Test accurately forecasts metastatic prostate cancer post-radical prostatectomy

A new prognostic test for prostate cancer outperformed existing risk assessment tools for predicting metastatic disease in patients following radical prostatectomy, researchers reported.

The Decipher test (GenomeDx Biosciences Inc., San Diego) more accurately classified prostate cancer patients at risk of developing metastasis compared to conventional clinical variables such as Gleason score and PSA. In the study, the test was able to pinpoint a high-risk patient group that was four times more likely to have metastatic cancer. Conversely, the Decipher test reclassified 60% of clinically high-risk patients as low risk, who were 2.5 times less likely to have metastatic cancer.

Study findings are in press in the Journal of Urology (published online June 12, 2013).

"Currently, men with advanced or high-risk prostate cancer have heterogenous outcomes, and this test can help better risk-stratify these men after surgery," said principal author R. Jeffrey Karnes, MD, of Mayo Clinic, Rochester, MN. "In this study, we found the test was able to re-classify from high-risk to low-risk 60% of patients, including high Gleason score patients that, upon clinical follow-up, had exceptionally good outcomes and never needed secondary therapy."

In the prospectively designed study, 1,010 patients treated with radical prostatectomy between 2000 and 2006 were identified from the Mayo Clinic tumor registry for a case-cohort study design. Genome-wide expression profiles were generated on a random sample of an at-risk population (219 patients, including 69 patients who eventually developed metastasis on study follow-up). Decipher results were generated from the genome-wide profiles from these patients in a blinded fashion.

Results showed that Decipher could independently forecast which patients developed metastasis and showed better performance over any clinical variable or prediction model for metastasis. Sixty percent of patients who had a low Decipher result also had a low (2.4%) 5-year cumulative incidence of metastasis. In contrast, 20% of patients with a high Decipher result had a 22.5% incidence of metastasis after 5 years.

The test measures 22 genomic biomarkers associated with metastatic cancer to generate a result that indicates the likelihood of metastasis. The result is independent of PSA and other existing clinical variables.

"The study is based on a large patient cohort with long-term follow-up, and Decipher is forecasting an important clinical endpoint: metastasis-free survival," said co-author Peter Black, MD, of the University of British Columbia, Vancouver. "Decipher is shown to be especially helpful in telling us which patients likely need no further therapy because they have a lower risk of progression. It makes me confident that this will be the test that raises the bar on the kind of information required to make a post-RP decision."

This study was partly funded by GenomeDx Biosciences, and several study authors are employees of the company. Dr. Black and a co-author are consultants for GenomeDx Biosciences.