Construction of dysregulated long non-coding RNA-associated competing endogenous RNA network in uterine corpus endometrial carcinoma

Background: Competing endogenous RNAs (ceRNAs) render the functions of long non-coding RNAs (lncRNAs) more complicated during cancer processes. Potential lncRNA biomarkers as ceRNAs have not been clearly described for uterine corpus endometrial carcinoma (UCEC). In this research, we researched the functions and regulatory mechanisms of lncRNAs as ceRNAs in UCEC, and their potential applications in prognosis. Methods: The lncRNAs, mRNAs, and miRNAs expression profiles including 552 UCEC tissues and 35 non-tumor tissues were downloaded from The Cancer Genome Atlas (TCGA) portal. Differentially expressed mRNAs, miRNAs and lncRNAs were confirmed by R/Bioconductor package of edgeR (vertical bar log(2)FC vertical bar > 2, FDR <0.01). GO enrichment and KEGG pathway enrichment analyses were accomplished utilizing DAVID online tool. The bioinformatics generated from miRcode and miRTarBase was used to construct the dysregulated lncRNA-associated ceRNA network. Kaplan-Meier curve analysis was used to predict the survival analysis of DERNAs. Results: A total of 1,102 lncRNAs, 2,612 mRNAs and 189 miRNAs were detected to be dysregulated in UCEC. The newly identified ceRNA network includes 27 UCEC-specific miRNAs, 90 lncRNAs, and 74 mRNAs. Eleven mRNAs, 3 miRNAs (has-mir-425, has-mir-211 and has-mir-301b) and 6 lncRNAs (AC11049.1, ADARB2-AS1, C10orf91, GLIS3-AS1, LINC00237 and LINC00261) were found to be significantly correlated with overall survival in UCEC (P value <0.05). Conclusions: In our research process, we successfully constructed a lncRNA-associated ceRNA network which will provide a novel perspective for improving the understanding of UCEC. This study also assists in the identification of new potential biomarkers to be used as candidate prognostic biomarkers or potential therapeutic targets.