TMIC-04. GLIOBLASTOMA-ASSOCIATED MYELOID CELLS DISPLAY NONPOLARIZED M0 MACROPHAGE PHENOTYPE

Abstract

INTRODUCTION: The glioblastoma (GBM) microenvironment is commonly infiltrated with innate immune cells including microglia, macrophages, and myeloid-derived suppressor cells (MDSCs). Presumably these cells assume a tumor-supportive alternatively activated M2 phenotype, but no comprehensive phenotypic and/or genotypic studies have been conducted yet. METHOD(S): Flow cytometry analysis was utilized to determine the expression of M1 and M2 markers in the GBM-associated myeloid cells (GAMs). Whole-genome microarray analysis was performed using RNA isolated from GAMs and matched GBM-patient and healthy-donor blood myeloid cells. Nanostring Technology was used to profile immune system- and cancer-related genes in GAMs, matched GBM-patient blood myeloid cells, healthy-donor blood myeloid cells, normal human microglia, and nonpolarized M0 and M1-, M2a-, M2cpolarized macrophages. Gene Set Enrichment Analysis (GSEA) was used to identify hallmarks of biological states or processes in immune cells derived from GBM-patient blood and tissue compared to healthy donors. RESULT(S): Among CD11b+ cells, microglia and MDSCs constituted a higher percentage of GAMs compared to macrophages. In contrast, microglia were the most common CD11b+cells isolated from nonmalignant surgical brain samples, with macrophages constituting the remaining contributor to the brain parenchyma. Classical GBMs had significantly higher number of MDSCs than macrophages, while the mesenchymal subtype had more microglia than MDSCs and macrophages. The number of GAMs positively corresponded with the overall tumor size (precisely with edema) but not with overall patient survival. Flow cytometry studies and gene expression profile analysis revealed that GAMs express both anti-tumor M1 and tumor-supportive M2 markers. GSEA identified biological processes in these cells relative to matched blood cells, including signaling by KRAS, TGF-beta, and TNF-alpha, epithelial-mesenchymal transition, angiogenesis, and hypoxia. Unsupervised analysis of gene expression profiles showed that GAMs aligned closely with nonpolarized M0 macrophages. CONCLUSION(S): New immunotherapeutic strategies for GBM are needed to redirect nonpolarized M0 GAMs towards the M1 phenotype.