Genomics of Prostate Cancer

Principal Investigators - Colin C. Collins, PhD | Peter R. Carroll, MD

This project is fortunate to have pathologists, epidemiologists, computer scientists, physicians, geneticists, and experts in the human genome and cell biology all working together to solve a vexing problem in prostate cancer.

Accurate identification of prostate tumors that will progress remains problematic. Approximately one half of tumors from patients at intermediate risk of progression will progress to metastatic disease, while half will not. Additionally, a significant fraction of tumors from high-risk patients do not recur. The hypothesis guiding this project is that changes in chromosome structure can be identified that will allow genetic classification of tumors into progressors and non-progressors and thus aid in treatment decisions. Further, genes identified in these chromosomal regions may be evaluated as therapeutic targets.

To test this hypothesis, we obtained a cohort of 100 prostate tumors comprised mostly of tumors from intermediate risk patients. The patients have up to 16 years of follow-up, and half progressed following surgery. A second cohort of high- and low-grade tumors has also been collected for analysis. We are employing a new technology called array comparative genomic hybridization (aCGH) to identify chromosome changes that can be used to classify prostate tumors. In essence, tumor DNA is applied to a biochip containing approximately 3,000 carefully selected pieces of the human genome so that we can map the pathologic topography of the human genome at very high resolution. The genomic location of each piece of DNA on the chip is known due to the human genome project. Thus, by dissecting tumors, and applying the DNA to these biochips, we are able to identify chromosome aberrations that correlate with progression. Using software developed in our lab, we then link the biochips to the underlying human genome sequence to identify genes involved in tumor progression. We also integrate these topographical maps with expression array data linking changes in chromosome structure to alterations in gene expression -- localizing the criminal to the crime scene.

To date we have identified numerous chromosomal changes that correlate with progression and identified one that may predict progression independent of tumor stage and grade, and we have identified numerous genes that may be involved in driving tumor progression, some of which may make excellent drug targets. Future work will be aimed at confirming and significantly extending these findings using CGH, gene expression chip, and epidemiology in additional intermediate/high risk patients, as well as in metastatic tumors. Further, we will explore the possibility of using our findings to define how lifestyle habits influence development of prostate cancer, and whether it can modulate progression in men with the disease using our unique resources and capabilities.