Division of Endocrinology, Department of Medicine, Harbor-UCLA Medical Center and Research and Education Institute, Torrance, California 90509
Address all correspondence and requests for reprints to: Christina Wang, M.D., Harbor-UCLA Medical Center, 1000 West Carson Street, Torrance, California 90509-2910. E-mail: . wang{at}gcrc.rei.edu
How does T work on target organs?
T is a remarkable hormone that has complex mechanisms of action. It may act directly on the nuclear ARs, be converted by the 5 reductase enzymes to metabolites such as 5
dihydrotestosterone (DHT) before acting on the AR, or aromatized to estrogens (E2) that act on the nuclear ERs. The steroid-steroid receptor complex induces transcriptional regulation through the steroid response elements on the promoter region of the target gene. A complex scheme of steroid-specific coactivators and corepressors modulate this transcriptional activity. In addition, T and its metabolites may act on the cell surface in a nuclear receptor-independent fashion to exert acute nongenomic actions. The multiple mechanisms by which T can act provide a high degree of tissue and cellular flexibility and an opportunity to use androgens with selective properties to exert the desired effects.
What about T treatment for the andropause?
It is now generally agreed that male aging is associated with a slow and progressive decrease in serum T concentrations (1, 2, 3). The decreases in serum T may be accompanied by a constellation of symptoms including sexual dysfunction, lack of energy, loss of muscle and bone mass, increased frailty, loss of balance, cognitive impairment, and decreased general well beinga condition termed "andropause" or "androgen deficiency of aging men" (2). Some of these clinical symptoms are relieved by replacement therapy with intramuscular T injections or transdermal T applications. Short-term studies have reported that T replacement resulted in variable improvement in sexual function, muscle and bone mass, and quality of life in older men (4, 5, 6). In a placebo-controlled randomized clinical trial, a transdermal T patch administered for 3 yr decreased body fat and increased lean body mass but failed to demonstrate significant changes in muscle strength or bone mass. This study has been criticized by some for its design limitations, including its inclusion of men with T levels within the normal range for young healthy men. Nevertheless, the authors provided further analyses demonstrating that bone mass was increased in men with the lower serum T levels and lower pretreatment bone mineral density (7, 8). Unpublished data from another 3-yr placebo-controlled study using T enanthate injections showed significant improvement in body composition and bone mineral density. In these studies, serum prostate-specific antigen (PSA) concentrations and clinical evidence of prostate disease were not different between the placebo- and T-treated groups. In aging men, the benefits of androgen replacement must be weighed against the potential risks (9). The important question of whether chronic T replacement in aging men with partial androgen deficiency will induce prostate cancer or promote conversion of histological to clinically evident prostate cancer remains unanswered. The clinical trials conducted, to date, have insufficient power to address this question.
Should we use 5 reducible or non-5
reducible androgens to treat andropause?
The natural potent androgen DHT is a selective androgen because it cannot be aromatized to estrogens. DHT binds to ARs more avidly than T. DHT is the product of conversion from T through the 5 reductase enzymes 1 and 2. Different tissues have different amounts of these two enzymes and, thus, can create local tissue environments that are low or high in DHT concentrations. The phenotypic appearance of men with congenital 5
reductase 2 deficiency has demonstrated the critical role DHT exerts on major androgen target tissues such as the external genitalia, prostate, and skin (10, 11, 12). Men with this congenital enzyme deficiency have very small prostate glands and ambiguous external genitalia but increase muscle mass at puberty and are not osteoporotic. Because the concentrations of DHT in the prostate are higher than that of T in normal men, it is generally believed that intraprostatic DHT may be the relevant androgen for stimulating prostate gland growth and development. Recognizing that T stimulates prostate growth, congenital 5
reductase 2 deficiency results in poor prostate development, and 5
reductase inhibitors administered to adult men with benign prostatic hyperplasia reduce prostate size, one might speculate that DHT administration might be a more potent prostate-stimulating pharmacological agent than its precursor T. It would seem paradoxical, therefore, to suggest that DHT might be used as a selective androgen that might exert androgenic effects with relatively less prostate stimulation. The observations suggesting that administered DHT may be relatively prostate sparing and the possible mechanisms for such a proposal will be discussed further.
Designer androgens with selective androgen receptor modulating (SARM) effects are under development. One goal of SARM development is to obtain steroidal or nonsteroidal compounds that have potent androgenic effects on muscles, bone, and probably also on the brain, but with minimal effects on the prostate and lipids. An androgen that cannot be 5-reduced but can undergo aromatization, 7
-methyl-19 nortestosterone (MENT), has been shown to have more potent androgenic effects on the gonadotrophs and muscle and less on the prostate (13, 14). Moreover, MENT has been shown to suppress serum gonadotropins in normal men (15) and maintain sexual function in hypogonadal men (16). It is proposed that MENT may have less stimulating effects than T on prostate growth when administered to men. MENT is currently being evaluated as a treatment of hypogonadal men and as a potential male contraceptive agent.
What do we know about the clinical use of DHT in boys and men?
DHT has been formulated as an injectable DHT heptanoate (enanthate) in experimental studies (17). In some countries in Europe, DHT is marketed as 2.5% DHT in a percutaneously applied hydroalcoholic gel (Andractim; Laboratories Besins Iscovesco, Paris, France). When applied transdermally, DHT is absorbed through the skin into the dermis and releases DHT into the circulation where serum DHT levels are maintained at relatively stable levels (18, 19, 20). Evidence that DHT acts as an androgen at the hypothalamic-pituitary level comes from the demonstration that transdermal DHT suppresses serum FSH and LH secretion in men (21, 22, 23).
Because DHT is not converted to estrogens, DHT would not cause gynecomastia while providing androgenic effects. It is an ideal replacement therapy for patients with 5 reductase 2 deficiency (11, 12). It is potentially useful for the treatment of pubertal gynecomastia (24, 25), microphallus (26), and constitutional delayed puberty in boys.
In hypogonadal men, transdermal DHT gel treatment has been reported to maintain sex characteristics, increase muscle mass, and improve sexual function without significant increases in prostate size (27, 28). In older men (5570 yr of age), DHT surprisingly resulted in improved sexual function and a small (15%) decrease in prostate volume (29). The effects on the prostate appear to be counter intuitive because DHT is the principal androgen required for the growth of the prostate. One hypothesis for the decrease in prostate size after DHT treatment of older hypogonadal men is based on the observation that estrogens may act synergistically with androgens in the prostate to promote prostate growth (30, 31, 32, 33, 34). DHT may lower tissue E2 levels by at least two mechanisms. First, it is not converted to estrogens. Second, by suppressing gonadotropins and endogenous T secretion, it provides less substrate for E2 production. Thus, DHT may decrease prostate growth because of the absence of synergistic effect of estrogens and androgens on the prostate gland. An alternate hypothesis is that the administered DHT may either be less well transferred from the circulatory compartment to the prostate gland than T or that exogenous T may actually provide higher intraprostatic DHT tissue levels than an equivalent amount of administered DHT. The latter possibility has not been tested because the effect of exogenous administration of DHT or T on intraprostatic androgens has not been studied. Thus, the effects of a nonaromatizable androgen such as DHT in hypogonadal men, especially in older men (because of the association of aging with prostate disease, osteoporosis, serum lipid abnormalities, and cognitive dysfunction) is of considerable clinical importance.
More recently, the pharmacokinetics of a reformulated DHT gel (0.7%; Laboratories Besins Iscovesco) were studied at three doses (approximately 16, 32, and 74 mg DHT delivered in the gel per day) for 14 d. Serum DHT rose in a dose-proportional manner with concomitant decreases in serum LH, FSH, T, and E2 concentrations. The calculated serum total androgens (T plus DHT) were maintained within the normal male adult range. After daily application of DHT gel, steady serum concentrations were maintained (35).
Using this newly formulated DHT gel, Ly et al. (36) reported in 2001 in this journal a double-blind, placebo-controlled, randomized clinical trial of DHT treatment in older men with partial androgen deficiency. Transdermal 0.7% DHT gel was administered daily (about 70 mg per day for 3 months) to 18 men aged 60 yr or older with an entry serum T equal to or less than 15 nmol/liter (normal adult range is usually above 10 nmol/liter). Compared with the placebo group with 19 subjects, the DHT-treated group had significantly elevated serum DHT and lower total and free T, LH, and FSH levels without significant changes on SHBG and E2 concentrations. The lack of suppression of SHBG and E2 in this study could be related to the small sample size of older men in this study. Body weight and lean mass showed no significant change whereas fat mass was decreased. Only the dominant knee flexion strength was increased without significant changes in other muscle strength testing. Daily functional tests, cognitive assessment, vascular reactivity, and quality of life measures remained unchanged. As anticipated, DHT treatment increased hematocrit and hemoglobin. Treatment with DHT decreased serum total and low-density lipoprotein (LDL)-cholesterol without affecting high-density lipoprotein (HDL)-cholesterol concentrations. Prostate disease markers such as serum PSA, symptom score, and central or peripheral prostate volume were not changed with DHT replacement. The study recruited men with partial androgen deficiency. It was not clear how many of the subjects had serum T levels below the normal adult male range, but the mean serum T in the DHT group was 15 nmol/liter, suggesting that a significant proportion of the treated subjects had serum T within the normal young adult male range. Moreover, at baseline, the DHT-treated group had high serum T levels and lower fat mass than the placebo group. The inclusion criteria in this study did not take into account any clinical symptoms of andropause. This study showed that DHT treatment for 3 months in a group of older men with serum T at the lower normal range had limited effect on knee flexion and fat mass.
In this issue of JCEM, Kunelius et al. (37) reported beneficial effect of DHT on erectile function in a large group of older men (mean, 58; range, 5070 yr). Because DHT has been proposed as a possible therapy for more global aspects of androgen deficiency in older men, this manuscript deserves careful scrutiny. First, the study population deserves consideration. It is well known that SHBG levels rise with age in men and serum-free T levels fall more steeply than total T concentrations (1, 2, 3). Based on this observation of the aging-related increase in SHBG, which in turn will amplify the decrease in free T in older men, the authors used a biochemical inclusion criteria based on either serum T less than 15 nmol/liter (normal range, 932) [similar to the study of Ly et al. (36)] and/or serum SHBG greater than 30 nmol/liter (normal range, 1462). It should be noted that these threshold criteria for enrollment were fairly liberal in terms of hormone levels, allowing men to be included in the study who had serum hormone concentrations within the normal range for both T and SHBG. The investigators did, however, require symptomatic evidence of androgen deficiency. In contrast to the study of Ly et al. (36), to be included in this study all men had andropause symptoms defined as decreased libido, erectile dysfunction, urinary disorders, asthenia, or depressed mood. Moreover, the subjects must have decreased nocturnal penile tumescence to less than once per week. Thus, the subjects entering into this study had relatively normal serum T levels (only 5 of 120 subjects had serum T below the normal adult range), normal serum SHBG (only 22 of 120 had serum SHBG levels above the adult range), but a relatively low calculated free androgen index. The DHT gel was applied transdermally, and the dose ranged between 125 and 250 mg/d using the 2.5% DHT gel formulation with dose adjustment at 3 months based on serum DHT levels attained. The serum DHT concentrations achieved at 3 and 6 months were similar whether the subjects applied 125, 187.5, or 250 mg/d of the gel. The results suggest either that the bioavailability of DHT is not proportional to the dose applied or that adjusting the dose of DHT based on serum DHT levels was not useful in attaining the desired serum DHT levels. Consistent with previous reports, administration of DHT suppressed serum T, E2, SHBG, FSH, and LH concentrations significantly. Of all the psychosexual parameters measured using modifications of the Psychological General Well Being and International Index of Erectile Function, the investigators demonstrated a statistically significant increase in two parameters, early morning erection (from 3 to 3.9) and ability to maintain an erection (from 2.3 to 3.2). Red cell parameters increased as expected with administration of an androgen but without significant changes in lipid profile, serum PSA, prostate weight, or symptom score. The authors remarked that the improvement in the erectile function score was small and their clinical significance uncertain. Unlike the studies of Ly et al. (36), body composition, bone, cognitive, daily functional, and quality of life parameters were not assessed in this study. Thus, the three reported studies on older men (29, 36, 37) with or without symptoms of andropause who had serum T or calculated free T in the lower normal range demonstrated possible small improvements in fat mass, knee flexion, and erectile dysfunction. It is not known whether the changes in these measures are clinically significant. In all the studies in older men, hematocrit/hemoglobin concentrations are increased but rarely to levels of clinical concern, and the changes in total LDL and HDL levels are variable.
What are the unanswered questions on the therapeutic role of DHT in androgen deficiency of older men?
Because of the relative short duration of the studies, the enrollment of men with relatively normal serum hormone concentrations, the small sample size, and the lack of appropriate outcome measures in some studies, the reported studies of DHT replacement in older men have not shown conclusively consistent or reproducible benefits. These studies demonstrated that transdermal DHT gel results in serum DHT levels that are significantly elevated above baseline (820 nmol/liter above baseline) and pharmacological suppression of endogenous T (free T), as well as E2 production. It should be noted that concern has been raised whether the use of pharmacological amounts of a nonaromatizable androgen such as DHT will compromise bone calcification by reducing tissue estrogen levels. This concern is based, in part, on recent evidence accumulated from studies on ER mutations and aromatase deficiency in mice and men that demonstrate significant osteopenia and delayed epiphyseal closure (38, 39, 40). These findings indicate that estrogens are important in maintaining bone mass. The important questions of whether the pharmacological administration of DHT, an androgen that cannot be aromatized to estrogens, will maintain bone mass and prevent fractures are critical and appropriate. Although serum E2 concentrations are suppressed with DHT treatment, the estrogen concentrations are not decreased to the nondetectable range as in congenital estrogen deficiencies. The role of estrogens in maintaining bone mass is well known in women and has been implied in men by epidemiological studies. In older men, although bone mineral density is dependent on serum androgen levels, the strongest correlation between bone mineral density and sex steroids is with serum-free E2 (41, 42). The studies reported by Ly et al. (36) and Kunelius et al. (37) published by this journal are of short duration. None of the studies reporting on DHT treatment of older men have measured bone turnover markers and bone mineral density. Thus, the question of whether DHT can maintain or increase bone mass remained unanswered. Future studies of DHT treatment in older men must be expanded to longer duration to allow, at a minimum, the assessment of bone mineral density.
It is also important to determine what effects DHT treatment will have on cognitive function. There is some evidence of putative protective effects of estrogens against dementia in women (43, 44). Androgens have been shown in improved visual-spatial function but not memory of verbal function in older men (45). Whether DHT or T can improve or maintain cognitive function is unknown.
The role of T and DHT treatment on the cardiovascular system is also important. Estrogens produce favorable lipid profiles with decreases in LDL-cholesterol and increases in HDL-cholesterol concentrations. Although DHT is not aromatizable and E2 levels fall, DHT treatment does not seem to have major effects on lipid profile (39, 40) or cause at most a seemingly cardiovascular protective small decrease in both total and LDL-cholesterol. It should be noted that there is a significant literature on other aspects of androgen action on the cardiovascular system that is not reviewed here. The effects of DHT on hematocrit and hemoglobin seem to be similar to those of T.
Muscle mass and fat mass are responsive to T. In both hypogonadal older and young men, lean body mass increased and fat mass decreased with T replacement (6, 7). Such marked changes in body composition have not been demonstrated with DHT therapy. In one study, DHT caused a decrease in fat mass (36), and in another study no decrease in visceral fat occurred (46). Recently, it has been shown that the increase in lean muscle mass and decrease in fat mass after T therapy are dose dependent in young men (47). It is not known whether DHT will lead to a dose response in improvement in body composition. Moreover, it has not been shown that the dose response of lean or fat mass to androgens also occurs in old men.
These DHT replacement studies have also not conclusively shown whether serum DHT will decrease prostate growth and prostate dysfunction and is more prostate-friendly than T because of the lack of conversion of DHT to estrogens or through other mechanisms (48). The reported studies showed, however, that even at the relatively high doses of DHT administered, there are no apparent, acute adverse effects on the prostate as measured by symptoms, serum PSA levels, or prostate volumes. Thus, DHT studies of longer duration with larger sample size are required and justified. These studies should be conducted in older men with androgen deficiency (with serum T levels below the normal adult range) and with concomitant symptoms of andropause. These studies should not only be randomized and placebo controlled but should include a group of men who will be treated with T to give comparatively similar levels of total androgens. Only when such studies are completed can we judge whether DHT therapy has a useful place in the therapy for older men with androgen deficiency. Such studies may also answer whether DHT therapy (a nonaromatizable androgen) has any advantage over T replacement on the long-term risks vs. benefits of androgen replacement therapy for older men.
Acknowledgments
We thank Sally Avancena, M.A., for assistance with the preparation of the manuscript.
Footnotes
This work was supported by NIH Grants M01-RR00425 and R01-DK61006.
Abbreviations: DHT, 5 Dihydrotestosterone; HDL, high-density lipoprotein; LDL, low-density lipoprotein; MENT, 7
-methyl-19 nortestosterone; PSA, prostate-specific antigen.
References