Comparing Sequencing of Abiraterone and Enzalutamide in Men With Metastatic Castration-Resistant Prostate Cancer: A Retrospective Study.

BACKGROUND There are no validated clinical decision tools to aid optimal treatment selection in men with metastatic castration-resistant prostate cancer (mCRPC). Frequently, abiraterone and enzalutamide are used prior to chemotherapy in mCRPC, given the more favorable safety profiles. However, there is no published data on clinical outcomes regarding best sequencing of these two agents. METHODS A retrospective analysis of consecutive mCRPC patients treated with enzalutamide and abiraterone at Johns Hopkins was conducted. Patients were treated with sequential enzalutamide and abiraterone, in either order. The combined progression-free survival (PFS: PFS1 + PFS2) of abiraterone-to-enzalutamide was compared to the reverse sequence, where PFS1 and PFS2 represented clinical/radiographic progression-free survival on the first and second agents, respectively. Overall survival (OS) from the start of the first therapy to death and PSA response rates (defined as ≥50% PSA declines at any time) were also compared between groups. Outcomes were adjusted using propensity score-weighted multivariable Cox analyses. RESULTS Eighty-one patients who satisfied our entry criteria were identified: 65 in the abiraterone-to-enzalutamide group and 16 in the enzalutamide-to-abiraterone group. There were no significant baseline differences between groups. Multivariable analysis suggested a difference between groups favoring the abiraterone-to-enzalutamide sequence with respect to combined PFS (HR 0.37, 95%CI 0.22-0.64, P < 0.001). There was no statistical difference in OS between the groups after multivariable adjustment (HR 0.57, 95%CI 0.29-1.11, P = 0.098), although OS was numerically superior in the abiraterone-to-enzalutamide group. CONCLUSIONS We observed differences suggesting improved outcomes favoring the abiraterone-to-enzalutamide sequence in men with mCRPC, with statistical confirmation in terms of PFS but not OS. Prospective studies are required to verify these hypothesis-generating findings. Further evaluation of biomarkers to inform optimal treatment sequencing in men with mCRPC is urgently needed. Prostate 77:33-40, 2017. © 2016 Wiley Periodicals, Inc.