Abstract
BACKGROUND: Ewing's sarcoma (ES) is the most common malignant primary bone tumor in children and adolescents. This study is aimed at developing new prognostic markers and building a microRNA-mRNA network in the development of ES. METHOD: GSE80201 and GSE39262 were downloaded from the Gene Expression Omnibus (GEO) database. Bioinformatics analysis was used to download and process data. The coexpression of differentially expressed microRNAs (DEMs) and genes (DEGs) was selected by using R software. The FunRich database was utilized to perform cellular component (CC), molecular function (MF), and biological process (BP) enrichment analysis. Cytoscape and ClueGO were used to perform Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis and construct the mRNA-microRNA network. The Kaplan-Meier Plotter was used to perform prognosis analysis between the expression level of genes we selected and overall survival (OS) of patients with ES. Univariate analysis and multivariate analysis were carried out to research the prognostic value of identified mRNA expression in ES according to TCGA database. RESULTS: By using bioinformatics analysis, 10 DEMs and 5 target mRNAs were identified. Based on the KmPlot software, COL1A2, COL3A1, and TGFBI were significantly related to the OS of patients with ES. High COL3A1 mRNA expression was correlated with distant metastasis, margin status, and poor overall survival of ES. Besides, multivariate analysis indicated that COL3A1 was an independent risk factor for ES patients. CONCLUSIONS: In conclusion, our results suggest that COL3A1 and its related molecules may be a potential diagnostic and prognostic biomarker for patients with ES.