Comprehensive analysis of competitive endogenous RNAs network associated with head and neck squamous cell carcinoma.

Long non-coding RNAs (lncRNAs) can regulate gene expression directly or indirectly through interacting with microRNAs (miRNAs). However, the role of differentially expressed mRNAs, lncRNAs and miRNAs, and especially their related competitive endogenous RNAs (ceRNA) network in head and neck squamous cell carcinoma (HNSCC), is not fully comprehended. In this paper, the lncRNA, miRNA, and mRNA expression profiles of 546 HNSCC patients, including 502 tumor and 44 adjacent non-tumor tissues, from The Cancer Genome Atlas (TCGA) were analyzed. 82 miRNAs, 1197 mRNAs and 1041 lncRNAs were found to be differentially expressed in HNSCC samples (fold change ≥ 2; P < 0.01). Further bioinformatics analysis was performed to construct a lncRNA-miRNA-mRNA ceRNA network of HNSCC, which includes 8 miRNAs, 71 lncRNAs and 16 mRNAs. Through survival analysis based on the expression profiles of RNAs in the ceRNA network, we detected 1 mRNA, 1 miRNA and 13 lncRNA to have a significant impact on the overall survival of HNSCC patients (P < 0.05). Finally, some lncRNAs, which are more important for survival, were also predicted. Our research provides data to further understand the molecular mechanisms implicated in HNSCC.