Metabolism of adrenal androgen and its impacts on prostate cancer after castration

Abstract

With the extensive utilization of PSA, digital rectum examination and transrectal ultrasound for screening in the aging population, the diagnosis of prostate cancer in China has markedly increased during the past years, particularly in developed regions. It was reported that the prevalence of prostate cancer in Shanghai city from 1973 to 1975 was 1.6/100 000 persons and rose to 7.7/100 000 persons by 2000.1 Most prostate cancers are, at least initially, androgen dependent, which is the basis for endocrine treatment. However, the favorable response time is on average only 18 to 24 months after androgen deprivation therapy (ADT) before the progress of local advanced prostate cancers or metastatic disease.2 This phenomenon is referred to as the transformation of “androgen-independent” prostate cancer. The mechanism of this transformation is currently unknown, but the evidence for a contribution by adrenal androgen during the process is very compelling. There are a number of questions to be answered to understand this adrenal androgen contribution. Is the adrenal androgen responsible for the continuous growth of prostate cancer post-castration? If so, when does this occur? Is it correlated with the transformation process? Furthermore, does the adrenal androgen have any impact on stimulating the growth of the progressive type of prostate cancer? In attempts to improve the efficacy of ADT and the quality of life of patients, numerous basic and clinical studies investigating these issues have been conducted. It is essential to recapitulate and analyze this information, which will benefit our research and future clinical treatment. PHYSIOLOGIC METABOLISM OF ADRENAL ANDROGEN3 The circulating androgen in adult males is formed mainly in the testis and adrenal glands, and to a lesser extent in peripheral tissues4 such as the prostate and skin, although there is no direct evidence proving this. Recently, it was shown that the key steroidogenic enzymes were present in the skeletal muscle of rats and were capable of generating sex steroid hormones,5 thus suggesting that skeletal muscle may also contribute to circulating testosterone. The metabolism pathways and forms of adrenal androgen are complicated and have been demonstrated briefly in Figure. The follicular zone and reticular zone cells in the adrenal cortex synthesize three major inactive prohormones of androgen via cytochrome p450 enzymes, namely dehydroepiandrosterone (DHEA), androstenedione and dehydroepiandrosterone sulfate (DHEAS), the former two have more significance in physiology. DHEA is converted to testosterone by a two-step reduction reaction, in which 3β-hydroxysteroid dehydrogenase (3β-HSD) initially converts DHEA to androstenedione in peripheral tissue, which is then reversibly reduced to testosterone via 17β-hydroxysteroid dehydrogenase (17β-HSD), which is the major pathway. It may also be reversibly reduced to 5-androstenediol by 17β-HSD via the minor pathway. Subsequently, 5-androstenediol is irreversibly reduced to testosterone by 3β-HSD. DHEA and DHEAS can be reversibly interchanged through the catalysis of sulfotransferase, as well as steroid sulfatase. The DHEAS concentration in the plasma of healthy males is 100 to 500 times higher than other androgens, but its metabolism is very slow with its average clearance rate for 10 L/24 hours.6 DHEAS is not a very potent androgen.Figure.: Metabolism pathways of adrenal androgens. DHEA-ST: DHEA-sulfotrasnferase; STS: steroid sulfatase; 3β-HSD: 3β-hydroxysteroid dehydrogenase; 17β-HSD: 17β-hydroxysteroid dehydrogenase.Androstenedione is produced in the plasma of healthy males at a rate of 2 to 6 mg/day, with 20% of it being generated by peripheral metabolism from other steroids, giving rise to a low serum concentration of (150±54) ng/dl. The peripheral tissues take up androstenediones and convert them to other complexes immediately after they are secreted into the blood; they are reduced to testosterone chiefly by 17β-HSD. Testosterone is reduced to dihydrotestosterone (DHT) within prostate cells via type 2 of 5α-reductase, the latter is natures most potent androgen in humans with a 7 times higher affinity for the androgen receptor (AR) than that of testosterone. In fact, sulfatase, 3β-HSD, 17β-HSD and a reductase of type 5 are present in prostate cells; therefore, the reactions mentioned above can also occur in prostate cells following adrenal prohormones being taken up; Labrie7 termed this intracrinology. It was observed in early clinical practice that the prostate involuted almost completely, with the total cell mass reduced more than 90% after castration of healthy male humans or animals. This reveals that under normal physiologic conditions, androgen arising from the adrenal is a very weak stimulant for the growth of the prostate. Meanwhile, it was also noticed that hyperfunction of the adrenal cortex resulting from neoplasm or hyperplasia of the adrenal gland can give rise to abnormal virilism in immature males or hirsutism in women,8 indicating that overproduction of adrenal steroids due to hyperstimulation of the adrenal cortex can stimulate the growth of the prostate. More importantly, only the adrenal cortex of primate animals can exclusively produce significant amounts of androgen steroids, thus being completely different from that of many lab animals, including rats, mice and guinea pigs, whose secretion rates are very low (if any).7 STIMULATION OF ADRENAL ANDROGENS OF THE GROWTH OF PROSTATE CANCER Majority of androgens in prostate cancer tissue arising from adrenals after castration A previous study used labeled DHT that could be recovered in prostatic tissue from patients who had received tritium (3H)-labeled androstenedione or DHEAS 0.5 hour before total prostatectomy for treatment of benign prostate hyperplasia (BPH).9 The labeled DHT in the prostate represented 9% and 2%, respectively, of the total labeled steroids found in the prostate, which showed that adrenal androgen contributed to the androgen in prostatic tissue. Obviously, the contribution is not equal to the total residual percentage of androgen in prostatic tissue, or to the residual percentage of labeled androgen in prostatic tissue; it is yet unknown at present. Kashiwagi’s animal experiment also confirmed the above point. They found the androgen level in male rats declined significantly after castration; the serum testosterone was reduced to 4% of that measured in intact male rats, while DHT in prostatic tissue was reduced to 3%.10 This indicates that the adrenal glands are the main source of androgen in vivo after castration, given that an average of 20% of androgen measured in intact men remained after castration.11 Besides, it suggested that peripheral tissues may also produce some androgens. Considering that the half-life of serum testosterone is short (lesser than 30 minutes), and that the androgen level in plasma declines immediately after castration, it can be concluded that there is no androgen stored in vivo under normal physiological conditions. Nishiyama et al12 investigated the change of androgen concentration in both serum and prostatic tissue of 30 prostate cancer patients before and after ADT. They found that the DHT level in serum was correlated with both adrenal androgen and testosterone levels in serum (r=0.466 for Androstenedione, r=0.577 for DHEA, r=0.480 for DHEAS, respectively). These findings suggested that serum testosterone after ADT came mostly from adrenal androgens converted in the prostatic cells. It also indirectly revealed that the DHT in cancer tissue after ADT was derived from the conversion of adrenal androgens in peripheral tissues, including the prostate. It is probable that the prostate is the major DHT-producing organ, and the level of DHT in prostatic tissue is correlated with the level of adrenal androgens and testosterone in prostatic tissue. Futhermore, Page et al11 found that there was a strong relationship between serum DHEA and prostatic tissue androgen after medical castration (r=0.94, P=0.06 for serum DHEA/prostatic tissue T; r=0.99, P=0.006 for serum DHEA/prostatic tissue DHT), providing supporting evidence to the above proposal although the sample size was small. Mohler13 and Labrie7 expressed similar opinions; the latter believes that about 50% of andogen in prostatic tissue originates from DHEA. Dependence of progressive prostate cancer on androgen After castration or ADT, the DHT level in prostatic tissue decreases sharply with 80% order of magnitude,11 being similar to that in prostate cancer tissue, which was 75%-91%,12-15 although a substantial amount of androgen still remains. Thus, some changes in cancer cells are occurring due to the alternation of the microenviroment in which prostate cancer cells grow and these changes are listed in the following section. Amplification of AR gene16-21and increasing expression of AR (5.8 times)18-20,22-24 It was shown that 98% (52/53) of the tumors contained AR positive malignant epithelial cells and that prostate cancer patients with high AR levels (greater than 64% AR positive nuclear area) in the malignant epithelial cells recurred more frequently and earlier.22 Hypersensitivity of the AR to androgens16,25-27 Recurrent prostate cancer lines CWR-R1 and LNCaP-C4-2 had an increased sensitivity to the growth promoting effects of DHT, which was 3-4 orders of magnitude lower than that required for androgen-induced growth of androgen sensitivity cancer cells, and the increased sensitivity of the AR to androgen was not related to the increment of the AR affinity. 25,26 Persistence of high levels of androgen in prostate cancer tissue For normal males, the average value of the DHT content in their prostatic tissue was 1.9 ng/g,11 although there is no consensus value for prostate cancer patients. Various studies report the DHT concentration in CaP tissue as 1.45 pmol/g,13 1.25 pmol/g,14 0.69 ng/g (2.13 pmol/g),15 1.35 ng/g or 4.65 pmol/g.12 The variation may be the result of different pretreatments and measurement methods, which are reported in Figure. However, all of them were far higher than the lowest concentration required for stimulating the growth of prostate cancer cells. Increasing of the affinity of AR with the DHT Gregory et al25,26 documented that the lowest concentration of the DHT to stimulate the growth of recurrent CaP, CWR-R1 and LNCaP-C-42 cell lines was only 10-14 mol/L, which is equivalent to 10-5 pmol/g, 104 orders of magnitude lower than that required for androgen sensitive cancer cell lines, suggesting that the androgen level remaining in prostate cancer tissue after castration is sufficient to activate the androgen receptor. Amplification of multiple genes responsible for the metabolism of androgen in prostate cancer cells It was reported that the expression levels of HSD3B2, AKR1C3, SRD5A1, AKR1C1, AKR1C2 and UGT2B5 were significantly increased in androgen-independent prostatic tissues, 23 meanwhile, heightened expression of type 2 5α-reductase and type 5 17β-hydroxysteroid dehydrogenase induced the upregulation of intracellular conversion activity of androstenedione and DHT in situ.28 Mohler et al13 found that the ratio of testosterone versus DHT in prostate cancer tissue was increasing compared with that in BPH, suggesting the activity of 5α-reductase was increasing. At the same time, these findings also accounted partly for the persistent high level of DHT in prostate cancer tissue after castration. Maitenance of major transcription in androgen-independent cancer Although the andogen level in cancer tissue decreased after castration, that corresponds with a 2 to 3 fold decrement of the expression of genes regulated by androgen.23 It was found that the protein encoded by the PSA gene, regulated by androgen, was present in recurrent and/or progressive prostate cancer tissue, despite the fact that its content was only 9% in BPH tissue, which was 97.6 nmol/g;13 suggesting that AR was stabilized by combinding with its ligand and, in turn, activated by androgen in prostatic tissue. These above findings show that recurrent and/or progressive prostate cancer during ADT was not really “androgen-independent”, but continue growing depending on androgen. Those data also fully highlight that adrenal androgen is the major factor to stimulate the growth of prostate cancer after castration, associating intimately with the progression of CaP during ADT, or the transformation of “androgen-independence” of CaP. Thus, we should block the production of androgen derived from the adrenal after castration, or disrupt the function of adrenal androgen in prostate cancer patients when treating CaP in the clinic; a suggestion that is consistent with the conclusion of the following clinical epidemiological investigations. Better treatment effect of combining castration with suppression of adrenal androgen over that of simple castration Early studies employing bilateral surgical adrenalectomy or hypophysectomy to ablate adrenal androgen production resulted in pain relief in a majority of patients and objective responses in up to 30%.23 However, there was significant morbidity associated with this therapy, and it was replaced by medical adrenalectomy using aminoglutethimide or ketoconazole. Response rates of about 44% have been reported with ketoconazole (usually given with hydrocortisone) in men with androgen-independent prostate cancer.29 Significantly, this treatment only partially suppresses adrenal androgen production (by about 50%)30 and responses have been correlated with the level of adrenal suppression. Small et al30 also measured adrenal androgen levels in men with prostate cancer before and after ADT, finding that the DHEA and androstenedione concentrations in serum dropped an average 50% after ADT, decreasing to 1.0 ng/ml and 0.30 ng/ml, respectively, but increased with prostate cancer progression, although not by a statistically significant level. Unfortunately, the investigation did not determine corresponding DHT levels in prostate cancer tissue to discover the impact of suppression of adrenal androgen production on the androgen level in prostate cancer tissue, and the percentage contribution of adrenal androgen to DHT in prostatic tissue. Similar results have been found by Kruit et al.31 Recently, Ueno et al32 reviewed the initiating hormone treatment situation of 628 prostate cancer patients staged T1c to T3. Among them, 399 patients selected the treatment of combined androgen blockade (CAB, medical castration in addition to antiandrogen), another 229 patents were assigned to simple castration through medication or surgery. At 8 years of follow-up, the disease-specific survival rate of CAB treated patients was 95.3%, significantly higher than 80.5% for the castration treated patients. The progression-free survival rate of patients in the CAB group was 87.3%, whereas that in the medical or surgical castration group was 79.7%, the difference between them was not statistically significant. A large-scale meta-analysis of initial hormone management of prostate cancer, which was launched by the American Society of Clinical Oncology, pointed out that CAB conferred a statistically significant clinical improvement in survival over orchiectomy or LHRH monotherapy,33 which was supported by other meta-analysis papers in this field.34,35 These clinical studies showed “androgen-independent” prostate cancer still depended on androgen after castration, and suppression of the adrenal will benefit the therapy of prostate cancer, proposing a new target of treatment. However, it is not defined whether or not complete suppression of adrenal androgen following castration can inhibit the growth of prostate cancer in the long run. METABOLISM STATUS OF ADRENAL ANDROGEN AFTER CASTRATION Adrenal androgen can stimulate the growth of prostate cancer after castration, but is there any predictability? That is, the androgenic action only last for a while after castration. In terms of this point, when is the appropriate time to suppress the adrenal androgen? Otherwise is it continuously functioning? These issues are important aspects for treating prostate cancer by endocrinology in the clinic. Unfortunately, the relevant studies in this field are limited. Several recent studies involved androgen metabolism after castration, as summarized in Table.10-15,36 Not all of these papers are emphasizing the metabolism alteration of adrenal androgen after castration, but comparing the androgen concentration in serum and prostatic tissue prior to and after ADT with that of BPH as controls or self-control. They also sampled at only one or two time points, in addition, the time interval between castration and sampling might be too long to obtain an accurate changing profile of the metabolism of adrenal androgen. It was reported that the decrease of apoptosis and the increase of the proliferation of prostate epithelial cancer cells became most significant at 3 days after castration, and then returned to baseline levels at 10 days after castration.37 Thus it can be inferred that the metabolism of adrenal androgen may greatly fluctuate within several days after castration, than remain at a relative stable level. Page et al11 researched the castration situation in normal men, but they did only one biopsy for prostatic tissue and took 3 serum samples. Their sampling time interval after medication with the castration drug was long (28 days), the sample size was small (4 men in each group, 13 men in total), all of these weakness limit this paper’s usefulness for defining the metabolism profile of adrenal androgen after castration. Kashiwagi et al10 evaluated the technique of measuring the steroid hormone via an animal study, however the rats did not produce significant amounts of androgen in their adrenal glands, thus this study failed to answer the question. Regardless of those elements, some fragmentary impression can be achieved in terms of the metabolic status of adrenal androgen after castration according to the current available data.Table: Androgen metabolism after castrationIn another study, 22 prostate cancer patients were treated with ADT (4 cases of LHRH + flutamide, 5 cases of LHRH, 11 cases of surgical castration, 1 case of hypophysectomy, 1 case of DES), the androstenedione and DHEA levels in their cancer tissue decreased approximately 50% after therapy (no details were provided), tissue level of DHEAS was also reduced (no details were provided).13 Furthermore, tissue androgen levels were similar among patients who received flutamide plus castration and those who received castration alone, suggesting the secretion of adrenal androgen may not be affected by antiandrogen therapy. Nishiyama et al 12 found that among 30 cases of prostate cancer who received ADT (25 cases of LHRH plus flutamide, 5 cases of surgical castration plus flutamide) with self-control, the serum levels of androstenedione, DHEA and DHEAS before treatment were 0.81 ng/ml, 2.03 ng/ml, 1194.90 ng/ml, respectively; reduced to 0.42 ng/ml, 1.22 ng/ml and 761.30 ng/ml, respectively. That is to say, androstenedione and DHEA content decreased by approximately 50%; this is consistent with the results in Mohler’s paper.13 But Oka et al36 argued that surgical or medical castration, especially an LHRH-agonist, may simultaneously reduce androgens derived not only from the testis, but also adrenals. Taken together these data suggested that no significant changes occur in the average concentration of adrenal androgen in serum and prostatic tissue after castration. Page et al11 had also proven this point. What is the time course of metabolism changes to adrenal androgen after castration, either in serum or in prostatic tissue? The information on this subject is still limited. Probably it is due to multiple forms of the metabolites, trace content in serum and prostatic tissue and the inconvenience of establishing a platform to detect various steroid hormones simultaneously. Under most circumstances, ethical and moral considerations preclude consecutive biopsies of prostatic tissue in a very short time; and the fact that the adrenals of non-primate animals do not secrete significant amounts of androgen precursors is proved in animal models. Thus it is not currently clear how the metabolism of adrenal androgen changes after castration, and when adrenal androgens have a role in growth stimulation. But there is no doubt that these issues are crucial in determining the endocrine treatment of prostate cancer, namely when is the proper time to suppress the adrenals after castration. SUMMARY Adrenal androgen is the predominant source of residual androgens in prostatic tissue after castration and exerts a role in stimulating the continuous growth of prostate cancer. But it is not yet determined whether or not the prostate tumor ceases to grow or progresses after the complete suppression of the adrenal androgen after castration; or after totally blocking the function of androgens arising from the testis and adrenals. When is the proper time to suppress the adrenals? Answers to all of these problems rely on studies of the metabolism of adrenal androgen after castration. Currently, with the extensive application and maturity of chromatography/mass spectrometry analysis methods, there should be no problem in developing a novel approach platform to simultaneously detect various steroid hormones in both serum and prostatic tissue; monkeys can first be used as a lab animal model. Based on the data of monkey models, several cut-off time points of metabolism changes can be chosen to sample the prostate cancer tissue in CaP patients receiving ADT, the results will be collected and accumulated over time to characterize the metabolic profile.