A personalised mRNA vaccine cuts melanoma relapse after surgery

A cancer vaccine built from each patient's own tumour has cut the risk of melanoma coming back after surgery in a phase 3 trial. In INTerpath-001, people who had a high-risk melanoma removed and then received the personalised mRNA treatment intismeran autogene alongside standard immunotherapy were 42% less likely to relapse or die during follow-up than people given immunotherapy and a placebo. At two years, 82.2% of the vaccine group were free of recurrence, against 70.0% of the placebo group.

Georgina Long of Melanoma Institute Australia presented the results as the first late-breaking abstract (LBA1) of the Presidential Symposium at the ESMO Congress 2026 in Madrid on 24 October. Merck (known as MSD outside the US and Canada) and Moderna, which developed the treatment together and ran the trial, said in a press release on 22 October that the results had been accepted by the New England Journal of Medicine. MedicalDaily, citing OncLive, reported that the paper is slated for 11 November. Until it appears, the numbers here come from the congress presentation, the companies' statements and medical press coverage.

What was tested

Melanoma is the most serious common form of skin cancer. When it's caught at stage IIB or later, removing it surgically isn't always the end of the story, because cancer cells may already have spread. Since the late 2010s, people in this position have often been offered a year or so of immunotherapy after surgery, called adjuvant treatment, to lower the chance of relapse. Pembrolizumab (brand name Keytruda) is one of the standard options. It blocks PD-1, a brake on the immune system's T cells.

Intismeran autogene (also known as V940 or mRNA-4157) adds a personalised element. After surgery, the removed tumour is genetically sequenced and compared with the patient's normal DNA to find mutations that only the cancer carries. Those mutations produce altered proteins, called neoantigens, that the immune system can learn to recognise as foreign. A computer selects up to 34 of the most promising ones, and Moderna manufactures an mRNA molecule encoding them, a different product for every patient. Once injected, the mRNA instructs the body's cells to make the neoantigens, training T cells to hunt down cancer cells carrying them. Pembrolizumab then keeps those T cells from being switched off.

In the trial, patients received intismeran at 1 mg as an injection into the muscle every three weeks for up to nine doses, or a placebo injection. Everyone also had pembrolizumab at 400 mg through a drip every six weeks for up to nine cycles. The trial was double-blind, so neither patients nor their doctors knew who had the vaccine.

Who took part

The trial randomised 1,137 people, two to the vaccine group for every one to placebo: 757 had intismeran and 380 had placebo. All had cutaneous melanoma (melanoma of the skin) at stage IIB, IIC, III or IV under the AJCC eighth edition system, and all had had their cancer completely removed. None had had any previous drug treatment for melanoma, they had to be well enough to carry on with normal daily activity, and they needed a tumour sample good enough to sequence. CancerNetwork reported a median age of 61, with about 60% of participants male.

The sample requirement turned out to be a real filter. According to the full presentation, as reported by MedicalDaily, 302 of 1,749 people screened (17.3%) were excluded because their tumour sample wasn't adequate. That's about one in six.

The registry entry is NCT05933577 on ClinicalTrials.gov.

Results in plain numbers

The main measure was recurrence-free survival, the time until the melanoma came back anywhere or the patient died, as judged by the treating investigators. If that was positive, the trial went on to test distant metastasis-free survival (the time until melanoma spread to distant organs, or death) and then overall survival, in that order.

The data cut-off was 30 June 2026, with a median follow-up of 24.6 months.

Sources: Merck press release (22 October 2026); event counts and 24-month rates from OncoDaily's summary of the ESMO abstract; overall survival from Targeted Oncology.

The hazard ratio of 0.58 is the source of the "42% lower risk" figure. In absolute terms, after two years about 12 more people in every 100 were still free of melanoma with the vaccine than without it. For distant spread, which is the form of relapse most likely to threaten life, the gain was about eight people per 100.

Overall survival is the number everybody wants, and it isn't there yet. The hazard ratio of 0.77 leans in favour of the vaccine, but the confidence interval runs from 0.45 (a large benefit) to 1.31 (some harm), and the result didn't meet the trial's threshold for significance (P = 0.33). With most patients still alive and well, that's expected at two years. It simply means the survival question is open.

The first six months

One detail from the presentation explains a lot about how the treatment works. A piecewise analysis, reported by Targeted Oncology, found no difference at all between the groups during the first six months (HR 0.99), and a strong difference afterwards (HR 0.42). MedicalDaily reported that the median time to the first vaccine dose was 64 days, because each batch has to be made after surgery. It takes time to manufacture the vaccine and then for the immune response to build, so any benefit appears late.

Who benefited

The benefit on relapse showed up in both lower-risk stages (IIB to IIIB, HR 0.60) and higher-risk stages (IIIC to IV, HR 0.57), according to OncoDaily. For distant spread, the lower-risk group's confidence interval crossed 1 (HR 0.65, 95% CI 0.41 to 1.02), so that result is less certain. MedicalDaily noted one US-only subgroup result for distant spread that went the other way, but it rested on just 10 events, too few to interpret.

Side effects

Nearly everyone in both groups had some side effect, which is normal for immunotherapy.

Sources: Merck press release; CancerNetwork report of the ESMO presentation.

The vaccine's own side effects were mostly the kind you'd expect after any vaccination: fever, a sore arm and flu-like symptoms. Investigators described these as low grade and short-lived. Serious side effects were only slightly more common with the vaccine, and the vaccine didn't make pembrolizumab's immune-related effects (such as thyroid problems or colitis) any more frequent.

More people did stop at least one study drug because of side effects on the vaccine arm (14.0% against 8.7%). There were two treatment-related deaths in the vaccine group, both attributed to pembrolizumab, and none attributed to intismeran.

Limitations

The full peer-reviewed paper isn't out yet. Figures from congress presentations occasionally change slightly in the final publication.

Overall survival is unknown. Recurrence-free survival is a sensible early measure in this setting, but a delay in relapse doesn't automatically translate into longer life, especially when people who relapse can still receive effective treatment.

The comparator was pembrolizumab alone. In recent years, giving immunotherapy before surgery (neoadjuvant treatment) has become an option for some people with stage III melanoma. MedicalDaily, citing AJMC, reported that analysts at RBC Capital Markets questioned the vaccine's value against neoadjuvant immunotherapy, which may achieve similar results at lower cost. INTerpath-001 didn't test that comparison.

Only people with a good tumour sample could take part. In practice, about one in six people screened weren't eligible on that basis alone. If the treatment is approved, how tumour tissue is handled at surgery will matter.

The trial was sponsored and run by the two companies developing the vaccine.

The cost hasn't been announced. A treatment made individually for each patient is unlikely to be cheap, which will matter for NHS and insurer decisions.

What it means for patients

This is the first phase 3 trial to show that a personalised mRNA cancer vaccine lowers the chance of cancer returning. Long told the congress, as reported by Targeted Oncology, that INTerpath-001 "sets a new benchmark for adjuvant therapy in melanoma". The companies say they are talking to regulators. No approval exists anywhere, and no regulatory decision date had been announced at the time of writing.

For anyone with stage IIB to IV melanoma, the practical points are:

• Standard treatment remains standard. Nobody should delay recommended care waiting for an unapproved product.
• Ask whether your tumour tissue is stored in a way that would allow sequencing later. Tissue that's used up or poorly preserved can rule people out.
• Ask whether neoadjuvant immunotherapy is an option for your stage, since it's a different route that INTerpath-001 didn't compare against.
• Ask about trials. Intismeran is being tested in other cancers too, and ClinicalTrials.gov lists trial sites.

Overall survival follow-up continues, and the NEJM paper should add detail when it appears.

This article explains published research and isn't medical advice. Talk to your own specialist team about treatment decisions.