Identification of prognostic biomarkers in glioblastoma using a long non-coding RNA-mediated, competitive endogenous RNA network.

Yuze Cao,Xia Li,Wenbiao Xiao,Peng Wang,Bo Xiao,Shangwei Ning

doi:10.18632/oncotarget.9569

Abstract

Glioblastoma multiforme (GBM) is a highly malignant brain tumor associated with a poor prognosis. Cross-talk between competitive endogenous RNAs (ceRNAs) plays a critical role in tumor development and physiology. In this study, we present a multi-step computational approach to construct a functional GBM long non-coding RNA (lncRNA)-mediated ceRNA network (LMCN) by integrating genome-wide lncRNA and mRNA expression profiles, miRNA-target interactions, functional analyses, and clinical survival analyses. LncRNAs in the LMCN exhibited specific topological features consistent with a regulatory association with coding mRNAs across GBM pathology. We determined that the lncRNA MCM3AP-AS was involved in RNA processing and cell cycle-related functions, and was correlated with patient survival. MCM3AP-AS and MIR17HG acted synergistically to regulate mRNAs in a network module of the competitive LMCN. By integrating the expression profile of this module into a risk model, we stratified GBM patients in both the The Cancer Genome Atlas and an independent GBM dataset into distinct risk groups. Finally, survival analyses demonstrated that the lncRNAs and network module are potential prognostic biomarkers for GBM. Thus, ceRNAs could accelerate biomarker discovery and therapeutic development in GBM.

Highlights

Glioblastoma multiforme (GBM) is an aggressive, malignant brain tumor [1]
Previous studies have indicated that long non-coding RNA (lncRNA) transcripts compete with endogenous mRNAs to bind miRNAs [25], which can be identified using current miRNA target prediction methods [26,27,28]
We found that lncRNA nodes usually had more degrees and betweenness centrality (BC) compared to mRNA nodes (Figure 2D, 2E)

Summary

Introduction

Glioblastoma multiforme (GBM) is an aggressive, malignant brain tumor [1]. There are approximately 10,000 new cases of high-grade glioma each year [2]. There is an urgent need for suitable molecular biomarkers for GBM diagnosis and for predicting patient prognosis. Long non-coding RNAs (lncRNAs) have received considerable attention [6]. These RNAs are non-proteincoding and are greater than 200 nucleotides in length. The expression of the oncogenic lncRNA MALAT1 was correlated with metastasis and patient survival in lung cancer [9] and GBM [8]. The lncRNA H19 was shown to promote cancer development and invasion in glioma [10]. Over-expression of HOTAIR has been shown to promote epithelial ovarian cancer metastasis and was predictive of poor patient prognosis [12]

Methods

Results

Conclusion