Identification of a survival-related signature for sarcoma patients through integrated transcriptomic and proteomic profiling analyses.

Abstract

Sarcoma (SARC) represents a group of highly histological and molecular heterogeneous rare malignant tumors with poor prognosis. There are few proposed classifiers for predicting patient's outcome. The Cancer Proteome Atlas (TPCA) and The Cancer Genome Atlas (TCGA) databases provide multi-omics datasets that enable a comprehensive investigation for this disease. The proteomic expression profile of SARC patients along with the clinical information was downloaded. 55 proteins were found to be associated with overall survival (OS) of patients using univariate Cox regression analysis. We developed a prognostic risk signature that comprises seven proteins (AMPKALPHA, CHK1, S6, ARID1A, RBM15, ACETYLATUBULINLYS40, and MSH6) with robust predictive performance using multivariate Cox stepwise regression analysis. Additionally, the signature could be an independent prognostic predictor after adjusting for clinicopathological parameters. Patients in high-risk group also have worse progression free intervals (PFI) than that of patients in low-risk group, but not for disease free intervals (DFI). The signature was validated using transcriptomic profile of SARC patients from TCGA. Potential mechanisms between high- and low-risk groups were identified using differentially expressed genes (DEGs) analysis. These DEGs were primarily enriched in RAS and MPAK signaling pathways. The signature protein molecules are candidate biomarkers for SARC, and the analysis of computational biology in tumor infiltrating lymphocytes and immune checkpoint molecules revealed distinctly immune landscapes of high- and low-risk patients. Together, we constructed a prognostic signature for predicting outcomes for SARC integrating proteomic and transcriptomic profiles, this might have value in guiding clinical practice.