Skip to content
GLP Loss
← Research
Evidence

GLP-1 and Fatty Liver: What the MASH Biopsies Showed

ESSENCE resolved steatohepatitis in 62.9% of patients against 34.3% on placebo. The earlier trial resolved it in 59% against 17% and missed its fibrosis endpoint at P = 0.48 — two questions the same drug answered differently.

Owen Castellanos9 min read
ESSENCE: liver biopsy at week 72Patients meeting each histological endpointResolution, semaglutide62.9%Resolution, placebo34.3%Fibrosis, semaglutide36.8%Fibrosis, placebo22.4%Resolution gap: 28.7 points, 95% CI 21.1 to 36.2.Fibrosis gap: 14.4 points, 95% CI 7.5 to 21.3.Interim read of the first 800 of 1,197 randomized.

Fatty liver is the condition most likely to be discovered by accident on the way to a weight-loss prescription: an ultrasound ordered for something else, a raised ALT on a routine panel, a note that the liver looks bright. The disease behind those findings now has a drug in this class licensed against it, which is a genuine first and a much narrower claim than it sounds. What the trials measured was tissue under a microscope, and the tissue answered two different questions differently. That distinction matters more than anything on the semaglutide price boards.

Two endpoints, not one

Metabolic dysfunction-associated steatotic liver disease, or MASLD, is fat in the liver. Metabolic dysfunction-associated steatohepatitis, or MASH, is fat plus inflammation and injured cells, and it is the form that scars. Every trial in this field therefore scores a biopsy twice: resolution of steatohepatitis, which is the inflammation clearing, and fibrosis stage, which is the scarring receding. A drug can move one and not the other, and the first semaglutide trial did exactly that.

Fibrosis is the endpoint that predicts what happens to a person. Scarring stages from F0 to F4, and F4 is cirrhosis. Resolution of inflammation is the endpoint that moves first and the one a press release leads with.

The phase 3 result

ESSENCE randomized 1,197 patients with biopsy-defined MASH and stage 2 or 3 fibrosis, in a 2:1 ratio, to once-weekly semaglutide 2.4 mg or placebo for 240 weeks. A planned interim analysis at week 72 covering the first 800 patients is what has been published.[1]

Resolution of steatohepatitis without worsening of fibrosis occurred in 62.9% of 534 patients on semaglutide against 34.3% of 266 on placebo, an estimated difference of 28.7 percentage points (95% CI, 21.1 to 36.2). Reduction in liver fibrosis without worsening of steatohepatitis occurred in 36.8% against 22.4%, a difference of 14.4 percentage points (95% CI, 7.5 to 21.3). Both endpoints were met at P < 0.001. Mean weight change was −10.5% against −2.0%.[1]

One prespecified outcome did not separate: mean changes in bodily pain scores did not differ significantly between the groups.[1] Gastrointestinal adverse events were more common on semaglutide, the same profile described in the side-effect article.

The earlier trial resolved inflammation and left the scarring alone

The phase 2 trial is the reason the fibrosis number in ESSENCE is worth reading twice. It randomized 320 patients with biopsy-confirmed steatohepatitis and stage F1, F2 or F3 fibrosis to once-daily semaglutide at 0.1, 0.2 or 0.4 mg or matching placebo for 72 weeks.[2]

Resolution with no worsening of fibrosis reached 59% in the 0.4 mg group against 17% on placebo (P < 0.001). The fibrosis endpoint did not follow: improvement of at least one stage occurred in 43% on the highest dose and 33% on placebo, P = 0.48.[2] Mean weight loss was 13% against 1%. One trial, one cohort, one dose: the inflammation endpoint cleared placebo by 42 percentage points and the scarring endpoint by 10 points that could not be distinguished from chance.

That trial also recorded something no summary should drop. Malignant neoplasms were reported in three semaglutide recipients (1%) and in none on placebo, and neoplasms of any kind in 15% of semaglutide patients against 8% on placebo, with no pattern in any specific organ.[2] It is a small imbalance in a 320-patient trial rather than a finding, and it is not the sort of number a page that quotes the 59% gets to omit.

In cirrhosis the result points the other way

The population most likely to assume a liver drug is for them is the one it has failed in. A phase 2 trial enrolled 71 patients with biopsy-confirmed steatohepatitis-related compensated cirrhosis — stage F4 — and a body-mass index of 27 or more, randomized 2:1 to once-weekly semaglutide 2.4 mg or placebo for 48 weeks.[3]

Improvement of one fibrosis stage or more without worsening of steatohepatitis occurred in five of 47 patients (11%) on semaglutide and seven of 24 (29%) on placebo — an odds ratio of 0.28 (95% CI, 0.06 to 1.24; p = 0.087). Resolution did not differ either (p = 0.29).[3] The confidence interval crosses 1, so this is a null result rather than evidence of harm; what it is not is a smaller version of ESSENCE. The AASLD practice guidance states plainly that semaglutide is not approved to treat MASH cirrhosis.[4]

What the approval actually covers

The Wegovy formulation received accelerated approval in August 2025 for MASH with moderate-to-advanced fibrosis, corresponding to stages F2 and F3, on the strength of the ESSENCE interim histology. Final approval awaits long-term outcome data.[4] Accelerated approval rests on a surrogate endpoint — biopsy appearance — and not yet on decompensation, transplant or death.

Patient selection under that guidance runs through non-invasive tests rather than biopsy: transient elastography between 8 and 15 kPa, magnetic resonance elastography between 3.1 and 4.4 kPa, or an enhanced liver fibrosis score between 9.2 and 10.5.[4] None of those is a questionnaire, and none of them is ordered at checkout. The guidance also names the risks to monitor, including gallbladder disease and pancreatitis — covered in the gallbladder article — and lean mass loss, covered in the muscle article.

Tirzepatide has the histology and not the indication

SYNERGY-NASH randomized 190 participants with biopsy-confirmed MASH and stage F2 or F3 fibrosis to once-weekly tirzepatide at 5, 10 or 15 mg or placebo for 52 weeks, with 157 evaluable biopsies at the end.[5]

Resolution without worsening of fibrosis was 10% on placebo, 44% at 5 mg, 56% at 10 mg and 62% at 15 mg, all at P < 0.001. Fibrosis improvement of at least one stage was 30% on placebo against 55%, 51% and 51%, with confidence intervals whose lower bounds sat at 5, 1 and 1 percentage points.[5] That is a phase 2 dose-finding trial, not a registration trial, and tirzepatide carries no MASH indication. Anyone weighing the two molecules against each other should read the molecule comparison with that asymmetry in front of them.

Why the headline percentages cannot be ranked

Four trials, four placebo groups, four different rates of getting better on nothing: 34.3% in ESSENCE,[1] 17% in the phase 2 semaglutide trial,[2] 10% in SYNERGY-NASH[5] and 9.7% in the phase 3 trial of resmetirom.[6] Biopsy scoring is subjective, entry criteria differ, and a placebo arm that improves in a third of its patients is not measuring the same thing as one that improves in a tenth.

Resmetirom is the other approved agent and it is not in this class at all. Its phase 3 trial randomized 966 analyzable patients with stage F1B, F2 or F3 fibrosis to 80 mg, 100 mg or placebo; resolution reached 25.9% and 29.9% against 9.7%, and fibrosis improvement 24.2% and 25.9% against 14.2%, all at P < 0.001.[6] Set against ESSENCE those percentages look far smaller, and the two placebo arms differ by twenty-five points. The guidance notes that the combination of resmetirom with semaglutide has not been studied.[4]

What a cash-pay buyer is actually holding

Every figure above was produced with branded, FDA-approved product at labeled doses, in patients enrolled on the strength of a liver biopsy. Most sellers in this market dispense compounded semaglutide or tirzepatide. Compounded drugs are not FDA-approved, and are not reviewed by the FDA for safety, efficacy or quality before they are dispensed — the distinction set out in the compounding article. A MASH indication granted to one branded formulation is not a property a vial acquires by sharing a molecule name.

The practical reading is narrow and worth stating. Suspected fatty liver is a reason to be assessed by someone who can order elastography and read it, not a reason to buy faster. The dose that carries the histology is 2.4 mg weekly, which is the top of the ladder in the titration article, and the trials ran it for 72 weeks. Whether an online program can reach that dose and hold it for that long is a question about the program, and the way this site establishes that sort of figure is described in the methodology.

Frequently asked

Does semaglutide treat fatty liver disease?
One formulation does, for one part of the disease. The Wegovy formulation received accelerated FDA approval in August 2025 for MASH with stage 2 or 3 fibrosis, based on the ESSENCE interim analysis at week 72. Final approval awaits long-term outcome data, and the approval does not extend to MASH cirrhosis.
Does it reverse liver scarring or just inflammation?
Both endpoints moved in ESSENCE, by different amounts. Resolution of steatohepatitis without worsening of fibrosis reached 62.9% against 34.3% on placebo, while fibrosis reduction without worsening of steatohepatitis reached 36.8% against 22.4%. In the earlier phase 2 trial the fibrosis endpoint was not met at all, at 43% against 33% with P = 0.48.
What about people who already have cirrhosis?
That trial has been run and it did not show a benefit. Among 71 patients with compensated cirrhosis from steatohepatitis, fibrosis improved in 11% on semaglutide and 29% on placebo, an odds ratio of 0.28 with a confidence interval from 0.06 to 1.24. The AASLD guidance states that semaglutide is not approved for MASH cirrhosis.
Is tirzepatide approved for MASH?
No. SYNERGY-NASH was a phase 2 dose-finding trial in 190 participants, in which resolution without worsening of fibrosis reached 62% at 15 mg against 10% on placebo. Those are histology results from a trial that was not designed to support registration, and tirzepatide carries no MASH indication.
Can these percentages be compared between drugs?
Not directly, because the placebo arms are not comparable. Resolution on placebo was 34.3% in ESSENCE, 17% in the phase 2 semaglutide trial, 10% in SYNERGY-NASH and 9.7% in the resmetirom phase 3 trial. Biopsy scoring is subjective and entry criteria differ, so a raw percentage carries its own trial's denominator with it.
Does a compounded vial carry the liver indication?
No. Every histology figure in this literature came from branded product at labeled doses in biopsy-selected patients. Compounded drugs are not FDA-approved and are not reviewed by the FDA for safety, efficacy or quality before dispensing, so an indication granted to a branded formulation does not transfer to a compounded preparation.

Sources

  1. [1] Sanyal AJ, Newsome PN, Kliers I, et al. (2025). Phase 3 Trial of Semaglutide in Metabolic Dysfunction-Associated Steatohepatitis. N Engl J Med. PMID 40305708
  2. [2] Newsome PN, Buchholtz K, Cusi K, et al. (2021). A Placebo-Controlled Trial of Subcutaneous Semaglutide in Nonalcoholic Steatohepatitis. N Engl J Med. PMID 33185364
  3. [3] Loomba R, Abdelmalek MF, Armstrong MJ, et al. (2023). Semaglutide 2·4 mg once weekly in patients with non-alcoholic steatohepatitis-related cirrhosis: a randomised, placebo-controlled phase 2 trial. Lancet Gastroenterol Hepatol. PMID 36934740
  4. [4] Bansal MB, Patton H, Morgan TR, et al. (2026). Semaglutide therapy for metabolic dysfunction-associated steatohepatitis: November 2025 updates to AASLD Practice Guidance. Hepatology. PMID 41201884
  5. [5] Loomba R, Hartman ML, Lawitz EJ, et al. (2024). Tirzepatide for Metabolic Dysfunction-Associated Steatohepatitis with Liver Fibrosis. N Engl J Med. PMID 38856224
  6. [6] Harrison SA, Bedossa P, Guy CD, et al. (2024). A Phase 3, Randomized, Controlled Trial of Resmetirom in NASH with Liver Fibrosis. N Engl J Med. PMID 38324483

Where to get it

Best GLP-1 injections

Every injectable seller we can verify, with the price each one publishes and an honest read of what the trials measured.

Compare providers →

More in Evidence