Steroids in prostate cancer: The jury is still out⋯ and even more confused

In cancer patients, steroids are often used to palliate symptoms related to coexisting inflammation due to the presence of the tumor itself in some metastatic sites (eg, the spine) or to certain anticancer treatments (eg, radiation therapy). They are also used as adjuvants to painkiller drugs or to antiemetics, as anabolizing agents, as lympholytic agents (in some hematologic malignancies), or as replacement therapy (eg, in patients with hypocorticosurrenalism due to different causes). The most exploited mechanism of action (MOA) of steroids remains the anti-inflammatory MOA. However, from other medical specialties (eg, rheumatology), we have learned that the relative anti-inflammatory potency of steroids varies greatly. Taking cortisol (ie, hydrocortisone), for example, as the standard reference and considering just the oral route of administration, the glucocorticoid potency of dexamethasone is reportedly 25–80 but is just 4 for prednisolone. In castration-resistant prostate cancer (CRPC), steroids may act as anticancer drugs; in this case, the MOA of corticosteroids has been linked putatively to inhibition of the pituitary production of adrenocorticotropic hormone, leading to reduced synthesis of adrenal androgens [1]. However, the pharmacologic activity of glucocorticoids on cell proliferation is more complex and involves inhibition of transcription of multiple genes (including cyclooxygenase 2, cytokines, cell adhesion molecules, and nitric oxide synthase), blockage of genomic effects of vitamin D3, and increased synthesis of annexin-1. All of these effects are proportional to glucocorticoid potency. In addition, steroids are able to inhibit the production of vascular endothelial growth factor by tumor cells, thus reducing neoangiogenesis and lymphangiogenesis [2]. In this issue of the Platinum Journal, Venkitaraman and colleagues reported the results of a randomized phase 2 trial comparing continuous low-dose dexamethasone, intermittent high-dose dexamethasone, and prednisolone in CRPC patients [3]. With prostate-specific antigen (PSA) response rate as the primary end point of their study, the authors observed higher efficacy for dexamethasone in this setting [3]. Furthermore, the intermittent high-dose dexamethasone arm was dropped after no response was seen in the first patients treated [3], suggesting that intermittent dexamethasone (given as premedication or as an antiemetic) does not have significant antitumor activity in CRPC. Beyond some methodological limitations (mainly relative to the small size of the population studied), the main result of this study—the superiority of dexamethasone—is somewhat expected, especially if we consider the above-reported different potency of the two steroids. What this study does not help us understand is the role of corticosteroids in prostate cancer—“friend or foe,” as another recent editorial published in the Platinum Journal asked [4], without finding a clear-cut answer. What is difficult to understand is why, according to the study of Venkitaraman and colleagues, the two steroids tested demonstrated activity against CRPC in terms of confirmed PSA response rates, time to PSA progression, and overall response rates, at least in those patients with measurable disease [3]; in the subgroup analyses of the abiraterone and enzalutamide registrative studies, the use of corticosteroids was associated with poorer prognosis [5] and [6]. Several mechanisms have been proposed over time and have been the subjects of a number of recent, authoritative papers, but the contradiction outlined above remains unresolved. In fact, the multifaceted role of corticosteroids is underscored by evidence that they may also promote androgen-independent growth [7] and [8]. Indeed, induction of glucocorticoid receptor (GR) expression is a common feature of anti–androgen-resistant tumors in preclinical models and patient samples. GR may replace androgen receptor (AR) signaling and activate a number of genes associated with maintenance of the anti–androgen-resistant phenotype. Dexamethasone drives enzalutamide resistance, and short-term AR inhibition results in GR upregulation in prostate cancer cells due to AR-mediated negative feedback [7]. Moreover, Carreira et al [8] showed a temporal association between clinical progression and appearance of AR mutations selected by glucocorticoids in about 20% of patients progressing on abiraterone. Before the availability of novel agents such as abiraterone and enzalutamide, this issue could have been regarded as trivial. Today, this uncertainty is no longer acceptable. Translational research aimed at elucidating the complex effect of steroids on prostate cancer cells (eg, the potential to inhibit or promote prostate cancer cell proliferation), their role in modifying tumor milieu (with the coparticipation of other cell populations), and better understanding of the interaction between the AR and the GR pathways will be key to solving the above-mentioned contradiction and, more importantly, to understanding the results of recent and future trials of novel hormonal agents for CRPC.