PBK is a potential therapeutic target and prognostic biomarker in ovarian cancer: integrating bioinformatics and clinical analysis

Abstract Purpose Ovarian cancer (OC) remains the most lethal gynecological malignancy worldwide. The recrudescence and chemoresistance are main reasons for the high mortality of OC. Identifying effective therapeutic targets is an urgent need. Cancer testis antigens (CTAs) are attractive targets for cancer therapy because of their expression restriction. Methods Based on the RNAseq data of OC in The Cancer Genome Atlas (TCGA) database, we screened the differentially expressed genes of OC from the Cancer-Testis Database and created a protein–protein interaction (PPI) network to analyze co-expressed genes. Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) analyses were performed to explore the potential functions of genes. The 3D structure of the protein was simulated and domain interactions were analyzed. We evaluated the expression of candidate gene in OC via qRT-PCR and immunohistochemistry, its correlation with clinical indicators was also analyzed. Results Seven significant CTAs were found to be differentially expressed in OC, PBK was eventually selected as a prognostic factor. The established PPI network contained 93 co-expression genes with 966 protein interactions. Functional enrichment analysis showed that PBK was involved in regulating cell cycle, cell division and mitosis. PBK shares common domains with CDK1, CDK2, PLK1, CCNA2 and CCNB1. PBK was apparently up-regulated in OC, and its overexpression was correlated with the FIGO stage, lymph-node metastasis, and chemoresistance. Conclusion PBK was highly expressed in OC and related to poor prognosis, it functions as a crucial cell-cycle regulator. PBK could be a potential biomarker for OC therapies and prognosis prediction.