26 Oct·2 min read

Personalized mRNA cancer vaccine cuts melanoma recurrence risk by 42%

Merck and Moderna's vaccine, built from each patient's own tumor, beat Keytruda alone in a 1,137-person phase 3 trial.

Merck and Moderna's personalized cancer vaccine, intismeran autogene, cut the risk of melanoma coming back or the patient dying by 42% when given with Keytruda, compared with Keytruda alone. The full results from the phase 3 INTerpath-001 trial leaked online on October 22, before their planned presentation at ESMO 2026 in Spain. It is the first positive phase 3 result for an mRNA cancer vaccine. It is also the first for a treatment built around one patient's own tumor mutations.

The trial enrolled 1,137 people with stage IIB to IV melanoma whose tumors had already been removed by surgery. Two thirds got the vaccine every three weeks plus Keytruda every six weeks, for about a year. The rest got Keytruda alone. After a median follow-up of just over two years, 82.2% of the combination group were still recurrence-free, against 70.0% on Keytruda alone. The risk of the cancer spreading to distant organs, or death, fell by 38%. Side effects were mostly mild: fever, tiredness and swelling where the shot went in.

What I find most interesting is how each dose gets made. Doctors sequence the removed tumor and a blood sample. The blood gives your normal DNA and the tumor gives your cancer's DNA. Comparing the two shows which mutations only the cancer has. An algorithm ranks them and picks up to 34 neoantigens, the abnormal protein fragments only that tumor makes. Those get written into a single strand of mRNA and packed into lipid nanoparticles, tiny bubbles of fat that carry the mRNA into cells.

Once it's injected, your cells read the mRNA and make those fragments. That teaches killer T cells what the cancer looks like, so they can hunt down any cancer cells left behind. Keytruda is a checkpoint inhibitor, which means it takes off one of the brakes the immune system normally keeps on. The vaccine shows the T cells the target and Keytruda lets them go after it. Your dose would do nothing for anyone else, because nobody else has your mutations.

There are real caveats. The data hasn't been peer-reviewed yet, and the overall survival numbers are still coming in, so we don't know yet whether people live longer. Each dose takes weeks to build, around six in the trial and longer in some reports. That is a long wait when you have cancer. Investors had hoped for something closer to the earlier phase 2b result, and Moderna's stock fell 21% after the leak. RBC said, "While approval seems likely, commercial success is far less certain."

I think the stock drop misses what matters. Twelve more patients in every hundred still cancer-free at two years, on top of a drug that was already the standard treatment, is a big deal for anyone actually facing melanoma. Approval is expected in 2027, and trials are already running in bladder, kidney and lung cancer. If it holds up there, a vaccine written from your own tumor could become a normal part of cancer treatment.