Identification of Pathologic and Prognostic Genes in Prostate Cancer Based on Database Mining

Kun Liu; Yijun Chen; Pengmian Feng; Yucheng Wang; Mengdi Sun; Tao Song; Jun Tan; Chunyang Li; Songpo Liu; Qinghong Kong; Jidong Zhang

doi:10.3389/fgene.2022.854531

Identification of Pathologic and Prognostic Genes in Prostate Cancer Based on Database Mining

Front Genet. 2022 Mar 11:13:854531. doi: 10.3389/fgene.2022.854531. eCollection 2022.

Authors

Kun Liu¹, Yijun Chen¹, Pengmian Feng², Yucheng Wang¹, Mengdi Sun¹, Tao Song¹, Jun Tan³, Chunyang Li¹, Songpo Liu¹, Qinghong Kong⁴, Jidong Zhang^{1

5}

Affiliations

¹ Department of Immunology, Zunyi Medical University, Zunyi, China.
² School of Basic Medical Sciences, Chengdu University of Traditional Chinese Medicine, Chengdu, China.
³ Department of Histology and Embryology, Zunyi Medical University, Zunyi, China.
⁴ Guizhou Provincial College-based Key Lab for Tumor Prevention and Treatment with Distinctive Medicines, Zunyi Medical University, Zunyi, China.
⁵ Special Key Laboratory of Gene Detection and Therapy of Guizhou Province, Zunyi Medical University, Zunyi, China.

Abstract

Background: Prostate cancer (PCa) is an epithelial malignant tumor that occurs in the urinary system with high incidence and is the second most common cancer among men in the world. Thus, it is important to screen out potential key biomarkers for the pathogenesis and prognosis of PCa. The present study aimed to identify potential biomarkers to reveal the underlying molecular mechanisms. Methods: Differentially expressed genes (DEGs) between PCa tissues and matched normal tissues from The Cancer Genome Atlas Prostate Adenocarcinoma (TCGA-PRAD) dataset were screened out by R software. Weighted gene co-expression network analysis was performed primarily to identify statistically significant genes for clinical manifestations. Protein-protein interaction (PPI) network analysis and network screening were performed based on the STRING database in conjunction with Cytoscape software. Hub genes were then screened out by Cytoscape in conjunction with stepwise algorithm and multivariate Cox regression analysis to construct a risk model. Gene expression in different clinical manifestations and survival analysis correlated with the expression of hub genes were performed. Moreover, the protein expression of hub genes was validated by the Human Protein Atlas database. Results: A total of 1,621 DEGs (870 downregulated genes and 751 upregulated genes) were identified from the TCGA-PRAD dataset. Eight prognostic genes [BUB1, KIF2C, CCNA2, CDC20, CCNB2, PBK, RRM2, and CDC45] and four hub genes (BUB1, KIF2C, CDC20, and PBK) potentially correlated with the pathogenesis of PCa were identified. A prognostic model with good predictive power for survival was constructed and was validated by the dataset in GSE21032. The survival analysis demonstrated that the expression of RRM2 was statistically significant to the prognosis of PCa, indicating that RRM2 may potentially play an important role in the PCa progression. Conclusion: The present study implied that RRM2 was associated with prognosis and could be used as a potential therapeutic target for PCa clinical treatment.

Keywords: TCGA; biomarker; differentially expressed genes; prognosis; prostate cancer.