#0346

Enzalutamide (ENZA)±leuprolide (L) in biochemically recurrent (BCR) prostate cancer (PC): EMBARK post hoc analysis of Asia Pacific (APAC) population

H. Wu¹, S. Freedland^2,3, H. Woo^4,5, C. Kim⁶, A. Dhar⁷, M. Rosales⁸, M. Kalac⁹, Y. Tang¹⁰, N. Shore¹¹

¹China Medical University Hospital, Department of Urology, Taichung City, Taiwan
²Cedars-Sinai Medical Center, Samuel Oschin Comprehensive Cancer Institute, Los Angeles, United States
³Durham VA Medical Center, Durham, United States
⁴Sydney Adventist Hospital, The Prostate Centre of Excellence, Sydney, Australia
⁵Australian National University, The College of Health and Medicine, Canberra, Australia
⁶Ewha Womans University Mokdong Hospital, Seoul, Korea (Republic of)
⁷Astellas Pharma Singapore, Singapore, Singapore
⁸Astellas Pharma Global Development, Northbrook, United States
⁹Pfizer Inc., New York, United States
¹⁰Pfizer Inc., Global Product Development, San Francisco, United States
¹¹Carolina Urologic Research Center, Myrtle Beach, United States

Introduction:

In EMBARK (NCT02319837), ENZA+L and ENZA monotherapy (mono) prolonged metastasis-free survival (MFS) vs placebo (P)+L in patients (pts) with high-risk BCR non-metastatic hormone-sensitive PC. This post hoc analysis was performed to assess outcomes of pts in the APAC region (Australia, Taiwan, South Korea).

Material and methods:

Pts with high-risk BCR PC were randomized (1:1:1) to ENZA+L (double-blind), ENZA mono (open-label), and P+L (double-blind). Treatment was suspended at week 37 if prostate-specific antigen (PSA) level was 0.2ng/mL and restarted when PSA was ≥5.0ng/mL (no prior radical prostatectomy [RP]) or ≥2.0ng/mL (prior RP). Primary endpoint: MFS with ENZA+L vs P+L; secondary endpoints: MFS with ENZA mono vs P+L, time to PSA progression, time to new antineoplastic therapy use, PSA0.2ng/mL at week 36, treatment suspension duration, and safety. Nominal two-sided P-values were derived for time-to-event efficacy endpoints with log-rank test for between-group comparisons. Hazard ratios (HRs) and 95% confidence intervals (CIs) for treatment effects were assessed with Cox regression model. Intention-to-treat principle was applied to efficacy endpoints; safety was assessed in the as-treated population. All results were considered descriptively.

Results:

Of 1068 pts in EMBARK, 191 were from APAC (ENZA+L: 62; ENZA mono: 62; P+L: 67). APAC population outcomes were numerically better compared to the main study but CIs were wider due to small sample size (Table). Median follow-up was 61 months. Risk of metastasis or death: 68% lower with ENZA+L and 50% lower with ENZA mono vs P+L. Risk of PSA progression: 95% lower with ENZA+L and 67% lower with ENZA mono vs P+L. Risk of new antineoplastic therapy use: 77% lower with ENZA+L and 65% lower with ENZA mono vs P+L. A greater percentage of APAC pts had PSA0.2ng/mL at week 36 compared to the overall population. Median treatment suspension duration for ENZA mono and ENZA+L was similar to the overall population. A similar percentage of pts in the APAC region reported treatment-emergent adverse events (TEAEs) as in the overall population.