0021-972X/00/$03.00/0 Vol. 85, No.

1
The Journal of Clinical Endocrinology & Metabolism Printed in U.S.A.
Copyright © 2000 by The Endocrine Society

Comparison of Spironolactone, Flutamide, and

Finasteride Efficacy in the Treatment of Hirsutism: A
Randomized, Double Blind, Placebo-Controlled Trial*
PAOLO MOGHETTI, FLAVIA TOSI, ANTONELLA TOSTI, CARLO NEGRI,
COSIMO MISCIALI, FABRIZIA PERRONE, MARCO CAPUTO, MICHELE MUGGEO,
AND ROBERTO CASTELLO

Downloaded from https://academic.oup.com/jcem/article/85/1/89/2852067 by guest on 26 April 2024

Division of Endocrinology and Metabolic Diseases, University of Verona, I-37126 Verona; the Division
of Dermatology, University of Bologna (A.T., C.M.), I-40138 Bologna; and Laboratory of Clinical
Chemistry, Ospedale Maggiore (M.C.), I-37126 Verona, Italy

ABSTRACT diameters and F-G scores. After 6 months of therapy all groups of
To compare objectively the efficacies of spironolactone (100 mg/ subjects given active drugs showed reductions of their hair diameters,
day), flutamide (250 mg/day), and finasteride (5 mg/day) in the treat- without statistically significant differences among groups (mean
ment of hirsutism, 40 hirsute women were randomly assigned to change 6 SEM, 211.7 6 5.6%, 218.0 6 6.1%, and 212.6 6 6.7%,
double blind treatments with 1 of these 3 drugs or placebo for 6 respectively, in the spironolactone, flutamide, and finasteride
months. Before and at the end of treatment, hirsutism was quanti- groups). F-G scores were also significantly reduced in women receiv-
tatively measured in each subject by determination, by computer- ing antiandrogen drugs, again without differences among groups
assisted light microscopy, of the largest diameter of 5 hairs plucked (mean change, 241.0 6 5.5%, 238.9 6 7.2%, and 231.6 6 3.7%,
from the linea alba. These measurements were averaged to produce respectively). No significant changes from baseline values were re-
a mean hair shaft diameter. For each subject, baseline and posttreat- corded by either hair diameter (21.4 6 5.2%) or F-G score (15.4 6
ment assessments were carried out at the same time by an investi- 3.7%) assessment in the placebo group. In conclusion, spironolactone,
gator blinded to both time and type of therapy. In addition, a semi- flutamide, and finasteride are all effective in the treatment of hir-
quantitative clinical evaluation was carried out by a modification of sutism. After a 6-month course of therapy, the clinical efficacies of
the Ferriman-Gallwey (F-G) scoring method, performed by a single these drugs, at least at the doses used, are similar. (J Clin Endocrinol
investigator. At baseline the 4 groups of women had similar hair Metab 85: 89 –94, 2000)

H IRSUTISM is a very common clinical problem in endo-

crinological practice, with potentially serious psycho-
social consequences (1). Mechanical hair removal is effective in
irregularity. The progestinic activity of cyproterone acetate
requires this drug to be associated with estrogens.
Other antiandrogen drugs, such as flutamide and finas-
many hirsute women. Nevertheless, a pharmacological ap- teride, have been proposed in the treatment of hirsutism.
proach is often required in subjects with moderate to severe Flutamide is a nonsteroidal compound that seems to act only
hirsutism to suppress androgen production and/or action. at the androgen receptor site and is therefore considered a
Contraceptives are widely used for this purpose, but their ef- pure antiandrogen (6). However, some data suggest that
ficacy is limited in established hirsutism (2). Although GnRH flutamide might also reduce the synthesis of androgens (7)
analogs may be indicated in selected patients (3), the most and/or increase their metabolism to inactive molecules (8).
reliable therapeutic tool in these women is the use of antian- This drug is efficaciously used in the treatment of advanced
drogen drugs, usually in association with nonandrogenic oral prostatic carcinoma and was successfully evaluated in hir-
contraceptive therapy (2). sute women (9 –11). Liver toxicity is a rare but potentially
Spironolactone and cyproterone acetate are the antiandro- severe side-effect of flutamide (12).
gens most commonly used in the treatment of hirsutism. Both Finasteride is a very potent competitive inhibitor of the
of these steroidal compounds possess intrinsic hormonal type 2 isoenzyme of 5a-reductase, the enzyme responsible
activity and interfere with steroidogenesis (4, 5). In addition, for conversion of testosterone to the active metabolite dihy-
cyproterone acetate shows significant antigonadotropic ef- drotestosterone (13). The drug has recently been approved
fects. Side-effects of these drugs include frequent menstrual for the treatment of benign prostatic hyperplasia (14). As
increased 5a-reductase activity is considered a pathogenetic
mechanism of hirsutism (15), selective enzyme inhibition has
Received May 21, 1999. Revision received August 23, 1999. Accepted
September 10, 1999.
been proposed as a rational medical approach to this con-
Address all correspondence and requests for reprints to: Dr. Paolo dition as well. Consistently, studies in hirsute women gave
Moghetti, Divisione di Endocrinologia e Malattie del Metabolismo, Osped- promising results, without appreciable side-effects (16 –18).
ale Maggiore, I-37126 Verona, Italy. E-mail: moghetti@iol.it. Rigorous clinical trial methodology is very rarely encoun-
* Presented in part at the 80th Annual Meeting of The Endocrine Society,
New Orleans, Louisiana, June 24 –27, 1998. This work was supported by
tered in reports evaluating antiandrogenic therapies in hir-
grants from the Italian Ministry of Higher Education and Scientific Research sute women (1, 19, 20). Furthermore, the majority of re-
and the Regione del Veneto (DGRV 964 no. 652 and 693). searchers who previously assessed the effects of these drugs

89
90 MOGHETTI ET AL. JCE & M • 2000
Vol 85 • No 1

used subjective parameters, namely the Ferriman-Gallwey uation of hirsutism in each patient was obtained by averaging the largest
(F-G) (21) or other similar scoring systems, as the only meth- diameters of five hairs plucked, using regular facial tweezers, from a 2 3
ods to measure hirsutism, making comparison of data among 2-cm area in the linea alba. To avoid any hair damage, hairs were not
cut, shaved, or submitted to any cosmetic procedures in this specific area
different studies unreliable (19). In addition, in several stud- for 3 months beforehand. Plucking was immediately followed by shav-
ies antiandrogens were given in combination with oral con- ing, and no additional cosmetic measure was allowed in this body region
traceptives, making it difficult to quantify the therapeutic throughout the study. Hairs were gently embedded in Eukitt (mounting
efficacies of the antiandrogen compound and the estrogens medium O, Kindler GmbH & Co., Freiburg, Germany) on a glass mi-
separately. Until now only a few studies directly compared croslide, using a small anatomical forceps; they were mounted longi-
tudinally on the slide and protected by a cover glass previously bathed
the clinical efficacies of different antiandrogen drugs in hir- with xilol.
sute women. Although some of these studies used objective The slides were examined using a fully integrated, optical microscope
methods to measure hair growth (17, 22, 23), none of them (Leitz DMRB, Wetzlar, Germany), with a 12-V, 100-watt halogen illu-
was double blind or placebo controlled. Thus, the scale of minator for transmitted light and a computer workstation system, the
relative potency of these drugs remains undetermined. CAS 200/486 imaging analysis system (Cell Analysis Systems, Inc.,

Downloaded from https://academic.oup.com/jcem/article/85/1/89/2852067 by guest on 26 April 2024

Elmhurst, IL). This device included an IBM enhanced AT computer, two
To address this issue, the present study compared the full-color display monitors, a graphics and system control/menu selec-
clinical efficacy, in 40 hirsute women, of a 6-month course of tion monitor, and a digital image display for displaying real-time digital
double blind, placebo-controlled treatments with spirono- images. Internally, images are stored at 256 3 256 pixel resolution, with
lactone, flutamide, or finasteride. Hair growth was estimated the corresponding image resolution determined by the objective lens
by both a modification of the F-G score and the objective chosen (27). The image analysis program used was Micrometer version
measurement of shaft diameters of hairs plucked from the 0.7, application 1992 (Cell Analysis Systems, Inc., version 1.0). Hair shaft
diameter was measured at 310 magnification, and the Video Trace mode
linea alba. was used for drawing a line from one point to another. Vertical reso-
lution was 0.0766053/mm, and horizontal resolution was 0.0446314/mm.
Materials and Methods The hair shaft diameter was measured just above the keratogenous zone.
Subjects This zone appears as a dark area of the hair shaft located above the hair
bulb. In this area, hair matrix cells undergo complete keratinization, and
Forty young women (age, 20.4 6 0.5 yr; body mass index, 24.5 6 0.7 the hair fiber decreases in diameter by about 25%, mainly because of
kg/m2; mean 6 sem), consecutively referred to our division for mod- water loss. A mean hair shaft diameter was obtained for each patient by
erate to severe hirsutism, were included in the study. The mean modified averaging the measurements obtained from the anagen hairs present in
F-G score (24) was 17.5 6 0.7 (range, 11–27.5). Two subjects also suffered the sample.
from mild acne. All measurements relative to each patient were made at the same time
Twenty-one women had polycystic ovary syndrome, diagnosed by a single investigator blinded to both time and type of therapy. With
according to the presence of hyperandrogenism and chronic anovu- this method, both intra- and intersubject mean coefficients of variation
lation (25). All of them showed a 17-hydroxyprogesterone hyperre- were less than 2%.
sponse to GnRH agonist testing (26). Eighteen of these subjects had A semiquantitative, clinical evaluation of hirsutism was also per-
oligoamenorrhea. formed in these subjects by a modification of the F-G method (24). The
Cushing’s syndrome, adrenal enzyme defects, adrenal and ovarian hirsutism scores were determined twice in the pretreatment period, with
tumors, hyperprolactinemia, and thyroid dysfunction were excluded in
a 3-month interval between measurements, and subsequently at the end
each subject. No patient suffered from any other disease or had been
of the 6-month treatment period. Patients using cosmetic measures were
treated with oral contraceptives or antiandrogen drugs in the previous
requested not to depilate for at least 1 month before each evaluation. All
12 months.
evaluations were carried out by a single investigator, blinded to the
Patients were clearly informed of potential risks of the treatments,
with a particular caution to avoid pregnancy because of possible male ongoing therapy. The mean difference in hirsutism scores for each sub-
fetus feminization. Sexually active women were advised to use barrier ject between the two baseline evaluations was 0.4 (range, 0 –3.5).
contraceptive methods or intrauterine devices during the study. The The patients’ subjective opinion of the clinical outcome of therapy
study was conducted in accordance with the Declaration of Helsinki on (excellent, good, fair, or poor) was also obtained. In addition, before and
human experimentation. Each patient gave her written informed con- at the end of treatment each woman completed a questionnaire speci-
sent to the study protocol, which was approved by the local ethical fying any cosmetic measures for hair removal, with details of type
committee. (plucking, waxing, shaving, bleaching, or depilatory cream), site, and
frequency.
Protocol
Patients were randomly assigned to double blind treatments, once Endocrine assessments
daily orally as a wafer capsule, for 6 months with one of the following:
1) spironolactone (100 mg), 2) flutamide (250 mg), 3) finasteride (5 mg), A standard hormonal profile, including serum gonadotropins and
or 4) placebo. Basally and at the end of treatment, hirsutism and hor- androgens (total and free testosterone, dehydroepiandrosterone sulfate,
monal parameters were evaluated in each subject, as described below. androstenedione, and 3a-androstanediol glucuronide), was determined
Patients were instructed to report any untoward effect during the at baseline and at the end of treatment from a blood sample obtained at
treatment period. In addition, safety parameters were assessed before 0800 h. Twenty-four-hour urine was collected at the same time points for
treatment and at 2-month intervals during the study. The safety eval- C19 and C21 steroid metabolite determinations. In the eumenorrheic
uation comprised hematology tests (hemoglobin, red blood cells, white patients blood samples were collected in the early follicular phase of the
blood cells, platelets, and leukocyte differential count) and biochemistry menstrual cycle, whereas in the oligoamenorrheic subjects luteal phase
tests (plasma glucose, uric acid, liver and renal function, and serum was excluded by serum progesterone assay.
electrolytes). To be considered clinically significant, changes in safety
parameters had to either exceed the normal limits or double the baseline
values. Assays

Hirsutism assessment Serum hormones were measured by commercial kits, as previously

described (16). All samples from each patient were run in the same assay,
Hair growth assessment included both an objective method and a in duplicate. Urinary steroid metabolites were assayed by gas chroma-
hirsutism score, graded by a modified F-G method. An objective eval- tography, as previously described (16).
 ANTIANDROGENS IN HIRSUTISM 91

Statistics and calculations Safety parameters did not show significant changes in any
Results were analyzed by Student’s t test for paired and unpaired subject during the study period.
data, Wilcoxon’s rank sum test, and ANOVA. All tests of significance
were two tailed, and P # 0.05 was considered significant. Etiochol-
anolone/androsterone and tetrahydrocortisol/allotetrahydrocortisol Hirsutism
urinary metabolite ratios were calculated as indexes of 5a-reductase At baseline, mean shaft diameters of hairs plucked from
activity. Data were expressed as the mean 6 sem.
the linea alba and modified F-G scores were comparable in
Results the four groups of hirsute women (Table 1). After 6 months
of therapy, all groups of subjects given active drugs showed
Tolerability
significant decreases in hair diameter (Fig. 1), without dif-
Metrorrhagias were reported by five women, all given ferences among groups (144 6 12 vs. 164 6 8, 139 6 10 vs.
spironolactone. This side-effect was transient in four subjects 172 6 13, and 146 6 8 vs. 172 6 9 mm, respectively, in the
and sustained in one, but not so severe as to require inter- spironolactone, flutamide, and finasteride groups; all P ,

Downloaded from https://academic.oup.com/jcem/article/85/1/89/2852067 by guest on 26 April 2024

ruption of treatment. On the other hand, previous menstrual 0.01 vs. changes in the placebo group). Hirsutism scores also
abnormalities improved after treatment in five patients showed significant improvements in women receiving an-
(three in the flutamide group and one in each of the other tiandrogen drugs (Fig. 1), without statistically significant
groups receiving active drugs). differences among groups (10.0 6 1.1 vs. 16.9 6 0.9, 11.1 6
One patient receiving flutamide complained of sleepiness 1.8 vs. 17.5 6 1.5, and 13.0 6 1.3 vs. 18.4 6 1.3 in the three
and hyporexia, which spontaneously disappeared after the groups; all P , 0.001 vs. placebo). Similar improvements
first month of therapy. One woman in the finasteride group were found in women with polycystic ovary syndrome and
complained of a transient sensation of being “swollen.” Fi- in those with nonovarian hyperandrogenism considered as
nally, one patient in the placebo group reported mild, tran- a whole, regardless of treatment group (changes in hair di-
sient headache and nausea. ameter, 211.9 6 4.8% vs. 215.7 6 5.2%; changes in hirsutism
TABLE 1. Baseline clinical characteristics of hirsute women divided according to treatment

Spironolactone Flutamide Finasteride Placebo Pa

No. 10 10 10 10
Age (yr) 19.7 6 0.7 20.2 6 1.1 19.8 6 0.8 21.8 6 1.3 NS
BMI (kg/m2) 25.3 6 1.4 23.6 6 1.0 23.3 6 0.7 25.8 6 2.0 NS
Modified Ferriman-Gallwey score 16.9 6 0.9 17.5 6 1.5 18.4 6 1.3 17.2 6 1.6 NS
Hair diameter (mm) 164 6 8 172 6 13 172 6 9 153 6 7 NS
Menses (irregular/regular) 2/8 8/2 4/6 4/6 NS
PCOS/other 4/6 8/2 4/6 5/5 NS
Cosmetic measures (yes/no) 5/5 4/6 8/2 7/3 NS
a
By ANOVA or k2, as appropriate.

FIG. 1. Changes after therapy (D%) in modified F-G score and mean hair shaft diameter in the four treatment groups.
92 MOGHETTI ET AL. JCE & M • 2000
Vol 85 • No 1

score, 241.6 6 4.3% vs. 231.0 6 4.8%). No significant changes tone as well as those given placebo did not show any changes
from baseline values were found by either hair diameter in endocrine parameters.
(156 6 11 vs. 153 6 7 mm) or hirsutism score (18.0 6 1.7 vs.
17.2 6 1.6) in the placebo group.
Before treatment 24 women were using cosmetic mea- Discussion
sures. These consisted of waxing (n 5 20), waxing and pluck- The vast majority of the published work concerning hir-
ing (n 5 2), waxing and shaving (n 5 1), or depilatory cream sutism treatment shows considerable shortcomings, such as
(n 5 1), with variable frequency (2–30 days). Changes in lack of a control group, assessment of results only by sub-
these procedures were reported only by patients given active jective methods, or too short a duration in relation to the
drugs. Four of these subjects (2 in the spironolactone and 2 physiology of hair growth (1, 19, 20). These limitations make
in the flutamide groups) stopped waxing. All of the other 13 it at present impossible to establish a scale of relative potency
women given antiandrogens reported a reduction in fre- among antiandrogen drugs. Indeed, some of the few con-
quencies of hair removal procedures, without substantial trolled studies were even unable to demonstrate differences

Downloaded from https://academic.oup.com/jcem/article/85/1/89/2852067 by guest on 26 April 2024

differences among groups (data not shown). On the other in hair growth during antiandrogen therapy vs. placebo ad-
hand, no change was reported by women receiving placebo. ministration (4). Furthermore, until now only a limited num-
Patients’ self-evaluations of clinical outcome at the end of ber of studies directly compared the clinical efficacies of
the study were also consistent with changes in hair diameter different antiandrogen drugs in the treatment of hirsute
and hirsutism score (Table 2). women (17, 22, 23, 28 –34). Although some of these studies
were randomized (17, 22, 28, 31), only a few used objective
Clinical and endocrine features measures of hair growth (17, 22, 23), and none of them was
Tables 1 and 3 show the main clinical and endocrine fea- double blind or placebo controlled.
tures of women in the four groups, before and during the The present study is the first to evaluate the efficacy on
trial. At baseline all groups showed similar characteristics. hirsutism of three different drugs, spironolactone, flut-
The frequency of menstrual irregularities was higher in the amide, and finasteride, by a rigorous clinical trial meth-
flutamide group, but this difference was not statistically odology of a double blind, placebo-controlled, random-
significant. ized study. Furthermore, hair growth changes were
After treatment a significant reduction of serum dehydro- measured not only by the subjective F-G scoring, com-
epiandrosterone sulfate was found in the flutamide group, monly used in clinical practice as a semiquantitative mea-
whereas 3a-androstanediol glucuronide levels, an index of sure of degree of hirsutism, but also by an objective
5a-reductase activity, were significantly reduced in the fin- method, measurement of hair shaft diameters. This
asteride group. Consistently, women given finasteride also method was previously validated and used in clinical tri-
showed significant increases in C19 and C21 urinary 5b/5a- als (19), although its sensitivity, at least using simple op-
steroid metabolite ratios (data not shown). Free testosterone tical microscopy, is not high (3, 19). To increase both the
showed a tendency to decrease in the flutamide group and sensitivity and reproducibility of this procedure, in the
to increase in the finasteride group, but these differences did present study a computer-assisted measurement was per-
not reach statistical significance. Women given spironolac- formed. On the other hand, although other procedures
have been proposed for the objective measurement of hair
growth, at present none of them may be considered a gold
TABLE 2. Patients’ self-evaluation of clinical outcome of the
therapy standard (19). Our data were further supported by con-
sistent results of patients’ self-evaluation of clinical out-
Spironolactone Flutamide Finasteride Placebo come and by changes in frequency of recourse to mechan-
Excellent 3 4 1 0 ical hair removal.
Good 1 4 4 0 In this study the changes in F-G score were 2–3 times
Fair 5 1 4 1 greater than those in hair diameter. This observation is not
Poor 1 1 1 9
surprising, as subjective perception of hair growth excess

TABLE 3. Standard hormonal profiles, before and after treatment, in the four treatment groups

Spironolactone Flutamide Finasteride Placebo

Basal After Basal After Basal After Basal After
treatment treatment treatment treatment
Free testosterone 3.47 6 0.31 3.43 6 0.24 3.36 6 0.45 2.78 6 0.27 3.50 6 0.41 4.24 6 0.42 3.21 6 0.29 3.25 6 0.32
(pg/mL)
DHEA-S (mg/L) 2177 6 319 2336 6 270 1962 6 231 1349 6 188a 2380 6 175 2079 6 183 2158 6 229 1707 6 134
3a-Androstanediol 4.24 6 0.56 5.64 6 0.66 4.05 6 0.71 3.54 6 0.40 6.40 6 1.23 2.86 6 0.78a 5.05 6 0.80 5.06 6 0.64
glucuronide
(ng/mL)
LH (IU/L) 7.7 6 1.4 8.3 6 1.4 5.5 6 0.9 8.3 6 1.7 6.5 6 1.1 6.6 6 1.5 7.4 6 1.1 7.1 6 1.7
FSH (IU/L) 4.9 6 0.2 5.3 6 0.2 5.1 6 0.3 4.8 6 0.4 4.9 6 0.2 4.7 6 0.5 5.5 6 0.4 5.6 6 0.5
Values are the mean 6 SEM.
a
P , 0.01 vs. basal.
 ANTIANDROGENS IN HIRSUTISM 93

depends not only on hair diameter, but also on the length and of tolerability, finasteride, devoid of appreciable side-effects,
density of terminal hairs. seems to be the current best choice. It should be borne in
The present controlled trial demonstrates that spironolac- mind that all antiandrogens imply the need to avoid a preg-
tone, flutamide, and finasteride are effective in the treatment nancy, given the potential risk of feminization of male
of hirsutism, supporting conclusions of previous, almost all fetuses.
uncontrolled, studies (17, 22, 23, 28 –30, 32). However, the Finally, at the doses used in this study the retail costs of
most interesting finding of the present study is that, in a a 1-month course of therapy are $21.3, $95.4, and $56.1, re-
population of unselected women with moderate to severe spectively for spironolactone, flutamide, and finasteride
hirsutism, the clinical efficacies of these drugs were similar (source of data: drugstore.com web site, July 26, 1999). This
despite their differing mechanisms of action. aspect should also be considered in the choice of an antian-
Spironolactone has been given to hirsute women in pre- drogen therapy, particularly as the efficacies of these drugs
vious studies in doses ranging from 50 – 400 mg/day (4). The are similar.
large majority of researchers used daily doses of 100 mg, as In conclusion, this double blind, placebo-controlled study

Downloaded from https://academic.oup.com/jcem/article/85/1/89/2852067 by guest on 26 April 2024

in the present trial, because an increased frequency of side- demonstrates that spironolactone, flutamide, and finasteride
effects is associated with higher doses (35). Flutamide, too, are all effective in the treatment of hirsutism. Moreover, after
was previously given to hirsute women in a large range of a 6-month course of therapy the clinical efficacies of these
doses, from 250 –750 mg/day (10, 11, 36, 37). We chose the drugs are similar. Further research should investigate the
lowest of these amounts to minimize any potential risk of potential for synergic effects of combined therapies with
liver toxicity. Furthermore, a dose-range study reported sim- drugs acting at different levels in androgen secretion and/or
ilar improvements in hirsutism with 250 vs. 500 mg flutamide action.
(37). On the other hand, 5 mg finasteride is the dose used in
all published trials with this drug, although similar effects on Acknowledgments
skin androgens were found with 1 mg (38). We thank Ms. Alessandra Rossi and Ms. M. Grazia Zanotti for their
We cannot exclude that by using different doses of drugs invaluable assistance. We also extend sincere thanks to Mr. Luciano
or by extending the duration of treatments it might be pos- Meneghelli for his excellent secretarial support.
sible to identify some differences in clinical efficacy among
these drugs. Nevertheless, differences of clinical relevance References
are unlikely. This opinion is supported by the observation 1. Jeffcoate W. 1993 The treatment of women with hirsutism. Clin Endocrinol
that the extent of improvements in F-G score in this study (Oxf). 39:143–150.
2. Rittmaster RS. 1995 Medical treatment of androgen-dependent hirsutism.
was roughly comparable with previously reported results at J Clin Endocrinol Metab. 80:2559 –2563.
higher doses of these antiandrogens (4, 36, 37), although in 3. Azziz R, Ochoa TM, Bradley EL, Potter HD, Boots LR. 1995 Leuprolide and
estrogen vs oral contraceptive pills for the treatment of hirsutism: a prospective
a short term (3-month) trial Lobo et al. (39) showed greater randomized study. J Clin Endocrinol Metab. 80:3406 –3411.
reduction of hair shaft diameters with 200 than 100 mg/day 4. McMullen GR, Van Herle AJ. 1993 Hirsutism and the effectiveness of spi-
spironolactone. Another potential bias is the different per- ronolactone in its management. J Endocrinol Invest. 16:925–932.
5. Neumann F. 1977 Pharmacology and potential use of cyproterone acetate.
centage of ovarian vs. nonovarian hyperandrogenism in our Horm Metab Res. 9:1–13.
treatment groups, although this difference did not reach 6. Simard J, Luthy I, Guay J, Belanger A, Labrie F. 1986 Characteristics of
statistical significance. However, considering the whole pop- interaction of the antiandrogen flutamide with the androgen receptor in var-
ious target tissues. Mol Cell Endocrinol. 44:261–270.
ulation of women receiving active drugs, we did not find any 7. Ayub M, Levell MJ. 1987 Inhibition of rat testicular 17a-hydroxylase and
difference in improvement of hirsutism between subjects 17,20-lyase activities by anti-androgens (flutamide, hydroxyflutamide,
RU23908, cyproterone acetate) in vitro. J Steroid Biochem. 28:43– 47.
with polycystic ovary syndrome and those with other forms 8. Brochu M, Belanger A, Dupont A, Cusan L, Labrie F. 1987 Effects of flutamide
of androgen excess. and aminoglutethimide on plasma 5a-reduced steroid glucuronide concen-
The tolerability of antiandrogen drugs examined in the trations in castrated patients with cancer of the prostate. J Steroid Biochem.
28:619 – 622.
study was good, with the noticeable exception of polymen- 9. Cusan L, Dupont A, Belanger A, Tremblay RR, Manhes G, Labrie F. 1990
orrhea in 50% of subjects given spironolactone. This is a well Treatment of hirsutism with the pure antiandrogen flutamide. J Am Acad
known adverse effect of the drug. However, in most of these Dermatol. 23:462– 469.
10. Couzinet B, Pholsena M, Young J, Schaison G. 1993 The impact of a pure
women the side-effect was transient, resolving within 3 anti-androgen (flutamide) on LH, FSH, androgens and clinical status in idio-
months. No patient receiving flutamide showed liver toxic- pathic hirsutism. Clin Endocrinol (Oxf). 39:157–162.
11. Moghetti P, Castello R, Negri C, et al. 1995 Flutamide in the treatment of
ity. This side-effect of flutamide showed a low incidence hirsutism: long-term clinical effects, endocrine changes, and androgen receptor
(,0.5%), but was occasionally fatal in large populations of behavior. Fertil Steril. 64:511–517.
men with advanced prostate carcinoma treated with 750- 12. Wysowski DK, Freiman JP, Tourtelot JB, Horton ML. 1993 Fatal and nonfatal
hepatotoxicity associated with flutamide. Ann Intern Med. 118:860 – 864.
1500 mg/day (12). We and others previously also observed 13. Jenkins EP, Andersson S, Imperato-McGinley J, Wilson JD, Russell DW.
occasional mild, transient hepatotoxicity in women given 1992 Genetic and pharmacological evidence for more than one human steroid
low doses of the drug (375–500 mg/day) (11, 40). As a whole, 5a-reductase. J Clin Invest. 89:293–300.
14. Rittmaster RS. 1994 Finasteride. N Engl J Med. 330:120 –125.
these observations suggest that the use of this compound for 15. Serafini P, Lobo RA. 1985 Increased 5a reductase activity in idiopathic hir-
the treatment of hirsutism should be carefully challenged in sutism. Fertil Steril. 43:74 –78.
16. Moghetti P, Castello R, Magnani CM, et al. 1994 Clinical and hormonal effects
each subject. On the other hand, Diamanti-Karamandis et al. of the 5a-reductase inhibitor finasteride in idiopathic hirsutism. J Clin Endo-
recently reported that flutamide has favorable effects on lipid crinol Metab. 79:1115–1121.
profile in women with polycystic ovarian syndrome (41); this 17. Wong IL, Morris RS, Chang L, Spahn M, Stanczyk FZ, Lobo R. 1995 A
prospective randomized trial comparing finasteride to spironolactone in the
effect is of clinical interest in subjects who frequently show treatment of hirsute women. J Clin Endocrinol Metab. 80:233–238.
several metabolic abnormalities (42). From the point of view 18. Castello R, Tosi F, Perrone F, Negri C, Muggeo M, Moghetti P. 1996 Outcome
94 MOGHETTI ET AL. JCE & M • 2000
Vol 85 • No 1

of long-term treatment with the 5a-reductase inhibitor finasteride in idiopathic of oral contraceptive vs. cyproterone acetate-estrogen regimens in the treat-
hirsutism: clinical and hormonal effects during a 1-year course of therapy and ment of hirsutism. Fertil Steril. 66:216 –219.
1-year follow-up. Fertil Steril. 66:734 –740. 31. Grigoriou O, Papadias C, Konidaris S, Antoniou G, Karakitsos P, Giannikos
19. Rittmaster RS. 1995 Gonadotropin-releasing hormone (GnRH) agonist and L. 1996 Comparison of flutamide and cyproterone acetate in the treatment of
estrogen/progestin replacement for the treatment of hirsutism: evaluating the hirsutism: a randomized controlled trial. Gynecol Endocrinol. 10:119 –123.
results [Editorial]. J Clin Endocrinol Metab. 80:3403–3405. 32. Erenus M, Yucelten D, Durmusoglu F, Gurbuz O. 1997 Comparison of fin-
20. Barth JH. 1996 How robust is the methodology for trials of therapy in hirsute asteride vs. spironolactone in the treatment of idiopathic hirsutism. Fertil Steril.
women? Clin Endocrinol (Oxf). 45:379 –380. 68:1000 –1003.
21. Ferriman D, Gallwey JD. 1961 Clinical assessment of hair growth in women. 33. Zemtsov A, Wilson L. 1997 Successful treatment of hirsutism in HAIR-AN
J Clin Endocrinol. 21:1440 –1447. syndrome using flutamide, spironolactone, and birth control therapy. Arch
22. O’Brien RC, Cooper ME, Lamurray RML, Seeman E, Thomas AK, Jerums G. Dermatol. 133:431– 433.
1991 Comparison of sequential cyproterone acetate/estrogen vs. spironolac- 34. Kelestimur F, Sahin Y. 1998 Comparison of diane 35 and diane 35 plus
tone/oral contraceptive in the treatment of hirsutism. J Endocrinol Metab. spironolactone in the treatment of hirsutism. Fertil Steril. 69:66 – 69.
72:1008 –1013. 35. Conn JJ, Jacobs HS. 1997 The clinical management of hirsutism. Eur J Endo-
23. Falsetti L, De Fusco D, Eleftheriou G, Rosina B. 1997 Treatment of hirsutism crinol. 136:339 –348.
by finasteride and flutamide in women with polycystic ovary syndrome. 36. Marcondes JAM, Minnani SL, Luthold WW, Wajchenberg BL, Samojlik E,
Gynecol Endocrinol. 11:251–257. Kirschner MA. 1992 Treatment of hirsutism in women with flutamide. Fertil
24. Hatch R, Rosenfield RL, Kim MH, Tredway D. 1981 Hirsutism: implications, Steril. 57:543–547.

Downloaded from https://academic.oup.com/jcem/article/85/1/89/2852067 by guest on 26 April 2024

etiology, and management. Am J Obstet Gynecol. 140:815– 830. 37. Muderris II, Bayram F, Sahin Y, Kelestimur F. 1997 A comparison between
25. Zawadzky JK, Dunaif A. 1992 Diagnostic criteria for polycystic ovary syn- two doses of flutamide (250 mg/d and 500 mg/d) in the treatment of hirsutism.
drome: towards a rational approach. In: Dunaif A, Givens JR, Haseltine FP, Fertil Steril. 68:644 – 647.
Merriam GR, eds. Current issues in endocrinology and metabolism: polycystic 38. Dallob AL, Sadick NS, Unger W, et al. 1994 The effect of finasteride, a
ovary syndrome. Boston: Blackwell; 377–384. 5a-reductase inhibitor, on scalp skin testosterone and dihydrotestosterone
26. Ehrmann DA, Barnes RB, Rosenfield RL. 1995 Polycystic ovary syndrome as concentrations in patients with male pattern baldness. J Clin Endocrinol Metab.
a form of functional ovarian hyperandrogenism due to dysregulation of an- 79:703–706.
drogen secretion. Endocr Rev. 16:322–369. 39. Lobo RA, Shoupe D, Serafini P, Brinton D, Horton R. 1985 The effects of two
27. Bacus JW, Grace LJ. 1987 Optical microscope system for standardized cell doses of spironolactone on serum androgens and anagen hair in hirsute
measurements and analyses. Appl Optics. 26:3280 –3293. women. Fertil Steril. 43:200 –205.
28. Cusan L, Dupont A, Gomez JL, Tremblay RR, Labrie F. 1994 Comparison of 40. Wallace C, Labor EA, Click CL. 1993 Hepatotoxicity complicating flutamide
flutamide and spironolactone in the treatment of hirsutism: a randomized treatment of hirsutism [Letter]. Ann Intern Med. 119:1150.
controlled trial. Fertil Steril. 61:281–287. 41. Diamanti-Kandarakis E, Mitrakou A, Raptis S, Tolis G, Duleba AJ. 1998 The
29. Erenus M, Gurbuz O, Durmusoglu F, Demircay Z, Pekin S. 1994 Comparison effect of a pure antiandrogen receptor blocker, flutamide, on the lipid profile
of the efficacy of spironolactone vs. flutamide in the treatment of hirsutism. in the polycystic ovary syndrome. J Clin Endocrinol Metab. 83:2699 –2705.
Fertil Steril. 61:613– 616. 42. Dunaif A. 1997 Insulin resistance and the polycystic ovary syndrome: mech-
30. Erenus M, Yucelten D, Gurbuz O, Durmusoglu F, Pekin S. 1996 Comparison anism and implications for pathogenesis. Endocr Rev. 18:774 – 800.