摘要

Background/Aims: Long non-coding RNAs (lncRNAs) act as competing endogenous RNAs (ceRNAs) to compete for microRNAs (miRNAs) in cancer metastasis. Head and neck squamous cell carcinoma (HNSCC) is one of the most common human cancers and rare biomarkers could predict the clinical prognosis of this disease and its therapeutic effect. Methods: Weighted gene co-expression network analysis (WGCNA) was performed to identify differentially expressed mRNAs (DEmRNAs) that might be key genes. GO enrichment and protein-protein interaction (PPI) analyses were performed to identify the principal functions of the DEmRNAs. An lncRNA-miRNA-mRNA network was constructed to understand the regulatory mechanisms in HNSCC. The prognostic signatures of mRNAs, miRNAs, and lncRNAs were determined by Gene Expression Profiling Interactive Analysis (GEPIA) and using Kaplan-Meier survival curves for patients with lung squamous cell carcinoma. Results: We identified 2,023 DEmRNAs, 1,048 differentially expressed lncRNAs (DElncRNAs), and 82 differentially expressed miRNAs (DEmiRNAs). We found that eight DEmRNAs, 53 DElncRNAs, and 16 DEmiRNAs interacted in the ceRNA network. Three ceRNAs (HCG22, LINC00460 and STC2) were significantly correlated with survival. STC2 transcript levels were significantly higher in tumour tissues than in normal tissues, and the STC2 expression was slightly upregulated at different stages of HNSCC. Conclusion: LINC00460, HCG22 and STC2 exhibited aberrant levels of expression and may participate in the pathogenesis of HNSCC.