AUTHOR=Fu Min , Wang Qiang , Wang Hanbo , Dai Yun , Wang Jin , Kang Weiting , Cui Zilian , Jin Xunbo 

TITLE=Immune-Related Genes Are Prognostic Markers for Prostate Cancer Recurrence

JOURNAL=Frontiers in Genetics

VOLUME=Volume 12 - 2021

YEAR=2021

URL=https://www.frontiersin.org/journals/genetics/articles/10.3389/fgene.2021.639642

DOI=10.3389/fgene.2021.639642

ISSN=1664-8021

ABSTRACT=Background: Prostate cancer (PCa) is an immune-responsive disease and this study was conducted to identify a robust immune-related prognostic gene signature for PCa.
Methods: The tumor Genome Atlas (TCGA) database and GSE46602 database were downloaded for the least absolute shrinkage and selection operator (LASSO) cox regression model, and immune related genes (IRGs) data were downloaded from import dataset. 
Results: In the weighted gene co-expression network analysis (WGCNA), nine functional modules are related to the biochemical recurrence of PCa, including 259 IRGs were identified. Univariate regression analysis and survival analysis identified 35 IRGs which were related to the prognosis of PCa. Using a LASSO Cox regression model, we constructed a risk prognosis model consisting of 18 IRGs. Multivariate regression analysis showed that the risk score was an independent predictor of the prognosis of PCa. A nomogram comprising a combination of this model and other clinical features showed good accuracy in predicting the prognosis of PCa. Further analysis found that different risk groups harbored different gene mutations, transcriptome expression and immune infiltration. There were more gene mutations in patients belonging to the high-risk group than those belonging to the low-risk group, especially high-frequency mutations in TP53. Immune infiltration analysis revealed that in the high-risk group, M2 macrophages were significantly enriched, which affected the prognosis of PCa. In addition, some immunostimulatory genes differentially expressed in risk group.
Conclusion: In summary, a risk prognosis model based on IGRs was developed and a nomogram comprising a combination of this model and other clinical features showed good accuracy in predicting the prognosis of PCa. This model may provide a basis for personalized treatment of PCa and help clinicians make effective treatment decisions.