mRNA cancer vaccines worked against melanoma. Now the field is asking the harder question.

An mRNA cancer vaccine — Merck and Moderna's adjuvant therapy for melanoma — improved the time before cancer came back and spread in a 1,137-patient Phase 3 trial, the first positive Phase 3 readout for an individualized mRNA cancer vaccine. Two months later, a BioNTech mid-stage trial of a similar personalized vaccine in colorectal cancer was halted after a survival imbalance was observed between arms. The pair is sharpening a real question: which cancers can this platform reach, and which ones can't it?

The Merck/Moderna result, announced Aug. 19, 2026, was the readout the field had been waiting for. INTerpath-001 enrolled patients with completely resected stage IIB-IV melanoma — meaning the tumor had been surgically removed but the risk of recurrence was real — and randomized them 2:1 to receive the personalized mRNA vaccine (intismeran autogene, branded as V940/mRNA-4157) plus Keytruda, or Keytruda alone. The vaccine arm met both its primary endpoint of recurrence-free survival and the key secondary endpoint of distant metastasis–free survival. Adjuvant Keytruda is itself a high bar in this setting, and adding the personalized vaccine pushed past it.

Karen Knudsen, CEO of the Parker Institute for Cancer Immunotherapy, told the field's trade press the community was waiting "with bated breath" for this readout. The wait, at least for melanoma, is over.

How the vaccine works explains the mixed results elsewhere. Each dose is built from a tumor sample: a sequencing machine reads the mutations unique to that patient's cancer, a bioinformatic pipeline picks the mutations most likely to be visible to the immune system, and an mRNA molecule encoding those neoantigens is manufactured and injected. The immune system then learns to recognize the tumor's specific signature. COVID taught the manufacturing and delivery side of mRNA; the harder part has always been choosing the right targets and matching them to the right cancer.

The harder part is showing up in colorectal cancer, where BioNTech's mid-stage Phase 1/2 trial was halted after a survival imbalance was observed between arms. The details — which direction the imbalance went, whether it reflected futility or harm — are not yet in public filings. Treating it as either a safety crisis or a clean negative would be guessing. The honest read is that the same platform that beat pembrolizumab's bar in melanoma ran into a wall in a different tumor type, and the wall is exactly the kind of thing Phase 3 is supposed to find.

That is not collapse. It's sorting. A platform that worked in one cancer is now being asked, in trial after trial, whether it works in others. The melanoma result says the technology can do something real; the colorectal halt says the biology is going to decide, cancer by cancer, where it lands. AJMC's coverage and ecancer's independent summary both frame the INTerpath-001 readout as the first Phase 3 win in a category the FDA has never approved.

The next readouts on the platform's calendar are non-small-cell lung cancer for Moderna's mRNA-4157 and renal cell carcinoma for BioNTech's BNT122. The melanoma answer is in. The colorectal answer is "not yet." Each new readout will tell the field and patients which cancers the platform can actually reach.