Introduction

Enzalutamide and abiraterone are frequently used for castration-resistant prostate cancer (CRPC), because of improved tolerability of them compared to taxanes. However, there are no predictive biomarkers to make a decision of how to best use abiraterone or enzalutamide. The aim of this study was to clarify the impact of serum testosterone in a decision-making of treatment selections (enzalutamide or abiraterone) for CRPC.

Methods

We retrospectively evaluated consecutive CRPC patients treated at our institution and other satellite hospitals between 2013 and May 2016. A total of 115 patients received enzalutamide or abiraterone for CRPC and had serum testosterone measure at the initiation of the treatment were included in this study. The patients were divided into two groups by serum testosterone before enzalutamide or abiraterone: 54 in the testosterone <5 ng/dl group and 61 in the testosterone ≥5 ng/dl group. Prostate-specific antigen (PSA) response rates (defined as ≥50% PSA declines), PSA progression-free survival (PSA-PFS), and overall survival (OS) were compared between groups.

Results

A total of 72 patients were treated with enzalutamide and 43 with abiraterone. In the testosterone <5 ng/dl group, the PSA response rates were significantly lower with enzalutamide than that with abiraterone (32% vs. 62%, p=0.033), whereas there was no difference in the testosterone ≥5 ng/dl group (81% vs. 93%, p=0.429). During the median follow-up period of 12 months, 68 men (59%) had PSA relapse. In the testosterone <5 ng/dl group, the median PSA-PFS was significantly lower with enzalutamide than that with abiraterone (2.8 months vs. 6.4 months, p=0.004) (Figure). Likewise, it was significantly lower with enzalutamide than that with abiraterone in the testosterone ≥5 ng/dl group (7.6 months vs. no available, p=0.004) (Figure). Multivariate analysis reveals that testosterone ≥5 ng/dl was an independent predictive factor for PSA-PFS (HR 3.1, p<0.001). However, there was no significant difference in the median OS according to the different testosterone groups.

Conclusions

This result suggests that serum testosterone level is a useful biomarker in a decision-making of treatment selections in the novel hormonal therapy for CRPC.

Funding

none