Network meta-analysis ranks FGF21 analogs and SGLT2 inhibitors highest for MASH fibrosis

Photo by National Cancer Institute / Unsplash

Frontiers in Medicine Published June 4, 2026 Medically reviewed June 4, 2026 DOI ↗ By Dr. Amelia Tan, PhD · Internal Medicine & Chronic Disease

Key Takeaway

FGF21 analogs and SGLT2 inhibitors show highest efficacy for MASH fibrosis improvement in network meta-analysis.

A network meta-analysis of 34 Phase II/III randomized controlled trials evaluated distinct pharmacological mechanisms for metabolic dysfunction-associated steatohepatitis (MASH) in adults with biopsy-proven disease. The primary outcome was fibrosis improvement, defined as at least one stage reduction without MASH worsening.

FGF21 analogs ranked highest for fibrosis improvement with a relative risk of 2.22 (95% CI 1.40-3.54). SGLT2 inhibitors also showed strong efficacy, numerically outperforming the FDA-approved THR-β agonist and GLP-1 receptor agonists. For MASH resolution, GLP-1/GIP dual agonists demonstrated the highest efficacy (RR 5.21), followed by FGF21 analogs (RR 3.52) and SGLT2 inhibitors (RR 2.92).

Safety data were limited, but discontinuation due to adverse events was noted. Cluster analysis suggested SGLT2 inhibitors occupied an optimal zone balancing fibrosis efficacy and tolerability. Sensitivity analyses excluding small studies confirmed the robustness of these rankings.

Limitations include reliance on limited histological data and preliminary evidence from a single pivotal trial for SGLT2 inhibitors. The findings require confirmation in adequately powered Phase 3 trials before clinical adoption.

Study Details

Study typeMeta analysis

EvidenceLevel 1

PublishedJun 2026

View Original Abstract ↓

BackgroundMultiple pharmacological agents targeting distinct pathways (e.g., GLP-1, FGF21, THR-β) show promise for metabolic dysfunction-associated steatohepatitis (MASH). However, while recent meta-analyses have established the efficacy of GLP-1 RAs and FGF21 analogs, they largely predate critical histological trials on SGLT2 inhibitors, leaving the comparative position of this highly accessible oral therapy undefined. We aimed to systematically rank the efficacy and tolerability of distinct pharmacological mechanisms for MASH and characterize their benefit–risk profiles through cluster analysis.MethodsWe searched PubMed, Embase, Web of Science, Scopus, and Cochrane Library for Phase II/III randomized controlled trials (RCTs) involving adults with biopsy-proven MASH. The primary outcome was fibrosis improvement (≥1 stage reduction without MASH worsening). Secondary outcomes included MASH resolution and safety (discontinuation due to adverse events). A frequentist network meta-analysis (NMA) was performed.ResultsWe included 34 studies from 33 unique publications. For fibrosis improvement, FGF21 analogs (RR 2.22, 95% CI 1.40-3.54) and SGLT2 inhibitors (RR 2.27, 95% CI 1.22-4.17) ranked highest. Notably, SGLT2 inhibitors numerically outperformed the FDA-approved THR-β agonist (RR 1.61) and GLP-1 RAs (RR 1.51, 95% CI 1.09-2.10). For MASH resolution, GLP-1/GIP dual agonists (RR 5.21) and FGF21 analogs (RR 3.52) showed the highest efficacy, while SGLT2 inhibitors also yielded significant benefits (RR 2.92, 95% CI 1.18-7.26). In the two-dimensional cluster analysis evaluating the balance between fibrosis efficacy and tolerability, SGLT2 inhibitors numerically occupied the “Optimal Zone,” though this finding is based on limited histological data. Sensitivity analyses excluding small-scale studies confirmed the robustness of these rankings.ConclusionFGF21 analogs (RR 2.22, 95% CI 1.40-3.54) demonstrated the most robust evidence for fibrosis improvement. Preliminary evidence from a single pivotal trial suggests SGLT2 inhibitors (RR 2.27, 95% CI 1.22-4.17) may represent a promising oral option, though this requires confirmation in adequately powered Phase 3 trials. Further adequately powered Phase 3 trials are warranted to validate the magnitude of these histological benefits.Systematic review registrationhttps://www.crd.york.ac.uk/prospero/, identifier CRD420261292638.