17-Gene Assay Is a Significant Predictor of Prostate Cancer Aggressiveness

Get Permission

Eric A. Klein, MD

Peter Carroll, MD, MPH

Results from three studies published recently in European Urology indicate that the 17-gene Oncotype DX Genomic Prostate Score is a significant predictor of disease aggressiveness at the time of diagnosis before intervention with radiation or surgery.1 The test provides precise and individualized risk assessment and may help physicians identify the most appropriate treatment for their patients.

The Genomic Prostate Score, developed and validated by Genomic Health in collaboration with the Cleveland Clinic and University of California, San Francisco, addresses the challenges of local prostate cancer diagnosis and treatment by revealing the underlying tumor biology and using genes from multiple biologic pathways to predict the aggressiveness of prostate cancer at diagnosis.

“Our extensive development studies addressed the key challenges inherent to prostate cancer and helped overcome tumor heterogeneity and biopsy undersampling and understaging to develop a prostate cancer test that improves risk assessment at the time of diagnosis,” said Eric A. Klein, MD, Chairman of Glickman Urological and Kidney Institute, Cleveland Clinic, and Principal Investigator for the original development studies.

The two development studies performed at the Cleveland Clinic analyzed gene expression in prostate cancer tissue from both radical prostatectomy samples (441) and very small needle-biopsy specimens (167). Out of more than 700 candidate genes, the final analysis yielded 17 genes from four biologic pathways that had strong analytic performance and predicted outcomes associated with aggressive prostate cancer, including clinical recurrence, biochemical recurrence, prostate cancer death, and adverse pathology.

Validation Study

“Experience to date suggests that refining a man’s risk for having adverse pathology using the Genomic Prostate Score, despite information such as biopsy grade and tumor volume that suggest less aggressive disease, helps them make the most appropriate choice between active surveillance and immediate definitive treatment,” said Peter Carroll, MD, MPH, Professor and Chair, Department of Urology, UCSF, and Principal Investigator of the validation study of the 17-gene Genomic Prostate Score.

In the UCSF-led clinical validation study of diagnostic biopsies from 395 men who were candidates for active surveillance, the Genomic Prostate Score predicted the presence of adverse pathology (high-grade and/or high-stage disease) within the entire prostate prior to intervention (P = .002). Importantly, the 17-gene score predicted adverse pathology (P < .005) after controlling for established clinical factors such as prostate-specific antigen, biopsy Gleason Score, and age in multivariable analysis. Finally, the use of the Genomic Prostate Score in conjunction with National Comprehensive Cancer Network categories or UCSF Cancer of the Prostate Risk Assessment ­(CAPRA) score2 was shown to significantly improve risk assessment and potentially increase the number of men who should consider active surveillance. ■

Disclosure: Dr. Klein is a paid consultant for Genomic Health. Drs. Klein and Carroll are corresponding authors for the European Urology article.


1. Klein EA, et al: European Urology. May 16, 2014 (early release online).

2. Cooperberg, MR, et al: Cancer 107:2384–2391, 2006.