A personalized mRNA cancer vaccine produced a long-lasting immune response in some pancreatic-cancer patients, and seven of the eight responders were still alive four to six years after surgery, according to extended follow-up data from a small phase 1 trial.
The experimental vaccine, called autogene cevumeran, was designed separately for each patient using genetic information taken from that person’s tumor. It was given after surgery as part of a treatment program that also included immunotherapy and chemotherapy.
The findings are encouraging because pancreatic cancer remains extremely difficult to treat. However, the trial included only 16 vaccinated patients, was not randomized and cannot yet prove that the vaccine itself caused the longer survival.
Seven of Eight Vaccine Responders Were Still Alive
Researchers at Memorial Sloan Kettering Cancer Center reported that the vaccine generated a measurable tumor-specific immune response in eight of the 16 patients.
Among those eight responders, seven, or 87.5%, were still alive four to six years after surgery. In comparison, two of the eight patients who did not generate the measured immune response were alive at the latest follow-up, and the nonresponders had a median survival of 3.4 years.
The results were presented at the 2026 annual meeting of the American Association for Cancer Research. They extend earlier follow-up findings showing that responders also experienced substantially longer periods without the cancer returning.
Readers can review the detailed results in Memorial Sloan Kettering’s report on the pancreatic-cancer vaccine trial.
This Was a Treatment Vaccine, Not a Preventive Shot
The vaccine was not designed to stop healthy people from developing pancreatic cancer.
It is a therapeutic cancer vaccine intended for patients who have already been diagnosed and undergone surgery. Its purpose is to train the immune system to recognize and destroy microscopic cancer cells that may remain after the visible tumor has been removed.
These remaining cells can be difficult to detect with scans but may later grow and cause the cancer to return.
Unlike vaccines against infections, a personalized cancer vaccine must teach the immune system to distinguish malignant cells from the patient’s healthy tissue. That is challenging because cancer originates from the person’s own cells rather than from an external virus or bacterium.
How the Personalized mRNA Vaccine Was Created
After surgery, researchers genetically sequenced each patient’s tumor to identify mutations found in the cancer cells.
Some mutations create abnormal proteins known as neoantigens. Because these neoantigens are present on cancer cells but not normal cells, they can act as targets for the immune system.
BioNTech used each patient’s tumor information to manufacture an individualized mRNA vaccine encoding as many as 20 selected neoantigens. The finished vaccine was then returned to New York and infused into the patient.
The mRNA provided temporary instructions that helped immune cells recognize the tumor’s unique molecular markers. It did not alter the patient’s DNA.
The American Association for Cancer Research’s explanation of the trial describes why mRNA was selected: it can support relatively rapid manufacturing while targeting several patient-specific mutations simultaneously.
The Vaccine Trained Cancer-Killing T Cells
The researchers focused heavily on CD8-positive T cells, which can identify and destroy abnormal cells.
Eight patients developed strong expansions of T cells that recognized neoantigens encoded in their personalized vaccines. Most of those immune-cell populations had not been detected in patients’ blood or tissue before vaccination, suggesting that the treatment generated new antitumor responses rather than merely amplifying existing ones.
The T cells were also durable. Researchers found that vaccine-induced immune cells remained detectable for years and retained the ability to respond to the targeted cancer markers.
The peer-reviewed study published in Nature reported that approximately 86% of tracked vaccine-induced T-cell clones remained present at meaningful levels around three years after vaccination. Some were estimated to have exceptionally long potential lifespans.
The latest follow-up indicates that these cancer-targeting CD8-positive T cells have persisted for as long as six years without obvious evidence that their activity is fading.
Why Pancreatic Cancer Is So Difficult to Treat
Pancreatic ductal adenocarcinoma is the most common form of pancreatic cancer and frequently causes few recognizable symptoms during its early stages.
By the time it is diagnosed, the disease may already have spread or become too advanced for surgery. Even among patients whose tumors can be surgically removed, microscopic cancer cells often remain and eventually cause recurrence.
Pancreatic tumors can also create an environment that suppresses immune activity, making many existing immunotherapies less effective than they are against cancers such as melanoma or lung cancer.
The five-year survival rate across pancreatic-cancer patients is approximately 13%, although individual outcomes vary greatly according to the stage, tumor biology and whether surgery is possible.
The survival results among vaccine responders are therefore notable, but they should not be directly compared with the overall national survival rate without considering that trial participants had surgically removable disease and were healthy enough to receive several treatments.
Patients Received More Than the Vaccine
The trial did not test the vaccine by itself.
Participants first underwent surgery to remove their tumors. They then received a dose of atezolizumab, an immunotherapy medicine that blocks the PD-L1 pathway, followed by several vaccine doses and modified FOLFIRINOX chemotherapy.
The vaccine was therefore one component of a combined treatment program.
Because there was no randomly assigned group receiving the same treatment without the vaccine, the phase 1 study cannot determine how much of the outcome resulted from surgery, chemotherapy, immunotherapy, the vaccine or differences between the patients themselves.
The fact that survival and recurrence outcomes differed between immune responders and nonresponders supports further study, but it does not prove that generating the measured response will always produce longer survival.
Why Only Half of the Patients Responded
Eight of the 16 vaccinated patients developed the strong T-cell response researchers were seeking, while the other eight did not.
Scientists are still trying to understand why.
Differences in tumor biology, immune-system function, vaccine targets and the quality of the selected neoantigens may influence whether a patient responds.
The timing of vaccine production may also matter. Personalized vaccines must be designed and manufactured after tumor sequencing, creating logistical challenges in a disease where treatment cannot be delayed unnecessarily.
Researchers are studying biological markers that might help predict which patients are most likely to benefit and how the vaccine could be improved for those who do not initially respond.
The Results Do Not Mean Pancreatic Cancer Has Been Cured
Seven long-term survivors among eight responders is an important result, but the trial is far too small to establish a cure rate.
Only 16 vaccinated patients were evaluated, and the participants were treated at a single specialist cancer center. A few unusually good or poor outcomes can strongly affect the percentages in such a small group.
The trial was also designed primarily to assess feasibility, safety and immune response—not to definitively prove improved survival.
Researchers acknowledged the small sample size and single-arm design as major limitations.
The term “responder” also refers to a laboratory-confirmed immune response rather than proof that every cancer cell had been eliminated. Some responders experienced recurrence, although the group overall had longer recurrence-free survival than the nonresponders.
A Larger Randomized Trial Is Underway
The early findings have led to a global randomized phase 2 study.
That trial is comparing standard modified FOLFIRINOX chemotherapy with a treatment program that adds autogene cevumeran and atezolizumab. It is enrolling a larger group of patients with surgically removable pancreatic cancer at multiple medical centers.
Randomization will allow researchers to compare similar patients and determine more reliably whether adding the personalized vaccine reduces recurrence or extends survival.
The larger study will also provide more information about adverse effects, manufacturing timelines and whether the treatment can be delivered consistently outside a single leading cancer center.
Until those results are available, autogene cevumeran remains investigational and is not an approved standard treatment for pancreatic cancer.
Personalized Vaccines Could Be Difficult and Expensive to Scale
Each vaccine must be custom-built using material from an individual patient’s tumor.
The process involves surgery, tumor sequencing, computer analysis, selection of suitable neoantigens, vaccine manufacturing, quality testing and delivery back to the treatment center.
That level of personalization may increase expense and make access difficult in hospitals without specialized genomic and manufacturing resources.
Researchers are working to shorten production times and expand the infrastructure needed to make mRNA cancer vaccines. Memorial Sloan Kettering has said it is developing in-house manufacturing capacity to support broader testing of personalized vaccines for pancreatic and other difficult cancers.
Other teams are studying ready-made vaccines targeting common mutations such as KRAS. Those products could potentially be manufactured in advance, although they would not offer the same level of individual customization.
The Vaccine’s Side Effects Appeared Manageable in the Early Trial
The phase 1 program previously found that the personalized vaccine could be administered successfully as part of the larger treatment regimen.
Patients may experience temporary immune-related effects after mRNA vaccination, including fever, chills, fatigue or reactions related to infusion. However, the small trial was not large enough to identify uncommon safety problems.
Some participants described the chemotherapy as more difficult than the vaccine, but individual experiences cannot replace systematic safety data from larger controlled studies.
Future trials must determine whether any added benefit justifies the additional treatment burden, cost and possible side effects.
What the Six-Year Findings Really Show
The strongest conclusion is that a personalized mRNA vaccine can generate durable, cancer-specific T cells in some patients with pancreatic cancer.
Those immune responses remained detectable for several years and were associated with longer recurrence-free survival and encouraging long-term survival.
The findings do not yet establish that the vaccine will work for most pancreatic-cancer patients. They also do not show whether it can treat inoperable or metastatic disease, since the trial involved patients whose tumors could be removed surgically.
Still, keeping seven of eight immune responders alive four to six years after treatment is a meaningful signal in a disease where recurrence is common and long-term survival remains difficult to achieve.
The ongoing randomized trial will determine whether that signal develops into reliable evidence for a new treatment.
