Prostate cancer is the most common form of cancer in American men. More than 230,000 men will be diagnosed with prostate cancer this year, and 30,000 will die from the disease. In 2005, our team at Michigan was fortunate enough to make the landmark discovery that 40%-80% of prostate cancer harbors gene fusions, which may be the cause of this highly prevalent disease. This finding was partially supported by the Department of Defense. Specifically, we used a computational approach to nominate cancer-causing genes from DNA microarray data. A testosterone-regulated gene called TMPRSS2 was found to be fused to a family of cancer causing genes called ETS factors. Fusing of these genes likely causes the development of prostate cancer. Previous to this discovery, it was thought that gene fusions only cause cancers in leukemias and lymphomas, but not common solid tumors such as prostate cancer. Our group is working on ways to take advantage of this finding to help in the diagnosis, prognosis, and treatment of men with prostate cancer.

In this proposal, we extend the computational approach further to study gene fusion negative prostate cancer. Interestingly, we identified a gene called SPINK1 that appears to be expressed only in gene fusion negative prostate cancer. Prostate cancers that had high levels of SPINK1 had a worse prognosis than those that did not. Thus, SPINK1 may represent a novel molecular subtype of prostate cancer that is distinct from those that harbor TMPRSS2-ETS gene fusions. SPINK1 has the potential to be utilized as a tissue marker to identify patients who require more aggressive treatments as opposed to watchful waiting. In addition, since we are able to detect SPINK1 in the urine of men with prostate cancer, it may be used in conjunction with urine gene fusion levels to develop a noninvasive screening test for prostate cancer. SPINK1 may also play a role in the development of gene fusion negative prostate cancer and thus could serve as a therapeutic target.