GLP-1 Receptor Agonists (Semaglutide, Liraglutide) for Alcohol Use Disorder

A Research-Grade Clinical Knowledge Base Article

Overview

Glucagon-like peptide-1 (GLP-1) receptor agonists — a class of medications that includes semaglutide (brand names Ozempic and Wegovy) and liraglutide — are FDA-approved for type 2 diabetes and obesity. They are not FDA-approved for alcohol use disorder (AUD). As of 2026, that distinction matters enormously.

What has changed is the weight of evidence pointing toward a real signal. A growing body of research — from animal studies, to large observational datasets, to a landmark randomized controlled trial published in 2026 — suggests these medications may meaningfully reduce heavy drinking in people with AUD, particularly those who also have obesity. The signal is real. The evidence is encouraging. But it is not yet definitive enough to call these medications a standard treatment for AUD.

This article synthesizes the full evidence base as it stands: what we know, what we don't, where the research is headed, and how to talk honestly with patients about all of it.

What GLP-1 Agonists Are

GLP-1 receptor agonists were originally developed to treat type 2 diabetes. They work by mimicking a natural gut hormone — glucagon-like peptide-1 — that is released after eating. This hormone tells the pancreas to release insulin, slows digestion, and signals the brain that the body is full.

When researchers and clinicians noticed that people taking these medications were losing significant amounts of weight, the drugs were studied and eventually approved for obesity as well. Semaglutide at a higher dose (2.4 mg weekly) is marketed as Wegovy for weight management; at a lower dose (0.5–2 mg weekly), it is marketed as Ozempic for diabetes.

The side effect that opened the door to addiction research was this: many people on GLP-1 agonists reported that they simply stopped wanting things they used to crave — not just food, but alcohol, cigarettes, and other substances. That observation, combined with what scientists already knew about how these drugs work in the brain, launched a new line of research.

Mechanism Relevant to Addiction

To understand why GLP-1 agonists might reduce alcohol use, it helps to understand how alcohol affects the brain's reward system.

Alcohol activates the brain's dopamine reward pathway — particularly a circuit running from the ventral tegmental area (VTA) to the nucleus accumbens (NAc). This is the same pathway involved in all addictive substances. When alcohol triggers dopamine release in this circuit, it creates the feeling of reward that drives continued drinking.

GLP-1 receptors are present throughout this reward circuitry. Preclinical studies — meaning studies in rodents and non-human primates — have consistently shown that GLP-1 agonists reduce alcohol self-administration and blunt the dopamine response to alcohol. This is the biological foundation for the hypothesis that these drugs might help people with AUD.

Critically, the corpus contains early human evidence supporting this mechanism. The 2022 exenatide RCT by Klausen et al. included a neuroimaging substudy and found that exenatide significantly attenuated fMRI alcohol cue reactivity in the ventral striatum and septal area, and that dopamine transporter availability was lower in the exenatide group compared to placebo ^[1]. A subsequent meta-analysis reinforced this, noting that neuroimaging studies reported attenuated alcohol cue reactivity and dopaminergic signaling with GLP-1 agonist use ^[2]. This is not a rodent extrapolation — it is direct human neurobiological evidence of reward circuit modulation.

However, the central reward pathway is not the only mechanism in play. A pilot study by Quddos et al. found that people taking GLP-1 agonists showed a delayed rise in breath alcohol concentration and subjective alcohol effects after an alcohol challenge — and this effect was not explained by nausea ^[3]. This points to a peripheral mechanism: GLP-1 agonists slow gastric emptying, which in turn slows how quickly alcohol is absorbed into the bloodstream. Slower absorption means lower peak blood alcohol levels and reduced intoxication.

A third mechanistic thread involves metabolism. A preliminary study by Harvanek et al. found that in people with both AUD and obesity, glucose levels and insulin resistance (measured by HOMA-IR) were significantly associated with provoked alcohol craving — but this relationship was not present in people with AUD who did not have obesity ^[4]. This suggests that metabolic dysregulation may amplify alcohol craving in people with comorbid obesity, and that GLP-1 agonists — which directly address insulin resistance — may reduce craving partly through metabolic normalization in this specific population.

The honest summary on mechanism: Three plausible pathways exist — central reward attenuation, peripheral alcohol pharmacokinetics, and metabolic modulation. The human evidence supports all three as possibilities. No mediation analysis from any adequately powered human study has yet determined which pathway drives the clinical effect. The SEMALCO trial protocol planned neuroimaging substudies to address this directly ^[5], but those results are not yet available in the published literature.

Observational Evidence

Before any randomized trials were conducted in AUD populations, researchers noticed a signal in large real-world datasets: people prescribed GLP-1 agonists for diabetes or obesity appeared to drink less and have fewer alcohol-related health events.

The Swedish nationwide cohort is among the most compelling observational studies. Using a within-individual design across 227,866 patients with AUD, researchers found that semaglutide was associated with an adjusted hazard ratio (aHR) of 0.64 (95% CI: 0.50–0.83) for AUD hospitalization — a 36% reduction in risk. Notably, this effect exceeded that of all FDA-approved AUD medications, which showed an aHR of 0.98 (essentially no effect) in the same dataset [lähteenvuo-2025-repurposing-semaglutide-liraglutide]. The within-individual design — comparing each person's outcomes during periods on versus off the medication — reduces some confounding concerns, though it cannot eliminate them entirely.

Wang et al. conducted a retrospective cohort study of over 83,000 patients with obesity and found that semaglutide was associated with a 50–56% lower risk of both the incidence and recurrence of AUD over 12 months compared to other anti-obesity medications [lähteenvuo-2025-repurposing-semaglutide-liraglutide]. This is a striking effect size, though observational studies of this kind carry unmeasured confounding that propensity matching cannot fully address.

A VA-based target trial emulation by John et al. found GLP-1 agonist use associated with lower odds of a positive AUDIT-C screen (aOR 0.75, 95% CI: 0.68–0.82) and a lower composite liver-related mortality outcome (aHR 0.70) ^[6]. The same study found a dose-response signal: each 1 mg/week increase in semaglutide dose was associated with reduced composite liver-related outcomes (aHR 0.50, 95% CI: 0.29–0.88) and death (aHR 0.33, 95% CI: 0.19–0.58) ^[6].

Liver outcomes deserve specific mention. Multiple studies demonstrate that GLP-1 agonist use is associated with reduced alcohol-related liver disease development (HR: 0.62) ^[7], reduced cirrhosis and decompensation ^[7], and lower composite liver-related mortality ^[6]. For people with AUD who already have liver damage, this is a clinically meaningful signal.

What observational evidence can and cannot tell us: These datasets generate hypotheses and identify signals. They cannot prove that GLP-1 agonists caused reduced drinking. People prescribed these medications differ systematically from those who are not — in health-seeking behavior, access to care, metabolic health, and many other ways. The consistent direction of effect across multiple large, independent datasets is meaningful. The magnitude of effect should be interpreted cautiously.

Pilot and Early RCT Evidence

The Exenatide Trial (2022)

The first randomized controlled trial of a GLP-1 agonist specifically in AUD was conducted by Klausen et al. using exenatide 2 mg once-weekly in 127 patients over 26 weeks ^[1]. The trial failed to meet its primary endpoint — it did not show a significant reduction in heavy drinking days in the overall population.

However, two secondary findings were important. First, the neuroimaging substudy showed significant attenuation of fMRI alcohol cue reactivity in the ventral striatum — the human mechanistic bridge described above. Second, in an exploratory analysis of the subgroup with obesity (BMI >30), exenatide did show a significant reduction in heavy drinking days. This BMI-dependent signal would prove to be a critical design insight for the next trial.

A separate biomarker substudy of the exenatide trial found no significant changes in pro-inflammatory markers (IL-6, hsCRP, FGF-21) after 26 weeks of treatment ^[8], suggesting the mechanism is more likely central or pharmacokinetic rather than anti-inflammatory.

The SEMALCO Trial (2026) — Current Best Evidence

The SEMALCO trial is the most important clinical trial in this evidence base. It was a 26-week, single-center, randomized, double-blind, placebo-controlled trial of semaglutide 2.4 mg subcutaneously once-weekly — the full obesity-approved dose — in 108 patients with moderate-to-severe AUD and comorbid obesity (BMI ≥30) ^[9].

The primary result: semaglutide reduced heavy drinking days by 41.1 percentage points from baseline, compared to 26.4 percentage points in the placebo group. The estimated treatment difference was −13.7 percentage points (95% CI: −22.0 to −5.4; p=0.0015) ^[9]. This is the first adequately powered RCT to demonstrate a statistically significant reduction in heavy drinking days with a GLP-1 agonist. The trial also reported "substantial effects on multiple secondary alcohol-related and somatic outcomes" ^[9], and the protocol confirmed secondary endpoints included total alcohol consumption, phosphatidylethanol (PEth) as an objective biomarker, quality of life, and FIB-4 liver fibrosis score ^[5]. Adverse events were predominantly mild-to-moderate gastrointestinal effects.

What SEMALCO does and does not establish: This is one trial, at one center, in a carefully selected population — people with both moderate-to-severe AUD and comorbid obesity, receiving cognitive behavioral therapy alongside medication. It cannot yet be generalized to the broader AUD population. It requires independent replication. It is, however, a genuine and meaningful advance.

Meta-Analytic Evidence

Two systematic reviews and meta-analyses synthesize the available evidence.

Sinha & Ghosal pooled three RCTs (N=430) — which did not include SEMALCO, published after their analysis — and found non-significant reductions in alcohol consumption (SMD: −0.24, 95% CI: −0.70 to 0.23) and craving (SMD: −0.14, 95% CI: −2.84 to 2.55) ^[10]. The craving confidence interval is so wide it is essentially uninformative. However, the same meta-analysis found that semaglutide specifically showed greater craving reduction (p=0.024) compared to other agents — a signal buried within the null headline result. Observational studies in the same meta-analysis showed reduced alcohol-related events (HR: 0.64, 95% CI: 0.59–0.69) ^[10] (Note: this specific figure could not be independently verified against the source abstract — the underlying study supports the general finding but the exact number should be confirmed before publication).

Eshraghi et al. pooled 14 studies and found a significant reduction in AUDIT scores (mean difference: −7.81 points, 95% CI: −9.02 to −6.60), though an I² of 87.5% signals substantial heterogeneity across studies ^[2].

De Faria Moraes et al. found no significant pooled effect on alcohol intake (SMD: −0.17; 95% CI: −0.39 to 0.04) across their meta-analysis ^[11].

Reconciling the meta-analytic null with SEMALCO's positive result: The apparent contradiction is largely explainable. The meta-analyses pooled heterogeneous compounds (exenatide, liraglutide, semaglutide at various doses), heterogeneous patient populations (with and without obesity), and heterogeneous outcome measures. Pooling a null exenatide result in unselected AUD patients with semaglutide data at lower doses mathematically dilutes a signal that appears to be dose- and population-dependent. SEMALCO used the highest approved semaglutide dose in an obesity-selected population — the pharmacologically optimized condition. The meta-analytic null and SEMALCO's positive result are not contradictory; they reflect the importance of dose, agent, and patient selection as genuine moderators of effect.

Ongoing Pivotal Trials

SEMALCO, while landmark, is a single-center trial requiring replication. Larger, multi-site trials are underway. The NIAAA has funded multi-site randomized controlled trials of semaglutide in AUD, with readouts expected over the next one to three years. These trials will be essential for determining whether SEMALCO's findings hold across more diverse populations and clinical settings.

Preclinical data on tirzepatide — a dual GLP-1/GIP receptor agonist — are promising ^[2], though no completed human RCTs in AUD exist as of 2026.

Real-World Comparative Evidence

One real-world study compared GLP-1 agonist users to people taking FDA-approved AUD medications (naltrexone, acamprosate, disulfiram) and found GLP-1 agonists associated with a lower one-year relapse rate (IRR: 0.55) ^[12]. This is a provocative finding. It is also observational, and confounding by indication — the possibility that GLP-1 agonist users differ systematically from AUD medication users in ways that affect outcomes — must be acknowledged. A head-to-head randomized comparison with naltrexone does not yet exist.

A separate analysis using a DPP-4 inhibitor comparator design found that GLP-1 agonist effects on AUDIT-C were substantially larger in individuals with baseline AUD (difference-in-differences: 0.51) and baseline hazardous drinking (difference-in-differences: 1.38) ^[2], suggesting severity-dependent effects.

Side Effects

The side effect profile of GLP-1 agonists is well-established from their use in diabetes and obesity. In the context of AUD trials, adverse events have been predominantly mild to moderate.

Common side effects:
- Nausea — the most frequently reported side effect; usually occurs early in treatment and improves over time as the dose is titrated up gradually
- Vomiting
- Diarrhea and other gastrointestinal upset
- Constipation
- Decreased appetite (which is part of the intended effect in obesity but may be a concern in people with AUD who already have nutritional deficiencies)

Less common but serious:
- Pancreatitis — rare; patients should be counseled to seek care for severe, persistent abdominal pain
- Thyroid C-cell tumors — a boxed warning based on rodent data; the clinical relevance in humans is uncertain, but the warning exists

Drug interactions:
- Hypoglycemia — risk increases when combined with sulfonylureas or insulin; less relevant in AUD-only patients without diabetes

Psychiatric safety signals: Two documents in the corpus raise questions about psychiatric adverse effects, including suicidality and mood changes ^{[13] [14]}. These signals require monitoring and should be discussed with patients, particularly given the high rates of co-occurring depression and anxiety in people with AUD. The evidence on this question is preliminary and the corpus does not resolve it definitively.

Contraindications

Based on the established safety profile of GLP-1 agonists:

• Personal or family history of medullary thyroid carcinoma — absolute contraindication
• Multiple endocrine neoplasia type 2 (MEN-2) syndrome — absolute contraindication
• Pregnancy — not recommended; weight loss during pregnancy is not appropriate
• History of pancreatitis — relative contraindication; requires careful risk-benefit discussion
• Severe gastrointestinal disease — use with caution

Off-Label Prescribing for AUD

As of 2026, prescribing semaglutide or liraglutide specifically for AUD is off-label in the United States. There is no FDA indication for this use. Some clinicians are prescribing these medications off-label for patients with AUD, particularly those with comorbid obesity or type 2 diabetes, based on the pilot evidence. This is a clinical judgment call that requires honest informed consent.

Payer coverage for off-label GLP-1 use in AUD is variable and generally unlikely. Cost is a major access barrier (see below).

The FDA-approved medications for AUD — naltrexone, acamprosate, and disulfiram — remain the evidence-based first-line pharmacological options. These are dramatically underutilized: the Swedish nationwide cohort found that FDA-approved AUD medications were associated with an aHR of 0.98 for hospitalization [lähteenvuo-2025-repurposing-semaglutide-liraglutide], a near-null effect that likely reflects poor adherence and the well-documented treatment gap rather than drug inefficacy. Before pivoting to expensive off-label prescribing, the underutilization of proven treatments deserves attention.

Cost and Access

GLP-1 agonists are expensive. Brand-name semaglutide (Wegovy, Ozempic) costs approximately $900–$1,300 per month without insurance coverage. Insurance coverage for off-label use in AUD is unlikely in most plans.

This creates a serious equity problem. People with AUD have disproportionately high rates of poverty, housing instability, and insurance gaps. The population most likely to benefit from these medications — based on current evidence — is people with AUD and comorbid obesity, a group that already faces significant barriers to care. A case report in the corpus illustrates this concretely: a patient experienced relapse following insurance loss that interrupted GLP-1 agonist treatment ^[15].

A compounded semaglutide market exists, offering lower-cost alternatives, but quality and dosing consistency concerns are real. The corpus contains no health economics or cost-effectiveness analyses for GLP-1 agonists in AUD — a critical gap for health policy.

Combination with FDA-Approved AUD Medications

No randomized trials have tested GLP-1 agonists in combination with naltrexone, acamprosate, or disulfiram in AUD populations. The mechanisms are different enough to be theoretically synergistic: naltrexone works primarily through opioid receptor blockade to reduce the rewarding effects of alcohol, while GLP-1 agonists appear to work through dopaminergic reward modulation and/or peripheral pharmacokinetics. No safety signals against combination have been identified, but this has not been studied at scale. This remains an important evidence gap.

Signal for Other Substance Use Disorders

Observational data suggest GLP-1 agonists may reduce use of other substances beyond alcohol, including opioids, nicotine, and stimulants. This has generated a broader hypothesis that GLP-1 agonists may have a generalized effect on addictive behavior through reward pathway modulation. Active research is underway. As of 2026, the evidence for other substances is even more preliminary than for AUD — largely observational, with no completed pivotal RCTs. This is a promising area that warrants monitoring but should not be cited as established evidence.

Hype vs. Evidence: An Honest Assessment

GLP-1 agonists have received extraordinary media attention, including coverage framing them as a potential "cure" for addiction. Social media enthusiasm has significantly outpaced the evidence. This matters because overpromising to people with AUD — a population that has often experienced treatment failures and stigma — can cause real harm.

Here is what the evidence actually supports as of 2026:

What we know:
- Consistent observational signal across multiple large, independent datasets showing reduced alcohol-related events in people taking GLP-1 agonists [lähteenvuo-2025-repurposing-semaglutide-liraglutide] [lähteenvuo-2025-repurposing-semaglutide-liraglutide] ^[6]
- One well-designed, adequately powered RCT (SEMALCO) showing a statistically significant reduction in heavy drinking days in people with AUD and comorbid obesity ^[9]
- Human neuroimaging evidence of reward circuit modulation ^{[1] [2]}
- A plausible, multi-pathway biological mechanism

What we don't know:
- Whether the effect generalizes to people with AUD who do not have obesity
- Whether the effect is neurologically, metabolically, or peripherally mediated — or some combination
- How GLP-1 agonists compare head-to-head with naltrexone in a randomized design
- Long-term outcomes beyond 26 weeks
- Optimal dose and duration for AUD specifically
- Whether effects are maintained after stopping the medication

What is coming:
- Results from NIAAA-funded multi-site RCTs over the next one to three years
- Neuroimaging substudy data from SEMALCO
- Trials in populations without metabolic comorbidity

The honest framing: this is a promising signal with one positive landmark trial and a strong observational foundation. It is not yet a proven treatment. The next two to three years of trial data will be decisive.

How to Talk to Patients About GLP-1 Agonists for AUD

People with AUD are increasingly asking about these medications, often having read about them online or in the news. Here is a framework for honest, empathetic counseling:

1. Acknowledge the interest and the signal.
"You're right that there's real research on this, and some of it is genuinely encouraging. Let me tell you what we actually know."

2. Be clear about off-label status.
"These medications are FDA-approved for diabetes and obesity, but not for alcohol use disorder. Using them for AUD would be off-label — meaning it's legal for a doctor to prescribe, but it's not an approved use, and insurance is unlikely to cover it for this reason."

3. Describe the evidence level accurately.
"The best evidence we have right now is one well-designed trial of 108 people with AUD and obesity, which showed a meaningful reduction in heavy drinking days. That's an important finding, but it's one trial in a specific population. Larger trials are underway."

4. Discuss who the evidence applies to.
"The strongest evidence is in people who have both AUD and obesity. If that's you, the evidence is more relevant. If you don't have obesity, we honestly don't yet know if these medications would help."

5. Review the side effect profile.
"The main side effects are nausea, vomiting, and other GI symptoms — usually worst early on and improving over time. There are also some rarer risks we'd want to screen you for."

6. Address cost honestly.
"These medications cost over $1,000 a month without insurance coverage for this use. That's a real barrier we need to think through."

7. Discuss FDA-approved options first.
"We have medications that are FDA-approved for AUD — naltrexone and acamprosate — that have strong evidence and are more likely to be covered by insurance. I'd want to make sure we've considered those as well."

Evidence Gaps

The panel identified the following as the most critical unanswered questions:

1. Definitive RCT data in broader populations. SEMALCO enrolled people with AUD and comorbid obesity. Whether the effect generalizes to people with AUD who do not have obesity is unknown and untested in any RCT.

2. Mechanism mediation. No mediation analysis from any adequately powered human study has determined whether the clinical effect operates through central reward attenuation, peripheral alcohol pharmacokinetics, or metabolic normalization — or what proportion each contributes.

3. Head-to-head comparison with naltrexone. No randomized trial has directly compared a GLP-1 agonist to naltrexone in AUD. Real-world comparisons exist ^[12] but are observational.

4. Optimal dose and duration. The dose-response relationship for semaglutide in AUD is suggested by observational data ^[6] but has not been tested in a dose-ranging RCT.

5. Long-term outcomes. No RCT data exist beyond 26 weeks. What happens to drinking behavior after stopping the medication is unknown.

6. Cost-effectiveness and access. The corpus contains no health economics analyses for GLP-1 agonists in AUD. Given the cost of these medications and the socioeconomic profile of many people with AUD, this is a critical policy gap.

7. Psychiatric safety. Preliminary signals regarding suicidality and mood effects [corpus-gap] require larger-scale safety monitoring in AUD populations specifically.

8. Combination pharmacotherapy. No trials have tested GLP-1 agonists combined with naltrexone or other AUD medications.

Summary

GLP-1 receptor agonists represent a genuinely promising new direction in AUD pharmacotherapy. The evidence base has matured meaningfully: from consistent observational signals across hundreds of thousands of patients, to a landmark randomized trial showing a statistically significant reduction in heavy drinking days in people with AUD and comorbid obesity ^[9]. The biological rationale is sound, with human neuroimaging evidence supporting reward circuit modulation ^[1] and preliminary evidence for peripheral pharmacokinetic effects ^[3].

At the same time, the evidence has real limits. This is one positive trial in a selected population. The mechanism is not fully understood. Long-term data are absent. Cost and access barriers are substantial. FDA-approved AUD medications remain underutilized and should be considered first.

The next two to three years — as NIAAA-funded multi-site trials read out — will determine whether GLP-1 agonists become a standard part of AUD treatment. Until then, the honest message is: real signal, early evidence, promising but not proven, and off-label. Patients deserve to hear exactly that.

This article reflects the evidence base as of early 2026. The field is moving rapidly; readers should check for updated trial results and any changes in FDA approval status.

Verified References

• ^[11] de Faria Moraes, Bernardo, André Pedral Diniz Leite, Gabriel, André Pedral Diniz Leite, Gustavo et al. (2025). "Impact of glucagon-like peptide-1 receptor agonists on alcohol consumption and liver-related outcomes: A systematic review and meta-analysis.". Drug Alcohol Depend. DOI: 10.1016/j.drugalcdep.2025.112840 [abstract-verified: yes]
• ^[2] Edvardsson, Christian E, Adermark, Louise, Gottlieb, Sam et al. (2026). "Tirzepatide reduces alcohol drinking and relapse-like behaviours in rodents.". EBioMedicine. DOI: 10.1016/j.ebiom.2025.106119 [abstract-verified: partial]
• ^[2] Eshraghi, Reza, Ghadimi, Delaram J, Montazerinamin, Sara et al. (2025). "Effects of glucagon-like peptide-1 receptor agonists on alcohol consumption: a systematic review and meta-analysis.". EClinicalMedicine. DOI: 10.1016/j.eclinm.2025.103645 [abstract-verified: partial]
• ^[2] Farokhnia, Mehdi, Tazare, John, Pince, Claire L et al. (2025). "Glucagon-like peptide-1 receptor agonists, but not dipeptidyl peptidase-4 inhibitors, reduce alcohol intake.". J Clin Invest. DOI: 10.1172/jci188314 [abstract-verified: partial]
• ^[12] Gougol, Amir, Kwo, Paul, Pike, William et al. (2026). "Real-World Alcohol Use Disorder Outcomes in Patients With Concurrent Metabolic Dysfunction: GLP-1 Receptor Agonists Versus FDA-Approved AUD Medications.". Aliment Pharmacol Ther. DOI: 10.1111/apt.70596 [abstract-verified: yes]
• ^[4] Harvanek, Zachary M, Milivojevic, Verica, Hart, Rachel et al. (2026). "Alterations in glucose and insulin resistance following stress and alcohol cues predict alcohol craving in those with AUD and obesity: A preliminary study.". Alcohol Clin Exp Res (Hoboken). DOI: 10.1111/acer.70283 [abstract-verified: yes]
• ^[15] Hill, Olivia, Hughes, Sarah, Singh, Aakanksha et al. (2025). "The role of dulaglutide in the treatment of alcohol use disorder: a case report.". Front Psychiatry. DOI: 10.3389/fpsyt.2025.1420316 [abstract-verified: partial]
• ^[8] Hviid, Malthe E B, Christoffersen, Lea A N, Klausen, Mette K et al. (2025). "Effect of the GLP-1 receptor agonist exenatide on pro-inflammatory and metabolic biomarkers in individuals with alcohol use disorder: Post hoc results from a randomized, double-blinded, placebo-controlled clinical trial.". Alcohol Clin Exp Res (Hoboken). DOI: 10.1111/acer.70110 [abstract-verified: yes]
• ^[7] Hwang, Soo Young, Hsieh, Pinghsin, Díaz, Luis Antonio et al. (2026). "Glucagon-like peptide-1 receptor agonist reduces risk of alcohol-associated cirrhosis in type 2 diabetes and alcohol use disorder patients.". Eur J Gastroenterol Hepatol. DOI: 10.1097/meg.0000000000003162 [abstract-verified: partial]
• ^[6] John, Binu V, Bastaich, Dustin, Marchetti, Daniella et al. (2026). "Association of Glucagon-Like Peptide-1 Receptor Agonists With Liver-Related Outcomes and All-Cause Mortality in Patients With Harmful Alcohol Use: A Target Trial Emulation Study.". Am J Gastroenterol. DOI: 10.14309/ajg.0000000000003585 [abstract-verified: yes]
• ^[1] Mette Kruse Klausen, Mathias Ebbesen Jensen, Marco Møller et al. (2022). "Exenatide once weekly for alcohol use disorder investigated in a randomized, placebo-controlled clinical trial.". JCI insight. DOI: 10.1172/jci.insight.159863 [abstract-verified: yes]
• ^[5] Klausen, Mette Kruse, Kuzey, Tugba, Pedersen, Julie Niemann et al. (2025). "Does semaglutide reduce alcohol intake in Danish patients with alcohol use disorder and comorbid obesity? Trial protocol of a randomised, double-blinded, placebo-controlled clinical trial (the SEMALCO trial).". BMJ Open. DOI: 10.1136/bmjopen-2024-086454 [abstract-verified: partial]
• ^[9] Klausen, Mette Kruse, Justesen, Signe Keller, Pedersen, Julie Niemann et al. (2026). "Once-weekly semaglutide versus placebo in patients with alcohol use disorder and comorbid obesity: a randomised, double-blind, placebo-controlled trial.". Lancet. DOI: 10.1016/s0140-6736(26)00305-3 [abstract-verified: yes]
• ^[7] Kuo, Chia-Chih, Li, Chun-Hsien, Chuang, Min-Hsiang et al. (2025). "Impact of GLP-1 Receptor Agonists on Alcohol-Related Liver Disease Development and Progression in Alcohol Use Disorder.". Aliment Pharmacol Ther. DOI: 10.1111/apt.70007 [abstract-verified: partial]
• [lähteenvuo-2025-repurposing-semaglutide-liraglutide] Lähteenvuo, Markku, Tiihonen, Jari, Solismaa, Anssi et al. (2025). "Repurposing Semaglutide and Liraglutide for Alcohol Use Disorder.". JAMA Psychiatry. DOI: 10.1001/jamapsychiatry.2024.3599 [abstract-verified: yes]
• ^[3] Quddos, Fatima, Fowler, Mary, de Lima Bovo, Ana Carolina et al. (2025). "A preliminary study of the physiological and perceptual effects of GLP-1 receptor agonists during alcohol consumption in people with obesity.". Sci Rep. DOI: 10.1038/s41598-025-17927-w [abstract-verified: yes]
• ^[14] Sa, Brianna, Maristany, Anthony, Subramaniam, Ashwin et al. (2026). "Retrospective chart review on psychiatric manifestations of GLP-1 agonist usage.". J Psychiatr Res. DOI: 10.1016/j.jpsychires.2026.01.042 [abstract-verified: partial]
• ^[13] Sharafshah, Alireza, Lewandrowski, Kai-Uwe, Gold, Mark S et al. (2025). "In Silico Pharmacogenomic Assessment of Glucagon-like Peptide-1 (GLP1) Agonists and the Genetic Addiction Risk Score (GARS) Related Pathways: Implications for Suicidal Ideation and Substance Use Disorder.". Curr Neuropharmacol. DOI: 10.2174/011570159x349579241231080602 [abstract-verified: yes]
• ^[10] Sinha, Binayak, Ghosal, Samit (2025). "The effects of glucagon-like peptide-1 receptor agonists (GLP1-RAs) on alcohol-related outcomes: a systematic review and meta-analysis.". Addict Sci Clin Pract. DOI: 10.1186/s13722-025-00637-z [abstract-verified: partial]
• [lähteenvuo-2025-repurposing-semaglutide-liraglutide] Wang, William, Volkow, Nora D, Berger, Nathan A et al. (2024). "Associations of semaglutide with incidence and recurrence of alcohol use disorder in real-world population.". Nat Commun. DOI: 10.1038/s41467-024-48780-6 [abstract-verified: partial]

Replacement Resolution Audit

Each REPLACE verdict from the adjudication pass was resolved by re-querying the indexed fulltext corpus and selecting the highest-scoring paper that the Level 3 verifier confirmed supports the claim.

• ^[16] → ^[17] (verifier: yes; score 0.84). Title: Glucagon-like peptide-1 receptor agonists, but not dipeptidyl peptidase-4 inhibitors, reduce alcohol intake.
• [lähteenvuo-2025-repurposing-semaglutide-liraglutide] → ^[18] (verifier: partial; score 0.66). Title: Associations of semaglutide with incidence and recurrence of alcohol use disorder in real-world population.
• ^[18] → [lähteenvuo-2025-repurposing-semaglutide-liraglutide] (verifier: yes; score 0.76). Title: Repurposing Semaglutide and Liraglutide for Alcohol Use Disorder.
• ^[18] → ^[6] (verifier: partial; score 0.81). Title: A systematic review on the role of glucagon-like peptide-1 receptor agonists on alcohol-related behaviors: potential the
• ^[19] → NO REPLACEMENT FOUND (considered 5 candidates; none verified)
• ^[19] → ^[6] (verifier: partial; score 0.81). Title: A systematic review on the role of glucagon-like peptide-1 receptor agonists on alcohol-related behaviors: potential the
• ^[20] → ^[7] (verifier: yes; score 0.54). Title: Association of topiramate prescribed for any indication with reduced alcohol consumption in electronic health record dat
• ^[21] → ^[7] (verifier: yes; score 0.54). Title: Association of topiramate prescribed for any indication with reduced alcohol consumption in electronic health record dat
• ^[17] → ^[2] (verifier: yes; score 0.75). Title: Tirzepatide reduces alcohol drinking and relapse-like behaviours in rodents.
• ^[22] → ^[13] (verifier: partial; score 0.70). Title: Medications for Alcohol Use Disorder Among Patients With Severe Alcohol-Related Liver Disease.
• ^[14] → NO REPLACEMENT FOUND (considered 5 candidates; none verified)
• ^[23] → ^[15] (verifier: partial; score 0.73). Title: An examination between treatment type and treatment retention in persons with opioid and co-occurring alcohol use disord