Promising Future for Diagnosing Prostate Cancer
Promising Future for Diagnosing Prostate Cancer
Cleveland Plain Dealer
Mar 1, 2011
CLEVELAND, Ohio – Researchers at the Cleveland Clinic and California-based Genomic Health, which creates diagnostic tests for cancer, last week presented study results that they say have moved them closer to developing a diagnostic test for localized (nonspreading) prostate cancer.
The scientific team presented its findings last week at the annual meeting of the Society for Urologic Oncology.
The team analyzed multiple prostate cancer tumor biopsy specimens from more than 400 men who underwent surgery at the Clinic between 1987 and 2004 to remove the prostate gland and surrounding tissue. Researchers used the same gene analysis technology that Genomic Health developed for its Oncotype DX breast and colon cancer tests. They screened more than 700 prostate-cancer-related genes and honed in on 295 genes strongly associated with cancer recurrence.
"The exciting thing is that we have identified the gene expression signature of [these genes] that we think predicts [cancer recurrence]," said urologist Dr. Eric Klein, chairman of the Glickman Urological and Kidney Institute at the Clinic.
"The biggest challenge we face in managing prostate cancer currently is determining whether someone with a newly diagnosed tumor will die or not," Klein said. "The reason the test is important is that a lot of low-grade tumors are found in routine PSA screening, in perfectly healthy men."
The National Cancer Institute estimated more than 217,00 new cases of prostate cancer in 2010, and more than 32,000 deaths that same year in the United States. More than 70 percent of men diagnosed are over age 65.
Along with a digital rectal exam, prostate cancer is detected by a test that measures prostate-specific antigen, or PSA, levels in the blood. The higher the PSA level, the greater the chance that cancer is present in the prostate. The presence of cancer is confirmed by a biopsy.
A biopsy is ordered so that a physician can examine tissue to see if it is cancerous.
"If it shows cancer, we would test for the gene expression," Klein said. "On a repeat biopsy, we could run that same sort of test and see when it switches on to a more aggressive state."
Results of the test, which would take about a week to process at Genomic Health's labs, would be able to determine whether or not a newly diagnosed patient would need to be treated right away or not, Klein said.
PSA sometimes leads to needless treatment
The PSA test has long frustrated men and their physicians. Ever since it became available in 1987, the test's effectiveness at saving lives continues to be scrutinized. It is widely thought that countless men are being needlessly treated for prostate cancer that is so slow-growing that it will never cause any problems, much less death.
If it's determined that a man's cancer is not life-threatening he can avoid radiation therapy or surgery and common treatment-related side effects such as loss of sexual function and incontinence.
The Clinic and Genomic Health have partnered on prostate cancer research for about six years, Klein said. Genomic Health had worked with other Clinic researchers as it developed the breast cancer test.
"They were looking for an academic partner to look at prostate, lung, kidney and colon cancer," Klein said.
Because the Clinic already had a sizable volume of frozen prostate cancer tumors in its pathology archives and a database of patients that stretched back a dozen years, the Clinic was well positioned to tap into that information, to figure out – without identifying them – which patients did well on various treatments and which ones didn't, Klein said.
Right now, a patient doesn't find out if he has any or all of the 295 identified genes.
"It's a statistical model that we're still working on," Klein said. "In order to turn this into a clinically useful diagnostic test, you pick out the most informative gene."
The team is currently analyzing all of the genes in order to eventually narrow the number of genes to a more manageable list for use in a diagnostic gene test, said Genomic Health's chief medical officer, Steven Shak.
Another study based on existing biopsy samples, as well as obtaining and analyzing new samples, will take about 18 months to complete.
"After that, the idea would be to validate those observations at a different institution," Klein said. "Now we can accelerate it [with] new samples and new patients. We know what genes to look for."
A different approach in Boston study
The Clinic and Genomic Health are identifying other centers interested in collaborating on a larger clinical study, which could start at the end of this year or in early 2012, to validate the findings, Shak said, adding, "The goal is to have a test available in 2013."
A similar study, published online Feb. 2 in the journal Nature, also sought to effectively answer how best to figure out how to proceed following a man's prostate cancer diagnosis.
Researchers at the Dana-Farber Cancer Institute in Boston approached the question a bit differently.
Their study started with mice models and then shifted to data from a study that has tracked the health of thousands of American physicians for nearly 30 years. They also looked at 300 genes and honed in on four that, when absent, allowed for the progression of prostate cancer. The Boston researchers say they hope their findings will lead to the development of a clinical test within a year or so.
Both the Boston and the Clinic-Genomic Health studies identified the same gene, SMAD4, as among the genes highly significant in predicting aggressiveness of cancer.
"It's getting at the same question," Klein said.
"The nice thing about our approach [is that it] can easily [test] for that particular gene that we're looking for."
Does this mean competition?
"Competition is better relegated to the commercial realm," Klein said. "On the scientific level, I don't see it as competition."