[Establishment of a prostate cancer prognostic risk model based on the TCGA database and inflammation-related genes]

Objective: To investigate the effect of inflammation-related genes on the prognosis of prostate cancer (PCa).

Methods: We downloaded PCa-related clinical data and mRNA sequencing data from the database Cancer Genome Atlas (TCGA) and inflammation-related pathway gene sets from MsigDB. Using univariate regression and LASSO regression analyses, we screened inflammation-related genes for the construction of a prognostic risk model and evaluated the performance of the model in predicting the prognosis of PCa by Kaplan-Meier and ROC analyses. Based on the nomogram, we calculated the risk scores of the patients, divided them into a high-risk and a low-risk group based on the median values of their risk scores, identified differentially expressed genes for enrichment analysis and verified the expression level of SPHK1 in the PCa tissue microarrays by immunohistochemical staining.

Results: Totally 19 inflammation-related genes were identified from 172 candidate genes for the construction of the prognostic risk model, including the risk genes CD14, PIK3R5, GABBR1, RELA, IRF7, SCARF1, MSR1, SPHK1, OSM and STAB1, and the protective genes AQP9, LPAR1, ATP2C1, NDP, CXCL6, P2RY2, DCBLD2, PCDH7, and IFNAR1. Kaplan-Meier analysis showed that the patients with high risk scores had a significantly lower recurrence-free survival and a worse prognosis than those with low risk scores. Differentially expressed genes were involved mainly in the activation of inflammatory response pathways. Immunohistochemical results indicated that the expression of SPHK1 was significantly higher in the tumorous than in the normal tissue and increased with the Gleason score. There was a correlation between the SPHK1 expression and envelope invasion.

Conclusion: The prognostic risk model of inflammation-related genes constructed based on the TCGA database can effectively predict the prognosis of PCa.