Androgenetic Alopecia in Women Vera H. Price University of California San Francisco, Department of Dermatology, San Francisco, California, USA
Androgenetic alopecia (AGA), also known in women as female pattern hair loss, is caused by androgens in genetically susceptible women and men. The thinning begins between ages 12 and 40 years, the inheritance pattern is polygenic, and the incidence is the same as in men. In susceptible hair follicles, dihydrotestosterone binds to the androgen receptor, and the hormone-receptor complex activates the genes responsible for the gradual transformation of large terminal follicles to miniaturized follicles. Both young women and young men with AGA have higher levels of 5a reductase and androgen receptor in frontal hair follicles compared to occipital follicles. At the same time, young women have much higher levels of cytochrome p-450 aromatase in frontal follicles than men who have minimal aromatase, and women have even higher aromatase levels
in occipital follicles. The diagnosis of AGA in women is supported by early age of onset, the pattern of increased thinning over the frontal/parietal scalp with greater density over the occipital scalp, retention of the frontal hairline, and the presence of miniaturized hairs. Most women with AGA have normal menses and pregnancies. Extensive hormonal testing is usually not needed unless symptoms and signs of androgen excess are present such as hirsutism, severe unresponsive cystic acne, virilization, or galactorrhea. Topical minoxidil solution is the only drug available for promoting hair growth in women with AGA. E⁄cacy has been shown in doubleblind studies using hair counts and hair weight. Key words: female pattern hair loss/miniaturization/5a reductase/ aromatase/topical monoxidil. JID Symposium Proceedings 8:24 ^27, 2003
scalp hair follicles in AGA. In genetically susceptible hair follicles, DHT binds to the androgen receptor, and the hormone-receptor complex then activates the genes responsible for the gradual transformation of large, terminal follicles to small, miniaturized follicles (Price, 1975; Uno et al, 1985; Messenger, 1993; Kaufman, 1996). Over successive hair cycles in AGA, the duration of anagen shortens and matrix size decreases, resulting in smaller follicles that produce shorter, ¢ner, miniaturized hairs that cover the scalp less and less well. These miniaturized hairs of various lengths and diameters are the hallmark of AGA (Frieden and Price, 1986; Whiting, 1993; Olsen, 1994). The number of follicles per unit area, however, remains the same (Whiting, 1993). In women with AGA, the extent of hair loss is generally less than in men. It has been assumed that the hormonal basis for AGA is similar in women as in men although earlier studies did not include female subjects (Price, 1975; Frieden and Price, 1986). In order to compare and clarify the underlying hormonal basis, a study was conducted in 12 young women (ages 14^33) and 12 young men (ages 18^30) with AGA (Sawaya and Price, 1997). Androgen receptor, type I and type II 5a reductase, and cytochrome p- 450 aromatase, were measured in hair follicles from scalp biopsies of these young subjects. Both young women and young men had higher levels of type I and type II 5a-reductase and androgen receptors in frontal hair follicles compared to occipital hair follicles; however, the levels in women were approximately half the levels in men (Sawaya and Price, 1997). At the same time, young women had much higher levels of cytochrome p- 450 aromatase in frontal follicles than men who had minimal aromatase, and women had even higher aromatase levels in occipital follicles. The di¡erences in aromatase, which is capable of converting testosterone to estradiol, are particularly notable. The ¢ndings of this study suggest that the milder expression of AGA in women may in part be the result of lower levels of 5a reductase and an-
ndrogenetic alopecia (AGA), also known in women as female pattern hair loss, is a common and distressing cause of hair loss in women and men. In spite of its prevalence, many clinicians and clinical investigators have di⁄culty making the diagnosis in women. The hair thinning begins between the ages of 12 and 40 years in both sexes (Hamilton, 1951; Trancik et al, 2001;1) and approximately half the population expresses this trait to some degree before the age of 50 years (Venning and Dawber, 1988; Olsen, 1994). Inheritance pattern of AGA is polygenic (complex inheritance) and is inherited from either or both sides of the family (Kuster and Happle, 1984; Sreekumar et al, 1999). PATHOPHYSIOLOGY
AGA is induced by androgens in genetically susceptible women and men. Hair follicles of women and men with AGA have increased 5a-reductase activity and increased levels of dihydrotestosterone (DHT) (Schweikert and Wilson, 1974; Dallob et al, 1994; Kaufman, 1996; Sawaya and Price, 1997). DHT, which is formed by the peripheral conversion of testosterone by 5a-reductase, is thought to be responsible for the characteristic miniaturization of Accepted for publication February 1, 2003 Reprint requests to: Vera H. Price, Department of Dermatology, University of California San Francisco, 350 Parnassus Avenue, Suite 404, San Francisco, CA 94117 USA; E-mail: [email protected]
1 Trancik RJ, Spindler JR, Cuddihy RV, Doren P: Minoxidil topical solution in the treatment of androgenetic alopecia in patients under 18 years of age. Poster presentation at the 3rd Intercontinental Meeting of Hair Research Societies; June 13^15, 2001, Tokyo, Japan. Abbreviations: AGA, Androgenetic alopecia; CTE, Chronic telogen e¥uvium.
0022-202X/03/$15.00 . Copyright r 2003 by The Society for Investigative Dermatology, Inc. 24
VOL. 8, NO. 1 JUNE 2003
ANDROGENETIC ALOPECIA IN WOMEN
drogen receptors in frontal follicles of women compared to levels in men; additionally, higher levels of aromatase in women may result in increased local formation of estradiol from testosterone, and less formation of 5a-reduced products such as DHT. Although hair thinning in women with AGA may be di¡use, nevertheless there is a pattern in women as there is in men, with the frontal and parietal scalp generally showing the greatest thinning and the occipital scalp showing the greatest hair density. From a clinical perspective, hair follicles of the occipital scalp behave di¡erently from frontal and parietal scalp follicles: in AGA, occipital follicles are typically spared by the hormonal in£uences; and in alopecia areata, occipital follicles a¡ected by the ophiasis pattern are typically more resistant to regrowth. These di¡erent behaviors of occipital and frontal/parietal follicles may result from the embryological derivation of the dermis in these two regions. It is known from avian embryology that the dermis of the frontal/parietal scalp is of neural crest origin, whereas the dermis of the occipital scalp is of mesodermal origin (Ziller, 1996). Regulatory genes, growth factors, and cell^cell interactions may in£uence hair follicles di¡erently in the two regions based on their dermal origin, and this may explain the patterning that is frequently seen in scalp hair loss. CLINICAL FEATURES OF AGA IN WOMEN
Women ¢rst notice hair thinning over the frontal area, and gradually the scalp becomes more visible (Fig 1). Over time the thinning, although it is often di¡use and may involve most of the scalp, nevertheless is usually patterned with most marked thinning over the frontal and parietal scalp, and with greater density over the occipital scalp. Women typically retain a rim of hair along the frontal hairline (Figs 1, 2), even when the scalp is visible behind the hairline (Ludwig, 1977). Miniaturized hairs, the shorter and ¢ner hairs of various lengths and diameters, are the hallmark of AGA (Frieden and Price, 1986; Whiting, 1993) and result from the shortening of anagen phase and reduction in matrix size. Increased spacing between hairs makes the central part appear wider over the frontal scalp compared to the occipital scalp. Less commonly, hair density appears normal proximally, but the hair no longer grows to its previous length, resulting in wispy distal ends (Fig 3); in this case, the shortening of anagen progresses more rapidly than matrix reduction. The patient may note that her ‘pony tail’ is smaller in girth and the hair is cut to shorter lengths in order to give a fuller appearance. Rarely, advanced thinning occurs with loss of the frontal hairline, but this is invariably associated with markedly elevated circulating androgens (Ludwig, 1977). Androgenetic alopecia in women is not usually accompanied by increased shedding. However, in some instances, an episode
Figure 2. Drawing of scalp regions. The frontal hairline refers to the rim of hair at the anterior margin of the frontal scalp.
of telogen e¥uvium following childbirth, major illness or other causes may uncover a latent predisposition to AGA. Sometimes the number of hairs shed is misleading in women who seldom comb their hair after shampooing, leave it to dry naturally; and later when their hands ‘comb’ through the hair many loose hairs appear on the ¢ngers. LABORATORY EVALUATION
Extensive hormonal testing is usually not needed unless symptoms and signs of androgen excess are present. Most women with AGA have normal menses and pregnancies. Women who require endocrine evaluation are identi¢ed in the o⁄ce with careful inquiry regarding menses, history of infertility, and the presence of hirsutism, severe unresponsive cystic acne, virilization, or galactorrhoea. If any one of these is present, laboratory measurement of serum total or free tesosterone, dehydroepiandrosterone sulfate and prolactin are indicated. Other common causes of hair loss are ruled out by measurement of serum thyrotropin, iron studies including serum iron and ferritin, complete blood count, and RPR. DIFFERENTIAL DIAGNOSIS
Figure 1. AGA in a 29-year-old woman. Note thinning in the frontal/parietal region and sparing of the frontal hairline.
The diagnosis of AGA in women is supported by early age of onset, the pattern of increased hair thinning over the frontal/parietal scalp with greater hair density over the occipital scalp, retention of the frontal hairline, and the presence of miniaturized hairs. When these features are present, the diagnosis is not problematic. However, AGA is common and other causes of hair loss
JID SYMPOSIUM PROCEEDINGS
(Mirmirani and Price, 2000). Both patchy and di¡use alopecia areata may coexist with AGA and sometimes present a challenging di¡erential which is resolved by a scalp biopsy showing a peribulbar lymphocytic in¢ltrate around anagen hair bulbs. An adult with loose anagen syndrome has a history of decreased hair density since early childhood, and the pull test is strongly positive (Price and Gummer, 1989). Low power microscopy of the easily extracted hairs shows misshapen anagen bulbs, a ‘‘ru¥ed’’ cuticle just distal to the bulb, and absent inner root sheath.
Figure 4. Mean change in hair weight in women using 2% topical minoxidil solution twice daily for 32 weeks.
Figure 3. AGA showing the less common presentation of full proximal hair density and wispy distal density; in these cases, the shortening of anagen progresses more rapidly than matrix reduction.
may coexist and need to be recognized or excluded. Scalp biopsy usually solves the di¡erential diagnosis. Biopsies are taken from active, representative sites, and horizontal sectioning is preferred because of the larger number of follicular structures that can be studied (Headington, 1984; Whiting, 1990). In AGA, biopsy shows increased numbers of miniaturized hairs, abundant and even enlarged sebaceous glands, and minimal in£ammation though the latter is controversial (Lattanand and Johnson, 1975; Whiting, 1998). Chronic telogen e¥uvium (CTE) may present a di⁄cult differential diagnosis even though the features of CTE seem distinct. In CTE, women in the fourth to sixth decade, with above average hair density, describe sudden onset of marked shedding from the entire scalp. Hair pull test may extract increased numbers of telogen hairs easily, yet scalp hair density appears normal or minimally decreased even though the shedding may be prolonged. Miniaturized hairs are not seen. Horizontal sections of a scalp biopsy distinguish CTE from AGA: the ratio of terminal hairs to miniaturized hairs in CTE is 9 : 1, in AGA is 2 : 1, and in a normal scalp is 7 : 1 (Whiting, 1996). Acute telogen e¥uvium may occur in a woman with long-standing AGA or unmask latent AGA. A careful history will usually identify the cause of profuse hair shedding such as high fever, severe dietary protein de¢ciency, or chronic blood loss as in women with prolonged heavy menses. Relevant laboratory tests are noted above. Categories of drugs that may cause hair loss include anticancer drugs, anticoagulants, anticonvulsants, antithyroid drugs, beta blockers, tricyclic antidepressants, and progestins with androgenic e¡ects
Figure 5. AGA in a 33-year-old woman before (top) and after (bottom) applying 2% topical minoxidil solution for 12 months.
VOL. 8, NO. 1 JUNE 2003
ANDROGENETIC ALOPECIA IN WOMEN
MEDICAL TREATMENT OF AGA IN WOMEN
Topical minoxidil solution is the only approved drug available for promoting hair growth in women with AGA (Price, 1999). E⁄cacy has been shown in double-blind, placebo-controlled studies using hair counts and hair weight as the primary end points (Price, 1999). Minoxidil-treated women had signi¢cantly higher hair counts and an increase in hair weight compared to women who received placebo. A double-blind study of women ages 22^41 years using 2% topical minoxidil solution showed a significant increase over placebo in mean change in hair weight at 16 weeks (Price and Menefee, 1990) (Fig 4). It should be noted that clinical perception of improved scalp coverage may take longer, up to 6^12 months, until hair length and mass are increased su⁄ciently (Fig 5). Minoxidil solution must be applied twice daily to a dry scalp. Finasteride is a competitive inhibitor of type II 5a-reductase, and is contraindicated in women who are or may become pregnant, because 5a-reductase inhibitors may cause abnormalities of the external genitalia of a male fetus. Finasteride was not e¡ective in postmenopausal women in a placebo-controlled study (Price et al, 2000). Women are devastated by their thinning hair (Cash et al, 1993), and need thoughtful evaluation and management, as well as reassurance that they may safely use hair cosmetics to make their hair appear fuller. REFERENCES Cash TF, Price VH, Savin RC: Psychological e¡ects of androgenetic alopecia on women. Comparisons with balding men and with female control subjects. J Am Acad Dermatol 29:568^575, 1993 Dallob AL, Sadick NS, Unger W, et al: The e¡ect of ¢nasteride, a 5a-reductase. inhibitor, on scalp skin testosterone and dihydrotestosterone concentrations in patients with male pattern baldness. J Clin Endocrinol Metab 79:703^706, 1994 Frieden IJ, Price VH: Androgenetic alopecia. In: Thiers BH, Dobson RL (eds). Pathogenesis of Skin Disease. New York: Churchill Livingstone, 1986: p 41^55 Hamilton JB: Patterned loss of hair in man: Types and incidence. Ann N Y Acad Sci 53:708^728, 1951 Headington JT: Transverse microscopic anatomy of the human scalp. Arch Dermatol 14:449^456, 1984 Kaufman K: Androgen metabolism as it a¡ects hair growth in androgenetic alopecia. Dermatol Clin 14:697^711, 1996 Kuster W, Happle R: The inheritance of common baldness: Two B or not two B? J AM Acad Dermatol 11:921^926, 1984 Lattanand A, Johnson WC: Male pattern alopecia. A histopathologic and histochemical study. J Cutan Pathol 2:58^70, 1975
Ludwig E: Classi¢cation of the types of androgenetic alopecia (common baldness) occuring in female sex. Br J Dermatol 97:247^254, 1977 Messenger AG: The control of hair growth: An overview. J Invest Dermatol 101, (1 Suppl):4S^9S, 1993 Mirmirani P, Price VH: Top 10 misconceptions about androgenetic alopecia in women. Cosmetic Dermatology 30^31 December, 2000 Olsen EA: Androgenetic alopecia. In: Olsen EA (ed). Disorders of Hair Growth: Diagnosis and Treatment. New York: McGraw-Hill, 1994; p 257^283 Price VH: Testosterone metabolism in the skin. A review of its function in androgenetic alopecia, acne vulgaris, and idiopathic hirsutism including recent studies with antiandrogens. Arch Dermatol 111:1496^1502, 1975 Price VH: Treatment of Hair Loss. N Engl J Med 341:964^973, 1999 Price VH, Gummer CL: Loose anagen syndrome. J Am Acad Dermatol 20:249^256, 1989 Price VH, Menefee E: Quantitative estimation of hair growth I. androgenetic alopecia in women: e¡ect of minoxidil. J Invest Dermatol 95:683^687, 1990 Price VH, Roberts JL, Hordinsky M, et al: Lack of e⁄cacy of ¢nasteride in postmenopausal women with androgenetic alopecia. J Am Acad Dermatol 43:768^776, 2000 Sawaya ME, Price VH: Di¡erent levels of 5a-reductase type I and II, aromatase, and. androgen receptor in hair follicles of women and men with androgenetic alopecia. J Invest Dermatol 109:296^300, 1997 Schweikert HU, Wilson JD: Regulation of human hair growth by steroid hormones. I. Testosterone metabolism in isolated hairs. J Clin Endocrinol Metab 38:811^819, 1974 Sreekumar GP, Pardinas J, Wong CQ, et al: Serum androgens and genetic linkage. analysis in early onset androgenetic alopecia. J Invest Dermatol 113:277^279, 1999 Trancik RJ, Spindler JR, Cuddihy RV, et al: Clinician survey evaluating monovidil topical solution in the treatment of androgenetic alopecia in patients under 18 years of age. Poster presented at 3rd Intercontinental Meeting of the Hair Research Societies, June 13^15, 2001: Tokyo, Japan, p129. Uno H, Cappas A, Schlagel C: Cyclic dynamics of hair follicles and the e¡ect of minoxidil on the bald scalps of the stumptailed macaques. Am J Dermatopathol 7:283^297, 1985 Venning VA, Dawber RPR: Patterned androgenic alopecia in women. J Am Acad Dermatol 18:1073^1077, 1988 Whiting DA: Male pattern hair loss: Current understanding. Int J Dermatol 37:561^ 566, 1998 Whiting DA: Chronic telogen e¥uvium: Increased scalp hair shedding in middle aged women. J Am Acad Dermatol 35:899^906, 1996 Whiting DA: Diagnostic and predictive value of horizontal sections of scalp biopsy. Specimens in male pattern androgenetic alopecia. J Am Acad Dermatol 28: 755^763, 1993 Whiting DA: The value of horizontal sections of scalp biopsies. J Cutan Aging Cosmet Dermatol 1:165^173, 1990 Ziller C: Pattern formation in neural crest derivatives. In: Dvan Neste, Randall VA (eds). Hair Research for the Next Millenium. Amsterdam, The Netherlands: Elsevier Science B, 1996; p 19^23