A personalized neoantigen cancer vaccine triggered tumor- and vaccine-specific immune responses in a phase 1 trial of 10 patients with high-grade epithelial ovarian cancer, according to a study published in Nature Medicine. The vaccine, called NeoVax, was tailored to mutations in each patient’s tumor and combined with delayed PD-1 blockade. Biomarker analyses suggested biologic activity in 6 of 10 vaccinated patients, but the trial was small and designed primarily to assess feasibility and safety rather than prove that the vaccine improves survival.
The vaccine was built from mutations unique to each patient’s tumor
NeoVax is a personalized neoantigen vaccine. Instead of targeting a protein shared by most tumors, researchers sequence an individual patient’s cancer and identify mutations that create abnormal protein fragments, or neoantigens, that are not present in normal cells.
Selected neoantigens are then manufactured as synthetic long peptides and combined with an immune-stimulating compound called poly-ICLC.
The goal is to teach T cells to recognize several tumor-specific targets at once.
That approach is conceptually similar to other personalized cancer vaccines now being tested across melanoma, kidney cancer and glioblastoma. Bites of Bio previously covered a personalized mRNA cancer vaccine in melanoma, although NeoVax uses synthetic peptides rather than an mRNA platform.
Ovarian cancer presents a difficult test for personalized vaccines
Ovarian cancer has generally been more difficult to treat with immunotherapy than cancers such as melanoma.
One challenge is that many ovarian tumors carry fewer straightforward point mutations that generate obvious neoantigens. They can also develop mechanisms that prevent tumor antigens from being effectively displayed to T cells.
That made ovarian cancer an important test case for whether a personalized vaccine could generate meaningful antitumor immunity even in a disease that is not typically considered highly responsive to checkpoint blockade.
The trial included newly diagnosed and recurrent disease
The open-label phase 1 trial, registered as NCT04024878, enrolled patients with high-grade epithelial ovarian, fallopian tube or primary peritoneal cancer.
The study included two cohorts. Cohort A included patients with newly diagnosed advanced disease receiving neoadjuvant platinum-based chemotherapy. Cohort B included patients being treated for their first platinum-sensitive recurrence.
The vaccine was manufactured from tumor tissue collected before chemotherapy. Patients then received standard platinum-based treatment while the personalized vaccine was being designed and produced.
Ultimately, 10 patients were vaccinated: one newly diagnosed patient from Cohort A and nine patients from Cohort B.
Manufacturing feasibility differed sharply between the two cohorts
The study’s feasibility results were very different depending on when patients entered the trial.
Among 12 evaluable patients in the first-recurrence cohort, 9 received the vaccine, corresponding to a feasibility rate of 75%.
Among 9 evaluable patients in the newly diagnosed cohort, only one ultimately received vaccination, corresponding to 11%.
That difference is important because highly personalized therapies can fail before treatment begins if suitable tissue, sequencing data, manufacturing time or clinical circumstances do not line up.
The study therefore tested not only whether the vaccine could stimulate immunity, but also whether a bespoke treatment could realistically be produced and delivered within the timeline of ovarian cancer care.
Most selected vaccine targets were still present months later
A personalized vaccine is only useful if the mutations selected at the time of biopsy are still present when the vaccine is eventually administered.
That is not guaranteed. Tumors evolve under treatment, and chemotherapy can change which tumor clones survive.
The researchers therefore used cell-free DNA in blood to track the mutations selected as vaccine targets.
More than 70% of selected targets remained detectable despite a median interval of 6.5 months between the diagnostic biopsy and vaccine administration, a period during which patients received standard chemotherapy.
That finding supports the practical idea that many neoantigen targets can remain relevant even after months of treatment.
The vaccine was followed by delayed PD-1 blockade
NeoVax was combined with delayed administration of the PD-1 inhibitor nivolumab.
PD-1 is an immune checkpoint that can restrain activated T cells. Blocking the pathway can help restore antitumor T-cell activity, but checkpoint inhibitors alone have historically had limited effectiveness in many ovarian cancers.
The rationale for the combination is that the vaccine first expands T cells capable of recognizing patient-specific tumor antigens, while PD-1 blockade may then help those T cells remain active against the tumor.
This general strategy of pairing vaccination with checkpoint inhibition is also being explored in other cancers. Bites of Bio has covered how immune stimulation may interact with checkpoint immunotherapy and why maintaining functional T cells matters for cancer treatment.
Six of 10 patients showed biomarker evidence of biologic activity
Across the 10 vaccinated patients, biomarker analyses were suggestive of biologic activity in six.
Those six patients also showed broader tumor- and vaccine-specific immune responses in blood and tumor samples after vaccination.
The wording matters. The study did not report that six patients experienced a conventional radiographic tumor response caused by the vaccine.
Instead, the investigators used integrated immune and molecular measurements to identify evidence that the vaccine was exerting biological pressure on the cancer.
That distinction is especially important in a small phase 1 trial in which patients had already received standard chemotherapy before vaccination.
The immune system appeared to recognize multiple vaccine targets
One advantage of personalized neoantigen vaccines is that they can target many mutations simultaneously.
A tumor may escape an immune response directed at a single antigen by losing or altering that target. A broader T-cell response makes that route of escape more difficult.
The Nature Medicine study found that stronger biologic activity was associated with broader immune recognition of both vaccine-selected neoantigens and other tumor-associated targets.
This supports the idea that the breadth of the immune response may be as important as its intensity.
Cell-free DNA revealed how resistant tumor clones escaped
One of the most interesting parts of the study came from analyzing tumor evolution during and after vaccination.
The investigators used cell-free DNA and tumor sequencing to follow individual cancer clones over time.
Resistant clones showed diverse changes affecting antigen presentation, the cellular machinery that displays protein fragments to T cells.
If a tumor cell stops presenting a neoantigen effectively, vaccine-primed T cells may no longer be able to recognize it even if the underlying mutation remains present.
The authors interpreted these changes as evidence of immune-selective pressure: the vaccine may have made certain antigen-presenting tumor cells less fit, leaving behind clones better able to hide from the immune response.
Antigen presentation may determine who benefits
The study’s central biological conclusion was that vaccine efficacy appears to depend on two interacting factors.
First, the tumor must continue to present enough of the selected neoantigens. Second, vaccination must generate a sufficiently broad immune response against those targets.
A strong vaccine-induced T-cell response may have little effect if the tumor no longer displays the relevant antigen. Conversely, a tumor that retains good antigen presentation may still escape if the immune response is too narrow.
This interplay may help explain why personalized vaccines work better in some patients than others.
The trial was designed for feasibility and safety, not survival
The biggest limitation is the study design.
Only 10 patients received the vaccine, there was no randomized control group, and the trial was not powered to determine whether NeoVax improves progression-free survival or overall survival.
Because patients also received platinum-based chemotherapy and nivolumab, any clinical outcome cannot be attributed to the vaccine alone.
The study therefore should not be interpreted as evidence that a personalized ovarian cancer vaccine has been proven to prevent recurrence.
No unexpected safety signals were reported
The authors reported no unexpected safety signals, and no vaccinated patient discontinued treatment because of toxicity.
That is reassuring for an early-stage trial, but the sample was too small to detect uncommon adverse events.
Safety will need to be evaluated in larger studies, especially when personalized vaccines are combined with checkpoint inhibitors that can cause immune-related toxicities.
Personalization creates a manufacturing challenge
NeoVax is not an off-the-shelf vaccine.
Each patient’s tumor must be sequenced, candidate neoantigens identified, peptides selected and manufactured, and quality-control testing completed before treatment.
That complexity takes time and makes personalized vaccines more difficult to deliver than conventional drugs.
Dana-Farber has previously estimated that designing and manufacturing a personalized neoantigen vaccine can take roughly six to 12 weeks, depending on the platform and study.
The feasibility difference between the two ovarian cancer cohorts illustrates why manufacturing logistics are not a minor issue. They are part of whether the treatment can work in real clinical practice.
The results fit a broader push toward individualized cancer immunotherapy
Personalized cancer vaccines are increasingly being tested alongside checkpoint inhibitors and other immune therapies.
The basic strategy is to generate a large pool of tumor-specific T cells and then help those cells remain functional inside an immunosuppressive tumor environment.
That challenge overlaps with other research Bites of Bio has covered, including how MEK signaling may contribute to T-cell exhaustion in cancer and how scientists are trying to make antitumor T cells more durable.
The ovarian cancer study adds another piece: even when the immune system is successfully trained, the tumor can escape by changing which antigens it presents.
The next question is whether the immune signal translates into longer disease control
The phase 1 results establish that personalized vaccination is feasible in at least a subset of patients with ovarian cancer, can generate tumor-specific immunity and can impose measurable selective pressure on tumor clones.
What they do not establish is whether those immune effects translate into fewer recurrences or longer survival.
Answering that question will require larger prospective studies with clinical efficacy endpoints and an appropriate comparison group.
For now, the strongest conclusion is narrower but still important: a personalized neoantigen vaccine can provoke measurable antitumor immunity in a cancer type that has historically been difficult to treat with immunotherapy, and the study reveals specific mechanisms tumors may use to escape that pressure.
Funding and disclosures
The study was supported by the National Cancer Institute, the Dana-Farber/Harvard Cancer Center Ovarian Cancer SPORE, the Mark Foundation, the Breast Cancer Research Fund, Friends of Dana-Farber and several philanthropic funds and foundations. Bristol Myers Squibb supplied nivolumab.
The paper reports multiple author relationships with biotechnology and pharmaceutical companies, including equity, advisory roles, licensed patents and research funding. These disclosures are detailed in the Nature Medicine article.
References
- Sellars MC, Frank JA, Shim B, et al. Personalized neoantigen-targeting cancer vaccines in ovarian cancer: a phase 1 trial. Nature Medicine. 2026. DOI: 10.1038/s41591-026-04717-8.
- Dana-Farber Cancer Institute. Open-label, Phase I study of NeoVax with Nivolumab in Patients with Ovarian Cancer. Clinical trial NCT04024878. Trial information.
- Dana-Farber Cancer Institute. NeoVax Cancer Vaccine. NeoVax research overview.