Personalized Cancer Vaccines
A vaccine designed to specifically target an individual's unique tumor. Built with our design and manufacturing partner, Glyphic Biotechnologies, and administered at our clinic in Stratham, New Hampshire.
What is a personalized neoantigen vaccine?
Most vaccines prevent illness in healthy people. This one is given to someone who already has cancer, and its job is to teach the immune system to find that cancer and attack it.
As cancer cells divide, they accumulate mutations, and some of those mutations produce proteins found nowhere else in the body. These are neoantigens—markings that belong to the tumor alone. Because they are absent from healthy tissue, they are considered ideal targets: an immune response aimed at them should spare normal cells.
How yours would be made
1Tumor tissue is collected
From a biopsy or surgery. While we can often use tissue collected during a prior procedure, it is best for us to work with you and your treating team to coordinate collection, so that it is preserved in a way that allows our partner to design the best vaccine possible.
This is the step that makes the whole therapy possible, and the reason someone without a tumor cannot receive it.
2Its DNA, RNA, and protein are analyzed
DNA encodes RNA, which in turn encodes proteins. Our partner analyzes all three to search for mutations unique to the tumor.
Healthy tissue, usually blood, is also analyzed, so that the mutations belonging only to the tumor can be told apart from the normal ones in your body.
3Targets are chosen
A tumor may carry thousands of mutations, and most would make poor targets. Analysis narrows them to the handful of neoantigens most likely to be visible to your particular immune system.
This is where personalization actually happens. Two people with the same type of cancer will almost never have exactly the same targets.
4The vaccine is designed, manufactured, and administered
Protein fragments matching those targets are manufactured into a therapy for you alone, then given by injection alongside an immune adjuvant. Our partner aims to create a vaccine targeting up to 30 neoantigens.
Designing and manufacturing takes roughly ten weeks. This is the longest part of the process and the stretch where patients most often feel they are waiting without news. We will tell you where things stand rather than leaving you to ask.
Dosing then runs about nineteen weeks. Doses are spaced closely at first and then further apart, on days 1, 4, 8, 15, 22, 78, and 134, so the immune system gets an initial concentrated series followed by two later reinforcements. Each visit includes an hour of observation afterward before you go home. Monitoring continues quarterly after that.
Those protein fragments are called synthetic long peptides. A therapy built for a single person is called an n-of-1 therapy: a treatment with exactly one patient.
Why the peptides are never given alone
A peptide tells the immune system what to look for. On its own it rarely produces a strong response. What turns recognition into a real immune reaction is what the peptides are delivered with—an immune adjuvant, mixed into the same injection.
We use poly-ICLC, sold as Hiltonol. It mimics viral genetic material, which signals to the immune system that a threat is present and worth responding to.
This is not our invention, and it is not new. Poly-ICLC has been the adjuvant in personalized neoantigen peptide vaccine research since the first-in-human trial at Dana-Farber in 2017, and it has carried through the studies that followed: the phase Ib glioblastoma trial, the 173-patient glioblastoma series, the PGV001 trial at Mount Sinai across five cancer types, and the 2025 Dana-Farber melanoma trial. Across those studies its safety record is consistent: side effects have been mild, most often injection-site reactions and flu-like symptoms.
The peptide is the introduction. The adjuvant is what makes the immune system pay attention.
▸Adjuvants vs. immunotherapy drugs—what's the difference?
An adjuvant is part of the vaccine, in the same syringe. It makes your immune system notice the peptides.
Immunotherapy drugs—such as checkpoint inhibitors—are separate cancer medicines from your oncologist, on their own schedule. They release the brakes on immune cells.
Both involve the immune system. They are not the same treatment, and some patients receive both.
Who is this most, and least, likely to help?
Two things drive whether this kind of therapy can help: how much cancer is present, and how much time there is. A vaccine does not attack a tumor directly. It trains the immune system to do so, and that training takes weeks to months.
First, who can be considered at all
Before any of the situations below apply, two basic conditions have to be met.
- There has to be a tumor sample. The vaccine is designed from your tumor's cellular and genetic data, so tissue from a surgery or biopsy is required. Without it, there is nothing to design from.
- You have to qualify for expanded access. This is the FDA pathway that allows investigational treatments outside a clinical trial. The requirements are set out in part four below.
If those hold, the question becomes how well your situation fits what the research has shown. Ideal candidates have a lower burden of cancer and a promising prognosis. The situations below follow from that principle. They are illustrations rather than a checklist, and none is a determination about any individual. We rely on the guidance of our Medical and Scientific Advisory Boards to help determine the eligibility of individuals who may not fall neatly into these categories.
▸After surgery, when scans are clearBest studied
The visible tumor has been removed and imaging shows no disease. But microscopic cancer cells can remain behind, and those are what cause recurrence years later. Doctors call this minimal residual disease.
This is the situation the research has studied most, and where results have been most encouraging.
Why: almost nothing left to clear, and time to build a response before anything can grow back.
▸Keeping cancer away after chemotherapy workedFavorable
Chemotherapy has substantially reduced the cancer, and the goal now is to keep it from coming back. This is sometimes called maintenance.
Why: same logic as above—little disease present, and time available before any return.
▸Already on an immunotherapy drug like Keytruda or OpdivoUnder study
These drugs are called checkpoint inhibitors—the immunotherapy drugs mentioned in part one. Tumors can switch off the immune cells that come after them; a checkpoint inhibitor switches those cells back on. What it cannot do is tell them what specifically to look for.
That is the gap a vaccine fills. One releases the brakes, the other supplies the target, which is why most of the strongest published research gives both together rather than testing a vaccine on its own.
If you are already receiving one of these drugs, that is generally a point in favor rather than a conflict.
Why it is not yet settled: because the studies combine both treatments, it is hard to measure how much of the benefit comes from the vaccine itself.
▸Large or fast-growing cancerLeast favorable
This is the hardest situation, and we will say so plainly.
Three things work against each other here. There is a great deal of cancer for an immune system to clear. Large tumors actively suppress the immune cells around them. And designing and manufacturing the therapy takes weeks that a fast-growing cancer may not wait through.
This does not make the conversation pointless. It means expectations should be set very differently than in the situations above.
▸High cancer risk, but no cancerNot a candidate
The vaccine is built from a tumor sample. Someone carrying a BRCA or Lynch syndrome mutation—inherited gene changes that raise lifetime cancer risk—or a firefighter with years of occupational exposure, may face genuinely elevated risk—but with no tumor, there is nothing from which to design a vaccine.
This is not a preventive vaccine and cannot be used as one. But preventing cancer in people at high risk is an active field of research, and it is moving. These are three recent examples rather than a complete list—there are many more, in many cancers.
In Lynch syndrome, an off-the-shelf vaccine called Nous-209 was given to 45 people carrying the mutation. Published in Nature Medicine in 2026, it produced immune responses in every evaluable participant with no serious side effects.
In people with precancerous pancreatic cysts, a peptide vaccine called mKRAS-VAX was tested in 20 high-risk individuals at Johns Hopkins. Responses lasted up to two years, and cysts shrank or resolved more often in vaccinated people than in a comparison group—though the researchers say plainly that larger studies are needed to show the vaccine caused it.
In breast cancer, a MUC1 vaccine is being studied in women with DCIS—ductal carcinoma in situ, abnormal cells still confined within a breast duct—to see whether it can stop those early changes from becoming invasive.
All of these are early-stage. None is approved, and none is available outside a study. If this interests you, trial listings are public, and a genetic counselor can help you find what is open.
How strong is the evidence, really?
Coverage of cancer vaccines tends to quote the best number from the best study and stop there. The more useful thing to see is how much these studies differ in the weight they can carry. Six patients with no comparison group is not the same kind of evidence as 157 patients with one.
A control arm is a second group of patients, assigned at random, who do not receive the vaccine. It is the only way to know whether the people who did well would have done well regardless. Just one of the studies below has one.
Some trials compare patients within the study instead—for example, those whose immune systems responded to the vaccine against those whose did not. That is worth knowing, but it is weaker evidence, because those two groups were not chosen at random. People whose immune systems mount a strong response may have been healthier or had less aggressive disease from the start.
got the vaccine Comparison group Follow-up Type of evidence
▸
INTerpath-001mRNA · melanomaMerck & Moderna, topline, Aug 2026
confirms KEYNOTE-942
1,137randomized 2:1
Yespembrolizumab alone
Interim
Randomized · phase 3
What the study asked, and what it found
Whether adding a personalized mRNA vaccine to pembrolizumab keeps melanoma from returning after surgery, tested at the scale required for approval—1,137 patients with completely resected stage IIB to IV melanoma, randomized, double-blind, placebo-controlled. At a planned interim analysis the trial met its primary endpoint of recurrence-free survival and its key secondary endpoint of distant metastasis-free survival, both described as statistically significant and clinically meaningful against pembrolizumab alone. Safety was consistent with earlier studies, with no new signals. This is the first positive phase 3 result for any individualized neoantigen therapy, and the first for an mRNA-based cancer therapy.
It is the confirmatory step for the trial below. Same vaccine, same drug pairing, same setting—but a new and much larger group of patients. A small trial can produce an encouraging result by chance, and the only way to find out whether it holds is to run it again in different people. It held.
Where it stops short
As of August 2026 this is a topline announcement from the companies, not a published paper. No effect sizes, confidence intervals, or survival curves have been released, so how large the benefit is remains unknown. Overall survival—whether people live longer, rather than staying cancer-free longer—is still being followed and has not been reported. The results have not been peer-reviewed or presented at a medical meeting, and regulators have not yet reviewed them. And this is an mRNA vaccine, not a peptide one.
▸
KEYNOTE-942mRNA · melanomaWeber et al., The Lancet, 2024
confirmed by INTerpath-001
107received the mRNA vaccine
Yes—randomized50 on pembrolizumab alone
~2 years
Randomized trial · phase 2b
What the study asked, and what it found
Whether adding a personalized mRNA vaccine to the immunotherapy drug pembrolizumab keeps melanoma from coming back after surgery. At 18 months, 79% of people on the combination had no recurrence, against 62% on pembrolizumab alone. This is the only study here that randomly assigned people to get the vaccine or not, which is what makes it possible to credit the difference to the vaccine itself.
Where it stops short
The result met the success threshold the researchers set in advance, but that threshold was deliberately more lenient than the one used in large, final-stage trials. With 157 people, the range of outcomes still statistically consistent with this data runs from a substantial benefit down to roughly no difference at all. Everyone knew which treatment they were getting. Severe treatment-related side effects were somewhat more common in the vaccine group (25% vs 18%), though serious events were similar in both. At five years, follow-up of these same patients showed a 49% reduction in the risk of recurrence or death. The confirmatory phase 3, INTerpath-001, has since read out positively in a separate and much larger group—see the row above.
▸ Glioblastoma real-world seriespeptide · glioblastomaLatzer et al., Nature Communications, 2024 17370 vaccinated before regrowth, 103 after Not randomized159 matched patients from earlier published studies 2015–2023 Observational · not a trial
What the study asked, and what it found
A look back at the records of 173 people with glioblastoma who received a custom peptide vaccine outside any study, in Germany. In the matched comparison, half the vaccinated patients were still alive 31.1 months after diagnosis, against 22.7 months in the comparison group. Of the 97 people whose immune systems were monitored, 87 responded to at least one target, and side effects were almost all mild—4 of 173 had a serious reaction.
Where it stops short
Patients did not get their first dose until a median of 10.3 months after diagnosis, so only people already living longer than most survived long enough to be vaccinated at all. That fact alone can make a treatment look effective when it is not. The researchers tried to correct for it, but the correction is imperfect. Patients also had to afford the vaccine and travel to Germany, and each received a different mix of other treatments. The authors state plainly that this shows correlation, not cause, and that a randomized trial is needed.
▸ MSK pancreaticmRNA · pancreatic cancerRojas et al., Nature, 2023; 3-year follow-up 2025 16of the 28 who had surgery on the trial NoneEveryone received the vaccine ~3 years Early-phase · phase 1, single arm
What the study asked, and what it found
The stated main goal was safety. It was safe, could be delivered on time after major pancreatic surgery, and in 8 of the 16 people produced a strong T-cell response. Roughly three years on, 6 of those 8 responders still had no sign of cancer returning, while 7 of the 8 non-responders had recurred.
Where it stops short
There was no comparison group. That striking difference comes from comparing people within the trial whose immune systems happened to respond against those whose did not, which is not the same as comparing vaccine against no vaccine. The researchers did test the most obvious alternative explanation: responders and non-responders mounted equally strong responses to a COVID-19 vaccine given at the same time, so responders did not simply have healthier immune systems. Even so, everyone received an immunotherapy drug and 15 of 16 received chemotherapy. With 16 people, these are preliminary findings; a larger randomized trial is underway.
▸ PGV001peptide · five cancer typesSaxena, Marron et al., Cancer Discovery, 2025 13of 14 enrolled; 11 finished all 10 doses NoneEveryone received the vaccine up to 5 years Early-phase · phase 1, single arm
What the study asked, and what it found
Whether a single vaccine-design process could work across very different cancers. Thirteen people—head and neck, multiple myeloma, lung, breast, and bladder—were vaccinated after finishing their main treatment, at high risk of recurrence. All 13 developed T-cell responses against targets in their vaccine. Every side effect was mild, with no grade 3 or 4 reactions.
Where it stops short
Thirteen people across five cancers with no comparison group cannot tell us whether the vaccine changes survival, and the authors say so directly. Patients received a range of other treatments alongside it. At five years, 6 of the 13 were alive and 3 of those 6 remained tumor free; 6 had died—one from infection and one from a heart event unrelated to cancer—and 1 was lost to follow-up.
▸ NeoVax multi-adjuvantpeptide · melanomaBlass, Keskin et al., Cell, 2025 10of 11 enrolled; 9 finished all 4 doses NoneEveryone also received ipilimumab and nivolumab ~2 years Early-phase · phase 1, single arm
What the study asked, and what it found
A technical question: can adding extra immune-stimulating ingredients produce a stronger response than earlier versions managed? The formal primary goal was safety, with immune response secondary, and on both counts it delivered. All 9 who completed the course developed T cells against most of their targets, and 6 of 9 developed the killer CD8 T cells earlier versions had struggled to generate. No one had a severe side effect from the vaccine.
Where it stops short
This was not built to test whether the vaccine helps people live longer or stay cancer-free—with 10 participants and no comparison group it cannot answer that, and the authors say so. Everyone also received two standard immunotherapy drugs, either of which can produce immune activity on its own.
▸ NeoVaxpeptide · melanoma · first in humansOtt et al., Nature, 2017 6of 10 enrolled NoneEveryone received the vaccine ~2 years Early-phase · first in humans
What the study asked, and what it found
The study that showed the idea was possible at all. Could a vaccine be built from one person's own tumor mutations, manufactured in time, and given without serious harm? For all 6 people it could. Side effects were mild, and the vaccine generated T cells against most of its targets.
Where it stops short
Six people, no comparison group, and no design capable of testing whether the vaccine prevents recurrence. Four of the six had no recurrence about two years after vaccination; the two who did both then responded completely to a checkpoint inhibitor. Those numbers are far too small to draw conclusions from. The value lies in proving the approach was workable.
Peptide and mRNA vaccines are different approaches to the same idea, and results from one do not automatically transfer to the other. Our therapy is peptide-based. Tap any study to see what it found and where it stops short.
▸How to read the "type of evidence" column
Randomized trial. People were assigned by chance to receive the vaccine or not. This is the only design that lets researchers say a difference in outcomes was caused by the vaccine rather than by differences between the groups.
Observational study. Researchers looked at what happened to people who received the vaccine outside a trial, and compared them with records of other patients. Useful for spotting patterns and confirming safety, but it cannot rule out that the two groups differed in ways that explain the result.
Early-phase trial. A small first study, usually 6 to 20 people, where everyone receives the vaccine. Designed to answer whether it can be made, whether it is safe, and whether it activates the immune system—not whether it works against cancer.
A note on the numbers. Survival and recurrence figures from small studies carry wide uncertainty, and a result that looks dramatic in 8 or 16 people may shrink or disappear in a larger trial. None of the studies here, including KEYNOTE-942, were designed to measure whether these vaccines help people live longer—the randomized trial measured whether cancer came back.
What is actually established
These vaccines reliably do something biologically. In the Dana-Farber melanoma trial, all nine patients who completed the full series developed immune responses against most of their targets. In the glioblastoma series, a response was detected in 87 of 97 patients monitored. That reproduces across platforms, cancers, and research groups: the immune system does learn the targets it is shown.
Safety is equally consistent. Across these trials there were no dose-limiting toxicities, and side effects were mostly mild—injection-site reactions, fatigue, low-grade fever.
And as of August 2026, something larger is established: the approach itself works. A phase 3 trial has now shown that adding a personalized vaccine to standard immunotherapy keeps melanoma from returning longer than the immunotherapy alone. Until this month that was an open question. It is no longer.
Separately, the safety of this vaccine format is an older question with a much longer record. Synthetic long peptides were proposed as a cancer treatment strategy in 2008 and tested in patients in a 2009 New England Journal of Medicine trial, well before personalization was possible.
What the randomized trials show
Two studies here randomly assigned patients to receive the vaccine or not. In the phase 2b melanoma trial, 79% of people receiving the vaccine plus pembrolizumab had no recurrence at 18 months, against 62% on pembrolizumab alone. The result met the threshold researchers had set in advance, though that threshold was deliberately more lenient than a final-stage trial would use.
That trial was designed to be confirmed rather than relied upon, and in August 2026 the confirmation arrived. INTerpath-001, a 1,137-patient phase 3, met both its primary endpoint of recurrence-free survival and its key secondary endpoint of distant metastasis-free survival. It is the first positive phase 3 result for any individualized neoantigen therapy, and it settles a question that had been genuinely open: whether this approach works at all.
Two things about it are worth holding onto. The announcement is topline only—no effect sizes have been published, so how large the benefit is remains unknown, and whether patients live longer rather than simply staying cancer-free longer is still being followed. And it used an mRNA vaccine.
What is genuinely missing
No personalized neoantigen cancer vaccine is approved by the FDA. Two therapeutic cancer vaccines are approved in the United States—sipuleucel-T for prostate cancer and T-VEC for melanoma—but neither is a personalized neoantigen vaccine. In a 2026 review of 78 registered trials, more than 90 percent were still at the earliest stage of human testing.
Both randomized trials above used mRNA vaccines. No randomized trial of a peptide vaccine like ours has been completed. Every peptide study in the table is single-arm. What has been shown for this approach is that it can be manufactured, given safely, and produce the immune response you would want to see. Clinical benefit has not been demonstrated for peptide vaccines specifically.
The phase 3 result matters even so. It establishes that teaching an immune system to recognize a patient's own tumor can change what happens to them—which is the principle underlying our therapy too. What it does not establish is that our particular version delivers the same benefit.
Response is also not universal. Where some patients mounted strong immune responses and others did not, benefit concentrated among the responders, and predicting who will respond remains unsolved.
The idea has now been proven. Our particular version of it has not. Those are different statements, and we will not blur them.
What we do, and what we do not do
What we do
We work with our partner, Glyphic, to design, manufacture, and administer personalized peptide neoantigen vaccines. We handle eligibility review, the regulatory pathway, coordination of tissue and processing, and the treatment visits. We communicate with your oncologist throughout, and we tell you where things stand—including during the manufacturing wait, when there is often no news to give.
What we do not do
We do not take over your cancer care. Your oncologist remains your oncologist. We do not diagnose, stage, or direct your chemotherapy, surgery, or radiation, and we will not ask you to stop treatment that is working.
We also do not tell you this will work. It is an investigational therapy. We can tell you what the research shows, what it does not, and where your situation sits relative to what has been studied.
How access works
Because these therapies are investigational, they are provided through expanded access, a regulated FDA pathway sometimes called compassionate use. The FDA sets out the criteria as follows:
Patient has a serious or immediately life-threatening disease or condition.
There is no comparable or satisfactory alternative therapy to diagnose, monitor, or treat the disease or condition.
Patient enrollment in a clinical trial is not possible.
Potential patient benefit justifies the potential risks of treatment.
Providing the investigational medical product will not interfere with investigational trials that could support a medical product's development or marketing approval for the treatment indication.
—U.S. Food and Drug Administration
A licensed healthcare provider requests treatment on your behalf, and an institutional review board—an independent committee that exists to protect patients—reviews the plan and consent documents before anything begins.
Right to Thrive is a cash-pay clinic. Not everyone who contacts us will be eligible, and we would rather say so early and plainly than late.
Questions worth asking your oncologist
- What is the current stage of my cancer?
- Is there banked tumor tissue from a previous surgery or biopsy? If so, how was it preserved, how much is left, and who do I contact to get access to it?
- Is there an open clinical trial I would qualify for? Trials provide treatment at no cost.
- How would this fit alongside my current treatment, and would anything need to pause?
- Realistically, do I have the weeks this takes to design and manufacture (~10 weeks)?
- Would you be willing to communicate with the team at Right to Thrive if we decide a personalized cancer vaccine is worth pursuing?
If you have read this far, you are probably carrying something heavy, for yourself or for someone you love.
We wish we could tell you this will work. We won't, because you have likely been promised things before, and you deserve better than another promise.
What we can tell you is that this clinic exists because the people who built it sat where you are sitting. They watched someone they love run out of options, and could not accept that the care that might have helped was somewhere else, out of reach.
You are entitled to the best care that exists, when you need it. Whatever you decide from here, we hope you find it. And if it turns out not to be us, we will tell you honestly, and help you keep looking.
Start the conversation
If you are a patient, a family member, or a healthcare provider with a question about whether this therapy may be appropriate, we would like to hear from you.
Start the conversation →