POPDC3 is a potential biomarker for prognosis and radioresistance in patients with head and neck squamous cell carcinoma.

Radiotherapy is the primary means of treatment for patients with head and neck squamous cell carcinoma (HNSCC); however, radioresistance-induced recurrence is the primary cause of HNSCC treatment failure. Therefore, identifying specific predictive biomarkers of the response to radiotherapy may improve prognosis. In the present study, to identify the potential candidate genes associated with radioresistance in patients with HNSCC, the microarray datasets GSE9716, GSE61772 and GSE20549 were downloaded from the Gene Expression Omnibus database. The original CEL files were preprocessed using the Affymetrix package and quantile normalization and background correction were conducted using the Core package in Bioconductor. The GSE9716 dataset, consisting of 18 irradiated and 16 non-irradiated samples, was divided into two groups according to their exposure to irradiation: i) Non-irradiation group, which included 8 radioresistant samples and 8 radiosensitive samples; and ii) post-irradiation group, which included 9 radioresistant samples and 9 radiosensitive samples. The two groups were treated as separate datasets and screened. A total of 86 differentially expressed genes (DEGs) were identified in the non-irradiation group and 405 DEGs in the post-irradiation group. Gene Ontology enrichment and Kyoto Encyclopedia of Genes and Genomes pathway analysis detected several significant functions associated with the DEGs. In the co-expression analysis, 76 hub genes in the light green module and 917 hub genes with a high connectivity were selected for further analysis. Finally, overlapping the DEGs and hub genes from the two groups yielded a map of 13 shared differentially expressed genes. The 13 genes showed significantly different expression in radioresistant samples compared with the radiosensitive samples before and after irradiation. Out of these genes, popeye domain-containing protein 3 (POPDC3) was highly expressed in the post-irradiation group compared with the non-irradiation group. In survival analysis, high POPDC3 expression correlated with poor a prognosis for patients with HNSCC. The independent prognostic factors were identified using univariate and multivariate Cox analyses based on The Cancer Genome Atlas database. These were incorporated into a nomogram to predict 3- and 5-year overall survival. Receiver operating characteristic curves were used to estimate the accuracy of the nomogram. Together these studies suggest that POPDC3 may serve as a potential predictive biomarker for prognosis and radioresistance of patients with HNSCC as well as clinical diagnosis and treatment of patients.