Prostate cancer is the most commonly diagnosed cancer and the third cause of cancer death in Canadian men. Some prostate tumors remain indolent for many years, while others progress very rapidly, spread to other organs (metastases) and become fatal. Unfortunately, we do not have good tools to predict (prognosis), at the time of diagnosis, which patients are going to need immediate treatment from those who will need only surveillance. It is well known that prostate cancer cells have alterations in their genome (DNA). Our recent research results suggest that aggressive tumors harbor genomic alterations that differ from indolent ones. These changes include genes that could be linked to the ability of tumors to spread to others organs. We think that those alterations, generally gains or losses of DNA, can serve as molecular markers similar to ¿fingerprints¿ to identify the type of tumor. We have developed a technique to test clinical samples for these molecular markers. In this project, we propose to identify the genomic alterations associated with metastases that could eventually serve as markers. We plan to evaluate the prognostic value of the new molecular markers by analyzing hundreds of prostate tumor samples with complete clinical follow-up. We will evaluate if there is a link between the presence of a marker and a specific clinical outcome (remission or disease recurrence). This work should lead to the development of new clinical tests. We will also study the role of genes affected by genomic alterations in disease progression by using in vitro and in vivo prostate cancer models. Understanding their role in metastases formation should enable the design of new therapeutics agents for the future.