Evaluating the prognosis of prostate cancer through metabolic pathways.

BACKGROUND: The pathogenesis of prostate cancer (PCa) is strongly influenced by metabolism. Thus, we explored candidate genes with metabolism-related functions that can be used to predict PCa prognosis. METHODS: To create a training set and two validation sets, RNA data and clinical parameters were downloaded from The Cancer Genome Atlas (TCGA) and the Gene Expression Omnibus (GEO). Risk score (RS) was constructed based on metabolism-related gene signatures. The predictive power of the RS was evaluated. A nomogram related to biochemical recurrence-free survival (BCRFS) was built and evaluated. Finally, an enrichment analysis using Gene Set Enrichment Analysis (GSEA) was performed to identify enriched Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) categories. The Human Protein Atlas (HPA) platform was employed to identify the expression of important genes at the protein level. RESULTS: Five gene signatures from 860 metabolism-related genes were selected. RS were constructed using the five-gene signatures, which showed high prognostic power for biochemical recurrence (BCR). The nomogram effectively predicted BCRFS. According to GO analysis, DNA damage is primarily associated with genes involved in metabolism. The five-gene signatures were primarily enriched in sulfur metabolic pathways, as analyzed by KEGG. With the progression of prostate cancer malignancy, the expression of HAGHL, and INPP5E also increased. CONCLUSION: The study establishes a robust, metabolism-based prognostic model for prostate cancer.