# GLP-1s and Fatty Liver: What the Phase 3 Trial Showed

By Nora Bissett, Pricing Editor (https://glpwatchdog.org/authors/nora-bissett) | Published: 2026-08-19 | Updated: 2026-11-01

Canonical page: https://glpwatchdog.org/research/glp1-and-fatty-liver-mash

> Semaglutide resolved steatohepatitis in 62.9% of patients, against 34.3% on placebo. The difference is 28.7 points. In the one trial in compensated cirrhosis, it produced neither resolution nor fibrosis improvement.

## Key points

- ESSENCE: steatohepatitis resolved in 62.9% of the semaglutide arm; placebo reached 34.3%.
- A THIRD of the placebo arm resolved, so the meaningful number is the 28.7-point difference.
- In MASH-related compensated cirrhosis, the one trial found neither resolution nor fibrosis improvement against placebo.
- It is an interim result: week 72 of a 240-week trial, first 800 of 1,197 patients.

[Fatty liver disease](https://glpwatchdog.org/conditions/fatty-liver-disease) is the condition most likely to be sitting underneath a GLP-1 prescription without anyone having mentioned it. In 2025 a phase 3 trial reported that semaglutide resolved steatohepatitis without worsening fibrosis in **62.9%** of patients.[1] That figure has been quoted everywhere. The one quoted far less often is the placebo arm: **34.3%**.

## What the trial found

ESSENCE randomized 1,197 patients with biopsy-confirmed MASH and fibrosis at stage 2 or 3, two to one, to weekly semaglutide 2.4 mg or placebo. It is designed to run 240 weeks. What has been published is a planned interim analysis at week 72, covering the first 800 patients.[1]

*ESSENCE part 1, at week 72. Figures from the published report. The difference column is the comparison that matters.*

| Endpoint | Semaglutide | Placebo | Difference |
| --- | --- | --- | --- |
| Steatohepatitis resolved, fibrosis not worse | 62.9% | 34.3% | 28.7 points (95% CI 21.1–36.2) |
| Fibrosis reduced, steatohepatitis not worse | 36.8% | 22.4% | 14.4 points (95% CI 7.5–21.3) |
| Both together | 32.7% | 16.1% | 16.5 points (95% CI 10.2–22.8) |
| Mean body weight change | −10.5% | −2.0% | −8.5 points |

> **Look at the placebo column.** A third of people assigned to placebo had their steatohepatitis resolve. That is what happens when people with liver disease enter a study, get monitored, and change how they eat and drink. It is also why a single-arm number is close to meaningless here, and why the headline figure should be 28.7 points rather than 62.9%.

## What FDA did with it

FDA acted on this interim result. On August 15, 2025 it approved a new Wegovy indication under its accelerated pathway.[4] The Wegovy injection now carries a label for adults with MASH who have moderate to advanced fibrosis (consistent with stages F2 to F3) but not cirrhosis. Accelerated approval means the license rests on biopsy improvement rather than on liver outcomes, and the label says continued approval may be contingent on a confirmatory trial showing clinical benefit.[3] Heart outcomes are a separate record: an analysis of the SELECT trial found semaglutide cut cardiovascular events in people whose blood tests suggested fibrosis, though [the largest of its three figures was the one that missed significance](https://glpwatchdog.org/research/the-biggest-number-is-the-weakest).

Two limits are written into it. The indication belongs to the injection, not to the Wegovy tablets. And it excludes cirrhosis, which matches the evidence set out below.[3] No tirzepatide product is labeled for MASH.[5] Survodutide is not approved for anything. That is why our [Wegovy for fatty liver board](https://glpwatchdog.org/best/semaglutide-providers/for-fatty-liver) lists only sellers of the Wegovy pen or injection.

## The wider evidence agrees

A 2025 meta-analysis in Liver International pooled 13 phase 2 and phase 3 randomized trials covering 1,811 participants. Among people with MASH and moderate-to-advanced fibrosis, GLP-1 drugs (semaglutide 2.4 mg especially) beat placebo for resolution, with a pooled odds ratio of 3.48, and for fibrosis improvement, at 1.79. Neither estimate showed measurable heterogeneity between trials.[2]

Liver fat measured by MRI fell by a pooled 4.50 percentage points across nine trials, although that estimate varied enormously between studies.[2] Consistency across independent trials is the strongest form this kind of evidence takes, and on the resolution and fibrosis endpoints it is there.

## Where it stops working

This matters most to the people with the most at stake, and it rarely makes the coverage. In the same meta-analysis, among people whose MASH had already progressed to compensated cirrhosis, semaglutide produced **neither resolution nor fibrosis improvement** compared with placebo.[2]

> The benefit shows up before cirrhosis. In compensated cirrhosis, the randomized results so far have not shown semaglutide reversing fibrosis.

That finding rests on one small randomized trial, and it is no longer the only one. A phase 2 trial in 457 people with compensated cirrhosis, NCT04971785, posted results on ClinicalTrials.gov in November 2025. At 72 weeks, fibrosis improved without worsening steatohepatitis in 13.7% on semaglutide plus cilofexor and firsocostat against 8.3% on placebo, which missed significance (P = .23). Semaglutide alone reached 15.6%, but the posted results do not test it against placebo. Steatohepatitis did resolve more often on the combination, 57.3% against 22.4% among those with data, a secondary result.[6] So the cirrhosis evidence is a signal rather than a settled answer, and it may change. But it points the same way as most of hepatology: fibrosis becomes progressively harder to reverse the further it has gone, which is the case for acting early.

## What this does and does not change for a reader

- **If you are on a GLP-1 for weight, this is a plausible bonus and not a treatment plan.** MASH is diagnosed [by biopsy or by imaging](https://glpwatchdog.org/research/a-trial-about-the-measuring-instrument), and nobody can tell you what your liver is doing from your weight.
- **The trial population was specific:** biopsy-confirmed MASH at fibrosis stage 2 or 3. Results in that group do not automatically describe simple fatty liver without inflammation, which is a different and much more common situation.
- **Weight loss and liver improvement traveled together** here, with a mean loss of 10.5% against 2.0%.[1] Untangling how much of the liver benefit is the drug and how much is the weight is not something this trial settles, and [pooling six biopsy trials](https://glpwatchdog.org/research/the-line-is-drawn-by-surgery) does not settle it either, because the link there is driven mostly by bariatric surgery arms.
- **It is an interim result.** ESSENCE runs to 240 weeks; what has reported is week 72 in the first 800 of 1,197 patients.[1] Long-term outcomes are still an open question.

> One finding cuts against the marketing. The groups showed no significant difference in mean change on bodily pain scores.[1] A trial can move a liver biopsy substantially without moving how someone feels.

If liver disease is part of why a GLP-1 was suggested to you, that is a conversation for the clinician holding your imaging and bloodwork. Whether these drugs can themselves injure the liver is a different and much rarer problem, set out in [GLP-1 liver damage and liver enzymes](https://glpwatchdog.org/research/glp1-liver-damage-and-liver-enzymes). Our [evidence-duration piece](https://glpwatchdog.org/research/how-long-is-the-evidence-actually) covers how far the randomized record extends, and the [trial tracker](https://glpwatchdog.org/research/trials) follows what is still running, including newer drugs such as [survodutide](https://glpwatchdog.org/research/survodutide-what-the-evidence-shows), which pairs GLP-1 with glucagon and is in phase 3 for MASH, and [pemvidutide](https://glpwatchdog.org/research/pemvidutide-what-the-evidence-shows), which uses the same pairing.

## Frequently asked questions

### Does semaglutide treat fatty liver disease?

In a phase 3 trial of people with biopsy-confirmed MASH and fibrosis at stage 2 or 3, steatohepatitis resolved without worsening fibrosis in 62.9% on semaglutide against 34.3% on placebo, a difference of 28.7 percentage points. Since August 2025 the Wegovy injection has carried an accelerated-approval indication for noncirrhotic MASH with stage F2 to F3 fibrosis. That indication requires the diagnosis. It does not cover simple fatty liver without inflammation, and it does not cover cirrhosis.

### Why does the placebo group matter so much here?

Because a third of them got better. Entering a trial means monitoring, dietary change and reduced drinking, and those move liver disease on their own. Quoting 62.9% without 34.3% makes the drug look about twice as effective as the controlled comparison supports.

### Does it help if I already have cirrhosis?

The available evidence says no. In a 2025 meta-analysis, among people with MASH-related compensated cirrhosis, semaglutide produced neither resolution nor fibrosis improvement against placebo. That rests on a single small trial. A larger phase 2 trial (NCT04971785, 457 people) posted results in November 2025: at 72 weeks, fibrosis improved in 13.7% on semaglutide plus two other drugs and 8.3% on placebo, which was not significant. Semaglutide alone reached 15.6%, but the posted results do not test it against placebo. So this points the same way as the rest of hepatology, though it is not yet a final answer.

### Is the fatty liver benefit just from losing weight?

The trial does not separate them. Weight fell by a mean 10.5% on semaglutide against 2.0% on placebo, and the liver improvements traveled alongside that. How much is the drug acting on the liver and how much is the weight loss is an open question.

### How solid is this result?

Reasonably solid and not final. It is a planned interim analysis at week 72 covering the first 800 of 1,197 patients in a trial designed to run 240 weeks, and a separate meta-analysis of 13 randomized trials in 1,811 people points the same way with no heterogeneity between studies.

## References

1. Sanyal AJ, Newsome PN, Kliers I, et al. Phase 3 Trial of Semaglutide in Metabolic Dysfunction-Associated Steatohepatitis. New England Journal of Medicine. 2025. PMID: 40305708. https://pubmed.ncbi.nlm.nih.gov/40305708/
2. 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 International. 2025. PMID: 40736113. https://pubmed.ncbi.nlm.nih.gov/40736113/
3. Novo Nordisk Pharmaceutical Industries, LP WEGOVY (semaglutide) injection and tablets — prescribing information, SPL version 19 (section 1, Indications and Usage). DailyMed, U.S. National Library of Medicine. 2026. https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=ee06186f-2aa3-4990-a760-757579d8f77b
4. U.S. Food and Drug Administration Drugs@FDA: WEGOVY (semaglutide), NDA 215256 — efficacy supplement 024 approved August 15, 2025. FDA Drugs@FDA database. 2025. https://www.accessdata.fda.gov/scripts/cder/daf/index.cfm?event=overview.process&ApplNo=215256
5. Eli Lilly and Company ZEPBOUND (tirzepatide) injection, for subcutaneous use — prescribing information, SPL version 40 (section 1, Indications and Usage). DailyMed, U.S. National Library of Medicine. 2026. https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=487cd7e7-434c-4925-99fa-aa80b1cc776b
6. Gilead Sciences Study of Semaglutide, and Cilofexor/Firsocostat, Alone and in Combination, in Adults With Cirrhosis Due to Nonalcoholic Steatohepatitis (NASH), NCT04971785, posted results. ClinicalTrials.gov. 2025. https://clinicaltrials.gov/study/NCT04971785

---

Cite as: GLP Watchdog, Nora Bissett, "GLP-1s and Fatty Liver: What the Phase 3 Trial Showed", published 2026-08-19, updated 2026-11-01. https://glpwatchdog.org/research/glp1-and-fatty-liver-mash

Canonical HTML page: https://glpwatchdog.org/research/glp1-and-fatty-liver-mash

Medical disclaimer: Nothing here is medical advice. We report on this subject; we do not treat anyone. Whether to begin a medication, supplement, or treatment, whether to adjust it, and whether to stop are questions for a clinician licensed to answer them, who knows your history.
