Author information

¹Department of Biomedical Informatics, College of Medicine, The Ohio State University, Columbus, Ohio, USA.

²Center for Biostatistics, College of Medicine, The Ohio State University, Columbus, Ohio, USA.

³Department of Internal Medicine, Division of Medical Oncology, College of Medicine, The Ohio State University, Columbus, Ohio, USA.

⁴Department of Internal Medicine, Division of Gastroenterology, Hepatology & Nutrition, College of Medicine, The Ohio State University, Columbus, Ohio, USA.

⁵Digestive Disease and Surgery Institute, Cleveland Clinic, Cleveland, Ohio, USA.

⁶Division of Gastroenterology and Hepatology, Rutgers Robert Wood Johnson Medical School, New Brunswick, New Jersey, USA.

⁷Center for Liver Diseases and Masses, Rutgers Robert Wood Johnson Medical School, New Brunswick, New Jersey, USA.

⁸Division of Population Sciences, The Ohio State University Comprehensive Cancer Center, Columbus, Ohio, USA.

⁹Department of Internal Medicine, Division of Cancer Prevention and Control, College of Medicine, The Ohio State University, Columbus, Ohio, USA.

¹⁰Department of Medicine, University of Pennsylvania Perelman School of Medicine, Philadelphia, Pennsylvania, USA.

Abstract

Background/aims: We examined the effects of glucagon-like peptide-1 receptor agonists (GLP-1RAs) initiation on long-term Adverse Liver Outcomes (ALO) in patients with Metabolic Dysfunction-Associated Steatotic Liver Disease (MASLD) cirrhosis and type 2 diabetes using real-world data from the MarketScan database.

Methods: We conducted a retrospective cohort study of patients with MASLD cirrhosis and type 2 diabetes between 2012 and 2020. Cox proportional hazard models examine the association between GLP-1RAs initiation, modelled as time-dependent, and the risk of ALO, a composite endpoint defined by the first occurrence of hepatic decompensation(s), portal hypertension, hepatocellular carcinoma (HCC) or liver transplantation (LT). We used Overlap Propensity Score Weighting (OPSW) to account for confounding. The study included 459 GLP-1RAs and 4837 non-GLP-1RAs patients.

Results: The non-GLP-1RAs patients presented with 1411 (29%) ALO over 7431.7 person years, while GLP-1RAs patients had 32 (7%) ALO over 586.6 person years - risk rate difference 13.5 (95% CI: 11.4-15.7) per 100 person-years. The OPSW-adjusted risk of ALO was reduced by 36% (hazard ratio [HR]: 0.64; 95% CI: 0.54-0.76) in patients with vs. without GLP-1RAs initiation. GLP-1RAs initiation was associated with significant reductions in the adjusted risk of hepatic decompensation (HR: 0.74; 95% CI: 0.61-0.88), portal hypertension (HR: 0.73; 95% CI: 0.60-0.88), HCC (HR: 0.37; 95% CI: 0.20-0.63) and LT (HR: 0.24; 95% CI: 0.12-0.43).

Conclusion: The use of GLP-1RAs was associated with significant risk reductions in long-term adverse liver outcomes, including hepatic decompensation, portal hypertension, HCC and LT, in MASLD cirrhosis patients with type 2 diabetes.