Is 3 alpha-androstanediol a marker of peripheral hirsutism?

Efficacy of second versus third generation oral contraceptives in the treatment of hirsutism

OBJECTIVE

To compare second versus third generation combination oral contraceptives (OCs) in the treatment of hirsutism.

METHODS

Women with hirsutism, as defined by a minimum Ferriman-Gallwey score of 10, were randomized in a double-blind fashion to receive an OC containing either ethinyl estradiol/desogestrel or ethinyl estradiol/levonorgestrel for 9 months of treatment. Ferriman-Gallwey scores, androgen levels and sex hormone-binding globulin were measured at baseline and every 3 months for the duration of the study. Hormones were measured in duplicate by radioimmunoassay.

RESULTS

Of the 47 women enrolled, 24 were randomized to ethinyl estradiol/desogestrel and 23 were randomized to ethinyl estradiol/levonorgestrel. Mean sex hormone-binding globulin increased significantly in subjects using the desogestrel-containing contraceptive compared with the levonorgestrel-containing contraceptive. Ten subjects completed the 9 months of treatment in the levonorgestrel group and 11 completed the study in the desogestrel group. Mean free testosterone and 3alpha-androstanediol glucuronide decreased significantly in the group receiving ethinyl estradiol/desogestrel but not in the ethinyl estradiol/levonorgestrel group. Mean Ferriman-Gallwey scores decreased significantly in both treatment groups. Improvement in mean Ferriman-Gallwey score was 35.7 +/- 38.1% (p < 0.001) for the ethinyl estradiol/desogestrel arm and 33.4 +/- 27.3% (p < 0.001) for the ethinyl estradiol/levonorgestrel arm. There were no statistically significant differences found in the improvement of Ferriman-Gallwey scores between the two treatment arms, although the power to detect a difference was limited by the small sample size.

CONCLUSIONS

Treatment of hirsute women with third generation OCs containing desogestrel results in a significant increase in sex hormone-binding globulin and decrease in free testosterone and 3alpha-androstanediol glucuronide. Both second and third generation OCs were clinically effective in treating hirsutism.

The clinical usefulness of 3 alpha-androstanediol glucuronide in premenopausal women with hirsutism

A biochemical parameter correlating with the clinical assessment of the severity of hirsutism and changing appropriately with the clinical response to treatment would be extremely useful. Preliminary reports of androstanediol glucuronide indicated that it was a peripherally-derived androgen and had a high correlation with clinical gradings of hirsutism. More recent reports have cast doubts on this association. This paper presents an evaluation of the clinical usefulness of androstanediol in 121 consecutive premenopausal patients with hirsutism. Androstanediol had a positive correlation with the clinical grading of hirsutism (p < 0.02) and the BMI (p < 0.01) but a negative correlation with age (p < 0.01). After adjustment for the effects of age and BMI there was no significant association between the degree of hirsutism and the level of androstanediol.

Two different pathogenetic mechanisms may play a role in acne and in hirsutism

OBJECTIVE

Acne is one of the most common skin disorders. Androgens are known to play an important and possibly central role. Androgens secreted from ovaries and adrenal glands (androstenedione, dehydroepiandrosterone and its sulphate, testosterone) and target tissue-produced androgens (testosterone and its 5 alpha-reduced metabolite, dihydrotestosterone) have been implicated. Although the sebaceous gland and the hair follicle form a single morphological entity, the pilosebaceous unit, acne and hirsutism do not always appear concomitantly, thus leading to the supposition that these two structures may have different degrees of sensitivity to similar androgenic stimulation.

DESIGN AND PATIENTS

To determine whether acne and hirsutism are the clinical expression of a different androgen metabolism at target tissue levels we studied 90 randomly selected patients who came to our Out-patient Department for diagnosis and treatment during the last 2 years with isolated acne of mild to severe degree and 52 patients with idiopathic hirsutism without acne or history of acne. Twenty-four women without acne or hirsutism and without a history of endocrine disease were studied as controls.

MEASUREMENTS

In both groups of patients, plasma levels of sex hormone binding globulin, of dihydrotestosterone, and of 3 alpha-androstanediol and of its glucuronide were evaluated. In all patients the percentage of free testosterone and the testosterone/sex hormone binding globulin ratio were also calculated.

RESULTS

Patients with acne and those with isolated hirsutism showed significantly decreased sex hormone binding globulin plasma levels. The values of the percentage free testosterone and those of the testosterone/sex hormone binding globulin ratio were, on the contrary, higher with respect to the controls, although there were no statistically significant differences between the two groups. Significantly increased plasma levels of dihydrotestosterone with respect to the controls were observed in patients with acne or in those with hirsutism. However, while all patients with hirsutism showed increased plasma values of 3 alpha-androstanediol and its glucuronide, all patients with acne showed plasma levels within the normal range, independently of the precursor plasma levels.

CONCLUSIONS

Our results demonstrate that dihydrotestosterone is further reduced to 3 alpha-androstanediol and its glucuronide only in hirsute patients but not in acne patients. These results suggest that dihydrotestosterone may undergo different metabolic pathways at skin levels and support the hypothesis that the two clinical manifestations may be the expression of the different metabolic fate of dihydrotestosterone itself. Moreover, our results demonstrate that 3 alpha-androstanediol and its glucuronide cannot be used as plasma markers of target-tissue produced androgens in all hyperandrogenic conditions.

Sex hormone levels and body hair growth in !Kung San and Kavango men from Namibia

The relation between hair growth and levels of sex hormones in serum and saliva was investigated in 256 !Kung San and Kavango men (ages 18 to 39 years) from Namibia/Southern Africa. Serum concentrations of total testosterone (Tser), 5 alpha-dihydrotestosterone (DHT), and estradiol (E2) as well as the level of bioavailable non-SHBG-bound testosterone in the saliva (Tsal) were determined by radioimmunoassay. The distribution and density of scalp and facial hair as well as the development of terminal hair on the chest, abdomen, pubic area, arms, fingers, and legs were categorized using objective criteria. Covariance analyses revealed marked differences in the distribution of body hair in the San and the Negro sample. This is partly explained by a significant influence of androgen and estrogen levels on the growth of terminal hair. DHT and the ratio DHT/Tser are significantly positively related to midphalangeal hair growth and negatively to pubic hair development. Tsal, the bioavailable fraction of total testosterone, exerts a weak positive influence on the degree of arm and leg hair growth; the most significant positive effect on the growth of abdominal, arm, and leg hair in our samples is caused by E2. The ratio Tser/E2 correlates significantly negatively with the arm and leg hair development and the ratio DHT/E2 with the degree of abdominal, pubic, arm, and leg hair, whereas lower DHT concentrations occur in men with stronger hair development.

Plasma 3 alpha-androstanediol glucuronide in precocious adrenarche

The aim was to study 3 alpha-androstanediol glucuronide (3AG) plasma levels and its relationship with 5-ene and 4-ene steroids in children with the benign form of precocious pubarche (precocious adrenarche). Sixty-five children with precocious adrenarche (PA), aged 3.6-8.2 yr (55 girls and 10 males) and 15 normal age-matched children were studied. We evaluated plasma androstenedione (A), dehydroepiandrosterone (DHA), its sulfate (DHA-S), testosterone (T), dihydrotestosterone (DHT), its glucuronide (DHTG), 3 alpha-androstanediol (3Ad) and its glucuronide (3AG) in all subjects. All androgens are expressed as mean +/- SD. We found significantly higher plasma levels not only in glandular androgens but also in peripheral androgens (A, 2.4 +/- 1.5 nM vs 0.79 +/- 0.46 nM, p < 0.001; DHA, 9.8 +/- 4.9 vs 2.7 +/- 0.76 ng/dl, p < 0.001; DHA-S, 3.4 +/- 2 microM vs 2.4 +/- 0.65 microM, p < 0.05; T, 0.74 +/- 0.5 nM vs 0.4 +/- 0.1 nM, p < 0.001; DHT, 0.36 +/- 0.13 nM vs 0.12 +/- 0.05 nM, p < 0.001; 3Ad, 0.13 +/- 0.1 nM vs 0.054 +/- 0.03 nM, p < 0.001; DHTG, 0.5 +/- 0.3 nM vs 0.26 +/- 0.09 nM p < 0.01). As far as the plasma 3AG levels are concerned we found significantly higher values in PA with respect to controls (1.17 +/- 0.7 nM vs 0.61 +/- 0.04 nM, p < 0.01), suggesting that 3AG may be considered a marker of skin androgen utilization.(ABSTRACT TRUNCATED AT 250 WORDS)

Is 3 alpha, 17 beta-androstanediol-glucuronide a diagnostic marker in women with androgenic manifestations?

3 alpha, 17 beta-androstanediol-glucuronide (Adiol-G) has been described as a marker of local androgen excess due to the increased activity of 5 alpha-reductase in the cells of the hair follicles. In order to test the diagnostic value of Adiol-G, the serum level was compared to that of testosterone, free testosterone, dehydroepiandrosterone sulfate (DHEA-S), androstenedione and to the body mass index in 44 women with androgenic symptoms (Group I), 27 women with menstrual disturbances but no androgenic symptoms (Group II), and 48 healthy women (Group III) who served as controls. Adiol-G was significantly higher (7.8 +/- 5.1 nmol/l) in women with androgenic symptoms than in the other groups, but there was a considerable overlap. Serum testosterone was also found to be higher in Group I than in Groups II and III, respectively. There was a significant correlation between Adiol-G and testosterone, and Adiol-G and DHEA-S. No significant correlation could be shown to exist between androstenedione and Adiol-G. When Adiol-G and testosterone were simply classified as 'normal' or 'increased' (Adiol-G 9.4 nmol/l; testosterone greater than 2.4 nmol/l), higher than normal values of the former were found in the presence of normal testosterone in only 4% of the cases. It is concluded that the level of Adiol-G generally parallels that of testosterone. Consequently, it does not seem to be an effective marker of peripheral androgen excess.

Bioactive and peripheral androgens in prepubertal simple hypertrichosis

Prepubertal simple hypertrichosis is characterized by excessive growth of vellus hair in children without other signs of endocrinopathies. The aetiology is unknown and it is not clear if it is an abnormal entity or an extreme form of the normal range of hair growth. Sixteen girls (aged 7 +/- 1.9 years, mean +/- SD) with prepubertal simple hypertrichosis and, as controls, 12 normal age-matched girls were studied. All patients were in preadrenarchal age and in all patients an ACTH test was performed to exclude non-classical forms of congenital adrenal hyperplasia. Testosterone (T), androstenedione (A), dehydroepiandrosterone sulphate (DHA-S), dihydrotesterone (DHT), 3 alpha-androstanediol (3Ad) and its glucuronide (3AG), and sex hormone binding globulin (SHBG) were evaluated and free testosterone (FT) and T/SHBG ratio were calculated in all subjects. In all patients we found T, A and DHA-S plasma levels comparable to controls, excluding an increased glandular androgen secretion. No significant differences, compared to controls, were observed for SHBG, FT and T/SHBG ratio, suggesting a normal T bioavailability. DHT plasma levels were significantly increased with respect to controls (0.42 +/- 0.04 vs 0.11 +/- 0.03 nmol/l; P less than 0.002) whereas 3Ad and 3AG were comparable to controls. The very significant increase in DHT plasma levels, without a parallel increase in 3Ad and 3AG found in our cases with prepubertal simple hypertrichosis, is difficult to explain. The clinical and biochemical significance of the high DHT plasma values needs more investigation.

Circulating dihydrotestosterone may not reflect peripheral formation

We compared the blood (PBDHT) and urine (PUDHT) production rate of dihydrotestosterone (DHT) in normal men and women to determine whether peripheral formation was totally reflected in blood. PBDHT was similar when measured at both sites in men (674 +/- 79 vs. 788 +/- 207 SE micrograms/d); however, PUDHT was greater than PBDHT in women (174 +/- 55 vs. 55 +/- 8 micrograms/d, P less than 0.02). Excretion rates of DHT and 3 alpha-androstanediol (3 alpha diol) were similar in both sexes despite major differences in blood levels. However, between sexes large differences were present in 3 alpha diol glucuronide (3 alpha diolG) in both plasma and urine. These observations indicate that peripheral (renal) formation of DHT and probably 3 alpha diol were not accurately determined by measurement of these steroids in blood. The large difference between blood and urine production rates in women suggests an important role of non-testosterone precursors of 5 alpha-reduced steroids. Measurements of 3 alpha diolG may provide more insight into these peripheral events.

Direct radioimmunoassay of free testosterone in the evaluation of androgenetic manifestations in women

The determination or calculation of the non-protein bound fraction of testosterone in serum had been suggested to be superior to the measurement of total testosterone levels in the evaluation of women with clinical signs of hyperandrogenism. We therefore compared the serum concentrations of free testosterone, as determined directly by radioimmunoassay, with the levels of various serum parameters in 317 women with hirsutism or acne. Total testosterone was elevated in 33%, free testosterone in 59%, and DHEA-S in 65% of the patients. Among the women with high levels of free testosterone, 24% showed serum concentrations both of total testosterone and SHBG within the normal range, indicating that the calculation of the testosterone:SHBG ratio (free androgen index) would be of limited value for androgen diagnostics. Serum free testosterone correlated positively with total testosterone and negatively with SHBG. No significant correlation was found between DHEA-S and free or total testosterone, and between total testosterone and SHBG. The stimulation of the adrenal cortex by ACTH resulted only in a significant rise in 17 alpha-hydroxyprogesterone levels. During the dexamethasone suppression test there was a significant suppression of the serum concentrations of total and free testosterone, DHEA-S, and 17 alpha-hydroxyprogesterone, while SHBG did not change. In several of the women with normal testosterone levels, free testosterone decreased by more than 50%, although the suppression of total testosterone was only slight. This was possibly due to the suppression of adrenal androgens which may compete with testosterone for the binding sites of SHBG. The direct measurement of free testosterone appears to be a valuable approach to the evaluation of women with androgenetic manifestations.

Lack of correlation between sex hormone binding globulin and free testosterone in some cases of "idiopathic" hirsutism

Plasma 3 alpha-androstanediol levels are increased in a high percentage of hirsute women, including those with normal plasma levels of its precursors, repeatedly evaluated during the menstrual cycle. To better ascertain the origin of this form of hirsutism, so called "idiopathic", 17 hirsute women (aged 19.5 +/- 4.8 SD yr), with normal bw, showing an isolated increase in plasma 3 alpha-androstanediol (0.65 +/- 0.18 SD nmol/l) were studied in basal conditions (on day 6, 7, 17, 18 of the menstrual cycle). A hirsute female (aged 14 yr) with congenital adrenal hyperplasia due to 21-hydroxylase deficiency was also included in the protocol. Sex hormone binding globulin was evaluated using a saturation method (dihydrotestosterone- 3H 0.5-25 nM) with and without 200x cold dihydrotestosterone. Maximum number of binding sites and Ka were calculated in all cases with Scatchard analysis. Free testosterone was evaluated by the equilibrium dialysis technique. The results obtained showed a significant increase in Ka values in 10 out of the 17 cases; low plasma levels of sex hormone binding globulin in 13 cases and normal levels in 4 cases; high levels of free testosterone fraction in 5 cases and in the upper limits of the normal range or normal in the remaining 12. These findings demonstrate that some cases of "idiopathic" hirsutism may present an alteration in the binding affinity of the sex hormone binding globulin which may be genetically transmitted, being found to be increased in two sisters and in their brother.

Morphological studies of polycystic mouse ovaries induced by dehydroepiandrosterone

Morphological alterations induced by dehydroepiandrosterone (DHA) were studied in polycystic mouse ovaries (PCO). Treated mice showed ovulatory failure and cystic changes; cysts and follicles in various stages of growth and atresia were present although corpora lutea were absent. The levels of testosterone, dihydrotestosterone, 3 alpha- and 3 beta-androstanediol, estrone and androstenedione increased, whereas estradiol was not detectable. The ultrastructure of granulosa cells in healthy and atretic follicles was similar to that of control animals, although the membrana granulosa in cysts was reduced to a monolayer of flattened cells. The theca interna of healthy and atretic follicles and ovarian cysts showed ultrastructural signs of abnormal steroidogenic stimulation. No significant differences (0.7 less than P less than 0.8) were found between the extensive surface area of gap junctions of healthy follicles of control and DHA-treated animals. On the P-face of granulosa cells of large healthy follicles, meandering strands of tight junctional particles were observed; their average length was significantly longer than those in healthy follicles of control animals (P less than 0.001). This increase was probably related to the large amounts of androgens present in the treated animals. Theca interna cells possessed small gap junctions; no significant differences (P greater than 0.9) in gap-junction surface area were observed between DHA-treated and control animals. These results suggest that the size of gap junctions is probably unrelated to the steroidogenic activities of theca cells.

The effect of endogenous progesterone on serum levels of 5 alpha-reduced androgens in hirsute women

This study was to examine indirectly the effect of endogenous progesterone, a known competitor for 5 alpha-reductase, on androgen metabolism in target organs in hirsute women. Serum levels of progesterone, testosterone (T), androstenedione (A), dihydrotestosterone (DHT) and 5 alpha-androstane 3 alpha 17 beta-diol (3 alpha-diol) and sex hormone binding globulin (SHBG) were assessed serially over a four week period in normal women, six hirsute women with regular menstrual cycles, eight hirsute women with oligomenorrhoea (and presumptive polycystic ovaries) and seven non-hirsute women with oligomenorrhoea. Serum T and A levels were significantly higher than normal in both hirsute and non-hirsute women with oligomenorrhoea, while serum SHBG was significantly lower than normal in the two groups of hirsute women. The calculated free T level was higher than normal in all three groups of patients. DHT levels were not significantly different from normal in any of the three groups of patients. The 3 alpha-diol level showed considerable overlap with normal in all groups of patients and was only significantly higher than normal in hirsute women with oligomenorrhoea (P less than 0.05). There was a small fall in DHT in the late luteal phase of the cycle of those women with a sustained rise in serum progesterone in the second half of the cycle, but no change in serum 3 alpha-diol. These studies suggest that serum 3 alpha-diol may not be as good an indicator of peripheral androgen metabolism in hirsute women as previously reported and that a rise in serum progesterone has only a minimal effect on circulating levels of the active 5 alpha-reduced androgen metabolites. Although in vitro 3 alpha-diol has been shown to be a potent inhibitor of 5 alpha-reductase this casts doubt on its role in this regard in vivo.

Androgen levels during cyproterone acetate and ethinyl oestradiol treatment of hirsutism

Androgen levels were measured before and during 12 months of treatment with cyproterone acetate (50 mg 5th to 16th day of cycle) and ethinyl oestradiol (50 micrograms 5th to 26th day of cycle) in 26 patients suffering from dysfunctional hirsutism. Patients were included in whom androgen levels after a dexamethasone suppression test fell above the range found in normal non-hirsute women on dexamethasone. Eighty percent of patients showed a substantial improvement of their Ferriman and Gallwey score after 12 months of treatment. During the treatment the mean level of testosterone was 64%, androstenedione 55%, testosterone oestradiol binding globulin 365%, apparent free testosterone 30%, dihydrotestosterone 74%, 5-androstanediol 43%, dehydroepiandrosterone 44%, 5-androstenediol 67% and dehydroepiandrosterone sulphate 74% of basal levels. From the results it is concluded that this treatment is effective both clinically and biochemically in reducing hyperandrogenism.