Histological efficacy of anti‐diabetic agents in MASH and the mediating role of weight loss: A network meta‐analysis

Background Metabolic dysfunction‐associated steatohepatitis (MASH) is a progressive liver disease with rising global prevalence, closely linked to type 2 diabetes. While several anti‐diabetic agents show promise, a comprehensive analysis comparing their efficacy on biopsy‐confirmed histological outcomes, including dose‐dependent effects and the mediating role of weight loss, remains unexplored. Methods A frequentist random‐effects network meta‐analysis (NMA) was conducted to explore histological efficacy of anti‐diabetic agents in biopsy‐confirmed MASH without cirrhosis. The primary outcome was fibrosis improvement (≥1 stage) without worsening steatohepatitis; the secondary outcome was MASH resolution without fibrosis worsening. Incretin‐based agents were dose‐stratified. Treatment ranking used surface‐under the cumulative‐ranking curve (SUCRA). Meta‐regression investigated the impact of percentage weight loss and baseline covariates on the proportion of individuals achieving histological end‐points. Results Data from five RCTs ( N = 1667) were included. All active treatments, including Dapagliflozin 10 mg, Survodutide (2.4 mg/wk, 4.8–6 mg/wk), Tirzepatide (5 mg/wk, 10–15 mg/wk), and Semaglutide (0.7–1.4 mg/wk, 2.4 or 2.8 mg/wk), improved fibrosis versus placebo ( I 2 = 0%). For MASH resolution, dose‐dependent effects led to significant heterogeneity ( I 2 = 73%), with lower‐dose Semaglutide demonstrating no benefits and Dapagliflozin showing benefits in the F2–F3 subgroup only on sensitivity analysis. Survodutide exhibited the highest ranking (SUCRA = 0.822–0.849), followed by Tirzepatide (SUCRA = 0.622–0.681) and higher‐dose Semaglutide (SUCRA = 0.327) for MASH resolution. Meta‐regression using data from 16 interventions, including placebo arms, showed that weight loss significantly explained heterogeneity in treatment effects on fibrosis improvement ( R 2 = 54.26%) and MASH resolution ( R 2 = 78.16%). Conclusions SGLT2 inhibitor and incretin‐based agents improved fibrosis in MASH, with weight loss being a significant mediator. Targeting multiple incretin pathways, especially involving glucagon receptors, may offer greater MASH resolution. Dose‐dependent effects were more prominent for MASH resolution than fibrosis improvement, indicating potential weight‐loss‐independent anti‐fibrotic pathways.