In this abstract, however, the authors specifically assess how the addition of a genomic classifier, such as a Decipher (based on biopsy pathology), to NCCN clinical staging would have changed clinical management.
Per guidelines, NCCN low risk, favorable intermediate risk, unfavorable intermediate risk, or high risk would be recommended for active surveillance, active treatment (RP or RT alone), active treatment (RP or RT/short term ADT) or active treatment (RP or RT/long-term ADT).
In these groups, patients who were NCCN low but genomic intermediate/high may be changed to active treatment; patients who were NCCN favorable intermediate but genomic high risk may be changed to RT + ADT; patients who were NCCN unfavorable intermediate but genomic low risk may be reduced to RT alone; and patients who were NCCN high risk but genomic low risk may be reduced to short-term ADT with RT. They don’t specifically look at RP, as there is less variability in RP treatment.
They evaluated 4,474 consecutive patients with PCa who received the Decipher Biopsy test and had all clinical variables necessary for NCCN classification. Of these, there were 927 NCCN low-risk, 2,427 intermediate, and 1,120 high-risk patients. 1212 patients had insufficient data for NCCN classification.
They do note some slight association with Decipher score with age – older patients had slightly higher Decipher scores. Decipher scores were also moderately positively correlated with Biopsy grade, % positive biopsy cores, PSA and clinical T-stage. It was strongly associated with NCCN clinical stage and CAPRA score.
Among NCCN low-risk, the incidence of genomic low, intermediate, and high risk was 58.7%, 25.0%, and 16.3% respectively.
For NCCN intermediate, the incidence of genomic low, intermediate, and high risk was 36.5%, 27.6%, and 35.8%, respectively.
For NCCN high risk, the incidence of genomic low, intermediate, and high risk was 15.9%, 17.1%, and 67.1%, respectively.
In terms of patients in whom genomic classification would have modified treatment due to discrepancy in classification, there were 41.3% of NCCN low risk patients with intermediate or high genomic risk, 26.7% of favorable intermediate risk patients who had high genomic risk, 32.4% of unfavorable intermediate risk patients with low genomic risk, and 15.9% of high risk patients with low genomic risk. Approximately 1/3 of the entire cohort would have had treatment change.
However, where the future of this analysis should go is whether or not the treatment change was associated with any clinical sequelae – if outcomes are similar, perhaps then no change is needed in clinical practice? If outcomes are much improved, then there is obvious need for incorporation into clinical classification.
Limitations / Discussion Points:
1. This was, to some degree, funded by GenomeDx, the parent company for the Decipher test
Speaker: Paul Nguyen, MD
Co-Authors: Jingbin Zhang, Kasra Yousefi, Elai Davicioni, Robert Benjamin Den, Felix Y Feng, Daniel Eidelberg Spratt
Institution(s): Brigham and Women's Hospital/ Dana-Farber Cancer Institute, Boston, MA; GenomeDx Biosciences Inc., Vancouver, BC, Canada; Sidney Kimmel Medical College at Thomas Jefferson University, Philadelphia, PA; University of California San Francisco, San Francisco, CA; Memorial Sloan Kettering Cancer Center, New York, NY
Written by: Thenappan Chandrasekar, MD, Clinical Fellow, University of Toronto, Twitter: @tchandra_uromd at the 2018 American Society of Clinical Oncology Genitourinary (ASCO GU) Cancers Symposium, February 8-10, 2018 - San Francisco, CA