Researchers have identified several genes that are consistently merged, or fused, in some prostate tumors and could potentially be used to detect the disease. The discovery is the first example of gene rearrangements recurring in a solid tumor, although such changes are a hallmark of some blood cancers.

The findings, reported in the October 28 Science, suggest that prostate cancer is not a special case and that other common cancers such as lung, breast, and colon may involve recurrent gene rearrangements. The study was completed in less than 4 months, and the initial results surprised even the researchers themselves.

"We were surprised because these types of gene rearrangements have been associated with leukemia and lymphoma but not with solid tumors," says Dr. Arul Chinnaiyan of the University of Michigan Medical School, who led the study. "To find this change in a majority of prostate cancers suggests that it is important in the disease."

The researchers estimate that between 60 and 80 percent of prostate cancers have the rearrangement. They are developing techniques to detect the change in urine and blood.

When the rearrangement occurs, one of two cancer genes, ETV1 or ERG, fuses with part of another gene, TMPRSS2. As a result of this fusion, the fused genes, which control other genes, become regulated by the hormone androgen and are at risk of stimulating too much genetic activity in the tumor cell.

"This is fantastic work," comments Dr. William Isaacs, professor of urology and oncology at Johns Hopkins University School of Medicine. "The results need to be independently replicated, but I have every reason to think this will happen rapidly."

The rearrangement may have gone undetected until now because solid tumors involve an overwhelming number of nonspecific, random aberrations.

To address this problem, two graduate students in Dr. Chinnaiyan's laboratory, Scott Tomlins and Daniel Rhodes, developed an algorithm that sifts through data on gene activity to find genes that are highly active in subsets of tumors.

Using the algorithm, called Cancer Outlier Profile Analysis, the team determined that ETV1 and ERG were highly active in some prostate tumors.

Further study revealed that one but not both of these genes frequently fuses with TMPRSS2 in prostate tumors. "This was a clue that the rearrangement played an important role in the development of prostate cancer," says Mr. Tomlins, noting that single fusion events typically cause some types of blood cancer.

Drugs could potentially be developed to inhibit the mutant genes, although this could take years. The drug imatinib (Gleevec), for instance, targets the gene fusion that causes chronic myelogenous leukemia.

"There are profound implications for diagnosis and treatment if it can be shown that this rearrangement occurs at the earliest stages of prostate cancer," says Dr. Sudhir Srivastava, chief of NCI's Cancer Biomarkers Research Program and director of the Early Detection Research Network, one of the NCI programs supporting the study.

The study does not demonstrate cause and effect, but "we know from other diseases that gene rearrangements are one of the major mechanisms in cancer," says Dr. Jacob Kagan, program director of NCI's Cancer Biomarkers Research Group. "We would now expect that there would be recurrent gene rearrangements in other common cancers as well."

By Edward R. Winstead