NeoBridge: Biophysics-informed neoantigen prioritization and clonal architecture profiling in pancreatic cancer.

NeoBridge achieves a 0.82 AUC-ROC in predicting neoantigens in pancreatic ductal adenocarcinoma, surpassing established tools and demonstrating strong cross-cancer generalization.

• Why it matters: Accurate neoantigen identification is crucial for personalized cancer vaccines, especially in low-mutation-burden tumors like PDAC, where current methods face significant challenges.
• What they did: Researchers developed NeoBridge, a hybrid ensemble model trained on 11,188 validated neoantigens, integrating features such as HLA binding, differential agretopicity, neoantigen abundance, and TCR recognition potential.
• The result: Subclonal neoantigen load in 118 PDAC patients correlates with worse progression-free survival (p < 0.001) and overall survival (p = 0.038), suggesting potential prognostic value and guiding future validation efforts.