ZNF154 is a promising diagnosis biomarker and predicts biochemical recurrence in prostate cancer.

OBJECTIVE:To screen the methylated genes for early diagnosis and biochemical recurrence (BCR) prediction in prostate cancer (PCa) patients. METHODS:Differentially methylated CpG sites (DMCs) of PCa were screened out from The Cancer Genome Atlas (TCGA) and the Gene Expression Omnibus (GEO) database. Combined with TCGA RNA sequencing data and clinical information, the DMCs associated genes with different expression and related to BCR were selected as candidate genes. Then, the expression level of the best candidate gene ZNF154 was validated by quantitative real-time polymerase chain reaction (qRT-PCR). Finally, the prognosis potential of the hypermethylation gene ZNF154 was assessed by Kaplan-Meier, univariate and multivariate cox regression analysis. RESULTS:A total of 87 candidate genes were screened out. Compared to benign prostate (BP) tissues, ZNF154 has three hypermethylation sites (cg03234186, cg12506930, cg26465391) in the promoter region in PCa tissues. qRT-PCR results showed that ZNF154 expression level was reduced in PCa tissues than in BP tissues (P = 0.004). Besides, the ZNF154 methylation level was negatively correlated with mRNA expression (r = -0.766, P < 0.001), and was highly cancer-specific in PCa (area under the curves (AUCs) = 90.030%). In addition, Kaplan-Meier analysis showed ZNF154 methylation level was associated with BCR (P = 0.005), and ZNF154 could be an independent factor for BCR prediction in PCa by using univariate and multivariate cox regression analysis (P = 0.035, HR = 8.218). CONCLUSIONS:87 PCa specific genes were obtained. Further analysis gave the evidence that ZNF154 can be used as a specific maker for PCa diagnosis. Hypermethylation level of ZNF154 lead to gene expression inhibition and function loss, which contribute to the development and poor outcomes in PCa. In addition, the mean methylation level of ZNF154 can be used as an independent risk factor to predict BCR.

KEYWORDS: {{ getKeywords(articleDetailText.words) }}