| Background: Due to very high disease-specific survival following prostate cancer treatment, disease progression (metastasis) and treatment-related complications may significantly affect a patient's life trajectory. Using distinct epidemiologic methodologies, this thesis sought to (1) identify novel microRNA predictors of metastasis following radical prostatectomy; (2) examine the association between local treatment modality (surgery or radiotherapy) and androgen deprivation therapy (ADT) on non-prostate cancer mortality; and (3) study the association between radiotherapy for prostate cancer and secondary malignancies.;Methods: We conducted a matched-case control study of 38 patients who underwent radical prostatectomy using bootstrapping with automated backward selection to identify miRNA sequences which were significantly associated with metastasis. To examine the association between treatment modality and non-prostate cancer mortality, we performed a propensity-score matched, population-based retrospective cohort study of 10,786 men treated for non-metastatic prostate cancer in Ontario between 2002 and 2009. We used the Fine and Gray method with generalized estimating equation survival models with a sandwich variance estimator to calculate the sub-distribution hazard ratio of treatment effect, accounting for ADT exposure in a time-varying manner. To assess the association between radiotherapy for prostate cancer and the development of secondary cancers, we performed a systematic review and meta-analysis utilizing random-effects models and Mantel-Haenszel weighting.;Results: We identified a panel of five microRNA which were associated with metastasis following surgery (AUC 89.5%, 95% CI 79.5-99.5). Treatment with radiotherapy was independently associated with an increased risk of non-prostate cancer mortality (HR 1.57, 95% CI 1.35-1.83) though ADT exposure was not (p = 0.26 - 0.87 depending on analytic strategy). Radiotherapy was associated with an increased risk of bladder (aHR 1.67, 95% CI 1.55-1.80), colorectal (aHR 1.79, 95% CI 1.34-2.38) and rectal cancers (aHR 1.79, 95% CI 1.34-2.38) but not hematological (aHR 1.64, 95% CI 0.90-2.99) or lung (aHR 1.45, 95% CI 0.70-3.01) cancers.;Conclusions: Varied epidemiologic techniques may be used to characterise outcomes following prostate cancer treatment. These data may inform patients and physicians when making decisions regarding prostate cancer treatment choice. |
