What is the evidence behind using GLP-1 agonists to treat alcohol use disorder in patients with cirrhosis of the liver?

Please summarize the evidence behind use of GLPs to treat alcohol use disorder in patients with cirrhosis of the liver

Comment by InpharmD Researcher

Limited evidence, primarily derived from mechanistic evaluations and small trials, suggests that glucagon-like peptide-1 receptor agonists (GLP-1RAs) may reduce alcohol intake, cravings, heavy drinking days, and relapse. Findings indicate these effects are likely mediated through modulation of central reward pathways, stress responses, satiety, and cognitive control, with early human studies and observational data showing potential benefits, particularly in individuals with obesity or metabolic comorbidities. However, given the small sample sizes and observational nature of most evidence, caution in interpretation is warranted. Importantly, there is a lack of studies specifically evaluating GLP-1RA use for alcohol use disorder in patients with established cirrhosis, leaving their efficacy, safety, and impact on liver-related outcomes in this specific population unclear.

A targeted literature search was conducted in PubMed using combinations of terms related to GLP-1 receptor agonists (e.g., GLP-1, semaglutide, liraglutide), alcohol use disorder (e.g., alcohol dependence, alcohol consumption, craving), and liver disease (e.g., cirrhosis, hepatic impairment, alcoholic liver disease). Supplementary searches were performed in general search engines (e.g., Google Scholar) using similar keywords to identify additional clinical, observational, and mechanistic evidence.

Background

Numerous review articles indicate that glucagon-like peptide-1 receptor agonists (GLP-1RAs), initially developed for type 2 diabetes and obesity, may reduce alcohol intake, craving, heavy drinking days, and relapse in both preclinical models and early human studies. Preclinical studies consistently demonstrate that GLP-1RAs decrease alcohol consumption, reduce motivation to drink, and prevent relapse, likely through modulation of central reward pathways, stress responses, satiety signaling, and cognitive control. Early clinical trials and small randomized studies in humans suggest reductions in alcohol use, particularly among individuals with obesity or metabolic comorbidities, although results are variable and sample sizes are limited. Observational and retrospective studies further report associations between GLP-1RA use and lower rates of alcohol-related hospitalizations, incident AUD, and other adverse alcohol-related outcomes, but these data are subject to confounding. Importantly, the available evidence does not specifically evaluate GLP-1RAs for AUD treatment in patients with established cirrhosis, and most studies involve metabolic cohorts or general AUD populations without stratification by cirrhosis status, leaving their efficacy, safety, and impact on liver-related outcomes in cirrhotic patients unestablished. [1], [2], [3], [4], [5]

A 2025 systematic review and meta-analysis investigated the impact of GLP-1RAs on alcohol consumption and liver-related outcomes. A total of 11 studies involving 263,616 patients (124,884 on GLP-1RAs, 138,732 controls) were included, with eight studies providing sufficient data for the meta-analysis. The standardized mean difference (SMD), incidence rate ratio (IRR), and hazard ratio (HR) with 95% confidence intervals (CI) were utilized as measures of pooled effect size. The findings revealed that GLP-1RAs were associated with a reduction in alcohol intake, binge drinking, AUDIT scores, and all-cause mortality, although results regarding heavy drinking were inconclusive. While the pooled analysis of alcohol intake demonstrated no significant difference with GLP-1RA use (SMD -0.17; 95% CI -0.39 to 0.04; I²= 0.0%), a noteworthy association was identified in reducing alcohol-related events (HR 0.56; 95% CI 0.48 to 0.65; I²= 63.6%) and liver-related outcomes (IRR 0.65; 95% CI 0.50 to 0.85; I² = 0.0%). Progression to cirrhosis showed no statistical difference, and most adverse events were of mild severity. These results underscore the potential of GLP-1RAs to diminish alcohol-related and liver-related events, although no significant impact on pooled alcohol intake was observed. Based on these findings, the authors emphasized that further high-quality studies are needed to robustly establish the efficacy of GLP-1RAs in reducing alcohol use and mitigating liver damage. [6]

Literature Review

A search of the published medical literature revealed 4 studies investigating the researchable question:

Please summarize the evidence behind use of GLPs to treat alcohol use disorder in patients with cirrhosis of the liver

Level of evidence

C - Multiple studies with limitations or conflicting results Read more→

Please see Tables 1-4 for your response.

GLP-1 Receptor Agonists and Risk for Cirrhosis and Related Complications in Patients With Metabolic Dysfunction-Associated Steatotic Liver Disease
Design	Retrospective cohort study with an active comparator, new-user design N= 16,058
Objective	To determine whether use of GLP-1 RAs is associated with lower risk of developing cirrhosis and its complications, including decompensation and hepatocellular cancer (HCC), among patients with metabolic dysfunction-associated steatotic liver disease (MASLD)
Study Groups	GLP-1 RA users without cirrhosis (n= 14,606) GLP-1 RA users with cirrhosis (n= 1,452) DPP-4i users without cirrhosis (n= 14,606) DPP-4i users with cirrhosis (n= 1,452)
Inclusion Criteria	Patients with MASLD and diabetes who initiated either a GLP-1 RA or DPP-4i between January 1, 2006, and June 30, 2022, at Veterans Health Administration hospitals
Exclusion Criteria	Patients <18 or >75 years old, those with liver transplant or HCC before baseline
Methods	Each GLP-1 RA new user was propensity score matched in a 1:1 ratio to a patient who initiated a DPP-4i during the same month. Separate analyses were conducted among patients without and with cirrhosis at baseline. Outcomes were assessed using ICD codes and VHA data
Duration	January 1, 2006, to December 31, 2022
Outcome Measures	Progression to cirrhosis, cirrhosis complications (composite and individual), decompensation, HCC, all-cause mortality
Baseline Characteristics		GLP-1 RA users without cirrhosis (n= 14,606)	DPP-4i users without cirrhosis (n= 14,606)	GLP-1 RA users with cirrhosis (n= 1,452)	DPP-4i users with cirrhosis (n= 1,452)
	Age, years	60.56 ± 10.31	60.47 ± 10.31	66.99 ± 7.09	67.01 ± 7.07
	Male	89.1%	89.4%	93.7%	93.9%
	Diabetes duration, years	7.77 ± 4.89	7.67 ± 4.70	9.54 ± 4.79	9.04 ± 4.78
	BMI	36.05 ± 6.66	33.55 ± 6.47	35.40 ± 6.08	35.14 ± 6.53
	Hypertension	89.8%	89.2%	95.0%	94.2%
	Chronic kidney disease	16.0%	15.1%	30.4%	26.9%
	Alcohol Use Disorders Identification Test 0 1 2 3	9412 ± 64.4 3815 ± 26.1 1049 ± 7.2 330 ± 2.3	9432 ± 64.6 3775 ± 25.8 1053 ± 7.2 346 ± 2.4	1048 ± 72.2 316 ± 21.8 68 ± 4.7 20 ± 1.3	1058 ± 72.9 311 ± 21.4 64 ± 4.4 19 ± 1.3
Results		GLP-1 RA users without cirrhosis	DPP-4i users without cirrhosis	GLP-1 RA users with cirrhosis	DPP-4i users with cirrhosis
	Cirrhosis incidence per 1000 person-years	9.98	11.10	-	-
	Composite outcome incidence per 1000 person-years	1.89	2.55	18.64	15.31
	Decompensation incidence per 1000 person-years	1.80	2.26	14.45	13.32
	HCC incidence per 1000 person-years	0.24	0.27	5.32	3.20
	All-cause mortality incidence per 1000 person-years	21.77	24.43	54.75	61.91
Adverse Events	No specific adverse events related to GLP-1 RA use were reported in the study.
Study Author Conclusions	GLP-1 RA use was associated with a lower risk of progression to cirrhosis and mortality among patients with MASLD and diabetes. The protective association was not seen in patients with existing cirrhosis, underscoring the importance of treatment earlier in the disease course.
Critique	The study's strengths include a large sample size and the use of a well-matched comparator group, which helps mitigate confounding. However, the retrospective design may introduce bias, and the study's reliance on administrative data could lead to misclassification of outcomes. The findings may not be generalizable to populations outside the Veterans Health Administration, and the study did not assess potential harms associated with GLP-1 RA use.

References:
[1] Kanwal F, Kramer JR, Li L, et al. GLP-1 Receptor Agonists and Risk for Cirrhosis and Related Complications in Patients With Metabolic Dysfunction-Associated Steatotic Liver Disease. JAMA Intern Med. 2024;184(11):1314-1323. doi:10.1001/jamainternmed.2024.4661

Impact of GLP-1 Receptor Agonists on Alcohol-Related Liver Disease Development and Progression in Alcohol Use Disorder
Design	Two retrospective cohort studies using the TriNetX Research Network N= 7132 (Alcohol-use disorder [AUD] cohort) N= 1896 (alcohol-related liver disease [ArLD] cohort)
Objective	To investigate the association between GLP-1 receptor agonists and the development and progression of ArLD in patients with alcohol use disorder
Study Groups	AUD cohort: GLP-1RA (n= 3566) DPP-4i (n= 3566) ArLD cohort: GLP-1RA (n= 948) DPP-4i (n= 948)
Inclusion Criteria	Patients aged 18 years or older with T2D and AUD, initiated on either GLP-1RAs or DPP-4is between January 2010 and December 2023
Exclusion Criteria	Use of medication indicated for AUD, history of bariatric surgery, prior exposure to GLP-1RAs or DPP-4is before the study period, history of ArLD prior to the index event
Methods	Retrospective cohort study using the TriNetX Research Network, propensity score matching with a 1:1 ratio, stratified into GLP-1RA and DPP-4i groups, follow-up for outcomes including ArLD incidence and hepatic decompensation.
Duration	AUD cohort: median follow-up 63.2 months ArLD cohort: median follow-up 28.2 months
Outcome Measures	Primary: Incident ArLD in the AUD cohort, hepatic decompensation in the ArLD cohort Secondary: All-cause mortality
Baseline Characteristics		GLP-1RA (n= 3566)	DPP-4i (n= 3566)
	Age at index, mean (SD), years	57.8 (12.0)	57.6 (13.5)
	Female	852 (23.9%)	847 (23.8%)
	Male	2601 (72.9%)	2603 (73.0%)
	White	2080 (58.3%)	2060 (57.8%)
	BMI, mean (SD)	32.9 (7.3)	31.9 (7.0)
Results	Treatment with GLP-1RAs was associated with significantly improved liver-related and survival outcomes compared with dipeptidyl peptidase-4 inhibitors (DPP-4is). Among patients with AUD without baseline ArLD (n= 7132), GLP-1RA use was associated with a lower incidence of ArLD over a median follow-up of 63.2 months (6.0 vs. 8.7 per 1000 person-years), corresponding to a 38% relative risk reduction (hazard ratio [HR] 0.62; 95% confidence interval [CI] 0.44 to 0.87; p= 0.006). GLP-1RA therapy was also associated with significantly reduced all-cause mortality in this cohort (HR 0.53; 95% CI 0.45 to 0.62; p< 0.001). In a separate matched cohort of patients with established ArLD (n= 1896), GLP-1RA use was associated with a lower risk of hepatic decompensation over a median follow-up of 28.2 months (39.5 vs. 51.4 per 1000 person-years; HR 0.66; 95% CI 0.51 to 0.85; p= 0.001) and a significantly reduced risk of all-cause mortality (HR 0.53; 95% CI 0.41 to 0.68; p< 0.001), while no significant differences were observed for hepatocellular carcinoma or liver transplantation. These associations were consistent across multiple stratified and sensitivity analyses, supporting the robustness of the findings.
Adverse Events	No specific adverse events reported in the study.
Study Author Conclusions	Our study demonstrates that GLP-1RA treatment is associated with significant clinical benefits in patients with AUD and T2D, including reduced risk of ArLD incidence, decreased ArLD progression, lower rates of hepatic decompensation and improved survival compared to DPP-4i treatment. These findings suggest that GLP-1RA could be a promising therapeutic option for patients with co-existing AUD and T2D, particularly given the limited pharmacological treatments currently available for AUD. While alcohol abstinence remains the cornerstone of AUD/ArLD management, GLP-1RA may offer additional benefits in reducing disease progression and mortality.
Critique	The study's strengths include the use of a large real-world database and propensity score matching to balance groups. However, limitations include reliance on administrative data, potential residual confounding, and lack of detailed clinical information such as alcohol consumption patterns and severity of liver disease.

References:
[1] Kuo CC, Li CH, Chuang MH, Huang PY, Kuo HT, Lai CC. Impact of GLP-1 Receptor Agonists on Alcohol-Related Liver Disease Development and Progression in Alcohol Use Disorder. Aliment Pharmacol Ther. 2025;61(8):1343-1356. doi:10.1111/apt.70007

Association of Glucagon-Like Peptide-1 Receptor Agonists With Liver-Related Outcomes and All-Cause Mortality in Patients With Harmful Alcohol Use: A Target Trial Emulation Study
Design	Observational target trial emulation study using electronic health records N= 16,080 (8,040 GLP-1 RA initiators and 8,040 controls)
Objective	To assess liver-related outcomes and mortality among individuals with harmful alcohol use who received GLP-1 RAs
Study Groups	GLP-1 RA initiators (n= 8,040) Controls (n= 8,040)
Inclusion Criteria	Patients aged 18 years or older with a positive AUDIT-C score, included from 1/3/2017 to 9/30/2024
Exclusion Criteria	Patients with substance use disorders other than alcohol and tobacco, on GLP-1/GIP agonists or other AUD treatments, fewer than 180 days from baseline to end of follow-up, or with HCC, decompensation, or death within 6 months of baseline
Methods	Clinical, laboratory, pharmacy, and imaging data were obtained from electronic health records within a national Veterans Affairs cohort. Exposure was defined as initiation of a glucagon-like peptide-1 receptor agonist with sustained prescription use. Participants were matched 1:1 to untreated controls using exact and propensity score matching across demographic variables, alcohol use severity, liver disease features, cardiometabolic comorbidities, medication exposures, and laboratory parameters. Outcomes were ascertained longitudinally using validated diagnostic codes, death registry data, and serial AUDIT-C assessments. Time-to-event analyses were conducted for liver-related outcomes and mortality with multivariable adjustment for residual imbalances. Alcohol use severity during follow-up was evaluated using repeated AUDIT-C measurements and modeled with adjusted regression analyses. Subgroup and dose–response analyses were performed, including evaluation of semaglutide-specific effects.
Duration	Follow-up until outcomes or study end (09/30/2024)
Outcome Measures	Primary: Time to composite liver-related outcomes (decompensation, hepatocellular carcinoma, liver-related death) and all-cause mortality Secondary: Proportion of patients with positive AUDIT-C scores
Baseline Characteristics		GLP-1 (n= 8,040)	Control (n= 8,040)
	Age, years (IQR)	64.0 (16.0)	67.0 (15.0)
	Female	706 (8.8%)	706 (8.8%)
	Race White Black Hispanic/Latino Other/unknown	4,926 (61.3%) 1,391 (17.3%) 539 (6.7%) 877 (10.9%)	4,955 (61.6%) 1,443 (18.0%) 473 (5.9%) 831 (10.3%)
	BMI 27.1 to 29.9 >= 30	1,027 (12.8%) 6,243 (77.7%)	1,021 (12.7%) 6,243 (77.7%)
	Cirrhosis or elevated FIB-4	485 (6.0%)	485 (6.0%)
	Steatosis	2,876 (35.8%)	2,876 (35.8%)
Results	Initiation of glucagon-like peptide-1 receptor agonists (GLP-1 RAs) was associated with significantly improved clinical outcomes compared with matched nonusers. Over a median follow-up of approximately three years, GLP-1 RA exposure was associated with a 30% reduction in the risk of the composite liver-related outcome, which included hepatic decompensation, hepatocellular carcinoma, and liver-related death (adjusted hazard ratio [aHR] 0.70, 95% confidence interval [CI] 0.56 to 0.87; p= 0.001), as well as a 57% reduction in all-cause mortality (aHR 0.43; 95% CI 0.37 to 0.49; p< 0.0001). Subgroup analyses demonstrated that these benefits were primarily driven by semaglutide use, which was associated with a greater reduction in composite liver-related outcomes (aHR 0.52; 95% CI 0.40 to 0.67) and mortality (aHR 0.20; 95% CI 0.16 to 0.25), with a significant dose–response relationship observed whereby each 1-mg weekly increase in semaglutide dose was associated with further reductions in liver-related outcomes and death. In contrast, non-semaglutide GLP-1 RAs were not associated with mortality benefit and were linked to higher rates of liver-related outcomes. With respect to alcohol use, GLP-1 RA therapy was associated with a lower likelihood of persistent harmful drinking during follow-up, as evidenced by reduced odds of a positive AUDIT-C score between 1 and 12 months compared with controls (adjusted odds ratio 0.75; 95% CI 0.68 to 0.82; p< 0.0001), an effect that was consistent across most metabolic and liver disease subgroups but without a clear dose–response relationship.
Adverse Events	Not specifically reported in the study.
Study Author Conclusions	In this target trial emulation study of Veterans with harmful alcohol use, GLP-1 RA use was associated with reduced composite liver outcomes, all-cause mortality, and reduced likelihood of harmful alcohol use during follow-up.
Critique	The study's strengths include a large sample size and the use of a target trial emulation design, which enhances the validity of the findings. However, the observational nature may introduce residual confounding, and the predominantly male veteran population may limit generalizability to other groups. Additionally, the study did not directly measure alcohol consumption in grams, relying instead on AUDIT-C scores.

References:
[1] John BV, Bastaich D, Marchetti D, et al. Association of Glucagon-Like Peptide-1 Receptor Agonists With Liver-Related Outcomes and All-Cause Mortality in Patients With Harmful Alcohol Use: A Target Trial Emulation Study. Am J Gastroenterol. Published online June 10, 2025. doi:10.14309/ajg.0000000000003585

Association of Glucagon-Like Peptide-1 Receptor Agonist with Progression to Liver Cirrhosis and Alcohol-Related Admissions in Patients with Alcohol Use Disorder and Diabetes: A Retrospective Cohort Study
Design	Retrospective propensity score-matched cohort study N= 29,653
Objective	To compare the risk of progression to liver cirrhosis and alcohol-related hospital admission after initiation of glucagon-like peptide-1 receptor agonists (GLP-1RA) versus dipeptidyl peptidase-4 inhibitors (DPP4i) in patients with type 2 diabetes mellitus and alcohol use disorder (AUD)
Study Groups	GLP-1RA users (n= 9,965) DPP4i users (n= 19,688)
Inclusion Criteria	Adults aged 35 years or older, regular Veteran Healthcare Administration (VHA) users, diagnosed with AUD or alcohol-associated liver disease, and had diabetes
Exclusion Criteria	Concomitant users of GLP-1RA and DPP4i, no follow-up after medication initiation, diagnosis of hepatocellular carcinoma, malignant metastases to the liver, or liver transplantation at baseline
Methods	Retrospective cohort study using data from the VHA from 2006 to 2021. Propensity score matching was used to match 7,302 pairs on 79 characteristics. Outcomes were assessed using ICD9 or ICD10 codes for cirrhosis and alcohol-related admissions.
Duration	2006 to 2021
Outcome Measures	Primary: Progression to cirrhosis, alcohol-related hospital admission Secondary: Components of progression to cirrhosis (compensated and decompensated cirrhosis)
Baseline Characteristics		DPP4i users (n= 7,302)	GLP-1RA users (n= 7,302)
	Age at index date, years	61 ± 10	61 ± 10
	Male	6,914 (94.7%)	6,918 (94.7%)
	Race Caucasian African American	4446 (61%) 2,101 (29%)	4687 (64%) 1,925 (26%)
	Ethnicity Hispanic/Latino Non-Hispanic/Latino	674 (9%) 6,405 (88%)	668 (9%) 6,401 (88%)
	Vital signs Mean systolic blood pressure, mmHg Mean diastolic blood pressure, mmHg	135 ± 12 79 ± 7	135 ± 11 79 ± 7
	Body mass index <25 kg/m2 25 to <30 kg/m2 30 to <35 kg/m2 35 to <40 kg/m2 40 to <45 kg/m2 ≥45 kg/m2	329 (4.5%) 1,628 (22.3%) 2,583 (35.4%) 1,633 (22.4%) 736 (10.1%) 393 (5.4%)	339 (4.6%) 1,650 (22.6%) 2,591 (35.5%) 1,644 (22.5%) 728 (10.0%) 350 (4.8%)
	Compensated liver cirrhosis	616 (8.4%)	612 (8.4%)
	Ascites	124 (1.7%)	126 (1.7%)
	Bleeding esophageal varices	9 (0.1%)	10 (0.1%)
	Hepatic encephalopathy	18 (0.3%)	17 (0.2%)
	Abbreviations: DPP4i, dipeptidyl peptidase-4 inhibitor; GLP-1RA, glucagon-like peptide-1 receptor agonist. Baseline characteristics are presented after propensity score matching. After propensity score matching, the prevalence of nonalcoholic fatty liver disease/non-alcoholic steatohepatitis (NAFLD/NASH), alcohol-associated liver disease, viral hepatitis, compensated cirrhosis, and other liver conditions was similar between groups. Baseline laboratory values were comparable, with approximately 85% having normal ALT, 89% with platelet counts >150 K/mm³, and similar distributions of AST, Fibrosis-4 (FIB-4), Hepatic Steatosis Index (HSI), and Aspartate aminotransferase to Platelet Ratio Index (APRI). Body mass index categories, diabetes complications, metformin use (81%), and healthcare utilization were also balanced. Mean exposure durations to GLP-1RA and DPP4i therapy were comparable between groups.
Results		DPP4i (n= 7,302)	GLP-1RA (n= 7,302)	OR (95% CI)	p-value
	Progression to cirrhosis	441 (6.0%)	484 (6.6%)	1.1 (0.97–1.26)	0.14
	Compensated cirrhosis	393 (5.4%)	435 (6%)	1.1 (0.97–1.28)	0.13
	Decompensated cirrhosis	195 (2.7%)	203 (2.8%)	1.0 (0.85–1.27)	0.68
	Alcohol-related hospital admission	123 (1.7%)	105 (1.4%)	0.85 (0.65–1.11)	0.23
	Abbreviations: CI, confidence interval; DPP4i, dipeptidyl peptidase-4 inhibitor; GLP-1RA, glucagon-like peptide-1 receptor agonist; OR, odds ratio. In the secondary analysis of patients with preexisting liver disease, progression to cirrhosis occurred in 531 GLP-1RA users (23.3%) and 1,126 DPP4i users (25.1%) (OR 1.09; 95% CI 0.95–1.26), and alcohol-related hospital admission occurred in 45 GLP-1RA users (2.0%) and 186 DPP4i users (4.2%) (OR 0.85; 95% CI 0.59–1.24). Across duration-based subgroups, outcomes did not differ significantly; for example, among patients with ≥3 years of medication use, progression to cirrhosis occurred in 7.4% of DPP4i users and 7.6% of GLP-1RA users (OR 1.2; 95% CI 0.97–1.59), and alcohol-related hospital admission remained similar (OR 0.66; 95% CI 0.39–1.12).
Adverse Events	No specific adverse events reported in the study
Study Author Conclusions	Use of GLP-1RA in patients with AUD was not associated with beneficial effect on progression to cirrhosis or alcohol-related hospital admission.
Critique	The study's strength lies in its large sample size and use of propensity score matching to minimize confounding. However, the retrospective design limits causal inference, and the study population was predominantly white males, which may affect generalizability. Additionally, the study did not quantify alcohol consumption, which could be a significant confounder in assessing the outcomes.

References:
[1] Al-Moussally F, Khan S, Katukuri V, Kinaan M, Mansi IA. Association of Glucagon-Like Peptide-1 Receptor Agonist with Progression to Liver Cirrhosis and Alcohol-Related Admissions in Patients with Alcohol Use Disorder and Diabetes: A Retrospective Cohort Study. Drugs. 2025;85(6):813-825. doi:10.1007/s40265-025-02177-x