A consortium approach identified key parameters of tumor epitope immunogenicity, filtering out 98% of non-immunogenic peptides while maintaining precision above 0.70.
98%of non-immunogenic peptides filtered out by the prediction model
What the researchers found
An integrated model combining peptide presentation and recognition features filtered out 98% of non-immunogenic peptides with precision above 0.70, and was validated in an independent cohort of 310 epitopes.
Why it matters
Personalized cancer vaccines depend on correctly identifying which tumor mutations will provoke an immune response. This consortium-derived model dramatically improves neoantigen prediction, potentially making immunotherapy more targeted and effective.
The numbers in context
608 epitopes assessed; ~8% confirmed immunogenic; peptide-MHC stability identified as top predictor.
How the study worked
Global consortium where participants predicted immunogenic epitopes from shared tumor sequencing data. 608 epitopes were assessed for T cell binding in patient-matched samples, with validation in an independent 310-epitope cohort.
Who was studied
Tumor-derived epitopes from shared sequencing data assessed by global consortium
What this study cannot tell us
The model was developed and validated using specific tumor types and HLA alleles, which may limit generalizability. T cell binding assays measure recognition potential but don't guarantee clinical anti-tumor responses.
How to read the evidence
Strong in vitro evidence with multi-center consortium validation in an independent cohort, though clinical translation remains to be demonstrated.
When this study was published
Published in 2020. Neoantigen prediction has continued to advance, but this consortium dataset remains a key reference resource.
The bigger picture
As cancer immunotherapy moves toward personalized neoantigen vaccines, the ability to accurately predict which tumor peptides will trigger immune responses is foundational. This work provides both a predictive framework and an open data resource that could accelerate vaccine development across cancer types.
Questions still open
- How well does this model perform across diverse HLA types and tumor mutational burdens?
- Can this predictive framework be integrated into clinical trial design for neoantigen vaccines?
- What additional features beyond presentation and recognition could further improve prediction accuracy?
Common questions
What is a neoantigen?
Why is predicting immunogenic epitopes so difficult?
Read the original research
Key Parameters of Tumor Epitope Immunogenicity Revealed Through a Consortium Approach Improve Neoantigen Prediction.
Cell, 183(3), 818-834.e13
Citation
Wells, Daniel K; van Buuren, Marit M; Dang, Kristen K; Hubbard-Lucey, Vanessa M; Sheehan, Kathleen C F; Campbell, Katie M; Lamb, Andrew; Ward, Jeffrey P; Sidney, John; Blazquez, Ana B; Rech, Andrew J; Zaretsky, Jesse M; Comin-Anduix, Begonya; Ng, Alphonsus H C; Chour, William; Yu, Thomas V; Rizvi, Hira; Chen, Jia M; Manning, Patrice; Steiner, Gabriela M; Doan, Xengie C; Merghoub, Taha; Guinney, Justin; Kolom, Adam; Selinsky, Cheryl; Ribas, Antoni; Hellmann, Matthew D; Hacohen, Nir; Sette, Alessandro; Heath, James R; Bhardwaj, Nina; Ramsdell, Fred; Schreiber, Robert D; Schumacher, Ton N; Kvistborg, Pia; Defranoux, Nadine A. (2020). Key Parameters of Tumor Epitope Immunogenicity Revealed Through a Consortium Approach Improve Neoantigen Prediction.. Cell, 183(3), 818-834.e13. https://doi.org/10.1016/j.cell.2020.09.015