Pancreatic Cancer Vaccine Triggered T-Cell Responses in 89 of 90 Patients, but Missed Its Main Endpoint

Mutations in the KRAS gene are found in about 90 percent of pancreatic ductal adenocarcinomas, the most common form of the disease, and the cancer often returns after surgery and chemotherapy. A vaccine built to teach the immune system to recognize seven of those mutations has now been tested against observation in a randomized phase 2 trial. The vaccine worked as an immune stimulant. It did not, by the trial's own primary measure, delay recurrence.

The results appeared in Nature Medicine, and Mayo Clinic announced them on October 23. Of 90 vaccinated patients whose blood could be evaluated, 89 developed T-cell responses against mutated KRAS. Median disease-free survival was 23.8 months with the vaccine and 21.2 months under observation, a hazard ratio of 0.85 with a P value of 0.5329. The authors state plainly that the primary endpoint was not met.

The trial and what it measured

The study, called AMPLIFY-7P, enrolled 144 patients across 24 U.S. centers with stage I to III KRAS-mutated pancreatic cancer. All had finished surgery and chemotherapy and showed no cancer on imaging. Researchers randomized 96 to the vaccine, called camupepimut or ELI-002 7P, and 48 to observation. Median follow-up at data cutoff in April 2026 was 19.4 months.

Recurrence or death had occurred in 42 of 96 vaccinated patients (43.8 percent) and 26 of 48 observed patients (54.2 percent). That gap points the right way, but with a P value above 0.5 it is compatible with chance.

A near-universal immune response

The immune data are the clearest result. Among the 89 responders, the median T-cell response was 44.3 times baseline, and 87.6 percent (78 of 89) responded to the specific mutation in their own tumor. Mayo reported that 95.5 percent of patients evaluable at least 46 weeks in still had a detectable KRAS-specific response, so the effect lasted past dosing.

Ben George, a Mayo Clinic medical oncologist and the study's lead author, said in the release that the team generated "a strong immune response against KRAS mutations in nearly every patient we evaluated." He also called the link to disease-free survival an exploratory finding. Safety looked similar between arms: grade 3 or higher adverse events occurred in 10.4 percent of vaccinated patients (10 of 96) and 14.6 percent of observed patients (7 of 48).

Where the encouraging numbers come from

The strongest-looking figure comes from inside the vaccine arm. Patients whose T-cell response exceeded 9.17 times baseline, 74 of 90, had median disease-free survival that was not yet reached. Those below the cutoff had 6.08 months, a hazard ratio of 0.22. That cutoff was not set in advance for this trial. It came from the phase 1 study, where an exploratory analysis picked the value that best separated outcomes.

A comparison within one arm cannot separate the vaccine's effect from the patient's own condition. Mayo noted that patients who had received radiation had fewer lymphocytes and weaker vaccine responses. Weak responders may simply be patients with weaker immune systems or more aggressive disease, and the trial cannot tell those explanations apart.

The paper reports a few subgroup signals as well. Patients with no detectable circulating tumor DNA at baseline had a hazard ratio of 0.51, and patients with no prior radiation had 0.59. The authors say these analyses were not adjusted for multiple comparisons and need validation. They also report that ctDNA-positive patients, who carry a much higher relapse risk, were overrepresented in the vaccine arm, which may have masked a benefit. That is the authors' explanation, and it has not been tested.

What the earlier trial predicted

The phase 1 trial of a two-mutation version of the vaccine treated 25 patients and reported a hazard ratio of 0.12 for relapse-free survival. At that size, a result that dramatic often shrinks in a larger randomized comparison. Here the headline effect fell to 0.85. Nobody can yet say whether the vaccine has a modest real effect that this trial was too small to show, or none.

Who paid, and what comes next

Elicio Therapeutics sponsored and funded the trial and designed it with the academic authors, and company employees contributed to the work. The study was open-label, so doctors and patients knew who received the vaccine, and it used a two-sided alpha of 0.1. Overall survival results are not available. A phase 3 trial is planned, and the paper says the data will inform a refined strategy for it.