First-line chemotherapy (CT) versus anti-androgen therapy (AT) for patients (pts) with metastatic castration-resistant prostate cancer (mCRPC): Individual patient analysis of the Veterans Health Administration (VHA) dataset.

Abstract

156 Background: The outcomes of pts receiving first-line CT vs AT for mCRPC are unclear. Using the VHA dataset, we compared outcomes with first-line CT vs AT in pts with mCRPC and assessed the impact of prior androgen deprivation therapy (ADT) duration and known prognostic factors. Methods: Pts with mCRPC initiating first-line AT (abiraterone, enzalutamide) or CT (taxane) from Oct 2012 to Sept 2014 were identified. The impact of AT vs CT on overall survival (OS) and time to discontinuation (TTD) was assessed using Cox proportional hazard models, adjusting for prior ADT duration, known prognostic factors (hemoglobin [Hb], albumin [alb], alkaline phosphatase [ALP], prostate-specific antigen [PSA]), Charlson Comorbidity Index (CCI) and chronic disease score (CDS). Results: Overall, 1445 pts were evaluable; 1108 received AT (abiraterone 996, enzalutamide 112) and 337 received CT (docetaxel). The overall median duration of prior ADT was 464 days. On multivariable analysis, prior ADT duration, CCI, CDS, Hb, Alb, ALP and PSA were associated with OS, but AT vs CT was not (HR: 1.041 [95% CI: 0.853–1.270], p = 0.6943). PSA levels, prior ADT duration, and ALP was associated with TTD, and TTD was shorter for CT vs AT (HR: 2.339 [95% CI: 1.969–2.779], p < 0.0001). Longer prior ADT duration was associated with longer OS (HR: 0.566, p < 0.0001) and TTD (HR: 0.831, p = 0.0363) in the AT cohort, but not in the CT cohort. Treatment-free interval (TFI) (between first- and second-line treatment) was longer for CT vs AT (mean: 53 vs 39 days; p = 0.0303). Conclusions: To our knowledge, this is one of the largest mCRPC datasets analyzed at the individual pt level comparing first-line CT vs AT. OS was not significantly different for first-line CT vs AT after adjusting for key prognostic factors, despite shorter TTD with CT and longer TFI after first-line CT. Prior ADT duration ≤ 464 days was associated with shorter OS and TTD in the AT cohort, but not the CT cohort, suggesting that such pts may benefit from receiving CT over AT. The results are hypothesis-generating and prospective validation is required. Funding: Sanofi Genzyme