Abstract 4989: T-distributed stochastic neighbor embedding (t-SNE) analysis of tumor infiltrating lymphocytes after treatment with a T cell activating therapy identifies a unique population of recruited CD8+ T cells and novel options for combination immunotherapy

Abstract

Abstract Immune therapies for cancer seek to alter the immune profile of the tumor microenvironment (TME). A suppressive TME can counteract active immune responses by CD8+ T cells by multiple mechanisms, including induction of checkpoint inhibitor (CPI) expression. DPX™ is a lipid nanoparticle-based platform that can be formulated with cancer targets to induce potent and sustained antigen-specific T cell responses. In preclinical and clinical testing previously published, combination of DPX with low dose intermittent cyclophosphamide (ldCPA) resulted in increased tumor infiltration of antigen-specific CD8+ T cells. These T cells express PD-1, and combination treatment with PD-1 blocking antibody was shown previously to result in better tumor control in preclinical tumor models. In this study, we used tSNE analysis to understand changes to the immune profile induced by DPX/ldCPA treatment in the C3 model. We prioritized the expression of CPIs on tumor infiltrating CD8+T cells to identify potential combinations with blocking monoclonal antibodies to enhance tumor CD8+ T cell activity. Mice (C57BL/6) bearing HPV16 E7-transformed C3 tumor cells were treated with ldCPA (20 mg/kg/day PO, 7 days), followed by subcutaneous injection with HPV16 E749-57 peptide in a DPX formulation (DPX-R9F). Tumors were collected for analysis 10 days after DPX-R9F treatment and a single cell suspension prepared by digestion. Immune profiling was performed by 12-color flow cytometry, to observe differences in infiltrating immune cells and CPIs. tSNE analysis of flow cytometry data was performed to analyze differences in cellular infiltrates and phenotype with treatment. We found CD8+ T cells that infiltrated tumors in mice treated with DPX-R9F/ldCPA expressed most CPIs tested (including PD-1, Tim-3, Lag-3, ICOS), and fewer expressed CTLA-4. Treatment induced infiltration of a unique population of CD8+ T cells that was PD-1+Tim-3+CTLA-4-, whereas CD8+ T cells present in untreated tumors were primarily PD-1+Tim-3+CTLA-4+. Tumor challenge was performed to evaluate combination of DPX-R9F/ldCPA treatment with anti-CTLA-4 (Clone 9D9) to potentially impact these resident CD8+ T cells. Mice treated with the triple combination had better control of C3 tumor growth compared to mice treated with DPX-R9F/ldCPA or anti-CTLA-4 separately. With tSNE analysis we were able to have a better model to observe differences in the CD8 population with DPX/ldCPA treatment and determine the mechanism for how certain checkpoint targets are impacting tumor infiltrating cells to further sculpt the anti-tumor response to increase efficacy of treatment. Citation Format: Ava Vila-Leahey, Alecia MacKay, Genevieve Weir, Marianne Stanford. T-distributed stochastic neighbor embedding (t-SNE) analysis of tumor infiltrating lymphocytes after treatment with a T cell activating therapy identifies a unique population of recruited CD8+ T cells and novel options for combination immunotherapy [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 4989.