Recently, Myriad defined a threshold for active surveillance combining the CCP and CAPRA scores.21 Myriad validated this cutoff in 2 cohorts of men initially managed conservatively. Although the model predicts up to 3.2% disease-specific mortality, there were no observed deaths during a decade of follow-up. Myriad reports that by using this cutoff in contemporary patients tested commercially with Prolaris, a health care system could increase the percentage of men who would fit current criteria for active surveillance from 36% to 60% with no increase in risk of disease-specific mortality.
The results of these 2 tests are presented in 2 different formats and provide risk estimates for different clinical endpoints, making it challenging for a clinician to directly compare them. Moreover, each genomic test is based on a different set of genes and uses different clinical risk criteria (AUA vs NCCN), which may result in different test output. Finally, and most relevant to the case described here, there is no evidence-based consensus on how to interpret these test results in the context of a BRCA2 mutation.
Based on the published literature reporting that BRCA2 mutations are associated with more aggressive disease, one prediction would be that test scores from genomic assays such as Oncotype DX and Prolaris would tend to be higher in BRCA2 carriers than those of the overall population of PCa patients. This has, in fact, been reported for the Oncotype DX Breast Cancer Assay recurrence score in women who are BRCA carriers.22 Further research is required to ascertain whether this will be true for Oncotype DX GPS and Prolaris CCP score in PCa. The mechanism of action that predisposes BRCA2 mutation carriers to develop a more aggressive variant of PCa may not be detectable by the genomic markers included in the Oncotype DX PCa and Prolaris tests. The degree to which a mutated BRCA2 gene may interact with the genes comprising these assays and the reported tumor aggressiveness is not yet understood but deserving of future study.
Treatment Recommendation and Patient’s Decision
After considering his test results, the veteran chose active surveillance. The sum of clinical, pathologic, and molecular factors, combined with the patient’s preference, determined his course of treatment. Because prostatectomy was not performed, it has not been positively determined whether or not the patient harbors aggressive disease. As the molecular test results place the patient at the high end of the low-risk group, the VA urologist recommended close monitoring and suggested a follow-up biopsy with magnetic resonance-ultrasound fusion guidance.
Conclusions
Molecular testing found that the patient’s PCa stage and grade are consistent with NCCN low risk (Oncotype DX) and that the disease-specific mortality risk is slightly higher than predicted by clinical features alone (Prolaris). Previous studies have shown that molecular testing in men with PCa provides information that influences clinical decisions. The findings reported here suggest that molecular testing may also be a vital component in the medical management of patients with complex clinical phenotypes and common chronic conditions. Additional studies are necessary to evaluate whether the finding reported here is typical of individuals diagnosed with PCa who also have a BRCA2 mutation.
For any new genomic test to be clinically useful, its results must have clinical actionability. In this case, the clinical decision point was whether to recommend immediate definitive treatment or active surveillance. For this patient, the Oncotype DX assay provided a likelihood of favorable surgical pathology of 71% (or conversely a 29% risk of unfavorable pathology); by comparison, the Prolaris CCP score provided a 3% estimate of PCaspecific
mortality at 10 years. A key question is: How do clinicians perceive the actionability of risk estimates for these different endpoints?
The current case illustrates the challenges that rapidly developing genomic medicine pose for physicians trying to optimize care and communicate results to patients in a meaningful and consistent manner. For example, some urologists find the different 2-tiered calculations confusing. When laboratories use proprietary scalesbased on internally develop algorithms, differing interpretations are to be expected. The risk-assessment tests described here use different algorithms, and their interpretations are based on clinical categories from different sets of guidelines. This underscores the need for better standardization of PCa care.23
Oncology and urology professional associations should collaborate to develop consistent guidelines for use of new technologies in the management of PCa. A positive example is the evolution of testing recommendations in lung cancer, which initially varied between professional entities. In April 2013, the College of American Pathologists, the International Association for the Study of Lung Cancer, and the Association for Molecular Pathology jointly issued a unified clinical practice guideline on molecular testing in patients with lung cancer.24 In October 2014, the American Society of Clinical Oncology issued an endorsement of the CAP/IASLC/AMP guideline.25 As the number of complex tests being used in PCa increases, it will be important for professional associations such as AUA and NCCN to collaborate in evaluating utility of innovations to make consistent recommendations
Author disclosures
Myriad Genetics and Genomic Health provided funding for research on their tests within the VA. Dr. Dash and Dr. Lynch are principal investigators of the Genomic Health study. Dr. Lowrance is the principal investigator of the Myriad Genetics study.
Disclaimer
The opinions expressed herein are those of the authors and do not necessarily reflect those of Federal Practitioner, Frontline Medical Communications Inc., the U.S. Government, or any of its agencies. This article may discuss unlabeled or investigational use of certain drugs. Please review complete prescribing information for specific drugs or drug combinations—including indications, contraindications, warnings, and adverse effects—before administering pharmacologic therapy to patients.
