Cortisol as biomarker for CYP17 inhibition in mCRPC patients treated with abiraterone acetate.

Abstract

5035 Background: Abiraterone acetate is an effective metastatic castration resistant prostate cancer (mCRPC) treatment, however, there is a high variability in response. It inhibits CYP17, thereby preventing the production of androgens. Abiraterone trough concentrations (Cmin) > 8.4 ng/mL have been associated with an increased progression free survival (PFS) (Eur J Cancer. 2017;72:54-61; Prostate Cancer Prostatic Dis. 2020;23(2):244-251). However, plasma levels do not directly provide information on the level of CYP17 inhibition. Ideally, testosterone levels should be measured, but these levels are below the detection limit of available assays. The synthesis of cortisol is also inhibited by abiraterone via CYP17 inhibition and might therefore be a biomarker for CYP17 inhibition. The objective of this study was to investigate if cortisol levels are related to abiraterone levels and to clinical response. Methods: An observational study was performed in mCRPC patients treated with abiraterone acetate. At the outpatient clinic, plasma samples were collected for pharmacokinetic (PK) monitoring at each hospital visit. Reference populations of healthy volunteers and mCRPC patients using enzalutamide were included to investigate the influence of mCRPC and abiraterone treatment on the circadian rhythm of cortisol. Abiraterone and cortisol levels were measured using validated liquid chromatography-mass spectrometry (LC-MS/MS) assays. Clinical (prostate specific antigen (PSA) independent) PFS and PSA response were evaluated. Results: In total, 117 mCRPC patients using abiraterone acetate, 100 mCRPC patients using enzalutamide and 12 healthy volunteers were included. A clear circadian rhythm of cortisol was described in healthy volunteers and unaffected in mCRPC patients using enzalutamide. Contrarily, a circadian rhythm of cortisol could not be identified in mCRPC patients using abiraterone acetate, due to continuous suppression throughout the day. Patients with an abiraterone Cmin > 8.4 ng/mL (n = 77) had a median cortisol concentration of 1.03 ng/mL vs. 2.59 ng/mL in patients with an abiraterone Cmin ≤8.4 ng/mL (n = 40) (p = 0.020). The median cortisol concentration in PSA responders (n = 63) was 1.02 ng/mL vs. 2.59 ng/mL for PSA non responders (n = 54) (p = 0.037). Patients in the highest cortisol tertile ( > 3.03 ng/mL) had a median PFS of 3.7 months (n = 39) vs. 13.8 months in patients with cortisol levels ≤3.03 ng/mL (n = 78) (p = 0.007). The median PFS was 16.3 months in patients with an abiraterone Cmin > 8.4 ng/mL and cortisol concentration < 3.03 ng/ml vs. 5.3 months in patients with an abiraterone Cmin > 8.4 ng/mL and cortisol concentration > 3.03 ng/ml (p = 0.02). Conclusions: This study demonstrates that cortisol levels are of additional value to abiraterone concentrations as a marker for abiraterone acetate efficacy. This might help be helpful to assess efficacy of abiraterone treatment.