A treatment can improve the appearance of liver tissue before researchers know whether it prevents liver failure. That distinction is central to ESSENCE, the phase 3 trial of semaglutide in metabolic dysfunction-associated steatohepatitis, or MASH. At 72 weeks, participants receiving semaglutide were more likely than those receiving placebo to meet two important biopsy-based goals: resolution of steatohepatitis without worsening scarring, and improvement in scarring without worsening steatohepatitis. These are substantial findings, but they answer a narrower question than whether treatment prevents transplantation or death.
Wegovy injection now has an FDA-approved MASH indication under the accelerated approval pathway; describing it as merely investigational for this use would be incorrect. The more useful question is how much confidence to place in each kind of benefit. This analysis uses the indexed primary report abstracts and available official FDA material, not a review of all full articles, protocols, or supplements. It separates tissue findings, patient-experienced outcomes, longer-term clinical events, and the practical implications of paying for ongoing care.
ESSENCE: two distinct liver-biopsy outcomes at week 72
Planned interim analysis of 800 adults with biopsy-defined MASH and stage F2/F3 fibrosis: 534 assigned weekly subcutaneous semaglutide 2.4 mg and 266 placebo. MASH resolution required no worsening of fibrosis; fibrosis reduction required no worsening of MASH.
The two endpoints overlap and must not be added. These biopsy results do not establish reductions in death or liver transplantation. Reported adjusted differences were 28.7 percentage points (95% CI 21.1–36.2) and 14.4 points (7.5–21.3); rounded arm percentages may not subtract to the adjusted estimate. Funded by Novo Nordisk.
Source 1Download the source-labeled chart (PNG) · Download the data (CSV)
| Study group | Reported value |
|---|---|
| MASH resolution: semaglutide | 62.9 |
| MASH resolution: placebo | 34.3 |
| Fibrosis reduction: semaglutide | 36.8 |
| Fibrosis reduction: placebo | 22.4 |
Sections in this article
The liver disease ESSENCE actually studied
MASH involves more than fat accumulation in the liver. Inflammation and tissue injury occur alongside fat buildup, and fibrosis means that scar tissue has formed. Progressive disease can lead to cirrhosis, loss of liver function, liver cancer, or transplantation. Yet many people have few symptoms before substantial damage develops. That makes objective measures valuable, while also explaining why symptom relief and disease modification are not interchangeable goals.
ESSENCE selected participants with biopsy-defined MASH and fibrosis stage F2 or F3: moderate or advanced scarring, rather than cirrhosis. Its findings therefore apply most directly to this noncirrhotic population. They should not automatically be extended to people with incidental liver fat, no established fibrosis, or cirrhosis. Those groups may have different baseline risks and different opportunities for benefit.
The intervention was once-weekly subcutaneous semaglutide, meaning an injection under the skin, at a study dose of 2.4 mg. Semaglutide activates the GLP-1 receptor. The FDA describes weight loss and potentially other incompletely understood mechanisms as contributors to liver improvement. ESSENCE does not establish exactly how much benefit came from weight loss versus other pathways.
Source 1 Source 2A planned interim analysis within a longer trial
ESSENCE is a multicenter, randomized, double-blind, placebo-controlled trial. Investigators assigned 1,197 participants in a 2:1 ratio to semaglutide or placebo for a planned 240 weeks. The reported part 1 analysis covered the first 800 participants at week 72: 534 assigned semaglutide and 266 assigned placebo. It was a planned analysis, not simply an unscheduled look after encouraging results.
Randomization strengthens the comparison by reducing systematic differences between treatment groups. Blinding limits the influence of knowing which treatment someone receives. According to the FDA account, both groups also received lifestyle intervention, and participants were on stable doses of lipid-lowering, glucose-management, and weight-loss medications. Thus, the comparison estimates the added effect of the trial treatment within that care setting, not a drug versus no care.
Novo Nordisk funded the trial. Sponsorship is relevant context, but it does not replace examination of design and endpoints. The indexed abstract does not provide the detailed missing-biopsy procedures, participant flow, or full safety tables needed for a complete methods appraisal. Our guide to reading a trial abstract explains why a strong headline result can coexist with unanswered methodological questions.
Source 1 Source 2Two primary endpoints capture different kinds of improvement
The first primary endpoint asked whether steatohepatitis resolved without fibrosis getting worse. It occurred in 62.9% of the semaglutide group and 34.3% of the placebo group. Resolution here means meeting the study’s tissue-based definition; it does not mean every aspect of liver health normalized or that future liver risk disappeared.
The second primary endpoint asked whether fibrosis decreased without steatohepatitis worsening. Results were 36.8% versus 22.4%. This is a distinct achievement: less active disease and less scar tissue are related, but one does not guarantee the other. The different response proportions show why calling all of these outcomes simply “liver improvement” loses clinically useful information.
Both primary comparisons had reported P values below 0.001. A secondary endpoint required both steatohepatitis resolution and fibrosis reduction, achieved by 32.7% versus 16.1%. These response categories overlap. They cannot be added to calculate a total success rate, because the same participant can satisfy more than one endpoint.
| Outcome | Semaglutide versus placebo | Estimated difference, percentage points (95% CI) |
|---|---|---|
| Steatohepatitis resolution without worsening fibrosis | 62.9% versus 34.3% | 28.7 (21.1 to 36.2) |
| Fibrosis reduction without worsening steatohepatitis | 36.8% versus 22.4% | 14.4 (7.5 to 21.3) |
| Both steatohepatitis resolution and fibrosis reduction | 32.7% versus 16.1% | 16.5 (10.2 to 22.8) |
Reading the differences without overstating precision
The reported 28.7-percentage-point difference is an absolute contrast between groups, not a 28.7% relative reduction. As an interpretation, it represents roughly 29 additional people per 100 meeting the resolution endpoint under the study comparison. It is not a prediction that exactly 29 additional people in every clinic will respond.
The displayed proportions subtract to 28.6 points, whereas the report’s estimated difference is 28.7. Likewise, the joint endpoint’s displayed values subtract to 16.6 points, while its reported estimate is 16.5. These are different presentations of the result; the published estimates should be retained rather than replaced with arithmetic on rounded percentages. The abstract does not supply enough detail to reconstruct the estimation procedure.
A 95% confidence interval describes statistical uncertainty around the estimated group difference. For fibrosis improvement, the interval runs from 7.5 to 21.3 percentage points: all values favor semaglutide, but the magnitude remains uncertain. These are not intervals predicting an individual’s response. The distinction also differs from the hazard ratios used in event trials, as explained in absolute differences versus relative hazards.
Source 1Weight changed substantially; pain did not clearly improve
Mean body weight declined 10.5% with semaglutide and 2.0% with placebo. The estimated between-group difference was −8.5 percentage points, with a 95% confidence interval of −9.6 to −7.4. These are group averages, not uniform experiences, and the negative sign indicates greater weight reduction with semaglutide.
Weight change, the joint biopsy endpoint, and bodily pain were the three secondary outcomes included in the plan to adjust for multiple testing. That matters because testing many outcomes increases the opportunity for chance-positive findings unless the analysis accounts for it. Mean bodily pain changes did not differ significantly between groups. The report therefore does not establish a pain benefit alongside the biopsy benefit.
The parallel weight and liver findings are biologically relevant but do not prove mediation: that weight loss caused a specified share of the liver response. A randomized treatment comparison can establish an effect of assignment without disentangling every pathway producing it. Nor does the average weight reduction tell us which participants gained fibrosis improvement or whether their weight and biopsy responses tracked together.
Source 1Accelerated approval leaves a specific clinical question open
FDA approved Wegovy injection for adults with MASH and moderate-to-advanced fibrosis through accelerated approval. This pathway permits approval for serious disease based on a surrogate endpoint, with additional evidence required to confirm meaningful clinical benefit. A surrogate is a measure used in place of directly observing how people feel, function, or survive.
Here, liver histology—the microscopic examination of biopsy tissue—supported approval. That is more disease-specific than weight change alone, but it is not the same as observing fewer liver transplants or deaths. The FDA describes continued follow-up to 240 weeks to determine whether the inflammation and scarring improvements translate into fewer deaths, transplants, and other liver-related events.
The reported 72-week analysis establishes biopsy benefit, while the longer-term liver-event question requires separate evidence. Regulatory approval and an ongoing confirmatory trial can coexist: approval recognizes the available findings, while continued follow-up tests whether they predict fewer serious clinical events. This distinction matters when assessing how study findings become marketing claims.
Source 1 Source 2Tolerability belongs beside the efficacy results
Gastrointestinal adverse events were more common with semaglutide in ESSENCE. The accessible abstract does not give their counts, severity distribution, or treatment-discontinuation rates. That limits a numerical benefit-harm comparison within this population; it does not justify describing the treatment as free of important tolerability problems.
The FDA approval announcement lists nausea, diarrhea, vomiting, constipation, abdominal pain, and other common adverse effects. It also identifies restrictions for people with a personal or family history of medullary thyroid cancer, multiple endocrine neoplasia type 2, or known sensitivity to semaglutide or product ingredients. This announcement is not a substitute for complete current prescribing information.
Safety percentages from another trial cannot fill ESSENCE’s missing safety table. For example, SELECT reported permanent discontinuation because of adverse events in 16.6% with semaglutide versus 8.2% with placebo. That is useful context for treatment burden, but SELECT studied a different population over longer follow-up. Its discontinuation rate is not an estimate of the rate among people receiving treatment for MASH.
Source 1 Source 2 Source 3Cardiovascular and kidney trials answer separate questions
SELECT and FLOW show why endpoint and population must travel with a drug’s name. SELECT studied adults with established cardiovascular disease and overweight or obesity, without diabetes. FLOW studied people with type 2 diabetes and chronic kidney disease defined by kidney function and urine albumin criteria. Neither was a trial of the ESSENCE biopsy-defined MASH population.
SELECT’s hazard ratio of 0.80 describes an estimated 20% lower hazard of a first composite cardiovascular event during follow-up, not a 20-percentage-point reduction. The observed event proportions were 6.5% versus 8.0%. FLOW’s primary hazard ratio was 0.76, with 331 versus 410 first events; its composite included kidney outcomes and cardiovascular death. A favorable composite does not establish that every component improved equally.
Both were randomized placebo-controlled trials funded by Novo Nordisk. FLOW stopped early following a prespecified interim review, and serious adverse events were reported in 49.6% versus 53.8%. These studies support benefits in their tested settings, not a transferable numerical promise for liver disease. Our guide to reading composite endpoints provides a framework for keeping those claims distinct.
| Trial and population | Intervention and follow-up | Primary endpoint and result |
|---|---|---|
| ESSENCE: MASH with F2/F3 fibrosis; 800 in interim analysis | Weekly subcutaneous 2.4 mg versus placebo; 72-week analysis within 240-week trial | Two biopsy endpoints; resolution 62.9% versus 34.3%, fibrosis improvement 36.8% versus 22.4% |
| SELECT: 17,604 adults with cardiovascular disease, BMI ≥27, no diabetes | Weekly subcutaneous 2.4 mg versus placebo; mean follow-up 39.8 months | Cardiovascular death, nonfatal heart attack, or nonfatal stroke: HR 0.80 (95% CI 0.72–0.90) |
| FLOW: 3,533 adults with type 2 diabetes and chronic kidney disease | Weekly subcutaneous 1.0 mg versus placebo; median follow-up 3.4 years | Kidney failure, at least 50% eGFR decline, or kidney-related/cardiovascular death: HR 0.76 (95% CI 0.66–0.88) |
An oral GLP-1 result is not a MASH equivalence study
ATTAIN-1 provides a useful boundary on class-wide claims. It tested orforglipron, an oral small-molecule GLP-1 receptor agonist, in 3,127 adults with obesity without diabetes. Participants were randomized to daily 6, 12, or 36 mg doses of the investigational formulation or placebo for 72 weeks. The primary endpoint was percentage change in body weight, not liver biopsy response.
Mean weight changes were −7.5%, −8.4%, and −11.2% across the active groups, versus −2.1% with placebo. Adverse events led to discontinuation in 5.3% to 10.3% of active-treatment participants versus 2.7% with placebo; gastrointestinal effects were most common. Eli Lilly funded this trial.
Similar-looking weight results across separate studies do not establish equivalent liver effects. The populations, molecules, routes, and endpoints differ, and there was no head-to-head MASH comparison. Our ATTAIN-1 analysis examines the oral trial on its own terms. ESSENCE supports the tested injectable semaglutide regimen, not automatic substitution with an oral drug or another incretin-based treatment.
Source 5 Source 1What the evidence means for a continuing care budget
Editorial interpretation: a biopsy benefit and a recurring care contract are different propositions. ESSENCE did not test provider memberships, compare pharmacy prices, or calculate a cost per transplant prevented. Its results cannot establish that a particular subscription is economical. A budget discussion should separate medication charges, clinical follow-up, testing arrangements, and cancellation terms rather than attaching a monetary value to an unmeasured clinical outcome.
Provider documentation can be reviewed independently of efficacy. Readers comparing written service and payment terms can consult the Ro review or WeightWatchers review. These links are invitations to examine documentary terms, not evidence that either service supplies the studied product, provides MASH-specific care, or reproduces ESSENCE outcomes.
Product identity also matters. FDA warns that compounded GLP-1 products do not undergo its premarket review for safety, effectiveness, and quality, and it has received reports involving dosing errors, some requiring hospitalization. Those concerns are separate from ESSENCE’s trial findings. Clinical results do not automatically transfer to compounded products, other formulations, or a provider simply because its advertising mentions semaglutide.
Source 1 Source 6Questions about the evidence
Does MASH resolution mean the liver is cured?
No. In ESSENCE, resolution was a biopsy-defined endpoint that also required no worsening of fibrosis. It did not require all scar tissue to disappear, and it did not establish elimination of future liver risk. The separate fibrosis and joint-response endpoints help show why resolution should not be treated as a complete cure.
Source 1Why did more than one-third of the placebo group meet the resolution endpoint?
The placebo group also received lifestyle intervention, and response reflects the full study setting rather than an untreated comparison. The available abstract does not determine how much of its response came from lifestyle changes, disease variation, biopsy assessment, or other factors. Randomized between-group differences are more informative about added treatment effect than the active group’s response alone.
Source 1 Source 2Has this analysis shown fewer liver transplants or deaths?
The reported 72-week analysis has not established those outcomes. Its primary endpoints were biopsy findings. FDA describes the trial’s planned 240-week duration as the period intended to determine whether tissue improvements translate into fewer deaths, transplants, and other liver-related events.
Source 1 Source 2Can the weight result explain all of the liver benefit?
The evidence summarized here cannot answer that. Semaglutide produced greater average weight loss and better biopsy outcomes, but parallel improvements do not quantify a causal pathway. Determining the share attributable to weight change would require additional analyses beyond the group averages in the indexed report.
Source 1 Source 2Does the MASH approval validate every semaglutide product?
No. The FDA action concerns Wegovy injection for the indicated population, and ESSENCE tested a specific weekly subcutaneous regimen. It does not establish MASH efficacy for compounded versions, oral formulations, or another GLP-1 drug. FDA separately warns that unapproved products have not undergone its premarket review for safety, effectiveness, and quality.
Source 1 Source 2 Source 6Editorial assessment
ESSENCE provides persuasive randomized evidence of improved liver histology at 72 weeks in noncirrhotic MASH with F2 or F3 fibrosis. Both primary endpoints favored semaglutide, with confidence intervals supporting a real treatment difference. The practical balance includes greater gastrointestinal burden and no demonstrated bodily pain advantage in this analysis. Accelerated approval recognizes the importance of the biopsy findings while preserving a distinct obligation to confirm clinically meaningful benefit. Editorially, neither a promise of longer survival nor a claim of financial value should be built from tissue responses alone.
Research access and limitations
This analysis uses indexed primary research abstracts and official FDA material. It does not represent examination of complete protocols, supplements, current prescribing information, or all subsequent publications. Detailed biopsy adjudication, missing-data procedures, subgroup estimates, and ESSENCE adverse-event counts are not available in the accessible abstract and are not inferred.
Source notes
Scientific sources checked October 21, 2026. Earlier research and provider sources retain their dated histories below. Provider pages document advertised terms. Study summaries identify the scope of our review.
- Phase 3 Trial of Semaglutide in Metabolic Dysfunction-Associated Steatohepatitis.Indexed original research abstract · Checked 2026-10-21
- FDA approves treatment for serious liver disease known as MASHOfficial FDA page read through web tool; historical approval announcement, not complete current labeling · Checked 2026-10-21
- Lincoff et al. — SELECT cardiovascular outcomes trialOriginal randomized trial; abstract reviewed · Checked 2026-10-21
- Effects of Semaglutide on Chronic Kidney Disease in Patients with Type 2 Diabetes.Primary research or official information · Checked 2026-10-21
- Orforglipron, an Oral Small-Molecule GLP-1 Receptor Agonist for Obesity Treatment.Indexed original research abstract · Checked 2026-10-21
- FDA — concerns with unapproved GLP-1 drugsRegulatory and safety context · Checked 2026-10-21