Identification of hub genes and construction of a survival prediction model for patients with nasopharyngeal carcinoma

The incidence of nasopharyngeal carcinoma (NPC) has remained static in southern provinces of China and poses a serious threat to public health. Biomarkers and prediction models that can accurately predict the survival of patients with NPC are lacking. In this study, the hub genes of NPC were identified using the gene expression datasets GSE61218 and GSE126683. The expression levels of the hub genes were subsequently determined in clinical samples, and the relationships between the expression levels of the hub genes and patient survival were analysed. Finally, a survival prediction model was constructed using clinical data and hub genes as variables, and the performance of the prediction model was evaluated. AURKA, AURKB, BUB1, BUB1B, CCNA2, CCNB2, and CDK1 were identified as hub genes, all of which were significantly upregulated in tumor tissues. The expression levels of AURKA, BUB1, and CDK1 were significantly upregulated in NPC samples from patients in the Death group. The results of the log-rank test suggested that the overall survival rate of patients with high expression levels of AURKA, BUB1, or CDK1 was significantly reduced. Finally, a survival prediction model was constructed using gender, age, T stage, N stage, M stage, BUB1 expression, and AURKA expression as variables. The results of the receiver operating characteristic (ROC) curve, area under the ROC curve, calibration plot, net reclassification index, integrated discrimination improvement index, and decision curve analysis revealed that the model had good discriminating ability, predictive ability, and clinical utility. In conclusion, AURKA, BUB1, and CDK1 are potential prognostic biomarkers of NPC, and a prediction model incorporating the expression levels of AURKA and BUB1 has good discriminating ability, predictive ability, and clinical utility.