The Vaccine That Starts From Scratch Every Single Time

Every patient who receives intismeran autogene gets a cancer vaccine that has never existed before and will never exist again.

By Carry and Conquer Publications

In the spring of 2022, Moderna and Merck announced something that had never been demonstrated in a randomized clinical trial: a personalized mRNA cancer vaccine that reduced the risk of recurrence or death in high-risk melanoma patients by nearly half. The data came from 157 patients. Each had received a completely different vaccine, synthesized in weeks from the unique mutational signature of their own tumor. What began as a proof-of-concept for an entirely new category of medicine has since become one of the largest clinical bets in oncology, with three active Phase 3 trials, a purpose-built manufacturing facility in Massachusetts, and five years of follow-up data that has not wavered.

From Biopsy to Vaccine in Weeks

The process begins with a knife. Once a surgeon removes the tumor, a sample goes to sequencing. AI algorithms scan the genetic data, identifying the somatic mutations unique to that patient's cancer and predicting which of them will produce peptides, called neoantigens, that the immune system is most likely to recognize and attack. From those predictions, up to 34 neoantigens are selected, encoded into a single strand of synthetic mRNA, encapsulated in lipid nanoparticles for delivery, and shipped back to the treating oncologist. The entire clock runs weeks, not months.

That turnaround is itself an engineering achievement. In early trials, production required approximately nine weeks from surgery to first dose. Advanced enzymatic DNA synthesis, automated closed-system manufacturing, and machine learning-guided process control have compressed that window to under four weeks. The dedicated Marlborough, Massachusetts facility that Moderna opened for intismeran, built with advanced robotics and scaled specifically for personalized batch production, began clinical-batch supply in September 2025. It is the industrial infrastructure of personalized medicine made concrete: a factory where no two products are the same.

The vaccine is then administered in combination with pembrolizumab (KEYTRUDA), Merck's anti-PD-1 checkpoint inhibitor, every three weeks for up to nine doses. The logic is synergistic: pembrolizumab takes the brakes off the immune system, while the vaccine tells it precisely what to attack.

The Numbers That Built the Phase III Case

The Phase 2b KEYNOTE-942 trial enrolled 157 patients with completely resected high-risk Stage III and IV melanoma, randomly assigned two-to-one to receive the combination or pembrolizumab alone. What followed was the first demonstration of clinical efficacy for an mRNA cancer treatment in a randomized controlled trial.

At a median follow-up of 34.9 months, the combination reduced the risk of recurrence or death by 49% compared to pembrolizumab alone, with a hazard ratio of 0.510. The 2.5-year recurrence-free survival rate was 74.8% in the combination arm versus 55.6% for pembrolizumab monotherapy. Distant metastasis risk fell by 62%. The benefit held across subgroups regardless of tumor mutational burden or PD-L1 status, which matters: it suggests the vaccine's personalized neoantigen approach works independent of the tumor characteristics that typically determine checkpoint inhibitor response. In January 2026, Moderna and Merck announced five-year follow-up data showing the 49% risk reduction remained stable, a durability finding that is rare in oncology and clinically meaningful.

The FDA, recognizing the strength of those findings, granted Breakthrough Therapy Designation to the combination for adjuvant high-risk melanoma. The European Medicines Agency extended Priority Medicines (PRIME) scheme designation. Both designations accelerate regulatory dialogue and reduce the runway to potential approval.

INTerpath: A Program Built for Multiple Fronts

KEYNOTE-942 was the proof of concept. The INTerpath program is the commercial bet.

INTerpath-001, the Phase 3 adjuvant melanoma trial, enrolled 1,089 patients with high-risk Stage IIB through Stage IV disease and completed enrollment in 2024. Phase 3 data are expected in 2026. INTerpath-009, the most recent expansion, targets non-small-cell lung cancer patients who did not achieve a pathological complete response after neoadjuvant chemotherapy plus pembrolizumab. That trial is enrolling 680 patients globally and randomizes them to the combination or placebo, with disease-free survival as the primary endpoint. The first patients began enrolling in Canada in late 2025.

Beyond the two flagship Phase 3 programs, Moderna and Merck have initiated Phase 2 trials in renal cell carcinoma and urothelial carcinoma, and a Phase 2/3 adaptive trial for cutaneous squamous cell carcinoma. Eight trials in total are now active across the INTerpath program, covering tumor types that together represent a substantial portion of global cancer incidence.

The Pancreatic Signal That Changes Everything

Lung cancer and melanoma are visible. Pancreatic cancer is something else.

Pancreatic ductal adenocarcinoma kills 88% of patients. It is largely resistant to checkpoint inhibitors, poorly immunogenic by conventional metrics, and almost universally fatal within two years of diagnosis. That historical record makes what happened in a Phase 1 trial using BioNTech's autogene cevumeran, a closely related personalized mRNA neoantigen platform, more striking than any melanoma number.

In a 16-patient cohort, eight patients mounted high-magnitude neoantigen-specific T-cell responses. At 3.2-year follow-up published in Nature, those responders showed median recurrence-free survival that had not been reached, compared to 13.4 months for non-responders. The vaccine-induced CD8+ T cell clones were estimated to have average lifespans of 7.7 years. An updated readout presented at the American Association for Cancer Research in April 2025 extended the horizon further: at six years, six of those eight responders were still alive, many remaining recurrence-free. In a cancer where two-year survival is an exceptional outcome, six-year survival data in early responders is not an incremental finding. It is a fundamental reexamination of what is possible.

The mechanistic insight deepened with it: benefit appeared to require coordinated engagement of both cytotoxic killer T cells and immune-memory helper T cells, a multi-compartment immune activation that may explain why mRNA-based personalized vaccines succeed where conventional checkpoint blockade fails in low-mutation-burden cancers.

The Industrial Problem Hiding Inside the Scientific Story

Intismeran is not a drug. It is a manufacturing process that produces a drug that is different every single time.

That distinction changes almost everything about how the investment thesis works. The per-patient cost of manufacturing currently exceeds $100,000, and the infrastructure required to sequence tumors, run AI neoantigen prediction algorithms, synthesize bespoke mRNA constructs, perform quality control on individualized batches, and deliver them within clinically meaningful timeframes is not a pharmaceutical operation. It is a data, logistics, and advanced manufacturing operation that happens to have a therapeutic output.

The Marlborough facility is the first physical manifestation of Moderna's answer to that challenge, but it is not the only one. Hybrid manufacturing approaches that pre-produce common mRNA elements while adding patient-specific neoantigen sequences at a final personalization step are in active development, with the potential to reduce timelines and per-patient costs through scale. Automated closed-system platforms with real-time quality monitoring, machine learning process control, and continuous production architectures are being implemented specifically to address the quality control paradox of a product where every batch is a unique first run.

Reimbursement presents a second layer of complexity. A product costing six figures to manufacture before markup, with no comparator for how payers price a truly individualized therapeutic, is a policy problem as much as a scientific one. The regulatory frameworks that govern batch-release for conventional pharmaceuticals were not designed for a medicine where the definition of a batch is one patient. The FDA, the European Medicines Agency, and Moderna itself are in active dialogue to develop the new frameworks required.

The Commercial Horizon

First commercial approvals are currently projected for 2029. Phase 3 data from INTerpath-001, the melanoma trial, are expected in 2026, which would enable an initial regulatory filing if results replicate the Phase 2b signal. That filing would follow the fastest path to market available: a Breakthrough Therapy product, in an indication with high unmet need, backed by five years of accumulating follow-up data and the first-ever Phase 3 read in the space.

The lung cancer program behind it opens a substantially larger commercial market. Non-small-cell lung cancer remains the leading cause of cancer death worldwide. A therapy that durably reduces recurrence in resected early-stage NSCLC, even for patients who failed to achieve complete pathological response to standard neoadjuvant therapy, would compete in one of the most commercially important pockets in oncology.

The scale of the partnership behind it is proportional to the ambition. Merck paid $250 million upfront in 2022 to exercise its option to jointly develop and commercialize intismeran, with subsequent milestone payments that extended the total potential commitment well beyond that figure. Moderna controls manufacturing. Merck controls KEYTRUDA, which is itself the world's best-selling oncology drug and is essentially required as the combination backbone. The incentive alignment is structural: Merck cannot win in this space without the vaccine; Moderna cannot win without the pembrolizumab combination.

What the Data Has Already Proven

The personalized mRNA cancer vaccine concept entered clinical development as a hypothesis. It now has a five-year randomized dataset in melanoma showing durable recurrence-free survival benefit, an FDA Breakthrough Therapy Designation, a purpose-built commercial manufacturing facility, three active Phase 3 trials across two tumor types, and a growing body of mechanistic evidence explaining exactly why the approach works.

The clinical question for the next three years is whether Phase 3 replicates Phase 2b in melanoma, and whether the lung cancer signal emerges on the same timescale. The manufacturing question is whether Moderna can compress costs and production times sufficiently to make commercial-scale personalized medicine economically sustainable. The regulatory question is whether new frameworks can accommodate a product category that has never existed before.

None of those questions have been answered. But a vaccine that codes for a unique set of neoantigens for every patient it has ever treated has now demonstrated that it can keep cancer from coming back for five years running. That is the foundation on which a new category of medicine is being built.