Scientists at the University of Michigan Medical School, in collaboration with researchers at Harvard's Brigham and Women's Hospital, have discovered a recurring pattern of scrambled chromosomes and abnormal gene activity that occurs only in prostate cancer.
In a paper being published in the Oct. 28 issue of Science, the research team indicates that these chromosomal rearrangements induce specific genes to merge, creating what scientists call a gene fusion. U-M researchers detected the unique molecular signature of these fused genes in the majority of prostate cancer tissue samples they analyzed, but found no evidence of gene fusion in benign prostate tissue or in prostate tissue with non-cancerous changes.
"The data in our study provides tantalizing evidence that gene fusion is the causative agent - the initiating event - in prostate cancer," says Arul M. Chinnaiyan, M.D., Ph.D., the S.P. Hicks Collegiate Professor of Pathology in the U-M Medical School, who directed the study. "It's what drives the aberrant over-expression of cancer-causing genes and is the first step in the progression of tissue changes leading to prostate cancer."
Because this particular gene fusion occurs only in prostate cancer, a diagnostic test to detect, in blood or urine, the fused genes or their protein products would be specific for prostate cancer and far more accurate than current screening tests, according to Chinnaiyan. And if scientists could find a way to block the gene, it could be the basis for a new, effective treatment for prostate cancer.
According to the American Cancer Society, more American men will be diagnosed with prostate cancer this year than with any other type of cancer. The American Cancer Society estimates that, in 2005, 232,000 men in the United States will be diagnosed with prostate cancer and 30,350 men will die from the disease. It is the second most common cause of cancer-related deaths in men.
"Studying gene alterations in prostate cancer is difficult, and as a result there has never been a clear identification of recurrent, non-random genetic rearrangements," says Jacob Kagan, Ph.D., program director for the Cancer Biomarkers Research Group at the National Cancer Institute, a sponsor of the research study. "This finding is an important advance, because it suggests that similar mechanisms may be involved in other epithelial cancers, such as breast, lung and colon."
The abnormal gene fusion associated with prostate cancer occurs when one of two genes, ERG or ETV1, merges with a prostate-specific gene called TMPRSS2. ERG and ETV1 are members of the ETS family of transcription factors, which are known to be involved in the development of a bone cancer called Ewing's sarcoma, and other types of cancer.
While rearrangements in chromosomes and fused genes have been detected in blood cell cancers like leukemia and lymphoma, and in Ewing's sarcoma, this is the first time they have been found in a common solid tumor like prostate cancer, which develops in epithelial cells lining the prostate gland.
"This is a paradigm shift for all epithelial tumors - such as cancers of the lung, breast, colon, ovary, liver and prostate - which are the most common types of cancer and account for most deaths due to cancer," says Chinnaiyan, who directs the Bioinformatics Core at the U-M Comprehensive Cancer Center. "We knew gene rearrangements were involved in hematologic malignancies and sarcomas. But finding this recurrent chromosomal rearrangement in prostate cancer suggests that other common epithelial cancers have their own recurrent chromosomal rearrangements. We just haven't found them yet."
A bioinformatics analysis method called Cancer Outlier Profile Analysis (COPA) developed by Scott A. Tomlins and Daniel R. Rhodes, U-M graduate students working in Chinnaiyan's laboratory, made it possible for the research team to detect extremely high expression levels of outlier genes, including ERG and ETV1, in 132 gene expression microarray datasets and six independent prostate cancer profiling studies.
U-M scientists also used laboratory analysis techniques and gene sequencing to detect gene fusions between TMPRSS2 and ERG or ETV1 in prostate cancer tissue samples. "Of 22 cases with high expression of ERG or ETV1, 91 percent showed evidence of a fusion with TMPRSS2," said Scott Tomlins, a U-M graduate student in pathology and first author of the Science paper. "Our results indicate that more than half of all prostate cancer cases have one of these two fusions."
Tomlins adds that TMPRSS2 may fuse to other members of the ETS family, in addition to ERG and ETV1. "We think there are more fusions. We just haven't found them yet," he says. "To develop a diagnostic test with full sensitivity, it will be important to identify all the molecular fusions, and we are working on that now."
"We are especially excited by the profound implications these findings have for the treatment of prostate cancer," Chinnaiyan says. "It will allow us to categorize prostate tumors by molecular sub-type, which could help determine the most effective treatment for each patient."
"This collaborative effort has yielded an important molecular biomarker that will help us better detect prostate cancer and ultimately help clinicians determine the risk of dying from the disease," says Mark A. Rubin, M.D., chief of urologic pathology at Brigham and Women's Hospital, who is also an associate professor of pathology at Harvard Medical School and a co-author on the paper.
In future research, U-M scientists hope to identify small molecule inhibitors for the genes involved in prostate cancer, which would be analogous to Gleevec - a drug designed to target the BCR-ABL gene fusion that causes chronic myelogenous leukemia. The U-M research team plans additional studies to verify that gene fusion induces prostate cancer in research animals. And they will search for recurrent chromosomal rearrangements in other common epithelial cancers.
The research was supported by the National Cancer Institute's Early Detection Research Network, the National Cancer Institute's Specialized Program of Research Excellence in Prostate Cancer, the American Cancer Society, the Department of Defense, the U-M Comprehensive Cancer Center Bioinformatics Core and the U-M Medical Scientist Training Program.
The University of Michigan has filed a provisional patent application on this research technology.
Additional research collaborators from U-M include research fellows Saravana Dhanasekaran, Ph.D., Rohit Mehra, M.D., and Sooryanarayana Varambally, Ph.D.; Xuhong Cao, research associate; James Montie, M.D., the Valassis Professor of Urological Oncology; Rajal B.Shah, M.D., clinical assistant professor of pathology; and Kenneth J. Pienta, M.D., professor of internal medicine and urology. Additional collaborators from Brigham and Women's Hospital include Sven Perner, Xiao-Wei Sun, Joelle Tchinda and Charles Lee. Rainer Kuefer from the University of Ulm in Germany also collaborated in the research study.