Non-Coding RNA Biomarkers in Prostate Cancer: Evidence Mapping and in Silico Characterization

Background: Non-coding RNAs (ncRNAs) have emerged as promising biomarkers for prostate cancer (PCa), yet evidence remains dispersed across heterogeneous studies and their regulatory context is seldom analyzed in an integrated manner. This study systematically maps ncRNAs reported as diagnostic biomarkers for PCa and characterizes their molecular interactions through in silico analyses. Methods: A comprehensive evidence-mapping strategy across major bibliographic databases identified 693 studies, of which 58 met eligibility criteria. Differentially expressed ncRNAs were extracted and classified by RNA type. Subsequently, miRNA–target prediction, miRNA–protein interaction network construction, and functional enrichment analyses were performed to explore the regulatory landscape of miRNA-associated proteins. Results: The final dataset included 4,500 participants (2,871 PCa cases and 2,093 controls) and reported 94 differentially expressed miRNAs, eight lncRNAs, and several circRNAs, snoRNAs, snRNAs, and piRNAs. In silico analyses predicted 13,493 miRNA–mRNA interactions converging on 4,916 unique target genes, with an additional 2,481 prostate tissue–specific targets. The miRNA–protein network comprised 845 nodes and 2,335 edges, revealing highly connected miRNAs (e.g., hsa-miR-16-5p, hsa-miR-20a-5p) and protein hubs (QKI, YOD1, TBL1XR1; prostate-specific CDK6, ACVR2B). Enrichment analysis showed strong overrepresentation of metabolic process–related GO terms and cancer-associated KEGG pathways. Conclusions: These findings refine the list of promising ncRNA biomarkers and highlight candidates for future clinical validation.