Can We Vaccinate Against Cancer?

 Cancer Vaccines Are Becoming Real — Could We One Day Vaccinate Against Cancer?

Cancer vaccines are no longer just a futuristic idea. Some already prevent virus-related cancers, while personalized vaccines are being designed to help the immune system fight specific tumors.

The strangest thing about one of the most promising cancer vaccines is that doctors cannot make it until after they have seen the tumor. They remove or biopsy the cancer, read its mutations, choose the most useful targets and manufacture an mRNA treatment for that one patient.

That is very different from the vaccines most of us know. We usually vaccinate before a disease appears. Cancer vaccines are now developing along two very different paths: some aim to prevent cancer before it starts, while others are therapeutic, training the immune system to recognize a tumor that has already existed.

In August 2026, the therapeutic approach crossed an important line. Merck and Moderna announced that a Phase III trial of intismeran autogene — previously known as V940 or mRNA-4157 — met its primary endpoint in patients with high-risk melanoma after surgery. The treatment was built from the mutational fingerprint of each patient’s tumor and used together with pembrolizumab, an established immunotherapy drug.

This is not a universal shot against cancer, and it is not yet an approved replacement for standard melanoma care. But it is one of the clearest signs so far that “cancer vaccine” is becoming a real medical category rather than a futuristic slogan.

First, What Does “Cancer Vaccine” Actually Mean?

The phrase “cancer vaccine” hides two very different strategies, and the distinction changes almost everything about how the technology works.

A preventive cancer vaccine is given before cancer appears. The best-known examples do not attack cancer cells directly; they block infections that can later cause cancer. HPV vaccination can prevent infections linked to cervical, anal, penile, vulvar, vaginal and oropharyngeal cancers. Hepatitis B vaccination prevents HBV infection, reducing the risk of serious liver disease and liver cancer.

So, in one important sense, we already vaccinate against cancer today — by preventing some of its causes.

A therapeutic cancer vaccine is different. It is given to someone who already has cancer, or who has had a tumor removed but remains at risk that microscopic cancer cells are still somewhere in the body. Instead of blocking an infection, it teaches immune cells to recognize features of the tumor and attack cells carrying those features.

Personalized neoantigen vaccines take this a step further. They are not designed for “melanoma patients” as a group. The target list can be different for every person.

Not all cancer vaccines mean the same thing. Some prevent cancer before it starts, while others are designed to train the immune system against an existing tumor.

Why Cancer Is So Much Harder to Vaccinate Against Than a Virus

A virus gives the immune system a relatively clean target. Viral proteins are foreign. Cancer is messier. A tumor begins as normal human tissue, which means most of what it contains still looks like “self” to the immune system.

That creates the central problem of cancer immunology: how do you direct a powerful immune attack at malignant cells without teaching the immune system to attack healthy tissue?

Mutations provide part of the answer. As cancer cells accumulate DNA changes, some of those mutations alter proteins. Small fragments of these abnormal proteins can appear on the surface of tumor cells. These mutation-created targets are called neoantigens.

Neoantigens are attractive because healthy cells generally do not carry them. In principle, they are molecular fingerprints that say: this cell is not normal.

But every tumor is different. Two people can both have melanoma and still carry very different sets of mutations. Even within one person, different parts of the same tumor may not be genetically identical. Cancer also changes over time, and immune pressure can select for cells that stop displaying a target.

That is why a universal injection against every non-viral cancer has remained such a difficult goal. Cancer is not one enemy wearing one uniform. It is thousands of evolving populations of abnormal cells.

A virus gives the immune system a clearer target. Cancer is harder because it comes from our own cells, changes over time, and often looks different from one patient to another.

How a Personalized mRNA Cancer Vaccine Is Made

The personalized approach tries to turn that diversity from a weakness into a target.

Imagine a patient has a high-risk melanoma removed by surgery. Doctors now have tumor tissue. They can sequence genetic material from the tumor and compare it with normal tissue from the same patient. The goal is to identify mutations found in the cancer but not in healthy cells.

Software then ranks the mutations most likely to produce useful neoantigens. A mutation is not automatically a good vaccine target: the resulting protein fragment has to be processed and displayed by HLA molecules — the cellular display system that shows protein fragments to T cells — and it has to trigger a meaningful immune response.

The selected targets are then encoded into an mRNA construct. In the current intismeran program, the therapy can encode up to 34 patient-specific neoantigens. The resulting product is manufactured for that individual patient and administered by injection.

Once inside specialized immune cells that act as antigen presenters, the mRNA provides temporary instructions for making those selected neoantigen fragments. The immune system is then shown what to look for, with the aim of expanding T cells that can recognize the same neoantigens on residual melanoma cells.

The mRNA does not rewrite the patient’s DNA and it does not alter the tumor’s genome. It is a short-lived instruction set used to train an immune response.

A personalized cancer vaccine begins with tumor sequencing. Researchers identify mutations, select neoantigens, build a custom mRNA vaccine, and train T cells to recognize cancer cells.

Why Pair the Vaccine With Pembrolizumab?

A vaccine can generate or expand tumor-specific T cells, but knowing the target is only part of the problem. Tumors can create an environment that suppresses the very immune cells trying to attack them.

Pembrolizumab is a checkpoint inhibitor that blocks PD-1, one of the braking systems that can restrain T cells. Put simply, the personalized vaccine helps T cells know what to look for; pembrolizumab helps stop those T cells from being switched off too easily.

That does not make pembrolizumab a minor add-on. Checkpoint inhibition is already an important standard treatment in melanoma. The real question has been whether a personalized vaccine can add meaningful protection beyond pembrolizumab alone.

The Melanoma Results That Changed the Conversation

The strongest evidence so far comes from the KEYNOTE-942 program in people whose high-risk melanoma had been completely removed by surgery.

In the randomized Phase IIb study, 157 patients received either personalized mRNA-4157/V940 plus pembrolizumab or pembrolizumab alone. The peer-reviewed results published in The Lancet in 2024 were encouraging: at 18 months, recurrence-free survival was about 79% with the combination and 62% with pembrolizumab alone. Grade 3 or higher treatment-related adverse events occurred in 25% of patients receiving the combination and 18% receiving pembrolizumab alone, while immune-mediated adverse events were similar between the groups.

The important question was whether that advantage would last. In the five-year update published in 2026, it did. The combination was associated with a 49% lower risk of recurrence or death and a 59% lower risk of distant metastasis or death. Overall survival also favored the combination numerically, but the confidence interval was wide, so that result is not yet definitive proof of a survival benefit.

Then the idea faced the test that matters most for changing clinical practice: a much larger Phase III trial.

On August 19, 2026, Merck and Moderna reported positive topline results from INTerpath-001. According to the companies, intismeran autogene plus pembrolizumab produced statistically significant and clinically meaningful improvements in both recurrence-free survival and distant-metastasis-free survival compared with pembrolizumab alone in patients with completely resected stage IIB–IV cutaneous melanoma.

If the full dataset supports the announcement, this will be a landmark result: the first positive Phase III readout for an individualized neoantigen therapy and for an mRNA-based cancer therapy.

The caution is simple but important. As of September 2026, the detailed Phase III numbers have not yet been publicly presented. We know the trial met its endpoints; we do not yet know the exact size of the benefit, how different subgroups performed, whether an overall-survival advantage will emerge or how regulators and guidelines will ultimately position the treatment.

Is This Really a Vaccine If the Tumor Has Already Been Removed?

Yes. More precisely, it is a therapeutic vaccine being used in the adjuvant setting — after the visible tumor has been removed, when treatment is aimed at lowering the risk of recurrence.

A clean scan does not guarantee that every malignant cell is gone. Microscopic cancer cells can remain and later seed a recurrence or metastasis. Adjuvant therapy tries to eliminate that residual disease before it becomes detectable.

That may actually be an unusually favorable setting for vaccination. Instead of asking the immune system to destroy a large, immunosuppressive tumor containing billions of cells, the treatment may be hunting a far smaller residual population.

The first major success of cancer vaccines, in other words, may not look like a dramatic injection that makes a large tumor vanish. It may be quieter: helping stop cancer from coming back after doctors have removed everything they can see.

What About a True Vaccine Given Before Cancer Exists?

This is the version most people picture when they hear “cancer vaccine”: get vaccinated while healthy, then have a lower chance of developing cancer years later.

For most non-viral cancers, that is still experimental. Before a sporadic tumor exists, we do not know which mutations it will acquire. Targeting normal human proteins too broadly could produce weak responses or unwanted autoimmunity, and cancers may take decades to develop, making prevention trials unusually slow and difficult.

Researchers are therefore starting where the biology gives them a more predictable target: people with inherited cancer syndromes.

Lynch syndrome is a good example. It is caused by inherited defects in DNA mismatch repair and is associated with a high lifetime risk of colorectal and several other cancers. Because mismatch-repair-deficient cells repeatedly generate certain classes of frameshift mutations, some abnormal targets are shared across different precancers and tumors.

In January 2026, researchers reported Phase Ib/II results for Nous-209 in people with Lynch syndrome. Unlike the personalized melanoma treatment, Nous-209 is not an mRNA vaccine; it uses viral vectors to deliver 209 shared frameshift-derived targets. In the 45-person cohort, neoantigen-specific immune responses were seen in all evaluable participants, and responses remained detectable one year later in 85%. The study was designed to test safety and immunogenicity, not whether fewer people actually developed cancer, so it is a biological proof of concept rather than proof of prevention.

The next experiment is even closer to the familiar idea of vaccination. In August 2026, the first participant received Moderna’s investigational mRNA-4194 vaccine in the UK INTERCEPT-Lynch trial. The aim is to train the immune system to recognize and eliminate precancerous cells in people with Lynch syndrome before invasive cancer develops.

That is genuine cancer interception — but for now it is being tested in a genetically high-risk group, not offered to the general population.

Could There Ever Be One Universal Cancer Vaccine?

Probably not in the simple sense of one injection that protects everyone from every cancer.

“Cancer” covers hundreds of diseases. Lung adenocarcinoma, melanoma, pancreatic ductal adenocarcinoma, glioblastoma and leukemia do not share one universal molecular identity. Even two tumors with the same diagnosis can evolve along different genetic paths.

The more likely future is a portfolio of different vaccines rather than one universal product.

Some will prevent cancer-causing infections. Others may target shared precancerous changes in people with inherited risk, or common antigens found across a subset of tumors. And at the most individualized end of the spectrum, some treatments may continue to be manufactured from the sequence of a single patient’s cancer.

That is less tidy than the idea of “the cancer vaccine,” but it is probably much closer to how cancer medicine will actually evolve.

The Hard Part Is Not Just the Biology

Personalization creates a practical problem ordinary mass-produced vaccines do not have: every treatment has to be designed, manufactured and quality-checked for one person. Tumor tissue has to be available, sequencing has to work, useful targets have to be selected and the final product has to be made quickly enough to fit into the patient’s treatment window.

That turns manufacturing into part of the therapy itself. If the process takes too long, fails for a particular sample or can only be done in a handful of specialized centers, a scientifically elegant vaccine may still be difficult to use at scale. Cost and access could become nearly as important as efficacy.

Even with perfect logistics, target selection remains an imperfect science. Algorithms have to predict which mutations will produce neoantigens that are actually displayed and recognized by T cells. Better models, larger immunological datasets and improved measurements of tumor evolution should help, but prediction errors will not disappear overnight.

Tumors can also escape. If the immune system aggressively attacks cells carrying one target, cells that lose or never displayed that target may gain a survival advantage. Including many neoantigens in one vaccine makes escape harder, but not impossible.

And melanoma may be a particularly favorable proving ground. It often carries many mutations and can be relatively visible to the immune system. Pancreatic cancer and other immunologically “cold” tumors create a much tougher environment, so success in melanoma cannot simply be copied across oncology.

What Else Is Being Tested?

Melanoma is the headline, but it is not the whole field. Personalized or neoantigen-directed vaccines are being studied in pancreatic, kidney, lung, bladder, colorectal and other cancers, using mRNA, peptides, viral vectors and other platforms.

Small studies in pancreatic and kidney cancer have already shown that personalized vaccination can generate durable tumor-specific T-cell responses. Those studies are far too small to prove a broad survival benefit, but they matter because they test the basic idea in cancers with very different biology from melanoma.

The field has also become large enough to attract a more coordinated national effort in the United States. In 2026, the National Cancer Institute and the Foundation for the National Institutes of Health outlined a National Therapeutic Cancer Vaccine Initiative focused on scalable vaccine platforms, antigen discovery, immune monitoring, manufacturing and trials in high-need cancers.

That shift is worth watching. The question is no longer only whether an individual experimental vaccine can work. Researchers are beginning to ask what infrastructure would be needed if personalized vaccination becomes a repeatable part of cancer care.

What Could Change in 2, 5 and 10 Years?

In the next 2 years

The immediate story is melanoma. The detailed INTerpath-001 results, regulatory review and eventual decisions about where intismeran autogene belongs alongside existing adjuvant therapies will determine whether personalized mRNA vaccination becomes a real treatment option rather than a promising trial result.

At the same time, manufacturers will be under pressure to shorten the interval between surgery, sequencing and the first individualized dose. In this field, speed is not a convenience; it is part of whether the treatment is clinically usable.

In about 5 years

Within roughly five years, the defining question should be whether the melanoma success travels. Large randomized trials in other tumor types will show whether personalized vaccination is a broad oncology platform or a powerful tool for a narrower set of cancers.

Blood-based minimal residual disease tests could also become an important partner. A future pathway might use a blood test to detect molecular evidence of remaining cancer, tumor sequencing to define targets and a personalized vaccine to help the immune system eliminate what is left.

In about 10 years

A decade out, the most consequential change may be the move from treating residual cancer to intercepting it before an invasive tumor fully forms.

People with inherited cancer syndromes are plausible early candidates because their tumors can develop through more predictable molecular routes. If prevention trials succeed there, researchers could move toward other high-risk groups defined by genetics, chronic inflammation, premalignant lesions or molecular screening.

A single universal cancer vaccine may still be unlikely. A mixed future is easier to imagine: some cancers prevented with off-the-shelf vaccines, others intercepted in high-risk people and still others treated with individualized vaccines generated from a tumor sequence.

The future of cancer vaccines may unfold in stages: first better personalized treatment, then expansion into more cancer types, and eventually earlier interception in people with inherited risk.

So, Will We One Day Vaccinate Against Cancer?

So, could we one day vaccinate against cancer? In narrow but important ways, the answer is already yes. HPV and hepatitis B vaccines prevent infections that can cause cancer, and personalized therapeutic vaccines are now showing that the immune system can also be trained against mutations inside an individual tumor.

What we do not have is one universal injection that makes “cancer” disappear as a category. The disease is too diverse for that simple story. The more plausible future is a collection of vaccines used at different moments: before cancer, after surgery, and perhaps one day at the first molecular signs that a precancerous clone is beginning to form.

The remarkable part is how quickly that future has moved closer. A treatment can now begin with a piece of one person’s tumor and end as an mRNA product designed around that tumor’s mutations. At the same time, early prevention trials are asking whether the immune system can be trained before cancer exists at all.