Merck and Moderna mRNA cancer vaccine intismeran cuts melanoma recurrence risk by 42% in phase 3 trial

For decades, the idea that a vaccine could be tailored to an individual patient's tumour and then deployed to stop that cancer from coming back has sat at the edge of oncology, more thought experiment than treatment plan. Melanoma, a skin cancer with a stubborn tendency to return after surgery, has long been the testing ground for that ambition because its mutation burden is high and its tumours are relatively accessible to sequencing. On October 22, 2026, that ambition moved a significant step closer to the clinic.

Merck & Co., known as MSD outside the United States and Canada, and Moderna announced detailed results from the phase 3 INTerpath-001 trial of intismeran autogene (also known as V940 or mRNA-4157), an mRNA-based individualized neoantigen therapy given with KEYTRUDA (pembrolizumab) as adjuvant treatment for completely resected stage IIB-IV melanoma. The headline number was a 42% reduction in the risk of recurrence or death compared with pembrolizumab alone.

The data were presented at the European Society for Medical Oncology Congress 2026 in Madrid as Presentation #LBA1, in a presidential symposium slot normally reserved for practice-changing results, and were accepted for publication in The New England Journal of Medicine. The path to that stage, however, was anything but smooth: the key results leaked online on the morning of October 22, two days before the planned presentation, forcing the companies to publish their release early and triggering a sharp selloff in Moderna shares.

Intismeran is not a preventive vaccine. It is made for each patient after surgery, using DNA sequenced from that person's tumour to identify mutations that give rise to neoantigens, the protein fragments that distinguish cancer cells from healthy tissue. The resulting synthetic mRNA codes for up to 34 of those neoantigens, training the immune system to recognise and attack any residual cancer cells carrying them.

Key Facts

At the primary pre-specified intention-to-treat analysis, after a median follow-up of 24.6 months, the combination of intismeran plus pembrolizumab reduced the risk of recurrence or death by 42% versus pembrolizumab alone, with a hazard ratio of 0.58 (95% CI 0.45-0.75; two-sided p<0.001). The risk of distant metastasis or death fell by 38% (HR=0.62, 95% CI 0.45-0.85; p=0.002). The trial enrolled 1,137 patients with completely resected stage IIB-IV melanoma across 157 centres, randomised 2:1 to intismeran 1 mg every three weeks for up to nine doses plus pembrolizumab 400 mg every six weeks for up to nine cycles, or placebo plus pembrolizumab, for about one year.

Two-year recurrence-free survival was 82.2% on the combination versus 70.0% on pembrolizumab alone, and two-year distant metastasis-free survival was 88.5% versus 80.2%. Median recurrence-free survival and median distant metastasis-free survival were not reached in either arm, a reflection of how few events had accrued by the cutoff. Improvements were generally consistent across prespecified subgroups. Overall survival data are not yet mature but showed a favourable trend.

Timing mattered. During the first six months, efficacy was comparable between arms for recurrence or death (HR=0.99; 95% CI 0.66-1.49) and for distant metastasis or death (HR=1.01; 95% CI 0.58-1.79). After the six-month immunization regimen was completed, a prespecified sensitivity analysis showed a 58% reduction in recurrence or death (HR=0.42, 95% CI 0.30-0.58) and a 50% reduction in distant metastasis or death (HR=0.50). In the modified intention-to-treat population the reductions were 49% and 45%.

Safety looked manageable and consistent with what is already known about both components. Treatment-related adverse events occurred in 95.8% of the combination arm versus 90.0% of the monotherapy arm, with grade 3-5 events in 14.2% versus 13.4%. The most common events were pyrexia (52.2%), injection site pain (51.0%) and fatigue (38.5%), and the companies reported no new safety signals.

The reaction from investigators was enthusiastic. Georgina Long, medical director of Melanoma Institute Australia and chair of Melanoma Medical Oncology at the University of Sydney, who served as principal investigator, called the results potentially a foundational shift in adjuvant melanoma treatment. Moderna chief executive Stephane Bancel described the data as a defining moment for mRNA cancer therapy, while Merck Research Laboratories president Dean Y. Li said the trial sets a new benchmark in the adjuvant setting.

Analysis

The clinical readout and the market readout told two very different stories about the same dataset. Nature reported on October 23, 2026 that independent oncologists regard a 10% risk reduction as substantial, and that a reduction above 40% would send many researchers over the Moon. Kristen Radford, an immunotherapist at Mater Research in Brisbane, and Seth Cheetham, a geneticist at the University of Queensland, made that point directly. By that standard, a 42% reduction in recurrence or death is not a marginal improvement. It is the kind of effect size that changes how an adjuvant regimen is designed.

Investors saw something else. AJMC reported on October 22, 2026 that the primary recurrence-free survival result fell at the low end of investor expectations, because investors had anchored on the phase 2b KEYNOTE-942 hazard ratio of 0.51. BioSpace reported on October 22, 2026 that analysts had hoped to see a hazard ratio below 0.60, so the 0.58 print cleared the clinical bar but missed elevated hopes. RBC Capital Markets wrote that while approval seems likely, commercial success is far less certain, raising questions about the value proposition relative to neoadjuvant immunotherapy, which produces similar efficacy at potentially much less cost. BMO Capital Markets cautioned against extrapolating the melanoma result to other tumour types, while noting the data supported a potential commercial opportunity.

The stock moves were brutal on one side and muted on the other. Shares of Merck fell just over 1% to $147.93, while Moderna dropped 21% to $193.65, according to BioSpace. Nature reported that Moderna's stock had risen since the August announcement that the trial met its primary endpoint but dropped by around 25% after the leak. BMO analyst Evan Seigerman said the reaction probably reflected inflated expectations rather than a true indictment of the therapy's efficacy, and William Blair noted that Moderna shares were still up 525% for the year even after the decline.

What this really means is that the scientific question and the investment question have decoupled. The trial delivered a statistically robust, clinically meaningful result in a disease where recurrence after surgery is the central fear, and it did so with a safety profile close to pembrolizumab alone. Whether that translates into a durable commercial franchise depends on questions the trial was never designed to answer: how the benefit holds beyond two years, whether it extends to other tumours, and how payers will weigh a bespoke manufacturing process against cheaper alternatives. Jessamy Tiffen, a cancer geneticist at the Centenary Institute in Sydney, made the first point plainly, cautioning that only time will tell how effective the treatment is beyond two years and that some people did not respond.

Why It Matters

The bigger picture here is that this is the first phase 3 win for an individualized mRNA cancer therapy, and the mechanism is the message. Each dose of intismeran is built from a patient's own tumour sequence, which means the manufacturing chain runs from surgery to sequencing to a custom mRNA batch to infusion, all within a treatment window that matters. Merck's Jane Healy, head of oncology early clinical development, told BioSpace that the results had met the company's bar for efficacy, calling the benefit not subtle and crediting genome sciences, AI and machine learning, and mRNA technology. That combination of tools is what makes a 34-neoantigen bespoke product feasible at trial scale.

For patients, the practical significance of the delayed separation is worth dwelling on. The curves were essentially superimposed for the first six months, when both arms were still receiving pembrolizumab, and only diverged after the immunization regimen finished. That pattern is consistent with an adaptive immune response that takes time to mature and then persists, and it is the kind of signal that supports the vaccine doing real immunological work rather than simply adding toxicity to an existing drug. Long noted that roughly 15 years ago people with advanced melanoma typically survived only months, while today nearly 55% are effectively cured with targeted drugs or immunotherapies.

For the field, the result also resets expectations for what a cancer vaccine trial can deliver. A 42% reduction in recurrence or death in resected stage IIB-IV melanoma gives the mRNA individualized neoantigen platform a registrational-grade dataset, and it does so in front of an audience at ESMO that has watched several vaccine programmes fail. The leak and the selloff are a reminder that scientific communication and market communication now run on different clocks, and that a result can be simultaneously a landmark and a disappointment depending on who is reading it.

Next Up

Merck and Moderna said they are engaging with regulatory authorities on the basis of the INTerpath-001 data. The companies are running nine phase 2 and phase 3 trials of intismeran, including three phase 3 trials in non-small cell lung cancer, which will test whether the melanoma result travels beyond a tumour type chosen partly for its high mutation burden. The full dataset was set to be presented on Saturday at ESMO, and publication in The New England Journal of Medicine is pending.

Long said participants have been followed for around 25 months and will be monitored for several years to determine overall survival, which remains the outcome that regulators and payers will scrutinise most closely. Until those mature data arrive, the 42% figure stands as the strongest evidence yet that a vaccine built around an individual tumour can meaningfully change the course of melanoma after surgery, and as an open question about how much of that value the market is willing to pay for.