Scientists have identified a gene that appears to encourage the spread of prostate cancer by switching off a host of other genes. The finding, reported in the 10 October issue of Nature, could lead to a biomarker that would help physicians identify which afflicted men need the most aggressive treatment.
Prostate cancer is usually diagnosed by testing levels of prostate-specific antigen (PSA). But the PSA test is not highly specific to prostate cancer, and it's not very helpful at indicating whether a tumor is likely to become metastatic and spread to other organs. Researchers are searching for molecules that may be more useful for prognosis. Now a promising candidate has been discovered by team led by Arul Chinnaiyan, a clinical pathologist University of Michigan Medical School, Ann Arbor.
Chinnaiyan and colleagues have been using DNA microarrays to observe 20,000 genes that might be expressed in normal and cancerous prostate tissue. Last year they discovered that some 480 genes are turned off in metastatic prostate tumors. That was odd because growing tumors usually put their genetic machinery into overdrive. Intrigued, the team decided to examine another set of 55 genes that had become more active in tumors to see if one of them might be silencing the other genes.
When Chinnaiyan's team put the busiest of these genes--one called EZH2--into prostate cancer cells, 163 genes fell silent. And when the researchers examined 1023 prostate samples, they found the most EZH2 protein in metastatic tumor tissue, less in localized tumors, and the least in benign tissue. This suggested that EZH2 might indicate the level of tumor aggressiveness, Chinnaiyan says.
The hunch was borne out by 278 additional samples taken from 64 men at the time of diagnosis and then 5 years later. The amount of EZH2 did a better job of predicting whether a tumor would metastasize than standard methods, Chinnaiyan says.
"This is the type of marker that is desperately needed," says William Isaacs, a prostate cancer researcher at Johns Hopkins School of Medicine in Baltimore. "It looks as strong if not stronger than any other markers I've seen," he says, although he cautions that prognostic markers like this must be evaluated in larger studies. If EZH2 does hold up, it will probably be most useful in a panel of a half dozen or so biomarkers, notes cell biologist Bruce Zetter of Children's Hospital in Boston. The finding may also be important for explaining the mechanism of metastasis, Zetter notes.