The Vaccine Made Just for You: How Scientists Are Training the Immune System to Hunt Cancer

Imagine being told that doctors are going to remove a piece of your cancer, read its genetic code and use the cancer’s own mutations to manufacture a vaccine against it.

Not a vaccine designed for an entire country.

Not even a vaccine designed for everyone with the same type of cancer.

A vaccine designed specifically for you.

The person sitting beside you in the hospital might have the exact same type of cancer, yet receive a completely different vaccine. This is because, at the genetic level, their cancer may be very different from yours.

It sounds more like science fiction than medicine. However, scientists are already testing these treatments in people with melanoma, pancreatic cancer, kidney cancer, lung cancer and several other tumours.

These vaccines do not directly poison cancer cells like chemotherapy, nor do they burn them away like radiation. Instead, they attempt something very different.

They give the immune system a detailed description of the enemy and tell it exactly what to hunt.

The vaccine does not act like a bullet.

It acts like a biological wanted poster.

But why does the immune system need a wanted poster in the first place? Shouldn’t it already know that cancer is dangerous?

To understand that, we first need to understand what makes cancer such a difficult enemy.

The Enemy Wearing Our Uniform

The human immune system is exceptionally good at detecting things that do not belong in the body.

When a virus enters us, its proteins are foreign. When bacteria invade a wound, their chemical structures are different from ours. Immune cells can identify these differences, attack the intruders and often remember them in case they return.

Cancer is much more difficult.

A cancer cell did not enter the body from the outside. It began as one of our own cells. It may have once been a skin cell, a lung cell or a pancreatic cell. Over time, changes in its DNA allowed it to grow uncontrollably, ignore the body’s instructions and sometimes spread into other organs.

This means that cancer is an enemy wearing our own uniform.

From the immune system’s point of view, many parts of a cancer cell still look normal. Attacking it without making a mistake is like trying to find one criminal in a stadium where everyone is wearing nearly identical clothing.

Cancer cells can also interfere with immune attacks. Some activate biological “brakes” that weaken T cells, which are among the immune system’s most important cancer-fighting cells. This is one reason a tumour can continue growing even when immune cells have entered the area around it.

But cancer cells are not perfect copies of normal cells.

Their DNA contains mistakes.

And those mistakes may expose them.

The Cancer’s Fingerprints

Cancer develops through genetic mutations, which are changes in the DNA instructions of a cell. Some mutations allow the cell to divide faster. Others help it avoid death, repair damage incorrectly or escape the immune system.

When mutated DNA is used to manufacture proteins, those proteins may contain unusual fragments that do not exist in healthy cells.

These abnormal fragments are called neoantigens.

The word can be separated into two parts:

“Neo” means new.

“Antigen” means something that the immune system may recognize and respond to.

Neoantigens are therefore new molecular features created by a tumour’s mutations. They can appear on the outside of cancer cells like tiny identification badges.

These badges are extremely valuable because they may allow the immune system to distinguish a cancer cell from the healthy cells surrounding it. Unlike many ordinary proteins found on both healthy and cancerous tissue, neoantigens can be specific to the tumour.

However, there is a problem.

A tumour may contain hundreds or thousands of mutations, but not every mutation creates a useful neoantigen. Some mutated proteins are never displayed to the immune system. Others are displayed but fail to create a strong response. A few may become excellent targets.

So how do scientists find the right ones?

They read the tumour.

Step One: Removing and Reading the Tumour

The process begins with a sample of the patient’s cancer, usually collected during surgery or a biopsy. Doctors may also collect healthy cells, often through a blood sample, so scientists can compare the patient’s normal DNA with the DNA found inside the tumour.

Powerful sequencing machines then read the genetic information.

Think of normal DNA as the original version of a very large book. The tumour’s DNA is another copy of the same book, but with spelling errors, missing sentences and altered instructions scattered across millions of letters.

Computers compare the two versions and identify the differences.

The tumour may contain an enormous number of mutations, so prediction algorithms are used to rank them. Scientists try to determine which mutations are actually producing abnormal proteins, which protein fragments are likely to appear on the surface of cells and which of those fragments have the greatest chance of attracting T cells.

This is one of the most difficult parts of the process. The computer is not simply searching for mutations. It is searching for mutations that can become useful targets.

For the experimental treatment now known as intismeran autogene, previously called mRNA-4157 or V940, computational systems can select up to 34 patient-specific neoantigens. Instructions for those targets are then placed into a single synthetic strand of messenger RNA.

At this point, the scientists have identified the cancer’s possible fingerprints.

Now they need to show those fingerprints to the immune system.

What Exactly Is mRNA?

To understand the next step, imagine that DNA is the master instruction book stored inside a protected library.

That library is the cell’s nucleus.

The DNA is extremely important, so cells generally do not carry the master book out whenever they need to build something. Instead, they produce a temporary copy of the required instruction.

That temporary copy is messenger RNA, or mRNA.

The mRNA travels to the cell’s protein-making machinery, where its instructions are read. Once its job is complete, it is broken down.

An mRNA vaccine uses this natural process. Scientists create a piece of mRNA containing instructions for selected antigens and deliver it into the body. The mRNA does not need to enter the nucleus, where DNA is stored, and it does not rewrite the patient’s genetic code.

In a personalized cancer vaccine, the instructions are not for a protein from a virus.

They are for carefully selected fragments of the patient’s own tumour.

This is where the technology becomes extraordinary.

Scientists are not taking a standard medicine from a shelf. They are manufacturing a new set of molecular instructions for one individual person.

Step Two: Building the Vaccine

Once the important tumour mutations have been selected, scientists design an mRNA sequence containing instructions for those neoantigens.

The treatment is manufactured, tested and prepared for injection.

After it enters the body, specialised cells known as antigen-presenting cells take up the mRNA. Dendritic cells, which are particularly important antigen-presenting cells, begin reading the instructions and producing the selected neoantigen fragments.

They then display these fragments on their surfaces.

Dendritic cells can be thought of as teachers inside the immune system. They collect suspicious material, carry it to other immune cells and effectively say:

“This is what the enemy looks like. Remember it.”

T cells that can recognize the displayed neoantigens are activated and begin multiplying. Some may travel through the body searching for cells carrying the same abnormal markers.

If they find a cancer cell displaying one of those neoantigens, they may attack and destroy it.

The vaccine therefore does not directly kill the cancer.

It creates a biological wanted poster and hands copies of it to the immune system.

A Vaccine After the Tumour Is Gone?

This may lead to an obvious question.

If doctors have already removed the tumour during surgery, why would the patient still need a cancer vaccine?

The answer is that removing everything visible does not always mean that every cancer cell has disappeared.

A few cells may have escaped before surgery. They may be travelling through the blood, hiding in another organ or remaining in such small numbers that scans cannot detect them.

These surviving cells can eventually begin dividing and cause the cancer to return.

Treatment given after surgery to reduce this risk is known as adjuvant therapy.

A personalized cancer vaccine may be especially useful during this period because the immune system is not being asked to destroy one enormous tumour. Instead, it may be hunting for a much smaller number of remaining cells.

In simple terms, it is usually easier for an army to hunt a few hidden soldiers than to attack an entire fortress.

However, the vaccine may still need help.

Even if T cells recognize the enemy, cancer can apply the brakes.

Removing the Immune System’s Brakes

T cells are extremely powerful. If they attacked everything without restriction, they could badly damage healthy tissue.

The body therefore has systems known as immune checkpoints. These checkpoints act like brakes, preventing immune responses from becoming dangerously aggressive.

One important checkpoint involves a protein called PD-1, which is found on T cells. When PD-1 connects with proteins such as PD-L1, the T cell receives a signal that reduces its attack.

This system normally helps protect healthy tissue.

Cancer cells, however, can take advantage of it.

Some tumours produce PD-L1 and use it like a fake security pass. The T cell approaches, receives a “do not attack” signal and becomes less effective.

Pembrolizumab, sold under the brand name Keytruda, is an immune checkpoint inhibitor. It attaches to PD-1 and blocks this suppressive signal, allowing T cells to continue attacking.

This explains why researchers are combining pembrolizumab with personalized vaccines.

The vaccine helps show the immune system what to attack.

Pembrolizumab helps remove a brake that can prevent the attack.

One provides the target.

The other gives the immune cells a better chance of reaching it.

Would this combination actually work in people?

To investigate that question, researchers conducted a trial involving patients with high-risk melanoma.

The Melanoma Trial That Changed the Conversation

Melanoma is a dangerous form of skin cancer. Even after it is completely removed, high-risk melanoma can return or spread to distant organs.

In the randomized Phase IIb KEYNOTE-942 trial, 157 adults with completely removed stage III or IV melanoma were divided into two groups.

A total of 107 patients received pembrolizumab combined with the personalized mRNA treatment then known as mRNA-4157 or V940. The remaining 50 patients received pembrolizumab alone.

The personalized treatment could encode up to 34 neoantigens selected from each patient’s tumour. Participants in the combination group were assigned nine doses of the individualized treatment, given once every three weeks. Pembrolizumab was given once every three weeks for up to 18 doses.

At the earlier analysis, the estimated percentage of patients who remained alive without their cancer returning at 18 months was 78.6% in the combination group, compared with 62.2% in the pembrolizumab-only group.

Researchers continued following the patients.

At a median follow-up of about five years, the combination was associated with a 49% lower relative risk of cancer recurrence or death than pembrolizumab alone. It was also associated with a 59% lower relative risk of the cancer spreading to a distant part of the body or causing death.

These results are important because the benefit did not disappear after a few months. The difference between the groups remained visible several years after treatment.

But the numbers must be understood correctly.

A 49% relative reduction in risk does not mean that 49% of all patients were cured. It compares how frequently recurrence or death occurred over time in one trial group against the other. It also does not prove that the treatment increases overall survival. The five-year overall-survival analysis showed an encouraging trend, but it was exploratory and statistically uncertain.

This was also a Phase IIb study involving only 157 people. That is large enough to be meaningful, but not large enough to settle every question.

That is why researchers moved to a much larger test.

The Bigger Test

A Phase III trial called INTerpath-001 is comparing intismeran autogene plus pembrolizumab with pembrolizumab alone in people with surgically removed, high-risk melanoma.

The study is designed to enrol approximately 1,089 participants, making it many times larger than KEYNOTE-942. It is intended to determine whether the results from the earlier study can be repeated across a much larger group of patients.

Until the Phase III evidence is available and reviewed by regulators, intismeran autogene remains an investigational treatment. It should not be described as an approved cancer cure.

This distinction is extremely important.

Medical history contains many treatments that looked promising in small studies but failed when tested in larger populations. Phase III trials are designed to reveal whether an apparent benefit is reliable, how large the benefit actually is and whether less common safety problems appear.

The personalized vaccine has passed an important test.

It has not passed every test.

What About the Side Effects?

Training the immune system is not always comfortable.

In the KEYNOTE-942 study, common effects attributed to the personalized treatment or combination included fatigue, pain at the injection site and chills. Serious or severe treatment-related side effects occurred in 25% of patients in the combination group and 18% of those who received pembrolizumab alone. No grade 4 or grade 5 events were attributed to mRNA-4157 in the published Phase IIb analysis.

Pembrolizumab also has its own risks.

Because checkpoint inhibitors release some of the immune system’s natural brakes, immune cells can sometimes begin attacking healthy organs. Depending on the person, this may affect the skin, lungs, intestines, liver, hormone-producing glands or other tissues.

This is why immunotherapy must be closely monitored by medical professionals. A treatment that strengthens an immune attack must remain strong enough to fight cancer without becoming uncontrolled.

Beyond Melanoma

Melanoma currently provides some of the strongest randomized evidence for a personalized mRNA neoantigen treatment, but it is not the only cancer being investigated.

In one small pancreatic-cancer study, scientists created individualized mRNA vaccines containing instructions for as many as 20 neoantigens. Eight of the 16 participants developed the immune response the researchers were seeking. Several years later, six of those responders remained cancer-free, although the study was far too small to prove that the vaccine caused this result.

A separate kidney-cancer trial tested personalized neoantigen vaccines in nine patients. None had experienced a recurrence when the results were reported, but there was no randomized control group, and some patients with surgically removed kidney cancer would be expected to remain cancer-free without the vaccine.

These results are interesting, but they remain early signals rather than final proof. Larger trials are required before researchers can determine whether the treatments genuinely delay recurrence.

Intismeran autogene is also being studied across other cancers, including non-small-cell lung cancer, bladder cancer and renal-cell carcinoma. Different tumours, however, have different numbers of mutations, different ways of suppressing the immune system and different levels of susceptibility to immunotherapy. A strategy that works in melanoma may not work equally well everywhere.

There may never be one universal cancer vaccine.

The future may involve many vaccines, designed for many patients, against many different collections of mutations.

The Challenges That Still Remain

The idea is brilliant, but making it work on a global scale will be extremely difficult.

Choosing the Correct Targets

Finding mutations is easier than identifying the ones that will produce a powerful immune response.

Prediction algorithms can estimate which neoantigens are likely to be displayed and recognized, but these predictions are not perfect. Selecting the wrong targets could produce T cells that never find the tumour.

It is like giving police officers an extremely detailed wanted poster of the wrong person.

Tumours Are Not Uniform

A tumour is not always one identical collection of cells.

Different areas can contain different mutations. A neoantigen found in one group of cancer cells may be missing from another. If the vaccine only targets the first group, the remaining cells could survive and continue growing.

Targeting multiple neoantigens is one possible solution, but tumour diversity remains a major obstacle.

Cancer Continues to Evolve

Cancer cells can change under pressure.

If immune cells destroy all cancer cells displaying a certain neoantigen, cells that do not display it may survive. Over time, those resistant cells can become dominant.

The wanted poster may eventually become outdated.

Manufacturing Takes Time

Each treatment must be designed and manufactured separately.

Scientists need to obtain a usable tumour sample, sequence it, analyse the mutations, select neoantigens, construct the mRNA, manufacture the doses and perform quality checks.

For a person with rapidly growing cancer, every week may matter.

Cost and Access

A factory producing millions of identical vaccine doses is very different from a factory producing thousands of separate medicines for thousands of individual patients.

Personalized manufacturing requires complex laboratory systems, secure movement of genetic information and a reliable method of ensuring that the correct treatment reaches the correct patient.

Even if the vaccines prove effective, making them fast, affordable and widely accessible may become one of the largest challenges facing the technology. Researchers continue to identify antigen selection, manufacturing, tumour variation and affordability as important barriers to widespread use.

Does This Mean Cancer Has Been Cured?

No.

Cancer is not one disease. It is a vast collection of diseases created by different mutations in different cells and organs.

Some cancers grow slowly. Others spread rapidly. Some contain many neoantigens. Others contain very few. Some respond strongly to immunotherapy, while others construct environments that keep immune cells out.

Personalized mRNA vaccines may therefore become one part of cancer treatment rather than a replacement for every existing method.

A patient may still require surgery to remove the main tumour, radiation to destroy local cancer cells, chemotherapy to attack rapidly dividing cells or targeted drugs aimed at a particular mutation.

The vaccine may then add something new.

It may train the immune system to continue patrolling the body after the visible tumour has disappeared.

This could be particularly valuable in preventing recurrence, because the immune system is not a medicine that immediately leaves the body. Some activated T cells can become memory cells, preserving information about their target and responding again if cells with the same neoantigens reappear.

Early studies in pancreatic and kidney cancer have detected neoantigen-reactive T cells several years after treatment, although scientists still need to determine how consistently this immune memory translates into protection from recurrence.

Conclusion

For most of medical history, cancer treatments were created for groups.

A treatment was designed for melanoma, lung cancer or pancreatic cancer, and every suitable patient received roughly the same medicine.

Personalized cancer vaccines introduce a completely different possibility.

The disease may have the same name, but the treatment can be built around the mutations inside one person’s tumour.

Doctors remove the cancer.

Sequencing machines read its DNA.

Computers search through its mutations.

Scientists manufacture an mRNA message.

Dendritic cells display the targets.

T cells begin the hunt.

What once appeared to be a tumour’s greatest strength, its ability to mutate, may also expose its greatest weakness. Every mutation that helps a cancer survive has the possibility of leaving behind a new fingerprint.

Scientists are trying to turn those fingerprints into targets.

The technology is not yet a universal cure, and important questions about effectiveness, manufacturing, safety, cost and accessibility remain. The larger Phase III trials will decide whether today’s encouraging results can become tomorrow’s standard treatment.

But the idea itself represents a major change in medicine.

Instead of asking only, “What type of cancer does this patient have?”, doctors may increasingly ask:

“What makes this patient’s cancer different from every other cancer?”

The answer could be hidden in the tumour’s genetic code.

And one day, that code may be used to manufacture the instructions that help destroy it.

References

• National Cancer Institute: Cancer Treatment Vaccines.
• Weber et al.: Individualised Neoantigen Therapy mRNA-4157 Plus Pembrolizumab Versus Pembrolizumab Monotherapy in Resected Melanoma, The Lancet.
• Carlino et al.: Five-Year Update of the KEYNOTE-942 Study, Journal of Clinical Oncology and ASCO 2026.
• ClinicalTrials.gov: KEYNOTE-942 and INTerpath-001.
• National Cancer Institute: Are New Immune-Based Treatments for Kidney and Pancreatic Cancer on the Horizon?