Abstract 3635: Association of reduction in serum androgens and change in serum PSA in patients treated with abiraterone acetate (AA): A subset analysis of the COU-AA-301 phase 3 randomized trial

• Published in: Cancer research (Chicago, Ill.) Vol. 72; no. 8_Supplement; p. 3635
• Main Authors: Ryan, Charles, Peng, Weimin, Kheoh, Thian, Scher, Howard I., Molina, Arturo
• Format: Journal Article
• Language: English
• Published: 15.04.2012
• ISSN: 0008-5472; 1538-7445
• DOI: 10.1158/1538-7445.AM2012-3635

Abstract Objective In study COU-AA-301 the androgen biosynthesis inhibitor AA significantly increased overall survival vs placebo in mCRPC post-docetaxel (HR=0.74) (Scher, ASCO 2011). The objective of the current study was to determine the activity of AA in decreasing androgens in a subset of patients in COU-AA-301 using an ultrasensitive assay that measured serum testosterone (T), androstenedione (A4), and dehydroepiandrostenedione (DHEA) below concentrations detectable with commercially available methods, to determine the ability of AA + prednisone (P), compared to placebo + P, to reduce androgen levels below the lower limits of quantification (LLOQ), and to correlate decreases in androgen levels with PSA response. Methods COU-AA-301 is a randomized double blind study of AA (1000 mg QD + prednisone [P] 5 mg po BID) vs placebo + P. Serum androgen levels were measured at baseline and week 12 in the AA + P and placebo + P arms by an ultrasensitive assay using LC/MS/MS. The LLOQs were: T, <0.05 ng/dL; A4, <0.10 ng/dL; DHEA, <0.10 µg/dL. In this post hoc exploratory analysis, the reduction in androgen levels from baseline to week 12 was compared between treatment arms using the t-test. Also evaluated was the androgen response, defined as the proportion of patients with week 12 androgen concentration below the respective LLOQ. The association between PSA response (PSA Δ50% baseline) and androgen response was analyzed by logistic regression. Results AA + P treatment led to significant reductions in T, A4, and DHEA levels from baseline to week 12 (T: –5.46 vs –2.74 ng/dL, P=0.0002; A4: –26.33 vs –8.81 ng/dL, P=0.0007; DHEA: –22.59 vs –10.40 µg/dL, P=0.0179; AA + P vs placebo + P, respectively). The number of patients with an androgen response at week 12 was substantially greater for AA + P compared with placebo + P (T: 34/72 [47%] vs 0/37; A4: 19/63 [30%] vs 0/31; DHEA: 46/80 [58%] vs 1/38 [3%]; AA + P vs placebo + P, respectively). Eleven patients showed androgen response on all three androgens, 2 patients on A4 and T only, 4 patients on A4 and DHEA only, and 14 patients on T and DHEA only. There was a positive association between PSA response and androgen response, adjusted for treatment effect, in favor of androgen responders (T: OR=1.54 [95% CI, 0.546, 4.347]; A4: OR=3.06 [95% CI, 0.975, 9.604]; DHEA: OR=1.08 [95% CI, 0.401, 2.899]). Conclusions: In study COU-AA-301, AA + P potently inhibited the production of serum androgens T, A4, and DHEA as measured by an ultrasensitive assay. Reduction in T and A4 appears to correlate with PSA response, providing further evidence that mCRPC remains dependent on androgen-mediated signaling. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 103rd Annual Meeting of the American Association for Cancer Research; 2012 Mar 31-Apr 4; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2012;72(8 Suppl):Abstract nr 3635. doi:1538-7445.AM2012-3635