Activity of abiraterone acetate in metastatic patients with castration-resistant prostate cancer (mCRPC) previously treated with ketoconazole: A prospective phase II study from the prostate cancer clinical trials consortium.

Abstract

53 Background: Abiraterone acetate (AA), like ketoconazole (keto), inhibits CYP17, the rate-limiting enzyme in androgen biosynthesis. Since patients (pts) with prior keto treatment were excluded from the pivotal phase III AA trials, the utility of AA after keto is not well understood. This prospective study evaluated the efficacy of AA in pts who had received prior keto. Methods: Pts with progressive castration-resistant prostate cancer (mCRPC), prior keto therapy 28 days or more, and normal baseline organ function (including ACTH stimulation) tests were treated with AA 1,000 mg PO daily and prednisone 5 mg PO BID. Pts with prior chemotherapy were excluded. Serum androgen levels, including dehydroepiandrosterone (DHEA), were measured by liquid chromatography/mass spectroscopy (LC/MS) at baseline and during treatment for exploratory analyses. Radiographic progression-free survival (rPFS) was defined as freedom from: death, radiographic progression, or unequivocal clinical progression. Results: Forty two pts were enrolled. Median age was 71. Median prostate-specific antigen (PSA) was 47.5ng/dL. Median duration of prior keto was 38 weeks (range 5 to 207). Treatment with AA resulted in 30% or greater decline in PSA at 12 weeks in 20 pts (48%, 95% CI, 32-63%), and 50% or greater decline in PSA at 12 weeks in 16 pts (38%, 95% CI 24-54%). Median time to PSA progression (TTPP) was 16 weeks (range 4 to 64). Median rPFS was 24 weeks (range 1 to 88). Baseline serum DHEA levels were measured in 40 pts. Nine pts had DHEA less than the limit of quantitation (LOQ, 0.250ng/mL), and 31 pts had DHEA greater than or equal to LOQ. One pt with DHEA less than LOQ (1 out of 9, 11%, 95% CI 0.6-49%) had PSA decline 30% or more at 12 weeks, compared to 17 pts (17 out of 31, 55%, 95% CI 36-72%) with DHEA greater than or equal to LOQ (p=0.028). Median time to pain progression (TTPP) was 8 weeks (range 4 to 32) for pts with DHEA less than LOQ, compared to 18 weeks (range 4 to 64) for pts with DHEA greater than or equal to LOQ (p=0.012). Median rPFS was 12 weeks (range 4 to 24) for pts with DHEA less than LOQ, compared to 36 weeks (range 1 to 88) for pts with DHEA greater tha or equal to LOQ (p = 0.0006). Six pts remain on AA. Conclusions: A significant proportion of pts with prior keto exposure demonstrate clinical response to AA. DHEA levels via LC/MS merits further study as a predictive biomarker in pts treated with androgen synthesis inhibitors. Clinical trial information: NCT01199146.