Moderna and Merck’s mRNA Cancer Vaccine Succeeds in Landmark Phase 3 Melanoma Trial

Moderna and Merck have reported a milestone: their personalised mRNA cancer vaccine, given alongside Merck’s immunotherapy Keytruda, succeeded in a pivotal Phase 3 melanoma trial — the first positive Phase 3 result for an mRNA-based, individualised cancer vaccine.

INTerpath-001 enrolled 1,137 patients whose high-risk melanoma (stage IIB–IV) had been surgically removed, randomising them to intismeran autogene plus Keytruda, or Keytruda alone.

How a personalised vaccine is made

These are neoantigen vaccines, and each is built for one patient.

A tumour is sequenced to identify mutations unique to that cancer. Some of those mutations produce altered proteins — neoantigens — that exist nowhere in healthy tissue, which makes them ideal immune targets: attacking them cannot damage normal cells because normal cells do not carry them.

Software selects which neoantigens are most likely to be displayed on the tumour’s surface and recognised by that individual’s immune system, and an mRNA vaccine encoding them is then manufactured — a bespoke product per patient, produced in weeks.

Why it is paired with a checkpoint inhibitor

The combination is not additive; the two components address different halves of the same problem.

A vaccine generates T cells recognising the tumour. A checkpoint inhibitor like Keytruda removes the brakes that tumours use to shut those T cells down once they arrive.

Either alone is incomplete. Vaccines have historically generated measurable immune responses that failed to control tumours, plausibly because the T cells were suppressed on arrival. Checkpoint inhibitors work only where T cells recognising the tumour already exist. Supplying both is the logical fix, and this trial tested it.

The results

The combination produced a statistically significant, clinically meaningful improvement in recurrence-free survival — the primary endpoint — and in distant metastasis-free survival, a key secondary endpoint, compared with Keytruda alone. No new safety signals were seen.

The distant metastasis endpoint matters as much as the primary one. Local recurrence is serious and frequently treatable; distant metastasis is what kills melanoma patients. Improving both suggests the effect is on the disease process rather than on where recurrence happens to be detected.

Why the adjuvant setting was chosen

These patients had their melanoma surgically removed — this is adjuvant treatment, aimed at the microscopic disease that may remain after surgery and cause recurrence months or years later.

That setting suits a vaccine particularly well. The tumour burden is minimal, so the immune system faces scattered residual cells rather than an established mass with its own suppressive environment. There is also time: generating an immune response takes weeks, which is unacceptable in advanced disease and entirely reasonable after surgery.

Why this result carries weight beyond melanoma

Neoantigen cancer vaccines have been an aspiration for years, and the field has produced a long run of encouraging early-phase results that did not survive rigorous testing.

This is the first randomised Phase 3 aimed at definitively proving the benefit. A properly powered randomised trial against an active comparator is a different standard from the single-arm and small randomised studies that preceded it.

It also revives the promise of mRNA technology beyond COVID-19. The platform’s pandemic success demonstrated it could be manufactured at scale and delivered safely; whether it could do therapeutic work was open. Individualised manufacturing in weeks per patient is exactly what mRNA enables and conventional vaccine production does not.

The manufacturing question

Which raises the practical constraint. Every dose is a distinct product requiring tumour sequencing, computational neoantigen selection, and manufacture of a unique mRNA construct — then quality testing and release for a batch of one.

Doing that for a trial of 1,137 patients is an achievement. Doing it for a commercial melanoma population, then for the other cancers this approach might address, is a different scale of problem, and the cost per patient will be substantial.

What remains

The results are a topline readout; full data and any approval decision are still to come, and the companies are expected to move toward regulatory filings.

Why melanoma was the right disease to try first

The choice of melanoma is not incidental, and it explains both why the result arrived here and why extending it elsewhere will be harder.

Melanoma carries one of the highest mutation burdens of any cancer, largely because ultraviolet damage generates mutations at enormous rates. A high mutation burden means many neoantigens, which gives a personalised vaccine plenty of material to work with. Cancers with few mutations offer correspondingly fewer targets.

Melanoma is also unusually responsive to immunotherapy generally — checkpoint inhibitors work better here than almost anywhere else, indicating an immune system already capable of engaging these tumours when unblocked.

So this was the setting most likely to succeed. Whether the approach transfers to lower-mutation cancers such as pancreatic or prostate, where checkpoint inhibitors largely fail and neoantigens are scarcer, is a genuinely open question that this result does not answer — and it is where most of the commercial value would sit.

The figures that matter are the effect size and how it holds over time — recurrence-free survival benefits can narrow with longer follow-up if the vaccine delays rather than prevents recurrence. Whether the benefit eventually translates into overall survival is the question adjuvant melanoma trials take years to answer. Clinical news, not medical advice.