Phd in NETWORK ONCOLOGY AND PRECISION MEDICINE

Thesis title: Network Medicine Applications for Prostate Cancer Early Detection

Background: Prostate cancer (PCa) remains the most frequently diagnosed malignancy in men and a major cause of morbidity and mortality worldwide. Despite the advances achieved with magnetic resonance imaging (MRI), current diagnostic pathways still face critical challenges, particularly the balance between overdiagnosis of clinically insignificant PCa (ciPCa) and missed detection of clinically significant disease (csPCa). The growing availability of molecular profiling techniques and computational methods offers new opportunities for precision diagnostics. Objective: This doctoral project aimed to develop and validate an integrated, non-invasive, and evidence-based diagnostic pathway for the early detection of csPCa through the combined use of clinical, imaging, and molecular data analyzed within a network medicine framework. Methods: The project was structured in three consecutive phases. A pilot proof-of-concept study established the feasibility of integrating mpMRI biomarkers, circulating microRNA (miRNA) profiles, and clinical data. Weighted Gene Co-expression Network Analysis (WGCNA) identified key miRNAs associated with disease aggressiveness, followed by functional enrichment analyses. A validation study was then conducted prospectively to confirm the diagnostic performance of the identified MRI biomarkers with the profiled miRNAs (miR-302a-5p and miR-367-3p) and clinical data. Results: The integrated pathway significantly improved the accuracy of csPCa detection compared with the standard pathway, MRI-only pathways and the miRNA-only pathway. Decision Curve Analysis demonstrated the highest net clinical benefit at low diagnostic thresholds (5–10%), corresponding to a 21–20% reduction in unnecessary biopsies without missing csPCa diagnoses. Conclusions: This work proposes and validates a novel, network-based integrated diagnostic model that combines imaging, molecular, and clinical biomarkers for precision detection of prostate cancer. The findings support the feasibility and clinical relevance of multi-omic integration in diagnostic decision-making, paving the way toward personalized, cost-effective, and non-invasive prostate cancer early detection strategies.