Abstract

This Lancet commentary by Hendershot and Klein reviews a double-blind, placebo-controlled, single-centre trial (Klausen et al., Lancet 2026) in which 108 treatment-seeking adults with alcohol use disorder and comorbid obesity were randomly assigned to subcutaneous semaglutide (titrated up to 2.4 mg/week) or placebo over 26 weeks. The primary endpoint was reduction in heavy drinking days at 26 weeks. Semaglutide outperformed placebo on the primary endpoint and on multiple secondary outcomes, including alcohol craving, drinks per drinking day, total alcohol consumption, and alcohol exposure biomarkers. The drug significantly increased the likelihood of achieving a 2-level reduction in WHO risk drinking level, a clinical trial endpoint recently qualified by the FDA, with a number-needed-to-treat of 4.3. The commentary situates these findings in the context of the large treatment gap for alcohol use disorder, the recent off-label prescribing surge for GLP-1 therapies in substance use disorders, and the critical unanswered questions regarding dose-response, post-treatment rebound, generalizability beyond patients with obesity, and real-world implementation.

Study Snapshot

Study Facts Table

What Researchers Actually Did

Klausen and colleagues conducted a double-blind, placebo-controlled, single-centre trial at a Danish site, randomly assigning 108 treatment-seeking adults who met criteria for both alcohol use disorder and comorbid obesity to either subcutaneous semaglutide or placebo over 26 weeks. Semaglutide was titrated from 0.25 mg/week up to a maximum of 2.4 mg/week, the dose approved for chronic weight management. The trial enrolled participants who were actively seeking treatment, distinguishing it from earlier semaglutide work in non-treatment-seeking individuals. The primary endpoint was reduction in the number of heavy drinking days assessed at the 26-week mark. The trial structure included weekly clinic visits and cognitive behavioural therapy, elements that reflect pragmatic clinical research conditions in a single-centre setting but may limit applicability to real-world primary care.

The commentary by Hendershot and Klein contextualizes these findings within the broader landscape of alcohol use disorder pharmacotherapy, the FDA’s recent qualification of the 2-level WHO risk drinking level reduction as a clinical trial endpoint, and the concerning trend of off-label GLP-1 prescribing for substance use disorders that has outpaced the supporting evidence base.

Key Findings: Primary Outcomes

• Semaglutide (up to 2.4 mg/week) significantly reduced the number of heavy drinking days at 26 weeks compared to placebo. Specific effect sizes, confidence intervals, and p-values for the primary endpoint are reported in the trial article (Klausen et al., Lancet 2026;407:1687–1698) and not reproduced numerically in this commentary.
• Semaglutide significantly increased the likelihood of achieving a 2-level reduction in WHO risk drinking level, the FDA-qualified clinical trial endpoint, with a number-needed-to-treat of 4.3.
• The commentary authors characterize the NNT of 4.3 as suggesting potentially greater efficacy relative to existing FDA-approved alcohol use disorder medications.

Key Findings: Secondary Outcomes

• Alcohol craving was reduced in the semaglutide group relative to placebo.
• Drinks per drinking day were lower in the semaglutide arm.
• Total alcohol consumption was reduced.
• Alcohol exposure biomarkers showed reductions consistent with self-reported consumption decreases.
• The trial identified individual variability in response: some participants showed little or no reduction in drinking, underscoring the absence of a uniform treatment effect.
• No subgroup analyses are described in the commentary. The trial population was 94% White, 49% female, and required comorbid obesity for enrollment, which are not subgroup findings but eligibility constraints that define the scope of generalizability.

Adverse Events and Safety Profile

The commentary characterizes the safety profile as favorable across the full semaglutide dose range (0.25–2.4 mg/week) over 26 weeks. Specific adverse event rates, dropout rates due to adverse events, and the nature of adverse events are not detailed in the commentary and must be reviewed in the primary trial publication (Klausen et al., Lancet 2026;407:1687–1698). The commentary notes that weight loss induced by semaglutide will constrain the alcohol use disorder population that can be treated safely, particularly as studies in participants without overweight or obesity have not yet been conducted.

Statistical Approach and Rigor

The primary analytic framework and statistical methods are reported in the full trial article rather than this commentary. What the commentary does convey is that the primary endpoint was pre-specified (heavy drinking days at 26 weeks), a 2-level WHO risk drinking level reduction was assessed as a secondary endpoint using the recently FDA-qualified criterion, and a number-needed-to-treat of 4.3 was calculated for that secondary endpoint. The trial was a single-centre study with 108 participants, a sample size that supports detection of moderate-to-large effect sizes but limits precision for subgroup analyses and may not capture lower-frequency adverse events. The commentary does not report intention-to-treat versus per-protocol analysis designations, attrition rates, or handling of missing data; readers must consult the primary publication for these details.

Clinical Takeaway

A 26-week, double-blind RCT found that semaglutide, titrated to 2.4 mg/week, reduced heavy drinking days, alcohol craving, and total consumption versus placebo in treatment-seeking adults with alcohol use disorder and comorbid obesity, with a number-needed-to-treat of 4.3 for a clinically meaningful WHO drinking risk level reduction. That NNT compares favorably on paper with currently approved medications. However, the trial enrolled a narrow population (obesity required, 94% White, single centre, weekly clinic visits, concurrent cognitive behavioural therapy), and whether these results translate to leaner patients, more diverse populations, or settings without intensive ancillary support is unknown. Off-label prescribing for alcohol use disorder is not yet supported by sufficient evidence. The correct clinical posture is watchful engagement: track the growing trial pipeline, avoid premature routine use, and recognize that this signal, though credible, requires replication in broader and more pragmatic settings before it changes prescribing practice.

Why This Matters Clinically

Alcohol use disorder affects hundreds of millions of people globally and carries a disease burden intersecting with cardiometabolic disease and alcohol-related liver disease. Fewer than 2% of Americans with alcohol use disorder receive pharmacotherapy, and the three medications with FDA approval — disulfiram, naltrexone, and acamprosate — are chronically underused. A medication already in broad clinical use, already accepted by patients and prescribers for metabolic indications, and with established supply chains could reshape treatment access in a way that purpose-built alcohol use disorder drugs have never achieved. Semaglutide’s documented benefits on liver and cardiometabolic outcomes also introduce the possibility of simultaneous treatment of comorbidities that co-occur at high rates in people with alcohol use disorder. The NNT of 4.3 reported here, if it replicates, would represent a meaningful therapeutic advance. That is a significant “if,” and clinicians should weight it accordingly.

CED Clinical Relevance

At CED Clinic, many patients using cannabis or other pharmacological approaches for symptom management also carry comorbid alcohol use patterns, cardiometabolic risk, or both. The intersection of GLP-1 pharmacology with substance use neurocircuitry is directly relevant to the integrative, evidence-grounded clinical framework we use. This trial provides preliminary evidence that a drug already in our pharmacological landscape may have utility for a second major substance use condition. For patients who are already candidates for semaglutide based on metabolic indications, the emerging alcohol use disorder data adds clinical texture to the risk-benefit discussion. For patients without obesity or overweight, no evidence yet supports this application. Prescribing semaglutide off-label for alcohol use disorder absent comorbid metabolic indication is not supported by the current evidence.

Fits What We Already Know

This trial extends prior evidence from a randomized clinical trial by Hendershot and colleagues (JAMA Psychiatry 2025;82:395–405), which showed that once-weekly low-dose semaglutide reduced alcohol use in non-treatment-seeking adults with alcohol use disorder who were not attempting reduction. The current trial advances that work by enrolling treatment-seeking participants, using the full weight-management dose range, and including concurrent behavioural therapy. Preclinical and early human evidence, cited by the commentary authors via Volkow and colleagues (Biological Psychiatry 2026) and Farokhnia and Leggio (JAMA Psychiatry 2026), has implicated central mechanisms involving GLP-1 receptor interactions with addiction neurocircuitry. The commentary also cites a prior consensus that GLP-1 receptor agonists remain “promising but unproven” for substance use disorders (Leggio et al., Nature Medicine 2023;29:2993–2995), a characterization that these new data partially — but not definitively — move past.

What This Study Teaches Us

A drug that targets metabolic pathways can simultaneously reduce alcohol craving and consumption, lending clinical plausibility to the GLP-1 receptor’s role in reward and impulse circuitry. The FDA-qualified endpoint of a 2-level WHO risk drinking level reduction provides a standardized anchor for comparing GLP-1 data against existing medications. An NNT of 4.3 — if replicated — would position semaglutide as competitive with, or superior to, naltrexone and acamprosate in terms of response rates for a clinically meaningful outcome. The existence of individual non-responders in this trial tells us that semaglutide is not a universal solution, and that identifying predictors of response will be essential for rational patient selection as this field matures.

What This Study Does Not Show

• It does not demonstrate efficacy in patients without obesity or overweight. Enrollment required comorbid obesity; extrapolation to leaner populations is not supported.
• It does not establish durability of benefit after treatment discontinuation. Whether gains persist or rebound post-treatment — as observed with metabolic endpoints in semaglutide trials — is unknown.
• It does not characterize the minimum effective dose. The titration reached 2.4 mg/week; lower doses may suffice or may not.
• It does not demonstrate generalizability beyond a predominantly White (94%), single-centre, Northern European sample with access to weekly clinical visits and cognitive behavioural therapy.
• It does not establish that semaglutide outperforms existing approved alcohol use disorder pharmacotherapies in a head-to-head comparison.
• It does not clarify whether observed reductions are driven by central reward mechanisms, peripheral satiety effects, nausea-related aversion, or a combination of these.
• It does not provide real-world effectiveness data in the absence of intensive ancillary support.

Fits the Broader Conversation

Eight incretin-class or metabolically related drugs are now under investigation for alcohol use disorder or alcohol-related liver disease, a number the commentary authors describe as extraordinary given that only three alcohol use disorder medications have received FDA approval in 75 years. Industry-sponsored trials have now entered the field alongside investigator-initiated studies, signaling the scale of investment and the realistic possibility of regulatory submissions. The FDA’s qualification of the 2-level WHO risk drinking level reduction as a clinical trial endpoint provides the field with a standardized, regulatorily accepted outcome measure that should facilitate cross-trial comparisons. This trial, positioned as one early brick in that edifice, moves the conversation from theoretical plausibility to provisional clinical evidence, while the commentary authors are appropriately careful to flag that an efficacy signal in a single centre trial is not equivalent to clinical effectiveness at scale.

What This Means for Families

Alcohol use disorder is one of the most common and least treated health conditions in the world. For families watching a loved one struggle, the idea that a medication already known for weight loss might also help reduce alcohol use is understandably exciting. This study showed that in a group of 108 people who were seeking treatment for both alcohol use disorder and obesity, a weekly injection of semaglutide reduced heavy drinking days, cravings, and overall alcohol intake more than a placebo did. That is a meaningful result. However, families should understand that this was a single study, conducted at one medical centre in Denmark, mostly in White participants, all of whom also had obesity. The study lasted 26 weeks, so what happens after stopping the medication is not known. The drug is not currently approved for alcohol use disorder. Hope is warranted; premature certainty is not.

What This Means for Clinicians

This trial provides the first full-dose (0.25–2.4 mg/week) semaglutide efficacy data in treatment-seeking adults with alcohol use disorder and comorbid obesity. The primary endpoint was met, multiple secondary endpoints were reduced, and the NNT of 4.3 for a 2-level WHO risk drinking level reduction — an FDA-qualified outcome — compares favorably with existing pharmacotherapy. The safety profile across the dose range was described as favorable. Key practice considerations: the trial required comorbid obesity for enrollment, so the evidence does not support use in patients without qualifying metabolic comorbidity. The trial infrastructure included weekly visits and concurrent cognitive behavioural therapy; whether the drug effect is robust independent of that support structure is untested. Dose-response data are not yet available to guide minimally effective dosing. Long-term durability post-discontinuation is unknown. This is a credible phase-shifting signal. It is not yet practice-changing evidence.

What This Means for the Cautious Reader

Hold the NNT of 4.3 loosely. It derives from a secondary endpoint, in a single-centre trial of 108 participants, in a demographic group (treatment-seeking, predominantly White, obese adults in Denmark with access to weekly clinical visits) that is far from representative of the global alcohol use disorder population. The commentary itself was co-authored by investigators with documented consulting relationships with pharmaceutical companies developing or marketing GLP-1 agents, though the acknowledged funding appears to be from federal and philanthropic sources. The trial was investigator-initiated, not industry-sponsored, which reduces one category of bias. Individual variability in response is acknowledged, and the mechanisms driving any benefit remain unresolved. Replicate, diversify the population, extend the follow-up, and study real-world contexts before this evidence drives formulary or prescribing decisions.

Why This Matters for Families

Alcohol use disorder remains one of the most stigmatized and undertreated conditions in medicine. Fewer than 2 out of every 100 people with alcohol use disorder in the United States receive any medication for it. A drug that is already widely used, already familiar to patients and doctors, and already covered by insurance for metabolic conditions could change that access equation dramatically if it receives regulatory approval for alcohol use disorder. For families, this research represents a credible step toward a future where their loved one’s doctor can offer a medication conversation as routinely as one about blood pressure or cholesterol. That future is not here yet, but this trial suggests it may be closer than it appeared a few years ago.

Why This Matters for Clinicians

The treatment gap in alcohol use disorder pharmacotherapy represents a genuine public health failure. Three approved medications, all chronically underused, leave the vast majority of affected patients without pharmacological support. Semaglutide’s pre-existing clinical infrastructure — prescriber familiarity, patient acceptance, pharmacy availability, and insurance coverage for metabolic indications — means that regulatory approval for alcohol use disorder would translate into accessible treatment at a scale no purpose-built alcohol use disorder drug has achieved. Additionally, semaglutide’s documented benefits on liver disease and cardiometabolic outcomes offer the prospect of simultaneously addressing comorbidities that are extremely prevalent in the alcohol use disorder population. Clinicians should monitor the forthcoming trial pipeline, including industry-sponsored programmes now underway, and engage in informed shared decision-making with patients who ask about this emerging application.

Why This Matters for Policy

The FDA’s recent qualification of the 2-level WHO risk drinking level reduction as a clinical trial endpoint creates a regulatory pathway that this trial has now meaningfully engaged. Eight incretin-class or related metabolic agents are currently under investigation for alcohol use disorder or alcohol-related liver disease, and industry-sponsored programmes are now active, signaling potential regulatory submissions on a years-long horizon. Coverage and access policy will need to anticipate a scenario in which semaglutide or related agents receive alcohol use disorder indications: existing GLP-1 reimbursement frameworks are anchored to metabolic criteria, and whether payers will extend those frameworks to behavioral health indications requires proactive planning. In low-income and middle-income countries, where access to any alcohol use disorder medication is extremely limited, a GLP-1 approval could either amplify or entrench disparities depending on pricing and access structures. Policy engagement now, before approval, will determine whether the public health potential of these drugs is realized equitably.

Quality Gate Alerts

• • Single-centre design: All participants were enrolled at one site in Denmark. Centre-specific practice effects, patient selection, and cultural context cannot be ruled out as contributors to the observed results.
  • Small sample size (N=108): Adequate for detecting moderate-to-large primary endpoint effects but insufficient for reliable subgroup analyses, lower-frequency adverse event detection, or nuanced dose-response characterization.

Physician-Led, Whole-Person Care

A doctor who takes the time to truly understand you.

Personal care that starts with listening and is guided by experience and ingenuity.

Health, Longevity, Wellness

One-on-One Cannabis Guidance

Metabolic Balance

Leave a Message
Metabolic Care
Medical Consulting
Cannabis Care