Immunotherapy clinical trials have not shown efficacy in glioblastoma (GBM), arguably the deadliest of all cancers. A microenvironment, characterized by a paucity of T cells recognizing tumor peptides displayed on major histocompatibility complexes of tumor cells is one of the main reasons for GBM’s evasion. Virotherapy provides the opportunity to reshape the GBM TME towards pro-inflammatory phenotype. Data from a recent clinical trial with herpes simplex-1 based oncolytic virus (oHSV) have shown that the administration of these agents leads to a rapid influx of CD8 T cells, and activation of myeloid cells, suggesting that oHSV may offer an in-situ vaccination approach to increase availability of actionable antigens. C57BL/6-derived GBM models do not mimic the outcome observed with patient derived tumors in terms of virus replication. To decipher the impact of oHSV on tumor antigen specific responses, better immunocompetent models are needed. To assess whether oHSV treatment modulates the immunopeptidome profile (IP) of tumor, we used two transplantable GEMM-derived GBM models, 1620 and 1694. Both have wild type human EGFR, loss of Pten and Cdkn2a in BALB/c background. These models were more receptive to infection than CT2A. In vivo, virus persisted in tumors for up to 6 days, but not for 4 days in CT2A. IP analyses revealed that oHSV promotes epitope presentation on tumor cells and in TME. Moreover, low dose of virus efficiently induced antigen availability in both models though it differentially modulated the MHC ligandome. Using two new GBM models, we showed that an oncolytic virus used in clinical trial with recurrent GBM (NCT03152318) modulates IP of tumors, and low dose of virus is sufficient to induce the presentation of tumor epitopes. This project is supported by Bridge Fund given to Drs. Chiocca and White and P01 CA236749 to Dr. Chiocca.

Abstract Despite the success in treating many solid cancers, immunotherapy has not shown significant efficacy in patients with glioblastoma (GBM). The lack of tumor-infiltrating T cells recognizing tumor antigens displayed with major histocompability complex I (MHC I) is one of the reasons for GBM’s evasion from immunotherapy. Our preclinical GBM models and an ongoing clinical trial with oncolytic herpes simplex virus-1 (oHSV) at our institution (NCT03152318) demonstrated that oHSV treatment leads to an increase influx of CD8+ and CD4+ T cells into tumor. However, it remained unclear whether these T cells are specific to the tumor antigens (neoantigens) or reactive to only viral antigens. To determine whether infiltrating CD8 T cells target tumor antigens, we first performed whole-exome and RNA sequencing of CT2A, a preclinical model of GBM impervious to immunotherapies for identification of neoantigens. We screened immunogenicity of identified neoantigens by ELISPOT assay. TILs from the oHSV-treated (but not from control group) reacted to two of the neoantigens that were identified with our pipeline and this activation was similar in magnitude to the reaction to the immunodominant HSV antigen. FACS analyses of TILs seven days after virotherapy showed that oHSV-treated tumor had higher level of TILs than vehicle control tumors, a finding that mirrored what we observed in our phase 1 clinical trial with oHSV. High dose of oHSV treatment not just induces the infiltration of viral specific T cells, but infiltration of T cells targeting neoantigens that were not presented without OV infection. Phosphoproteomic analysis of bulk tumor revealed activation of key inflammatory pathways, including that of T cells receptor signaling upon virus treatment. We evaluated the efficacy of therapeutic neoantigen vaccination in combination with oncolytic intratumoral virotherapy. Survival analysis showed that combination therapy not just delays the tumor progression but also provides survival advantage compared to monotherapies.

Background: Patients with malignant brain tumor, glioblastoma (GBM), do not benefit from promising immunotherapies that use checkpoint inhibitors, such as PD-1 and CTLA-4 blockade antibodies. Several reasons may underlie this clinical result, particularly because of immunogenically cold nature of the GBM microenvironment, which is characterized by the low level of infiltrating cytotoxic CD8+ T cells and by an enriched myeloid suppressor cell population. We and other groups hypothesize that oncolytic virus (OVs)-based treatments can overcome this problem. Our preclinical GBM models and an ongoing clinical trial with oncolytic herpes simplex virus-1 (oHSV-1) at our institution (NCT03152318) demonstrated that OV treatment leads to an increase in the influx of CD8+ and CD4+ T cells into TME. However, it remained unknown whether these T cells are specific to the tumor. Here, we evaluated T-cell responses against newly identified neoantigens from murine syngeneic glioma line in response to the intratumoral OV treatments. Results: To evaluate the effect of oHSV-1 treatment on neoantigen-specific responses, we first performed whole-exome and RNA sequencing of murine GBM for the identification of tumor-specific mutations. Putative MHC I antigens were predicted with netMHCpan V4 and immunogenicity of candidates, that were selected based on affinity to MHC I molecules, were checked by IFN-gamma enzyme-linked immunospot (ELISpot) assay. With our workflow, 91 H2-Db-restricted and 77 H2-Kb-restricted peptides were identified. From this group, 8 Db- and 9 Kb-restricted peptides were selected based on their highest MHC affinity. We then asked which of these neoantigens would show enhanced presentation upon oHSV infection in vivo. To determine that we first implanted GBM tumors in the brains of syngeneic mice. At Day7 post-tumor implantation, the oHSV-1 (or vehicle control), was in situ stereotactically administered. Seven days later, splenocytes, PBMCs, and TILs were harvested to be screened. TILs from the oHSV-1 treated mice, but not those from the control group reacted to the putative neoantigens, and this activation was similar in magnitude to the response to the immunodominant HSV peptide. We are currently performing immunopeptidome analyses to confirm the binding of these neoepitopes to MHC upon oHSV treatment and in vivo studies to determine how T cells specific to these MHC-I restricted neoepitopes contribute to the anti-tumor effect. Conclusion: Our data with the preclinical model of GBM show for the first time that infection with an oHSV-1 does indeed lead to novel TIL activation against tumor neoantigens that were not presented without OV infection. Thus, oncolytic viruses may offer an in-situ vaccination approach to stimulate neoantigen-specific immune responses. Citation Format: Raziye Piranlioglu, Junfeng Liu, Anan Islam, Alex Ling, Naoyuki Shono, William F. Goins, Hiroshi Nakashima, E. Antonio Chiocca. The effect of oncolytic virus therapy on neoantigen specific immune responses [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 6388.