Causal association and shared mechanisms between Graves' disease and prostate cancer: insights from Mendelian randomization, machine learning, and comprehensive bioinformatics.

BACKGROUND: Observational studies link hyperthyroidism to increased prostate cancer (PCa) risk, but causality and mechanisms remain unclear. Graves' disease (GD), the primary cause of hyperthyroidism, involves chronic immune dysregulation that may influence PCa through shared immune pathways. METHODS: We performed bidirectional two-sample Mendelian randomization (MR) using IEU Open GWAS data, then integrated differential expression analysis, weighted gene co-expression network analysis (WGCNA), and machine learning on Gene Expression Omnibus (GEO) datasets to identify shared gene, validated by ROC curves, and analyzed immune profilesusing ssGSEA. RESULTS: MR analysis indicated that genetic predisposition to GD significantly reduced PCa risk (OR = 0.997, 95% CI = 0.996-0.999, p = 0.004), with consistentsensitivity and no reverse causality. Four key genes (BTG2, JUN, JUNB, FOS) were identified as robust shared genes with high predictive accuracy in external validation. Immune profiles analysis revealed disease-specific associations of these genes: BTG2 and JUNB correlated with memory CD8 T cells in GD, whereas all four genes correlated with dendritic cells, mast cells and NK cells in PCa. CONCLUSION: This study provided novel insights into the protective effect of GD against PCa and identified shared genes and immune mechanisms, offering a deeper understanding of the common mechanisms between GD and PCa.