Abstract Background: MicroRNAs (miRNAs) are central regulators of gene expression, and rare pathogenic mutations in miRNA genes can cause Mendelian disease. However, the contribution of germline miRNA variants to prostate cancer (PCa) risk and clinical outcomes remains unclear, in part due to limited coverage of miRNA genes in previous exome, GWAS, and population sequencing datasets. Methods: We performed ultra-deep, targeted germline sequencing and rigorous variant calling of over 500 miRNA genes (including flanking regions) in 1, 502 PCa patients from hereditary and sporadic cohorts, representing both high- and low-grade disease. Variants were systematically annotated for functional region (stem, seed, mature, loop) and predicted impact on pre-miRNA folding (ΔG). Frequencies of recurrent variants (2 carriers) were compared to gnomAD whole-genome sequencing (WGS) and whole-exome sequencing (WES) populations using Fisher’s exact test within ancestry-matched groups. Results were cross-referenced with miRNASNP-v4, miRNA-seq data from prostate cancer cell lines, and TCGA-PRAD expression/outcome data. Per-gene miRNA burdens and individual variants were assessed for association with clinical features (family history, age, grade, metastasis, progression, PSA, cause of death). Results: We generated a comprehensive, high-confidence annotation resource of miRNA gene variants in germline DNA from prostate cancer patients, identifying 48 recurrent miRNA gene variants significantly enriched in PCa cases versus gnomAD (p 1×10^−8), including both known and novel polymorphisms. A subset displayed significant pre-miRNA folding change (ΔG 2 or 5 kcal/mol) and/or mapped to seed or mature regions. Several variants overlapped miRNA genes robustly detected in PCa cell lines and differentially expressed in TCGA-PRAD, with recurrent hits in known risk regions such as 8q24 (notably enriched among African/African American patients). Conclusions: This study establishes the deepest and most extensive annotation of miRNA gene variants in prostate cancer to date, integrating germline miRNA variation, functional prediction, and population ancestry. We observed significant discrepancies in miRNA variant allele frequencies when comparing gnomAD WGS to WES data, underscoring the critical need to use WGS datasets for reliable estimation of miRNA gene variant frequencies. Our results identify multiple miRNA gene variants that are enriched in our PCa cohort compared to gnomAD WGS reference populations and are predicted to impact miRNA biogenesis, function, and/or PCa cell biology. While these findings are novel and present strong candidates for further study, they remain preliminary and should be validated in independent cohorts for association with PCa risk and biological effects. Citation Format: Shawn E. Lupold, Jun Wei, Siqun Zheng, Jun Luo, William Isaacs, Jianfeng Xu. Comprehensive Evaluation of Recurrent microRNA Gene Variants in Prostate Cancer: Integrative Association with Risk, Predicted Function, and Clinical Outcome abstract. In: Proceedings of the AACR Special Conference in Cancer Research: Innovations in Prostate Cancer Research and Treatment; 2026 Jan 20-22; Philadelphia PA. Philadelphia (PA): AACR; Cancer Res 2026;86 (2Suppl): Abstract nr A041.
Lupold et al. (2026) studied this question.