Aims: AA, a selective androgen biosynthesis inhibitor, blocks the action of CYP17 and inhibits adrenal and intratumoral androgen production. AA has demonstrated improved overall survival (OS) by 4.6 months versus placebo (PL) (HR=0.74) in mCRPC patients previously treated with docetaxel.

Methods: COU-AA-301 is a randomized double-blind study (N = 1195) of AA(1 g)+P(5 mg po BID) vs PL+P administered to mCRPC patients post-docetaxel. Sixteen Australian sites enrolled 101 patients. Primary endpoint was OS. In these post-hoc exploratory analyses, we evaluated the impact of timing of investigator-reported first and last dose of docetaxel and reason for docetaxel discontinuation on OS.

Results: At randomization, treatment arms were balanced with respect to baseline characteristics, prior docetaxel use, and reasons for discontinuation. In both arms, almost half (45%) of patients discontinued docetaxel due to progressive disease (PD); remainder discontinued docetaxel after completing all planned cycles (37%), due to toxicity (12%), or for other reasons (5%) per investigator. Median OS from both first docetaxel dose (AA+P 32.6 versus PL+P 27.6 months, HR(95%CI): 0.75(0.65, 0.88), p=0.0002) and last docetaxel dose (AA+P 23.3 versus PL+P 19.4 months, HR(95%CI): 0.74(0.64, 0.86), p≤0.0001) was longer with AA+P. Median OS in patients discontinuing docetaxel (calculated from randomization) for PD (AA+P 14.2 versus PL+P 10.5 months HR(95%CI): 0.77(0.62, 0.97), p=0.0222) and for all other reasons (AA+P 17.0 versus PL+P 12.6 months HR(95%CI): 0.73(0.59, 0.89), p =0.0025) was longer with AA+P.

Conclusions: Patients treated with AA+P had a prolonged median OS of >32 months from time of initial docetaxel therapy. These exploratory analyses suggest the duration from first or last dose of docetaxel or whether or not patients discontinue docetaxel for PD does not alter the OS benefit of AA in mCRPC. Lack of dependence on docetaxel timing further confirms the robustness of the primary survival result.