Recently, Richard et al. developed a new computational pipeline, Ancer, to predict neoantigens from genomic data derived from the urothelial cancer cohort of The Cancer Genome Atlas (TCGA-BLCA). The study was published in Scientific Reports. Ancer applied three steps 1) predict neoepitopes using EpiMatrix algorithm, 2) compared mutated and matched normal sequences of neoantigen 3) remove self-like neoepitopes using the JanusMatrix algorithm which eliminates putative inhibitory or cross-reactive epitopes.
The investigators found that patients with high Ancer-derived neoepitope burdens were associated with longer overall survival (OS). The difference in median OS observed when stratifying patients using Ancer-derived neoepitope was double that of NetMHCpan-derived neoepitope. The five-year OS prediction accuracy and negative predictive value for Ancer neoepitope burden were higher than NetMHCpan neoepitope burden (65% vs. 61% and 88% vs. 86%, respectively).
In a multivariate analysis model adjusting for TMB, age, and disease stage, Ancer neoepitope burden was a significant predictor of OS (Hazard ratio HR: 0.64 95%CI: 0.41-1.011, p= 0.049). Using same model, the NetMHCpan neoepitope burden was not found to be a significant predictor of OS (0.98, 95%CI: 0.59- 1.63, p= 0.936).
The discordance rate between Ancer and TMB occurred in 13% of the cohort. Thus, it is unclear if an Ancer-derived neoepitope burden would perform better than TMB in predicting response to an immune checkpoint inhibitor. The cutoffs defining high neoepitope burden were derived from the median number of neoepitopes identified in the TCGA BLCA cohort. Further research is needed to improve neoantigen prediction and prospectively test the value of these predictions as biomarkers of response to immunotherapy.
Written by: Bishoy M. Faltas, MD, Director of Bladder Cancer Research, Englander Institute for Precision Medicine, Weill Cornell Medicine, New York City, New York
- Richard G, De Groot AS, Steinberg GD, Garcia TI, Kacew A, Ardito M, et al. Multi-step screening of neoantigens’ HLA- and TCR-interfaces improves prediction of survival. Sci Rep. 2021;11:9983. PMID: 33976291