Laird Harrison

May 08, 2013

SAN DIEGO, California — A genetic test can predict the aggressiveness of prostate cancer, a new study shows.

The 17-gene test predicts whether tumors will progress to high-grade or postsurgical stage III disease.

"We think this will allow patients and doctors to make more confident decisions," study coauthor Eric Klein, MD, told Medscape Medical News. "That will be a boon to everyone and it will save the system money."

The results of the study were presented here at the American Urological Association 2013 Annual Scientific Meeting.

The researchers used reverse-transcription polymerase chain reaction to quantify gene expression from manually dissected prostate cancer tissue.

They began with 732 candidate genes from 441 patients obtained after radical prostatectomy, and calculated the association between these genes and the occurrence of high-grade or stage III (pT3) cancer.

They identified 288 genes predictive of clinical recurrence, regardless of Gleason score patterns.

From these 288 genes, they quantified 81 in a needle biopsy study of 167 patients. This confirmed the strong association between the genes and adverse pathology.

They used multivariate analysis to narrow the genes of interest to 17 from multiple biologic pathways, including stromal response, cellular organization, androgen, proliferation, and reference genes.

We think this will allow patients and doctors to make more confident decisions.

The researchers then conducted a validation study to confirm the predictive value of their 17-gene test. They assessed biopsies from 395 patients suitable for active surveillance with tumors as small as 1 mm.

They found that they could predict high-grade and/or pT3 disease independent of Cancer of the Prostate Risk Assessment (CAPRA) or other standard pretreatment factors (P = .002).

The test enabled them to increase the number of patients in their study population identified as very low risk — and therefore suitable for active surveillance — from less than 10% to 26%.

In other words, they were able to reclassify more than a third of patients, originally classified as low risk on the basis of clinical factors, as very low risk. This enabled patients to pursue active surveillance.

The researchers also identified aggressive disease in need of immediate treatment in about 10% of patients who had been classified as low risk or very low risk.

There are already some gene tests for prostate cancer, but they are only useful for determining the risk for cancer recurrence in patients who have already been treated for prostate cancer.

Our 17-gene test "is useful in patients who have been newly diagnosed with prostate cancer," said Dr. Klein. "Right now, we don't have good tools to distinguish cancers that are going to do well from those that will need treatment."

Genomics Health began selling the 17-gene test, known as the Oncotype Dx Genomic Prostate Score, today. It will cost $3820 per patient, said Steven Shak, MD, chief medical officer of Genomic Health.

He said he expects insurance to pay for the test because similar Oncotype Dx tests for use in breast and colon cancer are now reimbursed. He argued that it could save money by reducing unnecessary treatment.

"We expect it to be cost-effective," he said. "It should triple the number who choose active surveillance."

This test is one of several promising new technologies for determining which prostate cancers pose the greatest risk, said Jeffrey Hahn Reese, MD, a clinical professor of urology at Stanford University in Palo Alto, California, who was not involved in the study.

"It's going to have to be validated at a number of institutions," he told Medscape Medical News. "What we have seen in the past is that some of these fly and some of them don't. I think you're going to see a lot more of these because it's what we need."

This study was funded by Genomics Health. Dr. Klein is a paid consultant to the company and to other companies developing genetic tests for prostate cancer. Dr. Reese has disclosed no relevant financial relationships.

American Urological Association (AUA) 2013 Annual Scientific Meeting. Abstract 2131. Presented May 8, 2013.