Skip to content

GLP-1 and incretin therapy

TL;DR — Semaglutide 2.4 mg weekly is the second drug conditionally approved for MASH. In the phase 3 ESSENCE interim analysis (first 800 of 1,197 randomised patients, MASH with F2–F3, 72 weeks), resolution of steatohepatitis without worsening of fibrosis occurred in 62.9% versus 34.3% placebo (difference 28.7 points, 95% CI 21.1–36.2, p<0.001), and fibrosis reduction without worsening of steatohepatitis in 36.8% versus 22.4% (difference 14.4 points, 7.5–21.3, p<0.001); mean weight change was −10.5% versus −2.0% (Sanyal 2025, PMID 40305708, NCT04822181). The Wegovy formulation received accelerated FDA approval in August 2025 for MASH with F2–F3 fibrosis, with final approval awaiting long-term outcomes (Bansal 2026, PMID 41201884). Note the placebo arm: 34.3% resolution and 22.4% fibrosis improvement, against 9.7% and 14.2% in MAESTRO-NASH — a reminder that cross-trial comparison of these drugs is unsound. The class is broadening fast: tirzepatide (GIP/GLP-1) achieved MASH resolution in 44–62% versus 10% placebo in phase 2 (Loomba 2024, PMID 38856224), survodutide (glucagon/GLP-1) 47–62% versus 14% (Sanyal 2024, PMID 38847460), and glucagon co-agonism appears to add hepatic fat clearance beyond weight loss — efinopegdutide reduced liver fat 72.7% versus semaglutide's 42.3% head-to-head (p<0.001) at similar weight loss (Romero-Gómez 2023, PMID 37355043). The class fails where it matters most: in MASH cirrhosis, semaglutide did not improve fibrosis (11% vs 29% placebo; OR 0.28, 95% CI 0.06–1.24, p=0.087) (Loomba 2023, PMID 36934740).

The trials

Agent (mechanism) Trial n Population MASH resolution, no fibrosis worsening ≥1-stage fibrosis improvement, no MASH worsening
Semaglutide 2.4 mg weekly (GLP-1) ESSENCE phase 3, 72-week interim (PMID 40305708, NCT04822181) 800 of 1,197 biopsy MASH, F2–F3 62.9% vs 34.3% (Δ28.7, 21.1–36.2, p<0.001) 36.8% vs 22.4% (Δ14.4, 7.5–21.3, p<0.001)
Semaglutide 0.4 mg daily (GLP-1) phase 2, 72 weeks (PMID 33185364, NCT02970942) 320 (230 F2–F3) biopsy NASH F1–F3 59% vs 17% placebo (p<0.001); 0.1 mg 40%, 0.2 mg 36% 43% vs 33% placebo, p=0.48 (not significant)
Semaglutide 2.4 mg weekly phase 2 in cirrhosis, 48 weeks (PMID 36934740, NCT03987451) 71 biopsy NASH-related compensated cirrhosis, BMI ≥27 no significant difference (p=0.29) 11% (5/47) vs 29% (7/24); OR 0.28 (0.06–1.24), p=0.087
Liraglutide 1.8 mg daily (GLP-1) LEAN phase 2, 48 weeks (PMID 26608256, NCT01237119) 52 (45 with end-of-treatment biopsy) overweight with NASH 39% (9/23) vs 9% (2/22); RR 4.3 (1.0–17.7), p=0.019 fibrosis progression in 9% vs 36%; RR 0.2 (0.1–1.0), p=0.04
Tirzepatide 5/10/15 mg weekly (GIP/GLP-1) SYNERGY-NASH phase 2, 52 weeks (PMID 38856224, NCT04166773) 190 randomised, 157 evaluable biopsy MASH F2–F3 44% / 56% / 62% vs 10% placebo (Δ34, 46, 53 points; p<0.001 all) 55% / 51% / 51% vs 30% placebo (Δ25, 22, 21 points)
Survodutide 2.4/4.8/6.0 mg weekly (glucagon/GLP-1) phase 2, 48 weeks (PMID 38847460, NCT04771273) 293 biopsy MASH F1–F3 47% / 62% / 43% vs 14% placebo (p<0.001 for quadratic dose-response) 34% / 36% / 34% vs 22%
Pemvidutide 1.2/1.8/2.4 mg weekly (GLP-1/glucagon) phase 1b/2a, 12 weeks (PMID 39002641, NCT05006885) 94 BMI ≥28, liver fat ≥10% by MRI-PDFF — (liver-fat endpoint) —
Efinopegdutide 10 mg weekly (GLP-1/glucagon) vs semaglutide 1 mg phase 2a, 24 weeks, active comparator (PMID 37355043, NCT04944992) 145 NAFLD, liver fat ≥10% — (liver-fat endpoint) —
Retatrutide 1/4/8/12 mg weekly (GIP/GLP-1/glucagon) phase 2a substudy, 24-week liver-fat readout (PMID 38858523, NCT04881760) 98 MASLD with liver fat ≥10% — (liver-fat endpoint) —

Liver fat, where histology was not the endpoint

Agent Relative liver-fat reduction Comparator Notes
Retatrutide 12 mg −82.4% placebo +0.3% (p<0.001) 86% achieved normal liver fat (<5%) at 12 mg vs 0% placebo; 79% at 8 mg, 52% at 4 mg, 27% at 1 mg (PMID 38858523)
Retatrutide 8 mg −81.4% as above reductions significantly related to changes in body weight, abdominal fat and insulin-sensitivity measures
Efinopegdutide 10 mg −72.7% (90% CI 66.8–78.7) semaglutide 1 mg −42.3% (36.5–48.1), p<0.001 weight loss 8.5% vs 7.1%, p=0.085 — i.e. the liver-fat gap was not explained by weight (PMID 37355043)
Pemvidutide 1.8 mg −68.5% (95% CI −84.4 to −52.5) placebo −4.4% 94.4% achieved ≥30% and 72.2% ≥50% reduction; 55.6% normalised (≤5%); weight −4.3%; ALT −13.8 IU/L; cT1 −75.9 ms (PMID 39002641)
Semaglutide (pooled) mean difference −4.50% absolute (95% CI −6.60 to −2.40), I²=95.9% across 9 MR-based trials placebo Mantovani 2025, PMID 40736113

The efinopegdutide head-to-head is the cleanest evidence in this field that glucagon receptor co-agonism reduces liver fat by a mechanism additional to weight loss: glucagon acts directly on hepatocytes to stimulate fatty-acid oxidation and suppress lipogenesis, whereas GLP-1 works chiefly through reduced energy intake. Pemvidutide's authors make the same argument explicitly (PMID 39002641), and it is the design rationale for survodutide and retatrutide. See pathogenesis for the supply-versus-disposal framing.

Meta-analysis

Thirteen phase 2/3 randomised trials (n=1,811), diagnosis by biopsy in 4 and MR-based techniques in 9 (Mantovani 2025, PMID 40736113):

Outcome Pooled effect Heterogeneity
MASH resolution (3 RCTs, F2–F3, regardless of diabetes status) OR 3.48 (95% CI 2.69–4.51) I²=0%
Fibrosis improvement OR 1.79 (95% CI 1.37–2.35) I²=0%
MR-measured liver fat (9 RCTs) mean difference −4.50% (−6.60 to −2.40) I²=95.9%
MASH-related compensated cirrhosis (1 RCT) no benefit for resolution or fibrosis —

The I²=0% for both histological endpoints is unusual and reassuring; the I²=95.9% for liver fat reflects the heterogeneity of agents, doses and durations in that pool.

In the wider network meta-analysis of 29 MASH trials (n=9,324), survodutide (SUCRA 90.87) and tirzepatide (84.70) ranked second and third behind pegozafermin for MASH resolution, and survodutide, tirzepatide and semaglutide all beat placebo for fibrosis regression (Souza 2025, PMID 39903735).

Observational evidence on hard hepatic outcomes

No incretin trial has yet reported liver events. Four large target-trial emulations and propensity-matched cohorts have, and they agree on direction while disagreeing on which agent and which endpoint:

Study Design Comparison Result
Henney 2025, PMID 40980971 three emulated target trials in a >150M-patient EHR network; T2D without prior major adverse liver outcome; 1:1 propensity-matched; 2-year follow-up tirzepatide (n=10,165), semaglutide (n=56,702), liraglutide (n=8,301) vs DPP-4 inhibitor incident MALO: tirzepatide HR 0.53 (0.40–0.71); semaglutide HR 0.81 (0.72–0.90); liraglutide HR 1.04 (0.79–1.36), null. Head-to-head: tirzepatide vs liraglutide HR 0.56 (0.39–0.79); tirzepatide vs semaglutide HR 0.83 (0.63–1.09), ns. Driven by compensated and decompensated cirrhosis rates
Choi 2025, PMID 41250965 target trial emulation, Mass General Brigham 2010–2023, MASLD + T2D with baseline FIB-4 <2.67, 2,238 matched pairs GLP-1RA vs DPP-4i progression to FIB-4 >2.67: 3.25 vs 4.29 per 100 person-years, HR 0.75 (0.65–0.87); consistent in per-protocol (0.80) and landmark analyses and in the BMI <30 subgroup. Composite cirrhosis/decompensation/HCC/transplant: HR 0.98 (0.72–1.34), null
Kuo 2025, PMID 40536520 TriNetX, MASLD, new GLP-1RA prescriptions Dec 2017–Sep 2023, 20,384 matched per arm semaglutide vs other GLP-1RAs composite mortality/MACE/kidney/liver: 31.8 vs 36.6 per 10,000 person-years, aHR 0.86 (0.80–0.93); all-cause mortality aHR 0.68 (0.59–0.80); major adverse liver outcomes aHR 0.79 (0.66–0.94); vs dulaglutide 0.88 (0.81–0.96), vs liraglutide 0.83 (0.71–0.97)
Kanwal 2024, PMID 39283612 see MASLD and type 2 diabetes GLP-1RA vs comparator in MASLD reduced risk of cirrhosis and related complications

Two things are worth reading carefully. First, Henney found lower point estimates for tirzepatide than semaglutide and a null result for liraglutide, while Kuo found semaglutide favourable to pooled other GLP-1RAs and to dulaglutide or liraglutide. These observational comparisons are directionally compatible with tirzepatide and semaglutide outperforming older agents, but they do not establish a stable within-class ranking. Second, Choi's dissociation matters: GLP-1RA use was associated with slower FIB-4 progression (HR 0.75) but not with the composite of cirrhosis, decompensation, HCC or transplant (HR 0.98, 0.72–1.34) in the same matched cohort. Follow-up or endpoint validity could explain the difference. All four studies are observational and remain vulnerable to confounding by indication, adherence and access.

Approval and how the licensed population is identified

Semaglutide (Wegovy) received accelerated FDA approval in August 2025 for MASH with moderate-to-advanced fibrosis (F2–F3), based on the ESSENCE interim results; final approval awaits long-term outcomes. AASLD issued an implementable practice update in November 2025 (Bansal 2026, PMID 41201884):

Element Guidance
Candidates MASH with stage 2–3 fibrosis, identified by non-invasive tests — VCTE 8–15 kPa, MRE 3.1–4.4 kPa, or ELF 9.2–10.5 — rather than biopsy, which is "impractical and unnecessary for most patients"
Grey zone VCTE 15–20 kPa, MRE 4.4–5 kPa, ELF 10.5–11.3: individualised decision, requiring exclusion of cirrhosis by a second confirmatory NIT, or cross-sectional imaging without nodular contour or portal-hypertension signs, or platelets <150,000/mm³
Not approved MASH cirrhosis (VCTE >20 kPa, MRE >5.0 kPa, ELF >11.3, or evidence of portal hypertension); those on semaglutide for another indication with compensated cirrhosis should be monitored carefully
Hepatic safety Favourable in ESSENCE; no discontinuations for liver-enzyme elevation; routine hepatic panels only as clinically indicated
Monitored risks acute kidney injury from dehydration, symptomatic gallbladder disease, pancreatitis, thyroid C-cell tumours, retinopathy progression, lean mass loss
Response markers No NIT reliably predicts histological response individually, but 72-week reductions suggestive of benefit: ALT ≥17 U/L or ≥20%; VCTE LSM ≥30%; MRE LSM ≥20%; ELF ≥0.5
Combination Resmetirom plus semaglutide 2.4 mg has not been studied
Concomitant Lifestyle modification remains the cornerstone alongside semaglutide

This is the most operationally specific guidance in the field, and it is worth noting what it concedes: the drug is licensed on a histological definition, will be prescribed on non-invasive thresholds that were not the trial entry criteria, and no non-invasive marker predicts individual response.

Adverse effects

Gastrointestinal events dominate and are dose-related. In the semaglutide phase 2, nausea 42% versus 11% placebo, constipation 22% versus 12%, vomiting 15% versus 2% (PMID 33185364). In survodutide, nausea 66% versus 23%, diarrhoea 49% versus 23%, vomiting 41% versus 4%, with serious adverse events 8% versus 7% (PMID 38847460). In LEAN, gastrointestinal disorders affected 81% versus 65%, including diarrhoea 38% versus 19%, constipation 27% versus 0% and appetite loss 31% versus 8% (PMID 26608256).

One signal deserves recording rather than dismissal: in the semaglutide phase 2, malignant neoplasms were reported in 3 semaglutide patients (1%) and none on placebo, and neoplasms of any kind in 15% versus 8%, with no pattern by organ (PMID 33185364). The trial authors report it without interpretation; it has not been replicated as a signal in later trials but has also not been formally refuted.

Lean-mass loss is flagged by AASLD as a monitored risk (PMID 41201884), which matters given the association between sarcopenia and MASLD outcomes (cardiovascular and extrahepatic outcomes). It has now been quantified across the class. In a network meta-analysis of 22 RCTs (n=2,258) with body-composition measurement, GLP-1 receptor agonists reduced total weight by 3.55 kg (95% CI −4.81 to −2.29), fat mass by 2.95 kg (−4.11 to −1.79) and lean mass by 0.86 kg (−1.30 to −0.42) — lean mass comprising approximately 25% of the total weight lost, although relative lean mass (percentage of body weight) was unchanged (Karakasis 2025, PMID 39719170). Tirzepatide 15 mg and semaglutide 2.4 mg were the most effective for weight and fat loss and among the least effective at preserving lean mass, while liraglutide was the only agent achieving significant weight reduction without significant lean-mass loss. Cross-trial comparisons cannot establish that greater hepatic efficacy is purchased with greater muscle loss; no MASH trial has reported histological outcomes stratified by lean-mass change.

The cirrhosis failure

The semaglutide cirrhosis trial is the most informative negative result in the class. In 71 patients with biopsy-confirmed NASH-related compensated cirrhosis (mean age 59.5, mean BMI 34.9, 75% with diabetes, 69% female), 48 weeks of semaglutide 2.4 mg produced fibrosis improvement in 11% versus 29% on placebo (OR 0.28, 95% CI 0.06–1.24, p=0.087), with no difference in NASH resolution (p=0.29). There were no decompensating events or deaths and hepatic and renal function remained stable (Loomba 2023, PMID 36934740). Two readings are possible and neither is excluded by the data: fibrosis at F4 may be less reversible than at F2–F3, or the trial (n=71, 48 weeks) may simply have been too small and short. Efruxifermin's phase 2b in compensated MASH cirrhosis found the same pattern — no significant fibrosis reduction at 36 weeks (19% vs 13%, p=0.52), with a larger difference emerging by week 96 (29% vs 11%; difference 16 points, 95% CI 2–30) (Noureddin 2025, PMID 40341827) — which supports the "too short" reading. See cirrhosis and decompensation.

Open questions

  • Do these drugs prevent hepatic clinical events? ESSENCE runs to 240 weeks; the published analysis is the 72-week histological interim in the first 800 patients (PMID 40305708). Accelerated approval is explicitly conditional on those outcomes (PMID 41201884). The same surrogate-validity problem applies as for resmetirom.
  • How much of the effect is weight loss, and how much is direct? Semaglutide's histological benefit accompanies −10.5% weight (PMID 40305708), close to the lifestyle threshold at which 90% of patients achieved NASH resolution (lifestyle and weight loss). Efinopegdutide's superiority to semaglutide on liver fat at equivalent weight loss (PMID 37355043) shows the two are separable, but no trial has decomposed the histological effect.
  • Why did the phase 2 semaglutide trial miss fibrosis while the phase 3 hit it? 43% versus 33%, p=0.48 at 72 weeks in phase 2 (PMID 33185364) versus 36.8% versus 22.4%, p<0.001 in phase 3 (PMID 40305708). Sample size (230 vs 800 with F2–F3), dosing regimen (0.4 mg daily vs 2.4 mg weekly) and placebo response all differ. The phase 2 placebo fibrosis-improvement rate of 33% is remarkable in itself.
  • Is cirrhosis unreachable, or just slow? One negative GLP-1 trial at 48 weeks (PMID 36934740) and one FGF21 trial negative at 36 weeks but positive-trending at 96 weeks (PMID 40341827). No adequately powered trial in compensated MASH cirrhosis has run to a hard endpoint.
  • Does combining resmetirom with semaglutide add? AASLD states plainly that it has not been studied (PMID 41201884), while a MAESTRO-NASH secondary analysis found no interference from background GLP-1 therapy (resmetirom and thyromimetics).
  • Is the neoplasm imbalance real? 15% versus 8% any neoplasm in the semaglutide phase 2 (PMID 33185364), unexplained and, so far as retrieved searches show, not formally addressed in a pooled safety analysis. Query run 2026-09-02: (NASH OR MASH OR NAFLD OR MASLD) AND (semaglutide OR liraglutide OR tirzepatide OR survodutide OR retatrutide OR efinopegdutide OR "GLP-1") — 1,231 records; no dedicated MASH-population neoplasm safety analysis retrieved.
  • Does lean-mass loss offset the hepatic benefit? Flagged as a monitored risk (PMID 41201884) and quantified at ~25% of total weight lost across the included body-composition trials (PMID 39719170); no MASH trial has reported body-composition-stratified histological outcomes.
  • Is the within-class ordering real, or confounding? Tirzepatide > semaglutide > liraglutide for major adverse liver outcomes appears in independent EHR networks (PMIDs: 40980971, 40536520), with liraglutide null despite being the first agent with positive histology (PMID 26608256). Prescribing of newer agents tracks insurance status, adherence and secular trends in MASLD care; no randomised head-to-head incretin comparison with hepatic endpoints exists.
  • Does slowing FIB-4 progression prevent liver events? GLP-1RA use reduced progression to FIB-4 >2.67 (HR 0.75) while leaving the cirrhosis/decompensation/HCC/transplant composite unchanged (HR 0.98) in the same matched cohort (PMID 41250965). Either the follow-up is too short or the non-invasive surrogate does not transfer treatment effects — and the two are not currently distinguishable.

References

  1. Sanyal AJ, Newsome PN, Kliers I, et al. Phase 3 Trial of Semaglutide in Metabolic Dysfunction-Associated Steatohepatitis. N Engl J Med. 2025;392(21):2089-2099. PMID 40305708
  2. Newsome PN, Buchholtz K, Cusi K, et al. A Placebo-Controlled Trial of Subcutaneous Semaglutide in Nonalcoholic Steatohepatitis. N Engl J Med. 2021;384(12):1113-1124. PMID 33185364
  3. Loomba R, Abdelmalek MF, Armstrong MJ, et al. Semaglutide 2·4 mg once weekly in patients with non-alcoholic steatohepatitis-related cirrhosis: a randomised, placebo-controlled phase 2 trial. Lancet Gastroenterol Hepatol. 2023;8(6):511-522. PMID 36934740
  4. Armstrong MJ, Gaunt P, Aithal GP, et al. Liraglutide safety and efficacy in patients with non-alcoholic steatohepatitis (LEAN): a multicentre, double-blind, randomised, placebo-controlled phase 2 study. Lancet. 2016;387(10019):679-690. PMID 26608256
  5. Loomba R, Hartman ML, Lawitz EJ, et al. Tirzepatide for Metabolic Dysfunction-Associated Steatohepatitis with Liver Fibrosis. N Engl J Med. 2024;391(4):299-310. PMID 38856224
  6. Sanyal AJ, Bedossa P, Fraessdorf M, et al. A Phase 2 Randomized Trial of Survodutide in MASH and Fibrosis. N Engl J Med. 2024;391(4):311-319. PMID 38847460
  7. Harrison SA, Browne SK, Suschak JJ, et al. Effect of pemvidutide, a GLP-1/glucagon dual receptor agonist, on MASLD: A randomized, double-blind, placebo-controlled study. J Hepatol. 2025;82(1):7-17. PMID 39002641
  8. Romero-Gómez M, Lawitz E, Shankar RR, et al. A phase IIa active-comparator-controlled study to evaluate the efficacy and safety of efinopegdutide in patients with non-alcoholic fatty liver disease. J Hepatol. 2023;79(4):888-897. PMID 37355043
  9. Sanyal AJ, Kaplan LM, Frias JP, et al. Triple hormone receptor agonist retatrutide for metabolic dysfunction-associated steatotic liver disease: a randomized phase 2a trial. Nat Med. 2024;30(7):2037-2048. PMID 38858523
  10. Mantovani A, Morandin R, Fiorio V, et al. Glucagon-Like Peptide-1 Receptor Agonists Improve MASH and Liver Fibrosis: A Meta-Analysis of Randomised Controlled Trials. Liver Int. 2025;45(9):e70256. PMID 40736113
  11. Bansal MB, Patton H, Morgan TR, et al. Semaglutide therapy for metabolic dysfunction-associated steatohepatitis: November 2025 updates to AASLD Practice Guidance. Hepatology. 2026;83(5):1326-1340. PMID 41201884
  12. Souza M, Al-Sharif L, Antunes VLJ, et al. Comparison of pharmacological therapies in metabolic dysfunction-associated steatohepatitis for fibrosis regression and MASH resolution: Systematic review and network meta-analysis. Hepatology. 2025;82(6):1523-1533. PMID 39903735
  13. Noureddin M, Rinella ME, Chalasani NP, et al. Efruxifermin in Compensated Liver Cirrhosis Caused by MASH. N Engl J Med. 2025;392(24):2413-2424. PMID 40341827
  14. Newsome PN, Ambery P. Incretins (GLP-1 receptor agonists and dual/triple agonists) and the liver. J Hepatol. 2023;79(6):1557-1565. PMID 37562748
  15. Henney AE, Riley DR, Anson M, et al. Target Trial Emulations of GLP-1 and Dual GLP-1/GIP Agonists to Reduce Major Adverse Liver Outcomes in Type 2 Diabetes. Liver Int. 2025;45(10):e70367. PMID 40980971
  16. Choi J, Kamath T, Nguyen VH, et al. GLP-1RA and Liver Fibrosis Progression in MASLD and Type 2 Diabetes: Target Trial Emulation Using Propensity Score Matching. Liver Int. 2025;45(12):e70447. PMID 41250965
  17. Kuo CC, Li CH, Chuang MH, et al. Semaglutide versus other GLP-1 receptor agonists in patients with MASLD. Hepatol Commun. 2025;9(7):e0747. PMID 40536520
  18. Karakasis P, Patoulias D, Fragakis N, et al. Effect of glucagon-like peptide-1 receptor agonists and co-agonists on body composition: Systematic review and network meta-analysis. Metabolism. 2025;164:156113. PMID 39719170
  19. Kanwal F, Kramer JR, Li L, et al. GLP-1 Receptor Agonists and Risk for Cirrhosis and Related Complications in Patients With Metabolic Dysfunction-Associated Steatotic Liver Disease. JAMA Intern Med. 2024;184(11):1314-1323. PMID 39283612