The Decipher (GenomeDX Biosciences) test differentiates men at higher risk for prostate cancer metastasis after prostatectomy from those at lower risk and adds prognostic value in comparison with widely used clinical and pathologic risk factors, a new meta-analysis shows.
The Decipher classifier is "absolutely" ready for "prime time" or widespread use to help prostate cancer patients and clinicians decide about the need for additional treatment, such as radiotherapy, said lead author Daniel Spratt, MD, University of Michigan, in Ann Arbor, when asked by Medscape Medical News.
In addition to being a clinical decision-making aid, Decipher could be used in the design of future clinical trials, he added.
"Decipher has been used on I think about 20,000 men or so in the United States already, and it's been approved by Medicare for their molecular diagnostic program, and it's also listed as part of our national guidelines," Dr Spratt said.
The meta-analysis was published online March 30 in the Journal of Clinical Oncology.
However, in an accompanying editorial, Anthony D'Amico, MD, Brigham and Women's Hospital and the Dana-Farber Cancer Institute, Boston, Massachusetts, decidedly does not agree about the test's utility and its prime time status.
Decipher ― or any other genomic test ― cannot currently determine whether adjuvant treatment following surgery for prostate cancer should be given or withheld, he argues.
"The problem with a prognostic test is that it doesn't necessarily tell you that if you treat the patient with radiation, hormonal therapy, or anything else, that it will reduce the risk of future recurrence or death from prostate cancer," Dr D'Amico told Medscape Medical News.
"A prognostic test like Decipher just tells you that a patient is at high risk for developing metastatic disease, but it doesn't tell you what needs to be done to reduce or avoid this risk," he underscored.
What oncologists really need is a predictive test ― in other words, a test that has been shown to identify people whose risk of developing metastatic prostate cancer will be reduced through adjuvant treatment. Such factors can only be determined in the context of a randomized controlled trial, Dr D'Amico noted.
Meta-analysis: Metastasis Risks at 5 and 10 Years
Decipher is a genomic classifier that indicates which men are most likely to develop metastatic disease following prostatectomy.
"The main reason why we want to predict which patients are most at risk for metastatic disease following surgery is because after surgery, there are multiple approaches that can be taken, including doing nothing, where you just observe the patient, or we can give adjuvant radiation therapy or radiation with potentially hormonal therapy, and there are even ongoing trials of potentially adding chemotherapy for men at very high risk of recurrence," said Dr Spratt.
"[But since] we have very limited predictors to tell us how aggressive a man's disease is, a genomic classifier can help us stratify patients to receive potentially more or less treatment in a more personalized manner," he added.
"And what this study shows is that when you take every factor we know of that predicts a man's recurrence ― their margin status, their tumor stage, their Gleason score, their prostate-specific antigen [PSA] score, their lymph node status — that even when you use everything available to us, a single genomic test outperforms all of them and provides independent prognostic value," Dr Spratt said.
Five studies involving 975 patients were included in the analysis. Patient-level data were available for 855 of those patients. The median follow-up was 8 years.
"During the study, 82 patients experienced metastasis," the investigators report, "and cumulative incidence curves demonstrated that Decipher categories significantly stratified risk of metastasis (P < .001)," they add.
At 5 years, the cumulative incidence of metastasis was 2.4% for patients whose Decipher scores indicated low risk. For those whose score indicated intermediate risk, the cumulative incidence of metastasis at 5 years was 5.8%, and for high-risk-scoring patients, the cumulative incidence of metastasis at 5 years was 15.2%.
At 10 years, cumulative incidence rates for metastatic disease were 5.5%, 15%, and 26.7% for men whose scores on Decipher indicated low, intermediate, and high risk, respectively, the investigators add.
When compared to clinicopathologic features, the investigators found only low to moderate correlation of Decipher scores with the Gleason score, extracapsular extension, and seminal vesical or lymph node invasion. Decipher did not significantly correlate with either preoperative PSA level or surgical margin status.
On the other hand, patients who scored as high-risk on Decipher had almost a 3.5-fold higher risk for metastasis, at a hazard ratio (HR) of 3.31, compared with patients who had a Gleason score of 8 to 10, at an HR of 3.23 (P < .001).
"We often view the Gleason score as the gold standard to help tell us the risk of recurrence, the aggressiveness of a man's disease," observed Dr Spratt.
"But in reality, the Gleason score has an AUC [area under the curve] of only about 0.6, so while it's better than flipping a coin, it's not that great," he added.
"And while we showed that as the Gleason score gets more aggressive, so does the genomic score. Our test has an AUC of closer to 0.7 or 0.75, so it dramatically outperforms the Gleason score," Dr Spratt elaborated.
Randomized Trial on the Way
"If the Decipher or any other genomic test says you're at high risk of a certain endpoint — and in terms of the Decipher test, it's metastatic disease after surgery ― that is not what's up for discussion. What needs to be determined is whether these tests can tell you who benefits from adjuvant therapy and who does not," Dr D'Amico further commented in his interview with Medscape Medical News.
Fortunately, the Decipher test is being subject to such a clinical trial to determine just that. In the study, men will be tested for their risk of developing metastatic disease at baseline, as determined by their Decipher score, and then they will be randomly assigned to receive either standard of care or standard of care plus something else.
"If that something else actually reduces the risk of metastatic disease, and the high-risk Decipher score can identify these men, then you've shown that Decipher is predictive, and only then could a high Decipher score (ie, >60) be used to recommend additional treatment proven in a clinical trial to reduce distant metastasis," Dr D'Amico noted.
"But until we have that information, here's the concern: Some physicians may be withholding treatment from men with high-risk clinical features because their Decipher score is low but in whom adjuvant radiation therapy has been proven to reduce the risk of recurrence, and given that some patients are looking for a way not to get treated after surgery, they are willing to use their Decipher score as a way to do that," he continued.
Conversely, if a man has low-risk clinical features and would not be considered for any further therapy, yet their Decipher score is high, should they be subjugated to further treatment simply on the basis of a high-risk Decipher score, especially when it has not been proven that treatment will reduce the risk for metastasis in these men?
"We shouldn't be using Decipher or any other similar test to guide treatment recommendations, given that treatment has potential side effects, nor should we be removing potentially curative treatments in settings where they are known to have benefit, until we have proven that these tests are able to identify men in whom additional therapy improves cancer control," he concluded.
The study was funded in part by the Prostate Cancer Foundation Young Investigator Award. Dr Spratt has disclosed no relevant financial relationships. Senior author Felix Feng, MD, has a leadership role with PFS Genomics and a consulting or advisory role with Medivation/Astellas, GenomeDx, Biosciences, Dendreon, Sanofi, and EMD Serono. Dr Feng also receives research funding from Varian Medical Systems and Celgene. Dr D'Amico has disclosed no relevant financial relationships.
J Clin Oncol. 2017;35:1991-1998. Abstract, Editorial
Follow Medscape Oncology on Twitter for more cancer news: @MedscapeOnc
Medscape Medical News © 2017
Cite this: To What End? Decipher Tells Prostate Cancer Metastasis Risk - Medscape - Jun 22, 2017.
Comments