BackgroundPDZ-binding kinase (PBK) encodes a serine/threonine protein kinase related to the dual specific mitogen-activated protein kinase kinase (MAPKK) family. There is evidence that overexpression of this gene is associated with tumorigenesis. However, the role of PBK in hepatocellular carcinoma (HCC) remain unclear. Therefore, we evaluated the prognostic role of PBK and its correlation with immune infiltrates in Hepatocellular Carcinoma.MethodsThe expression of PBK in pan-cancers was studied by Onconmine and TIMER. The expression of PBK in HCC patients and its relationship with clinicopathological characteristics were analyzed using The Gene Expression Profiling Interactive Analysis (GEPIA), The human protein atlas database (HPA), The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO) databases. Receiver operating characteristic (ROC) curve was used to determine the diagnostic value of PBK in HCC patients. The relationship between PBK and prognosis of HCC was performed by GEPIA and Kaplan Meier plotter web tool. The correlations between the clinical characteristics and overall survival were analyzed by Univariate Cox regression and Multivariate Cox hazards regression to identify possible prognostic factors for HCC patients. LinkedOmics was applied to investigate co-expression associated with PBK and to analyze Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways. The network map of PBK and related genes is constructed by GeneMANIA. Finally, TIMER and TISIDB were used to analyze the correlations between PBK and tumor-infiltrating immune cells.ResultsPBK was highly expressed in many types of tumors, including hepatocellular carcinoma, and was significantly related to tumor stage (P=0.0089), age (P=0.0131) and race (P=0.0024) of HCC patients. The receiver operating characteristic (ROC) curve analysis from 4 GEO databases showed that PBK is a predictive marker for HCC. Moreover, high expression of PBK correlated with poor prognosis. Multivariate Cox hazards regression showed that high expression of PBK may be an independent risk factor for overall survival in HCC patients (HR = 1.566, 95% CI=1.062-2.311, P= 0.024). The Protein–protein interaction network showed that PBK significantly interacted with LRRC47, ARAF, LGALS9B, TTK, MELK and other essential genes. Furthermore, enrichment analysis showed that PBK and co-expressed genes participated in many biological processes, cell composition, molecular functions and pathways in HCC. Finally, the immune infiltration analysis by TIMER and TISIDB indicated that PBK expression tightly correlated with the infiltration of tumor-infiltrating lymphocytes (TILs), chemokines, and receptors.ConclusionsHigh expression of PBK is significantly correlated with poor survival and immune infiltrates in hepatocellular carcinoma. Our study suggests that PBK can be used as a biomarker of poor prognosis and potential immune therapy target in hepatocellular carcinoma.