A network meta-analysis found GLP-1 RAs and pioglitazone most effective for MASH histological improvement, with weight loss mediating a significant portion of the fibrosis benefit.
GLP-1 RAs top rankedMost effective anti-diabetic agents for biopsy-confirmed MASH fibrosis improvement
What the researchers found
GLP-1 RAs and pioglitazone showed the greatest histological improvement in MASH, with weight loss mediating a significant portion of fibrosis benefit.
Why it matters
MASH is becoming a leading cause of liver transplantation. Identifying which diabetes drugs best improve liver histology could prevent progression to cirrhosis in millions.
How the study worked
Frequentist random-effects network meta-analysis of biopsy-confirmed MASH trials; primary outcome: fibrosis improvement; dose-response and weight loss mediation analyzed.
What this study cannot tell us
Network meta-analysis relies on indirect comparisons; included trials had varying durations and populations; biopsy-based outcomes have sampling variability.
How to read the evidence
Network meta-analysis of biopsy-confirmed trials — strong methodology for comparing therapies without head-to-head data.
When this study was published
Published 2026 in Diabetes, Obesity & Metabolism.
The bigger picture
This establishes GLP-1 RAs as leading MASH therapies, potentially rivaling or complementing dedicated MASH drugs like resmetirom.
Questions still open
- Should GLP-1 RAs be used specifically for MASH prevention in diabetic patients?
- How does tirzepatide compare to semaglutide for MASH fibrosis improvement?
Common questions
Can diabetes drugs treat fatty liver disease?
Is the liver benefit from the drugs themselves or from weight loss?
Read the original research
Histological efficacy of anti-diabetic agents in MASH and the mediating role of weight loss: A network meta-analysis.
Diabetes, obesity & metabolism, 28(1), 287-295
Citation
Banerjee, Mainak; Pal, Rimesh; Pal, Sandip. (2026). Histological efficacy of anti-diabetic agents in MASH and the mediating role of weight loss: A network meta-analysis.. Diabetes, obesity & metabolism, 28(1), 287-295. https://doi.org/10.1111/dom.70187