Abstract With abundant pro-tumorigenic myeloid cells and few tumoricidal tumor-infiltrating lymphocytes (<5%), GBM is representative of “immune cold” tumors. As such, many different types of immunotherapies have failed to show significant benefits for most glioma patients. Hence, a better understanding of drivers of the immune suppressive microenvironment in GBM and other immune cold tumors is urgently needed to guide future immunotherapy development and application. We recently analyzed 201,986 human glioma and immune cells from 44 tissue fragments from 18 human glioma patients, and present a comprehensive and high-resolution cellular, molecular, and spatial heterogeneity atlas of human glioma. We report an extensive spatial and molecular heterogeneity of glioma and immune cells within the same patient. In addition, we discovered that cell:cell communication between glioma:myeloid cells is considerably more robust than glioma:T-cells, indicating that myeloid cells form a communication hub in vivo. To gain a deeper understanding of these important immune cells, we analyzed 83,479 glioma-infiltrating myeloid cells and identified 9 molecularly distinct myeloid subtypes: 3 microglia subtypes, 3 bone marrow-derived macrophage (BMDM) subtypes, MDSCs, neutrophils, and dendritic cells. Notably, we found that five of these myeloid cell subtype gene signatures are significant predictors of glioma patient survival, independent of glioma cell mutational profiles or gene expression patterns. Leveraging our dataset, we also identified a novel immunotherapy target that is highly expressed in immune-suppressive macrophages and T cells but not in anti-tumor leukocytes: S100A4. We provide both in vitro and in vivo evidence that S100a4 deletion in stromal cells is sufficient to reprogram the immune microenvironment and significantly extend the survival of two independent glioma models. To broaden the potential impact of targeting S100A4 as a selective modulator of immune suppressive leukocytes, we compared the molecular signatures of glioma-associated myeloid cells to those from 12 other cancer types and peripheral blood myeloid cells. We found that S100A4 expression pattern is highly consistent among all tumor types, where its expression is highest in the monocytes and MDSCs and low in most DCs and tissue-resident macrophages. Our preliminary analysis also shows that myeloid cells in gliomas are molecularly distinct from corresponding cell types in other cancers, strongly indicating the role brain microenvironment in influencing the infiltrating BMDM maturation and polarization. Citation Format: Nourhan Abdelfattah, Parveen Kumar, Caiyi Wang, Jia-Shiun Leu, David Baskin, William Flynn, Ruli Gao, Kumar Pichumani, Omkar Ijare, Stephanie Wood, Suzanne Powell, David Haviland, Brittany Parker Kerrigan, Frederick Lang, Sujit Prabhu, Kristin Huntoon, Wen Jiang, Betty Kim, Joshy George, Kyuson Yun. Pan-cancer myeloid cell analysis at the single cell level reveals the influence of distinct organ sites in myeloid cell phenotypes and support targeting S100A4 to reverse immune suppression. [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 5871.
Read full abstract