Resmetirom and New Agents for Metabolic Dysfunction-associated Steatohepatitis: A Systematic Review of Phase 3 Data on Histologic and Clinical Outcomes

Abstract

Background: Metabolic dysfunction-associated steatohepatitis (MASH), formerly termed nonalcoholic steatohepatitis, can progress to advanced fibrosis, cirrhosis, hepatic decompensation, hepatocellular carcinoma, transplantation, and death. Resmetirom and semaglutide have reached clinical use through accelerated approval based on phase 3 histologic surrogate endpoints, while evidence for liver-related clinical benefit remains immature. Objective: To systematically evaluate phase 3 evidence for resmetirom and other pharmacotherapies in adults with MASH, with separate consideration of histologic efficacy, hard liver-related clinical outcomes, safety, risk of bias, and certainty of evidence. Methods: PubMed/MEDLINE, Europe PMC, ClinicalTrials.gov, Crossref, DOI.org, publisher pages, U.S. Food and Drug Administration records, National Institute of Diabetes and Digestive and Kidney Diseases records, and citation tracking were searched for records dated 1 January 2000 through 30 May 2026. Trial identifiers and linked primary sources were reconciled in a completed open-source verification process. Phase 3 adult interventional studies reporting histologic, clinical, or prespecified supportive safety and non-invasive-test outcomes were eligible. Risk of bias was assessed with RoB 2 or ROBINS-I, and certainty was judged narratively with GRADE domains. Owing to clinical, methodological, and estimand heterogeneity, synthesis followed SWiM principles without meta-analysis. Raw record and duplicate counts from unavailable subscription-database exports were neither inferred nor reported. Results: Twelve phase 3 studies represented in 11 primary reports or official registry-result records met the eligibility criteria: 11 histologic or clinical-efficacy studies and one supportive safety/non-invasive-test study. At week 52, resmetirom achieved MASH resolution without fibrosis worsening in 25.9% and 29.9% of participants receiving 80 and 100 mg, respectively, versus 9.7% with placebo; corresponding fibrosis improvement occurred in 24.2%, 25.9%, and 14.2%. At a planned week-72 interim analysis, semaglutide achieved MASH resolution without fibrosis worsening in 62.9% versus 34.3% and fibrosis improvement without MASH worsening in 36.8% versus 22.4%. Dapagliflozin improved several histologic outcomes in a 154-participant Chinese trial, but certainty was low. Vitamin E improved a historical composite histologic endpoint in adults without diabetes, without significant fibrosis improvement. Obeticholic acid improved fibrosis in precirrhotic disease but did not establish a successful clinical-outcome programme; selonsertib, cenicriviroc, and elafibranor did not establish meaningful phase 3 efficacy. Aramchol data were uncontrolled. No included agent had demonstrated a reduction in hepatic decompensation, transplantation, hepatocellular carcinoma, or mortality by the cut-off date. Conclusions: Resmetirom and semaglutide provide moderate-certainty evidence of improvement in regulatory histologic endpoints in noncirrhotic fibrotic MASH; their approved indications are restricted to F2-F3 disease. Dapagliflozin has promising but low-certainty single-trial evidence. Cross-trial response proportions should not be used to rank therapies. Whether histologic improvements translate into fewer liver-related events or improved survival remains unresolved, particularly in cirrhosis.