Intermittent (IAD) versus continuous androgen deprivation (CAD) in hormone sensitive metastatic prostate cancer (HSM1PC) patients (pts): Results of S9346 (INT-0162), an international phase III trial.

Abstract

4 Background: Castration resistance occurs in the vast majority of HSM1PC pts treated with AD, with a median survival of 2.5 years (y). It is in part an adaptive process with activation of genes resulting in the production of autocrine/paracrine growth factors that contribute to maintaining the viability of PC cells. Replacing androgens before castration resistance is hypothesized to maintain PC androgen-dependence. Preclinically IAD prolonged time to castration resistance and early clinical data indicated feasibility and potential for better quality of life. Methods: HSM1PC pts with performance status (PS) 0-2, PSA ≥ 5 ng/ml were treated with 7 months (m) of goserelin + bicalutamide. Pts achieving PSA ≤4 ng/ml on m 6 and 7 were stratified by prior neoadjuvant AD/finasteride, PS and disease extent (minimal, extensive) and randomized to CAD or IAD. Primary objective: To assess if overall survival (OS) with IAD is noninferior to CAD using a one-sided test with an upper bound hazard ratio=1.20, adjusting for stratification factors. Sample size: 756 pts/arm, type I and II error rates of 0.05 and 0.10. Results: 3,040 pts were accrued by SWOG, CALGB, ECOG, NCIC, and EORTC (5/95- 9/08). After 7 m of CAD, 1535 eligible pts achieved PSA ≤4.0 (median age 70 yrs, 4% PS 2, 48% extensive disease, 12% prior neoadjuvant AD) and were randomized to CAD (759 pts) or IAD (770 pts). Grade 3/4 related adverse events: IAD 30.3%, CAD 32.6%. Median follow-up was 9.2 yrs. Median and 10 yr OS: All eligible pts from study entry: 3.6 yrs, 17%; from randomization CAD: 5.8 yrs, 29%; IAD: 5.1 yrs, 23%, HR (IAD/CAD) = 1.09 (95% CI 0.95, 1.24). No interaction with therapy was significant (p>0.25) except suggestion with disease extent (p=0.08): extensive disease HR=0.96 (95% CI 0.79, 1.15, p=0.64); minimal disease: HR=1.23 (95% CI 1.02, 1.48, p=0.035). PC was cause of death in 56% of CAD and 64% IAD pts. OSby race was not different (p=0.44). Conclusions: In HSM1PC, IAD is not proven to be noninferior to CAD. For extensive disease pts IAD was noninferior; however, IAD was statistically inferior in minimal disease pts suggesting that CAD is the preferred treatment in this group.