Early use of glucagon-like peptide-1 receptor agonists was associated with fewer liver-related complications and lower transplant risk compared to metabolic bariatric surgery in patients with metabolic-dysfunction associated steatohepatitis (MASH) cirrhosis and obesity without diabetes, according to a large US database study.
“The incidence of [metabolic dysfunction-associated steatotic liver disease (MASLD)] is rising, and it is projected to be a leading cause of liver transplantation in the US,” study author Charanpreet Sasan, MD, told GI & Hepatology News in advance of Digestive Disease Week® (DDW) 2026, where the work was presented. “Given increased risks from surgical procedures, associated recovery times and cost, being able to provide a non-surgical effective alternative option like GLP-1s can help to expand our therapeutic toolkit significantly.”
For the analysis, Dr. Sasan, a resident physician at UMass Chan Medical School, Worcester, Mass., and colleagues evaluated outcomes in patients with MASH cirrhosis and obesity who did not have type 2 diabetes mellitus. They compared patients who initiated the glucagon-like peptide-1 receptor agonists (GLP-1s) semaglutide, tirzepatide, or liraglutide within one year of a cirrhosis diagnosis with those who underwent metabolic bariatric surgery (MBS) during the same period. Data were sourced from TriNetX, a network of 70 US health care organizations, using diagnostic codes to identify eligible patients while excluding those with hepatocellular carcinoma or cholangiocarcinoma.
The investigators used Kaplan-Meier methods to evaluate outcomes over time. Matching accounted for age, sex, race, ethnicity, body mass index, liver enzymes, serum creatinine, and cardiometabolic risk factors to reduce baseline differences between groups.
Before matching, the 2,475 patients treated with GLP-1s were older on average (51 vs 49 years), with slightly higher body mass index and higher lipid and glycemic markers than the 2,079 patients who underwent MBS. After propensity score matching to balance demographic, metabolic, and laboratory variables, the final cohort included 3,074 patients.
Over up to 10 years of follow-up, GLP-1 therapy was associated with lower rates of cirrhosis complications compared with surgery. Patients receiving GLP-1 had about 43% lower odds of developing portal hypertension. The risk reductions extended across multiple complications: ascites occurred at about 80% lower odds, hepatorenal syndrome at 89% lower odds, hepatic encephalopathy at 71% lower odds, and esophageal varices at 39% lower odds compared with MBS.
The need for liver transplantation also differed between groups. Patients treated with GLP-1 had about 55% lower odds of requiring transplant compared with those who underwent surgery.
“The most surprising finding was that the benefits of GLP-1 extended far beyond the reduction in decompensating events like ascites, variceal bleeding, hepatorenal syndrome and hepatic encephalopathy, particularly altering liver transplantation risk,” Dr. Sasan said. “Early initiation of the medication, within one year of a cirrhosis diagnosis, resulted in a significant decrease and delay in the need for transplantation further stressing their benefit.”
She added that the findings address a gap in evidence for patients with MASH cirrhosis who do not have diabetes, and advocate for a shift towards earlier and more proactive therapeutic intervention. “While prior studies have demonstrated the benefit of GLP-1 in the earlier stages of steatotic liver disease to prevent progression, our data shows that, even after development of cirrhosis, there are benefits from initiating GLP-1 therapy,” Dr. Sasan said. “The results emphasize that starting the GLP-1 therapy early, within the first year of a cirrhosis diagnosis, can help provide substantial protection against further MASH cirrhosis decompensations and decrease liver transplantation risk.”
During a question-and-answer session at the meeting, an attendee asked Dr. Sasan whether she and her colleagues had observed any differences in effects among the various GLP-1 drugs. “We had wondered about that, but we weren't able to run subgroup analyses because the N would get too small between the three GLP-1s that we included,” she said.
Despite the large sample size and long follow-up, the study has limitations. The retrospective design relies on coded data and may be subject to misclassification. Residual confounding is possible even after matching, particularly for factors such as medication adherence and lifestyle changes.
Dr. Sasan reported having no disclosures.
DDW is AGA’s annual meeting, jointly sponsored by AGA, AASLD, ASGE, and SSAT. Learn more at ddw.org.