Targeting Glucocorticoid Metabolism in Prostate Cancer

Abstract

In the treatment of metastatic prostate cancer, resistance to hormonal therapy is a major obstacle. With antiandrogen therapies that suppress androgen signaling through the androgen receptor (AR), the primary driver of prostate cancer, some malignancies are able take advantage of the closely related glucocorticoid receptor (GR). Escape from AR dependency often involves a simple functional switch from 1 steroid receptor to another. Recent research efforts have outlined the mechanism enabling this switch, which involves alterations in glucocorticoid metabolism that occur with antiandrogen therapy to increase tumor tissue glucocorticoids and enable GR signaling. Targeting this mechanism pharmacologically by blocking hexose-6-phosphate dehydrogenase shows promise in normalizing glucocorticoid metabolism and restoring responsiveness to antiandrogen therapy. This perspective reviews what we have learned about this resistance mechanism, examines potential implications, and considers how this knowledge might be harnessed for therapeutic benefit.

Prostate cancer is a disease primarily driven through activation of the androgen receptor (AR) by androgens. Suppression of AR signaling through androgen deprivation therapy (ie, medical or surgical castration) along with next-generation AR antagonists such as enzalutamide is a standard therapy for treating advanced-stage prostate cancer. Acquired resistance to these therapies, however, results in disease lethality and remains a major challenge in prostate cancer treatment. Resistance generally develops through restoration of AR signaling despite castration levels of circulating androgens. Numerous mechanistic pathways have been described that may contribute to this apparent sustainment of AR signaling, including AR amplification, mutations in AR, and bypass of AR activation through activation of the glucocorticoid receptor (GR). Tumor upregulation of GR has been demonstrated to be an especially prevalent mechanism of resistance (1). Recently, there have been some exciting advances in our understanding of this mechanism, with altered glucocorticoid metabolism occurring with AR suppression therapy underlying enhanced GR signaling and resulting in resistance (2, 3). A study from our group identified a pharmacologic target that normalizes glucocorticoid metabolism with AR suppression therapy in models of prostate cancer and reinstates enzalutamide responsiveness (4). This perspective highlights these recent advances, considers how more effective therapies might take aim at this metabolic resistance mechanism, and suggests ways in which the elucidation of the physiological resistance mechanisms to hormonal therapy in prostate cancer could help shed light on divergent aspects of physiology that are regulated by steroid hormones.

A Balance Between 2 Steroid Hormone Receptors

A vital role for GR in the development of resistance to antiandrogen therapy was demonstrated contemporaneously by 2 research groups. Arora et al. found that induction of GR expression occurs in a substantial subset of tumors resistant to enzalutamide and apalutamide (1). Induction of both the GR gene and protein was also found by Isikbay et al. to occur in AR-dependent prostate cancer cell lines treated with an AR antagonist and in tumor xenografts in response to castration alone (5). Support for a functional role of GR signaling in antiandrogen resistance was then demonstrated. In prostate cancer cell lines treated with a GR agonist, enzalutamide resistance developed, whereas treatment with a GR antagonist restored sensitivity to enzalutamide (1, 5). Furthermore, depletion of GR via shRNA delayed castrate-resistant tumor formation in mouse xenografts (5).

AR and GR are both members of the same intracellular receptor family of transcriptional regulators. The tumor-promoting effects of the activated androgen-AR complex occurs with binding to androgen response elements of target genes. In investigating the extent of overlap between the target genes of AR and GR using chromatin immunoprecipitation sequencing, Arora et al. found that more than one-half of AR binding sites were shared with GR, including KLK3, the gene encoding prostate-specific antigen (PSA), a serum biomarker for prostate cancer, and another canonical target in prostate cancer, TMPRSS2 (1). A targeted approach to this question confirmed upregulation of KLK3 with GR activation in addition to other AR targets (5). Taken together, these studies suggest that with antiandrogen therapy in prostate cancer, AR is functionally replaced by GR.

From a therapeutic standpoint, the major role of GR in sustaining AR target gene expression suggests 2 things. For one, dual targeting of GR and AR in advanced-stage prostate cancer may prove beneficial. There are potential problems with this approach, however, because total systemic GR antagonism may result in adrenal insufficiency. An ongoing clinical trial (NCT02012296) is testing the effect of mifepristone, a nonspecific GR antagonist, in combination with enzalutamide on PSA progression but initial findings do not indicate a beneficial effect of mifepristone. Another implication of these results is a need to reexamine how glucocorticoids are used in supplemental therapies for men with castration-resistant prostate cancer. Although glucocorticoids can aid in lowering adrenal androgen precursors through suppression of ACTH and reduce the side effects from systemic therapies (eg, abiraterone), their direct activation of GR is an alternative activity that should be considered.

An Adaptation in Glucocorticoid Metabolism With 11β-hydroxysteroid Dehydrogenase 2 Loss Allows for Enhanced GR Stimulation With AR Suppression

Activation of GR is determined not only by its expression that poises it for signaling, but also by the availability of its primary ligand in humans, cortisol. Because increased activation of GR was shown to confer resistance to antiandrogens, our group hypothesized that a metabolic mechanism that allows for sustained cortisol levels may be a critical aspect of GR activation in this context. Two isoenzymes of 11β-hydroxysteroid dehydrogenase (11β-HSD) catalyze the interconversion of the active glucocorticoid, cortisol, with the inactive glucocorticoid, cortisone. 11β-HSD1 catalyzes the generation of active cortisol and 11β-HSD2 catalyzes the inactivation of cortisol to cortisone. In examination of both cell line models of prostate cancer and patient tissues treated with enzalutamide, Li et al. identified profound loss of 11β-HSD2, which was shown to be mediated by the ubiquitin E3-ligase autocrine mobility factor receptor (2). This reduction in 11β-HSD2 with no accompanying change in 11β-HSD1 maintains high tumor cortisol levels and allows for GR takeover (2). It also appears that this alteration in glucocorticoid metabolism may not be limited to enzalutamide treatment. A neoadjuvant clinical trial showed that abiraterone treatment also increases prostate tissue cortisol concentrations that occurs alongside GR upregulation (6).

An alternative therapeutic strategy to GR antagonists for combatting GR-mediated antiandrogen resistance is targeting metabolism of glucocorticoids and thereby the availability of ligand for GR. Indeed, drugs targeting androgen metabolism, such as abiraterone, have had profound clinical success. In our initial study, reinstatement of 11β-HSD2 in mouse xenograft tumors reversed enzalutamide resistance (2). Therapeutic replacement of 11β-HSD2 is challenging, but approaches targeting this enzyme indirectly may be especially useful. With the use of glucocorticoids (eg, prednisolone, dexamethasone) as part of standard therapies, these findings also suggest that it may be important to consider whether pharmacologically administered glucocorticoids are inactivated by 11β-HSD2. For example, dexamethasone is impervious to inactivation by 11β-HSD2 and suppresses endogenous cortisol, which in contrast is a substrate and inactivated by this enzyme.

Enhanced Cortisol Availability Through Loss of 11β-HSD2 With AR Suppression Therapy Extends Beyond the Tumor

AR antagonists are systemic therapies, with potentially widespread reach beyond prostate cancer depending on the responsiveness of body tissues. Because 11β-HSD2 expression occurs in various tissues, Alyamani et al. investigated whether systemic glucocorticoid levels are altered in men undergoing AR suppression therapy. Through analysis of serum samples from patients undergoing AR antagonist treatment (enzalutamide or apalutamide), a rise in circulating cortisol/cortisone ratios was found with AR antagonist treatment (3). A major site of action for 11β-HSD2 is the distal tubule of the kidney, where it protects the mineralocorticoid receptor from exposure to agonistic activity of cortisol. This study identified AR expression in the kidney of men coincident with 11β-HSD2 expression (3), thus providing a plausible mechanism by which AR antagonists suppress renal 11β-HSD2, thereby increasing systemic cortisol levels. Curiously, higher cortisol/cortisone ratios in patients receiving enzalutamide were associated with better clinical outcomes (PSA and radiographic progression-free survival) (3). It is possible that higher cortisol/cortisone ratios are reflective of on-target AR suppression with enzalutamide and might be a pharmacodynamic biomarker.

An important conclusion drawn from this study is that glucocorticoid exposure is augmented across multiple tissues under AR antagonist therapy. The results of Alyamani et al. demonstrate elevated circulating cortisol/cortisone ratio but it is important to consider that local levels in peripheral tissues may be elevated to an even greater extent than circulating levels because of lowered local tissue expression of 11β-HSD2, similar to that which occurs in prostate tumors. These findings also potentially hold relevance in understanding the various adverse peripheral effects of AR antagonist treatment in men with prostate cancer. Hypertension, central nervous system effects, bone fractures, and immunosuppression are just a number of peripheral effects that may occur with 11β-HSD2 loss, and there is reason to believe that elevated glucocorticoid exposure may be in part responsible for some these effects in patients treated with potent AR antagonists.

Targeting Glucocorticoid Metabolism through Hexose-6-phosphate Dehydrogenase Holds Promise in Combatting GR-mediated Resistance to AR Antagonists

In another important step in understanding altered glucocorticoid metabolism with AR suppression therapy, a recently published study highlighted a role for the enzyme, hexose-6-phosphate dehydrogenase (H6PD) (4). H6PD within the endoplasmic reticulum generates the NADPH necessary for 11β-HSD1-mediated reduction of cortisone to cortisol (7). Li et al. found that H6PD was upregulated in prostate cancer tissues of men receiving enzalutamide treatment as well as in tumor xenografts (4). Genetic silencing of H6PD in human-derived enzalutamide-resistant cell line models shifted the 11β-HSD oxidoreductive balance through blocking the NADPH generation necessary for the activity of 11β-HSD1, effectively reinstating 11β-HSD2 function and normalizing glucocorticoid metabolism (4). Furthermore, pharmacologic blockade of H6PD through rucaparib, an inhibitor with modest efficacy, reversed enzalutamide resistance in xenograft models (4).

The results of this study raise hope in the utility of a therapeutic strategy that lowers H6PD activity in men with prostate tumors exhibiting antiandrogen resistance through glucocorticoid signaling. The existence of humans with loss-of-function mutations in H6PD, which causes apparent cortisone reductase deficiency, suggest that loss of H6PD is tolerable. Rucaparib is currently in clinical trials in combination with enzalutamide, but more potent inhibitors will likely provide more targeted, beneficial therapy.

Potential Implications on Generation of 11-Oxy Androgens and Related Physiology

The demonstrated changes in 11β-HSD isoenzyme activity under antiandrogen therapy raises questions for the course of prostate cancer progression with antiandrogen therapy because of how these changes in 11β-HSD isoenzymes might ultimately have some contradictory effects through altering androgen synthesis. As reviewed by Barnard et al., C11-oxy androgens are increasingly being recognized as important androgens, including in prostate cancer, and are also substrates for the 11β-HSD isoenzymes (8). Whether increased generation of potent androgens will have any measurable effect in the context of complete or near-complete AR antagonism is unclear, but the potential of this as an additional source of resistance deserves further study.

Beyond prostate cancer, a deeper understanding of how H6PD and 11β-HSD isoenzymes affect 11-oxy steroid metabolism could enhance our understanding of other aspects of physiology where 11-oxy androgens play a central role, particularly in conditions of androgen excess in women, including premature adrenarche and polycystic ovary syndrome. Additionally, a dominant role of 11-oxy androgens is well-studied in fish, which have 11β-OH-testosterone instead of testosterone as the major androgen. Because many of the mechanisms governing steroidogenesis are shared across species, future efforts should be placed on integrating the findings on this topic across vertebrate species.

Perspectives and Conclusions

Resistance to androgen suppression therapy is a crucial challenge we must overcome to successfully treat castration-resistant prostate cancer. The ease with which many tumors switch from AR-mediated tumor progression to a GR-mediated pathway has emerged as an important mechanism of resistance upon antiandrogen treatment. Recent progress in untangling this mechanism identifies an adaptation in glucocorticoid metabolism that results in elevated tumor cortisol levels within the tumor and allows for GR takeover. These changes in glucocorticoid metabolism occur systemically as well and may play an important role in promoting various adverse effects in men undergoing enzalutamide or apalutamide therapy. Shifting the activity of 11β-HSD isoenzymes responsible for the interconversion of glucocorticoids by blocking H6PD, an enzyme whose activity promotes the directionality of this reaction towards the production of cortisol, holds promise in combatting this resistance mechanism by depleting tumor glucocorticoids and suppressing GR signaling. Work remains, however, to effectively and safely block H6PD pharmacologically. Furthermore, an improved understanding of the potentially diverse, systemic effects of both glucocorticoids and the 11β-HSD enzymes is crucially needed to inform how therapies that modify glucocorticoid availability or metabolism, including those drugs currently used in supplemental therapies for prostate cancer, may ultimately affect patient outcomes. A deeper understanding of these mechanisms may additionally prove useful in other conditions aside from prostate cancer that are also driven by steroid metabolism, steroid hormones, and their respective receptors.

Additional Information

Disclosures: The authors declare no conflicts of interest.

Data Availability

Data sharing is not applicable to this article as no datasets were generated or analyzed during the current study.