LAUSR

The treatment of glioblastoma remains a challenge for modern therapy. Checkpoint blockade inhibitors (CBI), designed to block inhibitory T-cell signaling, represent attractive potential therapeutic interventions. However, glioblastoma response rates to CBI remain low. To this end, preclinical models are critical to studying multi-modal immune interventions to overcome CBI resistance. Therefore, we set out to identify endogenous neoantigens in a CBI resistant murine glioblastoma and assess the efficacy of neoantigen vaccination in combination with CBI treatment. Whole exome DNA and RNA sequencing was used to identify expressed, missense mutations in the C57BL/6 derived murine glioblastoma, CT2A. The pVAC-seq software suite was used to identify candidate neoantigens predicted to bind H2-Kb and H2-Db molecules. CD8 + T cells isolated from CT2A tumor-infiltrating lymphocytes (TIL) were screened for neoantigen reactivity by IFN-gamma enzyme linked immunospot assays. Survival analysis was performed on intracranial tumor bearing mice treated with neoantigen vaccination, anti-PD-L1, or combination therapy. In silico analysis identified 649 CT2A-derived candidate neoantigens predicted to bind H2-Kb or H2-Db molecules. Of the 40 top-ranking neoantigen candidates, 16 elicited CD8 + TIL responses in mice vaccinated with cognate peptides. Assessing for endogenous reactivity, we identified neoantigen specific CD8 + T cell responses in the intracranial TIL and draining lymph nodes to two H2-Kb restricted, Epb4 (H471L) and Pomgnt1 (R497L), and one H2-Db restricted neoantigen, Plin2 (G332R). Survival studies showed that therapeutic neoantigen vaccination with Epb4, Pomgnt1, and Plin2, in combination with anti-PD-L1 treatment was superior to anti-PD-L1 alone. We identified endogenous neoantigen specific CD8 + T cells within brain tumors and draining lymph nodes of a CBI resistant murine glioblastoma. Furthermore, we find that neoantigen vaccination significantly augments survival benefit in combination with anti-PD-L1 treatment. These observations provide important preclinical correlates for glioblastoma immunotherapy trials and support further investigation into the effects of multi-modal immunotherapeutic interventions on anti-glioma immunity.