PATIENTS WITH MAFLD TAKING GLUCAGON LIKE PEPTIDE-1 RECEPTOR AGONISTS DO NOT HAVE HIGHER INCIDENCE FOR HEPATOBILIARY CANCER BUT HAVE LOWER OVERALL MORTALITY: A MULTI-CENTER ANALYSIS.

Introduction
Metabolic dysfunction-associated fatty liver disease (MAFLD) affects one in four people globally and is the primary cause of chronic liver disease. MAFLD is associated with a nearly 1.5-fold to 2-fold increased risk of developing GI cancers. GLP-1 receptor agonists (GLP-1 RAs) have shown promising results in treating MAFLD. Therefore, we aim to evaluate the occurrence of a subgroup of GI malignancies (hepatobiliary cancers) and overall mortality of patients with MAFLD.
Methods
A retrospective cohort study utilizing the TriNetX platform was performed. Adult patients with diagnosis of MAFLD from 80 hospitals across the US between January 1, 2010, and October 31, 2023 were included. Using 1:1 propensity matching, the patients from this cohort who were using GLP-1 RAs (Semaglutide, Liraglutide, Dulaglutide) were matched with MAFLD patients not using GLP-1 RAs according to age, demographics, comorbidities, and medication. The primary endpoint was the occurrence of hepatobiliary cancers, including hepatic, biliary, and gallbladder malignancies. The secondary endpoint was overall mortality of patients with MAFLD who were using GLP-1 RAs during the study time. Cox proportional hazard models were used to estimate hazard ratios (HRs) and risk differences (RDs), and Kaplan-Meier analysis and log-rank tests were used to compare the outcomes across cohorts.

Results
We identified a total of 844 adult patients without history of cancer who were diagnosed with MAFLD and taking GLP-1 RAs and 16,004 adult patients without history of cancer who were diagnosed with MAFLD and not taking GLP-1 RAs. 830 patients from the GLP-1 RA group (mean [SD] age, 56.60 [11.30] years; 495 [59.64%] female) were matched with 830 individuals who were not on GLP-1 RAs (mean [SD] age, 57.00 [12.60] years; 518 [62.41%] female). There was no difference in the occurrence of hepatobiliary cancer in MAFLD patients regardless of their GLP-1 RAs treatment status. However, a lower risk for gallbladder cancer in GLP-1 RAs users was observed compared to non-GLP-1 RA users (RD, -1.21; 95% CI, -1.94, -0.46, p < 0.01). Lower overall mortality was found in the GLP-1 RA group when compared with the non-GLP-1 RA group (HR, 0.47; 95% CI, 0.328, 0.687, p < 0.01).
Discussion
GLP-1 RA use appears to reduce overall mortality among patients with MAFLD including in patients who develop hepatobiliary malignancies when compared to those not treated with GLP-1 RAs. However, no difference is seen in hepatobiliary cancer incidence among MAFLD patients based GLP-1 RA use status. Further prospective, randomized controlled studies are needed to confirm the mechanism for this clinical finding.