Abstract:

Although nasopharyngeal carcinoma (NPC) and oral squamous cell carcinoma (OSCC) are highly correlated clinical diseases, the underling molecular mechanisms to link the two diseases remain largely unknown. The aim of this study is to identify the shared functional modules for NPC and OSCC by using large-scale transcriptomic data. Gene expression profile datasets of NPC and OSCC were obtained from the GEO database. A total of 1279 differentially expressed genes (DEGs) of NPC and 1293 DEGs of OSCC were identified by fold change and empirical Bayes method, and 278 DEGs were common to these two diseases. These overlapped genes were translated into a primary network consisting of 1290 nodes (genes) and 1766 edges. The primary network was then decomposed into 15 compacted modules (subnets) with high modularity by Newman’s algorithm. Topological analysis of these modules identified a total of 58 hub genes, most of which (e.g., PCNA, CDK1, STAT1, CCL5, and MMP1) have been proved to be associated with NPC and/or OSCC, while the rest (e.g., MELK, NME1, RACGAP1, INHBA, and NID1) might be novel risk genes for the two diseases. Further bioinformatics analysis of KEGG databases revealed that these modules are involved in multiple pathogenic biological pathways for either NPC or OSCC (e.g., p53 signaling pathway, ECM-receptor interaction, focal adhesion, and cell cycle). This study demonstrates that NPC and OSCC have similar molecular bases, and the identified pleiotropic modules may shape the complicated molecular interplays underlying the two clinically correlated diseases.

Key words: nasopharyngeal carcinoma, squamous cell carcinoma, microarrays, network analysis, pleiotropism