A meta-analysis of 13 RCTs found GLP-1 receptor agonists were 3.5 times more likely to achieve MASH resolution and 1.8 times more likely to improve liver fibrosis — but did not help patients who had already progressed to cirrhosis.
OR 3.48 for MASH resolutionGLP-1 receptor agonists were 3.5 times more likely to achieve MASH resolution without worsening fibrosis (I²=0%), with fibrosis improvement also significant at 1.8x — but cirrhosis patients showed no benefit.
What the researchers found
Across 13 phase 2/3 RCTs with 1,811 participants:
- MASH resolution without worsening fibrosis (3 RCTs, biopsy-confirmed): OR 3.48 (95% CI: 2.69-4.51, I²=0%) — 3.5 times more likely with GLP-1RAs
- Fibrosis improvement without worsening MASH: OR 1.79 (95% CI: 1.37-2.35, I²=0%) — 1.8 times more likely
- Liver fat reduction by MRI (9 RCTs): -4.50% (95% CI: -6.60 to -2.40, I²=95.9%)
- Semaglutide 2.4 mg/week was the most studied and effective agent
- Critical exception: In 1 RCT of MASH-related compensated cirrhosis, semaglutide did NOT achieve MASH resolution or fibrosis improvement vs. placebo
Why it matters
This meta-analysis provides stronger evidence for GLP-1 drugs' liver benefits than previous analyses — critically, it shows fibrosis improvement (OR 1.79), which the earlier semaglutide-only meta-analysis did not find significant. The zero heterogeneity (I²=0%) for both histological outcomes increases confidence. However, the cirrhosis finding is sobering: once the liver has scarred irreversibly, GLP-1 drugs may be too late. This reinforces the importance of early intervention — using GLP-1 drugs before liver disease progresses to its most advanced stage.
How the study worked
Systematic search of three electronic databases through April 2025 identified 13 phase 2 or 3 RCTs examining GLP-1RAs in MASLD/MASH patients. MASLD/MASH was diagnosed by liver biopsy (4 trials) or MRI-based techniques (9 trials). Primary outcomes were MASH resolution without fibrosis worsening and fibrosis improvement without MASH worsening (biopsy-confirmed in relevant trials). Secondary outcome was MRI-measured liver fat reduction. Random-effects meta-analysis was used given the potential for between-study variability.
What this study cannot tell us
The meta-analysis includes only 1,811 participants across 13 trials — relatively small for a meta-analysis. Only 4 trials used liver biopsy (the gold standard), while 9 used MRI. The liver fat reduction outcome showed very high heterogeneity (I²=95.9%), likely due to different MRI techniques and patient populations. The cirrhosis finding is based on a single RCT. Semaglutide 2.4 mg dominated the evidence base, so conclusions may not generalize equally to other GLP-1RAs. Long-term effects on liver-related clinical events (liver failure, transplant, death) were not assessed.
How to read the evidence
This is a PRISMA-compliant meta-analysis of 13 phase 2/3 RCTs with histologically confirmed outcomes for the key endpoints. The zero heterogeneity for biopsy-confirmed outcomes is a notable strength. The relatively small total sample (1,811) and dominance of semaglutide in the evidence base are limitations.
When this study was published
Published in 2025 with literature through April 2025, this is among the most current meta-analyses of GLP-1 drugs for liver disease, published in Liver International — a leading hepatology journal.
The bigger picture
This meta-analysis is complementary to the earlier semaglutide-MASH analysis (RPEP-10843) in this processing batch, which found semaglutide doubled MASH resolution but didn't significantly improve fibrosis. The difference likely reflects study selection: this analysis focused on MASH-specific trials with biopsy-confirmed outcomes, while the other included broader metabolic trials. Together, they paint a nuanced picture: GLP-1 drugs clearly help fatty liver inflammation and may improve fibrosis when studied with proper histological endpoints — but they cannot reverse established cirrhosis.
Questions still open
- At what stage of fibrosis does GLP-1 therapy become ineffective — is there a point of no return before cirrhosis?
- Would combining GLP-1 drugs with fibrosis-targeted agents (like resmetirom) be effective even in cirrhosis patients?
- Do dual or triple peptide agonists (tirzepatide, retatrutide) show even stronger fibrosis improvement than semaglutide alone?
Common questions
Can GLP-1 drugs reverse liver scarring from fatty liver disease?
How is this different from the other semaglutide liver study?
Read the original research
Glucagon-Like Peptide-1 Receptor Agonists Improve MASH and Liver Fibrosis: A Meta-Analysis of Randomised Controlled Trials.
Liver international : official journal of the International Association for the Study of the Liver, 45(9), e70256
Citation
Mantovani, Alessandro; Morandin, Riccardo; Fiorio, Veronica; Lando, Maria Giovanna; Stefan, Norbert; Tilg, Herbert; Byrne, Christopher D; Targher, Giovanni. (2025). Glucagon-Like Peptide-1 Receptor Agonists Improve MASH and Liver Fibrosis: A Meta-Analysis of Randomised Controlled Trials.. Liver international : official journal of the International Association for the Study of the Liver, 45(9), e70256. https://doi.org/10.1111/liv.70256