Abstract 4434: Comprehensive bioinformatic analysis of immune composition and genes for survival prediction in sarcoma

Abstract

Abstract Introduction Sarcomas, a broad family of mesenchymal malignancies, exhibit remarkable histologic diversity. Previous study has reported that the infiltration of immune and stromal cells in tumor microenvironment contribute significantly to prognosis. ESTIMATE, an algorithm to calculate immune and stromal scores, predicts the infiltration of non-cancer components. The Cancer Genome Atlas (TCGA) database is available to grasp potential correlations between gene set prolife and overall survival of malignancies, including sarcomas. To better understand the proportions of immune cells in the tumor microenvironment, We used CIBERSORT deconvolution software and ssGSEA to infer the relative proportions of several distinct leukocyte cell types in the tumors from microarray gene expression data of sarcoma patients. By taking advantage of both TCGA database of sarcomas cohorts and ESTIMATE algorithm, we extracted a list of genes that predict poor outcomes in sarcoma patients. Finally, we validated these genes in an independent sarcoma cohort from the GSE17679. Thus, we obtained a list of tumor microenvironment-related genes that predict poor outcomes in sarcoma patients. Methods Gene expression profile and clinical data for sarcoma patients was obtained from the TCGA data portal. Immune scores and stromal scores were calculated by applying the ESTIMATE algorithm to the downloaded database. Batch adjusted data was subsequently analyzed using CIBERSORT to resolve the immune composition. For validation, gene expression profiles for Ewing sarcoma patients were obtained from the Gene Expression Omnibus dataset GSE17679. Results The immune and stromal score are associated with prognosis. The type 2 macrophages in sarcoma makes up the largest composition of all immune cells in the tumor microenvironment of sarcoma (except synovial sarcoma), which predicts poor outcomes. From functional enrichment analysis of TCGA database applied by ESTIMATE algorithm-based immune scores, we extracted that NR1H3, VAMP5, GIMAP2, GBP2, HLA-E and CRIP1 are highly expressed in the immune microenvironment, predicting good outcomes in sarcoma patients. Conclusion We extracted a list of tumor microenvironment related genes. These genes were validated in an independent sarcoma cohort and that may represent promising novel signatures for the diagnosis and prognosis prediction of sarcoma. The immune composition analysis could be useful for outlining the prognosis. Citation Format: Hongmin Chen. Comprehensive bioinformatic analysis of immune composition and genes for survival prediction in sarcoma [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr 4434.