Comparative risk of hepatocellular carcinoma and mortality among initiators of GLP-1 receptor agonists versus other glucose-lowering therapies: a target trial emulation.

INTRODUCTION: Glucagon-like peptide-1 receptor agonists (GLP-1 RA) may influence pathways involved in hepatocarcinogenesis, but evidence regarding their association with hepatocellular carcinoma (HCC) and mortality risk remains heterogeneous. We evaluated the association between GLP-1 RA initiation and HCC incidence compared with commonly used glucose-lowering medications, with secondary exploratory analyses of all-cause mortality following HCC diagnosis. METHODS: We emulated active-comparator, new-user target trials using Patient-Centered Clinical Research Network data from UT Southwestern Medical Center (PCORnet-UTSW; 2010-2025). Adults with T2D initiating GLP-1 RA were compared with initiators of metformin, insulin, sodium-glucose cotransporter-2 inhibitors (SGLT2i), dipeptidyl peptidase-4 inhibitors (DPP4i), and sulfonylureas. The primary outcome was incident HCC; a secondary analysis evaluated all-cause mortality after HCC diagnosis. Five-year risks and risk differences under the intention-to-treat (ITT) approach were estimated using pooled logistic regression with inverse probability of treatment weighting. Per-protocol analyses accounted for treatment adherence using inverse probability of censoring weights. RESULTS: GLP-1 RA initiators had lower five-year HCC risk than comparator groups. Under ITT, risk differences were - 0.37% (95% CI - 0.59 to - 0.08) versus metformin, - 0.52% (95% CI - 0.74 to - 0.35) versus insulin, - 0.37% (95% CI - 0.70 to - 0.06) versus DPP4i, and - 0.39% (95% CI - 0.64 to - 0.13) versus sulfonylureas. Findings were stronger in per-protocol analyses and consistent across subgroups, sensitivity analyses and by individual GLP-1 RA agent. Mortality analyses were limited by small sample sizes but suggested a potential survival benefit. CONCLUSIONS: GLP-1 RA initiation was associated with lower HCC risk versus several comparators. Mortality findings were inconclusive and exploratory, warranting confirmation in larger studies.