Summary
A phase 1 trial tested a personalized neoantigen vaccine after platinum chemotherapy in people with high-grade epithelial ovarian cancer. The researchers reported biological activity signals in six of 10 vaccinated patients, alongside immune responses in blood and tumour.
A phase 1 clinical trial reported that a personalized neoantigen vaccine could be made and administered to some people with high-grade epithelial ovarian cancer after platinum chemotherapy. In the 10 vaccinated patients, biomarker analysis suggested biological activity in six, accompanied by broader vaccine-specific and tumour immune responses. The study, published in Nature Medicine on 14 October 2026, was designed to assess feasibility and safety; its findings are early evidence of immune activity, not a measure of established clinical benefit.
How the personalized vaccine was used
The vaccine, called NeoVax, was made for each patient using synthetic long peptides targeting neoantigens selected from that person’s tumour. Neoantigens are altered protein fragments that can arise from cancer-cell mutations. When tumour cells display these fragments, immune cells may recognize them as targets. The vaccine peptides were mixed with poly-ICLC, an immune stimulant, and the treatment plan also included delayed administration of an anti-PD-1 antibody.
Participants had platinum-sensitive ovarian cancer and received the vaccine after standard-of-care platinum doublet chemotherapy. The report describes immune and biomarker findings in 10 vaccinated patients: one newly diagnosed patient and nine at first recurrence. Across the evaluable groups, vaccination was administered to 1 of 9 patients in the newly diagnosed cohort (11%; 90% CI 0.6–43%) and 9 of 12 in the recurrence cohort (75%; 90% CI 47–93%).
The safety findings were reassuring for this early-stage test: the researchers reported no unexpected safety signals, and no patient stopped treatment because of toxicity.
Immune activity and persistence of vaccine targets
Biomarker analysis was suggestive of biological activity in six of the 10 vaccinated patients. Those signals were associated with broader immune responses directed at the tumour and vaccine targets, measured in blood and tumour tissue after vaccination.
Personalized vaccines require time to develop from a tumour sample. In this trial, more than 70% of selected targets remained detectable through non-invasive cell-free DNA analysis across patients, despite a median of 6.5 months between diagnostic biopsy and vaccine administration. Patients received standard chemotherapy during that interval. The finding indicates that many selected targets could still be detected after the treatment and manufacturing period in this study.
The researchers also found that resistant tumour clones involved diverse changes affecting antigen presentation—the process by which cells display fragments for immune recognition. They interpreted these patterns as evidence consistent with immune-selective pressure from vaccination. This highlights a challenge for personalized cancer vaccines: tumours may vary in which targets they present, while the breadth of the immune response may influence how well those targets can be recognized.
The results provide early feasibility, safety and immune-response evidence for this approach in ovarian cancer. Whether these immune signals translate into durable clinical benefit remains to be determined in further clinical evaluation. The trial is registered as ClinicalTrials.gov NCT04024878.