Current understanding of androgenetic alopecia. Part II: clinical aspects and treatment

ARTICLE

In part one we described in detail the etiopathogenesis of androgenetic alopecia (AGA). In this part we will review some clinical aspects and treatment options of AGA in men and women.

Evolution and classification

The clinical hallmark of AGA in men is a patterned decrease in scalp hair density. As a rule AGA may start to develop with the onset of puberty in both men and women, although men tend to develop AGA earlier than females. AGA is distinct from the age-dependent thinning of scalp hair (senile balding) that occurs in both sexes by the seventh or eighth decade of life. The first sign of AGA in men is usually a bitemporal recession, which can be seen in nearly all sexually mature Caucasian males, including those men not destined to develop further hair loss. However, balding of the vertex or diffuse thinning of hair may be also seen. In contrast, AGA in women most often develops at the vertex as a diffuse thinning of hair and gradual reduction of hair density (Ludwig type). This progress typically starts in their third or fourth decade of life. In rare cases AGA may be seen in teenage boys or girls. The prevalence of progressive AGA approaches 50% of Caucasian men and women beyond the age of 40; whereas in Asian, native American and African-American men the prevalence is lower and AGA is less severe. Progression may worsen after the menopause with the decline in estrogen levels or in women taking aromatase inhibitors. The onset of AGA in women is in general particularly likely at times of hormonal change, such as the discontinuation of contraceptive pills, the postpartum and the postmenopausal periods. As in men, in most women with AGA there is no evidence of abnormal androgen production [1]. Women rarely go completely bald. The end result in women is a decrease in hair thickness and density, rather than baldness. All the hairs in an affected area may be involved in the miniaturization process and with hair miniaturization the production of pigment ceases. There is still controversy over whether the total number of hair follicles decreases during AGA. However, it can be assumed that some hairs in AGA are lost definitively, but the majority of hair shafts are still present as tiny, barely visible vellus hairs.

Psychological aspects of hair loss

The loss of hair is often trivialised, but hair loss may have profound effects on a patient's well-being and quality of life. Some degree of hair loss with aging is inevitable in both men and women, but the perception of hair loss differs in both sexes: most men have so far accepted that they may become bald, whereas women will not usually accept the progression of AGA. Studies about the psychosocial impact of hair loss have shown that younger men are often concerned at the onset of hair loss, how they will look in the future, and virtually all men regard hair loss to be an unwanted, distressing experience that interferes with their body image. In women AGA seems to have far more deleterious psychosocial effects. The vast majority of women will report that the onset of hair loss engendered considerable anxious preoccupation, helplessness, and feelings of diminished attractiveness. At times these symptoms may be exaggerated and it is essential to remember that a small but important minority of patients may have dysmorphophobia regarding their appearance. Especially in women, lessening patient's anxiety about hair loss by reassuring them that they will not become totally bald is often warranted and represents an effective form of patient management.

Diagnosis and differential diagnosis

Only exceptionally are laboratory tests or scalp biopsies needed to confirm the diagnosis of AGA. Therefore the clinical assessment of AGA is largely a matter of common sense and practice. Typical differential diagnoses include trichotillomania, post-menopausal frontal fibrosing alopecia, alopecia areata of the diffuse type, and acute or chronic telogen effluvium (TE). TE is due to a disturbance of the hair cycle in which a large proportion of anagen HF enter telogen in a rather coordinated fashion. This results in an increase in daily hair loss when HF re-enter anagen. TE occurs most commonly after pregnancy, severe and long lasting febrile diseases, drugs (e.g. beta-blockers, retinoids, heparin), and "crash diets". The two key clinical features of acute TE are a clear history of rapid increase in hair shedding and a 2-3-month delay between the precipitating factor and the onset of hair loss. Sometimes systemic illnesses such as malignancies or systemic lupus erythematodes may be associated with diffuse, chronic hair loss. Therefore a thorough clinical examination and history is very important and will nearly always establish or exclude the diagnosis of AGA. In cases of doubt histopathological examination of scalp biopsies and some laboratory investigations are useful.

Special examinations

Hair pull test

After a detailed inspection of the scalp some hairs are gently pulled with the thumb and forefinger in various sections over the entire scalp. In the vast majority only 2-5 telogen hairs will be seen. The telogen hair is easy to recognize because it appears to have a dry, hard, club-shaped bulb. Any broken hair or anagen hair is abnormal.

Trichogramm (Hair pluck test)

By using a rubber-armed forceps 60-100 hairs are plucked at two scalp locations (vertex versus occiput). This somewhat painful procedure allows the establishment of the anagen/telogen ratio or the assessment by microscopic examination of the hair shaft as to whether hair shaft dysplasias are present. The hair should not be washed for at least 3-5 days before. Repeated hair pluck tests are able to quantify the efficacy of treatment because the telogen hairs may decrease in number.

Computer-assisted hair count (CAHC)

CAHC is performed by clipping the hair of a small area to be counted that is marked with a tattoo spot. Standardized photographs are then taken after defined time points of treatment, centrally processed, monitored for technical adequacy, and counted with validated computer assisted methods. With this technique the total number of hairs before and after treatment can be calculated.

Computer-assisted anagen hair count (phototrichogramm)

This technique is a slight modification of the method mentioned above. The phototrichogram is a non-pluck-ing, non-invasive method. For this technique the hairs are clipped and the area is marked. After 2-3 days a photograph is taken again and this will show the number of actively growing (anagen) hairs in relation to the not growing (telogen) hairs. This procedure can be repeated after several months to monitor the effect of treatment. However, hair diameters from the phototrichogram are believed to be too unreliable to be of any practical use and analysis of the individual hair data revealed that light hair was much more difficult to evaluate than dark hair.

Laboratory investigations

Although not routinely recommendable, some simple investigations may be useful for the assessment of hair loss in general. For further reading several papers reviewed the rationale of laboratory testing [2] and guidelines of care for AGA have been published [3].

The literature on the subject of AGA and laboratory testing is vast and confusing. In fact, the development of AGA is almost always associated with normal systemic androgen levels in both men and women. Therefore, the mere presence of AGA with no additional sympton such as hirsitusm or acne (SAHA-syndrome), does not necessitate extensive laboratory examinations. Routine blood tests for elevated androgens are time consuming, expensive and worthless. If the condition readily responds to conventional therapy, the discovery of a mild hyperandrogenism would probably not change the therapeutic approach. However, in rare cases AGA might be the single cutaneous symptom of an underlying hyperandrogenism especially in early onset or severe AGA in women with no other signs of hyperandrogenemia such as increased body hair, deepened voice or cycle irregularities. In selected cases, analysis of FSH/LH-ratio, SHBG, total and free T, DHT, androstenedione, DHEA-S and prolactin might be useful.

Only in clinically atypical cases is the histopathological examination of a scalp biopsy useful and it will reveal miniaturized terminal hair follicles with associated increased numbers of vellus-like HF, perifollicular infiltrations, "streamers" and sometimes fibrosis. Diagnostic accuracy is not easily attainable in the histological evaluation of AGA and there is still some debate whether one or several punch biopsies are needed but classical vertical and horizontal sections are preferable.

Associated conditions

The association of baldness with an increased susceptibility or resistance to certain diseases has been claimed repeatedly, but the evidence is often unsatisfactory. AGA is defined as an androgen-dependent phenomenon and might be associated with acquired progressive kinking and twisting of the hair. An association between rapidly progressive, severe AGA of men and the risk of myocardial infarction has been reported by Lesko [4]. This finding, however, needs further evaluation, because it was not exluded that men with severe coronary heart disease may use various other medications such as a beta-blocker that may in itself induce hair loss, thus mimicking the progession of AGA. No association between AGA and prostatic cancer was found [5], but one group reported a correlation between severe AGA and the development of benign prostatic hyperplasia [6]. Again, this finding needs further investigation before this relationship can be taken as certain.

Treatment

The treatment of AGA is obscured by myths. Many products or procedures are advertized for the treatment of AGA such as vitamins, trace elements, exotic herbs, amino acids, "soft laser", scalp massage etc. Most of these techniques or substances have never been verified in sound clinical trials. Because of the psychosocial impact of hair loss, it is important to explain the evaluation to the patient and to inform them as to what they may expect in terms of continuing hair loss, and that response to any therapy may be slow and may include hair regrowth or only retardation of further thinning. A decision to have no treatment may be an appropriate option for certain patients. Nevertheless, patients with hair loss will often seek inappropriate or unproven therapies that are available in nonmedical settings and that may be rather costly. This review will not list substances with doubtful efficacy but focus on proven medical treatment options. Surgical approaches or the detailed description of special wigs are beyond the scope of this review.

The aim of AGA treatment is to reverse or to stabilize the process of HF miniaturization. This can be accomplished either by modifiers of the androgen signal transduction cascade or by other hair growth promoters. The first strategy tries to antagonize the androgen-induced processes within the hair follicles and is therefore a causative approach, whereas the latter strategy ignores the pathogenetic events leading to AGA. It is important to realize that AGA in men and women is treated differently.

Drugs interfering with the androgen-metabolism or the androgen receptor

Androgen receptor blockers (antiandrogens): Antiandrogens are in contrast to pure enzyme inhibitor substances that prevent androgens from expressing their activity at target sites. They compete with DHT or T for the specific AR thus inhibiting androgen-mediated processes. AR blockers are used in the treatment of various diseases such as cancer of the prostate. Because AGA is an androgen-mediated process, AR blockers are a rational approach for treatment. For obvious reasons, however, this type of treatment is contraindicated in men. In pregnant women antiandrogens may give rise to ambigous genitalia of male fetuses, therefore AR blockers are used in combination with estrogens to ensure contraception.

Cyproterone acetate (CPA) is widely used today throughout the world with the exception of the USA, but was mainly studied for the treatment of hirsutism. Some of these studies showed that CPA in combination with ethinyl estradiol may prevent further progression or reverse to some extent AGA in women [7-9]. However, rigorous clinical trials were not performed. In combination with 0.035 mg ethinyl estradiol CPA is typically used at 2 mg daily. In cases where androgenetic hair loss does not stop using this regimen, CPA can be added up to 50 mg each day for the first 15 days of the cycle.

Chlormadinone acetate (CMA) and dienogest are weaker antiandrogens used in combination with ethinyl estradiol for contraception. Both drugs may have some effect on AGA in women, but studies to prove this have not been performed.

Side effects of CPA or CMA include weight gain, breast tenderness, depression, loss of libido and nausea. Contraindications such as obesity, coagulopathies, chronic venous insufficiency or smoking should be considered.

Spironolactone is an aldosterone antagonist and a weak antiandrogen which acts both by blocking the AR and by direct inhibition of androgen production in the adrenals through the inhibition of 17alpha-hydroxylase and desmolase. As well as CPA, spironolactone is mainly used to treat hirsutism or acne in women and only small open trails have shown some clinical effects in AGA [10-12] showing that the further progression of AGA might be inhibited, but with no hair regrowth [13]. Anecdotal reports have claimed that cimetidine, a weak antiandrogen, may have a beneficial effect on AGA in women [14], but no strict clinical studies are available to validate those findings. Attempts to use antiandrogens topically were disappointing. There is weak experimental evidence that topical antiandrogens may affect the sebaceous gland but there is no real evidence of any effect on AGA.

Steroidogenic enzyme inhibitors: type 2 5alpha-reductase inhibitors

The first agent used for this purpose was finasteride, which is an orally active type 2 5alphaR inhibitor. Clinical studies conducted to treat benign prostatic hyperplasia established the excellent safety profile of this drug. Subsequently oral F has been shown to reverse balding in the stump-tailed macaque.

In theory F could be applied topically, but it should be noted that all lipophilic drugs will be absorbed. For example, it has been shown that unilateral application of progesterone ­ a natural competitive 5alpha-R inhibitor ­ resulted in inhibition of 5alpha-R activity on the untreated side as well [15]. The percutaneous absorption is hard to control because it differs from patient to patient thus making it difficult to monitor drug-related effects and side effects. Furthermore, type 25alphaR is expressed within the dermal papilla (DP) which is situated at the base of the HF in the subcutaneous fat, remote from the epidermal surface and enclosed by the connective tissue sheath and the matrix cells. Topical application of therapeutic substances such as antiandrogens might therefore reach their target cells only insufficiently. On the other hand, the DP is amply supplied with blood capillaries and is accessible to a systemic effect of oral drugs. Provided that undesirable side effects can be excluded, systemic inhibition of steroidogenic enzymes is an effective approach for the treatment of AGA.

Dose-finding studies were conducted to compare the efficacy of 5 mg versus 1 mg, 0.2 mg and 0.01 mg F to treat AGA in men, indicating that 1 mg F daily can be regarded as a safe and efficient approach for treatment. These preliminary data led to large-scale phase III placebo-controlled, cross-over designed clinical trials comprising 1,879 men showing that a two-year trial of oral F 1 mg/day promotes new hair growth in 66% (placebo 7%) and prevents further hair loss in 83% of patients [16]. There is no evidence that finasteride is efficient to treat AGA in postmenopausal women. Currently finasteride is being evaluated for the treatment of hirsutism [17]. Although it has been shown that finasteride is a well tolerated choice of treatment in this group, it should be borne in mind that the use of finasteride is contraindicated in women with child-bearing potential. In male fetuses, type 2 5alphaR-derived DHT plays an important role in normal prostatic and external genital differentiation. An inhibition of this enzyme in pregnant women taking finasteride may induce abnormalities in male embryos such as hypospadias, preputial adhesions to the glans, a small underdeveloped scrotum, a small penis, and a prominent midline raphe in male fetuses. Therefore, women should not come in contact with broken tablets. On the other hand it has been calculated that in men taking 1 mg/day the concentration of F in the ejaculate is so minimal that there is no risk for pregnant women and their embryos [18] and condoms need not be used during treatment.

1 mg/day F is well tolerated. The pharmacodynamic and -kinetic properties are reviewed by McClellan et al. [19]. F is a 5alpha-R inhibitor which has in vivo a much higher affinity for the type 2 than the type 1 isoenzyme. In vitro, F and T appear to compete for the same binding site of 5alpha-R, but F has no direct effect on the binding of T or DHT to the AR. Even a single dose of 12.5 mg F will lower systemic DHT levels for 56% by 72 hours. Hence it is understandable that F 1 mg/day lowers DHT concentrations in the prostate and in the scalp. The serum DHT concentrations are lowered by approximately 68% with an associated slight increase in serum T and E2 levels, which is still within the normal range and is therefore clinically irrelevant. F does not influence the hypothalamic-pituitary-testicular axis, nor has F a significant effect on serum levels of prolactin, SHBG, aldosterone or cortisol.

F in a dosage of 1 mg/day for men was not associated with any changes of sperm motility, morphology or total sperm number. As a result of a diminished function of seminal glands the total volume of the ejaculate was slighly reduced but still within the normal range. Serum lipid levels or bone metabolism are not affected by F. Some authors suggest that in older men PSA levels should be monitored before taking F. F has no effect on body hair. 7.7% of F versus 7% of placebo recipients reported mild to moderate treatment-related adverse events (1.4% and 1.6% withdrew). This difference was not statistically significant. The events reported more frequently on F as compared to placebo treatment were decreased libido (1.8% vs. 1.3%), ejaculation disturbances (1.2% vs. 0.7%) and erectile dysfunction (1.3% vs. 0.7%). It should be noted that these effects resolved in most men still on the drug and in all men who discontinued the treatment because of the adverse events.

Topical F inhibits the activity of the sebaceous gland in fuzzy rats and mice or that of the hamster flank organ and the prostate. For this reason attempts have been made to assess the efficacy of topical F in AGA. At present the results in humans are disappointing. It has been shown that topically applied F reduces DHT levels but will not prevent AGA in men.

Steroidogenic enzyme inhibitors: type 1 5alpha-reductase inhibitors

A type 1 5alpha-R inhibitor was used in animals for the treatment of AGA. These studies showed that MK-386 is able to reduce systemic DHT levels by 30-50% but without affecting hair growth [20]. It is believed that the type 1 5alpha-R is expressed in the sebaceous gland and in those HF affected by hirsutism. This raised the question of the possible use of specific 5alpha-R1 inhibitors in the treatment of hirsutism or acne.

Steroidogenic enzyme inhibitors: other 5alpha-Reductase inhibitors

Progesterone is a 5alpha-R inhibitor but topical application is ineffective in men because of rapid metabolization. Various synthetic progesterone derivatives (11a-hydroxy-progesterone; 11-OHP) have been studied in AGA. Sebum production has been reported to be inhibited by 11-OHP [21] but others were not able to confirm these results [22]. The effects of 11-OHP on AGA remain elusive. Small studies comprising less than 20 patients showed that topical 11OHP inhibits 5alpha-R activity in isolated HF and may work in acne vulgaris and may perhaps reduce further hair loss in AGA. Appropriate clinical studies have never been performed.

It must be noted that today many substances are said to possess an inherent 5alpha-R inhibitory capacity. However, most of these drugs were not tested on human cells and usually no distinction is made between the two 5alpha-R isoenzymes, which is of crucial importance because of the pivotal role of the type 2 5alpha-R in AGA. Furthermore, Voigt et al. [23] convincingly demonstrated that the essential requirement for an effective inhibitor of 5alpha-R is a delta⁴- 3-keto configuration. Most substances do not fit into this category, and therefore it is not surprising that they have only minimal effects on 5alpha-R activity. In addition, azasteroids are generally believed to be the most potent inhibitors of 5alpha-R [24, 25]. Several plant extracts are advertised for the treatment of BPH because of their capacity to inhibit 5alpha-R in vitro. In vivo studies were unable to prove such an action on 5alpha-R [26]. Therefore, commercial plant extracts advertised for the treatment of AGA because of their 5alpha-R inhibition should be appropriately analyzed to see whether they exert any effect in vivo.

Other steroidogenic enzyme inhibitors

Although theoretically attractive, selective and isoenzyme-specific inhibitors for type 3 17beta-HSD, type 2 3beta-HSD or steroid sulfatase are not at present available for the treatment of AGA.

Topical estrogens

Solutions containing either estradiol benzoate, estradiol valerate, 17beta- or 17a-estradiol (17beta-E, 17a-E) are commercially available in Europe. In Germany, some of these solutions contain corticosteroids, but the rationale of this approach is unclear. It should be noted that except for 17a-E most estrogens cannot be given to men because of the risk of developing gynecomastia. In Europe the synthetic 17a-estradiol is used and is believed not to exert estrogen receptor-mediated effects. Therefore, 17a-E can be given to men. One double-blind placebo controlled trial studied the effect of topical 0.025% 17a-E for a period of 6 months in 51 men and women with AGA [27]. In this study a significant reduction of telogen hairs and an increase in anagen hairs was seen.

The molecular pathways involved in estrogen-mediated induction of hair growth in AGA are unknown. Estrogens have been shown to prolong anagen in guinea pigs but in contrast topical application of 17beta-E to the clipped dorsal skin of mice arrested hair follicles in telogen and produced a profound and prolonged inhibition of hair growth while treatment with the biologically inactive stereoisomer, 17a-E had no effect [28, 29]. In contrast, other scientists were unable to confirm these findings [30] and at present it is not even clear whether topically applied estrogens are active locally or whether they exert their effects through percutaneous absorption. In this way the transdermal pathways of estrogens have been investigated and it has been reported that topically applied estrogens mainly remain in the stratum corneum and the sebaceous glands. However, estrogens may penetrate to the dermis via the epidermis and the hair canal and hair sheaths. Hence it is conceivable that as highly lipophilic substances, E might be absorbed via the skin. Whether topically applied E will reach the DP is unknown.

Estrogens are believed to increase the concentration of SHBG thus reducing the serum levels of free T and DHT but a causal relationship has not been shown. It is noteworthy that E2 is able to detach DHT from its binding sites on SHBG, thus making it even more bioavailable [31].

Some authors found that E might weakly inhibit 5alpha-R activity. Very high doses of E inhibit the T metabolism in rats [32, 33]. Groom [34] reported a direct suppression of 5alpha-R activity in canine prostate explants. E2 at very high concentrations (20 muM) has been found to be a non-competitive inhibitor of 5alpha-R in granulosa cells from rat ovaries [35]. Other groups did not confirm these data for genital skin fibroblasts [36] and an indirect effect such as metabolization of E2 to estrone via 17beta-HSD has never been excluded. Moreover, in nearly all experiments showing that E2 is able to inhibit 5alpha-R in vitro, an inhibition was only present at very high doses of estrogen (e.g. 20muM) that are not applicable in vivo.

One study indicated that 17a-E at very high concentrations (30 mug/100 mg liver tissues) inhibits 5alpha-R in liver tissues from female rats [37]. The type of 5alpha-R which was inhibited was not determined. The same authors failed to detect a similar effect when using male rat liver tissues. This was attributed to the much smaller total turnover of testosterone in male liver cells. By contrast, in another study the administration of low (physiologic) doses of estradiol to rats showed an increase in 5alpha-R activity in prostatic cells [38]. In the latter study it was concluded that in dosages used clinically and at the hormone concentrations found in male, estrogens have no direct inhibitory effect on the 5alpha-R in the prostate [38, 39] or skin slices [23]. Whether estrogens inhibit type 2 5alpha-R in human hair follicles has not been investigated so far.

In sum, the use of estrogens for the treatment of AGA can be regarded as a safe treatment option which is able to stop or reverse AGA, but the exact success rates and the underlying mechanisms are so far unknown.

Broad hair growth promoters

Several drugs such as benoxaprofen, corticosteroids, diazoxide, diphenylhydantoin have the intrinsic capacity to promote hair growth and this phenomenon is regarded as a side effect of systemic treatment. Some of these drugs are of theroretical interest, because the analysis of the underlying hair growth mechanism might help to find new hair growth promoters. Some immunophilin ligands (Ipl) are both potent immunosuppressants and effective modulators of hair growth. For example cyclosporin A (CSA) or FK 506 induce hair growth and inhibit hair follicle regression. Hence, the development of CSA analoga without the risk of inducing hypertrichosis would be of clinical importance.

Minoxidil

Minoxidil (M) is a drug commonly leading to hypertrichosis in approx. 70% of patients treated systemically. This is apparent after a few weeks of systemic treatment and the hair will fall out 2 months after discontinuation of the drug. M is a pyrimidine derivative (2,4-diamono-6-piperidinopyrimidine-3-oxyde) initially developed as a potent antihypertensive agent [40-42]. In addition to being a direct-acting vasodilatator, it was unexpectantly found to stimulate hair growth in vivo, and this side effect led to its clinical use in AGA. It is so far unclear why M induces hair growth. M stimulates a time-dependent increase in 3[H]-thymidine and 35[S]-cysteine incorporation in mouse vibrissa follicles and it has been suggested that this effect is mediated via the K⁺ channels and that a sulphatated metabolite of M exerts this effect via M-sulfotransferase which is present in HF. A number of large multicenter, double-blind trials have been conducted that compared topical 2% or 3% M versus placebo [43-46]. The use of M has been shown to induce a conversion of vellus to terminal hairs in approx. 30% of patients. These studies, however, were not as conclusive as hoped. Many studies were open-labeled rather than double-blind and enrolled a rather small number of patients. Many studies switched to M treatment only after 4 months of placebo treatment thus making it difficult to interpret the results. As a consequence the efficacy of M is still a matter of debate [47-49]. Many studies have shown that non-vellus hair regrew at the margins but complete covering of the bald areas was seen in less than 10% of responders [50]. Others found in 18% of treated patients good results which were present in only 6% after additional 12 months [51]. In sum, several studies have shown that approx. 15% will experience some sort of hair regrowth, while 50% have their hair loss delayed and approx. 35% will continue to loose hair [52].

Two decades ago effective drugs for the treatment of androgenetic alopecia (AGA) were nonexistent. Today a spectrum of different approaches can be offered to patients suffering from AGA. The approval of finasteride was a breakthrough for the treatment of AGA in men. Whether other 5-reductase inhibitors will be available for the treatment of AGA or other androgen-dependent diseases such as acne or hirsutism remains to be shown in the future.

With increasing knowledge of the follicular repertoire of isoenzymes of the androgen metabolism, new non-toxic and selective inhibitors may turn out to be fruitful therapeutic modalities. The same is true for selective androgen receptor blockers such as RU 58841, or those drugs interfering with the dihydrotestosterone-dependent signal transduction cascade in HF. Furthermore, by exploring the intrafollicular pathways involved in cyclosporin A or minoxidil-induced hypertrichosis new classes of drugs could be designed. We may even remember a statement by Van Scott and Ekel that was made more than 40 years ago: "If the assumption is made that the size of a hair depends on the size of its papilla, a search for factors controlling the size of the papilla would seem to be appropriate in further investigations of male type baldness" [53]. It will be fascinating to see what will be launched for the treatment of AGA within the next years. We will soon know the underlying genes, and this may give us the opportunity of a gene therapy by liposome entrapped genes targeting the hair follicle.

Article accepted on 2/3/00

REFERENCES

1. Hamilton JB. Male hormone stimulation as a prerequisite and an incitement in common baldness. Am J Anat 1942; 71: 451-3.

2. Sperling LC, Heimer WLd. Androgen biology as a basis for the diagnosis and treatment of androgenic disorders in women. II. J Am Acad Dermatol 1993; 28: 901-16.

3. Drake LA, Dinehart SM, Farmer ER, Goltz RW, Graham GF, Hordinsky MK, Lewis CW, Pariser DM, Webster SB, Whitaker DC, Butler B, Lowery BJ, Price VH, Baden H, DeVillez RL, Olsen E, Shupack JL. Guidelines of care for androgenetic alopecia. American Academy of Dermatology. J Am Acad Dermatol 1996; 35: 465-9.

4. Lesko SM, Rosenberg L, Shapiro S. A case-control study of baldness in relation to myocardial infarction in men [published erratum appears in JAMA 1993 May 19; 269: 2508] [see comments]. JAMA 1993; 269: 998-1003.

5. Demark-Wahnefried W, Lesko SM,
Conaway MR, Robertson CN, Clark RV, Lobaugh B, Mathias BJ, Strigo TS, Paulson DF. Serum androgens: associations with prostate cancer risk and hair patterning. J Androl 1997; 18: 495-500.

6. Oh BR, Kim SJ, Moon JD, Kim HN, Kwon DD, Won YH, Ryu SB, Park YI. Association of benign prostatic hyperplasia with male pattern baldness. Urology 1998; 51: 744-8.

7. Leo-Rossberg I, Laur S, Zielske F, Hammerstein J. Reversed sequential therapy of hirsutism using cyproterone acetate. I. Further clinical observations. Acta Endocrinol Suppl (Copenh) 1971; 152: 14.

8. Eschiletti MA, Goldim JR, Bakos L, Gross JL. Effect of the combination of cyproterone acetate and conjugated estrogens on hirsutism, alopecia and acne. AMB Rev Assoc Med Bras 1986; 32: 61-4.

9. Hoffmann E, Meiers HG, Hubbes A. The effects of oral contraceptives on male-pattern alopecia, seborrhoea, acne vulgaris and hirsutism. Anamnestic study on 325 women. Dtsch Med Wochenschr 1974; 99: 2151-4, 7.

10. Burke BM, Cunliffe WJ. Oral spironolactone therapy for female patients with acne, hirsutism or androgenic alopecia. Br J Dermatol 1985; 112: 124-5.

11. Dawber RP. Special symposium on dermatological therapy : V. Diseases of the appendages. Alopecia and hirsutism. Clin Exp Dermatol 1982; 7 : 177-82.

12. Adamopoulos DA, Karamertzanis M, Nicopoulou S, Gregoriou A. Beneficial effect of spironolactone on androgenic alopecia [letter]. Clin Endocrinol (Oxf) 1997; 47: 759-60.

13. Rubina VV, Korchevaia TA, Kravchenko L. Experience in treating androgen-dependent diseases (androgenetic alopecia, acne agminata, hirsutism) with the preparation Diane. Vestn Dermatol Venerol 1986 ; 21-5.

14. Aram H. Treatment of female androgenetic alopecia with cimetidine. Int J Dermatol 1987; 26 : 128-30.

15. Vermorken AJ, Goos CM, Roelofs HM. A method for the evaluation of the local antiandrogenic action of 5 alpha-reductase inhibitors on human skin. Br J Dermatol 1980; 102: 695-701.

16. Kaufman KD, Olsen EA, Whiting D, Savin R, DeVillez R, Bergfeld W, Price VH, Van Neste D, Roberts JL, Hordinsky M, Shapiro J, Binkowitz B, Gormley GJ. Finasteride in the treatment of men with androgenetic alopecia. Finasteride Male Pattern Hair Loss Study Group. J Am Acad Dermatol 1998; 39: 578-89.

17. Castello R, Tosi F, Perrone F, Negri C, Muggeo M, Moghetti P. Outcome of long-term treatment with the 5 alpha-reductase inhibitor finasteride in idiopathic hirsutism: clinical and hormonal effects during a 1-year course of therapy and 1-year follow-up. Fertil Steril 1996; 66: 734-40.

18. Rittmaster RS. Finasteride. N Engl J Med 1994; 330: 120-5.

19. McClellan KJ, Markham A. Finasteride: a review of its use in male pattern hair loss. Drugs 1999; 57: 111-26.

20. Rhodes L, Primka R, Berman C, Gaito G, Audette-Arruda J, Pikounis B, Matuszewska B, Harper J. Effects of 1 year treatment with oral MK386, an inhibitor of type 1 5alpha-Reductase, in the stumptailed macaques (Macaca arctoides). J Invest Dermatol 1995; 104: 658 (Abstr.).

21. Tamm J, Seckelmann M, Volkwein U, Ludwig E. The effect of the antiandrogen II alpha-hydroxyprogesterone on sebum production and cholesterol concentration of sebum. Br J Dermatol 1982; 107: 63-70.

22. Matias JR, Orentreich N, Malloy V, De Feo CPd, Matias L. The lack of effect of 11 alpha-hydroxyprogesterone on the flank-organ and ear sebaceous glands of adult male Syrian golden hamsters. Arch Dermatol Res 1984; 276: 346-8.

23. Voigt W, Fernandez EP, Hsia SL. Transformation of testosterone into 17 beta-hydroxy-5 alpha-androstan-3- one by microsomal preparations of human skin. J Biol Chem 1970; 245: 5594-9.

24. Rasmusson GH, Reynolds GF, Utne T, Jobson RB, Primka RL, Berman C, Brooks JR. Azasteroids as inhibitors of rat prostatic 5 alpha-reductase. J Med Chem 1984; 27: 1690-701.

25. Liang T, Heiss CE, Cheung AH, Reynolds GF, Rasmusson GH. 4-Azasteroidal 5 alpha-reductase inhibitors without affinity for the androgen receptor. J Biol Chem 1984; 259: 734-9.

26. Rhodes L, Primka RL, Berman C, Vergult G, Gabriel M, Pierre-Malice M, Gibelin B. Comparison of finasteride (Proscar), a 5 alpha reductase inhibitor, and various commercial plant extracts in in vitro and in vivo 5 alpha reductase inhibition. Prostate 1993; 22: 43-51.

27. Orfanos CE, Vogels L. Local therapy of androgenetic alopecia with 17 alpha-estradiol. A controlled, randomized double-blind study (author's transl). Dermatologica 1980; 161: 124-32.

28. Oh HS, Smart RC. An estrogen receptor pathway regulates the telogen-anagen hair follicle transition and influences epidermal cell proliferation. Proc Natl Acad Sci USA 1996; 93: 12525-30.

29. Smart RC, Oh HS. On the effect of estrogen receptor agonists and antagonists on the mouse hair follicle cycle. J Invest Dermatol 1998; 111: 175.

30. Stenn KS, Paus R, Filippi M. Failure of topical estrogen receptor agonists and antagonists to alter murine hair follicle cycling. J Invest Dermatol 1998; 110: 95.

31. Dechaud H, Goujon R, Claustrat F, Boucherat M, Pugeat M. In vitro influence of plasma steroid-binding proteins on androgen metabolism in human leukocytes. Steroids 1995; 60: 226-33.

32. Farnsworth WE. A direct effect of estrogens on prostatic metabolism of testosterone. Invest Urol 1969; 6: 423-7.

33. Jenkins JS, McCaffery VM. Effect of oestradiol-17beta and progesterone on the metabolism of testosterone by human prostatic tissue. J Endocrinol 1974; 63: 517-26.

34. Groom M, Harper ME, Fahmy AR, Pierrepoint CG, Griffiths K. The effect of oestrogen on the prostatic metabolism of testosterone in tissue culture. Biochem J 1971; 122: 125-6.

35. Payne DW, Packman JN, Adashi EY. Follicle-stimulating hormone inhibits granulosa cell 5 alpha-reductase activity. Possible role of 5 alpha-reductase as a steroidogenic pubertal switch. J Biol Chem 1992; 267: 13348-55.

36. Beckmann MW, Wieacker P, Dereser MM, Flecken U, Breckwoldt M. Influence of steroid hormones on 5 alpha-reductase activity in female and male genital skin fibroblasts in culture. Acta Endocrinol (Copenh) 1993; 128: 161-7.

37. Schriefers H, Wright MC, Rozman T,
Hevert F. Hemmung des Testosteron-Stoffwechsels durch 17 alpha-Estradiol in Rattenleberschnitten. Arzneimittelforschung 1991; 41: 1186-9.

38. Makela S, Santti R, Martikainen P, Nienstedt W, Paranko J. The influence of steroidal and nonsteroidal estrogens on the 5 alpha- reduction of testosterone by the ventral prostate of the rat. J Steroid Biochem 1990; 35: 249-56.

39. Usui T, Ishibe T, Nakamura K, Mizutani M, Yasukawa A. Effect of diethylstilbestrol diphosphate on activity of 5 alpha- reductase in human prostate. Arch Androl 1987; 19: 229-35.

40. Dargie HJ, Dollery CT, Daniel J. Minoxidil in resistant hypertension. Lancet 1977; 2: 515-8.

41. Devine BL, Fife R, Trust PM. Minoxidil for severe hypertension after failure of other hypotensive drugs. Br Med J 1977; 2: 667-9.

42. Watkins J, Dargie HJ, Bune A, Dollery CT. Reduction of beta-blocking drugs in hypertensive patients treated with minoxidil. Br Med J 1979; 1: 1400.

43. Connors TJ, Cooke DE, De Launey WE, Downie M, Knudsen RG, Shumack S, Eggleston AS. Australian trial of topical minoxidil and placebo in early male pattern baldness. Australas J Dermatol 1990; 31: 17-25.

44. de Groot AC, Nater JP, Herxheimer A. Minoxidil for male-pattern baldness. Lancet 1987; 2: 563.

45. de Groot AC, Nater JP, Herxheimer A. Minoxidil: hope for the bald? Lancet 1987; 1: 1019-22.

46. Koperski JA, Orenberg EK, Wilkinson DI. Topical minoxidil therapy for androgenetic alopecia. A 30-month study. Arch Dermatol 1987; 123: 1483-7.

47. de Groot AC, Nater JP. Minoxidil: no hope for bald people. Ned Tijdschr Geneeskd 1989; 133: 898-9.

48. Feldman SR. Is minoxidil (Rogaine) clinically effective? Arch Dermatol 1989; 125: 572.

49. Morgan JP, Zappa M. Topical minoxidil for hair loss. Ration Drug Ther 1987; 21: 1-4.

50. Simpson NB, Barth JH. Hair patterns: hirsuties and androgenetic alopecia, in Dawber R (ed): diseases of the hair and scalp (ed third). Oxford, Blackwell Science, 1997, pp. 67-122.

51. Tosti A. Topical minoxidil useful in 18% of patients with androgenetic alopecia: a study of 430 cases. Dermatologica 1986; 173: 136-8.

52. Sinclair R. Male pattern androgenetic alopecia. Bmj 1998; 317: 865-9.

53. Van Scott EJ, Ekel TM. Goemetric relationships between the matrix of the hair bulb and its dermal papilla in normal and alopecic scalp. J Invest Dermatol 1958; 31: 281-7.