Corticosteroids and Prostate Cancer: Friend or Foe?

Abstract

Corticosteroids have been a mainstay of treatment of men with metastatic prostate cancer (PCa) for decades. They are used for palliation of symptoms, to prevent adverse reactions or toxicity from other therapies, and for their antitumor effects mediated by way of the decrease of adrenal androgen production, transcription factors, and cytokines. Recently, two rationally designed oral drugs targeting the androgen receptor (AR) axis have become an important part of the armamentarium in treating men with castration-resistant PCa (CRPC). Abiraterone acetate, a CYP17A1 inhibitor, decreases androgen production in testicular, adrenal, and prostatic tissues; enzalutamide is a highly selective and effective AR antagonist that limits AR nuclear localization and downstream gene regulation. Both drugs improve survival in men with CRPC; however, unlike enzalutamide, abiraterone requires concurrent corticosteroid administration to mitigate its effect of increased mineralocorticoid production. Understanding the mechanisms of the inevitably acquired resistance to these drugs is vital, so the role of corticosteroids and the glucocorticoid receptor (GR) in CRPC is now being explored. CRPC progression can result from a number of ARmediated factors, including amplification of the AR, alternative splicing, and mutations of the AR that can lead to constitutive stimulation of the receptor, ligand-independent AR activation, and paradoxical stimulation of the receptor by other steroidal hormones and antiandrogens. As shown by the eventual clinical progression of CRPC despite highly effective AR-targeted therapies such as enzalutamide and abiraterone, there are undoubtedly other, non–ARmediated mechanisms of CRPC progression. One such posited mediator of AR-independent CRPC progression is the GR. The GR, which is activated by corticosteroids, is a nuclear hormone receptor in the same protein family as the AR, with structural homology and shared DNA response elements. Thus, the questions of how the GR and AR interact with corticosteroids and what their relationship to each other is in CRPC have been raised. GR protein levels in PCa tissue increase subsequent to AR blockade [1]. In addition, in preclinical models, corticosteroid activation of the GR can promote CRPC progression and resistance to effective second-generation AR antagonists such as enzalutamide [2,3]. A potential functional role for GR signaling in clinical CRPC progression is thus of great clinical interest. To address the role of theGR in CRPC, in thismonth’s issue of EuropeanUrology, a post hoc analysis byMontgomery et al. [4] examined whether baseline corticosteroid use in the phase 3 trial of abiraterone versus prednisone in men with metastatic CRPC after chemotherapy (COU-AA-301) was predictive of overall survival (OS) [5]. The study showed that the 33% of men who were receiving corticosteroids at baseline had universally worse disease characteristics at study entry than their counterparts, with inferior OS (16.1 vs 11.2mo; hazard ratio [HR]: 0.68; p < 0.0001) and an increased risk of death (HR: 1.48; p < 0.0001) on univariate analysis. However, in a multivariate model, baseline corticosteroid status did not predict OS when other clinical factors were included. These results suggest that baseline corticosteroid usewas an indicator of poor prognosis butwas not predictive of response to abiraterone. Several points should be highlighted. First, the use of abiraterone led to improvement in OS, prostate-specific antigen (PSA) progression, and progression-free survival regardless of baseline corticosteroid use. Second, all patients received prednisone once the trial commenced. Third, the duration of, and indications for, baseline EU RO P E AN URO L OG Y 6 7 ( 2 0 1 5 ) 8 7 4 – 8 7 5