Genetic Tests for Prostate Cancer

Prostate Cancer

Prostate Cancer continued Genetic Tests for Prostate Cancer Reviewed by Adam Kern, MD, Alan W. Partin, MD, PhD The Brady Urological Institute, Johns Hopkins Hospital, Baltimore, MD [Rev Urol. 2013;15(4):208-209 doi:10.3909/riu0597] © 2014 MedReviews®, LLC R ecent attention has focused on risk stratification for men for the early detection of prostate cancer insofar as triaging which men should undergo initial prostate-specific antigen (PSA) testing. However, risk stratification of men with biopsy-proven prostate cancer also remains an emerging field. Several nomograms exist to predict surgical pathology or the subsequent risk of posttreatment biochemical recurrence, including the Cancer of the Prostate Risk Assessment Postsurgical score (CAPRA-S), Memorial Sloan-Kettering PreTreatment Nomogram, and the Partin and Han tables. The rapidly declining cost of genetic analysis using offthe-shelf technology has ushered in a new generation of commercially available genetic assays that further delineate pre- and post-treatment prostate cancer risk for men with biopsy-proven disease. Here we review evidence validating three new genetic assays for prostate cancer that are recently US Food and Drug Administration (FDA) approved or pending approval. Validation of a Cell-cycle Progression Gene Panel to Improve Risk Stratification in a Contemporary Prostatectomy Cohort Cooperberg MR, Simko JP, Cowan JE, et al. J Clin Oncol. 2013;31:1428-1434. Cooperberg and colleagues validated a previously described genetic risk score based on quantification of cell cycle progression (CCP), in a cohort of patients’ status post-radical prostatectomy (RP). CCP is calculated as a function of gene expression of 31 CCP marker genes relative to 15 housekeeping control genes. This assay is now FDA approved and marketed under the trade name Prolaris by Myriad Genetics (Salt Lake City, UT). Recurrence was defined as two PSA levels $ 0.2 ng/mL or any salvage treatment. The CCP score was assessed for prognostic value beyond that of standard postoperative risk assessment using the CAPRA-S. In a cohort of 413 men, 82 (19.9%) experienced a recurrence. Adjusting for CAPRA-S, the hazard ratio (HR) for each unit increase in CCP score was 1.7 (95% confidence interval, 1.3-2.4). The authors also found that the combined CAPRA-S + CCP score consistently predicted outcomes across the range of clinical risk—including men with low-risk disease—and had superior performance to either individual score alone. Interestingly, at its extreme, CCP score remained highly predictive of clinical outcome regardless of CAPRA clinical risk group. No man with a low CCP score , 21 experienced a biochemical recurrence (BCR) during a 5-year period. Conversely, men with CCP score . 1.5% experienced BCR across all CAPRA risk subsets, including men with CAPRA-defined low-risk disease. The authors concluded that the independent CCP score serves as a predictor of posttreatment prostate cancer recurrence, and that models incorporating both CCP and CAPRA-S offer enhanced prognostic capability. Development and Validation of the Biopsy-based Genomic Prostate Score as a Predictor of High Grade or Extracapsular Prostate Cancer to Improve Patient Selection for Active Surveillance Cooperberg MR, Simko J, Falzarano S et al. American Urological Association Annual Meeting 2013, San Diego, CA. Abstract 2131. In this study also by Cooperberg and colleagues, a biopsy-based genomic prostate score (GPS) scheme was analyzed to assess its utility in predicting high-risk extracapsular prostate cancer in order to improve patient selection for active surveillance (AS) based on biopsy result. GPS is calculated based on reverse transcription polymerase chain reaction quantification of a panel of 17 prostate cancer-associated genes in the biopsy specimen. Genomic Health Inc. (Redwood City, CA) is currently developing the GPS assay under the trade name Oncotype DX®, which is pending FDA approval. The authors validated the 17-gene GPS panel using both RP specimens and needle biopsy specimens. During initial investigation, a panel of 732 candidate genes was narrowed to 288 genes potentially predictive of clinical recurrence, based on final RP pathology results. A subset of 81 of these 288 genes was carried forward into analysis using prostate biopsy cores from low/intermediate pretreatment risk patients. Multivariate analysis combining both populations parsed out 17 genes from multiple biologic pathways that were highly associated with highgrade and/or pT3 disease, forming the basis for GPS. 208 • Vol. 15 No. 4 • 2013 • Reviews in Urology 4004170006_RIULitra.indd 208 16/01/14 5:12 PM Prostate Cancer The results of this initial validation study are encouraging. The 17-gene GPS may be used in the pretreatment paradigm to predict high-risk disease based on biopsy tissue alone, and may prove particularly useful in identifying suitable candidates for AS. Validation of a Genomic Classifier That Predicts Metastasis Following Radical Prostatectomy in an At-risk Patient Population Karnes RJ, Bergstralh EJ, Davicioni E, et al. J Urol. 2013;190:2047-2053. This study aimed to develop a method to directly predict actual risk of prostate cancer metastasis based on tissue analysis of RP specimens. Men diagnosed with high-risk prostate cancer based on pretreatment clinical nomograms have heterogeneous outcomes, with imperfect correlation between ultimate tumor metastatic aggressiveness and observed clinical pathology after RP. Karnes and colleagues performed a validation study of a genomic classifier (GC) panel of 22 genes analyzed in a case-cohort study of 1010 post-RP specimens. The GC system is being developed by GenomeDx Biosciences (Vancouver, BC) under the trade name Decipher™. All patients were preoperatively assessed to be high risk based on either PSA . 20 ng/mL, Gleason $ 8, pT3b, or Gleason, PSA, seminal vesicle, and margin status (GPSM) score $ 10. GC scores were calculated for a subset of 219 patients and receiver-operating characteristics were calculated for GC assay performance. GC had an area under the curve of 0.79 for predicting 5-year metastasis post-RP, outperforming comparative clinical-only models of disease metastasis. Importantly, among patients with Gleason 7 tumors, 41% had GC $ 0.4 and 44% of these men identified as high-risk on GC ultimately developed clinical metastasis. Conversely, 36% of patients with clinical high-risk disease (Gleason $ 8) based on pretreatment parameters had low GC scores, and the majority of these men had favorable long-term clinical outcomes, with 77% being metastasis free and 85% of them alive at 5 years post-treatment. From these thought-provoking data, Karnes and colleagues concluded that GC assessment may identify men normally classified as “intermediate risk” who actually have a relatively high risk of prostate cancer metastasis. Even more importantly, GC may identify men labeled as pathologic “high-risk” based on standard parameters who actually have a low probability of developing metastases. Current guidelines recommend post-RP adjuvant treatment for men with high-risk features on post-RP pathology. Identification of this latter population of men using GC technology is particularly critical, potentially offering an avenue to spare them from adjuvant therapy and its associated side effects when there is little chance adjuvant treatment will offer a superior benefit over clinical observation. These three studies highlight several exciting recent advances in the genetic diagnosis of prostate cancer. Commercially available tissue assays are, or will soon be, available that may help with risk stratification based on either pretreatment biopsy core sampling or post-RP specimen analysis. These technologies may be particularly useful in the identification of men unlikely to benefit from adjuvant therapy following extirpative treatment for prostate cancer. The current state-of-theart is rapidly accelerating for the diagnosis of men with biopsy-proven prostate cancer. Vol. 15 No. 4 • 2013 • Reviews in Urology • 209 4004170006_RIULitra.indd 209 16/01/14 5:12 PM