receptors dampen reward responses to nicotine, and the secondary effect on weight (smokers often relapse when they gain weight after quitting) may help maintain abstinence.

Could Compounded Semaglutide Become an Addiction Treatment?

Not yet, and not for a long time on regulatory grounds. The FDA approves drugs for specific indications based on adequate trials. Off-label prescribing exists, but no addiction physician should be prescribing semaglutide for alcoholism today based on a 127-person trial that missed its primary endpoint.

The pipeline is moving. Novo Nordisk and Eli Lilly have both opened phase 2 programs in alcohol use disorder, and the National Institute on Drug Abuse is funding multiple academic trials. By 2028 we should have at least one phase 3 addiction trial readout with a GLP-1 or GIP/GLP-1 dual agonist.

For patients with weight loss indications who also struggle with alcohol or nicotine, the cardiometabolic benefits remain the primary reason to prescribe. Any addiction benefit is welcome but unproven at the individual level. TrimRx clinicians focus on the established metabolic indications and do not prescribe for off-label addiction treatment.

What About Food Addiction and Binge Eating?

Binge eating disorder is the third arm of the reward story, and probably the most clinically relevant for the GLP-1 prescribing population. The 2023 Hudson et al. study tested liraglutide 3.0 mg in 27 patients with binge eating disorder and obesity. Binge episodes dropped from a mean of 6 per week to 1.4 per week at 17 weeks, with parallel reductions on the Yale Food Addiction Scale.

Semaglutide phase 2 trials in binge eating disorder are ongoing. The mechanism is clearer here: slowed gastric emptying produces earlier satiety, central GLP-1 signaling reduces palatability-driven eating, and patients consistently report that food simply “stops calling to them.”

This food-noise reduction is what most GLP-1 patients describe spontaneously. Whether the same mechanism translates to alcohol craving, opioid craving, and stimulant craving in the same way is the open question the trials are answering.

What Are the Risks of Using GLP-1 in Addiction Medicine?

GLP-1 agonists in patients with active substance use carry several risks beyond the usual side effects. Patients in early alcohol withdrawal can’t tolerate nausea well, and adding semaglutide during detox could worsen vomiting and electrolyte disturbances. Opioid use disorder patients often have erratic eating patterns, and superimposing satiety drugs on already-poor nutrition could accelerate sarcopenia or hypoglycemia.

Suicidality signals in early FDA pharmacovigilance reports raised concerns. A 2024 BMJ meta-analysis of GLP-1 trials found no excess suicidality versus placebo (pooled hazard ratio around 1.0), but addiction populations have higher baseline suicide risk and the safety profile would need separate study.

Combining GLP-1 with naltrexone, buprenorphine, or methadone has not been studied at scale. Drug-drug interactions appear minimal pharmacokinetically, but pharmacodynamic interactions on reward circuitry are unknown.

How Do GLP-1 Agonists Compare with Existing Addiction Medications?

For alcohol use disorder, naltrexone and acamprosate are the main approved drugs. Both have modest effect sizes (numbers needed to treat around 12 for naltrexone, around 9 for acamprosate). If a phase 3 GLP-1 trial shows a 30% reduction in heavy drinking days, that would be competitive with naltrexone and probably better tolerated long-term, especially in obese patients.

For opioid use disorder, methadone and buprenorphine are still the most effective treatments, with naltrexone third-line. No GLP-1 drug will replace medication-assisted treatment for opioid use disorder, but as an adjunct to reduce craving in stabilized patients, it could find a role.

For nicotine, varenicline is the most effective drug but has its own neuropsychiatric concerns. A GLP-1 agonist combined with nicotine replacement might offer a useful new option, especially for smokers worried about post-cessation weight gain.

Key Takeaway: GLP-1 receptors live in the ventral tegmental area and nucleus accumbens, brain regions central to reward processing

Could GLP-1 Work for Gambling and Behavioral Addictions?

Anecdotally yes, and a few small studies have started in compulsive shopping and pathological gambling. The hypothesis is that any reward-driven behavior involving dopaminergic reinforcement could respond to GLP-1 modulation. Hayes and colleagues (2024) published a 28-patient open-label case series of semaglutide in pathological gambling with reduction in gambling time and Gambling Symptom Assessment Scale scores.

Behavioral addictions are harder to study than substance addictions because endpoints are softer and patients are less likely to enroll in trials. The signals are intriguing but exploratory.

What Does This Mean for Someone on GLP-1 WHO Used to Drink Heavily?

If you started semaglutide or tirzepatide for weight loss and noticed you stopped wanting alcohol, this is a real and documented effect, not your imagination. Many patients report this within the first month and the effect tends to persist as long as the drug does. Patients who stop the medication often see drinking interest return within weeks.

For a patient with controlled alcohol use disorder, this is welcome. For a patient with significant cardiovascular comorbidities, the alcohol reduction may amplify the metabolic benefits. For a patient with a fragile recovery history, the unintended drug effect on reward circuits is worth discussing with both an addiction specialist and the prescribing clinician on TrimRx’s personalized treatment plan.

What Trials Should We Be Watching?

Through 2026 and 2027, watch the Leggio NIH semaglutide trial in alcohol use disorder publication, the Novo Nordisk phase 2 of semaglutide in AUD, ongoing tirzepatide trials in behavioral addiction, and the larger pharmacovigilance datasets emerging from Optum, TriNetX, and Veterans Affairs records.

By 2028 the field will likely have answered the core question: does GLP-1 modulation of reward circuitry produce clinically meaningful, durable reductions in substance use disorder severity? The animal data and observational data both lean positive. The hardest evidence (large randomized phase 3 trials with sustained follow-up) is still to come.

How Does Reward Sensitivity Actually Change on GLP-1?

Functional MRI studies offer a window. The 2022 Hanssen et al. study scanned 20 obese patients on liraglutide and 20 on placebo while showing them food cues. Brain activation in the orbitofrontal cortex, ventral striatum, and amygdala dropped significantly in the liraglutide group. Smaller fMRI studies in alcohol cues, gambling cues, and nicotine cues showed the same pattern of dampened reward-circuit activation.

What patients describe subjectively maps onto what scanners show objectively. The “food noise” reduction patients talk about is measurable in cortical and subcortical activation patterns. Whether the same neural dampening fully explains the reduced drinking, smoking, and drug use signals is plausible but not proven at the imaging level for every substance.

Are There Populations Where GLP-1 Addiction Effects Might Backfire?

Yes. Patients with bulimia nervosa, severe anorexia history, or restrictive eating disorders should not be prescribed GLP-1 agonists for any indication without specialist eating disorder care. The drug’s anorexigenic effect could destabilize a fragile relationship with food.

Patients on chronic opioid therapy for cancer pain may have legitimate reasons their reward circuits should stay receptive. Dampening response to opioids in palliative care could undermine analgesia. The same logic applies to medication-assisted addiction treatment: buprenorphine and methadone work partly through reward modulation, and adding a GLP-1 agonist might interfere in ways nobody has rigorously studied.

For most patients in the metabolic prescribing population, none of these concerns apply. The reduced alcohol or nicotine craving is a clean unintended bonus.

What Is the Regulatory Pathway for GLP-1 in Addiction?

For semaglutide to gain an FDA approval in alcohol use disorder, Novo Nordisk would need a phase 3 trial with a clinically meaningful primary endpoint (typically percent heavy drinking days or abstinence) over at least 6 months. The trial would need to enroll patients without obesity to separate the addiction effect from the metabolic effect.

The economics matter. Semaglutide is approaching loss of patent exclusivity (2032 in major markets), so the financial incentive to fund a $200 million addiction trial is shrinking. Smaller phase 2 trials funded by NIDA and academic groups may be the main source of evidence for several more years. Generic semaglutide approved for addiction would be a public health win but a tough investment case for Novo Nordisk.

Bottom line: The first phase 2 semaglutide trial in alcohol use disorder (Leggio NIH) read out in 2025 with positive primary endpoint

FAQ

Does Ozempic® Make You Stop Drinking?

Many patients on Ozempic or Wegovy® report reduced alcohol craving and intake, and this has been documented in observational and small randomized studies. It is not an FDA-approved indication and the effect size in randomized trials so far has been modest, with the biggest effect in patients with both obesity and alcohol use disorder.

Can Semaglutide Treat Alcohol Use Disorder?

There is no FDA approval for semaglutide in alcohol use disorder. Phase 2 trials are ongoing. The Klausen 2024 JAMA Psychiatry trial of exenatide missed its overall primary endpoint but showed significant effects in the obese subgroup. The NIH Leggio semaglutide trial reported positive topline results in 2025 with full publication pending.

Will Tirzepatide Help Me Quit Smoking?

Tirzepatide has not been tested in dedicated smoking cessation trials. Older GLP-1 drugs (exenatide, liraglutide) have shown signals in small RCTs and observational data, with effect sizes concentrated in overweight smokers.

Are GLP-1 Drugs Being Used for Opioid Addiction?

Not as a standalone treatment. Methadone and buprenorphine remain the first-line treatments for opioid use disorder. Small phase 2 trials suggest GLP-1 agonists may reduce craving when added to medication-assisted treatment, but this is investigational.

Why Would a Diabetes Drug Affect Addiction?

GLP-1 receptors are expressed in brain reward circuits including the ventral tegmental area and nucleus accumbens. Activating these receptors appears to dampen drug-induced dopamine release in animal models, producing reduced self-administration of alcohol, cocaine, nicotine, and opioids.

Is the Reduced Alcohol Craving Permanent?

In most reported cases, the reduction in alcohol craving and intake reverses when patients stop the medication. The effect appears to be pharmacologically maintained rather than producing durable abstinence after drug discontinuation.

Should I Tell My Doctor I Drink Less on Semaglutide?

Yes. This is clinically relevant information that helps your prescriber understand your overall response and adjust monitoring. It may also matter for liver enzyme tracking, blood pressure, and overall metabolic improvement attribution.

Disclaimer: This content is for informational purposes only and does not constitute medical advice. It is not intended to diagnose, treat, cure, or prevent any disease or condition. Individual results may vary. Always consult a qualified healthcare professional before starting any weight loss program or medication.