Abstract
Recent studies highlight the role of uric acid in tumor development, but its impact on prostate cancer (PCa) remains underexplored. This study aimed to investigate how uric acid influences PCa prognosis by analyzing transcriptomic data on PCa and uric acid-related genes (UARGs) from public databases. Differential expression analysis, protein-protein interaction (PPI) network, univariate Cox regression, and machine learning were used to identify prognostic genes. A risk model was then constructed based on these genes. Six prognostic genes (AHSG, AOX1, APOC1, LPL, NKX2-2, NKX6-1) were identified through the analysis of 1 433 differentially expressed genes (DEGs) and 3 806 UARGs. The risk model showed strong predictive ability, with the high-risk group (HRG) exhibiting poorer prognosis. Additionally, 10 immune cell types were significantly different between risk groups, with the HRG showing higher tumor mutation burden. A total of 8 drugs were found to correlate with risk scores. Enrichment analysis revealed that AHSG, AOX1, and APOC1 were linked to oxidative stress and Parkinson's disease, while NKX2-2 and NKX6-1 were associated with RNA degradation. These findings suggest that oxidative stress may be a key mechanism in PCa progression. This study offers a novel perspective on PCa treatment by identifying 6 prognostic genes and providing a prognostic risk model.