Depil, S.; Baulu, E.; Bolon, A.; Etchegaray, E.; Raimundo, F.; Merienne, A.; Martin, J.; Grandsire, J.; Richard, T.; Tonon, L.; Dubois, C.; Estornes, Y.; Boulos, R.; Tabone, O.; Bonaventura, P.; Page, A.; Gardet, C.; Alcazer, V.; Hughes, S.; Gillet, B.; Gervois, N.; Labarriere, N.; Wang, Q.; Valladeau-Guilemond, J.; Chuvin, N.; Marcel, V.; Diaz, J.-J.

Cancer cells rely on alternative modes of translation for protein synthesis, promoting internal ribosome entry site (IRES)-dependent translation of mRNA encoding pro-oncogenic factors. Furthermore, ribosomes translate mRNA with lower fidelity in tumor cells. We proposed that these translational modifications in cancer produce shared tumor-specific epitopes derived from IRES-containing oncogenes. To identify such neoepitopes, we developed an in silico-based method that we applied to c-MYC. We showed that the non-canonical translation of c-MYC mRNA in cancer cells, involving a (+1) ribosomal frameshift, generates a shared neoepitope which induces high-avidity T cells able to kill tumor cells in vitro and in vivo while sparing normal cells. Our data provide preclinical rationale for developing immunotherapies targeting this c-MYC-derived neoepitope and validate a new type of shared translation-associated neoantigens.