Personalized mRNA Cancer Vaccine Cuts Melanoma Recurrence Risk by 42% in Phase 3 Trial

New clinical trial results support the efficacy of a personalized mRNA therapy for cutaneous malignant melanoma (skin melanoma), one of the more aggressive skin cancers. The phase 3 trial, INTerpath-001, enrolled adult patients with stage IIB to IV disease whose lesions had been completely removed by surgery. It tested a combination of the immune checkpoint inhibitor pembrolizumab and "intismeran autogene" (development codes: mRNA-4157/V940), a personalized mRNA neoantigen therapy designed from the genetic mutations in each patient's tumor.

The term "cancer vaccine" often brings to mind preventive shots given to healthy people, such as the HPV vaccine that protects against cervical and other cancers. This investigational drug is not that kind of vaccine and does not prevent cancer from developing. It is an "adjuvant therapy" given after surgery to patients who have already developed melanoma and had all visible disease removed. Its aim is to use the immune system to eliminate microscopic cancer cells that may remain in the body after surgery, thereby suppressing recurrence and distant metastasis.

The results were presented in a Presidential Symposium at the European Society for Medical Oncology congress (ESMO 2026). Although the paper has reportedly been accepted by the New England Journal of Medicine, the full peer-reviewed text and DOI have not been made public. At this stage, the available sources are the conference presentation materials, the trial registry entry (ClinicalTrials.gov: NCT05933577), and data released by the developers. Based on these, we examine what the widely reported "42% lower risk of recurrence or death" figure actually means, how large the real-world benefit is, and what questions remain about long-term survival.

A Personalized mRNA Therapy That Prevents Post-Surgery Recurrence, Not Cancer Itself

In adjuvant treatment of melanoma, pembrolizumab, an anti-PD-1 antibody, is one of the standard options. Cancer cells can evade immune attack by sending inhibitory signals through molecules such as PD-L1 to the PD-1 receptor on the surface of T cells. Pembrolizumab binds to PD-1 and blocks this inhibitory signal, making it easier for T cells to attack cancer cells. However, pembrolizumab is not a drug tailored to each patient, and it does not by itself present antigens unique to an individual's tumor to the immune system.

Intismeran autogene, by contrast, is a personalized therapy manufactured from the genetic information of the patient's own tumor. First, DNA and RNA information is obtained from tumor tissue removed in surgery and from normal tissue, and genetic mutations are analyzed with a next-generation sequencer. From these results, up to 34 "neoantigens" are selected: antigens that are absent from normal cells and derive from mutations in tumor cells. The information on the selected neoantigens is incorporated into a single synthetic mRNA, which is encapsulated in lipid nanoparticles (LNPs) and administered by intramuscular injection.

The administered mRNA is taken up by antigen-presenting cells such as dendritic cells, which synthesize the proteins corresponding to the neoantigens. These cells present fragments of those proteins on their surface via MHC molecules, inducing an immune response by T cells that recognize and attack cancer cells carrying the same antigens. By teaching the immune system to recognize the features of each patient's particular tumor, the therapy aims to prompt attacks on cancer cells that may remain after surgery.

In this combination, pembrolizumab releases the immune suppression acting on T cells, while intismeran autogene induces an immune response against the tumor antigens that serve as targets. By combining these distinct actions, the goal is to eliminate minimal residual disease (MRD) after surgery more effectively. However, even if such a mechanism of action is theoretically sound, it does not necessarily reduce recurrence in actual patients. Confirming clinical efficacy requires a rigorous trial against standard treatment.

Design and Patient Population of the Phase 3 INTerpath-001 Trial

Efficacy was tested in INTerpath-001, an international, randomized, double-blind, phase 3 trial. It enrolled 1,137 adult patients with stage IIB, IIC, III, or IV cutaneous melanoma that had been completely resected. Patients were randomly assigned in a 2:1 ratio to receive either intismeran autogene plus pembrolizumab or placebo plus pembrolizumab.

An important point is that the control group also received pembrolizumab, the existing standard treatment. In other words, the trial did not simply compare treated patients with untreated ones; it tested how much benefit is gained by adding a personalized mRNA product to standard care. Double blinding also limits the influence on assessment of patients and investigators knowing the treatment assignment.

Intismeran autogene or placebo was given at 1 mg intramuscularly every 3 weeks for up to 9 doses, and pembrolizumab at 400 mg intravenously every 6 weeks for up to 9 cycles. The treatment plan spans about 56 weeks, and dosing is discontinued if recurrence is confirmed or unacceptable toxicity occurs.

The primary endpoint was recurrence-free survival (RFS), which measures the time from randomization until the investigator confirms local recurrence, regional lymph node recurrence, or distant metastatic recurrence, or until death from any cause. Secondary endpoints included distant metastasis-free survival (DMFS), overall survival (OS), and safety.

There are, however, clear limits on the patients to whom the results apply. All participants were disease-free at enrollment, with no lesions detectable after complete resection, and in principle had no history of systemic therapy other than surgery. It therefore cannot be assumed that the same effect would be obtained in unresectable advanced melanoma or in conditions with different biological characteristics, such as mucosal melanoma or uveal melanoma arising in the eye.

Hazard of Recurrence or Death Down 42%; 2-Year Recurrence-Free Survival Up 12.2 Points

In the prespecified intention-to-treat (ITT) analysis at a median follow-up of 24.6 months, the combination group showed a hazard ratio of 0.58 for the primary endpoint, recurrence-free survival (RFS), versus the control group (95% confidence interval: 0.45–0.75; two-sided p < 0.001). The difference was statistically significant, showing that adding the personalized mRNA product lengthened the time to recurrence or death.

In general-audience coverage, this result is often described as a "42% reduction in the risk of recurrence or death." That does not mean the proportion of patients who relapse fell by 42 percentage points. A hazard ratio of 0.58 is a statistical measure indicating that the hazard of experiencing recurrence or death during follow-up was 42% lower in relative terms than in the control group. To understand the actual size of the treatment effect, it is necessary to look not only at such relative measures but also at the absolute difference in recurrence-free survival rates.

{
  "type": "bar",
  "title": "2年時点の推定無再発生存率（RFS）比較",
  "unit": "%",
  "categories": ["2年無再発生存率"],
  "series": [
    { "name": "intismeran併用群", "values": [82.2] },
    { "name": "ペムブロリズマブ単独群", "values": [70.0] }
  ],
  "caption": "追跡期間中央値24.6か月時点の推定値。絶対差は12.2パーセントポイント。",
  "source": "ESMO 2026発表資料およびスポンサー公表データ"
}

The estimated 2-year recurrence-free survival rate was 82.2% in the intismeran combination group, compared with 70.0% in the pembrolizumab-alone group, an absolute difference of 12.2 percentage points. Even with standard treatment alone, about 70% of patients remained free of recurrence for two years; adding the personalized mRNA product raised that proportion by a further 12.2 points. It is significant that a clinically meaningful difference was confirmed, not just a relative reduction in hazard.

The secondary endpoint of distant metastasis-free survival (DMFS) also favored the combination group. Distant metastasis or death occurred in 89 patients (11.8%) in the combination group and 71 patients (18.7%) in the control group. The hazard ratio was 0.62 (95% CI: 0.45–0.85; two-sided p = 0.002), indicating a 38% relative reduction in the hazard of distant metastasis or death. Estimated 2-year DMFS rates were 88.5% in the combination group and 80.2% in the control group, suggesting efficacy in suppressing distant metastasis, a key goal of adjuvant therapy.

The main efficacy results are summarized below.

The detailed results from Moderna and Merck also include a different figure, a "58% risk reduction." This comes not from the trial's primary analysis but from a prespecified piecewise hazard analysis (piecewise sensitivity analysis) covering the period from 6 months after the start of treatment. That analysis yielded a hazard ratio of 0.42 (95% CI: 0.30–0.58), but the absolute event rates in the two compared groups are not shown in the publicly available materials we could confirm. Because it is thought to take some time for an mRNA product to generate an antigen-specific immune response, the treatment effect may change over time. However, this result alone does not prove how the effect changes over time or why, and it should be regarded strictly as a sensitivity analysis.

Whether overall survival (OS) was extended, on the other hand, cannot yet be concluded. In the analysis at a median follow-up of 24.6 months, the hazard ratio for overall survival was 0.77 (95% CI: 0.45–1.31; two-sided p = 0.33), which does not meet the prespecified criteria for statistical significance. Few deaths have occurred in either group, so the data needed for an adequate assessment have not yet accumulated. Although the trial showed an effect in suppressing recurrence, whether this ultimately translates into longer patient survival must be confirmed in future follow-up.

Limited Increase in Side Effects, but Challenges Remain in Individualized Manufacturing

Alongside efficacy, the physical burden added by the extra treatment matters. According to Moderna's announcement, the incidence of grade 3–5 treatment-related adverse events (TRAEs) was 14.2% in the intismeran combination group, compared with 13.4% in the pembrolizumab-alone group, a difference of only 0.8 percentage points. The developers state that this is consistent with the safety profile seen in the phase 2 trial and that no new safety concerns were identified.

However, similar rates of severe adverse events do not necessarily mean the burden of side effects is the same. With mRNA products, acute reactions such as fever, fatigue, muscle pain, and injection-site reactions can be an issue, and immune-related adverse events (irAEs) from pembrolizumab also require attention. The frequency and duration of mild to moderate adverse events, and details of cases that led to discontinuation, cannot be adequately assessed from summary data alone. A more accurate safety judgment requires detailed reporting in a peer-reviewed paper and subsequent long-term follow-up.

In addition, individualized manufacturing, the defining feature of intismeran autogene, is a major challenge for introduction into actual clinical practice. Unlike conventional drugs, the same product cannot be mass-produced and supplied to medical institutions; tumor information that differs for each patient must be analyzed and a dedicated mRNA product manufactured for each one. Because the manufacturing process differs from patient to patient, sufficient production capacity must be maintained while ensuring quality and supply stability.

The first requirement is to secure tissue samples of sufficient quantity and quality from the resected tumor for genetic analysis. The process then proceeds through sequencing to identify mutations, algorithm-based selection of neoantigen candidates, mRNA synthesis, encapsulation in LNPs, and quality control testing before the product reaches the patient. If these steps take weeks to months, the key question is whether the product can be delivered without missing the limited treatment window after surgery.

Manufacturing individual products in parallel for patients around the world also requires manufacturing facilities, logistics, and quality management systems. Some speculate that individualized manufacturing could cost tens of millions of yen per patient, but little information has been published on actual manufacturing costs or future drug prices. How far manufacturing costs can be reduced, and how costs would be covered under each country's health insurance system, will also influence how widely the therapy spreads.

In adjuvant therapy, treatment must be started at an appropriate time with the risk of recurrence in mind. It is therefore also necessary to consider how to handle patients who relapse before their product is completed because manufacturing takes time. While this trial demonstrated the clinical efficacy of a personalized mRNA product, it did not resolve the practical question of whether it can be delivered reliably to many patients in routine care.

Effects on Long-Term Survival and Application to Other Cancers Still Need Testing

Following the ESMO 2026 presentation, researchers and the developers have expressed hope that personalized mRNA therapy could significantly change cancer treatment. Showing improved recurrence-free survival by adding treatment that targets a patient's unique tumor antigens to an existing immune checkpoint inhibitor is highly significant. However, the clinical effects confirmed so far need to be considered separately from the outcomes hoped for in the future.

First, as of this announcement, intismeran autogene is not a drug approved by regulators in Japan, the United States, Europe, or elsewhere. Showing efficacy in a phase 3 trial is not the same as becoming available in medical practice. Regulators must examine not only clinical efficacy and safety but also quality control and reproducibility in the process of manufacturing a different product for each patient. Whether approval will be granted, and which patient populations would be eligible if it is, must await future decisions.

Second, there is the limitation of a median follow-up of 24.6 months for long-term suppression of recurrence. In melanoma, recurrence can also occur more than two years after surgery, so it is unknown whether the treatment effect confirmed here will be maintained at 5 or 10 years. Resistance to treatment could also emerge over the long term, for example through "antigen loss," in which tumors lose the targeted neoantigens, or "immune exhaustion," in which T-cell function declines. Long-term follow-up is needed to clarify how often such phenomena actually occur and whether the recurrence-suppressing effect ultimately translates into improved overall survival.

Third, there is the question of whether the results can be applied to other solid tumors. Melanoma is known as a cancer with a relatively high tumor mutation burden (TMB), the number of genetic mutations carried by tumor cells, owing to factors such as DNA damage from ultraviolet radiation. The more mutations there are, the more neoantigens the immune system can recognize, which is one of the conditions under which immune checkpoint inhibitors work well. These favorable results may be related to such biological properties specific to melanoma.

Indeed, in the development of other companies' mRNA cancer vaccines for cancers such as colorectal and head and neck cancer, some randomized controlled trials failed to show the expected effect in suppressing recurrence, forcing a review of development plans. The success of the phase 3 trial in melanoma therefore does not mean that mRNA vaccines will work against every solid tumor. Efficacy must be confirmed in separate clinical trials for each cancer type, taking into account its mutational characteristics, tumor microenvironment, antigen-presentation mechanisms, and degree of immune suppression.

The INTerpath-001 trial is an important study showing that, in patients with high-risk cutaneous melanoma after complete resection, a personalized mRNA neoantigen therapy adds benefit to standard immune checkpoint inhibitor therapy. The 12.2-percentage-point difference in estimated 2-year recurrence-free survival is a major step toward the practical use of personalized cancer immunotherapy. At the same time, the effect on overall survival has not been established, and questions remain about long-term safety and the practicality of individualized manufacturing. Going forward, detailed data from the peer-reviewed paper, long-term follow-up results, and the course of regulatory review will be important in judging the clinical position of this therapy.