Abstract
Increasing evidence suggests that N6-methyladenosine(m6A) has a vital role in cancer progression. Therefore, we aimed to explore the prognostic relevance of m6A-related genes in oral squamous cell carcinoma (OSCC). First, Expression profiles were downloaded from The Cancer Genome Atlas (TCGA) and m6A-related genes were extracted afterwards. Then, cluster analysis and principal component analysis (PCA) were used to analyze m6A-related genes. And differentially-expressed analysis was performed in R software. Furthermore, a risk model was constructed, and crucial m6A genes were selected to explore its biological effects in OSCC cells. Total of 13 m6A-related genes were extracted and 8 differentially-expressed genes were identified. Subsequently, m6A-based clustering showed 2 subtypes with different clinical outcome. In addition, a risk model was successfully established. Of 13 m6A-related genes, only heterogeneous nuclear ribonucleoprotein C (HNRNPC) might be an independent biomarker and mean unfavorable overall survival in OSCC by univariate and multivariate cox regression analysis. Functional studies revealed that overexpression of HNRNPC promoted carcinogenesis of OSCC via epithelial- mesenchymal transition (EMT). In total, a risk model of m6A-related genes in OSCC was established. Subsequently, HNRNPC was proved to promote OSCC carcinogenesis and be an independent biomarker prognostic biomarker of OSCC, suggesting that it might be a new biomarker and therapeutic target of OSCC.