Elicit: GLP-1 receptor agonists for efficacy and safety in adults with alcohol use disorder: a joint synthesis of consumption, alcohol-related harms, and adverse events
GLP-1 receptor agonists for efficacy and safety in adults with alcohol use disorder: a joint synthesis of consumption, alcohol-related harms, and adverse events
Research reportView only
Create alertChat
May 4, 2026
What is the evidence for the efficacy and safety of GLP-1 receptor agonists in reducing alcohol consumption and alcohol-related harms in adults with alcohol use disorder?
RCT evidence demonstrates that semaglutide significantly reduces alcohol consumption, craving, and objective drinking biomarkers in adults with co-occurring AUD and obesity, while large observational studies consistently show GLP-1 RA use is associated with substantially lower rates of alcohol-related hospitalizations and diagnoses, with gastrointestinal side effects as the main safety concern and no signal for pancreatitis or suicidality; however, evidence for efficacy in individuals with AUD who do not have obesity is currently insufficient, and one trial suggested potential harm in lean participants.
Abstract
Ten studies—four RCTs and six observational studies—examine the efficacy and safety of GLP-1 receptor agonists in adults with AUD. The strongest trial evidence comes from a 26-week RCT of semaglutide 2.4 mg weekly in adults with AUD and co-occurring obesity, which demonstrated significant reductions in heavy drinking days (−13.7 percentage points vs. placebo; Cohen’s d = 0.57), total alcohol consumption (−467.5 g/30 days), craving (PACS −3.3 points; p = 0.0024), and the objective biomarker PEth (p < 0.0001). A phase 2 trial of low-dose semaglutide over 9 weeks similarly showed significant reductions in alcohol self-administration and craving. An earlier exenatide RCT failed on its primary endpoint in the overall sample but found significant reductions in heavy drinking days and total intake in a pre-specified obese subgroup (BMI >30), with a paradoxical increase in heavy drinking among lean participants, indicating that obesity may be a critical moderator of treatment response. Six large observational studies consistently report that GLP-1 RA use is associated with 28–75% lower rates of alcohol-related hospitalizations, alcohol intoxication events, and AUD diagnoses across populations with T2D or obesity, with stronger associations in more severe AUD and reductions in hepatic decompensation in those with alcohol-associated liver disease. Gastrointestinal adverse events are the primary safety signal, occurring more frequently with GLP-1 RAs than placebo; no pancreatitis or gallbladder events were observed, and available data do not suggest increased suicidality risk. Overall, the evidence supports the efficacy of semaglutide in reducing alcohol consumption and craving in adults with AUD and obesity, while evidence for benefit in AUD populations without obesity remains insufficient.
Methods
We analyzed 10 sources from an initial pool of 7469, using 4 screening criteria. Each paper was reviewed for 17 key aspects that mattered most to the research question. More on methods
Records from Elicit search
n = 10000
Records from PubMed
n = 96
Abstract screening: Human study, GLP-1 RA intervention, Alcohol outcome, Empirical data
n = 7469
Full-text screening: Human study, GLP-1 RA intervention, Alcohol outcome, Empirical data, Adults with AUD, Comparator present, Sample size, Extractable outcome data
n = 61
Screened out (abstract)
n = 7408
Human study: n = 964
GLP-1 RA intervention: n = 931
Alcohol outcome: n = 5254
Empirical data: n = 257
Other / below screening threshold: n = 2
Papers included for extraction
n = 10
Screened out (fulltext)
n = 51
GLP-1 RA intervention: n = 1
Adults with AUD: n = 23
No full text: n = 27
Press enter or space to select a node.You can then use the arrow keys to move the node around. Press delete to remove it and escape to cancel.
Press enter or space to select an edge. You can then press delete to remove it or escape to cancel.
Paper search
Elicit Corpus
We performed a semantic search across over 138 million academic papers from the Elicit search engine, which includes all of Semantic Scholar and OpenAlex.
We ran these queries:
- “GLP-1 receptor agonists for alcohol use disorder efficacy and safety”
- “GLP-1 receptor agonist alcohol-related harms hepatic hospitalization adverse events mortality”
The searches returned 10,000 total results from Elicit.
PubMed Corpus
We performed a keyword search across the PubMed corpus.
We ran this query:
(semaglutide[tiab] OR liraglutide[tiab] OR exenatide[tiab] OR dulaglutide[tiab] OR tirzepatide[tiab] OR "GLP-1 receptor agonist"[tiab] OR "glucagon-like peptide-1 receptor agonist"[tiab]) AND ("alcohol use disorder"[tiab] OR "alcohol dependence"[tiab] OR "alcohol consumption"[tiab] OR "alcohol craving"[tiab] OR "alcohol drinking"[tiab] OR alcoholism[tiab] OR "heavy drinking"[tiab]) NOT (steatohepatitis[tiab] OR NAFLD[tiab] OR NASH[tiab] OR MASH[tiab] OR MASLD[tiab])
The search returned 96 total results from PubMed.
We retrieved 7469 papers most relevant to the query for screening.
Screening
Abstract screening
We screened in sources based on their abstracts that met these criteria:
- Human study: The study is conducted in humans. Exclude rodent, non-human primate, in vitro, cell line, and ex vivo studies.
- GLP-1 RA intervention: The study evaluates a GLP-1 receptor agonist (semaglutide, liraglutide, exenatide, dulaglutide, lixisenatide, albiglutide) or a dual/triple agonist with a GLP-1 component (tirzepatide, retatrutide, pemvidutide, cagrilintide+semaglutide). Exclude studies of non-incretin drugs without a GLP-1 RA arm.
- Alcohol outcome: The study reports at least one alcohol-related outcome: alcohol consumption (drinks, AUDIT, %DA, heavy drinking days), craving, AUD diagnosis or symptoms, alcohol-related hospitalization or ED visit, alcohol-related liver disease progression, or alcohol-related mortality. Studies on non-alcoholic fatty liver disease (NAFLD/NASH/MASH/MASLD) without alcohol outcomes are excluded.
- Empirical data: The article reports primary empirical data (RCT, cohort, case-control, register study, target trial emulation, pharmacovigilance, case series with n>=10). Exclude narrative reviews, editorials, commentaries, mechanistic perspectives, conference abstracts without numerical results, and protocol papers without results. Tag systematic reviews and meta-analyses for citation tracking but do not extract from them.
Papers that failed any strict criterion were automatically excluded. For the remaining papers, we considered all screening questions together and made a holistic judgement about whether to screen in each paper.
61 papers passed abstract screening and moved to full-text screening.
At abstract screening, the number of papers excluded for each primary reason was:
- Human study: n = 964
- GLP-1 RA intervention: n = 931
- Alcohol outcome: n = 5254
- Empirical data: n = 257
- Other / below screening threshold: n = 2
Full-text screening
We then screened papers based on their full text using these additional criteria:
Papers that failed any strict criterion were automatically excluded. For the remaining papers, we considered all screening questions together and made a holistic judgement about whether to include each paper in the final analysis.
10 papers passed full-text screening and moved to data extraction.
At full-text screening, the number of papers excluded for each primary reason was:
- GLP-1 RA intervention: n = 1
- Adults with AUD: n = 23
- No full text: n = 27
Data extraction
We asked a large language model to extract each data column below from each paper. We gave the model the extraction instructions shown below for each column.
- Study design:
Classify the study design.
- Country and setting:
Country/countries and care setting (specialist addiction service, primary care, hospital, claims database, registry).
- Recruitment / data window:
Years participants were enrolled or data extracted.
- Population:
How AUD was defined (DSM-5, ICD-10/11, ICD codes in claims, self-report). Mean age, % female, key comorbidities (T2D, obesity, ALD, HIV).
- N enrolled / exposed and N analyzed:
Per-arm sample sizes at randomization or exposure ascertainment, and at primary analysis.
- GLP-1 RA: drug, dose, duration:
Specific drug, dose, route, treatment duration.
- Comparator:
Placebo, active comparator, or alternative drug class.
- Drinking outcomes:
Drinks/day, drinks/week, heavy drinking days/month, %DA, total alcohol grams. Report change from baseline with effect size and 95% CI.
- Craving:
OCDS, PACS, AUQ, VAS, or other validated craving measure. Report effect size and 95% CI.
- Biomarkers:
PEth, CDT, GGT, ALT, AST. Report direction and magnitude of change.
- AUD-related harms:
Alcohol-related hospitalization, ED visit, ALD progression, cirrhosis incidence, FibroScan/MELD change, alcohol-related mortality. Report HR/RR/OR with 95% CI.
- Safety: any AE / SAE:
Number and proportion with any AE and SAE per arm.
- Safety: GI / pancreatitis / gallbladder:
GI AEs (nausea, vomiting, diarrhea), pancreatitis, gallbladder events.
- Safety: suicidality:
Suicidal ideation, behavior, or self-injury events per arm.
- Discontinuation:
Discontinuation rate overall and due to AEs.
- Risk of bias tool and judgment:
RoB2 for RCTs, ROBINS-I for non-randomized. Overall judgment (Low / Some concerns / High; or Low / Moderate / Serious / Critical).
- Funding and conflicts:
Funding source(s) and any disclosed conflicts of interest, especially industry sponsorship by Novo Nordisk, Eli Lilly, AstraZeneca.
Results
Characteristics of Included Studies
The review identified 10 studies examining GLP-1 receptor agonists (GLP-1 RAs) in adults with alcohol use disorder (AUD). These comprised four randomized controlled trials (RCTs) or secondary analyses thereof, and six observational studies using retrospective cohort, case-control, or target trial emulation designs. The studies span multiple countries and settings, with data windows ranging from 2005 to 2025.
Study
Full text retrieved?
Study Design
Country and Setting
Data Window
GLP-1 RA (Drug, Dose, Duration)
Comparator
Population (AUD Definition, N, Mean Age, % Female, Key Comorbidities)
Risk of Bias
Hendershot et al., 2025
Yes
RCT (phase 2)
United States; university research setting
Sep 2022 – Feb 2024
Semaglutide SC, 0.25 mg (wk 1–4), 0.5 mg (wk 5–8), 1.0 mg (wk 9); 9 weeks
Placebo
DSM-5 AUD; N=48; mean age 39.9 y; 71% female; obesity prevalent; diabetes excluded
Low (RoB2)
Klausen et al., 2022
Yes
RCT
Denmark; specialist addiction outpatient clinics
Aug 2017 – Oct 2019
Exenatide SC 2 mg once weekly; 26 weeks
Placebo
DSM-5/ICD-10 AUD; N=127 (62 exenatide, 65 placebo); mean age 52 y; 40% female; diabetes excluded
Low (RoB2)
Jensen et al., 2025
Yes
Secondary analysis of RCT
Not specified
Not specified
Exenatide extended-release SC once weekly; 26 weeks
Placebo
AUD + obesity (BMI ≥30); N=30 (18 analyzed at wk 26); mean age 53 y; 30% female
Some concerns (RoB2)
Klausen et al., 2026
Yes
RCT (single-centre)
Denmark; specialist outpatient mental health/addiction service
Jun 2023 – Feb 2025
Semaglutide (Wegovy) SC, titrated 0.25–2.4 mg weekly; 26 weeks
Placebo (saline SC)
DSM-5/ICD-10 AUD + obesity (BMI ≥30); N=108 (54/arm); mean age 52.3 y; 49% female; 85% severe AUD; T2D excluded
Low (RoB2)
Lähteenvuo et al., 2024
Yes
Register-based cohort (within-individual design)
Sweden; national registry
2006–2021, follow-up to Dec 2023
Exenatide, liraglutide, dulaglutide, semaglutide; dose/duration not specified
Non-use of GLP-1 agonists; also AUD medications
ICD-10 F10 AUD; N=227,886 (6,276 GLP-1 exposed); mean age 40.0 y; 36.5% female
Not formally assessed
Wang et al., 2024
Yes
Retrospective cohort
United States; TriNetX EHR platform
Dec 2017 – Dec 2022
Semaglutide (2.4 mg Wegovy or 0.5–1 mg Ozempic); 12 months
Other anti-obesity or anti-diabetes medications
ICD-10 F10 AUD; 83,825 obesity / 598,803 T2DM; mean age 51.2 y; 65.9% female
High (ROBINS-I)
Qeadan et al., 2024
Yes
Retrospective cohort
United States; Cerner Real-World Data (~136 health systems)
Jan 2014 – Sep 2022
GIP/GLP-1 RAs (various agents); dose/duration not specified
No GIP/GLP-1 RA prescription
ICD-coded AUD; N=817,309 AUD patients; T2D and obesity as stratification variables
High (ROBINS-I)
Abegaz et al., 2026
Yes
Nested case-control
United States; NIH All of Us Research Program
Jan 2005 – Feb 2025
Liraglutide, semaglutide, exenatide, dulaglutide; dose/duration not specified
No GLP-1 RA use
SNOMED/ICD-coded AUD; N=22,652 (AUD cohort); mean age ~55.1 y; ~42–46% female; T2D or obesity required
Not formally assessed; likely Moderate or higher per ROBINS-I
Rodriguez et al., 2025
Yes
Target trial emulation (retrospective cohort)
United States; 30 US health care systems (Truveta EHR)
2018–2024
Semaglutide, tirzepatide; dose/duration not specified
Other anti-diabetic medications, anti-obesity medications, or AUD medications
ICD-10 F10 AUD; N=40,260 across 4 trials; mean age 49–60 y; 27–61% female; T2D or obesity
Moderate (ROBINS-I)
Rashid et al., 2024
Yes
Retrospective cohort
United States; IBM MarketScan claims database
2013–2020
Semaglutide, albiglutide, liraglutide, dulaglutide, lixisenatide, exenatide; dose not specified
Routine care without GLP-1 RA
ICD-9/10 coded ALD + T2DM; age <65 y
Moderate (ROBINS-I)
toof
Pageof
All 10 studies had full texts retrieved. Three RCTs (Hendershot et al., 2025; Klausen et al., 2022; Klausen et al., 2026) and one RCT secondary analysis (Jensen et al., 2025) provide direct evidence on efficacy endpoints including drinking outcomes, craving, and biomarkers. The six observational studies contribute evidence primarily on alcohol-related harms (hospitalizations, intoxication events, AUD diagnosis, and liver outcomes). Notable heterogeneity exists across populations: the RCTs enrolled treatment-seeking (or non-treatment-seeking in Hendershot et al.) adults with clinically diagnosed AUD, while observational studies defined AUD or alcohol-related diagnoses through administrative codes in populations with co-occurring T2D and/or obesity. Treatment durations ranged from 9 weeks to 26 weeks in RCTs, whereas observational studies assessed exposure windows from 12 months to several years of registry follow-up.
Effects
Drinking Outcomes
Study
Outcome
Effect Estimate (GLP-1 RA vs. Comparator)
95% CI
p-value
Notes
Hendershot et al., 2025
Drinks per calendar day
beta = −0.27
−0.63 to 0.09
0.17
Not significant
Hendershot et al., 2025
Drinks per drinking day
beta = −0.41
−0.73 to −0.09
0.04
Significant reduction
Hendershot et al., 2025
Heavy drinking days (IRR)
IRR = 0.84
0.71 to 0.99
0.04
Significant reduction
Hendershot et al., 2025
Total alcohol grams (lab session)
beta = −0.48
−0.85 to −0.11
0.01
Significant reduction
Klausen et al., 2022
Heavy drinking days (overall)
No significant difference
—
—
Primary endpoint null
Klausen et al., 2022
Heavy drinking days (BMI >30 subgroup)
−23.6 percentage points
−44.4 to −2.7
0.034
Significant in obese subgroup
Klausen et al., 2022
Total alcohol intake (BMI >30 subgroup)
−1,205 g
−2,206 to −204
0.026
Significant in obese subgroup
Klausen et al., 2022
Heavy drinking days (BMI <25 subgroup)
+27.5 percentage points
4.7 to 50.2
0.024
Increased HDD in lean subgroup
Jensen et al., 2025
Heavy drinking days
6.0 pp difference
−7.4 to 19.4
0.40
Not significant (self-report)
Jensen et al., 2025
Total alcohol consumption
−42.0 g
−507.7 to 423.7
0.90
Not significant (self-report)
toof
Pageof
View 5 more rows
The strongest trial-level evidence for efficacy on drinking outcomes comes from the Klausen et al. (2026) RCT, which used the highest dose of semaglutide (up to 2.4 mg weekly) in a population with both AUD and obesity over 26 weeks. This trial demonstrated clinically meaningful reductions in heavy drinking days (d = 0.57), total alcohol grams consumed per 30 days (d = 0.54), and drinks per drinking day (d = 0.45). By contrast, the earlier exenatide trial (Klausen et al., 2022) failed to demonstrate efficacy on its primary endpoint in the overall sample, though a significant benefit emerged in the pre-specified subgroup with BMI >30. Jensen et al. (2025), a secondary analysis of the same exenatide trial restricted to participants with BMI ≥30 (n = 30), found no significant difference on self-reported drinking outcomes but did demonstrate a significant delayed reduction in the objective biomarker PEth at week 26, suggesting that self-report measures may underestimate treatment effects in this population. The Hendershot et al. (2025) trial used lower doses of semaglutide over only 9 weeks and found significant reductions in drinks per drinking day, heavy drinking days, and total alcohol grams in a controlled laboratory self-administration paradigm, though the reduction in drinks per calendar day did not reach significance.
Craving
Study
Measure
Effect Estimate
95% CI
p-value
Hendershot et al., 2025
PACS
beta = −0.39
−0.73 to −0.06
0.01
Klausen et al., 2022
PACS
No significant difference at week 26
—
0.980
Klausen et al., 2026
PACS
−3.3 points
−5.5 to −1.2
0.0024
toof
Pageof
Two of three RCTs reporting craving demonstrated significant reductions on the Penn Alcohol Craving Scale. The Klausen et al. (2026) trial found a 3.3-point greater reduction with semaglutide 2.4 mg versus placebo, and Hendershot et al. (2025) reported a significant craving reduction with low-dose semaglutide. The exenatide trial (Klausen et al., 2022) found no difference in craving at week 26. The observational studies did not assess craving.
Biomarkers
Study
Biomarker
Direction and Magnitude
95% CI
p-value
Jensen et al., 2025
PEth
−0.9 micromol/L (exenatide vs. placebo, Week 26)
−1.6 to −0.1
0.03
Klausen et al., 2026
PEth
−0.28 micromol/L treatment difference
−0.41 to −0.15
<0.0001
Klausen et al., 2026
GGT
−24.2 U/L treatment difference
−33.4 to −15.1
<0.0001
Klausen et al., 2026
ALT
−3.3 U/L treatment difference
−11.0 to 4.4
0.40
toof
Pageof
Objective biomarker data corroborate the self-reported drinking reductions. PEth, a direct alcohol metabolite, was significantly reduced in both the exenatide-in-obesity secondary analysis (Jensen et al., 2025) and the semaglutide 2.4 mg RCT (Klausen et al., 2026). The latter also showed a substantial reduction in GGT (−24.2 U/L; p < 0.0001), suggesting improved hepatic status, though ALT changes did not reach significance. No RCT reported CDT or AST as a formal outcome.
Alcohol-Related Harms
Study
Outcome
Effect Estimate
95% CI
Population
Lähteenvuo et al., 2024
AUD hospitalization (semaglutide)
aHR = 0.64
0.50–0.83
AUD, Sweden registry
Lähteenvuo et al., 2024
AUD hospitalization (liraglutide)
aHR = 0.72
0.57–0.92
AUD, Sweden registry
Wang et al., 2024
AUD incidence (obesity, semaglutide vs. other AOM)
HR = 0.50–0.56
See source
Obesity or T2DM
Qeadan et al., 2024
Alcohol intoxication (GIP/GLP-1 RA vs. none)
aIRR = 0.50
0.40–0.63
AUD, US EHR
Abegaz et al., 2026
AUD diagnosis (any GLP-1 RA)
OR = 0.26
0.20–0.34
T2D/obesity, US
Abegaz et al., 2026
AUD diagnosis (semaglutide)
OR = 0.15
0.07–0.34
T2D/obesity, US
Rodriguez et al., 2025
Alcohol-related hospitalization (ADM trial)
HR = 0.70
0.59–0.83
AUD + T2D
Rodriguez et al., 2025
Alcohol-related hospitalization (AOM trial)
HR = 0.59
0.48–0.74
AUD + obesity
Rodriguez et al., 2025
Alcohol-related hospitalization (MAUD-T2D trial)
HR = 0.36
0.29–0.46
AUD + T2D (severe)
Rodriguez et al., 2025
Alcohol-related hospitalization (MAUD-obesity trial)
HR = 0.32
0.23–0.43
AUD + obesity (severe)
toof
Pageof
View 2 more rows
All six observational studies reported consistent directional reductions in alcohol-related harms associated with GLP-1 RA use, with effect estimates ranging from an adjusted hazard ratio of 0.32 to 0.72 depending on population and comparator. The Swedish within-individual registry study (Lähteenvuo et al., 2024) found significant reductions in AUD-related hospitalizations with both semaglutide (aHR 0.64) and liraglutide (aHR 0.72), with no increased risk of suicide attempts. Rodriguez et al. (2025) found that the strongest associations emerged in populations with markers of more severe AUD who would otherwise receive AUD-specific medications, with hazard ratios as low as 0.32. Importantly, their negative control outcome (non-alcohol-related hospitalization) showed no significant difference in two of the four trial emulations, supporting the specificity of the alcohol-related signal. Rashid et al. (2024) demonstrated that GLP-1 RAs were associated with a 44% reduction in hepatic decompensation among patients with co-existing alcohol-associated liver disease and T2DM, representing a clinically distinct harm-reduction endpoint.
Safety
Study
Any AE (GLP-1 RA arm)
Any SAE (GLP-1 RA arm)
GI AEs (Nausea)
Pancreatitis
Suicidality
Discontinuation Overall
Discontinuation due to AEs
Hendershot et al., 2025
Not reported
Not reported
Not reported
Not reported
Not reported
Not reported
Not reported
Klausen et al., 2022
Nausea 37.1%, vomiting 22.6%
24.2% (exenatide) vs 18.5% (placebo)
Nausea 37.1% vs 15.4%
0 cases
1 suicide (exenatide, post-withdrawal)
54.3%
≥6.5% (injection site reactions)
Jensen et al., 2025
Not reported
Not reported
Not reported
Not reported
Not reported
~40%
Not reported
Klausen et al., 2026
Not reported (per-type only)
1/54 (1.9%)
More frequent with semaglutide
0 cases
0 (semaglutide) vs 1 (placebo, ideation)
18.5% overall
7.4% semaglutide vs 1.9% placebo
toof
Pageof
Safety data are available primarily from the RCTs. Gastrointestinal adverse events were the most frequently reported class, occurring more commonly in GLP-1 RA arms than placebo in both the exenatide trial (nausea 37.1% vs. 15.4%) and the semaglutide 2.4 mg trial. These were generally transient and mild to moderate. No cases of pancreatitis were reported in any trial, and no gallbladder events were documented.
Regarding suicidality, one patient in the exenatide trial committed suicide 7 weeks after withdrawal from the study. In the semaglutide 2.4 mg trial, no suicidal thoughts occurred in the semaglutide group compared with one case in the placebo group. The Swedish registry study found no statistically significant increase in suicide attempts with GLP-1 agonist use (aHR 0.55; 95% CI 0.23–1.30), in contrast to conventional AUD medications which were associated with increased risk.
Discontinuation rates varied substantially. The exenatide trial had a 54.3% overall dropout rate, while the semaglutide 2.4 mg trial achieved 81% completion. In the latter, AE-related discontinuation was higher with semaglutide (7.4%) than placebo (1.9%). Observational studies generally did not report discontinuation; one noted that approximately one-third of patients prescribed GLP-1 RAs fail to complete treatment, and another found 49–53% discontinuation rates for GLP-1 RA initiators compared with 81–82% for AUD medication initiators.
Synthesis
The apparent contradiction between the null primary result of the exenatide trial (Klausen et al., 2022) and the positive findings of the semaglutide trials can be explained by three intersecting factors: drug potency/dose, BMI-dependent response, and treatment duration.
First, regarding dose and pharmacology, semaglutide at 2.4 mg weekly achieves substantially greater GLP-1 receptor engagement than exenatide 2 mg weekly, and the Klausen et al. (2026) trial used the highest approved dose of semaglutide. Even the low-dose semaglutide protocol in Hendershot et al. (2025), reaching only 1.0 mg, showed significant effects on laboratory-measured alcohol self-administration. This dose-response pattern suggests that higher-potency GLP-1 RA exposure produces more robust reductions in alcohol consumption.
Second, body mass index appears to be a critical moderator. The exenatide trial showed a striking BMI-by-treatment interaction: participants with BMI >30 showed significant reductions in heavy drinking days (−23.6 percentage points), while those with BMI <25 showed a paradoxical increase (+27.5 percentage points). Both the Jensen et al. (2025) secondary analysis and the Klausen et al. (2026) trial restricted enrollment to individuals with obesity (BMI ≥30), and both demonstrated significant benefits. The mechanism underlying this BMI-dependent effect may involve differential central GLP-1 receptor sensitivity, metabolic co-regulation of reward pathways, or the greater tolerability of GLP-1 RA-induced appetite suppression in individuals with excess adiposity. Several observational studies similarly identified consistent protective associations in populations with obesity or T2D, populations where GLP-1 RAs are primarily prescribed.
Third, the timing of effects merits attention. Jensen et al. (2025) found that the objective biomarker PEth showed no significant group difference at weeks 4, 12, or 20, with significance emerging only at week 26, suggesting a delayed pharmacological effect on drinking behavior that may be missed in shorter trials such as the 9-week Hendershot et al. (2025) study. This delayed onset aligns with the neurobiological hypothesis that GLP-1 RAs progressively modulate dopaminergic reward circuitry, as supported by the fMRI and SPECT findings from Klausen et al. (2022) showing attenuated alcohol cue reactivity in the ventral striatum and reduced dopamine transporter availability.
The observational evidence on alcohol-related harms is remarkably consistent across all six studies, despite heterogeneity in study design, data source, country, comparator, and analytic approach. Effect estimates for alcohol-related hospitalizations and related outcomes cluster between 0.32 and 0.72, with stronger associations observed in populations with more severe AUD. The use of active comparators (other anti-diabetic medications, anti-obesity medications, or AUD medications) in several studies rather than simply non-use mitigates healthy-user bias, though residual confounding by indication remains a concern, particularly in studies that compared GLP-1 RA users to non-users. The within-individual design of Lähteenvuo et al. (2024) effectively eliminates time-invariant confounding, strengthening the inference that GLP-1 RA exposure periods are associated with reduced AUD-related hospitalizations within the same individual.
Taken together, the evidence supports the following context-specific conclusions: in adults with co-occurring AUD and obesity, semaglutide at doses of 2.4 mg weekly for 26 weeks produces clinically meaningful reductions in heavy drinking days (d ≈ 0.57), alcohol consumption, and craving. For individuals with AUD but without obesity, the evidence is insufficient to conclude benefit and one trial suggested potential harm in lean individuals. In populations with T2D or obesity prescribed GLP-1 RAs for metabolic indications, there appears to be a substantial and consistent reduction in alcohol-related hospitalizations and diagnoses, though this evidence is observational and subject to confounding. GLP-1 RAs appear to be well tolerated in AUD populations, with GI adverse events as the primary safety signal, no observed pancreatitis, and no evidence of increased suicidality.
References
Download BIBDownload RISDownload TXT
Report
Version
May 4, 2026 3:39 AM
Status
Gather sources
7469 sources found
Details
Abstract screening
61 sources included
(report limited to 40)
Details
Full-text screening
10 sources included
Details
Extract data
170 data points extracted
Details
Generate report
Save PDF
BIBLaTeX, ZoteroRISZotero, MendeleyTXTAPA BibliographyPDFPDFDOCXMicrosoft Word
Chat
Got some follow-up questions?
Sign up or sign in to chat with this report.
Back
Exenatide once weekly for alcohol use disorder investigated in a randomized, placebo-controlled clinical trial
M. K. Klausen, M. E. Jensen, M. Møller, Nina le Dous, A. M. B. Jensen, Victoria A Zeeman, Claas-Frederik Johannsen, Alycia M. Lee, G. Thomsen, J. Macoveanu, P. Fisher, M. Gillum, N. Jørgensen, M. L. Bergmann, H. Enghusen Poulsen, U. Becker, J. Holst, H. Benveniste, N. Volkow, S. Vollstädt-Klein, K. Miskowiak, C. Ekstrøm, G. Knudsen, Tina Visboll, A. Fink-Jensen
JCI Insight·
2022·
147 citations
SourceDOI
Plain textPDF
Searching for PDF
Unable to find PDF from source
Back
Study design
RCT
Country and setting
Country: Denmark; Setting: Specialist addiction service (outpatient clinics)
Recruitment / data window
August 7, 2017, to October 1, 2019
Population
- Population: Treatment-seeking AUD patients - Mean age: 52 years - % female: 40% - Key comorbidities: Not mentioned (exclusions included severe mental disorder, other drug use disorder, diabetes, pancreatitis) - AUD definition: DSM-5 and ICD-10
N enrolled / exposed and N analyzed
- N enrolled/exposed: Exenatide: 62, Placebo: 65 - N analyzed: Exenatide: Not explicitly stated, Placebo: Not explicitly stated
GLP-1 RA: drug, dose, duration
Exenatide, 2 mg, subcutaneously, once weekly for 26 weeks
Comparator
Placebo
Drinking outcomes
- Overall: No significant reduction in heavy drinking days or total alcohol intake compared to placebo. - Obese patients (BMI > 30 kg/m2): Reduced heavy drinking days by 23.6 percentage points (95% CI, -44.4 to -2.7, P = 0.034); reduced total alcohol intake by 1,205 g (95% CI, -2,206 to -204, P = 0.026). - Non-obese patients (BMI < 25 kg/m2): Increased heavy drinking days by 27.5 percentage points (95% CI, 4.7 to 50.2, P = 0.024).
Craving
- Measure: Penn Alcohol Craving Scale score - Baseline: Significant difference between healthy controls and patients (P < 0.001) - Week 26: No significant difference between exenatide and placebo groups (P = 0.980) - Effect size and 95% CI: Not explicitly provided
Biomarkers
- PEth: Not mentioned - CDT: Not mentioned - GGT: Not mentioned - ALT/AST: Not mentioned - HbA1c: Decreased by 1.6 mmol/mol (95% CI, -2.8 to -0.4, P = 0.011) in the exenatide group - Urinary oxidative stress markers (-8-oxoGuo and 8-oxodG): Increased in the exenatide group
AUD-related harms
Not mentioned (the paper does not provide specific data on alcohol-related harms such as hospitalization, ED visits, ALD progression, cirrhosis incidence, FibroScan/MELD change, or alcohol-related mortality with HR/RR/OR and 95% CI)
Safety: any AE / SAE
- Adverse Events (AEs): - Exenatide: Nausea (37.1%), decreased appetite (24.2%), vomiting (22.6%), weight loss (67.7%), fatigue (12.9%), injection site reaction (41.0%) - Placebo: Nausea (15.4%), decreased appetite (9.2%), vomiting (7.7%), weight loss (40.0%), fatigue (4.6%), injection site reaction (0.0%) - Serious Adverse Events (SAEs): - Exenatide: 24.2% - Placebo: 18.5% - Notable SAEs: - Exenatide: One patient committed suicide 7 weeks after withdrawal. - Placebo: One patient died due to alcohol withdrawal symptoms.
Safety: GI / pancreatitis / gallbladder
- GI AEs: Nausea (37.1% vs. 15.4%), vomiting (22.6% vs. 7.7%), decreased appetite (24.2% vs. 9.2%) - Pancreatitis: No cases reported; no elevation of pancreas enzymes above upper limits - Gallbladder events: Not mentioned
Safety: suicidality
One patient in the exenatide group committed suicide 7 weeks after withdrawal from the trial. No other specific suicidal ideation, behavior, or self-injury events are detailed per arm.
Discontinuation
- Overall discontinuation rate: 54.3% - Discontinuation due to AEs: At least 6.5% due to injection site reactions; exact number for other AEs not specified
Risk of bias tool and judgment
Low
Funding and conflicts
Funding sources: Novavi Foundation; Research Foundation, Mental Health Services, Capital Region of Denmark; Research Foundation, Capital Region of Denmark; Ivan Nielsen Foundation; A.P. Moeller Foundation; Augustinus Foundation; Woerzner Foundation; Grosserer L.F. Foghts Foundation; Hartmann Foundation; Aase and Ejnar Danielsen Foundation; P.A. Messerschmidt and Wife Foundation; and Lundbeck Foundation. Conflicts of interest: A. Fink-Jensen received an unrestricted research grant from Novo Nordisk A/S for a different study. Tina Visboll has served on scientific advisory panels for, been part of speaker's bureaus for, served as a consultant to, and/or received research support from Amgen, AstraZeneca, Boehringer Ingelheim, Eli Lilly, Gilead, Mundipharma, MSD/ Merck, Novo Nordisk, and Sun Pharmaceutical Industries.
Background Alcohol use disorder (AUD) is a chronic, relapsing brain disorder that accounts for 5% of deaths annually, and there is an urgent need to develop new targets for therapeutic intervention. The glucagon-like peptide-1 (GLP-1) receptor agonist exenatide reduces alcohol consumption in rodents and nonhuman primates, but its efficacy in patients with AUD is unknown. Methods In a randomized, double-blinded, placebo-controlled clinical trial, treatment-seeking AUD patients were assigned to receive exenatide (2 mg subcutaneously) or placebo once weekly for 26 weeks, in addition to standard cognitive-behavioral therapy. The primary outcome was reduction in number of heavy drinking days. A subgroup also completed functional MRI (fMRI) and single-photon emission CT (SPECT) brain scans. Results A total of 127 patients were enrolled. Our data revealed that although exenatide did not significantly reduce the number of heavy drinking days compared with placebo, it significantly attenuated fMRI alcohol cue reactivity in the ventral striatum and septal area, which are crucial brain areas for drug reward and addiction. In addition, dopamine transporter availability was lower in the exenatide group compared with the placebo group. Exploratory analyses revealed that exenatide significantly reduced heavy drinking days and total alcohol intake in a subgroup of obese patients (BMI > 30 kg/m2). Adverse events were mainly gastrointestinal. Conclusion This randomized controlled trial on the effects of a GLP-1 receptor agonist in AUD patients provides new important knowledge on the effects of GLP-1 receptor agonists as a novel treatment target in addiction. Trial registration EudraCT: 2016-003343-11. ClinicalTrials.gov (NCT03232112). Funding Novavi Foundation; Research Foundation, Mental Health Services, Capital Region of Denmark; Research Foundation, Capital Region of Denmark; Ivan Nielsen Foundation; A.P. Moeller Foundation; Augustinus Foundation; Woerzner Foundation; Grosserer L.F. Foghts Foundation; Hartmann Foundation; Aase and Ejnar Danielsen Foundation; P.A. Messerschmidt and Wife Foundation; and Lundbeck Foundation.
Introduction
Alcohol use disorder (AUD) is an essential contributor to the burden of global disease (1). In Denmark, the cumulative all-cause 10-year mortality risk is almost 30% after a first-time hospital contact due to an alcohol problem (2). Only 3 medications are approved by the US Food and Drug Administration (FDA) to treat AUD: disulfiram, naltrexone, and acamprosate (3). About 40% of patients treated for AUD relapse within the first 3 years (4), and new targets for therapeutic interventions are urgently needed for this devastating chronic disease (1,3).
The endogenous glucagon-like peptide-1 (GLP-1) is a 30-amino acid peptide hormone produced in the intestinal L cells in response to food intake (5), as well as in the nucleus tractus solitarius of the medulla oblongata (6). GLP-1 stimulates insulin secretion, inhibits glucagon secretion, and, notably, dampens appetite and food intake (5). GLP-1 receptor agonists are approved by the European Medicines Agency (EMA) and FDA to treat type 2 diabetes and obesity (7). Since drugs of abuse and alcohol activate the same reward system that underlies food reward (8), it is conceivable that appetite-regulating peptides such as GLP-1 target areas associated with reward and addiction. In support of this hypothesis, several studies have reported expression of GLP-1 receptors in brain areas associated with reward and addiction (6,(9)(10)(11)(12)(13)(14)(15)(16). Furthermore, treatment with GLP-1 receptor agonists reduce alcohol intake and decrease relapse-like alcohol drinking in nonhuman primates (17) and rodents (18). In humans, a recent study reported that the GLP-1 receptor 168Ser allele variant was associated with increased alcohol intake in humans (19). However, the effects of a GLP-1 receptor agonist on alcohol consumption in humans remain unknown. To this end, we performed a randomized, placebo-controlled clinical trial lasting 26 weeks plus a long-term 6-month follow-up to evaluate the efficacy of the once-weekly GLP-1 receptor agonist exenatide (Bydureon) at a dose of 2 mg in patients diagnosed with AUD according to the Diagnostic and Statistical Manual of Mental Disorders (DSM-5). In total, 127 treatment-seeking AUD patients, who had a minimum of 5 heavy drinking days, i.e., 60/48 g of alcohol or more per day (men/women) in the past 30 days, were included. Since the pharmacodynamics and pharmacokinetics of a GLP-1 receptor agonist in patients with AUD have not been investigated, we chose a dosing regimen consistent with established tolerability and efficacy in treatment of type 2 diabetes, i.e., exenatide, 2 mg subcutaneously once weekly. Importantly, exenatide crosses the bloodbrain barrier (20), and a similar dosing regimen, i.e., 2 mg subcutaneously once weekly, has recently shown efficacy in other neuropsychiatric disorders, including nicotine dependence (21) and Parkinson's disease (20), suggesting a central engagement, possibly mediated, at least in part, by dopamine signaling (22).
The primary endpoint was reduction in heavy drinking days, recorded with the Time-Line Follow Back method (23). A subgroup of the patients had a functional MRI (fMRI) scan and a single-photon emission CT (SPECT) scan performed at baseline and at week 26. Using the fMRI technique, we investigated whether exenatide once weekly would reduce alcohol cue reactivity in brain areas involved in drug reward and addiction, and in top-down regulation of impulsivity (24), as preclinical and clinical evidence suggests that GLP-1 receptor stimulation may be associated with improved cognitive impairment (25). By use of the SPECT scan, we measured the availability of the striatal dopamine transporter (DAT), a key modulator of extracellular dopamine. Dopamine plays a pivotal role in the neurobiological underpinnings of reward (26), and a large body of evidence suggests that brain dopamine homeostasis changes following chronic alcohol intake (27).
Results
Characteristics of the patients. From August 7, 2017, to October 1, 2019, 152 patients were screened for eligibility, and 127 patients were enrolled; 62 were randomly assigned to the exenatide group, and 65 were assigned to the placebo group (Figure 1 ). Overall, the 2 treatment groups were balanced with respect to baseline characteristics (Table 1 ). All patients were White, with a mean age of 52 years. The majority of the patients were men (60%). On average, they had 17 heavy drinking days and an overall alcohol intake of JCI Insight 2022;7 (19):e159863 https://doi.org/10.1172/jci.insight.159863 2,400 g of pure alcohol over the last month, and 80% fulfilled the criteria for severe AUD, i.e., more than 5 symptoms, according to DSM-5 (see baseline characteristics and flowchart for the patients included in the brain imaging substudy in Supplemental Appendices 1 and 2; supplemental material available online with this article; https://doi.org/10.1172/jci.insight.159863DS1). Of the 127 patients included, a total of 58 patients completed the trial, i.e., participated in the last follow-up after 26 weeks of treatment; 25 patients finished prematurely, i.e., participated in a final examination before 26 weeks of treatment. Fifty-five patients participated in the long-term 6-month follow-up visit (Supplemental Figure 1 ), with the last visit held on the October 10, 2020. The mean (SD) number of injections was 22.6 (2.2) in the exenatide group and 22.1 (2.8) in the placebo group (Supplemental Table 1 ). There was no difference (P = 0.46) between the 2 groups in time to trial discontinuation (Figure 2 ). In addition, 25 healthy controls matched for sex, age, and educational status of the included patients were recruited for the fMRI substudy.
Efficacy. For both groups, the number of heavy drinking days (Table 2 and Figure 3 ) and total alcohol intake (Table 2 ) were strongly reduced, but there were no significant differences between the 2 groups. The exenatide group had a reduction in BMI of 0.95 (95% CI, -1.6 to -0.3, P = 0.006), glycated hemoglobin (HbA1c) of 1.6 mmol/mol (95% CI, -2.8 to -0.4, P = 0.011), and a worsening in Drug Use Disorders Identification Test (DUDIT) score of 0.96 points (95% CI, 0.7 to 1.3, P < 0.001) relative to the placebo group (Table 2 ). There were no group differences in FGF-21, phosphatidylethanol, or bone markers (Table 2 ); life quality measurements, i.e., 36-Item Short Form Health Survey (SF-36) (Supplemental Table 3 ) and Symptom Checklist-92 (SCL-92) (Supplemental Table 4 ); or cognition i.e., Screen for Cognitive Impairment in Psychiatry test (SCIP) (Supplemental Table 5 ). Exenatide once weekly increased urine oxidative stress parameters -8-oxoGuo of 0.24 nmol/mmol creatinine (95% CI, 0.04 to 0.44, P = 0.022) and 8-oxodG of 0.43 nmol/mmol creatinine (95% CI, 0.15 to 0.72, P = 0.003) -relative to placebo (Supplemental Table 2 ). In the exenatide group, the plasma level of exenatide was 45.6 pmol/L (95% CI, 16.5 to 74.7, P = 0.003), and the overall anti-exenatide antibody binding was 16.1% (95% CI, 6.9 to 25.3, P = 0.002) relative to the placebo group (Supplemental Table 2 ).
Exploratory analyses. Exenatide once weekly did not reduce the number of heavy drinking days in the prespecified subgroup analyses (baseline heavy drinking days, severity of DSM-5 criteria, and geography) (Supplemental Table 6 ). However, an exploratory subgroup analysis (Supplemental Table 7 ) including BMI subgroups revealed that in obese patients with a BMI greater than 30 kg/m 2 (n = 30), exenatide reduced heavy drinking days by 23.6 percentage points (95% CI, -44.4 to -2.7, P = 0.034) (Figure 4 ) and reduced total alcohol intake per 30 days by 1,205 g (95% CI, -2,206 to -204, P = 0.026) relative to placebo (Figure 5 ). In patients with a BMI less than 25 kg/m 2 (n = 52), treatment with exenatide increased number of heavy drinking days by 27.5 percentage points (95% CI, 4.7 to 50.2, P = 0.024) relative to the placebo group. However, in this subgroup (BMI < 25 kg/m 2 ) the total alcohol intake did not differ between treatment groups. Other exploratory post hoc subgroup analyses were performed to investigate whether some subgroups responded differently from others on the intervention. However, no significant differences were observed with respect to sex, baseline craving (Penn Alcohol Craving Scale score), baseline Alcohol Use Disorders Identification Test (AUDIT) score, baseline number of days without alcohol, baseline total alcohol consumption, fMRI subgroup (n = 22), and SPECT subgroup (n = 16).
Besides the exploratory subgroup analyses, we also looked at the reduction in WHO risk drinking levels (28). Both groups reduced their risk drinking levels, but there was no significant difference between the 2 groups (Supplemental Table 6 ).
To explore whether there was a correlation between change in HbA1c and change in heavy drinking days, the Pearson correlation coefficient was computed in the imputed data set (n = 127) to assess linear relationship. Here, we found a weak negative correlation between the 2 variables [r(12755) = -0.27, P = 0.001]. We also found a weak negative correlation between changes in HbA1c and changes in total alcohol intake [r(12755) = -0.36, P = 0.001]. Six-month long-term follow-up. There was no difference between the 2 groups at the 6-month follow-up after exenatide or placebo discontinuation (Table 2 and Supplemental Table 8 ), except for a higher AUDIT score (5.1 points; 95% CI, 0.9 to 9.3, P = 0.02) in the original exenatide group (adjusted from the end of treatment) compared with the placebo group.
fMRI alcohol cue reactivity. The predefined region of interest (ROI) masks were acquired from WFU Pick-Atlas (29). The analyses revealed a statistically significant interaction between treatment and time on the fMRI response in all 3 ROIs: ventral striatum [F(1,31) = 4.744, P = 0.037, partial η 2 = 0.133], dorsal striatum [F(1,31) = 6.124, P = 0.019, partial η 2 = 0.165], and putamen [F(1,31) = 4.730, P = 0.037, partial η 2 = 0.132], indicating reduced cue reactivity after 26 weeks of treatment with exenatide compared with placebo. The ROI analysis in the caudate and nucleus accumbens did not reveal any significant effects (Figure 6A ). At week 26, the cue-induced activity in ventral striatum was significantly lower in the exenatide group compared with placebo (mean difference [M] = -0.176, SEM = 0.075, P = 0.025). However, in the dorsal striatum (M = -0.142, SEM = 0.076, P = 0.073) and in the putamen (M = -0.123, SEM = 0.084, P = 0.153) no significant differences were observed. At baseline, cue-induced activity did not differ between the treatment groups. Within the exenatide group, cue-induced activity was significantly reduced from baseline to week 26 in ventral striatum (M = -0.254, SEM = 0.116, P = 0.044) and in dorsal striatum (M = -0.351, SEM = 0.156, P = 0.039), but not in putamen (M = -0.405, SEM = 0.202, P = 0.063). Within the placebo group, we found no statistically significant differences (Figure 6B ).
At baseline, the exploratory whole-brain analysis showed no significant difference in cue reactivity between the placebo group and the exenatide group. When cue reactivity in all patients was compared with that in healthy controls, significant differences were found in the left superior and middle frontal gyrus, caudate, and insula (P = 0.001). However, at the week 26 rescan, these differences were no longer significant. At the week 26 assessment, cue-induced activation was significantly reduced in the exenatide group compared with the placebo group in the following brain areas (Supplemental Appendix 1: Supplemental Table 13 ): left caudate and septal area (Figure 7A ) and right middle frontal gyrus (Figure 7B ). There was no significant change in cue
Clinical measures
Body weight, kg -0.5 (-1.8 to 0.9) -2.9 (-4.3 to -1.5) -2.4 (-5.0 to 0. 8 for full details of the subscales. G Linear mixed model, adjusted for intake of benzodiazepine at the time of the assessment. H See Supplemental Table 9 for full details of the subscales. I A censored regression model was used for this analysis due to zeroinflated values; scores range from 0 to 44, with a higher score indicating substance use disorder. J See Supplemental Table 10 for full details of the subscales.
C L I N I C A L M E D I C I N E
JCI Insight 2022;7(19):e159863 https://doi.org/10.1172/jci.insight.159863 reactivity in the placebo group at the rescan, but the exenatide group showed a significant reduction in cue-induced activation in the temporal lobe, hippocampus, and parahippocampus (rescans per protocol: Supplemental Appendix 1: Supplemental Table 14 , Supplemental Figure 5 ; rescans per protocol including premature rescans: Supplemental Appendix 1: Supplemental Table 15 , Supplemental Figure 6 ). Subjective craving results: fMRI Subjective craving results. The analysis showed a significant difference at baseline between the healthy controls and the patients (P < 0.001; mean ± SD: healthy controls, 8.8 ± 15.96; placebo group, 33.5 ± 26.9; exenatide group, 30.6 ± 28.6). At the week 26 follow-up, this was no longer significant (P = 0.50; mean ± SD: healthy controls, 8.8 ± 15.96; placebo group, 13.6 ± 12.0; exenatide group, 14.8 ± 23.07), and there were no significant differences between the exenatide and the placebo group (P = 0.980).
fMRI spatial working memory. The voxel-wise analysis showed a significant reduction in the exenatide group at the week 26 rescan compared with placebo in response to the 2-back > 1-back task in 2 clusters in the right frontal pole and right superior frontal gyrus, within the dorsolateral prefrontal cortex ROI (Figure 8 and Supplemental Appendix 1: Supplemental Table 16 ). The additional right dorsolateral prefrontal cortex ROI analysis showed no significant change in the exenatide group at week 26 Safety. Gastrointestinal (GI) symptoms, body weight loss, fatigue, and injection site reactions were the most common adverse events reported, and the incidence was higher in the exenatide compared with the placebo group (nausea, 37.1% vs. 15.4%; decreased appetite, 24.2% vs. 9.2%; vomiting, 22.6% vs. 7.7%; overall weight loss, 67.7% vs. 40.0%; fatigue, 12.9% vs. 4.6%; injection site reaction, 41.0% vs. 0.0%) (Table 3 ). The GI side effects reported lasted until the first 5 weeks of treatment, and the weight loss continued throughout the trial. The injection site reactions were typically small nodules of 1-2 cm, hard, mobile, skin-colored, and were reabsorbed within 6 weeks, leaving no scar. Serious adverse events were reported almost equally between the 2 groups (exenatide 24.2% vs. placebo 18.5%), and there were no cases of acute pancreatitis or elevation of pancreas enzymes above upper limits. One patient in the exenatide treatment group committed suicide 7 weeks after withdrawal from the trial. One patient in the placebo group was found dead after being hospitalized 3 times in one week for alcohol withdrawal symptoms.
Discussion
To our knowledge, this is the first randomized controlled trial (RCT) investigating the effects of exenatide in AUD patients. Treatment with exenatide once weekly was not superior to placebo in reducing the number of heavy drinking days in the prespecified analysis. The negative results could reflect the characteristics of the AUD patients included in our RCT. Data from preclinical trials showed that high-alcohol-consuming animals decreased their alcohol intake significantly more than low-alcohol-consuming animals when treated with a GLP-1 receptor agonist (23,31). In the present trial, 80% of the patients fulfilled the DSM-5 criteria for severe AUD. However, their severity profile, based on baseline alcohol intake and heavy drinking days (Table 1 ), was less severe than those observed in other AUD pharmacotherapy trials (32,33). Another explanation could be that the potent placebo response could have masked a possible beneficial effect of exenatide (Figures 3 and 4 ). The observed potent placebo response could be due to the standardized cognitive-behavioral therapy against AUD (34) offered to all participants in the study, but it could also be due to the lesser severity profile of the AUD patients included, which is typically linked to a higher placebo response (35). Large placebo responses are also reported in other clinical AUD trials and shown to be negatively correlated with the treatment intervention effect sizes (36).
Predefined fMRI brain ROI analysis found a reduced alcohol cue reactivity in the exenatide group compared with the placebo group in the ventral striatum, a region that plays a pivotal role in addiction and relapse (Figure 6 ). This finding is important because it implies that AUD subjects treated with exenatide lose the incentive salience of alcohol-associated cues. The exenatide-induced reduction in cue reactivity in the septal area (37) observed in the whole-brain analysis (Figure 7A ) is particularly intriguing as this is an area connected to reward (15), and a brain area where GLP-1 receptors are highly expressed (6). These findings are in accordance with a central effect of exenatide as mentioned in the Introduction. Future fMRI studies investigating the effects of GLP-1 receptor agonists on alcohol cue-induced activation should include the septal area as a region of interest.
Impairments in cognitive processes related to executive function in AUD patients (38) may negatively influence clinical outcomes owing to deficits in self-regulation (39). In the fMRI spatial working memory test, we found reduced cue reactivity in the dorsal prefrontal cortex in the exenatide group compared with the placebo group, possibly indicating a reduced effort to maintain cognitive performance (40).
The SPECT substudy revealed no significant differences in DAT availability at baseline between the AUD patients and healthy controls, which is in accordance with some earlier findings (41), but in discordance with others (42). After 26 weeks of treatment, the analysis revealed a significant reduction of DAT in the striatum, caudate, and putamen in the exenatide group compared with placebo, which might compensate for the decreased dopamine activity previously reported in AUD patients (43). Notably, this effect is most likely not acutely induced, since no change in DAT availability was observed after acute treatment with exenatide in healthy volunteers (44). Even though the results from the exploratory post hoc BMI subgroup analysis are preliminary, we think they are of substantial interest because overlapping dysfunctional brain circuits are observed in individuals with obesity or addiction (8), and deranged GLP-1 signaling is also reported in obese individuals (45). In addition, an fMRI study in obese versus lean individuals showed that exenatide infusions "normalized" the brain response to a food paradigm in obese patients with a BMI greater than 30 kg/m 2 compared with lean individuals (46). Moreover, several GLP-1 receptor agonists have recently been approved to treat obesity (BMI > 30 kg/m 2 ), and other compounds are under development (7). The reason why the number of heavy drinking days was increased in the subgroup of exenatide-treated patients with a BMI less than 25 kg/m 2 compared with placebo-treated patients could be that those lean individuals treated with exenatide experienced a larger decrease in blood sugar (47), and this might be associated with increased alcohol craving (48).
The significant increase in urinary oxidative stress markers in the exenatide group was previously reported in type 2 diabetes patients treated with exenatide ( 49), but the clinical significance of rising levels of urinary stress parameters 8-oxoGuo and 8-oxodG is currently unknown (50). Notably, increased urinary oxidative stress parameters in patients with type 2 diabetes are associated with increased mortality risk (51), and the clinical impact of these biomarkers should be further investigated.
GLP-1 receptor agonists have shown beneficial skeletal effects in rodents (52). However, in the present trial, no differences in bone turnover markers were observed between groups, indicating that bone-related adverse effects are not of concern in this patient population.
Both the exenatide group and the placebo group exhibited an overall reduction in DUDIT score after 26 weeks of treatment. However, the exenatide group had a significantly higher DUDIT score compared with placebo after 26 weeks of treatment (Table 2 ). An exclusion criterion was a diagnosis of any active substance use disorder (SUD) except for nicotine. Men with a DUDIT score greater than 6 and women with a DUDIT score greater than 2 were screened according to International Classification of Diseases, Tenth Revision (ICD-10), SUD criteria and, if diagnosed with SUD, excluded from the trial. Only 4 of the 25 included participants with a positive DUDIT score (range between 1 and 22 points) finished per protocol. This is essential information for a follow-up study, where it may be relevant to exclude all individuals with a positive baseline DUDIT score to increase study compliance.
The previously reported safety profile of exenatide once weekly is consistent with the present safety data. Our most significant safety concern was the risk of pancreatitis in patients with AUD (3) combined with the associated risk of exenatide treatment (53,54). Importantly, none of the patients experienced a rise in blood amylase above upper limits or developed pancreatitis. Surprisingly, the injection site reactions to exenatide were a bigger problem for the patients, due to unexpected concerns from their relatives, who might have been unaware of their AUD diagnosis. This led to a 6.5% withdrawal rate specifically due to injection site reactions in our AUD trial compared with only 0.5% in exenatide-treated patients with type 2 diabetes (55). The GI side effects, which are well recognized but typically transient (56), were in the exenatide group (44.1%) higher than reported in diabetes trials (57,58). Also, 23.6% of placebo-treated patients experienced GI side effects, indicating that this group of patients may have a GI vulnerability (59). Only a single RCT has investigated the effects of pretreatment with antiemetics, reporting a significant reduction in nausea and vomiting in exenatide-treated healthy subjects (60).
Large dropout rates are often observed in AUD intervention trials (61), and the present study -with 54.3% dropout -is no exception. Although our sensitivity analysis (Supplemental Table 9 ) confirmed the robustness of the results even with imputations of missing data, the present dropout rate (69 of 127) remains a concern in evaluating the reproducibility and reliability of the findings. Weekly visits for 26 weeks might have been a contributory factor. However, in accordance with the EMA guidelines (62), we chose a study duration of 26 weeks to see whether there was a sustained treatment effect, lasting longer than the 12 weeks often reported for alcohol RCTs (63).
The approved 2 mg dosing regimen for diabetes patients is reported as the maximally efficacious dose for glucose control, reduction in body weight, and tolerable side effects (64). Our data also show striatum (M = -0.351, SE = 0.156, P = 0.039), but not in putamen (M = -0.405, SE = 0.202, P = 0.063). Within the placebo group, no statistically significant differences were found. (A and B) ROI data were analyzed using a repeated-measures ANOVA including factors group and time and an independent sample 2-tailed t test comparing groups (placebo and exenatide). Placebo, n = 16; exenatide, n = 17. Boxes represent upper and lower quartiles, the line represents the median, and the X represents the mean. Horizontal lines indicate significant interactions between treatment and time (*P < 0.05), and brackets indicate significant simple effects (*P < 0.05). JCI Insight 2022;7(19):e159863 https://doi.org/10.1172/jci.insight.159863 that AUD patients obtain the same incretin response as diabetes patients with respect to improved glycemic control, weight loss, and side effects. Also preclinically, the standard exenatide dose used in preclinical food reward trials (65) has shown effects in preclinical alcohol self-administration experiments (66,67). We did report a central effect in the brain imaging substudies, but of course we cannot rule out that the standard dose given was too low to elicit a reduction in number of heavy drinking days. However, the mean plasma exenatide level in this study was 4 times as high as that reported as the minimal effective concentration in humans, about 50 pg/mL (68). Also, because of safety concerns in this vulnerable group of patients, we did not raise the dose above the registered dose for treatment of type 2 diabetes.
Previous studies in diabetes patients have reported that while 45% of individuals receiving exenatide generate low-titer anti-exenatide antibodies (69), there is no apparent correlation between antibody titers and the effect of exenatide on mean HbA1c (57,69). To the best of our knowledge, there is also no evidence of altered exenatide clearance in AUD patients. The renal elimination of exenatide ( 5) is an advantage in this group of patients, who typically have a heightened risk of hepatic injury (1).
One would expect a correlation between reduced brain alcohol cue reactivity and alcohol consumption. However, this was not the case in the present study, neither for the whole group of patients (n = 127) nor for the subgroup of patients that were fMRI-scanned (n = 22) or SPECT-scanned (n = 16). The sample size of the fMRI BMI subgroups with BMI less than 25 (n = 7) or BMI greater than 30 (n = 5) was too small to further explore whether the overall fMRI striatal responses were correlated with heavy drinking days in the overweight or obese subgroups. Only a few RCTs on AUD patients including fMRI measurements at baseline and follow-up have been performed (70), and most studies have been underpowered or had too much variation in study populations to report significant clinical treatment effects (71). Reduced cue-induced activation in the exenatide group compared with the placebo group after 26 weeks of treatment in the left caudate and septal area (x, y, z coordinates = 0, 0, 4) (A) and right middle frontal gyrus (x, y, z coordinates = 36, 20, 48) (B). A 2-sample 2-tailed t test was performed for the post hoc analyses to compare groups (placebo, exenatide) and within a group across time (placebo/exenatide: T1, T2). For the group comparisons, the contrast of interest used was 'alcohol > neutral stimuli' , where the probability of a family wise error (FWE) was set to 0.05 to control for multiple statistical testing. Using the AlphaSim (3dClustSim) method, a combined voxel wise threshold of P < 0.001 and a cluster extent threshold of 101 voxels were calculated (n = 22).
Methods
Trial design. This randomized, placebo-controlled, double-blinded clinical trial was conducted at 4 alcohol outpatient clinics in Copenhagen, Denmark. The main trial comprised a 26-week treatment period investigating the primary and secondary endpoints. To evaluate the potential long-term effects, a single follow-up visit was conducted 6 months after treatment (24). A subgroup of the participants also underwent an fMRI scan and a single-photon emission CT (SPECT) DAT scan at baseline and after 26 weeks of treatment.
Patients. All potential participants received oral and written information about the project. Before signing of the written consent form, the alcohol breath concentration had to be below 0.5‰, which is the same limit as for driving a motor vehicle in Denmark (72). Eligible patients were 18-70 years of age, diagnosed with AUD according to DSM-5 and alcohol dependence according to ICD-10, and seeking treatment. Inclusion criteria required a minimum of 5 heavy drinking days, i.e., 60/48 g (men/women) of alcohol or more per day, in the past 30 days, measured by the Time-Line Follow Back (TLFB) method (73). Key exclusion criteria included severe mental disorder, other drug use disorder, a history of diabetes, pancreatitis, alcohol withdrawal seizures, and current treatment with drugs against alcohol dependence (disulfiram, acamprosate, naltrexone, and nalmefene). Full inclusion and exclusion criteria are listed in Supplemental Table 10 . The healthy controls included in the fMRI substudy (n = 25) were matched by sex, age, and educational level. All patients were recruited from outpatient alcohol treatment facilities in the suburbs of Copenhagen or through our project webpage, and healthy controls via the project webpage. No patients were involved in setting the research question, planning the study, or interpreting or writing up the results. The results of the trial and the assigned intervention will be disseminated to all patients and healthy participants.
Procedures. The randomization was stratified in terms of sex, age (+/-40 years of age), and number of heavy drinking days at baseline (4 strata), and the patients were randomly assigned 1:1 by Research Electronic Data Capture (REDCap) (74) to receive 2 mg exenatide once weekly (Bydureon) or placebo subcutaneously. The weekly injections were administered by an unblinded project nurse who did not participate in any assessments or behavioral treatment sessions. No randomization was performed in the imaging subgroup, as all eligible patients were invited to participate.
Patients who participated in the brain imaging substudy were scanned before receiving the first injection and again after 26 weeks of treatment. Throughout the trial, patients received the assigned treatment while wearing blindfolds by an unblinded nurse at the outpatient clinic, to whom they also delivered their weekly alcohol diary. Patients were assessed by blinded project staff at the time of 11 and Supplemental Figure 2 ). At every assessment, weight and somatic symptoms or diseases since the last visit were recorded, and safety blood samples were collected. In case medical assistance was needed, a 24-hour phone line was available. As a safety precaution due to earlier reports of pancreatitis caused by GLP-1 receptor agonist treatment (75), blood pancreas amylase was measured at all assessments. Participants with initial severe GI side effects received injections every second week for the first 6 weeks to reduce GI symptoms. All harms were recorded up until 10 weeks after termination of the intervention -i.e., week 26.
Throughout the trial, all patients received the assigned treatment as an add-on to standard AUD behavioral treatment, which included therapy sessions every second week, with a combination of motivational interviewing, cognitive therapy, and family therapy with a blinded therapist. Patients discontinuing the trial after a minimum of 8 weeks were encouraged to participate in a premature final visit and rescan. Only patients completing the week 26 visit (premature + per protocol) were invited for the long-term 6-month follow-up visit. The healthy fMRI control group was assessed for eligibility before brain imaging at the Neurobiology Research Unit at Rigshospitalet, Copenhagen, Denmark. See Supplemental Appendices 1 and 2 for full details of the fMRI and SPECT substudies, respectively.
Outcomes. The primary endpoint was change in heavy drinking days, from baseline to week 26, as recorded by the TLFB method. Secondary endpoints included changes in total alcohol consumption; number of days with no alcohol consumption; Penn Alcohol Craving Scale score; Alcohol Use Disorders Identification Test (AUDIT) score; Drug Use Disorders Identification Test (DUDIT) score; Screen for Cognitive Impairment in Psychiatry (SCIP) test; Fagerström Test for Nicotine Dependence; blood γ-glutamyl transferase; blood alanine aminotransferase; blood phosphatidylethanol (PEth); mean cell volume; glycemic control parameters (HbA1c); body weight; blood pressure; heart rate; measures of health and life quality, i.e., 36-Item Short Form Health Survey (SF-36) and Symptom Checklist-92 (SCL-92); SPECT DAT specific binding ratio (BP ND ); blood oxygen level-dependent (BOLD) fMRI signal change; change in subjective craving assessed with an alcohol cue reactivity task; change in top-down regulation assessed with an fMRI spatial working memory task; and change in heavy drinking days at 6-month follow-up. Additional methodological details regarding the analysis of blood and urine samples are given in Supplemental Methods.
Data availability. The study protocol, statistical analysis plan, and deidentified individual participant data, except raw fMRI and SPECT data and alcohol diaries, are available at the Mendeley database (76). Criteria for access to data are a methodologically sound proposal with an approved aim directed to the corresponding author, and requestors will have to sign a data access agreement. Data will be available for 5 years.
Statistics. The study was designed to have 90% power to detect a 28-percentage point treatment difference between the 2 groups with an estimated dropout of 40%. We planned to include 114 patients, but owing to a 60% dropout, we extended enrollment until October 1, 2019, or until 144 patients were included, whichever came first. All continuous outcomes were analyzed with an ANOVA adjusted for baseline until the last observational endpoint, and missing data were imputed with the use of multiple imputations in the mice package (77) in R software version 3.6.0 (78), method = pmm (predictive mean matching), and the number of imputed data sets = 100.
No adjustment for covariates was performed. SCIP data were analyzed with a linear mixed model, adjusted for benzodiazepine intake at the time of the assessment. DUDIT data were analyzed with a censored regression model due to zero-inflated values. An exploratory subgroup analysis based on the WHO BMI categories (79) was performed to see whether the effect of the treatment was related to baseline BMI. The statistical analysis plan was uploaded to the ClinicalTrials.gov homepage (80), and the data set was locked before any analyses were performed. All statistical analyses, except the post hoc analysis regarding exenatide plasma levels, were performed blinded. The hypothesis test was 2-sided, the level of statistical significance was 5%, and a confidence interval of 95% was used. All efficacy and safety analyses were performed according to the intention-to-treat principle. Analyses were performed with R software version 3.6.0 (78). See Supplemental Appendices 1 and 2 for the complete statistical method for the fMRI and SPECT analyses.
Study approval. The protocol was approved by the Danish Ethics Committee of the Capital Region, Copenhagen, Denmark (H-17003043), the Danish Medical Agency (2017014028), and the Danish Data Protection Agency (RHP-2017-029). The trial was monitored by an independent study monitor (Good Clinical Practice unit, Copenhagen, Denmark). Protocol modifications performed after trial commencement are shown in Supplemental Table 12 . All participants provided written informed consent prior to study inclusion. The funding sources and the manufacturer of exenatide once weekly (Bydureon, AstraZeneca) had no influence on the trial design or data analysis. The trial was conducted according to the Declaration of Helsinki, and the authors assume responsibility for the accuracy of data, analysis, and overall fidelity to the trial protocol.
availability
C L I N I A L M E D I C I N E JCI Insight 2022;7(19):e159863 https://doi.org/10.1172/jci.insight.159863
Author contributions
Conceptualization was contributed by AFJ and TV. Data curation was performed by MKK and CTE. Statistical power analysis and statistical analysis plan was performed by MKK, AFJ and CTE. Clinical data were analyzed by MKK and CTE. SPECT data were analyzed by MKK and MEJ. fMRI N-back task were analyzed by JM. fMRI ALCUE data were analyzed by PMF, MKK, AL, and SVK. Plasma FGF-21 were analyzed and validated by MPG. Urine oxidative stress parameters were analyzed and validated by HEP. Plasma PINP, CTX, TRAP-5b were analyzed and validated by NRJ. Exenatide and antibody plasma levels
Acknowledgements
AcknowledgmentsWe thank all patients and healthy controls who participated in the trial; all the foundations who supported the study (see below); the staff at the Novavi alcohol outpatient clinics for help with the recruitment of participants; and especially all the project nurses who administered injections every week.We also thank Kerstin Kiis Antonsen for help with planning and preparing the trial, protocol, and ethics approval; Jeff Zarp Petersen for preprocessing and performing the first-level analysis of the spatial working memory task data; Jens Svenningsen for assisting with the FGF-21 analysis; Svitlana Olsen for helping with the SPECT scans; Peter Steen Jensen for managing the imaging data and technical support; Birgit Tang for managing logistics regarding the fMRI scans; Signe Düring for being prinicpal investigator at the Novavi outpatient clinic; and Birgitte Sonne Rasmussen for her guidance in Good Clinical Practice and for conducting the GCP inspections.This study was funded by the Novavi Foundation; the Research Foundation, Mental Health Services, Capital Region of Denmark; the Research Foundation, Capital Region of Denmark; the Ivan Nielsen Foundation; the A.P. Moeller Foundation; the Augustinus Foundation; the Woerzner Foundation; the Grosserer L.F. Foghts Foundation; the Hartmann Foundation; the Aase and Ejnar Danielsen Foundation; the P.A. Messerschmidt and Wife Foundation; and the Lundbeck Foundation.Address correspondence to: Anders Fink-Jensen, Psychiatric Center Copenhagen, Edel Sauntes Allé 10, 2100 Copenhagen, Denmark.Phone: 4522755843; Email: anders.fink-jensen@regionh.dk.
Funding
AFJ has received an unrestricted research grant from Novo Nordisk A/S to investigate the effects of GLP-1 receptor stimulation on weight gain and metabolic disturbances in patients with schizophrenia treated with an antipsychotic.TV has served on scientific advisory panels for, been part of speaker's bureaus for, served as a consultant to, and/or received research support from Amgen, AstraZeneca, Boehringer Ingelheim, Eli Lilly, Gilead, Mundipharma, MSD/ Merck, Novo Nordisk, and Sun Pharmaceutical Industries.HB has received honoraria from Washington University Seminar.JJH has received consulting fees from Novo Nordisk A/S and grants from the Novo Nordisk Foundation.GMK has received personal honoraria from Sage Biogen, H. Lundbeck A/S, and Sanos and serves as president of the European College of Neuropsychopharmacology (unpaid) and chair of the Science and Infrastructure Advisory Board for the Human Brain Project (personal honorarium).SVK has received grants from the German Research Foundation.JM has received honoraria from H. Lundbeck A/S.KWM has received honoraria from H. Lundbeck A/S and Janssen.The funding sources and the manufacturer of exenatide once weekly (Bydureon, AstraZeneca) had no influence on the trial design or data analysis.FUNDING.Novavi Foundation; Research Foundation, Mental Health Services, Capital Region Denmark; Research Foundation, Capital Region of Denmark; Ivan Nielsen Foundation; A.P. Moeller Foundation; Augustinus Foundation; Woerzner Foundation; Grosserer L.F. Foghts Foundation; Hartmann Foundation; Aase and Ejnar Danielsen Foundation; P.A. Messerschmidt and Wife Foundation; and Lundbeck Foundation.
Failed to load PDF:
StripeM-Inner
Paper sources
Abstract screening pilot
Abstract screening results
Full-text screening pilot
Full-text screening results
Extraction pilot
Extraction results
Research report
Modify setup
May 4, 2026 3:39 AM
May 4, 2026 3:19 AM