Bipolar androgen therapy sensitizes castration-resistant prostate cancer to subsequent androgen receptor ablative therapy

Abstract

Background: Cyclical, high-dose testosterone administration, termed bipolar androgen therapy (BAT), can induce clinical responses and restore sensitivity to androgen signalling inhibition in patients with previously treated castration-resistant prostate cancer (PCa) (CRPC). This trial evaluated whether BAT is a safe and effective first-line hormonal therapy for patients with CRPC.

Patients and methods: In cohort C of this single-centre, open-label, phase II, multi-cohort trial (RE-sensitizing with Supraphysiologic Testosterone to Overcome REsistance study), 29 patients with CRPC received first-line hormonal therapy with 400 mg of testosterone cypionate intramuscularly every 28 days concurrent with a luteinising hormone-releasing hormone agonist/antagonist. The primary end-point of the study was the PSA₅₀ response rate to BAT treatment.

Results: After treatment with BAT, four of 29 patients (14%; 95% confidence interval [CI]: 4-32%) experienced a PSA₅₀ response. The median radiographic progression-free survival to BAT was 8.5 months (95% CI: 6.9-15.1) for patients with metastatic CRPC. After progression on BAT, 17 of 18 patients (94%; 95% CI: 73-100%) achieved a PSA₅₀ response and 15 of 18 patients (83%; 95% CI: 59-96) achieved a PSA₉₀ response on abiraterone or enzalutamide. Twelve of 15 patients (80%; 95% CI: 52-96) with metastatic CRPC remain on abiraterone or enzalutamide with a median duration of follow-up of 11.2 months.

Conclusion: As first-line hormonal treatment for CRPC, BAT was well tolerated and resulted in prolonged disease stabilisation. After progression on BAT, patients had favourable responses to second-generation androgen receptor-targeted therapy.

Trial registration: ClinicalTrials.gov NCT02090114.

Keywords: Bipolar androgen therapy; Castration-resistant prostate cancer; RESTORE trial; Testosterone.

Figure 1.

Efficacy of BAT. (A) Waterfall plot of best PSA response to BAT. Fifty percent decrease in PSA (PSA₅₀) is shown with a dotted line. Increases greater than 100 percent are truncated at 100 percent. (B) Radiographic PFS to BAT for patients with mCRPC is shown as a Kaplan-Meier curve (solid line) with 95% confidence intervals shown with dotted lines. nmCRPC = non-metastatic castration-resistant PCa, mCRPC = metastatic castration-resistant PCa.

Figure 2.

Clinical benefit of sequential treatment with BAT followed by abiraterone or enzalutamide. (A) Waterfall plot of best PSA response to subsequent abiraterone or enzalutamide. Fifty percent and ninety percent decrease in PSA (PSA₅₀ and PSA₉₀) are shown with dotted lines. (B) PSA-PFS for subsequent abiraterone or enzalutamide for patients with mCRPC shown as a Kaplan-Meier curve (solid line) with 95% confidence intervals shown with dotted lines. (C) PSA-PFS2 for BAT followed by abiraterone or enzalutamide for patients with mCRPC shown as a Kaplan-Meier curve (solid line) with 95% confidence intervals shown with dotted lines. nmCRPC = non-metastatic CRPC, mCRPC = metastatic CRPC

Figure 3.

(A) Swimmer’s plot of all patients. Two patients remain on BAT at the time of data compilation. One patient died while on BAT. Seventeen patients were treated with second line abiraterone or enzalutamide (Abi/Enza) and 14 patients remain on this therapy at the time of data compilation with 4 of these patients demonstrating a PSA complete response (PSA CR). One patient has experienced prolonged disease stability off therapy after progression on BAT. One patient was lost to follow-up (LTF) after progression on BAT. Four patients with treated with non-hormonal second line therapy. (B) Genetic Sequencing data available from medical record for each patient aligned with swimmer’s plot duration of BAT.