Neoantigen Peptide Vaccines: Personalized Cancer Immunotherapy Comes of Age
Executive Summary
Neoantigen peptide vaccines — individualized cancer vaccines built from tumor-specific mutant peptides — have progressed from a scientific curiosity to a clinical reality in under a decade. In 2025, Moderna and Merck reported that their mRNA-4157 neoantigen vaccine combined with pembrolizumab reduced the risk of recurrence or death by 49% in high-risk melanoma. Peptide-based neoantigen vaccines, while lagging mRNA approaches in speed, offer advantages in manufacturing simplicity, regulatory precedent, and cost that position them for a distinct and potentially larger role in the immunotherapy landscape.
How Neoantigen Vaccines Work
The premise is elegant: sequence a patient’s tumor DNA, identify mutations that generate novel peptide sequences not present in normal tissue (neoantigens), synthesize peptides corresponding to those mutations, and administer them to train the patient’s immune system to recognize and destroy tumor cells bearing those mutations.
The workflow has four stages: tumor biopsy and sequencing (1–2 weeks) → bioinformatic neoantigen prediction — identifying which mutations generate peptides that bind the patient’s MHC molecules and are likely to be immunogenic (1 week) → peptide synthesis — typically 10–20 peptides, 15–30 amino acids each, synthesized by SPPS (2–4 weeks) → formulation and administration with an immune adjuvant such as poly-ICLC (1 week). Total turnaround time from biopsy to first dose is 5–8 weeks — longer than mRNA vaccines (4–6 weeks) but well within the window for adjuvant therapy in most solid tumor settings.
Clinical Evidence: 2020–2026
Three randomized trials have established proof-of-concept for neoantigen peptide vaccines:
NeoVax (Dana-Farber/Broad Institute). A Phase I trial of a personalized neoantigen peptide vaccine (up to 20 long peptides with poly-ICLC) in 8 patients with high-risk melanoma, published in Nature (2017), showed that all patients generated neoantigen-specific CD4+ and CD8+ T-cell responses. At 4-year follow-up (2021), 6 of 8 patients remained disease-free — a result that launched the field.
NEO-PV-01 (Neon Therapeutics/BioNTech). A Phase Ib trial combining a neoantigen peptide vaccine with nivolumab in advanced melanoma, NSCLC, and bladder cancer, reported in Cell (2020), demonstrated that vaccination induced de novo neoantigen-specific T-cell responses in all three tumor types. The epitope spreading observed — T-cell responses against neoantigens not included in the vaccine — suggested that peptide vaccination can broaden the anti-tumor immune response beyond the initially targeted mutations.
EVX-01 (Evaxion Biotech). A Phase IIa trial combining an AI-designed neoantigen peptide vaccine with pembrolizumab in metastatic melanoma reported a 67% objective response rate at ASCO 2025, compared to approximately 40% for pembrolizumab alone in historical controls. The trial was small (n=28) but provided the strongest efficacy signal to date for a peptide-based neoantigen approach.
Peptide vs. mRNA: A Strategic Comparison
| Parameter | Peptide Vaccines | mRNA Vaccines |
|---|---|---|
| Turnaround time | 5–8 weeks | 4–6 weeks |
| Peptides per vaccine | 10–20 | Up to 34 |
| Manufacturing | SPPS (well-established) | In vitro transcription (newer) |
| Cold chain | Lyophilized; ambient | −20°C to −80°C |
| Regulatory precedent | Extensive (peptide drugs) | Limited (COVID vaccines) |
| Cost per patient | $20,000–50,000 | $50,000–100,000+ |
The comparison reveals complementary strengths. mRNA vaccines offer faster turnaround and the ability to encode more neoantigens, but they require complex cold-chain logistics and carry higher manufacturing costs. Peptide vaccines are slower but benefit from a century of peptide chemistry experience, simpler regulatory pathways, and lyophilized formulations that can be shipped and stored at ambient temperature — a critical advantage for global deployment outside major academic medical centers.
Expert Insight: The Bioinformatic Bottleneck
The single greatest challenge in neoantigen vaccine development is not peptide synthesis or clinical trial design — it is neoantigen prediction. Current algorithms identify 100–500 candidate neoantigens per patient, but only 1–5% of predicted neoantigens are actually immunogenic. Every false-positive neoantigen included in a vaccine competes for immune attention with true immunogenic neoantigens, potentially diluting the therapeutic effect.
The prediction problem has three dimensions: MHC binding prediction (how well does the mutant peptide bind the patient’s MHC molecules?), TCR recognition prediction (will the patient’s T-cell repertoire recognize the MHC-peptide complex?), and tumor presentation prediction (is the mutant peptide actually presented on the tumor cell surface at sufficient density?). Current algorithms handle MHC binding well (AUC > 0.9) but perform poorly on TCR recognition (AUC ~ 0.6–0.7) and tumor presentation (AUC ~ 0.5–0.6). Companies that crack the TCR recognition problem — likely through AI trained on paired TCR–peptide–MHC data — will define the next generation of neoantigen vaccines.
Frequently Asked Questions
Are neoantigen vaccines curative?
Not yet. The most promising results — Moderna/Merck’s mRNA-4157 and Evaxion’s EVX-01 — show significant improvements in recurrence-free survival and response rates, but they are used as adjuvant or combination therapy alongside checkpoint inhibitors, not as standalone curative treatments. The field’s ambition is to achieve durable complete responses in metastatic disease, but this has not been demonstrated in randomized trials as of mid-2026.
Which tumor types are most amenable to neoantigen vaccination?
Tumors with high tumor mutational burden (TMB) — melanoma, non-small cell lung cancer, bladder cancer, and microsatellite-instable colorectal cancer — generate the most neoantigens and have shown the strongest responses to neoantigen vaccines. Tumors with low TMB (pancreatic, prostate, glioblastoma) present fewer neoantigen targets and have shown limited responses. Expanding neoantigen vaccines to low-TMB tumors will require better neoantigen prediction algorithms and combination strategies that enhance T-cell infiltration into immunologically “cold” tumors.
What is the regulatory pathway for personalized vaccines?
The FDA has indicated that personalized neoantigen vaccines will be regulated under a platform approach: the manufacturing process and bioinformatic pipeline are validated once, and individual patient-specific vaccines are released under that validated platform. This is analogous to the regulatory framework for CAR-T cell therapies, which are also patient-specific products manufactured under a common platform. The FDA’s 2025 guidance on ” platform-based drug development” provides a framework for this approach, though no neoantigen vaccine has yet completed the platform validation process.
Further Reading
- Peptide-Drug Conjugates — another peptide-based approach to oncology
- AI-Designed Peptides — the algorithms that could improve neoantigen prediction
Last reviewed: June 2026. Peptide Proof Editorial Team.