Five Years of Proof: The Personalized mRNA Cancer Vaccine That Halved Melanoma Recurrence

At ASCO 2026, five-year data from the KEYNOTE-942 trial confirmed what immunologists suspected: a personalized mRNA vaccine can train the immune system to remember cancer for half a decade, halving recurrence risk without added toxicity.

Presented By Our Partners

When Moderna’s mRNA technology platform emerged from the pandemic era, researchers immediately asked a bigger question: could the same molecular machinery that teaches the immune system to recognize a spike protein also teach it to hunt cancer? At the American Society of Clinical Oncology Annual Meeting in Chicago this June, a definitive five-year answer arrived. The personalized mRNA cancer vaccine intismeran autogene, combined with Merck’s checkpoint inhibitor pembrolizumab (Keytruda), sustained a 49% reduction in the risk of melanoma recurrence or death compared to Keytruda alone, with an overall survival rate of 92.2% versus 71.3% in patients receiving immunotherapy alone.

The results were presented by lead author Dr. Matteo Carlino, a medical oncologist at Westmead and Blacktown Hospitals and a faculty member at Melanoma Institute Australia, and were simultaneously published in the Journal of Clinical Oncology. They represent the longest follow-up yet for any personalized neoantigen cancer vaccine in a randomized clinical trial, and they answer a fundamental durability question that has shadowed the field since the first early clinical signals emerged.

Melanoma’s Rising Toll and the Limits of Standard Treatment

Melanoma is the most lethal form of skin cancer. While it accounts for only about 1% of all skin cancers, it causes the majority of skin cancer deaths. The American Academy of Dermatology projects 234,680 new melanoma diagnoses in the United States in 2026 alone, continuing a 30-year upward trend driven by UV exposure patterns, population aging, and improved detection.

The introduction of checkpoint inhibitor immunotherapy over the past decade transformed melanoma treatment. Pembrolizumab, which blocks the PD-1 receptor and frees the immune system’s T cells to attack tumor cells, became a cornerstone of both adjuvant treatment (following surgical resection) and advanced disease management. Five-year survival rates for metastatic melanoma, once below 10%, climbed toward 40% in patients responding to checkpoint therapy.

Yet a critical gap remained. Even among patients who undergo complete surgical resection of high-risk stage III or IV disease, melanoma recurs in a substantial proportion within five years. Standard immunotherapy treats cancer cells the immune system can already partially recognize. It does not direct the immune response toward the specific mutations that make one patient’s tumor distinct from every other. That specificity is precisely what personalized neoantigen vaccines are designed to provide.

The Science Behind Intismeran Autogene: Teaching Immunity One Tumor at a Time

Intismeran autogene, formerly known as mRNA-4157 and V940, belongs to a category of immunotherapy called individualized neoantigen therapy. Neoantigens are proteins displayed on the surface of cancer cells that arise from somatic mutations specific to each patient’s tumor. Because they are absent from normal cells, neoantigens are ideally positioned as targets for immune attack. They are, in effect, the tumor’s fingerprints.

Producing the vaccine begins with next-generation sequencing of the patient’s tumor DNA and matched normal tissue. Bioinformatic algorithms compare the two, identify mutations unique to the cancer, and predict which of those mutations are most likely to generate peptide fragments that the patient’s own immune molecules (HLA proteins) can present to T cells. The algorithms select up to 34 neoantigens per patient, encode them as a single synthetic mRNA sequence, and manufacture individualized vaccine doses for that patient alone. No two patients receive the same vaccine.

“The vaccine can be considered a personalized approach that is ‘tumor informed,’ meaning that neoantigens from a patient’s individual tumor are used in the vaccine design,” explained Dr. Janice Mehnert, the trial’s senior investigator and a professor at NYU Grossman School of Medicine. “This is given with the anti-PD-1 therapy, which is more of an empiric and standardized approach, to result in truly personalized therapy.”

The logic of combining the neoantigen vaccine with pembrolizumab is synergistic. Checkpoint inhibitors remove the molecular brakes that tumors use to suppress immune activity, but they are only as effective as the number of tumor-reactive T cells present to respond. The personalized vaccine expands and activates a precisely targeted army of T cells, while pembrolizumab removes the barriers that would otherwise quiet them.

KEYNOTE-942: What the Five-Year Data Show

The KEYNOTE-942 trial (also designated mRNA-4157-P201) was a randomized, open-label phase 2b study that enrolled 157 patients with high-risk, completely resected stage III or IV melanoma between 2019 and 2021 at sites in the United States, Australia, and Europe. Patients were assigned 2:1 to receive intismeran autogene (1 mg every three weeks for nine doses) plus pembrolizumab (200 mg every three weeks for up to 18 cycles), or pembrolizumab alone for approximately one year.

At the planned five-year follow-up with a median of 60.3 months, the combination arm demonstrated three headline findings. First, a 49% reduction in the risk of recurrence or death compared to pembrolizumab alone. This is the trial’s primary endpoint: recurrence-free survival. Critically, the 49% benefit at five years is numerically identical to the result reported at three years, confirming that the protective effect did not wane over time.

Featured Partner

Invest in the Infrastructure Behind Modern Medicine

As healthcare expands beyond hospital walls, the buildings and campuses supporting that shift are generating compelling returns for investors who move early. The Healthcare Real Estate Fund offers qualified investors direct access to a curated portfolio of medical office, outpatient, and specialty care facilities.

Learn More →

Second, a 59% reduction in the risk of distant metastasis or death, the trial’s key secondary endpoint. This figure reflects the vaccine’s ability to suppress the spread of melanoma to the lungs, brain, liver, and other organs, the clinical outcome that most directly correlates with long-term survival.

Third, an overall survival rate of 92.2% in the combination group versus 71.3% in the pembrolizumab-only group, an exploratory analysis showing an absolute survival advantage of nearly 21 percentage points at five years. The hazard ratio for death in this exploratory analysis was 0.471 (95% CI, 0.165 to 1.345), consistent with a clinically meaningful survival signal even in this relatively small phase 2 study. Seven patients in each arm died during follow-up, mostly due to disease progression.

The safety profile was consistent with earlier analyses. Immune-related adverse events occurred in over 45% of patients in the combination arm, but the combination did not increase the rate of serious immune toxicity compared to pembrolizumab alone. The vast majority of side effects were grade 1 or 2 in severity, primarily fatigue, injection-site pain, and chills related to the vaccine itself. No grade 4 or 5 adverse events were reported in the combination arm.

“Prior attempts to improve on single-agent anti-PD-1 with dual checkpoint inhibition led to higher rates of immune-related toxicities,” noted Dr. Carlino. “In our study the combination did not increase the rate of irAE, and was actually associated with a slightly lower rate of irAEs.” This distinction matters significantly in the adjuvant setting, where patients have had their tumors surgically removed and are receiving treatment to prevent recurrence rather than to control active disease.

The Durability Finding: Why Five-Year Consistency Changes the Calculus

In cancer immunotherapy, durability is everything. A treatment that produces dramatic early responses but fades quickly provides temporary reprieve rather than lasting protection. The history of adjuvant cancer therapy includes multiple trials that showed strong short-term results only to see benefit erode at longer follow-up.

The KEYNOTE-942 five-year data tell the opposite story. At two years, the combination reduced the risk of recurrence or death by 44%. At three years, that figure rose to 49%. At five years, it remains 49%. The benefit has not merely been maintained; the hazard ratio has stabilized, suggesting the vaccine-trained T cells are providing lasting immunologic surveillance rather than a one-time activation event.

The mechanistic explanation for this durability lies in the nature of T cell memory. Vaccine-generated T cells do not simply attack cancer cells once and disappear. When a vaccine successfully establishes a population of memory T cells specifically targeting the patient’s tumor neoantigens, those cells can persist for years, ready to respond when they encounter residual or recurrent disease. Translational studies embedded within KEYNOTE-942 directly support this model: the combination arm showed increased T cell clonality and a greater number of novel T cell clones compared to pembrolizumab alone, indicating the vaccine generates new and diversified anti-tumor immune responses rather than merely amplifying pre-existing ones.

“The benefits we have seen with respect to improved relapse-free survival have persisted over time, and the translational studies show proof of concept that intismeran is contributing to mounting a meaningful immune response,” said Dr. Mehnert.

Modern immunology identifies two T cell populations as particularly relevant to long-term protection. Late-differentiated effector memory T cells, which are primed and ready to immediately kill cancer cells upon recognition, and stem-cell-like memory T cells, which are capable of self-renewal and long-term persistence. Both populations have been identified in patients receiving neoantigen vaccines, and the balance between them may explain why vaccine-induced immunity can remain protective years after the last dose was administered.

From Phase 2 to Phase 3: The Road to Approval

As compelling as the KEYNOTE-942 five-year data are, they come from a phase 2b trial of 157 patients. The FDA’s bar for approval requires randomized phase 3 confirmation at a much larger scale, and Merck and Moderna have already crossed the enrollment finish line on the study designed to provide it.

INTerpath-001, the phase 3 adjuvant melanoma trial, is fully enrolled and event-driven, meaning data will be released once a pre-specified number of recurrence events occur. Moderna CEO Stéphane Bancel signaled at the 2026 J.P. Morgan Healthcare Conference that interim data readout is expected in 2026. Given the phase 2 results, and assuming the phase 3 arm balance and patient selection are comparable, the oncology community is watching for a confirmatory signal that could trigger regulatory submissions in the United States and Europe and potentially establish intismeran autogene as a new standard of care for high-risk resected melanoma.

Moderna has identified intismeran autogene as one of its primary commercial growth drivers for 2027. The program extends well beyond melanoma. Phase 3 trials are underway in two non-small-cell lung cancer settings, and phase 1 and 2 studies are investigating intismeran autogene in renal cell carcinoma, bladder cancer, and additional solid tumor types. The hypothesis, supported by the melanoma data, is that any cancer with sufficient mutational burden generating identifiable neoantigens could in principle respond to a tumor-informed mRNA vaccine.

The manufacturing challenge that once seemed prohibitive has been substantially addressed. When KEYNOTE-942 enrolled its first patients in 2019, the turnaround time from tumor biopsy to vaccine delivery was a source of clinical anxiety: each dose must be custom-built from an individual’s tumor sample, sequenced, designed by algorithms, synthesized, quality-tested, and shipped. That process has since been industrialized. By 2025, manufacturing capacity expanded to accommodate thousands of patients across multiple concurrent trials, and Moderna’s production infrastructure is being further scaled in anticipation of potential approval. “Significant efforts have already occurred to streamline and upscale manufacturing from the time of our study,” noted Dr. Carlino. “Many thousands of patients have now received intismeran on multiple clinical trials and capacity is being expanded to meet the demand once intismeran is approved.”

A New Paradigm: Vaccines That Follow the Tumor’s Lead

What intismeran autogene represents at a conceptual level is a fundamental shift in how oncologists think about the relationship between a patient’s immune system and their cancer. Traditional cancer vaccines attempted to target shared tumor antigens, proteins expressed on many patients’ cancers simultaneously, with disappointing clinical results because tumors can downregulate shared antigens and escape immune recognition entirely. Neoantigen vaccines abandon the shared-target hypothesis.

The approach accepts the heterogeneity of cancer rather than fighting it, using the tumor’s own mutations as the blueprint for immunotherapy. It is, in a very literal sense, the most precise expression of personalized medicine: a treatment derived entirely from the molecular characteristics of one individual’s disease, manufactured uniquely for that patient and no one else.

This philosophy connects to a broader transformation occurring across oncology. Genomic profiling, which once served primarily as a diagnostic and prognostic tool, is increasingly serving as a therapeutic input. The polygenic risk scores redefining cardiovascular risk stratification, the 50-gene expression profiles guiding chemotherapy decisions in breast cancer (as demonstrated in the recent OPTIMA trial), and now the somatic mutation signatures driving neoantigen vaccine design all reflect the same underlying shift: medicine is becoming personalized at the molecular level, and the mRNA platform is accelerating that transition.

The convergence of mRNA manufacturing technology with advances in tumor genomics and AI-driven neoantigen prediction has compressed what would once have required decades of development into the span of a single clinical trial program. Researchers note that AI-assisted HLA epitope prediction and neoantigen prioritization algorithms have substantially improved since the KEYNOTE-942 enrollment period began in 2019. Future generations of personalized vaccine programs will benefit from more accurate neoantigen selection than was possible six years ago, potentially expanding the population of patients who respond and deepening the magnitude of benefit.

What This Means For You

If you or someone you know has been diagnosed with melanoma, particularly high-risk stage III or IV disease following surgical resection, the KEYNOTE-942 five-year results are directly relevant to conversations with an oncologist today. Pembrolizumab is already standard of care in the adjuvant setting. The data suggest that combining it with a personalized neoantigen vaccine may substantially reduce the likelihood of recurrence, and the phase 3 INTerpath-001 trial represents a pathway to receiving this treatment outside of standard care at participating centers. Asking your oncologist about eligibility is a reasonable and informed next step.

For the broader population, these results mark a conceptual turning point worth understanding. Melanoma is, in important ways, a proving ground for approaches that oncologists intend to apply across many cancer types. The lung cancer and renal cell carcinoma phase 3 programs already underway represent the next expansion of the personalized vaccine platform. If INTerpath-001 data confirm the phase 2 findings and the FDA approves intismeran autogene for adjuvant melanoma, it will establish a regulatory and manufacturing template for personalized neoantigen vaccines in other high-mutational-burden cancers.

From a prevention perspective, melanoma risk remains meaningfully modifiable. Sun protection, annual full-body skin checks with a dermatologist, and avoidance of tanning beds remain the most effective primary prevention tools. For those with a family history of melanoma or prior skin cancers, genetic counseling to assess hereditary risk factors, including mutations in CDKN2A and CDK4, is appropriate.

The broader message from five years of KEYNOTE-942 follow-up is one of validated hope. Cancer immunotherapy, when precisely targeted to a patient’s own tumor biology, can generate immune memory that persists for years. The mRNA platform that protected millions from a respiratory pandemic is now being trained, one tumor at a time, on some of the most treatment-resistant diseases medicine faces. That the protective effect only sharpens with time is, in itself, a landmark finding.

Free Daily Briefing

The Latest Longevity Science.
Delivered Every Morning.

Join researchers, physicians, and health professionals getting daily breakthroughs in AI-driven medicine, epigenetics, and longevity research.

Support the research that powers this editorial

Subscribe Free

No spam. Unsubscribe anytime. We respect your inbox.