Abstract 6173: Single cell network-based analyses defines the effect of radiation on the tumor immune microenvironment, including the response by and molecular determinants of suppressive myeloid cells

Abstract

Abstract Background: More than half of all cancer patients undergo radiation therapy (RT), which can be curative or palliative to improve cancer-specific outcomes and quality of life. However, radioresistance emerges and a cure is not always achieved. To date, researchers have studied T-cells and lymphocytes to understand response to RT, but little is known about myeloid cells. This diverse group of cells helps mount immunological responses, but can also be immunosuppressive by lessening the strength of T-cells. We hypothesized that our network-based analyses of single cell proteogenomic data generated from tumor-infiltrating immune cells can identify druggable proteins in pathways driving unwanted immunosuppressive changes, particularly among myeloid cells. Methods: Proteogenomic (CITE-Seq) analysis was conducted on CD45+ immune cells from unirradiated and irradiated orthotopic 4T1 murine mammary tumors three- and ten-days after tumor irradiation. Based on select proteins’ surface expression, CITE-Seq generated eight distinct clusters, including three myeloid cell clusters — monocytes, macrophages, and granulocytes. We used VIPER (virtual inference of protein activity by enriched regulon analysis), a machine learning-based algorithm that infers protein activities by taking regulatory gene expression networks and calculating the weighted expression on each protein’s targets, to create functional subclusters of the eight immune-cell populations. For subclusters enriched by RT, we used OncoTreat to identify active, druggable proteins. Results: Lymphocytes were depleted with radiation, especially ten days post-irradiation. While monocytes appeared homogeneous, one of three subclusters of macrophages appeared more resistant (i.e. enriched with radiation) and had 4 druggable targets (PSMA4, HSP90AB1, PARP1, APEX1) from OncoTreat. Similarly, three of eight subclusters of granulocytes appeared resistant, with two potential targets (PSMA4, S100A9). Conclusion: Suppressive populations were identified among the immune cells, specifically myeloid cells. We plan to experimentally test whether drugging the aforementioned targets in macrophages and granulocytes improves the overall efficacy of RT by augmenting an anti-tumor immune response. Affiliations: Aparna Krishnan1, Aleksandar Obradovic1, Catherine Spina2, Andrea Califano1,3,4,5,6 1 Department of Systems Biology* 2 Department of Radiation Oncology* 3 Herbert Irving Comprehensive Cancer Center* 4 Department of Medicine* 5 Department of Biochemistry & Molecular Biophysics* 6 Department of Biomedical Informatics* * Columbia University Irving Medical Center, New York, NY 10032, USA Citation Format: Aparna Krishnan, Aleksander Obradovic, Catherine Spina, Andrea Califano. Single cell network-based analyses defines the effect of radiation on the tumor immune microenvironment, including the response by and molecular determinants of suppressive myeloid cells [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 6173.