Background: Prostate cancer is the most commonly diagnosed malignancy in men and the second most frequent cause of male cancer deaths. In 2007 approximately 218,000 men were diagnosed with prostate cancer and 27,000 deaths have occurred. With effective prostate cancer screening programs, such as testing for levels of the prostate specific antigen (PSA) or digital rectal exams (DREs) at hand, an increasing number of men are identified at early stages of prostate cancer and treated for the disease. However, in spite of effective therapies, such as surgical removal of the gland (prostatectomy) or targeting of tumor tissue by irradiation, a significant number of men will suffer recurrence of the disease, and this form of tumor is usually more resistant to therapy. Ultimately, treatment-resistant cancer will lead to metastasis of the tumor from the prostate to lymph nodes or bone. Unfortunately, as of today there are no predictive models that could identify men at high risk for disease recurrence and progression. The identification of high risk primarily relies on information on treatment response and on classical inspection of basic cellular features of the tumor under a microscope. Therefore, a lot of ongoing research is dedicated to the identification of the underlying genetic and molecular changes that either initiate the disease or that can be linked to causing aggressive tumors, which metastasize and are hard to treat. We are studying the gene PTEN (pronounced as "P-10"), which is mutated in up to 80% of advanced prostate cancers. To determine if it is causally linked to the process, we have mutated the gene in mouse and found that partial PTEN loss initiates prostate cancer while complete loss shuts down cellular proliferation through a process called senescence, but only until the p53-gene, the cell's major guardian for integrity, is lost. Thus, unexpectedly, cells have an intrinsic capacity to fight back the tumor with p53. We therefore defined to two distinct classes of PTEN mutant tumors: slow-growing prostate cancer with partial PTEN loss that does not trigger a p53-defensive response and fast-growing, aggressive cancers with complete PTEN and p53 loss.

Objective: In this study, we want to test if the many human prostate cancers with PTEN alterations also fall into these two categories. Our analysis in mouse has afforded us a number of genetic and molecular tests for unambiguous assignment of cancers in respect to our two categories. We are now carrying out these tests in collaboration with clinicians at Memorial Sloan-Kettering Cancer Center as part of their ongoing comprehensive prostate sample analysis program. In addition, we will identify novel genetic changes in mouse and correlate them with changes in human prostate cancer.

Impact on Basic Prostate Cancer Research: PTEN mouse models have been key for understanding the genetics of prostate cancer initiation. If we confirm that these models also faithfully recapitulate the genetics of disease progression, then this will provide the medical research community with an invaluable tool for studying prostate cancer as well as for treating the disease by using the well-defined PTEN mutant mice for drug trials.

Impact on Patient Care: Of the many men diagnosed with prostate cancer in the United States every year, over 150,000 are expected to have PTEN alterations. Our research will help these men by defining strict criteria for identification of those cases that are at high risk for recurrence and metastasis. This will have a profound effect on patient diagnosis and treatment choices as the costs for the molecular genetic analysis of biopsies are ever-decreasing.