ASSOCIATION OF GLUCAGON-LIKE PEPTIDE-1 AGONISTS WITH CIRRHOSIS AND HEPATOCELLULAR CANCER IN METABOLIC DYSFUNCTION ASSOCIATED STEATOTIC LIVER DISEASE (MASLD)

Background: Patients with MASLD and diabetes have an increased risk of cirrhosis and hepatocellular cancer (HCC). Recent studies show that glucagon-like peptide-1 (GLP-1) agonists, such as semaglutide, are effective in improving liver inflammation in MASLD. Whether these treatments prevent progression of liver disease and HCC remains unknown.

Methods: We conducted a retrospective cohort study of adult patients with type 2 diabetes and concomitant MASLD, defined based on the recent nomenclature, who sought care at 129 hospitals in U.S. Veterans Affairs (VA) healthcare system. We identified all patients who initiated GLP-1 (users) and used the date of first prescription as the index date for follow-up. Each user was matched with up to 50 patients who were not prescribed GLP-1 (non-users) from the same cohort. We defined the index date for non-users so that they matched the users on age, sex, and duration of care in the VA. For the primary analysis, we excluded patients with prevalent cirrhosis (defined based ICD codes or FIB-4>2.67) or HCC any time before or 6 months after index. We then applied propensity score weighting to balance potential confounders between users and non-users, including demographics, income, disability, eligibility for government aid, cardiometabolic risk factors and their treatments, comorbidity index, duration and severity of diabetes, and lab-based underlying liver disease severity. The outcome was the time from the index date to cirrhosis or HCC, whichever occurred first. We conducted a separate analysis among patients who had progressed to cirrhosis at the time of GLP-1 use; matched non-users were patients who had cirrhosis by matched index date. The outcome was time to HCC. For both analyses, we used cause-specific Cox proportional hazards models to study the causal effect of GLP-1 use.

Results: We identified 23,670 patients who used liraglutide (13%), semaglutide (70%), or both agents (17%) and 169,646 matched non-users, followed for median of 20 months (interquartile range, 10-39 months). The propensity score weighting achieved balance in all confounders. The annual incidence rate of cirrhosis or HCC was 0.82% among users and 2.3% among non-users (adjHR: 0.29; 95% confidence interval [CI] 0.26-0.32). GLP-1 agonists were also associated with a reduced risk of HCC (annual incidence 0.02% vs. 0.05%, adjHR: 0.36, 95%CI, 0.17-0.76). There was a trend toward HCC chemopreventive benefit with GLP-1 agonists in patients with cirrhosis (3706 users, 25763 non-users; annual incidence 0.35% vs. 0.52%, respectively; adjHR, 0.68, 95%CI, 0.43-1.09).

Conclusions: In a nationwide study of patients with MASLD and type 2 diabetes, use of semagludite or liraglutide was associated with a significantly lower risk of progression to cirrhosis or HCC than no use. These are the first data linking GLP-1 agonists to a reduced risk of HCC in MASLD.