Therapeutic efficacy, biomarker signatures, and translatability of semaglutide in the liver biopsy-confirmed GAN DIO-MASH mouse model.

Abstract

Glucagon-like peptide-1 (GLP-1) receptor agonists have emerged as promising therapeutic candidates for metabolic dysfunction-associated steatohepatitis (MASH). Importantly, semaglutide was recently approved as the first GLP-1-based treatment for people with MASH with moderate-to-severe fibrosis. Translational models that recapitulate human MASH are critical for guiding early-stage drug discovery, enabling rigorous efficacy evaluation, and facilitating the progression of drug candidates into clinical development. In this study, we investigated the efficacy of semaglutide across an extensive series of experiments in the liver biopsy-confirmed GAN diet-induced obese MASH (GAN DIO-MASH) mouse model, benchmarking outcomes against those from pivotal clinical trials of semaglutide in MASH. Treatment outcomes in the GAN DIO-MASH mouse closely mirrored clinical findings, particularly for hepatic steatosis and inflammation endpoints. Longer semaglutide treatment durations (≥16 wk) led to pronounced and consistent improvements in quantitative fibrosis histology across studies. In comparison, the response rate for fibrosis stage improvement with semaglutide was modest and largely independent of treatment duration. Notably, pooling data from long-term intervention studies revealed more favorable effects on fibrosis stage. Importantly, the GAN DIO-MASH mouse recapitulated many human MASH-associated changes in circulating proteins and semaglutide-responsive biomarkers. Collectively, these results support the therapeutic effects of semaglutide in MASH and underscore the reproducibility and clinical translatability of multiple disease-relevant features of the GAN DIO-MASH mouse model, highlighting its applicability as a robust platform for preclinical drug development.Translational preclinical models that faithfully recapitulate human MASH are essential for guiding early-stage drug discovery by enabling rigorous efficacy assessment. In this study, we demonstrate that the biopsy-confirmed GAN DIO-MASH mouse model replicates semaglutide treatment responses across metabolic, biochemical, and histological endpoints, as well as key biomarker signatures observed in pivotal clinical trials in people with MASH. These findings strongly support the model's clinical translatability to facilitate efficient progression of therapeutic candidates into clinical development.