Many attributes of the tumor microenvironment, such as the level of CD8 T-cells in the tumor, higher levels of pro-inflammatory cytokine network dominated by interferon signaling, antigen processing and presentation correlate with superior efficacy of checkpoint control inhibitors. Tumors lacking CD8 T-cells are less responsive to checkpoint control blockade, and therefore other therapeutic modalities for treating these tumors need to be explored. In this study, we set out to identify a set of core pathways associated with the absence or presence of specific immune cell types in tumors. These core pathways can be modulated to alter the immune profile of these unresponsive tumors and sensitize them to checkpoint control blockade. To this end, we created a robust bioinformatic solution OncoPeptTUME, to systematically investigate the immune landscape of tumors from RNA-seq data, using a set of proprietary immune cell type-specific gene expression signatures. Using a scoring method derived from Single Cell Gene Set Enrichment Analysis (ssGSEA), we quantitated the relative abundance of different immune cell types present in 9345 tumors across 33 cancers available in the TCGA dataset. We first validated our approach by selecting high and low CD8 T-cell containing tumors across many different cancers. Differential gene expression analysis between these two sets of tumors identified upregulated genes in PD-1 signaling, IFN-α, β and γ pathways, TCR signaling, antigen processing and presentation as previously reported in multiple studies. Similarly, tumors with high infiltration of myeloid derived suppressor cells (MDSCs) showed high level expression of a large number of inhibitory receptors associated with innate immune cells, demonstrating potential mechanism of immunesuppression in these tumors. Few studies, however, have investigated tumor intrinsic and extrinsic factors that favor infiltration of macrophages and induce them to differentiate into M1 and M2 functional states. The M1 and M2 macrophages regulate the inflammatory state of the tumor microenvironment by producing cytokines, chemokines and growth factors thereby making them susceptible or resistant to immune-mediated elimination. We applied M1 and M2-specific gene expression signatures on CD8 T-cell depleted tumors and identified 236 tumors having high or low M1 or M2 macrophages. Differential gene expression analysis reveal that tumors containing high M1 macrophages have significant upregulation of IFN-α/β signaling, tryptophan catabolism, IL1-β processing and CASP1 inflammosome activity. By contrast, tumors with low M1 infiltrated macrophages have upregulation of FGFR1c-Klotho pathway genes. Since differentiation of macrophages into M1 or M2 occurs in the tissue microenvironment, FGFR1 signaling may regulate macrophage phenotype rather than their migration into the tumor. In support of this hypothesis, we detected higher level of M2 macrophages in tumors that are depleted of M1 macrophages. We also analyzed genetic alterations in tumor cells that favor higher levels of M2 macrophages by examining non-synonymous somatic mutations in the coding sequences of genes. We observed loss- of-function mutations in p53 gene across many different cancers (breast, glioblastoma, stomach adenocarcinoma and lung squamous cell carcinoma) showing higher burden of M2 macrophages compared to the M1 type. Our analysis demonstrate that combining expression signatures with tumor mutanome analysis can provide a powerful tool to assess the tumor microenvironment and identify pathways that promote, or exclude infiltration/differentiation of specific immune cells.

Citation Format: Nitin Mandloi, Ashwini Patil, Rekha Sathian, Aparna Mohan, Malini Manoharan, Ravi Gupta, Hiranjith Govindamangalam, Amit Chaudhuri. Differential gene expression and tumor mutanome analysis reveal significantly enriched pathways associated with higher tumor burden of M1 and M2 macrophages. [abstract]. In: Proceedings of the AACR Special Conference on Tumor Immunology and Immunotherapy; 2016 Oct 20-23; Boston, MA. Philadelphia (PA): AACR; Cancer Immunol Res 2017;5(3 Suppl):Abstract nr A17.