RESTORE: A single-arm, open-label phase II trial of bipolar androgen therapy (BAT) in men with metastatic castration resistant prostate cancer (mCRPC)—A comparison of post-abiraterone (Abi) versus post-enzalutamide (Enza) patients (Pts).

Abstract

5576 Background: A paradoxical inhibition of cell growth has been observed in both androgen-sensitive and castration resistant prostate cancer cell lines following the addition of high-dose testosterone.We have conducted several clinical trials investigating a mode of supraphysiologic testosterone therapy termed, BAT, in which testosterone levels are rapidly driven to the supraphysiologic range followed by a return to near-castrate levels over 28-day treatment cycles with favorable results. We previously reported the efficacy of BAT in mCRPC pts that were progressing on enza. In this study, we compared the effect of BAT in mCRPC pts whose last therapy was abi vs. enza. In addition, we examined the benefit of abi or enza rechallenge after progression on BAT. Methods: 59 mCRPC pts (n = 29 post abi; n = 30 post enza) were enrolled and received at lease one dose of BAT monotherapy, 400mg intramuscularly every 28 days. After clinical or radiographic progression on BAT, pts were rechallenged with the AR targeted therapy to which they were most recently resistant. The co-primary endpoints were a 50% decline in PSA from baseline (PSA50) for BAT and for enza/abi rechallenge. Results: 5/29 (17.2%) of post-abi pts compared to 9/30 (30%) in the post enza group achieved a PSA50 response (P = 0.36). Post BAT rechallenge with abi (n = 19) or enza (n = 22) resulted in a PSA50 response rate of 15.8% (n = 3) and 68.2% (n = 15), respectively (P = 0.001). The total duration of benefit (i.e. PFS on BAT + PFS on rechallenge = “PFS2”) was significantly longer in the post enza vs. post-abi patients (Median PFS2: 12.75 vs. 8.125 months; P = 0.04. Lastly, AR-V7 negative (n = 42) pts has a significantly longer median PFS2 compared to AR-V7 positive (n = 10) pts. (10.3 vs. 7.1 months, P = 0.005). Conclusions: Our data suggest that BAT may be more effective at resensitizing mCRPC to direct AR antagonists (i.e. enza) compared to abi. Detection of AR-V7 portended a worse outcome on BAT/rechallenge. Further clinical study is warranted. Clinical trial information: NCT02090114 . [Table: see text]