Role of circulating MicroRNAs in prostate cancer diagnosis and risk stratification in the MCC Spain study

To identify circulating microRNAs (miRNAs) associated with prostate cancer and to develop predictive models capable of distinguishing cases from controls and stratifying patients by Gleason risk categories (low, intermediate, and high risk). This case-control study included 203 prostate cancer cases and 54 population-based controls. Serum samples were analyzed by RT-qPCR (performed at QIAGEN Genomic Services). Total RNA was extracted from 200 µl of serum using the miRNeasy Serum/Plasma Advanced Kit and reverse-transcribed with the miRCURY LNA RT Kit. A panel of 46 candidate miRNAs was profiled, and feature selection was performed using LASSO penalization. Logistic regression models were used to estimate age- and covariate-adjusted odds ratios (ORs) with 95% confidence intervals (CIs) for the association between miRNA expression and prostate cancer risk. Predictive performance was assessed using repeated 5-fold cross-validation with bootstrap resampling (10 repetitions; 1000 resamples), and summarized using the area under the receiver operating characteristic curve (AUC) with bias-corrected 95% CIs. Fourteen miRNAs were significantly associated with prostate cancer. Notably, miR-199a-5p (OR = 1.89, 95% CI: 1.13-3.15; p = 0.015) and miR-150-5p (OR = 0.20, 95% CI: 0.06-0.63; p = 0.006) showed consistent differential expression across all Gleason risk categories, with miR-199a-5p overexpressed and miR-150-5p underexpressed, suggesting a potential role in disease progression. miR-145-5p, miR-182-5p, and miR-93-5p were significantly associated with prostate cancer in both the overall model and in low- and intermediate-risk strata, highlighting their potential relevance in early-stage disease. In contrast, miR-24-3p was exclusively overexpressed in high-risk prostate cancer (OR = 2.93, 95% CI: 1.43-5.98; p = 0.003), indicating a possible link with aggressive tumor phenotypes. Predictive models demonstrated strong discriminatory performance, particularly for the low-risk group (AUC = 0.930, 95% CI: 0.882-0.979), followed by the intermediate-risk (AUC = 0.806, 95% CI: 0.728-0.883) and high-risk groups (AUC = 0.752, 95% CI: 0.658-0.848). The overall model achieved an AUC of 0.824 (95% CI: 0.756-0.892), reflecting robust performance in distinguishing cases from controls. This study identifies key circulating miRNAs associated with prostate cancer and demonstrates their potential in predictive models for risk stratification. The strongest discriminatory performance was observed in low-risk tumors (AUC = 0.930), followed by intermediate- (AUC = 0.806) and high-risk (AUC = 0.752) categories. These findings support the use of miRNAs as non-invasive biomarkers for diagnosis and personalized management of prostate cancer.