Identification of Novel Pyroptosis-Related Gene Signatures to Predict Prostate Cancer Recurrence

Chun Li; Jie Zhu; Hexi Du; Chaozhao Liang

doi:10.3389/fonc.2022.814912

Identification of Novel Pyroptosis-Related Gene Signatures to Predict Prostate Cancer Recurrence

Front Oncol. 2022 May 20:12:814912. doi: 10.3389/fonc.2022.814912. eCollection 2022.

Authors

Chun Li^{1

2

3

4}, Jie Zhu⁵, Hexi Du^{1

3

4}, Chaozhao Liang^{1

3

4}

Affiliations

¹ Department of Urology, The First Affiliated Hospital of Anhui Medical University, Hefei, China.
² Department of General Surgery, The First Affiliated Hospital of Anhui Medical University, Hefei, China.
³ Institute of Urology, Anhui Medical University, Hefei, China.
⁴ Anhui Province Key Laboratory of Genitourinary Diseases, Anhui Medical University, Hefei, China.
⁵ Central Hospital Affiliated to Shandong First Medical University, Jinan, China.

Abstract

Prostate cancer (PCa) is a common malignant type of urogenital tract tumor with poor prognosis. Despite therapeutic advances, the recurrence and mortality rates of PCa have continued to increase with poor prognoses. Pyroptosis, also known as inflammatory cell necrosis, is a recently identified type of programmed cell death that can regulate the invasiveness, differentiation, proliferation, and metastasis of tumor cells; thus, it has a profound effect on the prognosis of patients with tumors. However, the relationship between pyroptosis and PCa remains unclear. We first identified 25 pyroptosis-related genes (PRGs) that were differentially expressed between PCa tissues and matched normal tissues in The Cancer Genome Atlas (TCGA) cohort. Based on the expression levels of 25 PRGs, PCa patients were clearly divided into two clusters and 17 PRGs were found to be significantly different between the two clusters, suggesting probable roles for these genes in the progression and recurrence of PCa. Therefore, the GSE40272 dataset with recurrence follow-up information was used to verify their value. Univariate analysis suggested that 5/17 genes were associated with recurrence, the number of genes did not decrease after least absolute shrinkage and selection operator (LASSO) regression analysis, and 5 PRGs constituted the risk score formula. Low-risk and high-risk subgroups identified using the recurrence model showed different disease-free survival (DFS) times (P<0.001) and the risk score of five PRGs was a factor of independence for recurrence in patients with PCa. In addition, Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) analyses suggested that these pathways, and comprising PRGs might be closely related to carcinogenesis and invasion of tumors, tumor microenvironment, and immune response. In conclusion, the expression signatures of PRGs play an important role in predicting PCa recurrence.

Keywords: immunity; prognosis; prostate cancer; pyroptosis; recurrence.