Glucagon-like peptide-1 receptor agonist use is associated with a lower risk of major adverse liver-related outcomes: a meta-analysis of observational cohort studies

Abstract

Background Glucagon-like peptide-1 receptor agonists (GLP-1RAs) have shown promising effects on liver histology in phase 2 trials enrolling patients with metabolic dysfunction-associated steatotic liver disease. However, the impact of GLP-1RAs on the long-term risk of major adverse liver-related outcomes (MALOs) remains uncertain. Objective We performed a meta-analysis of observational cohort studies to quantify the magnitude and direction of the association between GLP-1RA use and MALOs in people with type 2 diabetes (T2D). Design We systematically searched eligible cohort studies comparing GLP-1RA new users versus users of other glucose-lowering medications. The primary outcome was the cumulative incidence rates of MALOs. Secondary outcomes included hepatic decompensation events, hepatocellular carcinoma (HCC) and liver-related mortality. Random-effects models were used to calculate incidence rate ratios (IRRs). Results 11 retrospective cohort studies with aggregate data on 1 467 220 patients with T2D (647 903 GLP-1RA new users, 819 317 non-users) were included. GLP-1RA use was significantly associated with a lower risk of MALOs (IRR 0.71, 95% CI 0.57 to 0.88) and hepatic decompensation (IRR 0.70, 95% CI 0.52 to 0.94). Association with reduced risk of HCC was also observed (IRR 0.82, 95% CI 0.61 to 1.11). Compared with other antidiabetic medications, GLP-1RAs showed superior effectiveness versus SGLT2 inhibitors in preventing MALOs (IRR 0.93, 95% CI 0.87 to 0.99), versus DPP-4 inhibitors in preventing hepatic decompensation (IRR 0.74, 95% CI 0.66 to 0.83) and versus insulin therapy in preventing HCC (IRR 0.32, 95% CI 0.13 to 0.80). Conclusions GLP-1RA use is associated with a lower risk of liver-related complications and hepatic decompensation in people with T2D. These findings suggest a role of GLP-1RAs in preventing liver-related complications beyond their beneficial cardiometabolic effects. Data are available on reasonable request. All supporting data of the meta-analysis are available within the article (and in the online-only supplementary material).