Background: SPARTAN evaluated APA vs PBO in pts with nmCRPC and a prostate-specific antigen doubling time of ≤ 10 mo receiving androgen deprivation therapy (ADT). At primary end point analysis of metastasis-free survival (MFS), APA significantly improved median MFS by 2 yrs, as well as time to metastasis, progression-free survival, and time to symptomatic progression vs PBO (Smith, et al. NEJM 2018); overall survival (OS) results were immature. SPARTAN was unblinded upon meeting the primary end point; pts still on PBO were allowed to cross over to APA. Final survival results are reported herein. Methods: 1207 nmCRPC pts were randomized 2:1 to APA (240 mg QD) or PBO plus ongoing ADT. At progression, pts could receive open-label sponsor-provided abiraterone acetate + prednisone. After the primary efficacy end point (MFS) was met, 76 PBO pts (19%) crossed over to APA. OS and time to cytotoxic chemotherapy (TTCx) were tested by group sequential testing procedure with O’Brien-Fleming (OBF)-type alpha spending function. Time-to-event end points were analyzed by Kaplan-Meier method and Cox model. A sensitivity analysis for OS, accounting for crossover using a naïve censoring approach, was conducted. Results: With follow-up of 52.0 mo, 428 (of 427 required) OS events had occurred. Median treatment duration: APA, 32.9 mo; PBO, 11.5 mo. Median OS was significantly longer with APA + ADT vs PBO + ADT (73.9 vs 59.9 mo), (hazard ratio [HR], 0.784, Table). APA significantly lengthened TTCx (HR, 0.629). Discontinuation rates (APA vs PBO) due to progressive disease were 42.7% vs 73.9%, and due to adverse events (AE) 15.2% vs 8.4%. Safety was consistent with previous reports; grade 3/4 treatment-emergent (TE) AEs of special interest were rash 5.2%, fractures 4.9%, falls 2.7%, ischemic heart disease 2.6%, hypothyroidism 0%, and seizures 0%. 1 TEAE leading to death (myocardial infarction) was considered potentially APA related. Conclusions: In pts with nmCRPC, APA + ADT significantly improved OS compared with PBO + ADT, with median OS > 6 yr in the APA + ADT group and 14 mo improvement over PBO + ADT. Benefit from APA was observed despite a 19% crossover from PBO. The safety profile of APA was consistent with prior interim analyses. Clinical trial information: NCT01946204.

^ap value from stratified log-rank test.

^bOBF required p value ≤ 0.046 to be considered statistically significant.

NR, not reached.