- Email: [email protected]
Clinical and endocrine effects of finasteride, a 5α-reductase inhibitor, in women with idiopathic hirsutism
FERTILITY AND STERILITY Copyright
©
Vol. 64, No.2, August 1995 Printed on acidfree paper in U. S. A.
1995 American Society for Reproductive Medicine
Clinical and endocrine effects of finasteride, a Sa-reductase inhibitor, in women with idiopathic hirsutism
Lilliana Ciotta, M.D. *t Antonio Cianci, M.D.t AIdo E. Calogero, M.D.* Marco Antonio Palumbo, M.D.t
Elvira Marletta, M.D.t Anna Sciuto, M.D.t Giuseppe Palumbo, M.D.t
University of Catania, Catania, Italy
Objective: To evaluate the effects of long-term administration of finasteride on hirsutism score, basal gonadotropin, and androgen secretion in women with idiopathic hirsutism. Design: Randomized single-blinded study. Patients: Eighteen patients with moderate-severe hirsutism were recruited for the study. Interventions: Nine hirsute patients received 7.5 mg/d oral finasteride for a period of 9 months whereas the other nine were treated with placebo. Hirsutism score, serum basal gonadotropin, androgens, estrogen, and sex hormone-binding globulin (SHBG) levels were evaluated in all patients before treatment and every 3 months during treatment. Results: Mter 6 and 9 months of treatment, the hirsutism score improved significantly in the patients receiving finasteride, whereas no significant modifications were observed in patients treated with placebo. The side effects observed were headache and depression of modest entity during the 1st month of treatments, whereas libido did not change. Serum levels of LH, FSH, androstenedione, unbound T, DHEAS, E 2 , 17a-hydroxyprogesterone, and SHBG did not change during therapy. Hirsute patients treated with finasteride exhibited a marked decrease of dihydrotestosterone and a significant increase of T serum levels from the 3rd and 6th months of treatment, respectively. Conclusion: Finasteride decreased the hirsutism score of patients affected by idiopathic hirsutism with few side effects during treatment. No modification oflibido was observed. Fertil Steril 1995;64:299-306 Key Words: Androgen, 5a-reductase activity, idiopathic hirsutism, finasteride
Androgens are necessary for sexual hair and sebaceous development. In conjunction with other regulatory factors, such as induction factors from the dermal papilla and insulin-like growth factors, they induce the prepuberal pilosebaceous unit in androgen-dependent areas to differentiate either into a terminal hair follicle or into a sebaceous follicle (1). Hyperandrogenism is a common condition in women characterized by excessive growth of terminal hair in a male pattern, hirsutism, and/or by exagger-
Received August 4, 1994; revised and accepted March 3, 1995. * Reprint requests: Lilliana Ciotta, M.D., Ospedale Santo Bambino, Via Torre del Vescovo, 95124 Catania, Italy (FAX: 003995-312001). t Department of Obstetrics and Gynaecology. :j: First Department of Internal Medicine. Vol. 64, No.2, August 1995
ated activity of sebaceous glands, which causes acne, seborrhea, and alopecia. Like obesity and/or anovulation, these symptoms are variably expressed manifestations of androgen excess and thus they may be cutaneous signs of an androgenic disorder. Female hyperandrogenism, in most cases, is dependent on an ovarian or adrenal overproduction of androgens or by hypersensitivity of the pilosebaceous unit towards normal levels of free androgens (idiopathic hirsutism) (2). Because circulating androgen levels are in most cases normal, several data indicate increased local formation of androgens (3) or increased sensitivity of the androgen receptor to normal levels of androgens. Consequently, antiandrogens have been used extensively in the treatment of hirsute women without androgen excess. These compounds act mainly to inhibit binding of androgens to the androgen receptor and they are able to Ciotta et al. Finasteride in the treatment of IH
299
reverse the pilosebaceous unit toward the vellus state (1). Previous studies have documented that hirsute women have a higher 5a-reductase activity than normal women (4), and 5a-reductase is the enzyme responsible for the irreversible conversion ofT to dihydrotestosterone (DHT) (5). It was shown previously that, within the androgen-dependent glandular cells, DHT but not T is the active intracellular androgen (6, 7). Consequently, the inhibitors of 5a-reductase may by used in the treatment ofhyperandrogenism in females. Finasteride is an orally active agent undergoing clinical trials for the treatment of benign prostatic hyperplasia (BPH) (8). This drug, a member of the 4-azasteroid family of compounds, is a competitive inhibitor of steroid 5a-reductase (9). Hence, as hypothesized (10, 11), it is possible that a specific compound with inhibitory activity on 5a-reductase, such as finasteride, may be able to reduce the skin hyperandrogenic symptomatology in affected women. Therefore this study was undertaken to gain information about the clinical and endocrine effects of finasteride in patients affected by idiopathic hirsutism. To accomplish this, we evaluated the effects of long-term treatment of finasteride on hirsutism score, basal gonadotropins, androgen, estrogen, and sex hormone-binding globulin (SHBG) serum levels in nine patients with idiopathic hirsutism. These data were compared with those obtained in other nine idiopathic hirsutism-matched patients treated with placebo. Table 1 Clinical and Endocrinologic Features of Idiopathic Hirsute Patients Before Treatment with Placebo or Finasteride* Parameterst
Placebo
Finasteride
Age (y) Body mass index (kg/m2) Ferriman-Gallwey score Menses LH (mID/mL) FSH (mID/mL) A (ng/mL) Total T (ng/mL) Unbound T (pg/mL) DHEAS (f.lg/mL) DHT (pg/mL) E2 (pg/mL) 17-0HP (f.lg/mL) SHBG (f.lg/mL)
20.2 :+: 0.62 20.7 :+: 0.47 21.8 :+: 0.81:1: Eumenorrhea 5.57:+: 0.42§ 4.39:+: 0.35§ 1.73:+: 0.16§ 0.53:+: 0.01§ 2.70:+: 0.08§ 2.08:+: 0.14§ 374.4 :+: 37.2§ 58.24:+: 2.84§ 384.2 :+: 36.6§ 2.22:+: 0.15§
20.9 :+: 0.61 20.8 :+: 0.36 19.0 :+: 1.57:1: Eumenorrhea 5.57:+: 0.47§ 3.62:+: 0.18§ 1.84:+: 0.12§ 0.53:+: 0.05§ 2.86:+: 0.11§ 2.02:+: 0.10§ 380.0 :+: 29.4§ 52.66:+: 4.45§ 408.9 :+: 23.9§ 2.17:+: 0.13§
* Values are means:+: :j: P
SEM.
< 0.001 versus normal women.
§ P = not significant versus normal women. t The following conversion factors to SI units are applicable for the values cited in the table: LH and FSH, 1.00; A, 3.49; Total or Unbound T, 3.47; DHEAS, 2.71; DHT, 3.45; E2, 3.67; 17-0HP, 3.03; and SHBG, 3.44. 300
Ciotta et al.
Finasteride in the treatment of IH
Table 2 Clinical and Endocrinologic Features of Normal Women* Parameterst Age (y) Body mass index (kg/m2) Ferriman-Gallwey score Menses LH (mID/mL) FSH (mID/mL) A (ng/mL) Total T (ng/mL) Unbound T (pg/mL) DHEAS (f.lg/mL) DHT (pg/mL) E2 (pg/mL) 17 -OHP (f.lg/mL) SHBG (f.lg/mL)
22.4 (19.0 to 27.0) 21.2 (19.0 to 23.0) 6.9 (4.0 to 10.0) Eumenorrhea 5.5 (4.0 to 7.0) 4.8 (3.4 to 6.4) 1.80 (1.35 to 2.20) 0.64 (0.42 to 0.76) 2.69 (2.50 to 2.95) 2.27 (1.60 to 3.00) 326.6 (180.6 to 430.2) 54.7 (45.6 to 68.4) 370.4 (270.6 to 430.5) 2.24 (1.80 to 3.00)
* Values are means with ranges in parentheses. Control group, n = 16.
t The following conversion factors to SI units are applicable for the values cited in the table: LH and FSH, 1.00; A, 3.49; Total or Unbound T, 3.47; DHEAS, 2.71; DHT, 3.45; E2, 3.67; 17-0HP, 3.03; and SHBG, 3.44.
MATERIALS AND METHODS Subjects
Eighteen women (age 20.6 :::':: 0.43 years; mean :::':: SEM) with moderate-severe hirsutism (mean score, according to the Ferriman and Gallwey classification [12], 20.4 :::':: 0.92) were recruited for the study. Their hirsutism score was significantly higher (P < 0.001) (Table 1) than the hirsutism score of 16 normal women with regular menses (control group) (Table 2). No patient had acne-seborrhea and/or other clinical signs of hyperandrogenism. All patients had regular menses, body mass index (mean value 20.7 :::':: 0.29 kg/m 2 ) and were otherwise healthy. Two blood samples were withdrawn during the follicular phase (days 5 to 8) in both hirsute and normal women for hormonal evaluations. Normal serum levels of LH, FSH, androstenedione (A), total and unbound T, DHEAS, DHT, E 2 , 17a-hydroxyprogesterone (17-0HP), SHBG (Table 1), total triiodothyronine, total T 4 , thyroid-stimulating hormone, and PRL were found in all patients. Their LH:FSH ratio was <2. The response of 17-0HP to ACTH stimulation test, performed during the follicular phase of the cycle, was normal in all patients. Ovarian volume was calculated in patients and control subjects according to the formula 4/37f(d l /2 X dJ2 X da/2), in which the diameters were determined by ultrasound (US) as the mean of the length, width, and depth of the ovary. Ovarian size was normal in both nonhirsute (control group) and hirsute women. Ovarian US morphology was normal in the control Fertility and Sterility
group and in 14 hirsute patients (77.8%), whereas multifollicular features was observed in the other 4 hirsute patients (22.2%). Consequently, a diagnosis of idiopathic hirsutism was posed in all women affected by hirsutism. None of the patients had received any hormonal treatment for 2:6 months before the study. After an informed consent was obtained from all patients, they were treated randomly and blindly with placebo (nine patients) or finasteride (Prostide; Sigma-Tau, Milan, Italy) at the dose of 7.5 mg/d orally (nine patients) for 9 months. All patients were urged to use contraception barrier methods or an intrauterine device during the study. Clinical Study
The evaluation of the hirsutism score was performed according to the Ferriman and Gallwey classification (12) before and after 3, 6, and 9 months of treatment with finasteride or placebo. Measurements were performed by a single examiner (L.C.). A general physical examination was performed at the same times during the study. The evaluation of libido was assessed by an interval scale for subjective evaluation (desire) and a semistructured talk (clinical) for objective assessment (frequency of coitus). Endocrine and Hematochemical Studies
The hormonal evaluation was carried out in all patients during their follicular phase (days 5 to 8 of the cycle). The serum concentrations ofLH, FSH, A, total and unbound T, DHEAS, DHT, E 2, 17-0HP, and SHBG were measured every 3 months during treatment with finasteride and placebo. Hematologic, renal, and liver functions were evaluated by measuring hemochrome, hematocrit, glycemia, azotemia, creatinemia, electrophoretic protidogram, total and fractioned bilirubinemia, transaminase, alkaline phosphatase, and gamma-glutamil-transpeptidase every 3 months during treatment. Hormone Determination
Serum levels ofLH, FSH, A, total T, and DHEAS (Sorin, Saluggia, Vercelli, Italy), unbound T and 17OHP (Pantex, Santa Monica, CA), DHT (Bioline, Bruxelles, Belgium), E2 (CIS, Santhia, Vercelli, Italy), and SHBG (Intertech, Strasen, Luxembourg) were measured with double-antibody RIA methods using commercially available kits. All samples were assayed in duplicate. The intra-assay coefficients of variation (CVs) ofLH and FSH were 3.8% and 5.8%, respectively, and the interassay CVs were 10.5% and 9.8%, respectively. The intra-assay and interassay Vol. 64, No.2, August 1995
O·······OPlacebo e-e Finasteride
25 w
20
0:::
o
fA
°T···············O·················O·
e ______ -
e ____ ! e _______*.9 e
15
(!)
lL..
·······0
10
5 oL---~--~--~---~-~
o
3
6
9
MONTHS OF TREATMENT
Figure 1 Ferriman and Gallwey (FG) hirsutism score of patients with idiopathic hirsutism treated with placebo or finasteride. *p < 0.05 versus 0; OP < 0.05 versus 3 months (Duncan multiple range test).
CVs for the other hormones were 7.2% and 10.5% for A, 8.2% and 10.8% for total T, 7.2% and 9.5% for unbound T, 5.6% and 9.5% for DHEAS, 6.2% and 8.2% for DHT, 7.0% and iO.6% for E 2, 6.4% and 8.7% for 17-0HP, and 4.8% and 5.6% for SHBG, respectively. Statistical Analysis
Results are presented as means ± SEM throughout the study. Data were analyzed by using one- or two-way analysis of variance (ANOVA) followed by the Duncan multiple range test. Significance was accepted for P values < 0.05. RESULTS Clinical Evaluation
Finasteride reduced significantly (P < 0.001, oneway ANOVA) the hirsutism score of the patients treated (Fig. 1). After 6 and 9 months of treatment, the mean hirsutism score decreased significantly with respect to pretreatment values (P < 0.05). In addition, the mean value observed after 9 months was significantly lower than that recorded after 3 months of therapy (P < 0.05). No significant modification of the hirsutism score was observed in idiopathic hirsutism patients treated with placebo. Accordingly, the hirsutism score of idiopathic hirsutism patients treated with finasteride was significantly lower than that observed in the idiopathic hirsutism patients treated with placebo (P < 0.01, two-way ANOVA). Hirsutism regression was characterized by a progressive slowing down of hair growth, thinning and lightening of the hairs, followed by a reduction and/or Ciotta et al. Finasteride in the treatment of IH
301
r
DPlacebo _ Finasteride
7.0 6.0 -:J E 5.0 '-... ::J
1 I
4.0 3.0
-'
2.0 1.0 0.0
'-
--
~
5.0 .-..... -' 4.0 E
'-... ::J
E
3.0
I
2.0
'-' (/)
i.L
1.0 0.0
o
-'--
'---
-
3 6 9 LENGTH OF TREATMENT (Mo)
Figure 2 Serum levels ofLH (top) and FSH (bottom) of patients affected by idiopathic hirsutism treated with placebo or finasteride.
disappearance of terminal hairs in the androgendependent body area. Hirsutism decline was seen first on the face (chin, cheeks, upper lip), around the mammary areola, upper limbs (forearm), and lower limbs (upper internal part of the thigh), while it was slower on the chest and lower abdomen. During the period of the study, all patients had regular cycles. In the group of patients treated with finasteride, headache was found in four patients (44.4%) during the 1st month of treatment, and depression in was found in two patients (22.2%) during the first 3 months of therapy. In the group of patients treated with placebo, anxiety and depression was found in four patients (44.4%) during the first 2 months of therapy, and headache was found in three patients (33.3%) during the 1st month. No significant modifications in the frequency of coitus and sexual desire were found during treatment with placebo or finasteride. Endocrine and Hematochemical Evaluation
No significant modifications ofLH, FSH (Fig. 2), A, unbound T, E 2 , 17-0HP, and SHBG (Table 3) serum levels were observed in all patients during the study period. Patients treated with finasteride showed a significant increase of serum total T levels (P 302
Ciotta et a!. Finasteride in the treatment of IH
< 0.001, one-way ANOVA). This increase became significant after 6 and 9 months of treatment (P < 0.05). An increase of total T levels also was observed in the group of idiopathic hirsutism treated with placebo (P < 0.01, one-way ANOVA). However, the increment was of modest entity and it was significant at the 3rd and 9th month of therapy (Fig. 3, top). Total T levels during finasteride treatment were significantly higher than those obtained in the placebo group (P < 0.01, two-way ANOVA). Serum levels of DHT showed a marked decrease in the group of patients treated with finasteride (P < 0.001, one-way ANOVA), whereas no significant modifications were observed in the placebo group (Fig. 3, bottom). Accordingly, DHT levels during finasteride treatment were significantly lower than those obtained in the placebo group (P < 0.01, two-way ANOVA). No significant variations in DHEAS serum levels were observed in both placebo and finasteride groups (Fig. 4). However, patients treated with finasteride had levels of DHEAS significantly lower than those recorded in patients treated with placebo (P < 0.01, two-way ANOVA). No variations were seen in the hematologic, renal, and liver functions. DISCUSSION
An increased sensitivity of the terminal hair to androgens has been hypothesized in hirsute women with normal circulating levels of androgens (2, 13). It generally is accepted that the free androgen fraction diffuses passively into the target cells where, under the influence of 5a-reductase, T is reduced to DHT (5). Both T and DHT bind to the same androgen receptor. However, in target tissue, the affinity of the receptor for DHT is much higher than that for T. Hence, in most target tissue, DHT is the active androgen (6, 7). Two 5a-reductase isoenzymes have been cloned recently (14, 15). The role of 5a-reductase-l is presently unknown. It is present in small quantities in the prostate, a gland rich of 5a-reductase-2, but its presence in other areas rich in 5areductase still is unknown. The 5a-reductase-2 is responsible for differentiation of the male external genitalia and prostate, and it is defective in male pseudohermaphrodites with 5a-reductase deficiency (15). 5a-Reductase-2 is essential in humans for hair growth (16, 17) and it has been shown that there is a significant correlation between the 5a-reductase and the Ferriman and Gallwey score in hirsute women (4, 18). Finasteride is a new antiandrogen that competitively inhibits 5a-reductase-2 (19), producing a profound decrease in serum DHT level. Finasteride itself possesses no androgenic, antiandrogenic, or Fertility and Sterility
Table 3
Hormone Serum Levels in Idiopathic Hirsute Patients Treated with Placebo or Finasteride* Finasteride
Placebo
o
Months A (ng/mL)t Unbound T (pg/mL)t E2
1. 73 ": 0.16
58.2
(pg/mL)§
17-0RP (l'g/mL)11 SRBG
2.8
":
384.2 ": 36.6
(l'g/mL)~
1.88 ": 0.1
1.84 ": 0.12
1.84 ": 0.12
1.82 ": 0.08
1.89 ": 0.09
1.83 ": 0.11
2.82 ": 0.1
2.92 ": 0.09
2.86 ": 0.11
2.97 ": 0.13
2.97 ": 0.11
3.04 ": 0.08
":
1.2
59.4
":
2.4
57.4
404.8 ": 38.4
385.9 ": 37.3
2.22 ": 0.15
* Values are means":
1.83 ": 0.09
60.6
9
1.95 ": 0.11
2.16 ": 0.15
2.2
52.6 ": 4.4
407.1 ": 33.1
408.9 ": 23.9
":
1.96 ": 0.16
t Conversion factor to SI unit. 3.47.
steroid hormone-related properties. Inhibition occurs without affecting the binding of T or DHT to the androgen receptor. The metabolism of fin asteride now has been defined in humans (20). The parent drug is the major circulating component in the plasma. In addition, two monohydroxy metabolites also possess some activity as inhibitors of 5a-reductase. The biologic effect of finasteride persists sig-
0.8
-
-:::::J
E
"0> S
0.6
Placebo Finasteride
*0
*0 *
* J...
~
I....J
~
0.4
0
I-
0.2 0.0
-~
57.4 ": 4.7 381.1
": 16.7
2.08 ": 0.07
6
58.7 ": 4.5 403.3
": 11.2
1.89 ":
0.12
57.9
":
3.7
405.5 ": 20.6 1.87 ": 0.08
§ Conversion factor to SI unit, 3.67. II Conversion factor to SI unit, 3.03. ~ Conversion factor to SI unit, 3.44.
SEM.
t Conversion factor to SI unit. 3.49.
0
2.17 ": 0.13
3
9
6
2.88 ": 0.14
2.7 ": 0.08
o
3
-
l...-'-
-~
nificantly longer than the serum half-life of 5 to 6 hours (20). Previous studies investigating 24 men of 12 different families affected by congenital 5a-reductase deficiency and consequently low DHT levels (21) showed that the female components in these families apparently were unaffected. Hence, a drug able to lower DHT levels and possessing low toxicity (8), such as finasteride, may provide an acceptable medical therapy for hirsute women. Indeed, it has been shown recently that short-term treatment with finasteride improves significantly the hirsutism score in nonselected hirsute patients (22) and that this improvement seemed to be slightly better than that obtained with other antiandrogens, such as fiutamide or cyproterone acetate (CPA) (22). In the present study, we elected to evaluate the clinical and hormonal efficacy and safety of a longterm therapy with finasteride in a group of patients
400 2.5
DPlacebo _ Finasteride
-:::::J
E 300
"01
*
~ I-
:r:
*
200
....J ----2.0 E "01
*
::t
'-'
0
1.5
en
i'3
100
:r:
1.0
0
0
o
~'-
'----
3
6
--
'---
LENGTH OF TREATMENT (Mo)
Figure 3 Serum levels of total T (top) and DHT (bottom) of patients affected by idiopathic hirsutism treated with placebo or finasteride. Serum levels of total T of patients treated with finasteride were significantly higher than those treated with placebo (P < 0.01, two-way ANOVA), whereas those of DHT were significantly lower (P < 0.01, two-way ANOVA). *P < 0.05 versus 0, OP < 0.05 versus 3 months (Duncan multiple range test). Vol. 64, No.2, August 1995
0.5
9
0.0
036 9 LENGTH OF TREATMENT (Mo)
Figure 4 Serum levels of DHEAS of patients affected by idiopathic hirsutism treated with placebo or finasteride. Serum levels of DHEAS of patients treated with finasteride were significantly lower than those found in patients treated with placebo (P < 0.01, two-way ANOVA).
Ciotta et al. Finasteride in the treatment of IH
303
p
affected by idiopathic hirsutism. In a preliminary study, a group of nons elected hirsute patients were treated long term with 5 mg/d offinasteride, the dose usually used for the treatment ofBPH in men. Albeit there was an improvement of their hirsutism score, it seemed slower than that obtained with other antiandrogens in the same patients. Given the low toxicity of finasteride (8), we thought to perform this study using the dose of 7.5 mg/d. Oral contraceptives were not prescribed to the patients enrolled in this study to provide a direct evaluation of finasteride as a single agent. To avoid feminization of a male fetus, secondary to reduction of DHT during embryogenesis (21), all patients were urged to use alternative, nonhormonal, contraceptive methods. As suggested previously (10, 11) and demonstrated recently (22), our study showed that finasteride is effective in the treatment of hirsutism because all patients had a significant reduction of their hirsutism score during treatment. This reduction became significant after 6 months and it further decreased after 9 months oftherapy. The improvement of the Ferriman-Gallwey score seemed to be similar in time and degree to that obtained with the use of other antiandrogens, such as flutamide, whereas it seemed to be better to that observed during treatment with CPA or spironolactone and oral estrogens in patients with a moderate-severe degree of hirsutism (unpublished observations). The administration of 7.5 mg/d of the drug did not seem to induce a better clinical response after 3 months of treatment in women with idiopathic hirsutism in comparison to that reported in nons elected hirsute women treated with a dose of 5 mg/d for 3 months (22). However, the observed further significant reduction ofthe hirsutism score in our patients after 9 months of treatment with respect to the values seen after 3 months and the small number of women tested previously (22) and in this study must be taken into account before any definitive conclusion can be drawn. According to the mechanism of action of fin asteride, a significant decrease of DHT serum levels already was observed after 3 months of treatment. This decrease preceded and paralleled the reduction of the hirsutism score. The other hormones did not show any significant change with the exception of total T, which increased significantly after 6 months of treatment. Accordingly, total T previously was found to be increased in normal men (9, 23), in patients with BPH (8), and in hirsute women (22) during treatment with finasteride. The fact that the increased total T serum levels did not modify the positive clinical response observed during therapy confirms the hypothesis that DHT is the essential 304
Ciotta et al. Finasteride in the treatment of IH
androgen for hair growth (19, 20). In our study, the increase of total T serum levels in the group of patients treated with placebo after 3 and 9 months was of modest entity and it may relate to seasonal hormonal variations. We did not observe any changes in basal gonadotropin levels in patients treated with finasteride, as reported previously in men (8,9,23) and in hirsute women (22). Recently, it has been reported that short-term therapy with finasteride does not modify both LH pulse frequency and amplitude in hirsute patients (22). In addition, finasteride treatment does not influence the release of LH and FSH in response to GnRH both in normal men (23) and in hirsute women (22). The unchanged levels of gonadotropins, even during the administration of 7.5 mg/d, suggest that the gonadotropin secretion is not influenced significantly by direct reduction of serum DHT levels or by indirect increase of total T serum levels induced by finasteride. However, it should be pointed out that LH levels in male pseudohermaphrodites with congenital5a-reductase deficiency, although elevated, are substantially lower than LH levels seen in castrated subjects (24), suggesting that, albeit DHT may playa role in the negative feedback effect, the major feedback on LH is exerted by T directly or by conversion to E2 in the general circulation or at the hypothalamic level. It also can be postulated that it is not the circulating DHT levels that exert a negative effect on LH but hypothalamic and/or pituitary DHT and that finasteride does not cross the blood-brain barrier sufficiently to alter these levels. Because of the lack of finasteride effects on both basal and stimulated gonadotropin release (22, 23), the increase of total T serum levels may not be ascribed to a more sustained LH-dependent ovarian theca-interstitial cell stimulation. Furthermore, an adrenal overproduction of T cannot explain the observed increase of total T serum levels in our patients because the unmodified DHEAS levels during treatment with finasteride suggest that the drug has no influence on the adrenal androgenesis. This datum confirms what was found previously in hirsute women, in whom the therapy with finasteride does not determine any significant modification in both basal and ACTH-induced release of A and DHEAS with respect to pretreatment values (22). On the contrary, in our study we observed a significant reduction of DHEAS serum levels in the group of patients treated with finasteride compared with values obtained in the placebo group. Therefore, it is possible that the selective blockade of a major metabolic pathway for T, resulting from the inhibition of 5areductase by finasteride rather than an increase of Fertility and Sterility
its secretion, may cause the increase of total T serum levels. In fact, 5a reduction is a major metabolic pathway for T in the prostate, and inhibition of this reaction probably may explain the observed accumulation of substrate T in the gland (25). However, if the increased T serum levels may be related to the reduced conversion of T to DHT, it is difficult to explain why we did not observe a parallel increase in A serum levels, that, in the target tissues, is converted by 5a-reductase to androstanedione (5). Furthermore, the conversion of A to T is catalyzed by the enzyme 17 -ketosteroid reductase and this reaction is reversible. The binding capacity of the SHBG remained unchanged, suggesting that inhibition of DHT formation does not affect SHBG biosynthesis by the liver. This datum could explain why the free T levels did not change significantly during finasteride treatment, despite the increase in T levels. Finasteride was tolerated well by our patients. No significant adverse effects and no modifications in the cycles were observed during treatment. The frequency of coitus and the sexual desire did not change under finasteride treatment, showing that, at least in women, libido is not modified by the reduction of DHT serum levels. The principal hematochemical parameters examined did not undergo any significant variation in our patients during treatment. This finding confirms that finasteride has low toxicity, as reported previously in men during long-term therapy (8). In conclusion, finasteride was effective in lowering the Ferriman-Gallwey score of idiopathic hirsutism patients with moderate-severe hirsutism. This positive clinical response was accompanied by reduction of DHT and increased total T serum levels, whereas no changes in the other hormones tested were recorded. Few, transient, and nonspecific side effects and no changes of the hematochemical parameters examined were observed during therapy with finasteride. Hence, finasteride could represent an alternative treatment of patients with idiopathic hirsutism, ensuing prescription of a contraceptive method. However, more prolonged and detailed studies are needed, particularly aimed to establish the optimal dosage of the drug.
Acknowledgment. The authors thank Mr. Giuseppe Tomaselli for technical assistance.
REFERENCES 1. Rosenfield RL. The ovary and female sexual maturation. In: Kaplan SA, editor. Clinical pediatric endocrinology. 2nd ed. Philadelphia: WE Saunders, 1989:259-323. Vol. 64, No.2, August 1995
2. Mauvais-Jarvis P. Androgen metabolism in human skin: mechanism of control. In: Martini L, Motta M, editors. Androgens and antiandrogens. New York: Raven Press, 1977: 229-33. 3. Labrie F. Intracrinology. Mol Cell Endocrinol 1991; 78: 113-8. 4. Serafini P, Ablan F, Lobo RA. 5a-Reductase activity in the genital skin of hirsute women. J Clin Endocrinol Metab 1985; 60:349-55. 5. Rittmaster RS, Zwicker H, Thompson DL, Konok G, Norman RW. Androstanediol glucuronide production in human liver, prostate, and skin. Evidence for the importance of the liver in 5a-reduced androgen metabolism. J Clin Endocrinol Metab 1993; 76:977 -82. 6. Siiteri PK, Wilson JD. Dihydrotestosterone in prostatic hypertrophy. I. The formation and the content of dihydrotestosterone in the hypertrophic prostate of man. J Clin Invest 1970;49:1737 -45. 7. Bruchovsky N, Wilson JD. The intranuclear binding oftestosterone and 5-androstan-17-,B-ol-3-one by rat prostate. J BioI Chern 1968;243:5953-60. 8. The MK-906 (Finasteride) Study Group (Bracken B, Geller J, Imperato-McGinley J, Vaughan D, Gormley GJ, et al.). One-year experience in the treatment of benign prostatic hyperplasia with finasteride. J Androl 1991; 12:372-5. 9. Gormley GJ, Stoner E, Rittmaster RS, Gregg H, Thompson DL, Lasseter KC, et al. Effects offinasteride (MK-906), a 5areductase-inhibitor, on circulating androgens in male volunteers. J Clin Endocrinol Metab 1990; 70:1136-41. 10. Brooks JR, Berman C, Hichens M, Primka RL, Reynolds GF, Rasmusson GH. Biological activities of a new steroidal inhibitor of delta 4-5a-reductase. Proc Soc Exp BioI Med 1982; 169:67-73. 11. Rittmaster RS. Androgen conjugates: physiology and clinical significance. Endocr Rev 1993; 14:121-32. 12. Ferriman D, Gallwey JD. Clinical assessment of body hair growth in women. J Clin Endocrinol Metab 1961;21:1440-7. 13. Horton R, Hawks D, Lobo R. 3a, 17,B-androstanediol glucuronide in plasma. A marker of androgen action in idiopathic hirsutism. J Clin Invest 1982;69:1203-6. 14. Andersson S, Russell DW. Structural and biochemical properties of cloned and expressed human and rat steroid 5a-reductases. Proc Natl Acad Sci USA 1990;87:3640-4. 15. Andersson S, Berman DM, Jenkins EP, Russell DW. Deletion of steroid 5a-reductase 2 gene in male pseudohermaphroditism. Nature 1991;354:159-61. 16. Diani AR, Mulholland MJ, Shull KL, Kubicek MF, Johnson GA, Schostarez HJ, et al. Hair growth effects of oral administration offinasteride, a steroid 5a-reductase inhibitor, alone or in combination with topical minoxidil in the balding stumptail macaque. J Clin Endocrinol Metab 1992;74:345-50. 17. Ebling F JG. Hair follicles and associated glands as androgen targets. J Clin Endocrinol Metab 1986; 15:319-39. 18. Miles RA, Cassidenti DL, Carmina E, Gentzschein E, Stanczyk FZ, Lobo RA. Cutaneous application of an androstenedione gel as in vivo test of 5a-reductase activity in women. Fertil Steril 1992;58:708-12. 19. Imperato-McGinley J, Gautier T, Cai LQ, Yee B, Epstein J, Pochi P. The androgen control of sebum production. Studies of subjects with dihydrotestosterone deficiency and complete androgen insensitivity. J Clin Endocrinol Metab 1993;76: 524-8. 20. Carlin JR, Hoglund P, Eriksson LO, Christofalo P, Gregoire SL, Taylor AM, et al. Disposition and pharmacokinetics of [14C] finasteride after oral administration in humans. Drug Metab Dispos 1992;20:148-55.
Ciotta et al. Finasteride in the treatment of IH
305
21. Walsh PC, Madden JD, Harrod MJ, Goldstein JL, MacDonald PC, Wilson JD. Familial incomplete male pseudohermaphroditism, type 2: decreased dihydrotestosterone formation in psedovaginal perineoscrotal hypospadias. N Engl J Med 1974;291:944-9. 22. Fruzzetti F, De Lorenzo D, Parrini D, Ricci C. Effects of finasteride, a 5a-reductase inhibitor, on circulating androgens and gonadotropin secretion in hirsute women. J Clin Endocrinol Metab 1994;79:831-5. 23. Rittmaster RS, Lemay A, Zwicker H, Capizzi TP, Winch S, Moore E, et al. Effect offinasteride, a 5a-reductase inhibitor,
306
Ciotta et al. Finasteride in the treatment of IH
on serum gonadotropins in normal men. J Clin Endocrinol Metab 1992; 75:484-8. 24. Imperato-McGinley J, Peterson RE, Gautier T, Cooper G, Danner R, Arthur A, et al. Hormonal evaluation of a large kindred with complete androgen insensitivity: evidence for secondary 5a-reductase deficiency. J Clin Endocrinol Metab 1982; 54:931-41. 25. McConnell JD, Wilson JD, George FW, Geller J, Pappas F, Stoner E. Finasteride, an inhibitor of 5a-reductase, suppresses prostatic dihydrotestosterone in men with benign prostatic hyperplasia. J Clin Endocrinol Metab 1992;74:505-8.
Fertility and Sterility
©
Vol. 64, No.2, August 1995 Printed on acidfree paper in U. S. A.
1995 American Society for Reproductive Medicine
Clinical and endocrine effects of finasteride, a Sa-reductase inhibitor, in women with idiopathic hirsutism
Lilliana Ciotta, M.D. *t Antonio Cianci, M.D.t AIdo E. Calogero, M.D.* Marco Antonio Palumbo, M.D.t
Elvira Marletta, M.D.t Anna Sciuto, M.D.t Giuseppe Palumbo, M.D.t
University of Catania, Catania, Italy
Objective: To evaluate the effects of long-term administration of finasteride on hirsutism score, basal gonadotropin, and androgen secretion in women with idiopathic hirsutism. Design: Randomized single-blinded study. Patients: Eighteen patients with moderate-severe hirsutism were recruited for the study. Interventions: Nine hirsute patients received 7.5 mg/d oral finasteride for a period of 9 months whereas the other nine were treated with placebo. Hirsutism score, serum basal gonadotropin, androgens, estrogen, and sex hormone-binding globulin (SHBG) levels were evaluated in all patients before treatment and every 3 months during treatment. Results: Mter 6 and 9 months of treatment, the hirsutism score improved significantly in the patients receiving finasteride, whereas no significant modifications were observed in patients treated with placebo. The side effects observed were headache and depression of modest entity during the 1st month of treatments, whereas libido did not change. Serum levels of LH, FSH, androstenedione, unbound T, DHEAS, E 2 , 17a-hydroxyprogesterone, and SHBG did not change during therapy. Hirsute patients treated with finasteride exhibited a marked decrease of dihydrotestosterone and a significant increase of T serum levels from the 3rd and 6th months of treatment, respectively. Conclusion: Finasteride decreased the hirsutism score of patients affected by idiopathic hirsutism with few side effects during treatment. No modification oflibido was observed. Fertil Steril 1995;64:299-306 Key Words: Androgen, 5a-reductase activity, idiopathic hirsutism, finasteride
Androgens are necessary for sexual hair and sebaceous development. In conjunction with other regulatory factors, such as induction factors from the dermal papilla and insulin-like growth factors, they induce the prepuberal pilosebaceous unit in androgen-dependent areas to differentiate either into a terminal hair follicle or into a sebaceous follicle (1). Hyperandrogenism is a common condition in women characterized by excessive growth of terminal hair in a male pattern, hirsutism, and/or by exagger-
Received August 4, 1994; revised and accepted March 3, 1995. * Reprint requests: Lilliana Ciotta, M.D., Ospedale Santo Bambino, Via Torre del Vescovo, 95124 Catania, Italy (FAX: 003995-312001). t Department of Obstetrics and Gynaecology. :j: First Department of Internal Medicine. Vol. 64, No.2, August 1995
ated activity of sebaceous glands, which causes acne, seborrhea, and alopecia. Like obesity and/or anovulation, these symptoms are variably expressed manifestations of androgen excess and thus they may be cutaneous signs of an androgenic disorder. Female hyperandrogenism, in most cases, is dependent on an ovarian or adrenal overproduction of androgens or by hypersensitivity of the pilosebaceous unit towards normal levels of free androgens (idiopathic hirsutism) (2). Because circulating androgen levels are in most cases normal, several data indicate increased local formation of androgens (3) or increased sensitivity of the androgen receptor to normal levels of androgens. Consequently, antiandrogens have been used extensively in the treatment of hirsute women without androgen excess. These compounds act mainly to inhibit binding of androgens to the androgen receptor and they are able to Ciotta et al. Finasteride in the treatment of IH
299
reverse the pilosebaceous unit toward the vellus state (1). Previous studies have documented that hirsute women have a higher 5a-reductase activity than normal women (4), and 5a-reductase is the enzyme responsible for the irreversible conversion ofT to dihydrotestosterone (DHT) (5). It was shown previously that, within the androgen-dependent glandular cells, DHT but not T is the active intracellular androgen (6, 7). Consequently, the inhibitors of 5a-reductase may by used in the treatment ofhyperandrogenism in females. Finasteride is an orally active agent undergoing clinical trials for the treatment of benign prostatic hyperplasia (BPH) (8). This drug, a member of the 4-azasteroid family of compounds, is a competitive inhibitor of steroid 5a-reductase (9). Hence, as hypothesized (10, 11), it is possible that a specific compound with inhibitory activity on 5a-reductase, such as finasteride, may be able to reduce the skin hyperandrogenic symptomatology in affected women. Therefore this study was undertaken to gain information about the clinical and endocrine effects of finasteride in patients affected by idiopathic hirsutism. To accomplish this, we evaluated the effects of long-term treatment of finasteride on hirsutism score, basal gonadotropins, androgen, estrogen, and sex hormone-binding globulin (SHBG) serum levels in nine patients with idiopathic hirsutism. These data were compared with those obtained in other nine idiopathic hirsutism-matched patients treated with placebo. Table 1 Clinical and Endocrinologic Features of Idiopathic Hirsute Patients Before Treatment with Placebo or Finasteride* Parameterst
Placebo
Finasteride
Age (y) Body mass index (kg/m2) Ferriman-Gallwey score Menses LH (mID/mL) FSH (mID/mL) A (ng/mL) Total T (ng/mL) Unbound T (pg/mL) DHEAS (f.lg/mL) DHT (pg/mL) E2 (pg/mL) 17-0HP (f.lg/mL) SHBG (f.lg/mL)
20.2 :+: 0.62 20.7 :+: 0.47 21.8 :+: 0.81:1: Eumenorrhea 5.57:+: 0.42§ 4.39:+: 0.35§ 1.73:+: 0.16§ 0.53:+: 0.01§ 2.70:+: 0.08§ 2.08:+: 0.14§ 374.4 :+: 37.2§ 58.24:+: 2.84§ 384.2 :+: 36.6§ 2.22:+: 0.15§
20.9 :+: 0.61 20.8 :+: 0.36 19.0 :+: 1.57:1: Eumenorrhea 5.57:+: 0.47§ 3.62:+: 0.18§ 1.84:+: 0.12§ 0.53:+: 0.05§ 2.86:+: 0.11§ 2.02:+: 0.10§ 380.0 :+: 29.4§ 52.66:+: 4.45§ 408.9 :+: 23.9§ 2.17:+: 0.13§
* Values are means:+: :j: P
SEM.
< 0.001 versus normal women.
§ P = not significant versus normal women. t The following conversion factors to SI units are applicable for the values cited in the table: LH and FSH, 1.00; A, 3.49; Total or Unbound T, 3.47; DHEAS, 2.71; DHT, 3.45; E2, 3.67; 17-0HP, 3.03; and SHBG, 3.44. 300
Ciotta et al.
Finasteride in the treatment of IH
Table 2 Clinical and Endocrinologic Features of Normal Women* Parameterst Age (y) Body mass index (kg/m2) Ferriman-Gallwey score Menses LH (mID/mL) FSH (mID/mL) A (ng/mL) Total T (ng/mL) Unbound T (pg/mL) DHEAS (f.lg/mL) DHT (pg/mL) E2 (pg/mL) 17 -OHP (f.lg/mL) SHBG (f.lg/mL)
22.4 (19.0 to 27.0) 21.2 (19.0 to 23.0) 6.9 (4.0 to 10.0) Eumenorrhea 5.5 (4.0 to 7.0) 4.8 (3.4 to 6.4) 1.80 (1.35 to 2.20) 0.64 (0.42 to 0.76) 2.69 (2.50 to 2.95) 2.27 (1.60 to 3.00) 326.6 (180.6 to 430.2) 54.7 (45.6 to 68.4) 370.4 (270.6 to 430.5) 2.24 (1.80 to 3.00)
* Values are means with ranges in parentheses. Control group, n = 16.
t The following conversion factors to SI units are applicable for the values cited in the table: LH and FSH, 1.00; A, 3.49; Total or Unbound T, 3.47; DHEAS, 2.71; DHT, 3.45; E2, 3.67; 17-0HP, 3.03; and SHBG, 3.44.
MATERIALS AND METHODS Subjects
Eighteen women (age 20.6 :::':: 0.43 years; mean :::':: SEM) with moderate-severe hirsutism (mean score, according to the Ferriman and Gallwey classification [12], 20.4 :::':: 0.92) were recruited for the study. Their hirsutism score was significantly higher (P < 0.001) (Table 1) than the hirsutism score of 16 normal women with regular menses (control group) (Table 2). No patient had acne-seborrhea and/or other clinical signs of hyperandrogenism. All patients had regular menses, body mass index (mean value 20.7 :::':: 0.29 kg/m 2 ) and were otherwise healthy. Two blood samples were withdrawn during the follicular phase (days 5 to 8) in both hirsute and normal women for hormonal evaluations. Normal serum levels of LH, FSH, androstenedione (A), total and unbound T, DHEAS, DHT, E 2 , 17a-hydroxyprogesterone (17-0HP), SHBG (Table 1), total triiodothyronine, total T 4 , thyroid-stimulating hormone, and PRL were found in all patients. Their LH:FSH ratio was <2. The response of 17-0HP to ACTH stimulation test, performed during the follicular phase of the cycle, was normal in all patients. Ovarian volume was calculated in patients and control subjects according to the formula 4/37f(d l /2 X dJ2 X da/2), in which the diameters were determined by ultrasound (US) as the mean of the length, width, and depth of the ovary. Ovarian size was normal in both nonhirsute (control group) and hirsute women. Ovarian US morphology was normal in the control Fertility and Sterility
group and in 14 hirsute patients (77.8%), whereas multifollicular features was observed in the other 4 hirsute patients (22.2%). Consequently, a diagnosis of idiopathic hirsutism was posed in all women affected by hirsutism. None of the patients had received any hormonal treatment for 2:6 months before the study. After an informed consent was obtained from all patients, they were treated randomly and blindly with placebo (nine patients) or finasteride (Prostide; Sigma-Tau, Milan, Italy) at the dose of 7.5 mg/d orally (nine patients) for 9 months. All patients were urged to use contraception barrier methods or an intrauterine device during the study. Clinical Study
The evaluation of the hirsutism score was performed according to the Ferriman and Gallwey classification (12) before and after 3, 6, and 9 months of treatment with finasteride or placebo. Measurements were performed by a single examiner (L.C.). A general physical examination was performed at the same times during the study. The evaluation of libido was assessed by an interval scale for subjective evaluation (desire) and a semistructured talk (clinical) for objective assessment (frequency of coitus). Endocrine and Hematochemical Studies
The hormonal evaluation was carried out in all patients during their follicular phase (days 5 to 8 of the cycle). The serum concentrations ofLH, FSH, A, total and unbound T, DHEAS, DHT, E 2, 17-0HP, and SHBG were measured every 3 months during treatment with finasteride and placebo. Hematologic, renal, and liver functions were evaluated by measuring hemochrome, hematocrit, glycemia, azotemia, creatinemia, electrophoretic protidogram, total and fractioned bilirubinemia, transaminase, alkaline phosphatase, and gamma-glutamil-transpeptidase every 3 months during treatment. Hormone Determination
Serum levels ofLH, FSH, A, total T, and DHEAS (Sorin, Saluggia, Vercelli, Italy), unbound T and 17OHP (Pantex, Santa Monica, CA), DHT (Bioline, Bruxelles, Belgium), E2 (CIS, Santhia, Vercelli, Italy), and SHBG (Intertech, Strasen, Luxembourg) were measured with double-antibody RIA methods using commercially available kits. All samples were assayed in duplicate. The intra-assay coefficients of variation (CVs) ofLH and FSH were 3.8% and 5.8%, respectively, and the interassay CVs were 10.5% and 9.8%, respectively. The intra-assay and interassay Vol. 64, No.2, August 1995
O·······OPlacebo e-e Finasteride
25 w
20
0:::
o
fA
°T···············O·················O·
e ______ -
e ____ ! e _______*.9 e
15
(!)
lL..
·······0
10
5 oL---~--~--~---~-~
o
3
6
9
MONTHS OF TREATMENT
Figure 1 Ferriman and Gallwey (FG) hirsutism score of patients with idiopathic hirsutism treated with placebo or finasteride. *p < 0.05 versus 0; OP < 0.05 versus 3 months (Duncan multiple range test).
CVs for the other hormones were 7.2% and 10.5% for A, 8.2% and 10.8% for total T, 7.2% and 9.5% for unbound T, 5.6% and 9.5% for DHEAS, 6.2% and 8.2% for DHT, 7.0% and iO.6% for E 2, 6.4% and 8.7% for 17-0HP, and 4.8% and 5.6% for SHBG, respectively. Statistical Analysis
Results are presented as means ± SEM throughout the study. Data were analyzed by using one- or two-way analysis of variance (ANOVA) followed by the Duncan multiple range test. Significance was accepted for P values < 0.05. RESULTS Clinical Evaluation
Finasteride reduced significantly (P < 0.001, oneway ANOVA) the hirsutism score of the patients treated (Fig. 1). After 6 and 9 months of treatment, the mean hirsutism score decreased significantly with respect to pretreatment values (P < 0.05). In addition, the mean value observed after 9 months was significantly lower than that recorded after 3 months of therapy (P < 0.05). No significant modification of the hirsutism score was observed in idiopathic hirsutism patients treated with placebo. Accordingly, the hirsutism score of idiopathic hirsutism patients treated with finasteride was significantly lower than that observed in the idiopathic hirsutism patients treated with placebo (P < 0.01, two-way ANOVA). Hirsutism regression was characterized by a progressive slowing down of hair growth, thinning and lightening of the hairs, followed by a reduction and/or Ciotta et al. Finasteride in the treatment of IH
301
r
DPlacebo _ Finasteride
7.0 6.0 -:J E 5.0 '-... ::J
1 I
4.0 3.0
-'
2.0 1.0 0.0
'-
--
~
5.0 .-..... -' 4.0 E
'-... ::J
E
3.0
I
2.0
'-' (/)
i.L
1.0 0.0
o
-'--
'---
-
3 6 9 LENGTH OF TREATMENT (Mo)
Figure 2 Serum levels ofLH (top) and FSH (bottom) of patients affected by idiopathic hirsutism treated with placebo or finasteride.
disappearance of terminal hairs in the androgendependent body area. Hirsutism decline was seen first on the face (chin, cheeks, upper lip), around the mammary areola, upper limbs (forearm), and lower limbs (upper internal part of the thigh), while it was slower on the chest and lower abdomen. During the period of the study, all patients had regular cycles. In the group of patients treated with finasteride, headache was found in four patients (44.4%) during the 1st month of treatment, and depression in was found in two patients (22.2%) during the first 3 months of therapy. In the group of patients treated with placebo, anxiety and depression was found in four patients (44.4%) during the first 2 months of therapy, and headache was found in three patients (33.3%) during the 1st month. No significant modifications in the frequency of coitus and sexual desire were found during treatment with placebo or finasteride. Endocrine and Hematochemical Evaluation
No significant modifications ofLH, FSH (Fig. 2), A, unbound T, E 2 , 17-0HP, and SHBG (Table 3) serum levels were observed in all patients during the study period. Patients treated with finasteride showed a significant increase of serum total T levels (P 302
Ciotta et a!. Finasteride in the treatment of IH
< 0.001, one-way ANOVA). This increase became significant after 6 and 9 months of treatment (P < 0.05). An increase of total T levels also was observed in the group of idiopathic hirsutism treated with placebo (P < 0.01, one-way ANOVA). However, the increment was of modest entity and it was significant at the 3rd and 9th month of therapy (Fig. 3, top). Total T levels during finasteride treatment were significantly higher than those obtained in the placebo group (P < 0.01, two-way ANOVA). Serum levels of DHT showed a marked decrease in the group of patients treated with finasteride (P < 0.001, one-way ANOVA), whereas no significant modifications were observed in the placebo group (Fig. 3, bottom). Accordingly, DHT levels during finasteride treatment were significantly lower than those obtained in the placebo group (P < 0.01, two-way ANOVA). No significant variations in DHEAS serum levels were observed in both placebo and finasteride groups (Fig. 4). However, patients treated with finasteride had levels of DHEAS significantly lower than those recorded in patients treated with placebo (P < 0.01, two-way ANOVA). No variations were seen in the hematologic, renal, and liver functions. DISCUSSION
An increased sensitivity of the terminal hair to androgens has been hypothesized in hirsute women with normal circulating levels of androgens (2, 13). It generally is accepted that the free androgen fraction diffuses passively into the target cells where, under the influence of 5a-reductase, T is reduced to DHT (5). Both T and DHT bind to the same androgen receptor. However, in target tissue, the affinity of the receptor for DHT is much higher than that for T. Hence, in most target tissue, DHT is the active androgen (6, 7). Two 5a-reductase isoenzymes have been cloned recently (14, 15). The role of 5a-reductase-l is presently unknown. It is present in small quantities in the prostate, a gland rich of 5a-reductase-2, but its presence in other areas rich in 5areductase still is unknown. The 5a-reductase-2 is responsible for differentiation of the male external genitalia and prostate, and it is defective in male pseudohermaphrodites with 5a-reductase deficiency (15). 5a-Reductase-2 is essential in humans for hair growth (16, 17) and it has been shown that there is a significant correlation between the 5a-reductase and the Ferriman and Gallwey score in hirsute women (4, 18). Finasteride is a new antiandrogen that competitively inhibits 5a-reductase-2 (19), producing a profound decrease in serum DHT level. Finasteride itself possesses no androgenic, antiandrogenic, or Fertility and Sterility
Table 3
Hormone Serum Levels in Idiopathic Hirsute Patients Treated with Placebo or Finasteride* Finasteride
Placebo
o
Months A (ng/mL)t Unbound T (pg/mL)t E2
1. 73 ": 0.16
58.2
(pg/mL)§
17-0RP (l'g/mL)11 SRBG
2.8
":
384.2 ": 36.6
(l'g/mL)~
1.88 ": 0.1
1.84 ": 0.12
1.84 ": 0.12
1.82 ": 0.08
1.89 ": 0.09
1.83 ": 0.11
2.82 ": 0.1
2.92 ": 0.09
2.86 ": 0.11
2.97 ": 0.13
2.97 ": 0.11
3.04 ": 0.08
":
1.2
59.4
":
2.4
57.4
404.8 ": 38.4
385.9 ": 37.3
2.22 ": 0.15
* Values are means":
1.83 ": 0.09
60.6
9
1.95 ": 0.11
2.16 ": 0.15
2.2
52.6 ": 4.4
407.1 ": 33.1
408.9 ": 23.9
":
1.96 ": 0.16
t Conversion factor to SI unit. 3.47.
steroid hormone-related properties. Inhibition occurs without affecting the binding of T or DHT to the androgen receptor. The metabolism of fin asteride now has been defined in humans (20). The parent drug is the major circulating component in the plasma. In addition, two monohydroxy metabolites also possess some activity as inhibitors of 5a-reductase. The biologic effect of finasteride persists sig-
0.8
-
-:::::J
E
"0> S
0.6
Placebo Finasteride
*0
*0 *
* J...
~
I....J
~
0.4
0
I-
0.2 0.0
-~
57.4 ": 4.7 381.1
": 16.7
2.08 ": 0.07
6
58.7 ": 4.5 403.3
": 11.2
1.89 ":
0.12
57.9
":
3.7
405.5 ": 20.6 1.87 ": 0.08
§ Conversion factor to SI unit, 3.67. II Conversion factor to SI unit, 3.03. ~ Conversion factor to SI unit, 3.44.
SEM.
t Conversion factor to SI unit. 3.49.
0
2.17 ": 0.13
3
9
6
2.88 ": 0.14
2.7 ": 0.08
o
3
-
l...-'-
-~
nificantly longer than the serum half-life of 5 to 6 hours (20). Previous studies investigating 24 men of 12 different families affected by congenital 5a-reductase deficiency and consequently low DHT levels (21) showed that the female components in these families apparently were unaffected. Hence, a drug able to lower DHT levels and possessing low toxicity (8), such as finasteride, may provide an acceptable medical therapy for hirsute women. Indeed, it has been shown recently that short-term treatment with finasteride improves significantly the hirsutism score in nonselected hirsute patients (22) and that this improvement seemed to be slightly better than that obtained with other antiandrogens, such as fiutamide or cyproterone acetate (CPA) (22). In the present study, we elected to evaluate the clinical and hormonal efficacy and safety of a longterm therapy with finasteride in a group of patients
400 2.5
DPlacebo _ Finasteride
-:::::J
E 300
"01
*
~ I-
:r:
*
200
....J ----2.0 E "01
*
::t
'-'
0
1.5
en
i'3
100
:r:
1.0
0
0
o
~'-
'----
3
6
--
'---
LENGTH OF TREATMENT (Mo)
Figure 3 Serum levels of total T (top) and DHT (bottom) of patients affected by idiopathic hirsutism treated with placebo or finasteride. Serum levels of total T of patients treated with finasteride were significantly higher than those treated with placebo (P < 0.01, two-way ANOVA), whereas those of DHT were significantly lower (P < 0.01, two-way ANOVA). *P < 0.05 versus 0, OP < 0.05 versus 3 months (Duncan multiple range test). Vol. 64, No.2, August 1995
0.5
9
0.0
036 9 LENGTH OF TREATMENT (Mo)
Figure 4 Serum levels of DHEAS of patients affected by idiopathic hirsutism treated with placebo or finasteride. Serum levels of DHEAS of patients treated with finasteride were significantly lower than those found in patients treated with placebo (P < 0.01, two-way ANOVA).
Ciotta et al. Finasteride in the treatment of IH
303
p
affected by idiopathic hirsutism. In a preliminary study, a group of nons elected hirsute patients were treated long term with 5 mg/d offinasteride, the dose usually used for the treatment ofBPH in men. Albeit there was an improvement of their hirsutism score, it seemed slower than that obtained with other antiandrogens in the same patients. Given the low toxicity of finasteride (8), we thought to perform this study using the dose of 7.5 mg/d. Oral contraceptives were not prescribed to the patients enrolled in this study to provide a direct evaluation of finasteride as a single agent. To avoid feminization of a male fetus, secondary to reduction of DHT during embryogenesis (21), all patients were urged to use alternative, nonhormonal, contraceptive methods. As suggested previously (10, 11) and demonstrated recently (22), our study showed that finasteride is effective in the treatment of hirsutism because all patients had a significant reduction of their hirsutism score during treatment. This reduction became significant after 6 months and it further decreased after 9 months oftherapy. The improvement of the Ferriman-Gallwey score seemed to be similar in time and degree to that obtained with the use of other antiandrogens, such as flutamide, whereas it seemed to be better to that observed during treatment with CPA or spironolactone and oral estrogens in patients with a moderate-severe degree of hirsutism (unpublished observations). The administration of 7.5 mg/d of the drug did not seem to induce a better clinical response after 3 months of treatment in women with idiopathic hirsutism in comparison to that reported in nons elected hirsute women treated with a dose of 5 mg/d for 3 months (22). However, the observed further significant reduction ofthe hirsutism score in our patients after 9 months of treatment with respect to the values seen after 3 months and the small number of women tested previously (22) and in this study must be taken into account before any definitive conclusion can be drawn. According to the mechanism of action of fin asteride, a significant decrease of DHT serum levels already was observed after 3 months of treatment. This decrease preceded and paralleled the reduction of the hirsutism score. The other hormones did not show any significant change with the exception of total T, which increased significantly after 6 months of treatment. Accordingly, total T previously was found to be increased in normal men (9, 23), in patients with BPH (8), and in hirsute women (22) during treatment with finasteride. The fact that the increased total T serum levels did not modify the positive clinical response observed during therapy confirms the hypothesis that DHT is the essential 304
Ciotta et al. Finasteride in the treatment of IH
androgen for hair growth (19, 20). In our study, the increase of total T serum levels in the group of patients treated with placebo after 3 and 9 months was of modest entity and it may relate to seasonal hormonal variations. We did not observe any changes in basal gonadotropin levels in patients treated with finasteride, as reported previously in men (8,9,23) and in hirsute women (22). Recently, it has been reported that short-term therapy with finasteride does not modify both LH pulse frequency and amplitude in hirsute patients (22). In addition, finasteride treatment does not influence the release of LH and FSH in response to GnRH both in normal men (23) and in hirsute women (22). The unchanged levels of gonadotropins, even during the administration of 7.5 mg/d, suggest that the gonadotropin secretion is not influenced significantly by direct reduction of serum DHT levels or by indirect increase of total T serum levels induced by finasteride. However, it should be pointed out that LH levels in male pseudohermaphrodites with congenital5a-reductase deficiency, although elevated, are substantially lower than LH levels seen in castrated subjects (24), suggesting that, albeit DHT may playa role in the negative feedback effect, the major feedback on LH is exerted by T directly or by conversion to E2 in the general circulation or at the hypothalamic level. It also can be postulated that it is not the circulating DHT levels that exert a negative effect on LH but hypothalamic and/or pituitary DHT and that finasteride does not cross the blood-brain barrier sufficiently to alter these levels. Because of the lack of finasteride effects on both basal and stimulated gonadotropin release (22, 23), the increase of total T serum levels may not be ascribed to a more sustained LH-dependent ovarian theca-interstitial cell stimulation. Furthermore, an adrenal overproduction of T cannot explain the observed increase of total T serum levels in our patients because the unmodified DHEAS levels during treatment with finasteride suggest that the drug has no influence on the adrenal androgenesis. This datum confirms what was found previously in hirsute women, in whom the therapy with finasteride does not determine any significant modification in both basal and ACTH-induced release of A and DHEAS with respect to pretreatment values (22). On the contrary, in our study we observed a significant reduction of DHEAS serum levels in the group of patients treated with finasteride compared with values obtained in the placebo group. Therefore, it is possible that the selective blockade of a major metabolic pathway for T, resulting from the inhibition of 5areductase by finasteride rather than an increase of Fertility and Sterility
its secretion, may cause the increase of total T serum levels. In fact, 5a reduction is a major metabolic pathway for T in the prostate, and inhibition of this reaction probably may explain the observed accumulation of substrate T in the gland (25). However, if the increased T serum levels may be related to the reduced conversion of T to DHT, it is difficult to explain why we did not observe a parallel increase in A serum levels, that, in the target tissues, is converted by 5a-reductase to androstanedione (5). Furthermore, the conversion of A to T is catalyzed by the enzyme 17 -ketosteroid reductase and this reaction is reversible. The binding capacity of the SHBG remained unchanged, suggesting that inhibition of DHT formation does not affect SHBG biosynthesis by the liver. This datum could explain why the free T levels did not change significantly during finasteride treatment, despite the increase in T levels. Finasteride was tolerated well by our patients. No significant adverse effects and no modifications in the cycles were observed during treatment. The frequency of coitus and the sexual desire did not change under finasteride treatment, showing that, at least in women, libido is not modified by the reduction of DHT serum levels. The principal hematochemical parameters examined did not undergo any significant variation in our patients during treatment. This finding confirms that finasteride has low toxicity, as reported previously in men during long-term therapy (8). In conclusion, finasteride was effective in lowering the Ferriman-Gallwey score of idiopathic hirsutism patients with moderate-severe hirsutism. This positive clinical response was accompanied by reduction of DHT and increased total T serum levels, whereas no changes in the other hormones tested were recorded. Few, transient, and nonspecific side effects and no changes of the hematochemical parameters examined were observed during therapy with finasteride. Hence, finasteride could represent an alternative treatment of patients with idiopathic hirsutism, ensuing prescription of a contraceptive method. However, more prolonged and detailed studies are needed, particularly aimed to establish the optimal dosage of the drug.
Acknowledgment. The authors thank Mr. Giuseppe Tomaselli for technical assistance.
REFERENCES 1. Rosenfield RL. The ovary and female sexual maturation. In: Kaplan SA, editor. Clinical pediatric endocrinology. 2nd ed. Philadelphia: WE Saunders, 1989:259-323. Vol. 64, No.2, August 1995
2. Mauvais-Jarvis P. Androgen metabolism in human skin: mechanism of control. In: Martini L, Motta M, editors. Androgens and antiandrogens. New York: Raven Press, 1977: 229-33. 3. Labrie F. Intracrinology. Mol Cell Endocrinol 1991; 78: 113-8. 4. Serafini P, Ablan F, Lobo RA. 5a-Reductase activity in the genital skin of hirsute women. J Clin Endocrinol Metab 1985; 60:349-55. 5. Rittmaster RS, Zwicker H, Thompson DL, Konok G, Norman RW. Androstanediol glucuronide production in human liver, prostate, and skin. Evidence for the importance of the liver in 5a-reduced androgen metabolism. J Clin Endocrinol Metab 1993; 76:977 -82. 6. Siiteri PK, Wilson JD. Dihydrotestosterone in prostatic hypertrophy. I. The formation and the content of dihydrotestosterone in the hypertrophic prostate of man. J Clin Invest 1970;49:1737 -45. 7. Bruchovsky N, Wilson JD. The intranuclear binding oftestosterone and 5-androstan-17-,B-ol-3-one by rat prostate. J BioI Chern 1968;243:5953-60. 8. The MK-906 (Finasteride) Study Group (Bracken B, Geller J, Imperato-McGinley J, Vaughan D, Gormley GJ, et al.). One-year experience in the treatment of benign prostatic hyperplasia with finasteride. J Androl 1991; 12:372-5. 9. Gormley GJ, Stoner E, Rittmaster RS, Gregg H, Thompson DL, Lasseter KC, et al. Effects offinasteride (MK-906), a 5areductase-inhibitor, on circulating androgens in male volunteers. J Clin Endocrinol Metab 1990; 70:1136-41. 10. Brooks JR, Berman C, Hichens M, Primka RL, Reynolds GF, Rasmusson GH. Biological activities of a new steroidal inhibitor of delta 4-5a-reductase. Proc Soc Exp BioI Med 1982; 169:67-73. 11. Rittmaster RS. Androgen conjugates: physiology and clinical significance. Endocr Rev 1993; 14:121-32. 12. Ferriman D, Gallwey JD. Clinical assessment of body hair growth in women. J Clin Endocrinol Metab 1961;21:1440-7. 13. Horton R, Hawks D, Lobo R. 3a, 17,B-androstanediol glucuronide in plasma. A marker of androgen action in idiopathic hirsutism. J Clin Invest 1982;69:1203-6. 14. Andersson S, Russell DW. Structural and biochemical properties of cloned and expressed human and rat steroid 5a-reductases. Proc Natl Acad Sci USA 1990;87:3640-4. 15. Andersson S, Berman DM, Jenkins EP, Russell DW. Deletion of steroid 5a-reductase 2 gene in male pseudohermaphroditism. Nature 1991;354:159-61. 16. Diani AR, Mulholland MJ, Shull KL, Kubicek MF, Johnson GA, Schostarez HJ, et al. Hair growth effects of oral administration offinasteride, a steroid 5a-reductase inhibitor, alone or in combination with topical minoxidil in the balding stumptail macaque. J Clin Endocrinol Metab 1992;74:345-50. 17. Ebling F JG. Hair follicles and associated glands as androgen targets. J Clin Endocrinol Metab 1986; 15:319-39. 18. Miles RA, Cassidenti DL, Carmina E, Gentzschein E, Stanczyk FZ, Lobo RA. Cutaneous application of an androstenedione gel as in vivo test of 5a-reductase activity in women. Fertil Steril 1992;58:708-12. 19. Imperato-McGinley J, Gautier T, Cai LQ, Yee B, Epstein J, Pochi P. The androgen control of sebum production. Studies of subjects with dihydrotestosterone deficiency and complete androgen insensitivity. J Clin Endocrinol Metab 1993;76: 524-8. 20. Carlin JR, Hoglund P, Eriksson LO, Christofalo P, Gregoire SL, Taylor AM, et al. Disposition and pharmacokinetics of [14C] finasteride after oral administration in humans. Drug Metab Dispos 1992;20:148-55.
Ciotta et al. Finasteride in the treatment of IH
305
21. Walsh PC, Madden JD, Harrod MJ, Goldstein JL, MacDonald PC, Wilson JD. Familial incomplete male pseudohermaphroditism, type 2: decreased dihydrotestosterone formation in psedovaginal perineoscrotal hypospadias. N Engl J Med 1974;291:944-9. 22. Fruzzetti F, De Lorenzo D, Parrini D, Ricci C. Effects of finasteride, a 5a-reductase inhibitor, on circulating androgens and gonadotropin secretion in hirsute women. J Clin Endocrinol Metab 1994;79:831-5. 23. Rittmaster RS, Lemay A, Zwicker H, Capizzi TP, Winch S, Moore E, et al. Effect offinasteride, a 5a-reductase inhibitor,
306
Ciotta et al. Finasteride in the treatment of IH
on serum gonadotropins in normal men. J Clin Endocrinol Metab 1992; 75:484-8. 24. Imperato-McGinley J, Peterson RE, Gautier T, Cooper G, Danner R, Arthur A, et al. Hormonal evaluation of a large kindred with complete androgen insensitivity: evidence for secondary 5a-reductase deficiency. J Clin Endocrinol Metab 1982; 54:931-41. 25. McConnell JD, Wilson JD, George FW, Geller J, Pappas F, Stoner E. Finasteride, an inhibitor of 5a-reductase, suppresses prostatic dihydrotestosterone in men with benign prostatic hyperplasia. J Clin Endocrinol Metab 1992;74:505-8.
Fertility and Sterility