Exploring Urinary Epigenetic Biomarkers for Identification of Aggressive Prostate Cancer

Prostate cancer (PCa) is the most common non-cutaneous malignancy in men. However, the majority of PCa diagnoses are localized, low-grade, clinically insignificant (CI-PCa) tumours that are unlikely to progress. The current clinical challenges for PCa patient management are to reduce the over-treatment of patients with CI-PCa and the early identification of patients with clinically significant cancer (CS-PCa). To address these challenges, most patients diagnosed with low-grade tumours are recommended enrollment into Active Surveillance (AS) protocols where they are frequently monitored with MRI if available, repeated prostate-specific antigen (PSA) tests and digital rectal exams (DREs), and periodic prostate biopsies. Around 20-30% of AS patients will show signs of progression within 5 years. They are then “reclassified” with higher-risk disease and offered definitive treatment. While AS reduces over-treatment of CI-PCa patients, monitoring via prostate biopsies is invasive and causes morbidity. Additionally, patients who will reclassify may benefit from earlier identification and treatment instead of monitoring until symptoms of progression are exhibited. In this regard, non-invasive biomarkers that can distinguish CI-PCa and CS-PCa patients upfront, are warranted. Epigenetic dysregulations, including DNA methylation and miRNAs, are hallmarks of PCa. They are beneficial to serve as potential biomarkers for PCa diagnosis and prognosis as they require small amounts of starting material, is very stable, and can be assessed in a variety of different sources including non-invasively collected liquid biopsies such as serum or urine. Previously, my lab has discovered and/or characterized several DNA methylation and miRNA biomarkers that are associated with CS-PCa. In this thesis, I assess these biomarkers in the urine samples of patients at varying stages of PCa. I developed and validated two DNA methylation biomarker panels that can differentiate between CI-PCa from CS-PCa (AUC=0.795) and predict AS patient reclassification (OR=2.695). Additionally, I developed a panel consisting of both urinary DNA methylation and miRNAs that could predict AS patient reclassification (OR=2.166). The findings presented in this thesis, with further validation, will hopefully lead to non-invasive urine-based tests that will drive clinical decision making to improve PCa patient management and reduce invasive biopsies and their associated morbidity.Ph.D