SYNERGY-NASH: Tirzepatide and Liver Disease
Last verified May 2026 · Phase 2 · Finished; headline results published June 2024 · NCT04166773 ↗
MASH — metabolic dysfunction-associated steatohepatitis, the condition formerly called NASH — is a liver disease driven by the same insulin resistance and obesity that GLP-1 drugs act on, and until recently it had essentially no drug treatment. What makes this trial credible is the endpoint. Histologic resolution scored on paired biopsies by centrally adjudicated pathologists is a much harder bar than an imaging surrogate or a blood marker, and it is the bar regulators have asked for. The trade is size: 190 participants is small, and a phase 2 result at this scale is a strong signal rather than a settled question.
- Enrollment
- 190
- Duration
- 52 weeks of treatment, with a liver biopsy at the start and another at the end to score the main result
- Drug
- Tirzepatide
- Population
- Adults 18 to 80 with a BMI between 27 and 50, weight stable for at least three months, and MASH confirmed by biopsy with stage F2 or F3 scarring — significant, but short of cirrhosis. Type 2 diabetes was allowed if HbA1c was 9.5% or below, and about 58% had it. ⚠ Exclusions were extensive: cirrhosis, any other liver disease, prior pancreatitis, substantially reduced kidney function, a family history of medullary thyroid cancer, a raised calcitonin, a recent cardiac event, and any cancer within five years. Average weight around 95 kg at a BMI near 36, about 57% women.
Primary endpoint
Liver inflammation resolved on biopsy at 52 weeks, with scarring no worse
Treatment arm
15 mg: 62% (95% CI 47-77); 10 mg: 56% (95% CI 41-71); 5 mg: 44% (95% CI 29-59)
Comparator
Placebo: 10% (95% CI 0-21)
Treatment difference: All three tirzepatide doses significantly superior to placebo (P<0.001 for each)
★ Understand what was measured, because it is unusually hard: two liver biopsies a year apart, scored by pathologists who did not know who got what. This is the standard regulators ask for, not a scan or a blood test, and it is why a 190-person trial carries weight it otherwise would not. The three doses lined up in order.
Secondary endpoints
| Endpoint | Treatment | Comparator | Difference |
|---|---|---|---|
| Liver scarring actually improved by at least one stage, without the inflammation getting worse ★ The more demanding of the two regulatory pillars — reversing damage rather than halting it — and roughly half the treated participants managed it against 30% on placebo. ⚠ Note the doses did NOT line up here: 5 mg did as well as 15 mg, which hints that the scarring benefit tops out earlier than the inflammation benefit. That is a hypothesis from 190 people, not a finding. | 15 mg: 51% (95% CI 35-66); 10 mg: 51% (95% CI 35-67); 5 mg: 55% (95% CI 39-71) | Placebo: 30% (95% CI 15-44) | Numerically higher on all tirzepatide doses; statistical significance not consistently reached across all dose comparisons in the primary analysis |
| How much body weight came off In line with what this drug produces in weight trials at the same doses, which sets up the question of whether the liver result is anything more than the weight result. | 15 mg: -15.6%; 10 mg: -13.3%; 5 mg: -10.7% | -2.3% | Placebo-subtracted: -13.3 (15 mg), -11.0 (10 mg), -8.4 (5 mg) percentage points |
| The combined liver damage score, adding up fat, inflammation and cell swelling Down 3 points on the higher doses against 1 on placebo, on a scale running 0 to 8. This score is the underlying arithmetic that produces the headline result. | 15 mg: -3.0; 10 mg: -3.0; 5 mg: -2.6 | -1.0 | Placebo-subtracted reductions of 1.6-2.0 points; all three doses significantly different from placebo |
| How much fat left the liver, measured by MRI ★ Down 59% at the top dose against 10% on placebo. A 30% reduction is generally enough to predict a biopsy response, so this comfortably clears it — and unlike biopsy, it is a measurement someone could actually get repeated. | 15 mg: -59%; 10 mg: -52%; 5 mg: -43% | -10% | All three doses significantly different from placebo |
| ALT, the liver enzyme most commonly tested ⚠ Down substantially on every dose. Useful as corroboration and not sufficient on its own — enzymes can normalize while disease persists, which is why regulators insist on tissue. | 15 mg: -28 U/L; 10 mg: -24 U/L; 5 mg: -28 U/L | -10 U/L | Placebo-subtracted reductions of 14-18 U/L across tirzepatide doses |
| AST, the other routinely tested liver enzyme Down roughly twice as much as placebo, moving with ALT. | 15 mg: -16 U/L; 10 mg: -14 U/L; 5 mg: -17 U/L | -7 U/L | Placebo-subtracted reductions of 7-10 U/L across tirzepatide doses |
| Three-month average blood sugar, in those who had diabetes Down about 1.6 points at the higher doses, matching what this drug does in dedicated diabetes trials. Those without diabetes moved about 0.4 points. | 15 mg: -1.6%; 10 mg: -1.6%; 5 mg: -1.3% | -0.3% | Placebo-subtracted reductions of 1.0-1.3 percentage points in the T2DM subgroup |
| Liver stiffness measured by a non-invasive scan ★ Improved on every dose. It matters because it agrees with the biopsy result using a completely different method — two independent measures pointing the same way is much harder to explain away than one. | 15 mg: -3.7 kPa; 10 mg: -3.1 kPa; 5 mg: -2.5 kPa | -0.5 kPa | All three tirzepatide doses produced larger reductions in liver stiffness than placebo |
Adverse events
| Event | Treatment rate | Comparator rate |
|---|---|---|
| Nausea Mostly mild to moderate and clustered while the dose was rising, following the same dose pattern seen in the weight and diabetes trials. | 5 mg: 18%; 10 mg: 34%; 15 mg: 32% | 13% |
| Diarrhea Higher than placebo without tracking the dose closely. | 5 mg: 23%; 10 mg: 13%; 15 mg: 23% | 13% |
| Constipation ⚠ Only modestly above placebo, and at the lowest dose actually below it — a reminder that small trials produce noisy safety numbers. | 5 mg: 8%; 10 mg: 17%; 15 mg: 19% | 13% |
| Vomiting Rose steadily with dose, against none at all on placebo. | 5 mg: 5%; 10 mg: 9%; 15 mg: 13% | 0% |
| Loss of appetite Recorded as a side effect, though it is how the drug works. | 5 mg: 18%; 10 mg: 11%; 15 mg: 21% | 0% |
| COVID-19 ★ Listed because the trial ran from 2020 to 2023. It is background noise from the pandemic, not a drug effect — and a good illustration of why a raw adverse-event list needs reading rather than counting. | 5 mg: 25%; 10 mg: 23%; 15 mg: 21% | 13% |
| Stopped the drug over a side effect ⚠ Only 3 to 6%, notably lower than this drug's weight and diabetes trials report. With 190 participants that may be chance rather than a genuinely easier experience. | 5 mg: 3%; 10 mg: 6%; 15 mg: 4% | 0% |
| Serious adverse events of any kind ★ No excess over placebo, and specifically no sign of livers decompensating — the thing you would most want ruled out when treating liver disease. | 5 mg: 8%; 10 mg: 6%; 15 mg: 8% | 10% |
| Pancreatitis, confirmed by independent review Not one case in any arm across the year. | 0% across all tirzepatide arms | 0% |
Subgroup analyses
- People who also had type 2 diabetes — about 58%: The same rates of liver improvement as everyone else
★ A meaningful check: if the drug were helping the liver purely by fixing blood sugar, the diabetes group should have improved more. They did not, which points at additional mechanisms.
- Moderate scarring against more advanced scarring: Broadly consistent improvement in both
★ The F3 group is the one that matters clinically — one step below cirrhosis and closest to serious complications. That the drug worked there too is the more important half of this finding, even though the numbers in each stratum are small.
Clinical significance
For anyone with fatty liver disease alongside obesity, this is among the most consequential findings in the class, and it sits alongside semaglutide's ESSENCE program pointing the same way. Read it with its phase and size in view — 190 people, phase 2, and a resolution rate that will be tested at scale before it becomes a label. ⚠ Note also that MASH resolution is not the same as fibrosis regression; the endpoint required fibrosis not to worsen, which is a lower bar than reversing it.
Who sells tirzepatide, and for how much
This trial tested Tirzepatide. Across the 556 sellers on this register, 253 publish a standing monthly cash price for compounded tirzepatide, from $99 a month, with a median of $240.
⛔ None of those sellers was in this trial. A compounded preparation is not the product studied here: no agency has reviewed that specific preparation for safety, effectiveness or manufacturing quality, and its concentration is set by whichever pharmacy filled it. The result above is the strongest evidence available for tirzepatide and it was generated on something else. Every seller, with its price record.
Frequently Asked Questions
References
- 1.Loomba R, Hartman ML, Lawitz EJ, et al.; SYNERGY-NASH Investigators. Tirzepatide for Metabolic Dysfunction-Associated Steatohepatitis with Liver Fibrosis N Engl J Med. 2024. PMID: 38856224.
- 2.Eli Lilly and Company. A Study of Tirzepatide (LY3298176) in Participants With Nonalcoholic Steatohepatitis (NASH) (SYNERGY-NASH, NCT04166773) ClinicalTrials.gov. 2024. https://clinicaltrials.gov/study/NCT04166773
- 3.Eli Lilly and Company. Lilly's tirzepatide significantly resolved metabolic dysfunction-associated steatohepatitis (MASH) in adults at risk of progressive liver disease in SYNERGY-NASH Phase 2 trial Lilly Investor Press Release. 2024. https://investor.lilly.com/news-releases/news-release-details/lillys-tirzepatide-significantly-resolved-non-alcoholic
- 4.Jastreboff AM, Aronne LJ, Ahmad NN, et al.; SURMOUNT-1 Investigators. Tirzepatide Once Weekly for the Treatment of Obesity (SURMOUNT-1) N Engl J Med. 2022. PMID: 35658024.
- 5.U.S. Food and Drug Administration. FDA Approves First Treatment for Patients with Liver Scarring Due to Fatty Liver Disease (resmetirom/Rezdiffra) FDA Press Announcement. 2024. https://www.fda.gov/news-events/press-announcements/fda-approves-first-treatment-patients-liver-scarring-due-fatty-liver-disease