Harshraj Leuva1, Keith Sigel2, Mengxi Zhou3, Julia Wilkerson3, David H Aggen3, Yeun-Hee Anna Park4, Christopher B Anderson5, Ta-Chueh Melody Hsu4, Erik Langhoff4, Glen McWilliams5, Charles G Drake3, Richard Simon6, Susan E Bates5, Tito Fojo7. 1. James J. Peters Bronx Veterans Affairs Medical Center, Bronx, NY; SUNY Downstate Health Sciences University, Brooklyn, NY. Electronic address: harshrajleuva@gmail.com. 2. Mount Sinai School of Medicine, New York, NY. 3. Columbia University Herbert Irving Comprehensive Cancer Center, New York, NY. 4. James J. Peters Bronx Veterans Affairs Medical Center, Bronx, NY. 5. James J. Peters Bronx Veterans Affairs Medical Center, Bronx, NY; Columbia University Herbert Irving Comprehensive Cancer Center, New York, NY. 6. R Simon Consulting, Potomac MD. 7. James J. Peters Bronx Veterans Affairs Medical Center, Bronx, NY; Columbia University Herbert Irving Comprehensive Cancer Center, New York, NY. Electronic address: atf2116@cumc.columbia.edu.
Abstract
BACKGROUND: With 1.3 million new cases in 2018 worldwide, prostate cancer remains a challenge. Development of novel therapies targeting the androgen pathway followed recognition of the continued importance of androgens in castrate-resistant prostate cancer. To assess abiraterone and enzalutamide efficacy we analyzed data from US Veterans Administration Medical Centers (VAMCs). METHODS: We used a novel method independent of assessment intervals and ideal for real-world analysis to estimate rates of tumor growth (g) and regression (d). FINDINGS: Using the VA Informatics and Computing Infrastructure, we collected data from 5,116 Veterans with castrate-resistant prostate cancer prescribed abiraterone, enzalutamide or both. We estimated values for g and d and demonstrated a correlation of g with overall survival (P < .0001). Abiraterone and enzalutamide slowed growth rates across age groups and across the entire VAMC system, although less so in Veterans previously treated with a taxane and those with Gleason grade group 5 tumors. Abiraterone and enzalutamide efficacy in first-line were comparable although abiraterone in first-line slowed growth rates significantly more in African Americans than in Caucasians; enzalutamide was a better salvage therapy. When abiraterone was first-line and g was low, switching to enzalutamide was associated with a faster g in 67%. INTERPRETATION: In the real-world g can be estimated using a novel analysis method indifferent to assessment intervals that correlates highly with OS. While we show excellent real-world outcomes with abiraterone and enzalutamide, 2 effective and tolerable therapies, our results in VAMCs suggest enzalutamide should follow abiraterone. Changing therapies may be detrimental and consideration should be given to continue monitoring of growth rates over time. Funding Support from the Prostate Cancer Foundation and the Blavatnik Family Foundation.
BACKGROUND: With 1.3 million new cases in 2018 worldwide, prostate cancer remains a challenge. Development of novel therapies targeting the androgen pathway followed recognition of the continued importance of androgens in castrate-resistant prostate cancer. To assess abiraterone and enzalutamide efficacy we analyzed data from US Veterans Administration Medical Centers (VAMCs). METHODS: We used a novel method independent of assessment intervals and ideal for real-world analysis to estimate rates of tumor growth (g) and regression (d). FINDINGS: Using the VA Informatics and Computing Infrastructure, we collected data from 5,116 Veterans with castrate-resistant prostate cancer prescribed abiraterone, enzalutamide or both. We estimated values for g and d and demonstrated a correlation of g with overall survival (P < .0001). Abiraterone and enzalutamide slowed growth rates across age groups and across the entire VAMC system, although less so in Veterans previously treated with a taxane and those with Gleason grade group 5 tumors. Abiraterone and enzalutamide efficacy in first-line were comparable although abiraterone in first-line slowed growth rates significantly more in African Americans than in Caucasians; enzalutamide was a better salvage therapy. When abiraterone was first-line and g was low, switching to enzalutamide was associated with a faster g in 67%. INTERPRETATION: In the real-world g can be estimated using a novel analysis method indifferent to assessment intervals that correlates highly with OS. While we show excellent real-world outcomes with abiraterone and enzalutamide, 2 effective and tolerable therapies, our results in VAMCs suggest enzalutamide should follow abiraterone. Changing therapies may be detrimental and consideration should be given to continue monitoring of growth rates over time. Funding Support from the Prostate Cancer Foundation and the Blavatnik Family Foundation.