Identification of angiogenesis-related genes in the diagnosis of benign prostatic hyperplasia using bioinformatics analysis.

BACKGROUND: Benign prostatic hyperplasia (BPH), the most prevalent male chronic disease, features typical lower urinary tract symptoms (LUTS) that compromise male health. Angiogenesis, the formation of new vascular networks from pre-existing vessels, is typically related to cancer development and inflammatory microenvironments. Identifying and diagnosing angiogenesis-related genes implicated in BPH is of critical importance for the health management of the BPH population. Therefore, this study aims to identify and validate angiogenesis-related genes associated with BPH to provide potential biomarkers for its diagnosis and treatment. METHODS: BPH-related datasets were from the Gene Expression Omnibus (GEO). Modules specifically linked to BPH diagnosis were identified via weighted gene co-expression network analysis (WGCNA). Hub genes were noted through machine learning (ML) approaches: random forest (RF) and support vector machine-recursive feature elimination (SVM-RFE). A protein-protein interaction (PPI) network was constructed, with diagnostic performance rated via receiver operating characteristic (ROC) curves and nomograms. RESULTS: The emerald module identified by WGCNA exhibited a strong correlation with BPH. ML models identified three hub genes (CARTPT, TOB1, and GLO1). Subsequent validation revealed that all three genes were significantly downregulated in BPH patients. The final analysis demonstrated that these three hub genes and the derived nomogram possess high diagnostic value, with area under the curve (AUC) values of 0.956 for CARTPT, 0.939 for TOB1, and 0.933 for GLO1. CONCLUSIONS: Our findings suggest that CARTPT, TOB1, and GLO1 might serve as potential angiogenesis-related diagnostic biomarkers for BPH. These genes could possibly predict the risk of BPH and may offer prospective molecular targets for the clinical diagnosis and treatment of the disease.