Clinicians may soon be able to improve the risk stratification of men with prostate cancer with the help of a genomic classifier or a biopsy-based reverse transcription polymerase chain reaction assay, according to the results of two studies presented at the Genitourinary Cancers Symposium in Orlando, FL.
Clinicians may soon be able to improve the risk stratification of men with prostate cancer with the help of a genomic classifier or a biopsy-based reverse transcription polymerase chain reaction (RT-PCR) assay, according to the results of two studies presented at the Genitourinary Cancers Symposium in Orlando, FL.
In the first study,presented by Paul L. Nguyen, MD, the genomic classifier Decipher was able to accurately predict metastasis and prostate-cancer specific mortality (PCSM) from a needle biopsy in men with intermediate- and high-risk prostate cancer.
“The Decipher classifier obtained from biopsy samples was associated with distant metastases and prostate-cancer specific mortality after radical prostatectomy or radiation therapy and a short course of hormone therapy. It provided significant prognostic information beyond clinical variables alone,” said Dr. Nguyen, associate professor of radiation oncology at Harvard Medical School, Boston.
Decipher is a 22-gene RNA-based signature developed to predict risk of metastasis after radical prostatectomy. Here, Dr. Nguyen and co-authors conducted a multi-institutional study of Decipher aimed at evaluating whether it could predict metastasis and prostate cancer-specific mortality earlier in the treatment process using needle biopsy tumor specimens.
The study included 175 patients who were treated with first-line radical prostatectomy (43%) or radiation therapy plus androgen deprivation therapy (57%). All patients had genomic expression profiles available.
The majority of the men in the study (85%) had intermediate- or high-risk disease. With a median follow-up of 6 years, 32 patients developed metastases and 11 patients died of prostate cancer.
Next: Test stratifies for risk of metastasis
Results from the study showed that the Decipher classifier scores did stratify for risk of metastasis. The risk for 5-year metastasis by Decipher score was 23.4% in high-risk patients, 9.3% in intermediate-risk patients, and 5.0% in low-risk patients.
“This multivariable model shows that the classifier does add valuable prognostic information to the clinical variables when predicting for distant metastases even when you adjust for PSA, Gleason score T category, and treatment type,” Dr. Nguyen said.
In both the univariable and multivariable analyses, the Decipher score was the only variable that was significant for distant metastases when accounting for the other clinical variables.
The Decipher score was also found to improve the C-index when predicting for distant metastases. Using National Comprehensive Cancer Network risk groups, the C-index was 0.66; however, when using NCCN risk groups added to the Decipher score, the C-index was 0.75.
“It enhances our ability to decide which patient are going to develop metastases,” Dr. Nguyen said.
Finally, Decipher also predicted for PCSM (HR=1.57 per 10% increase in Decipher score; 95% CI: 1.07-.240; p=.02). The risk for 5-year prostate cancer-specific mortality by Decipher score was 9.4% in high-risk patients, 0% in intermediate-risk patients, and 0% in low-risk patients.
Next: GPS evaluated as predictor of BCR
The second study examined the use of the Oncotype DX prostate cancer test as a predictor of biochemical recurrence in men with prostate cancer treated with radical prostatectomy in a community setting. The test’s results are provided as a 17-gene Genomic Prostate Score (GPS).
The biopsy-based RT-PCR assay had previously been validated as an independent predictor of adverse surgical pathology and biochemical recurrence in men with low-risk and low-volume disease. The updated data, presented by Stephen K. Van Den Eeden, PhD, a senior investigator at Kaiser Permanente Northern California, Oakland, showed that GPS was also associated with biochemical recurrence independent of other clinical factors in surgically treated men with prostate cancer.
“The study confirmed that increases in the GPS from biopsy samples are associated with an increased risk of biochemical occurrence among men who were treated with surgery,” Dr. Van Den Eeden told Urology Times.
The study included men treated with radical prostatectomy and diagnosed with NCCN very low-, low-, intermediate-, and high-risk prostate cancer between 1995 and 2010 in the Kaiser Permanente Northern California network. Archival biopsy tissue was assayed to yield a GPS (scale 0-100).
Tumor tissue was retrieved for 334 patients, 279 met all the study eligibility requirements, and 259 generated a valid GPS. There were 117 biochemical recurrence events.
Data from a univariable analysis showed that the GPS was strongly associated with biochemical recurrence (hazard ratio per 20 GPS units=2.5; p<.0001). After adjusting for PSA, clinical T stage, and central biopsy Gleason score, the hazard ratio was 2.1 (p=.002).
“The information from the GPS for the properly targeted patient can provide additional information regarding treatment decisions,” Dr. Van Den Eeden said. “The GPS, if used, needs to be part of the overall picture that takes into account the clinical factors already available, such as the diagnostic PSA level, Gleason score, and stage.”
Dr. Nguyen is a consultant/adviser for Ferring, GenomeDx, and Medivation and receives research funding from Astellas Pharma. Several of his co-authors are consultant/advisers for or are employees of GenomeDx, and several have a financial or other relationship with pharmaceutical companies. Dr. Van Den Eeden has received research funding from Abbott Molecular, Genomic Health, and Takeda. Several of his co-authors are employees of Genomic Health. One of his co-authors has received research funding from several pharmaceutical companies.
Subscribe to Urology Times to get monthly news from the leading news source for urologists.