A genomic test to distinguish between clinically indolent and aggressive prostate cancer may soon become a reality, suggests a study presented at the Genitourinary Cancers Symposium held in Orlando, Florida, in February.¹ Once the target genes are identified and the test is validated, it could be used to identify patients who can safely opt for active surveillance and those who should undergo primary treatment with surgery or radiation. The test is being developed by Genomic Health Inc—the same company that developed the Oncotype DX assays for breast cancer and colon cancer, two commercially available tests performed by the Genomic Health laboratory, which is regulated and certified under the Clinical Laboratory Improvement Amendments (CLIA) program of the Centers for Medicare & Medicaid Services.

Prognostic Uncertainty and Treatment Choice

Developing a genomic test for prostate cancer that predicts need for therapy is important, explained Eric A. Klein, MD, Chairman and Professor of Surgery at the Glickman Urological and Kidney Institute, Cleveland Clinic. Dr. Klein was first author of a prospectively designed study to measure gene expression in prostate cancer specimens obtained from radical prostatectomy and to link gene expression to outcomes.

“The biggest challenge facing men with newly diagnosed prostate cancer is uncertainty about the biologic potential of their disease,” Dr. Klein said. “About 90% of men with newly diagnosed low-grade cancer opt for treatment with radical prostatectomy or radiation therapy, but many of them have indolent disease and could be spared primary therapy.”

 Dr. Klein pointed out that autopsy studies show that a substantial number of older men who die have clinically silent prostate cancer and die of other causes. These men have indolent disease and don’t need treatment, he emphasized.

The study he presented at a poster session was based on 501 archived tumor samples from men who had radical prostatectomy at the Cleveland Clinic between 1987 and 2004 analyzed so as to be representative of all 2,600 men operated on during this period. Of the 501 samples, 127 had clinical recurrence and 374 had no recurrence during a long follow-up interval. A total of 51 tumor samples were excluded for insufficient tissue. The expression pattern of 732 genes previously suggested to be important drivers of prostate cancer progression were tested in these samples using an RT-PCR–based technique. The results indicated that the expression signatures of about 300 genes had the ability to predict whether or not a patient’s disease had recurred, Dr. Klein explained.

Another important finding is that the genes associated with clinical recurrence appear to be similar in both the predominant and the highest Gleason pattern from an individual patient’s biopsy, indicating that sampling tissues on biopsy would not limit the usefulness of the test, Dr. Klein noted.

Evolving Test

“This was a wonderful marriage of tissue samples and quantitative gene expression linked to outcomes. We started with 732 genes, analyzing the results to identify the smaller number of genes that gives us the best test performance. We expect to end up with a range of 10 to 25 genes,” said Steven Shak, MD, Chief Medical Officer of Genomic Health.

Once the final number of genes needed for the test is determined, future studies will be done to validate the test. The study reported at the Genitourinary Cancers Symposium lays the groundwork for determining genes that are predictive for important clinical outcomes in prostate cancer, Dr. Klein said.

“There is a tight correlation between gene expression and recurrence,” Dr. Klein emphasized. “The gene expression patterns we observed reveal an underlying biology that overcomes apparent heterogeneity by Gleason grading on biopsy,” he added.

Another important observation was that genes predicting clinical recurrence are involved in multiple pathways that drive prostate cancer metastasis.

This study took almost 5 years to conduct, reflecting the difficulty of doing a genomic study in prostate cancer. “This study provides new biologic information that may help us better understand the biologic mechanisms that drive prostate cancer growth, and identify which genes or pathways are appropriate targets for developing therapies,” Dr. Klein explained.

The study was supported by Genomic Health.

Tempered Enthusiasm

“Many people are working hard on developing gene signatures for prostate cancer, but I think the systems we have are pretty good. That includes using devices such as the Kattan nomogram, the Partin nomogram, and the Memorial Sloan-Kettering Cancer Center risk assessment that use clinical and laboratory measurements to determine whether a newly diagnosed patient has aggressive disease,” explained Oliver Sartor, MD, Director of the Prostate Cancer Program at Tulane University.

“To convince me that a genomic test is useful, the investigators would have to compare the test to the best nomograms that utilize readily available clinical and laboratory parameters,” he said. “Show me the data that this test is superior at risk stratification compared to nomograms as they are used at centers of excellence.”

A genomic test is a good step forward, said Nicholas Vogelzang, MD, US Oncology, Las Vegas. According to Dr. Vogelzang, even if a test can identify which prostate cancers are more aggressive, the “tool chest” has only a limited number of treatments for nonmetastatic prostate cancer. “We don’t have many options for these patients,” he stated.

“A test like this should be linked to improved outcome,” Dr. Vogelzang said. “On the other hand, it could spare people from needless therapy and overtreatment.” ■

Financial Disclosure: Dr. Klein disclosed direct research support from Genomic Health to support tissue processing and data extraction. Dr. Sartor and Dr. Vogelzang reported no potential conflicts of interest. Dr. Shak is the Chief Medical Officer of Genomic Health.

Reference

1. Klein EA, Falzarano SM, Maddala T, et al: Use of quantitative gene expression in primary and highest Gleason pattern cancers to identify genes associated with clinical recurrence after radical prostatectomy. Genitourinary Cancers Symposium. Abstract 39. Presented February 17, 2011.