# GLP-1 Drugs in MASH: Deaths, Ascites and a Small Liver Cancer Signal | This GLP A matched cohort of 168,733 adults with MASH found 2.2 fewer deaths per hundred on a GLP-1, and a liver-cancer signal too small to mean anything. Publisher: This GLP Editorial Team Canonical: https://thisglp.com/research/glp1-liver-cancer-signal-dismissed Updated: 2026-09-18 Published: 2026-09-16 [← Research](https://thisglp.com/research) Evidence # GLP-1 Drugs in MASH: Deaths, Ascites and a Small Liver Cancer Signal In 168,733 matched pairs of adults with MASH, GLP-1 users had 2.2 fewer deaths per hundred. A liver-cancer signal cleared significance with a risk difference of zero, and the authors set it aside. [This GLP Editorial Team](https://thisglp.com/about)September 20266 min readUpdated September 2026 Difference in percentage points, GLP-1 exposed against matched · ascites · -1.3 points · all-cause death · -2.2 points · liver cancer · 0.0 — and statistically significant · 168,733 matched patients per group. Follow-up window not stated. **Research, not medical advice.** This GLP is independent and educational. Nothing here is a substitute for individualized care from a licensed clinician, and reading it creates no doctor–patient relationship. Talk to your own provider before starting, stopping, or changing any medication, peptide, or therapy. See our [methodology](https://thisglp.com/methodology) and [terms](https://thisglp.com/terms). ## Key takeaways - A claims-database study matched 168,733 adults with MASH taking a GLP-1 against the same number who were not. - Ascites was lower by 1.3 percentage points and death by 2.2 points, both at p < 0.001. - Liver cancer was higher, at p = 0.025, with a risk difference of 0.000. - The authors set the cancer signal aside because its absolute size was negligible. That reasoning is right. - The same reasoning, run on the benefits, makes them real and small rather than transformative. This is a study with an unusual virtue and one blind spot, and they are the same feature seen from two sides. ## What it found Adults with metabolic dysfunction-associated steatohepatitis were pulled from a large real-world health database and split by whether they had a documented GLP-1 prescription. [\[1\]](https://thisglp.com/research/glp1-liver-cancer-signal-dismissed#ref-1) Propensity matching on demographics, coexisting conditions and concomitant drugs left 168,733 patients on each side, out of 168,740 exposed. Three outcomes were reported, all as absolute risk differences. Ascites was lower among GLP-1 users by 0.013, a 1.3-point difference, 95% CI 0.014 to 0.012 lower, p < 0.001. All-cause death was lower by 0.022, or 2.2 points, 95% CI 0.023 to 0.021 lower, p < 0.001. Liver cancer went the other way: higher among users, p = 0.025, risk difference 0.000, 95% CI 0.000 to 0.001. ## The good part Reporting all three on one scale is better practice than most papers manage. A relative figure would have made the cancer result sound alarming and the mortality result sound heroic, and both would have been the same numbers dressed differently — which is [how a statistical result usually gets oversold](https://thisglp.com/research/does-a-glp1-affect-surgery-outcomes). The authors then looked at the cancer signal, saw a difference of zero at three decimal places, and declined to treat a p-value of 0.025 as a clinical finding. In a matched sample this size, tiny differences clear significance easily. That judgment is correct. Run it in the other direction The same standard applied to the benefits gives 1.3 fewer cases of ascites and 2.2 fewer deaths per hundred people. Those are real. They are also modest, and the paper’s closing description of GLP-1s as beneficial and hepatically safe carries a weight the risk differences do not. Nothing in the arithmetic is inconsistent; the language around it is. ## Nobody said over how long The indexed abstract gives no follow-up duration. A risk difference without a window is not a rate. Two point two points of mortality across one year and the same figure across eight years describe very different drugs, and a reader cannot tell which this is. ## Prescribed is not randomized Everyone here chose, or was offered, their treatment. Being prescribed a GLP-1 for MASH and filling it repeatedly selects for people healthy enough to be considered, insured enough to afford it, and engaged enough to keep going. Matching on codes cannot see any of that. It is the standing weakness of every database comparison on this site, and it pushes in the direction that flatters the treated group — the same shape as [a conclusion stated more confidently than its design allows](https://thisglp.com/research/do-glp1-drugs-help-parkinsons-disease). ## What to do with it Read it as reassurance about liver safety rather than as a reason to buy. The liver-cancer result is the kind of signal that would matter if it were larger, and it is not. The survival result points the right way and is small. If liver disease is the reason someone is considering one of these drugs, the relevant page is [the trial evidence in that population](https://thisglp.com/research/semaglutide-and-liver-disease), not a claims cohort — and [which events a study counts](https://thisglp.com/research/which-glp1-has-the-worst-side-effects) decides most of what its numbers can mean. ## Frequently asked Do GLP-1 drugs cause liver cancer? This study found more cases among users at p = 0.025, but the absolute difference was 0.000 — nothing a person could act on. The authors treated it as clinically irrelevant, which is the right reading of a difference that size. How much did it reduce deaths? By 2.2 percentage points against matched patients, at p < 0.001. The abstract does not say over what period, so the figure has no rate attached to it. Is this as good as a trial? No. Treatment was not randomized, and being prescribed a GLP-1 selects for people already healthier and better insured. Matching on database codes cannot correct for that. What was the population? Adults with metabolic dysfunction-associated steatohepatitis in a multinational health database, 168,733 matched on each side of the comparison. ## Sources 1. \[1\] Khan SA, et al. (2026). Clinical outcomes associated with GLP-1 receptor agonist use in metabolic dysfunction-associated steatohepatitis: A multinational cohort study Medicine (Baltimore). [PMID 42736811](https://pubmed.ncbi.nlm.nih.gov/42736811/) ## 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 →](https://thisglp.com/best/compounded-semaglutide) ## More in Evidence - [ 150 minutes · Weekly activity goal · 2 days · Muscle strengthening · Adapt · Symptoms and ability ### Exercise on Ozempic: Building Strength Without Ignoring Symptoms ](https://thisglp.com/research/exercise-on-ozempic) - [ STEP 1, semaglutide 2.4 mg vs placebo, % change at 68 weeks · LDL cholesterol · −2.5% vs +1.3% · Total cholesterol · −3.3% vs +0.1% · HDL cholesterol · +5.2% vs +1.4% · Triglycerides · −21.9% vs −7.3% · Wegovy label, Table 13. Adults without diabetes. Ozempic doses are lower. ### Does Ozempic Lower Cholesterol? What the Lipid Results in the Trials Show ](https://thisglp.com/research/does-ozempic-lower-cholesterol) - [ Where the insulin reduction came from, units per day · week 4 · 13.4 total · week 26 · 21.9 · the drug itself · the weight loss · The drug’s own contribution barely moved: 11.1 then 11.4. · Everything added between weeks 4 and 26 came from weight. ### Does Semaglutide Help Type 1 Diabetes? What Two Small Trials Found ](https://thisglp.com/research/semaglutide-in-type-1-diabetes)