PROJECT SUMMARY/ABSTRACT The most common non-cutaneous cancer in American men, prostate cancer is characterized by a substantial degree of overdiagnosis and is consequently a significant source of morbidity and driver of unnecessary cost. There is therefore a significant need to identify prostate cancer patients who need aggressive therapy compared to those who need no treatment at all. The cancer-specific mortality of these tumors is attributable to metastatic disease. Adenocarcinoma of the prostate must interact with smooth muscle at multiple steps of the metastatic process, including intravasation and extravasation between the bloodstream and tissue. In addition, the cancer- prone peripheral zone of the human prostate gland is composed of a fibromuscular stroma and bounded by a smooth muscle capsule. While the current paradigm for most cancers is that local invasion is not predictive of aggressive disease, this is not the case with prostate cancer. Aggressive prostate cancer cells are directly observed as extending past the smooth muscle capsule in clusters, in a process called extracapsular extension (ECE), thereby escaping organ confinement. The presence of ECE defines the pT3a category in the pathologic staging of prostatic adenocarcinoma and is associated with an increased risk of biochemical recurrence, distant metastases and cancer-specific mortality. Although smooth muscle invasion is required for ECE, intravasation, and extravasation to distant sites, there are no markers of the switch from indolent to aggressive, muscle-invasive prostate cancer. This work is therefore concentrated on understanding the molecular events that regulate ECE to prevent this critical step in the metastatic process. The central hypothesis is that invasion through the smooth muscle capsule requires unique molecular properties that predict aggressive cancer. If correct, the concept will change current standard of care by providing new molecular discriminators of indolent versus aggressive disease. Objective measures of the aggressive disease transition will inform treatment decisions and help to direct the appropriate level of treatment for the degree of disease aggression.