Clinical Insight | CED Clinic

Schacht and colleagues tested oral semaglutide against placebo in 50 treatment-seeking adults with moderate to severe alcohol use disorder. The primary laboratory craving endpoint was not significantly different, and drinks per day also did not significantly improve. Several secondary alcohol outcomes favored semaglutide, while fewer cannabis-use days appeared in an exploratory analysis. The clinically responsible reading is mixed and preliminary: this is randomized human evidence that supports larger confirmatory trials, not an established treatment recommendation.

What This Study Teaches Us

The trial provides a mixed rather than uniformly positive result. Semaglutide did not significantly improve the primary laboratory cue-induced craving endpoint or drinks per day, but it was associated with fewer heavy-drinking days, fewer drinks per drinking day, lower naturalistic craving, fewer alcohol-related consequences, and fewer cannabis-use days in exploratory analysis.

Why This Matters

GLP-1 receptor agonists are attracting substantial interest in addiction medicine, but enthusiasm can easily outrun the evidence. This trial matters because it enrolled treatment-seeking adults, used random assignment and placebo control, and reported both null and favorable outcomes rather than reducing the study to a single positive headline.

Study Snapshot

Clinical Bottom Line

Oral semaglutide did not improve the trial’s primary laboratory craving endpoint, but several clinically relevant secondary alcohol outcomes favored treatment. The findings justify larger confirmatory trials, not routine semaglutide prescribing for alcohol use disorder.

The Primary Endpoint Was Negative

At week 6, oral semaglutide did not significantly reduce laboratory cue-elicited alcohol craving compared with placebo.

That null primary result must remain central. Positive secondary outcomes do not retroactively convert the primary endpoint into a success.

Several Secondary Outcomes Favored Semaglutide

Semaglutide was associated with fewer heavy-drinking days, fewer drinks per drinking day, lower naturalistic alcohol craving, and a faster reduction in alcohol-related consequences.

More participants receiving semaglutide reduced their World Health Organization risk drinking level by at least one category.

The Cannabis Finding Was Exploratory

Cannabis-use days declined relative to placebo, with a reported regression coefficient of -1.434 and a 95% confidence interval from -2.568 to -0.301.

Cannabis use was not the primary focus of the trial. The result should be treated as a signal for future study, not evidence that semaglutide treats cannabis use disorder.

Why Treatment-Seeking Participants Matter

The trial enrolled adults seeking treatment for moderate to severe alcohol use disorder, a clinically different population from people taking GLP-1 medicines primarily for diabetes or weight management.

Even so, 50 participants and eight weeks are not enough to establish long-term effectiveness, safety, relapse prevention, or comparative value against approved alcohol-use-disorder treatments.

What the Trial Cannot Establish

The abstract does not establish an optimal dose, treatment duration, durability after discontinuation, or which patients are most likely to benefit.

It also does not support self-directed use or a general conclusion that GLP-1 receptor agonists reduce every addictive behavior.

What This Paper Does Not Show

The trial does not prove that semaglutide is an established treatment for alcohol use disorder or cannabis use disorder. It does not establish long-term benefit, relapse prevention, comparative effectiveness, or a safe self-treatment approach.

How This Fits With the Broader Clinical Conversation

The emerging GLP-1 addiction literature now includes randomized, observational, and mechanistic work. Each design answers a different question, and small positive trials still require larger replication.

For patients, the practical discussion should include established alcohol-use-disorder treatments, metabolic indications, gastrointestinal and other medication risks, concurrent substance use, and the reason semaglutide is being considered.

Dr. Caplan’s Take

I would describe this as encouraging but mixed evidence. A negative primary endpoint deserves the same attention as the favorable secondary outcomes.

The exploratory reduction in cannabis-use days is especially interesting for CED readers, but it is not a treatment claim. It is a reason to design a cannabis-focused trial with a prespecified endpoint and adequate power.

What a Careful Reader Should Take Away

A careful reader should see a small randomized trial with a null primary endpoint, several favorable secondary alcohol outcomes, and an exploratory cannabis signal that needs direct replication.

Evidence Interpretation Guide

Four distinctions that matter