"GLP-1 and Addiction Circuitry: Alcohol, Nicotine, More (2026)",

The most surprising thing about Ozempic, Wegovy, Mounjaro and Zepbound isn't what they do to your stomach — it's what people keep reporting about their cravings. Patients on GLP-1 drugs describe drinking less, smoking less, and feeling the constant pull of food, alcohol, or even shopping and gambling quiet down. That isn't just marketing or wishful thinking: GLP-1 receptors sit squarely inside the brain's mesolimbic reward system, the dopamine circuit that drives wanting and reinforcement for everything from a slice of cake to a cigarette. Activating those receptors blunts the reward signal — which is exactly how these drugs curb appetite, and plausibly how they touch other reward behaviors too. The animal evidence is strong and consistent. The human evidence is genuinely promising but still early: one small published randomized trial in alcohol use disorder (Hendershot 2025 ^[1]), large but observational electronic-health-record studies (Wang 2024 ^[2][3]), and a handful of signals for nicotine, opioids, and cannabis. None of this is FDA-approved for addiction. This article explains the circuitry, walks through what is and isn't established, and is honest about the limits. For the deeper alcohol-specific evidence map, see GLP-1 drugs and alcohol use disorder; for how the same circuitry quiets food cravings, see the neuroscience of GLP-1 “food noise.”

The honest summary

• One mechanism, many targets. GLP-1 receptors are expressed in the brain's reward circuitry — the ventral tegmental area (VTA) and nucleus accumbens (NAc). Activating them dampens the dopamine surge that addictive substances and behaviors normally trigger, which is the leading explanation for effects on appetite and, plausibly, on alcohol and nicotine (Klausen 2022 review ^[4]).
• The animal evidence is robust. In rodents and nonhuman primates, GLP-1 drugs (exenatide, liraglutide, semaglutide) reliably reduce alcohol intake and blunt alcohol-induced dopamine release; semaglutide cut binge-like and dependence-driven drinking across multiple models (Chuong 2023 ^[5]).
• The best human alcohol data is one small RCT. Hendershot 2025 (n=48, 9 weeks, low-dose semaglutide) significantly reduced alcohol craving and lab-measured drinking versus placebo — and, in the 13 smokers, also reduced cigarettes per day ^[1].
• Large observational studies point the same way — but can't prove cause. In 83,825 adults with obesity, semaglutide was linked to ~50% lower incident and ~56% lower recurrent alcohol-use-disorder diagnoses (Wang 2024 ^[2]); a target-trial-emulation study found fewer tobacco-use-disorder encounters (Wang 2024 ^[3]).
• The signal extends beyond alcohol and nicotine. EHR studies report associations with lower opioid-overdose risk (Wang 2024 ^[6]) and reduced cannabis-use-disorder incidence/relapse (Wang 2024 ^[7]) — early, hypothesis-generating, not confirmatory.
• It is NOT approved for addiction. No GLP-1 drug is FDA-approved to treat any substance-use disorder. The standard of care for alcohol use disorder remains naltrexone, acamprosate, and disulfiram. Don't start or stop a GLP-1 to manage drinking or smoking without a prescriber.

How GLP-1 drugs reach the brain's reward circuitry

GLP-1 (glucagon-like peptide-1) is best known as a gut hormone that lowers blood sugar and slows gastric emptying. But the GLP-1 receptor is also expressed in the central nervous system — including the mesolimbic dopamine system, the circuit that assigns reward value and drives motivation. Two nodes matter most: the ventral tegmental area (VTA), where dopamine neurons originate, and the nucleus accumbens (NAc), where they project. This is the same circuit that lights up for food, alcohol, nicotine, opioids, and behavioral rewards like gambling. Addictive substances all converge on it by driving a dopamine surge that the brain reads as “do that again” (Klausen 2022 ^[4]).

When a GLP-1 receptor agonist activates receptors in or upstream of this circuit, it appears to turn the gain down on that reward signal. In animal studies, GLP-1 drugs blunt the alcohol-induced rise in accumbal dopamine and reduce the motivation to work for a reward. The leading hypothesis is that the same satiety-and-reward machinery that makes a cheeseburger feel less compelling on Wegovy also makes a drink or a cigarette feel less compelling — because they all share the dopamine pathway (Klausen 2022 ^[4]). That is why researchers describe GLP-1 receptor agonists as a potential transdiagnostic tool for reward-driven behaviors, rather than a drug specific to any one substance.

The animal evidence: strong, consistent, and mechanistic

The preclinical case is the most solid part of this story. Across rodents and nonhuman primates, GLP-1 receptor agonists — exenatide, liraglutide, and semaglutide — reduce alcohol intake and suppress the alcohol-triggered activation of the mesolimbic dopamine system (Klausen 2022 ^[4]). The clearest single study is Chuong 2023 ^[5], which showed that semaglutide dose-dependently reduced binge-like alcohol drinking in mice and cut both binge-like and dependence-induced drinking in rats. It also altered GABA neurotransmission in reward-related brain regions, offering a cellular-level mechanism beyond “dopamine goes down.” Similar work has shown GLP-1 receptor activity in regions like the nucleus accumbens and lateral septum reducing alcohol reinforcement.

Two caveats keep this honest. First, animal models of “addiction” capture intake and reinforcement, not the lived human experience of a substance-use disorder. Second, the doses and timeframes don't map cleanly onto human dosing. The animal data establish a plausible, well-characterized mechanism — they don't establish that the drugs work as addiction treatments in people. For that, you need human trials.

The human alcohol evidence: one small RCT, several large cohorts

The single most important human study is Hendershot 2025 ^[1], a phase 2, double-blind, randomized trial published in JAMA Psychiatry. It enrolled 48 non-treatment-seeking adults with alcohol use disorder and gave them low-dose semaglutide (titrated from 0.25 mg up to a possible 1.0 mg/week) or placebo for 9 weeks. The results were encouraging: semaglutide significantly reduced the amount of alcohol consumed in a controlled laboratory self-administration test (medium-to-large effect), reduced weekly alcohol craving (β −0.39; P = .01), and lowered drinks per drinking day. Strikingly, among the 13 participants who smoked, semaglutide also produced a greater relative reduction in cigarettes per day (β −0.10; P = .005) — a hint that the effect isn't alcohol-specific.

But the trial was small, short, used low doses, and enrolled people with moderate (not severe, treatment-seeking) AUD — the authors flag all of this. An earlier randomized trial of exenatide for alcohol use disorder (Klausen 2022 ^[8], n=127, 26 weeks) did not reduce heavy-drinking days overall, but did show a significant reduction in a prespecified subgroup of participants with obesity (BMI > 30). That mixed result is important: it suggests the effect may be larger in people with higher body weight, and that the headline “GLP-1 cures drinking” framing is premature.

The largest human numbers come from observational electronic-health-record studies. Wang 2024 ^[2] analyzed 83,825 adults with obesity and found semaglutide was associated with roughly 50% lower risk of a new alcohol-use-disorder diagnosis (hazard ratio 0.50) and 56% lower risk of recurrence (HR 0.44) over 12 months, versus other anti-obesity medications — with consistent results replicated in a population of nearly 600,000 patients with type 2 diabetes. A 2024 systematic review pulled the RCT subgroups and observational data together and concluded GLP-1 drugs may reduce alcohol consumption in some individuals (Subhani 2024 ^[9]).

Beyond alcohol: nicotine, opioids, and other reward behaviors

If the mechanism is a shared reward circuit, the effect shouldn't stop at alcohol — and the early data suggest it doesn't. For nicotine, Wang 2024 ^[3] used a target-trial-emulation design and found that, among patients with type 2 diabetes and tobacco use disorder, semaglutide was associated with significantly fewer tobacco-use-disorder-related medical encounters and fewer smoking-cessation-medication prescriptions than other antidiabetic drugs — and the Hendershot RCT's smoking subgroup pointed the same direction ^[1].

The signal extends further. For opioids, Wang 2024 ^[6] reported that among patients with type 2 diabetes and opioid use disorder, semaglutide was associated with substantially lower opioid-overdose risk than other antidiabetic medications. For cannabis, Wang 2024 ^[7] found semaglutide associated with lower incidence and lower relapse of cannabis use disorder in real-world populations. All of these are observational, hypothesis-generating, and limited to patients who already had the relevant diagnosis or risk — they justify dedicated trials, not clinical use. There is also emerging interest in behavioral rewards (compulsive shopping, gambling), but that rests almost entirely on anecdote and circuit logic, not trial data.

What this means for you — the practical upshot

• Reduced cravings are a plausible, real effect — not a guarantee. If you're on a GLP-1 and notice you want alcohol or cigarettes less, that's consistent with the mechanism and the data. But responses vary widely, and the effect was clearest in people with higher body weight in the trials.
• Don't use a GLP-1 as an addiction treatment on your own. It is not FDA-approved for alcohol, nicotine, opioid, or any substance-use disorder. Self-directing it for that purpose means taking a drug with real side effects for an unproven indication.
• If you have a substance-use disorder, use proven treatments. For alcohol, that means naltrexone, acamprosate, or disulfiram plus behavioral support; for nicotine, varenicline, bupropion, or nicotine-replacement therapy. Talk to a clinician — these have strong evidence behind them.
• Tell your prescriber about your drinking and smoking. Alcohol independently compounds some labeled GLP-1 risks (pancreatitis, hypoglycemia in people on insulin or a sulfonylurea, dehydration). The interaction is worth a conversation regardless of the craving question — see can you drink alcohol on Ozempic.
• Watch this space, but read the headlines skeptically. “Ozempic cures addiction” coverage runs far ahead of the evidence. The truth is more interesting and more measured: a well-characterized brain mechanism, strong animal data, and promising early human signals that now need confirmation in real trials.

Bottom line

GLP-1 drugs act on the brain's mesolimbic dopamine reward circuit — the same system that drives appetite and reinforces alcohol, nicotine, and other rewards — and dampening that signal is the most likely reason they affect cravings beyond food (Klausen 2022 ^[4]). The animal evidence is strong and mechanistic (Chuong 2023 ^[5]); the human evidence is promising but early, anchored by one small positive RCT in alcohol use disorder (Hendershot 2025 ^[1]), a mixed exenatide trial (Klausen 2022 ^[8]), and large observational signals for alcohol, nicotine, opioids, and cannabis (Wang 2024 ^[2][3][6][7]). None of it is FDA-approved for addiction, and observational associations cannot prove cause (Subhani 2024 ^[9]). The honest takeaway: a real, well-understood mechanism with genuinely encouraging early human data — and a clear instruction not to treat a GLP-1 as an addiction drug until the trials are done. Use proven therapies and coordinate with a clinician.

This article is educational and is not medical advice. Every claim above is sourced to a peer-reviewed randomized trial, mechanistic study, cohort study, or systematic review indexed in PubMed, verified against the live PubMed database before publication. No GLP-1 receptor agonist is FDA-approved for the treatment of any substance-use disorder. If you are concerned about alcohol, tobacco, or other substance use, talk to a clinician about evidence-based treatments.