Identification of potential prognostic long non-coding RNA signatures based on a competing endogenous RNA network in lung adenocarcinoma

Abstract

A number of experimental and computational studies have demonstrated the key roles of long non-coding RNAs (lncRNAs) acting as competing endogenous RNAs (ceRNAs) in the tumorigenesis of lung adenocarcinoma (LUAC). However, there remains a requirement for prognostic candidate biomarkers acting as ceRNAs for the prediction of overall survival in patients with LUAC. The main goal of the present study was to identify novel lncRNAs associated with LUAC overall survival and assess their prognostic values. The study analyzed coding RNA and ncRNA expression profiles of patients with LUAC by retrieving existing RNA-sequencing datasets from The Cancer Genome Atlas database, and 2,507 differentially expressed mRNAs, 1,633 lncRNAs and 113 miRNAs were screened from patients with LUAC compared with those of adjacent normal samples (P<0.01 and |logFC|>2). Of these LUAC-specific RNAs, 134 lncRNAs, 21 miRNAs and 34 mRNAs were used to build an lncRNA-mRNA-miRNA ceRNA network, among which 8 lncRNAs and 9 mRNAs were associated with overall survival in patients with LUAC by acting as ceRNAs. Next, an lncRNA-based prognostic signature was constructed by risk scoring approach based on the expression levels of 9 prognosis.associated lncRNAs using Cox's regression analysis. Moreover, the prognostic capacity of the 9-lncRNA signature was independent of known clinical prognostic factors. These results provide novel insight into the potential of lncRNA ceRNAs to be candidate biomarkers associated with LUAC overall survival.