Presentation Authors: Fred Saad*, Montréal, Canada, Shibu Thomas, Spring House, PA, Felix Y. Feng, San Francisco, CA, Michael Gormley, Spring House, PA, Angela Lopez-Gitlitz, Margaret K. Yu, Los Angeles, CA, Shinta Cheng, Raritan, NJ, Deborah S. Ricci, Spring House, PA, Oliver Brendan Rooney, High Wycombe, United Kingdom, Paul N. Mainwaring, Brisbane, Australia, David Olmos, Madrid, Spain, Simon Chowdhury, London, United Kingdom, Boris A. Hadaschik, Essen, Germany, Nick Fishbane, Elai Davicioni, Yang Liu, San Diego, CA, Eric J. Small, San Francisco, CA, Matthew R. Smith, Boston, MA
Introduction: The DECIPHER prostate test (GenomeDx Biosciences, Inc., San Diego, CA) is a clinical-grade mRNA-based test with a genomic classifier (GC) score that has been independently validated for predicting metastatic disease post radical prostatectomy in men with prostate cancer (Klein EA, et al. Eur Urol. 2013). Pts with high GC scores have significantly higher risk for metastasis than pts with low to average GC scores. APA, a next-generation androgen receptor inhibitor, improved metastasis-free survival (MFS) and other secondary end points in the SPARTAN study of pts with nmCRPC with prostate-specific antigen doubling time â‰¤ 10 months who were receiving androgen deprivation therapy (ADT). We evaluated the effect of APA on MFS in a subgroup of SPARTAN pts grouped by DECIPHER GC score.
Methods: A high-density Affymetrix GeneChip platform (1.4 million genomic loci) was used to assess gene expression in 233 archived primary tumors from pts enrolled in SPARTAN who had been randomized 2:1 to APA+ADT or ADT alone. Gene expression values of 22 markers were used to generate GC scores (â‰¤ 0.6 [low to average] vs > 0.6 [high]) using prevalidated algorithms. Cox proportional hazard model was used to assess association of GC scores with clinical outcomes.
Results: Among 233 pts in SPARTAN, 50% had a high DECIPHER GC score. 78 and 39 in APA+ADT and ADT arms, respectively, had high GC scores; 76 and 40 pts in these arms, respectively, had low to average GC scores. Pts with high GC score, who typically have poor prognosis with standard of care (SoC) treatment, including ADT, had improved MFS when treated with APA+ADT vs ADT alone (hazard ratio [HR], 0.21; p < 0.0001). Pts with low to average GC score also had improved MFS with APA+ADT vs ADT alone (HR, 0.46; p = 0.031). There was no difference in MFS with APA+ADT among pts with high GC scores vs those receiving APA+ADT who had low to average GC scores (HR, 1.11; p = 0.741). Similar correlations between GC scores and second progression-free survival, overall survival, and symptomatic progression were observed.
Conclusions: APA+ADT improved MFS and can overcome the negative prognosis associated with high GC scores. To our knowledge, this is the first report of application of the DECIPHER GC score at diagnosis to predict outcomes and therapeutic response in CRPC and the first SoC treatment to overcome negative prognosis.
Source of Funding: Janssen Research & Development