After Melanoma Surgery, a Vaccine Built From the Tumor Itself Cut the Risk of the Cancer Returning

CATEGORY: IMMUNE HEALTH

Surgery can take out a dangerous melanoma. What it can’t take away is the waiting that follows: the scans, and the question of whether the cancer will be back. For some patients, that question now has a better answer. In a randomized trial, people whose high-risk melanoma had been fully removed and who then got a vaccine custom-built from their own tumor, alongside the immunotherapy drug pembrolizumab, had a 49% lower risk of the cancer returning or of dying over five years than people who got the same immunotherapy alone.¹

A Wanted Poster Printed From the Tumor’s Own Mutations

Every melanoma carries its own set of genetic mutations, and some of those mutations produce protein fragments that sit on the cancer’s cells and nowhere else in the body. Researchers call these fragments neoantigens. For each patient in the trial, scientists read the DNA of the removed tumor, picked out up to 34 of them, and wrote them into a personalized mRNA vaccine. Think of it as a wanted poster. The vaccine shows the immune system exactly what this one person’s cancer looks like, so the hunt becomes specific. Everyone in the trial also received pembrolizumab, an infusion that releases the immune system’s natural brakes so its cells can attack the cancer. The trial, called KEYNOTE-942, enrolled 157 people with stage 3 or 4 melanoma that had been completely removed by surgery but was judged likely to return. Chance made the assignment, not doctors and not patients: two of every three got the vaccine plus the drug, and the rest got the drug on its own.

Five Years Later, the Gap Hadn’t Closed

Early results from this trial drew attention in 2023. The question hanging over them was whether a benefit built on brand-new immune cells would last. Five years on, it has. The vaccine group’s risk of recurrence or death stayed 49% lower, and their risk of the cancer spreading to distant organs, or of death, was 59% lower.¹ Fewer people in the vaccine group had died, too, though that difference was small enough that chance could explain it. Side effects were mostly the familiar vaccine kind: sore arms, fatigue and chills. Serious immune-related problems were no more common with the vaccine added, affecting 10.6% of that group versus 12% on pembrolizumab alone. The honest caveat is size. With 157 people, this was a midsized trial, and its five-year numbers describe what happened rather than prove it to a statistical standard.

A Far Larger Trial Already Pointed the Same Way

That proof may be close. In August 2026, Merck and Moderna announced that a final-stage trial of the same vaccine-plus-pembrolizumab combination in over a thousand people with resected melanoma met both of its main goals, with less recurrence and less distant spread. By the companies’ account it’s the first mRNA cancer vaccine ever to clear that bar. The detailed numbers haven’t been published in a journal yet, so the size of that benefit can’t be reported here. The personalized approach is also being tested in lung, kidney and bladder cancers. In pancreatic cancer, among the deadliest of all, a small first-stage study found a similar custom vaccine produced cancer-hunting immune cells that were still on patrol roughly three years later, and the eight patients who responded to it went longer without recurrence than the eight who didn’t.² The road hasn’t been smooth everywhere. A different company’s colorectal cancer vaccine trial was stopped this August after an early look suggested it wasn’t helping. Whether this idea works beyond melanoma is still an open question, and one that dozens of running trials are built to answer.

Key Takeaways

• Five years after treatment, people who got a vaccine built from their own melanoma alongside pembrolizumab had a 49% lower risk of recurrence or death, and a 59% lower risk of distant spread or death, than people on pembrolizumab alone.
• Both groups received the immunotherapy, so the difference is what the personalized vaccine added on top of it.
• The randomized trial was midsized, at 157 people, and its five-year comparisons are descriptive; a final-stage trial was announced as positive in August 2026, but its detailed results aren’t published yet.
• The approach hasn’t succeeded everywhere: a colorectal cancer vaccine trial was stopped early this year, and trials in lung, kidney, bladder and pancreatic cancer are still under way.

References

1. Khattak A, Carlino MS, Meniawy T, et al. Intismeran Autogene Plus Pembrolizumab Versus Pembrolizumab Alone in High-Risk Resected Melanoma: 5-Year Update of the Randomized Phase IIb KEYNOTE-942 Study. Journal of Clinical Oncology. 2026;44(26):2482–2489. link
2. Sethna Z, Guasp P, Reiche C, et al. RNA neoantigen vaccines prime long-lived CD8+ T cells in pancreatic cancer. Nature. 2025;639(8056):1042–1051. link