Explore this article's sources with AI
Follow PeptideMethods on Google
ESSENCE Japanese Subgroup Readout
The Japanese ESSENCE subgroup broadly matched the main semaglutide MASH result, but its fibrosis estimate remained imprecise.
Why we wrote this. National subgroup results can look more certain than their sample permits. We kept the Japanese estimates separate from the powered overall comparison.
In this article (5 sections)
A prespecified subgroup analysis of ESSENCE found that the main steatohepatitis-resolution result for semaglutide was directionally similar among 116 Japanese participants and the overall phase 3 trial population. The fibrosis endpoint was less certain in the Japanese subgroup because its confidence interval included no difference[1]. The trial was powered for the full population, not for this national subgroup.
The Japanese participants were older and had a lower body mass index than the overall population, which makes the consistency check clinically useful without showing that nationality itself changes treatment response. They remained part of the parent ESSENCE trial.
Where the subgroup came from
ESSENCE is an ongoing multicenter, randomized, double-blind, placebo-controlled phase 3 trial that assigned adults with biopsy-defined metabolic dysfunction-associated steatohepatitis (MASH) and fibrosis stage F2 or F3 to weekly semaglutide 2.4 mg or placebo. The full protocol runs for 240 weeks[3]. This report stops at week 72 and covers the first 800 randomized participants from the 1,197-person population described in the paper[2].
Of those 800 participants, 116 were Japanese: 78 assigned to semaglutide and 38 to placebo. At week 72, 72 and 37 participants, respectively, remained on treatment. The primary histology analysis used the intention-to-treat principle and multiple imputation for missing values[2]. The regimen describes the trial and is not a dosing instruction.
What the biopsy endpoints showed
Resolution of steatohepatitis without worsening fibrosis was estimated in 63.1% of Japanese participants assigned semaglutide and 36.8% assigned placebo. The estimated difference was 26.65 percentage points, with a 95% confidence interval from 7.43 to 45.86[1]. This endpoint was close to the overall ESSENCE estimates of 62.9% and 34.3%.
Fibrosis improvement without worsening steatohepatitis was estimated in 36.5% with semaglutide and 28.9% with placebo. The difference was 7.07 percentage points, with a 95% confidence interval from -11.48 to 25.62[1]. Because that interval crosses zero, this subgroup result does not establish a statistically clear fibrosis benefit. Precision is the issue. The semaglutide response was similar to the full trial, but the placebo response was higher in the Japanese subgroup.
Other outcomes and adverse events
Mean body weight changed by -11.41% with semaglutide and -2.59% with placebo, an estimated treatment difference of -8.81 percentage points. The secondary analysis also covered liver enzymes and several non-invasive liver tests. Those measures moved more favorably with semaglutide, but the subgroup analyses were largely descriptive and the paper states that most secondary endpoints had no formal statistical analysis[2].
The safety set included 141 Japanese participants from the wider treated population, with adverse events reported in 93.7% assigned semaglutide and 84.8% assigned placebo and serious adverse events in 21.1% and 13.0%, respectively. Common events with semaglutide included constipation, nausea and diarrhea. Gastrointestinal disorders were the most common serious-event category[2]. These subgroup counts are small, and some zero-cell data were withheld to preserve blinding in the ongoing trial.
How to interpret a national subgroup
Japanese participants were more often 65 or older and more often had a body mass index below 25 than participants in the overall trial. Baseline liver enzymes, diabetes prevalence, fibrosis-stage distribution and liver stiffness were broadly similar[2]. The authors did not find a meaningful pattern by age or body-mass-index category, but several strata contained very few people.
A subgroup can check whether the full-trial result points in the same direction in a clinically relevant population. It cannot prove identical effects across countries or ethnic groups. Small samples produce wide intervals. Baseline imbalances can also remain after adjustment, while the unexplained higher placebo fibrosis response makes the Japanese estimate especially imprecise.
Our earlier ESSENCE overview covers the overall interim trial and its regulatory context. This subgroup paper adds a narrower answer about Japanese participants, while the main randomized comparison remains the trial's primary efficacy result. The paper contains no head-to-head comparison with another medicine.
What we do not know yet
The clinical-outcomes result is pending. ESSENCE continues to week 240 with cirrhosis-free survival as a primary outcome, so the week-72 biopsy analysis does not yet show whether semaglutide prevents cirrhosis, liver decompensation, transplantation or liver-related death[3]. The subgroup paper also notes that treatment-revealing data remain undisclosed while part 2 stays blinded[2].
For the Japanese subgroup, larger post-approval datasets may clarify uncommon adverse events and whether age, body size or other baseline factors modify outcomes. The analysis cannot separate nationality from the many clinical and care-setting differences that may travel with it. Nor can the absence of an obvious pattern across small strata prove that treatment response is uniform. Readers comparing these results with the wider semaglutide evidence should keep the hierarchy clear: the randomized full-population estimate is the main efficacy result, while this subgroup is a narrower check on consistency. Subgroup uncertainty provides no evidence of either equivalence or a country-specific effect. Patients should not infer eligibility or expected benefit from these averages. Diagnosis, fibrosis staging and treatment decisions require a clinician who can interpret the individual's liver and metabolic risk.
Medical disclaimer: This article is for educational and journalistic purposes only and does not constitute medical advice. Peptides discussed may be classified as prescription medicines or research chemicals depending on your jurisdiction. Always consult a qualified healthcare professional before using any peptide product. PeptideMethods.com does not sell, distribute, or facilitate the sale of any peptide product.
Frequently asked
How many Japanese participants were in the ESSENCE subgroup?
The interim efficacy subgroup included 116 Japanese participants: 78 assigned to semaglutide and 38 to placebo. The parent ESSENCE interim analysis included 800 participants.
Did semaglutide improve MASH resolution in the Japanese subgroup?
The model-based estimate for MASH resolution without worsening fibrosis was 63.1% with semaglutide and 36.8% with placebo. The estimated difference was 26.65 percentage points, with a 95% confidence interval from 7.43 to 45.86.
Was the fibrosis result statistically clear in Japanese participants?
No. Fibrosis improvement without worsening MASH was estimated at 36.5% with semaglutide and 28.9% with placebo. The confidence interval for the difference crossed zero, and the trial was not powered for this subgroup.
Sources
- [1]Nakajima A et al. Semaglutide in Japanese participants with MASH: a subgroup analysis of ESSENCE. PMID 42663669.Tier 1 · primary↩
- [2]Nakajima A et al. Open-access full text at Journal of Gastroenterology. DOI 10.1007/s00535-026-02507-0.Tier 1 · primary↩
- [3]ClinicalTrials.gov. ESSENCE phase 3 trial record (NCT04822181).Tier 1 · primary↩
No revisions yet. First published .