The Influence of Androgen Deprivation Therapy on Dihydrotestosterone Levels in the Prostatic Tissue of Patients with Prostate Cancer

Since the observation of Huggins and Hodges (1) that disseminated prostate cancer reacts favorably to castration or the administration of estrogenic hormones, first-line hormonal therapy has been used to impair the production or activity of androgens or both.

Testosterone is converted to dihydrotestosterone by 5-α reductase in the prostate. There have been several reports that examined in detail the method for quantitative analysis of the tissue dihydrotestosterone concentrations of the prostate (2, 3, 4, 5). Belanger et al. (5) and Labrie et al. (6) stated that after the elimination of testicular androgens, the intraprostatic concentration of dihydrotestosterone remains at ∼40%. These data indicate that a substantial level of dihydrotestosterone remains in the prostate after castration. Belanger et al. (5) and Labrie et al. (6) also suggested that dihydrotestosterone completely disappears from the prostate after androgen deprivation therapy with castration and flutamide. The influence of androgen deprivation therapy on dihydrotestosterone levels in the prostatic tissue in prostate cancer, however, is not fully known. Changes in dihydrotestosterone levels in the prostatic tissue during androgen deprivation therapy for prostate cancer in the same patients have not been reported. One of the reasons is that the detectable quantity of dihydrotestosterone involved in the prostatic tissue collected from needle biopsy samples is minute. We, however, have developed a detection system for minuscule quantities of dihydrotestosterone with liquid chromatography/electrospray ionization-mass spectrometry after polar derivatization of dihydrotestosterone (7).

Therefore, we analyzed dihydrotestosterone levels in prostatic tissue and endogenous hormone levels in serum both in patients with prostate cancer and those without prostate cancer who underwent prostatic biopsy. The patients diagnosed with clinically localized prostate cancer, furthermore, were treated with androgen deprivation therapy in a neoadjuvant setting for 6 months. We then carried out rebiopsy or prostatectomy 6 months after androgen deprivation therapy treatment to analyze dihydrotestosterone levels in prostatic tissue and endogenous hormone levels in serum.

Between April 2000 and October 2002, 103 patients suspected of having prostate cancer underwent prostatic biopsy. Those patients diagnosed with clinically localized prostate cancer were given androgen deprivation therapy (castration and flutamide) in a neoadjuvant setting for 6 months. Baseline patients’ characteristics are listed in Table 1. This research was reviewed and approved by the Institutional Review Board. Written informed consent was obtained from all participants.

To determine dihydrotestosterone levels, the samples of prostatic tissue were obtained from the midlateral region of the prostate with a 16-gauge biopsy needle; alternatively, prostatectomy specimens were used. Serum samples for endocrine study were collected from the patients between 9:00 and 12:00 p.m. (noon). In all patients who underwent ultrasound-guided biopsy or radical prostatectomy, serum samples were obtained before the respective interventions. Serum samples were stored at −20°C until additionally processed. All biopsies and prostatectomy specimens were analyzed by conventional pathological examination. Tissue samples were stored at −80°C until additional processing.

The prostate-specific antigen and hormones were quantified by commercially available immunoassays: prostate-specific antigen [TOSOH-II (PA)], luteinizing hormone, and follicle-stimulating hormone. All hormones were quantified by automated fluorescence polarization assays on Tosoh equipment (Tosoh Corporation, Tokyo, Japan). Serum levels of testosterone, adrenocorticorticotropic hormone (ACTH), cortisol, androstenedione, dehydroepiandrosterone (DHEA), DHEA sulfate (DHEA-S), and prolactin were determined by radioimmunoassay (BML, Tokyo, Japan).

The dihydrotestosterone levels in prostatic tissue and serum were analyzed by liquid chromatography/electrospray ionization-mass spectrometry after polar derivatization of dihydrotestosterone, as described previously (7). The polar derivatization method for electrospray ionization was developed and applied to the sensitive analysis of dihydrotestosterone. Dihydrotestosterone in prostatic tissue was dissolved in alkaline solution and extracted via a solid-phase column and derivatized to N-methylpyridinium-dihydrotestosterone as a polar derivative. N-Methylpyridinium-dihydrotestosterone was purified by Bond Elut C18 and determined with a semi–micro-liquid chromatography/electrospray ionization-mass spectrometry with selected reaction monitoring. The calibration graph was linear from 5 to 100 pg/tube. The lowest dihydrotestosterone level in this method was 5 pg/tube.

Statistical comparison of hormonal levels in patients with prostate cancer before treatment and noncancer patients at diagnosis was carried out with the Mann-Whitney U test. Statistical comparison of DHEA level in patients with prostate cancer before treatment and noncancer patients at diagnosis was also carried out with a multivariate analysis with logistic regression after forcing age in the model. Statistical comparison of the influence of androgen deprivation therapy on hormonal levels was carried out with the Wilcoxon’s signed rank test. The correlation between the dihydrotestosterone levels or ACTH and ACTH or other androgens was analyzed with the Spearman rank correlation coefficient. The test was two-sided, and a P value of <0.05 was considered statistically significant. Statistical analyses were carried out with SPSS software v.11.0 for PC (SPSS, Inc., Chicago, IL).

Sixty-nine patients were diagnosed as having prostate cancer and 34 as having a nonmalignant prostate condition. The patients’ characteristics are listed in Table 1. Thirty of the 69 patients were treated with androgen deprivation therapy with a luteinizing hormone-releasing hormone agonist (goserelin acetate or leuprolide acetate) or bilateral orchiectomy and flutamide in a neoadjuvant setting for 6 months. Eight of the 30 patients were withdrawn from flutamide treatment because of adverse effects during the following dosing periods: 1 month, 1 patient; 2 months, 1 patient; 4 months, 3 patients; and 5 months, 3 patients. Six patients were withdrawn because of liver dysfunction and two because of diarrhea.

In this study, the serum DHEA level was significantly lower in patients with prostate cancer by comparison with noncancer patients using the Mann-Whitney U test (P = 0.037; Table 2). However, there is no statistical association between prostate cancer and DHEA level using results of the logistic regression model (P = 0.762). There were no statistically significant differences between the patients with prostate cancer and the patients without prostate cancer in LH (P = 0.165), follicle-stimulating hormone (P = 0.206), prolactin (P = 0.169), ACTH (P = 0.788), cortisol (P = 0.770), testosterone (P = 0.539), androstenedione (P = 0.509), and DHEA-S (P = 0.404), including dihydrotestosterone levels in serum (P = 0.602) and prostatic tissue (P = 0.302) in this study. There were no statistically significant differences between the patients with prostate cancer and the patients without prostate cancer with respect to the ratios of testosterone to serum dihydrotestosterone (P = 0.772) and dihydrotestosterone in prostatic tissue (P = 0.191).

The level of dihydrotestosterone in prostatic tissue before androgen deprivation therapy was not correlated with the serum level of testosterone (rs = 0.010, P = 0.923; Table 3). The level of dihydrotestosterone in prostatic tissue was correlated with the serum levels of DHEA (rs = 0.243, P = 0.014) and DHEA-S (rs = 0.239, P = 0.015). There was a small correlation between the serum level of dihydrotestosterone and the level of dihydrotestosterone in prostatic tissue (rs = 0.229, P = 0.025, y = 0.001x + 5.0165). The serum level of dihydrotestosterone was correlated with the serum levels of testosterone (rs = 0.425, P < 0.001) and DHEA (rs = 0.305, P = 0.003).

The serum levels of ACTH (P < 0.001), testosterone (P < 0.001), androstenedione (P < 0.001), DHEA (P = 0.001), DHEA-S (P < 0.001), and dihydrotestosterone (P < 0.001) and the level of dihydrotestosterone in prostatic tissue (P < 0.001) significantly declined after androgen deprivation therapy (Table 4). The dihydrotestosterone levels in prostatic tissue after androgen deprivation therapy, however, remained at ∼25% of those measured before androgen deprivation therapy. The dihydrotestosterone levels in serum decreased to ∼7.5% after androgen deprivation therapy. Testosterone levels decreased to ∼2.7% after androgen deprivation therapy, and serum hormone levels were reduced to 59% for ACTH, 52% for androstendione, 60% for DHEA, and 64% for DHEA-S. The decrease in adrenal androgens in the flutamide withdrawal cases was less significant than that in the flutamide cases. The prolactin level (P = 0.737) and cortisol level (P = 0.148) in serum did not decline after androgen deprivation therapy.

The level of dihydrotestosterone in prostatic tissue was correlated with the serum level of testosterone (rs = 0.390, P = 0.033; Table 5). The level of dihydrotestosterone in prostatic tissue was not correlated with the serum level of androgens other than testosterone. The serum level of dihydrotestosterone was correlated with the serum levels of androstenedione (rs = 0.466, P = 0.009), DHEA (rs = 0.577, P = 0.001), and DHEA-S (rs = 0.480, P = 0.007). There was no correlation between the serum level of dihydrotestosterone and the level of dihydrotestosterone in prostatic tissue (rs = 0.013, P = 0.869). There was no correlation between the serum level of ACTH and the serum levels of androgens and the level of dihydrotestosterone in prostatic tissue.

Our results showed that after androgen deprivation therapy with castration and flutamide, the dihydrotestosterone level in prostatic tissue remained at ∼25% of the amount measured before androgen deprivation therapy in the same patients. Previous reports revealed that the mean dihydrotestosterone levels in the prostate tissue treated with androgen deprivation therapy were between 10 and 40% of those of untreated prostrate tissue (2, 3, 4, 5). Mohler et al. (8) showed that the dihydrotestosterone level in recurrent prostate cancer tissue was decreased to 18% of the level in benign prostate tissue. Belanger et al. (5) and Labrie et al. (6) indicated that androgen deprivation therapy with castration and flutamide decreases intraprostatic dihydrotestosterone to the point where it is undetectable. Our data, however, indicates that flutamide acts to suppress the binding of the residual dihydrotestosterone to androgen receptors, not to decrease intraprostatic dihydrotestosterone to undetectable levels.

It is not clear to what extent the testosterone and dihydrotestosterone in prostate tissue derives from adrenal androgens or other steroid precursors. Previous reports showed that persistent levels of prostatic dihydrotestosterone after castration are derived from adrenal androgens in the prostate (3, 5, 8). A sulfatase is present in human prostate that converts DHEA-S to DHEA (9). The plasma concentration of DHEA-S is 100 to 500 times higher than that of testosterone. Koh et al. (10, 11) revealed that prostate cancer cells have the ability to convert adrenal androgens to dihydrotestosterone intracellularly. Mohler et al. (8) revealed that recurrent prostate cancer tissue levels of adrenal androgens were ∼50% the levels in benign prostate. In our data, the level of dihydrotestosterone in prostatic tissue before androgen deprivation therapy was not correlated with the serum level of testosterone (Table 3). The level of dihydrotestosterone in prostatic tissue after androgen deprivation therapy was only correlated with the serum level of testosterone (Table 5). The level of dihydrotestosterone in prostatic tissue before androgen deprivation therapy was correlated with the serum level of adrenal androgens other than androstenedione (Table 3). The serum dihydrotestosterone level after androgen deprivation therapy was correlated with serum levels of adrenal androgen (Table 5). These findings suggest that serum testosterone after androgen deprivation therapy mostly comes from adrenal androgens converted in the prostatic cells. These findings could also suggest that serum dihydrotestosterone after androgen deprivation therapy comes from adrenal androgens converted in the peripheral tissues, including the prostate. It is possible that the prostate is the major dihydrotestosterone-producing organ, and the level of dihydrotestosterone in prostatic tissue is correlated with the level of adrenal androgens and testosterone in prostatic tissue. These results reveal that the source of dihydrotestosterone in prostatic tissue after androgen deprivation therapy involves intracrine production within the prostate to convert adrenal androgens to dihydrotestosterone.

The serum hormone levels were reduced to ∼60% in ACTH, androstenedione, DHEA, and DHEA-S after androgen deprivation therapy with castration and flutamide in our study. The mechanism causing the decrease of adrenal androgens after androgen deprivation therapy has yet to be determined (12, 13, 14). Several investigators have shown the effects of flutamide on the plasma levels of adrenal androgens (12, 14). Flutamide allegedly decreases adrenal androgens after treatment by castration and flutamide (13, 14). Our results also showed that castration and flutamide reduced adrenal androgens to ∼60%. The serum ACTH level after androgen deprivation therapy was not correlated with serum adrenal androgen levels and dihydrotestosterone levels in serum and prostatic tissue. Prolactin and cortisol in serum did not decline after androgen deprivation therapy. The mechanism of the suppression of adrenal androgens is speculated to be by flutamide having an inhibitory effect on human adrenal microsomal 17α-hydroxylase and 17,20-lyase activities (14).

A recent collaborative meta-analysis has shown that the addition of a nonsteroidal antiandrogen (flutamide or nilutamide) to castration reduced highly significantly the risk of death (all causes of death) by 8% (95% confidence interval, 3–13; P = 0.005), which translates to a small but significant improvement in 5-year survival of 2.9% over castration alone (15). Most meta-analyses show a positive result with nonsteroidal antiandrogens (16). The percentage of PSA responses has been shown to be significantly higher among patients receiving androgen deprivation therapy composed of castration and flutamide than among patients undergoing castration only (17). Labrie et al. (18) showed long-term and continuous androgen deprivation therapy could offer the possibility of long-term control or possible cure of localized prostate cancer. It is established that 5α-reductase inhibitor such as finasteride can reduce intraprostatic levels of dihydrotestosterone (19). Visakorpi et al. (20) showed that amplification of the androgen receptor gene is increased during androgen deprivation therapy. Gregory et al. (21) showed that the androgen receptor is transcriptionally active in recurrent prostate cancer and can increase cell proliferation at the low levels of androgen that occur after androgen deprivation therapy. Zegarra-Moro et al. (22) revealed that therapies that target the androgen receptor directly with an androgen receptor antibody or androgen receptor ribozymes inhibited growth of both androgen-sensitive and androgen-refractory prostate cancer in vitro. Chen et al. (23) revealed that the increase in androgen receptor mRNA and protein was both necessary and sufficient to convert prostate cancer growth from a hormone-sensitive to a hormone-refractory stage and was dependent on a functional ligand-binding domain. Increased levels of androgen receptor confer resistance to antiandrogens by amplifying signal output from low levels of residual ligand and by altering the normal response to antagonists (23). Leibowitz and Tucker (24) revealed that triple androgen blockade therapy followed by finasteride maintenance appears to be a promising alternative for the management of patients with clinically localized or locally advanced prostate cancer. These findings and our results suggest that new therapies that target androgen receptor and prevent formation of androgens within prostate cancer cells such as treatment with a combination of antiandrogens and 5α-reductase inhibitors can block the stimulation from adrenal androgens that contributes ∼25% of total dihydrotestosterone when they are combined with testicular suppression of androgens and may offer the most effective androgen deprivation therapy to prolong remission of prostate cancer as of now.