Metabolic dysfunction-associated steatohepatitis, often called MASH, is a form of fatty liver disease that goes beyond simple fat accumulation. In MASH, the liver becomes inflamed, and over time that inflammation can leave behind scar tissue, a process called fibrosis. When fibrosis progresses far enough, the liver becomes cirrhotic, meaning its architecture is severely disrupted. Researchers have been searching for treatments that can actually reverse that scarring, not just slow it down.
A large phase 2 clinical trial published in The Lancet Gastroenterology and Hepatology set out to answer whether combining two peptide-based compounds could outperform either one alone in reducing liver fibrosis in people with MASH. The trial enrolled 698 participants across 187 clinical sites in 22 countries, making it one of the more geographically broad efforts of its kind. The results were more nuanced than many researchers may have hoped, and they raise important questions about combination strategies in liver disease research.
What the trial was designed to test
The study focused on zalfermin, a fibroblast growth factor 21 analogue, combined with semaglutide, a GLP-1 receptor agonist peptide. Both compounds have shown some activity in metabolic and liver-related endpoints in earlier research. The hypothesis was that their combined mechanisms might work together to produce greater improvements in liver fibrosis than either peptide could achieve separately.
Participants had confirmed MASH with clinically significant liver fibrosis, ranging from stage F2 through F4c on the standard NASH CRN fibrosis scale. The F4c designation means compensated cirrhosis, the most advanced stage included. All participants were adults, and the trial ran for 52 weeks with once-weekly subcutaneous injections.
Seven treatment arms were tested: three doses of zalfermin combined with semaglutide, zalfermin alone, semaglutide alone, a combination of cagrilintide with semaglutide, and a placebo-only group. The primary endpoint was a one-stage or greater improvement in liver fibrosis without any worsening of the underlying MASH condition, assessed at the end of the 52-week period.
Primary endpoint results
The main combination arm, zalfermin 30 mg plus semaglutide, did not produce a statistically significant improvement over placebo on the primary endpoint. In that combination group, 24 percent of participants achieved the fibrosis improvement threshold, compared with 16 percent in the placebo group. The estimated difference was about eight percentage points, but the confidence interval crossed zero and the p-value was 0.19, meaning the result could plausibly have occurred by chance alone.
Lower doses of zalfermin combined with semaglutide also failed to show substantially different results from placebo. The exploratory arm combining cagrilintide with semaglutide similarly did not distinguish itself from the placebo group on this measure.
Zalfermin alone at 30 mg reached a 22 percent responder rate, again not statistically different from placebo. So neither the combination strategy nor zalfermin as a single agent met the primary bar the researchers had set.
The semaglutide signal
The most notable finding in the trial was actually in the semaglutide-alone arm. Thirty percent of participants in that group achieved the primary endpoint, producing a nominally statistically significant difference from placebo with a p-value of 0.024 and a confidence interval that did not include zero.
The authors use the phrase 'nominally significant' deliberately. In a multi-arm trial like this one, the statistical threshold for any single comparison is adjusted to account for the number of tests being run. The semaglutide arm was not the primary comparison, so this result requires careful interpretation rather than a firm conclusion.
Still, the trial authors noted that semaglutide might warrant further clinical assessment as a potential disease-modifying therapy, including specifically in the F4c cirrhosis population. This is a population where treatment options have historically been very limited, and even a nominally significant signal in a well-controlled phase 2 trial is considered worth pursuing in subsequent research.
Safety observations across the arms
Adverse events were common across all groups, but the trial authors characterised most of them as non-serious and mild to moderate in severity. Gastrointestinal side effects were the most frequently reported. In the highest-dose combination group, 80 percent of participants reported gastrointestinal events. The semaglutide-alone group saw a 73 percent rate, and the placebo group reported gastrointestinal events in 51 percent of participants, which is a notable baseline rate likely reflecting the underlying metabolic disease.
Serious adverse events occurred in 7 percent of participants in the zalfermin 30 mg plus semaglutide group, 13 percent in the zalfermin-alone group, 10 percent in the semaglutide-alone group, and 5 percent in the placebo group. Five deaths occurred during the trial. One death, in the zalfermin-alone group, was assessed as possibly related to the study drug and involved heart failure.
The safety profile did not reveal unexpected signals beyond what was already known for these compound classes, but the relatively higher serious adverse event rate in the zalfermin-alone arm and the one drug-related death are findings that researchers and regulatory reviewers will consider carefully in any future programme involving this peptide.
Why combination strategies can be complicated
One of the broader lessons from this trial is that layering two active compounds does not automatically produce additive or synergistic results. In some cases, combining agents can actually dilute the measurable effect of the more active compound, or introduce interactions that alter how each peptide behaves in the body.
In this trial, the semaglutide-alone arm numerically outperformed every combination arm on the primary endpoint. That is counterintuitive if you expect more mechanisms to equal more benefit. Researchers have several hypotheses about why this might happen, including dose competition, pharmacokinetic interactions, or simply the fact that one mechanism may dominate the relevant biology in this disease context.
Early-phase trials like this one are specifically designed to surface these kinds of findings. A phase 2 study is not expected to provide a definitive answer; it is meant to inform what to test next and how. In that sense, the data here do exactly what they are supposed to do, even if the headline result did not confirm the primary hypothesis.
Context for the research community
MASH with significant fibrosis represents an area of active and urgent research. Liver fibrosis has long been considered difficult to reverse, and the histological confirmation required in this trial, actual liver biopsy analysis, is a rigorous standard for measuring biological change rather than just a surrogate marker.
The fact that 30 percent of participants in the semaglutide arm showed fibrosis improvement without MASH worsening after 52 weeks is a meaningful data point, even with the statistical caveats attached to it. For context, only 16 percent of placebo participants met the same bar, and the placebo response in fibrosis trials is often higher than the general public might expect because lifestyle changes during a trial can themselves produce some liver improvement.
Researchers studying the FGF21 pathway, which zalfermin is designed to activate, are likely to analyse the subgroup data from this trial carefully to understand whether there are specific patient profiles in which the combination might still show benefit. The cagrilintide plus semaglutide exploratory arm, while not the focus of the primary analysis, also adds to the growing dataset around amylin-based and GLP-1-based combinations in metabolic liver disease. Future trials will need to determine whether longer durations, different doses, or different patient selection criteria produce different outcomes.
