Background
The 5-year overall survival rate in metastatic prostate adenocarcinoma (PRAD) is extremely low. Genomic studies of PRAD have improved our understanding of disease biology. However, the role of immune checkpoint genes (ICGs) in PRAD remains unclear.
Conclusions
Our results provide new targets for predicting tumor metastasis in PRAD and treating patients with metastatic PRAD.
Methods
Univariate and multivariate analyses were used to analyze genes associated with metastasis-free survival (MFS) in The Cancer Genome Atlas (TCGA)-PRAD dataset. The expressions of ADORA2A and TNFRSF18 were detected via immunohistochemical assay and real-time fluorescence quantitative PCR (RT-PCR) assay in our in-house cohort. The expression of long non-coding RNAs (lncRNAs) AL139287.1, SLC9A3-AS1, and SNHG12 were detected via RT-PCR assay in our in-house cohort. Stepwise regression, Cox regression, and nomogram analyses were used to evaluate the prognostic role of these genes in both the TCGA dataset and in-house cohort. The "pRRophetic" R package was used to evaluate drug sensitivity in the TCGA cohort according to the gene mRNA expression level.
Results
In our study, univariate and multivariate analyses revealed that the mRNA expressions of two ICGs, ADORA2A and TNFRSF18, were independent factors affecting MFS in PRAD patients. A prognostic 2-ICG model predicted the MFS of PRAD patients with medium-to-high accuracy in the TCGA dataset and in-house cohort. The expressions of AL139287.1, SLC9A3-AS1, and SNHG12 were correlated with ADORA2A and TNFRSF18. A prognostic lncRNA-ICG model predicted the MFS of PRAD patients with medium-to-high accuracy in the TCGA dataset and in-house cohort. In addition, correlation analyses between the sensitivity of doxorubicin, erlotinib, gemcitabine, or vinorelbine and AL139287.1, SLC9A3-AS1, SNHG12, ADORA2A, and TNFRSF18 were conducted. Conclusions: Our results provide new targets for predicting tumor metastasis in PRAD and treating patients with metastatic PRAD.