Experimental Gerontology 37 (2002) 981–990

www.elsevier.com/locate/expgero

Mini-Review

Molecular mechanisms of androgenetic alopecia

Ralph M. Trüeb*
Department of Dermatology, University Hospital of Zurich, Gloriastr. 31, 8091 Zurich, Switzerland
Received 10 June 2002; received in revised form 17 June 2002; accepted 18 June 2002

Abstract
Androgenetic alopecia (AGA) is hereditary and androgen-dependent, progressive thinning of the scalp hair that follows a
defined pattern. While the genetic involvement is pronounced but poorly understood, major advances have been achieved in
understanding principal elements of the androgen metabolism involved: androgen-dependent processes are predominantly due
to the binding of dihydrotestosterone (DHT) to the androgen receptor (AR). DHT-dependent cell functions depend on the
availability of weak androgens, their conversion to more potent androgens via the action of 5alpha-reductase, low enzymatic
activity of androgen inactivating enzymes, and functionally active AR present in high numbers. The predisposed scalp exhibits
high levels of DHT, and increased expression of the AR. Conversion of testosterone to DHT within the dermal papilla plays a
central role, while androgen-regulated factors deriving from dermal papilla cells are believed to influence growth of other
components of the hair follicle. Current available treatment modalities with proven efficacy are oral finasteride, a competitive
inhibitor of type 2 5a-reductase, and topical minoxidil, an adenosine-triphosphate-sensitive potassium channel opener which
has been reported to stimulate the production of vascular endothelial growth factor in cultured dermal papilla cells. Since the
clinical success rate of treatment of AGA with modulators of androgen metabolism or hair growth promoters is limited,
sustained microscopic follicular inflammation with connective tissue remodeling, eventually resulting in permanent hair loss, is
considered a possible cofactor in the complex etiology of AGA. q 2002 Elsevier Science Inc. All rights reserved.
Keywords: Androgenetic alopecia; Androgen metabolism; Androgen receptor; Polygenic inheritance; Follicular microinflammation; Permanent
hair loss

Androgenetic alopecia (AGA), also referred to as androgen-dependent, and occurs in a defined pattern.
male-pattern hair loss or common baldness in men, It is assumed that the genetically predisposed hair
and as female-pattern hair loss in women, affects at follicles are the target for androgen-stimulated hair
least 50% of men by the age of 50 years, and up to follicle miniaturization, leading to gradual replace-
70% of all males in later life (Norwood, 1975). ment of large, pigmented hairs (terminal hairs) by
Estimates of its prevalence in women have varied barely visible, depigmented hairs (vellus hairs) in
widely, though recent studies claim that six percent of affected areas (Paus and Cotsarelis, 1999). The result
women aged under 50 years are affected, increasing to is a progressive decline in visible scalp hair density.
a proportion of 30– 40% of women aged 70 years and While male pattern AGA is characterized by its
over (Norwood, 2001). The hair loss is heritable, typical bitemporal recession of hair and balding
vertex, female pattern AGA is set apart by its diffuse
* Tel.: þ41-1255-3471; fax: þ41-1255-4549. thinning of the crown and intact frontal hairline.
E-mail address: ramitru@derm.unizh.ch (R.M. Trüeb). While prerequisites are thus a genetic predisposition
0531-5565/02/$ - see front matter q 2002 Elsevier Science Inc. All rights reserved.
PII: S 0 5 3 1 - 5 5 6 5 ( 0 2 ) 0 0 0 9 3 - 1
982 R.M. Trüeb / Experimental Gerontology 37 (2002) 981–990

and androgens, clinical practice has shown us that whether shedding of the telogen hair (teloptosis ) is
simply blocking androgens does not result in the also an active, regulated process or represents a
conversion of miniaturized follicles to terminal ones passive event that occurs at the onset of subsequent
in advanced alopecia. On histologic examination of anagen, as the new hair grows in (Paus and Cotsarelis,
scalp biopsies, the miniaturization of terminal hairs is 1999; Pierard-Franchimont and Pierard, 2001). There
frequently associated with perifollicular lymphocytic are considerable variations in length of these stages
infiltration, and eventually fibrosis (Jaworsky et al., depending on the body site location, with the duration
1992; Whiting, 1993). Therefore it is conceivable that of anagen determining the type of hair produced,
the role of this microscopic follicular inflammation particularly its length (Paus and Cotsarelis, 1999). On
causing fibrosis below the shortened balding follicle the scalp, hairs remain in anagen for a 2– 7-year
has been underestimated, though it seems likely that period of time, whereas that of telogen is 100 days,
this would prevent the follicle to reform a terminal leading to a ratio of anagen to telogen hairs of
hair follicle. approximately 9:1. On average the amount of new
It is the aim of this paper to review the molecular scalp hair formation essentially matches the amount
mechanisms resulting in AGA, as far as androgens, that is lost due to shedding (approximately 100/day),
genetics, and inflammatory phenomena are involved. thereby maintaining a consistent covering.
Hair growth control: The controls that underlie the
hair cycle reside within the hair follicle itself, and are
1. Hair-follicle cycling and signaling molecules believed to result from changes in the intra- and
controlling hair growth perifollicular expression of specific regulatory mol-
ecules and their receptors (Paus et al., 1999). Much
The hair-growth cycle: The hair follicle is subject circumstantial evidence suggests that the dermal
to constant turnover in the course of perpetual cycles papilla which is composed of specialized fibroblasts
through various stages of proliferation (anagen ), located at the base of the follicle, determines hair
involution (catagen ), and resting (telogen ), with follicle growth characteristics, especially the regu-
regeneration in the successive hair cycle (Fig. 1). It lation of cell proliferation and differentiation of hair
is a major characteristic of anagen that not only the follicle matrix: without papilla fibroblasts and an
hair shaft is growing but that most epithelial hair intimate contact with hair matrix keratinocytes
follicle compartments undergo proliferation, with the anagen cannot be sustained. Also, hair follicle
hair matrix keratinocytes located around the dermal morphogenesis can be induced by implanting dermal
papilla showing the highest proliferative activity. papilla cells under an appropriately receptive epi-
Also, the newly formed hair shaft is pigmented by the thelium (Jahoda et al., 1984). Finally, it has been
follicle pigmentary unit (Paus and Cotsarelis, 1999). shown that implanting few cells of follicle dermal-
During the following catagen stage of the hair cycle, sheath tissue from the scalp from an adult human male
hair follicles enter a highly controlled process of is sufficient to form new dermal papillae and induce
involution that is characterized by a burst new hair follicles in the skin of a genetically unrelated
of programmed cell death (apoptosis) in the majority female (Reynolds et al., 1999). There is substantial
of follicular keratinocytes, termination of pigment evidence from bioassays that cultured dermal papilla
production, substantial extracellular matrix-remodel- cells can secrete a number of cytokines, growth
ing, and condensation of the dermal papilla (Paus and factors and other, yet unidentified bioactive molecules
Cotsarelis, 1999). The resulting shortening of the that influence growth in other dermal papilla cells,
regressing epithelial strand is associated with an outer root sheath cells, keratinocytes, and endothelial
upward movement of the dermal papilla within the cells (Stenn et al., 1996). Finally, the hair cycle is
connective tissue sheath of the follicle. In telogen the subjected to cycle modulation by numerous extrinsic
hair shaft matures into a club hair, which is held influences, such as androgens (Paus, 1996).
tightly in the bulbous base of the follicular epithelium, Pathobiology of AGA: AGA is characterized by
before it is eventually shed from the follicle, usually progressive shortening of the duration of anagen with
as a result of combing or washing. It is still unresolved successive hair cycles, leading to decreased numbers
 R.M. Trüeb / Experimental Gerontology 37 (2002) 981–990 983

Fig. 1. The hair-growth cycle. Abbreviations: HS ¼ hair shaft; OHRS ¼ outer hair root sheath; IHRS ¼ inner hair root sheath; SG ¼ sebaceous
gland; Bu ¼ bulge; Me ¼ Melanocytes; HM ¼ hair matrix; DP ¼ dermal papilla.

of hair in anagen at any given time, and progressive does the response vary from stimulation to
follicular miniaturization with conversion of terminal inhibition of hair growth depending on the body
to vellus-like follicles (Paus and Cotsarelis, 1999). site, but androgen sensitivity also varies within
The result is increased shedding of short-lived telogen individual areas, i.e. regression in AGA occurs in
hairs (telogen effluvium), while the affected hair a patterned, progressive manner. Since many
follicles produce shorter, finer hairs that cover the extrinsic hair growth-modulatory factors, such as
scalp poorly. Since AGA involves a process of androgens (Randall et al., 1992), apparently
premature termination of anagen associated with operate at least in part via the dermal papilla,
premature entry into catagen, it is critically important research is currently also focused on identifying
to dissect the molecular controls of the anagen– androgen-regulated factors deriving from dermal
catagen transformation of the hair cycle (Paus, 1996). papilla cells.
Catagen has been suggested to occur as a consequence Of the several factors that have been suggested to
of decreased expression of anagen maintaining play a role in hair growth, so far only insulin-like
factors, such as insulin-like growth factor 1 (IGF-1), growth factor (IGF-1) has been reported as altered in
basic fibroblast growth factor (bFGF), and vascular vitro by androgens (Itami et al., 1995), and stem cell
endothelial growth factor (VEGF), and increased factor (SCF) has been found to be produced in higher
expression of cytokines promoting apoptosis, such as amounts by androgen-dependent beard cells than in
transforming growth factor beta 1 (TGFb 1), inter- control non-balding scalp cells, presumably also in
leukin-1alpha (IL-1a), and tumor necrosis factor response to androgens (Hibberts et al., 1996). Since
alpha (TNFa). SCF is the ligand for the cell surface receptor c-kit on
Responses to androgens are obviously also melanocytes, this may also play a role for hair
intrinsic to the individual hair follicle: not only pigmentation.
984 R.M. Trüeb / Experimental Gerontology 37 (2002) 981–990

2. Androgens, androgen metabolism, and the most target organs testosterone can be metabolized to
androgen receptor DHT by the enzyme steroid 5a-reductase. Based on
its affinity for the androgen receptor, DHT is fivefold
Androgens: Of various hormones that affect hair more potent than testosterone. DHT is implicated in
growth, the most studied are the androgens, particu- the pathogenesis of several disorders, including
larly as they pertain to AGA. Since Aristotle first benign prostatic hyperplasia, prostate cancer, hirsut-
noted that ‘maleness’ and sexual maturity were ism, acne vulgaris, and AGA.
required for balding, it was not until 1942 that Androgen metabolism within skin: The skin and
Hamilton’s observations on men deprived of testicular pilosebaceous unit are enzymatically equipped for
androgens by castration established beyond doubt that local metabolism and conversion of sex steroids
androgens, in the form of testosterone or its (Kaufman, 1996). The skin is capable of synthesizing
metabolites, were prerequisites for development of active androgens from the systemic precursor DHEA-
common baldness. Hamilton observed that men who sulfate (DHEA-S). The first step is the desulfatation of
were castrated before puberty did not develop AGA, DHEA-S by the enzyme steroid sulfate (STS). The
and that AGA can be triggered in castrated men by principal pathways involved in conversion of weak
injecting testosterone (Kaufman, 1996). androgens like DHEA to more potent androgens are
Androgen metabolism: Androgen metabolism through activity of the enzymes 3b-hydroxysteroid
comprises glandular and extraglandular production, dehydrogenase-D5!4-isomerase (3b-HSD), 17b-
transport, target cell metabolism, and cellular hydroxysteroid dehydrogenase (17b-HSD), and 5a-
response. While androgen biology in the adrenals reductase. Once formed, potent androgens, such as
and gonads, and the influence of the pituitary axis go testosterone and DHT, can be removed by conversion
beyond the scope of this review, androgen metabolism back to the weaker 17-ketosteroids, or are metab-
within the skin, as it pertains to hair growth and its olized via other enzymatic pathways, including
disorders, is the focus (Kaufman, 1996). The andro- aromatase, which converts androgens to estrogens,
gen metabolism pathway begins with pregnenolone, a and 3a-hydroxysteroid dehydrogenase to form
21 carbon steroid substrate, converted from choles- androsterone and androstanediol. The latter can be
terol. Following a-hydroxylation at the C-17 position, glucuronidated to form androgen conjugates that are
the action of the enzyme C17 – 20 lyase cleaves distal more rapidly cleared from the circulation. Remark-
carbon moieties, leaving a C19 carbon steroid with a ably, some target tissues, such as the hair follicle,
C-17 ketone in the distal ring. These ‘17-ketosteroids’ show enhanced androgen metabolism and androgen
make up a group of weak androgens, such as sensitivity. The activity of enzymes involved in
dehydroepiandrosterone (DHEA), defined by a low androgen metabolism within the skin has been studied
affinity for the androgen receptor. These weak in a variety of tissue preparations. The sebaceous
androgens, however, can be enzymatically converted glands in balding skin have been shown to express
to more potent androgens with greater affinity for the increased 3b-HSD activity when compared to non-
androgen receptor, such as testosterone. Testosterone balding scalp areas (Sawaya et al., 1988). Earlier it
is the major circulating androgen. In women, systemic was shown that plucked human hair follicles or hair
levels of testosterone are low compared with men, but follicles from balding stumptailed macaques express
the more abundant weak androgens serve as a source considerable 17b-HSD activity (Takashima et al.,
of precursors for potent androgens, which provide the 1970). In a study of plucked hair follicles from young
physiologic or pathophysiologic androgen activity. adults not yet expressing AGA but with a strong
Only a small fraction of androgens exists as free family history of baldness, two populations were
steroids in the circulation, with an equilibrium found, one with high 17b-HSD activity and one with
between free hormones and protein-bound androgens. low enzyme activity (Hodgins et al., 1985). The study
The most important protein for androgen binding is suggested that low enzyme activity may be related to
sex-hormone binding globulin (SHBG). Normally lesser degrees of balding. More recently, both men
70% of testosterone is bound to SHBG, and 19% to and women with AGA were shown to have higher
albumin. The remainder is circulating unbound. In levels of 5a-reductase enzyme activity in frontal
 R.M. Trüeb / Experimental Gerontology 37 (2002) 981–990 985

follicles than in their own occipital follicles, whereas exposing DNA-binding sites, and then bind to specific
higher levels of aromatase were found in their hormone response elements in the DNA, promoting
occipital follicles (Sawaya and Price, 1997). the expression of specific hormone-regulated genes.
Steroidogenic enzyme mutations: Since STS con- The AR is believed to be responsible for determining
verts DHEA-S to DHEA that is eventually metab- the sensitivity of cells to androgens. Besides androgen
olized to more potent androgens in the periphery, and insensitivity, various mutations have been described
elevated plasma levels of DHEA-S and DHEA have in the gene encoding the AR in a variety of diseases,
been reported to correlate with balding in young men, including spinal and bulbar muscular atrophy (Ken-
the hypothesis was advanced, that men with genetic nedy’s disease), and prostate cancer (Gottlieb et al.,
STS deficiency (X-linked recessive ichthyosis, XRI) 1998). Some of these are associated with functional
do not or only develop minor forms of AGA. A survey changes in AR expression. Expression of the AR has
of patients with XRI showed that this was not the case, also been found to be increased in balding scalp
since these men also showed advanced AGA (Trüeb (Randall et al., 1992; Sawaya and Price, 1997). Most
and Meyer, 2000). In genetically determined recently, polymorphism of the AR gene has been
deficiencies of the enzymes 3b-HSD, or 17b-HSD, found to be associated with male pattern baldness
respectively, the presence or absence of AGA has not (Ellis et al., 2001).
been investigated so far (Hoffmann and Happle,
2000).
The description of an unusual form of incomplete 3. Genetic involvement
male pseudohermaphroditism, due to a genetic
deficiency of the type 2 steroid 5a-reductase by The genetic involvement is pronounced, and the
Imperato-McGinley et al. (1974), implicated DHT as importance of genes concurs with marked racial
principal mediator of androgen-dependent hair loss. differences in prevalence of AGA; non-Caucasians
Affected men, who are homozygous for mutation of often exhibit significantly less balding. While
the gene, do not develop AGA. major progress has been done in the understanding
Mutations of the human gene encoding aromatase of androgen metabolism, the genetic predisposition
(CYP19) are rare and result in aromatase deficiency. to AGA remains poorly understood. A very high
Affected girls show pseudohermaphroditism at birth, frequency of AGA has complicated attempts to
and at puberty develop virilization and hirsutism due establish a mode of inheritance. Moreover, it is
to an androgen excess, pubertal failure with no signs not clear whether AGA is genetically homo-
of estrogen action, hypergonadotropic hypogonadism, geneous; some authorities suggest that female
polycystic ovaries, and a tall stature. Males are rather pattern hair loss is not the female counterpart of
tall with eunuchoid skeletal proportions. In theory, male AGA, and not androgen-dependent (Orme
females and males might develop early onset of AGA et al., 1999). The genes for type 1 and type 2 5a-
(Hoffmann and Happle, 2000). Consistent with the reductase have been shown not to be associated
role of aromatase in avoiding androgen-mediated with the inheritance of AGA (Ellis et al., 1998).
effects on androgen-dependent hair follicles, is the Polymorphism of the AR gene is associated with
observation that women taking aromatase inhibitors male pattern baldness (Ellis et al., 2001), however,
for the treatment of breast cancer often experience an the AR gene is located on the X chromosome and
AGA-like hair loss. does not explain the relatively strong concordance
Androgen receptor (AR): Finally, the absence of of the degree of baldness in fathers and sons. No
balding in individuals with the androgen-insensitivity specific gene has been identified so far, though
syndrome who lack functional AR clearly demon- single gene mutations, such as abnormality of the
strates the need for AR for AGA to occur (Quigley, AR, might be necessary, but not sufficient for the
1998). All steroid hormones act by diffusing through phenotype (Ellis et al., 2001). We probably deal
the plasma membrane into the target cell and binding with a polygenic inheritance, dependent on a
to specific intracellular receptors. The hormone- combination of mutations, e.g. in or around the
receptor complex undergoes conformational changes, AR gene affecting the expression of the AR, and
986 R.M. Trüeb / Experimental Gerontology 37 (2002) 981–990

other genes controlling androgen levels. Inter- observation of a perifollicular infiltrate in the
actions between such genes might account for the upper follicle near the infundibulum suggests that
tissue-specific and developmental stage-specific the primary causal event for the triggering of
expression of the AR that is necessary to explain inflammation might occur near the infundibulum
the characteristic anatomic and temporal patterns (Mahé et al., 2000). On the basis of this
of AGA. Other genes relevant to androgens, localization and the microbial colonization of the
including those on the Y chromosome might also follicular infundibulum with Propionibacterium
be examined (Ellis et al., 2001). sp., Staphylococcus sp., Malassezia sp., or other
members of the transient flora, one could specu-
late that microbial toxins or antigens could be
4. Hair follicle microinflammation involved in the generation of the inflammatory
response. The production of porphyrins by Pro-
The limited success rate of treatment of AGA pionibacterium sp. in the pilosebaceous duct has
with hair growth promoters or modulators of also been considered to be a possible cofactor of
androgen metabolism means that further patho- this initial pro-inflammatory stress (Mahé et al.,
genic pathways may be taken into account. The 2000). Alternatively, keratinocytes themselves may
implication of microscopic follicular inflammation respond to chemical stress from irritants, pollu-
in the pathogenesis of AGA has recently emerged tants, and UV irradiation, by producing radical
from several independent studies (Jaworsky et al., oxygen species and nitric oxide, and by releasing
1992; Mahé et al., 2000; Whiting, 1993). An early intracellularly stored IL-1a. This pro-inflammatory
study referred to an inflammatory infiltrate of cytokine by itself has been shown to inhibit the
activated T cells and macrophages in the upper growth of isolated hair follicles in culture
third of the hair follicles, associated with an (Philpott et al., 1996). Moreover, adjacent kerati-
enlargement of the follicular dermal-sheath com- nocytes, which express receptors for IL-1, start to
posed of collagen bundles (perifollicular fibrosis), engage the transcription of IL-1 responsive genes:
in regions of actively progressing alopecia mRNA coding for IL-1b, TNFa, and IL-1a, and
(Jaworsky et al., 1992). Horizontal section studies for specific chemokine genes, such as IL-8,
of scalp biopsies indicated that the perifollicular and monocyte chemoattractant protein-1 (MCP-1)
fibrosis is generally mild, consisting of loose, and MCP-3, themselves mediators for the recruit-
concentric layers of collagen that must be ment of neutrophils and macrophages, have been
distinguished from cicatricial alopecia (Whiting, shown to be upregulated in the epithelial compart-
1993). The term ‘microinflammation’ has been ment of the human hair follicle (Mahé et al.,
proposed, because the process involves a slow, 2000). Besides, adjacent fibroblasts are also fully
subtle, and indolent course, in contrast to the equipped to respond to such a pro-inflammatory
inflammatory and destructive process in the signal. The upregulation of adhesion molecules for
classical inflammatory scarring alopecias (Mahé blood-borne cells in the capillary endothelia,
et al., 2000). The significance of these findings has together with the chemokine gradient, drive the
remained controversial. However, morphometric transendothelial migration of inflammatory cells,
studies in patients with male pattern AGA treated which include neutrophils through the action of
with minoxidil showed that 55% of those with IL-8, T cells and Langerhans cells at least in part
microinflammation had regrowth in response to through the action of MCP-1. After processing of
treatment, in comparison to 77% in those patients localized antigen, Langerhans cells, or alterna-
without inflammation and fibrosis (Whiting, 1993). tively keratinocytes, which may also have antigen
Inflammatory phenomena: An important ques- presenting capabilities, could then present antigen
tion is how the inflammatory reaction pattern is to newly infiltrating T lymphocytes and induce T-
generated around the individual hair follicle. cell proliferation. The antigens are selectively
Inflammation is regarded as a multistep process destroyed by infiltrating macrophages, or natural
which may start from a primary event. The killer cells.
 R.M. Trüeb / Experimental Gerontology 37 (2002) 981–990 987

Perifollicular fibrosis: On the occasion that the Therapeutic challenges: The aim of therapy is to
causal agents persist, sustained inflammation is the increase hair coverage of the scalp and to retard
result, together with connective tissue remodeling, progression of hair thinning. Currently, two FDA
where collagenases, such as matrix metalloproteinase approved drugs are available for this purpose, oral
(also transcriptionally driven by pro-inflammatory finasteride, at a dose of 1 mg per day, and topical
cytokines) play an active role (Mahé et al., 2000). solution of minoxidil (Price, 1999). Finasteride is a
Collagenases are suspected to contribute to the tissue competitive inhibitor of type 2 5a-reductase and
changes in perifollicular fibrosis. inhibits the conversion of testosterone to DHT. The
Permanent alopecia: Generally, permanent alope- rationale for the use of finasteride to treat AGA in
cia is the result of irreversible damage to the putative men is based on the absence of AGA in men with
site of follicular stem cells in the ‘bulge’ area of the congenital deficiency of type 2 5a-reductase, and the
outer-root sheath in the superficial portion of the hair presence of increased 5a-reductase activity and DHT
follicle (Lavker et al., 1993). In most of inflammatory levels in balding scalp (Kaufman et al., 1998).
scarring alopecias, e.g. lichen planopilaris, lupus Finasteride is contraindicated in women who are or
erythematosus, and pseudopelade (Brocq), the inflam- may become pregnant, because 5a-reductase inhibi-
mation involves this area. In the recently described tors may cause malformation of the external
fibrosing alopecia in a pattern distribution (Zinker- genitalia of male fetuses. Minoxidil promotes hair
nagel and Trüeb, 2000), patients with AGA have growth through increasing the duration of anagen. It
additional clinical and histological features of inflam- causes hair follicles at rest to grow, and enlarges
mation and fibrosis limited to the area of androgenetic suboptimal follicles. While minoxidil was developed
hair loss. A lichen planopilaris-type inflammation for treatment of hypertension, and this feature of the
involving the bulge area presumably irreparably drug’s action is best understood, its mechanism of
damages follicle stem cells. The preference of this action on hair growth is poorly understood. Minox-
site of the follicle for the immunologic attack may be idil is a potassium-channel opener and vasodilatator,
related to the fact, that in contrast to the proximal hair and has been reported to stimulate the production of
follicle, the isthmus and infundibulum area do not VEGF in cultured dermal papilla cells (Lachgar et al.,
bear any immune privilege (Paus, 1997). 1998). There is evidence that this effect is mediated
by adenosine and sulfonylurea receptors, which are
well-known target receptors for adenosine-tripho-
5. Concluding remarks sphate-sensitive potassium channel openers (Li et al.,
2001). Topical solutions of 2 and 5 percent
Clinical and investigative advances have helped minoxidil are available for treatment of AGA in
us to understand some of the pathogenic steps men and women. Unfortunately, the efficacy of
leading to androgenetic hair loss (Fig. 2). Besides minoxidil is variable and temporary, making it
androgens and genetic imbalance, additional difficult to predict the success of treatment on an
pathogenic factors are suspected, such as individual basis. Estrogens and antiandrogens are
microbial flora, endogenous and exogenous stress, used in women with AGA, although no controlled
microinflammation, and possibly others. While studies have been done. When a combination of
further suspects are likely to be exposed, individ- estrogen and a progestin is prescribed for oral
ual diversity of causal agents, as well as of the contraception or hormonal replacement therapy in
sequence of events, or combined factors, must be women with AGA, care should be taken to select a
kept in mind, when addressing the biological progestin with no androgenic, or preferably with
conditions contributing to AGA. The large number antiandrogenic activity, e.g. cyproterone acetate.
of therapeutic molecules currently claimed to be Women with this condition should also avoid
active and patented in this field and their limited androgens and their precursors, such as DHEA,
efficacy in offering a definitive cure of AGA, since these may exacerbate hair loss (Price, 1999).
confirm that the mechanism of AGA is highly So far, the inflammatory component has not been
complex. included in treatment protocols for AGA. Finally, it
988 R.M. Trüeb / Experimental Gerontology 37 (2002) 981–990

Fig. 2. Androgenetic alopecia: pathogenic mechanisms and therapeutic strategies.

has been proposed that gene therapy may offer yet introduction of genes seems feasible (Li and Hoff-
another approach on condition that the genes man, 1995), though the large amounts of genetic
responsible for alopecia are identified (Paus and material and the need to re-apply the agent at
Cotsarelis, 1999). Given the accessibility of the hair intervals on a continued basis would make commer-
follicle and the availability of liposomal preparations cial use very expensive and impractical (Sawaya and
that selectively target the follicle, the topical Shapiro, 2000).
 R.M. Trüeb / Experimental Gerontology 37 (2002) 981–990 989

References lurea receptor 2B as a target of minoxidil. J. Invest. Dermatol.

117, 1594–1600.
Mahé, Y.F., Michelet, J.F., Billoni, N., Jarrousse, F., Buan, B.,
Ellis, J.A., Stebbing, M., Harrap, S.B., 1998. Genetic analysis of
Commo, S., Seint-Leger, D., Bernard, B.A., 2000. Androgenetic
male pattern baldness and the 5alpha-reductase genes. J. Invest.
alopecia and microinflammation. Int. J. Dermatol. 39, 576–584.
Dermatol. 110, 849– 853.
Norwood, O.T., 1975. Male-pattern baldness. Classification and
Ellis, J.A., Stebbing, M., Harrap, S.B., 2001. Polymorphism of the incidence. South Med. J. 68, 1359– 1370.
androgen receptor gene is associated with male pattern baldness. Norwood, O.T., 2001. Incidence of female androgenetic alopecia
J. Invest. Dermatol. 116, 452–455. (female pattern alopecia). Dermatol. Surg. 27, 53–54.
Gottlieb, B., Lehvaslaiho, H., Beitel, L.K., Lumbroso, R., Pinsky, Orme, S., Cullen, D.R., Messenger, A.G., 1999. Diffuse female hair
L., Trifiro, M., 1998. The androgen receptor gene mutations loss: are androgens necessary? Br. J. Dermatol. 141, 521–523.
database. Nucleic Acids Res. 26, 234 –238. Paus, R., 1996. Control of the hair cycle and hair diseases as cycling
Hibberts, N.A., Messenger, A.G., Randall, V.A., 1996. Dermal disorders. Curr. Opin. Dermatol. 3, 248 –258.
papilla cells derived from beard hair follicles secrete more stem Paus, R., 1997. Immunology of the hair follicle. In: Bos, J.D., (Ed.),
cell factor (SCF) in culture than scalp cells or dermal fibroblasts. Skin Immune System. Cutaneous Immunology and Clinical
Biochem. Biophys. Res. Commun. 222, 401–405. Immunodermatology, CRC Press, Boca Raton.
Hodgins, M.B., Murad, S., Simpson, N.B., 1985. A search for Paus, R., Cotsarelis, G., 1999. The biology of hair follicles. N. Engl.
variation in hair follicle androgen metabolism which might be J. Med. 341, 491 –497.
linked to male pattern baldness (abstract). Br. J. Dermatol. 113, Paus, R., Müller-Röver, S., Botchkarev, V.A., 1999. Chronobiology
794. of the hair follicle: hunting the ‘hair cycle clock’. J. Invest.
Hoffmann, R., Happle, R., 2000. Current understanding of Dermatol. Symp. Proc. 4, 338–345.
androgenetic alopecia. Part I: etiopathogenesis. Eur. Philpott, M.P., Sander, D.A., Bowen, J., Kealey, T., 1996. Effects of
J. Dermatol. 10, 319 –326. interleukins, colony stimulating factor and tumour necrosis
Imperato-McGinley, J., Guerrero, L., Gautier, T., Peterson, R.E., factor on human hair follicle growth in vitro: a possible role for
1974. Steroid 5a-reductase deficiency in man: an inherited form interleukin-1 and tumour necrosis factor-a in alopecia areata.
of male pseudohermaphroditism. Science 186, 1213–1215. Br. J. Dermatol. 135, 942 –948.
Itami, S., Kurata, S., Takayasu, S., 1995. Androgen induction of Pierard-Franchimont, C., Pierard, G.E., 2001. Teloptosis, a turning
follicular epithelial cell growth is mediated via insulin-like point in hair shedding biorhythms. Dermatology 203, 115–117.
growth factor I from dermal papilla cells. Biochem. Biophys. Price, V.H., 1999. Treatment of hair loss. N. Engl. J. Med. 341,
Res. Commun. 212, 988–994. 964 –973.
Jahoda, C.A., Horne, K.A., Oliver, R.F., 1984. Induction of hair Quigley, C.A., 1998. The androgen receptor: physiology and
growth by implantation of cultured dermal papilla cells. Nature pathophysiology. In: Nieschlag, E., Behre, H.M. (Eds.),
311, 560 –562. Testosterone: Action, Deficiency, Substitution, Springer, Berlin,
Jaworsky, C., Kligman, A.M., Murphy, G.F., 1992. Characterisation pp. 33–106.
of inflammatory infiltrates in male pattern alopecia: implication Randall, V.A., Thornton, M.J., Messenger, A.G., 1992. Cultured
for pathogenesis. Br. J. Dermatol. 127, 239 –246. dermal papilla cells from androgen-dependent human hair
Kaufman, K.D., 1996. Androgen metabolism as it affects hair follicles (e.g. beard) contain more androgen receptors than those
growth in androgenetic alopecia. Dermatol. Clin. 14, 697–711. from non-balding areas of the scalp. J. Endocrinol. 133,
Kaufman, K.D., Olsen, E.A., Whiting, D., Savin, R., DeVillez, R., 141 –147.
Bergfeld, W., Price, V.H., Van Neste, D., Roberts, J.L., Reynolds, A.J., Lawrence, C., Cserhalmi-Friedman, P.B., Chris-
Hordinsky, M., Shapiro, J., Binkowitz, B., Gormley, G.J., tiano, A.M., Jahoda, C.A., 1999. Trans-gender induction of hair
1998. Finasteride in the treatment of men with androgenetic follicles. Human follicle cells can be inducted to grow in an
alopecia. Finasteride male pattern hair losss study group. J. Am. incompatible host of the other sex. Nature 402, 33 –34.
Acad. Dermatol. 39, 578–589. Sawaya, M.E., Honig, L.S., Garland, L.D., Hsia, S.L., 1988. Delta
Lachgar, S., Charveron, M., Gall, Y., Bonafe, J.L., 1998. Minoxidil 5-3 beta-hydroxysteroid dehydrogenase activity in sebaceous
upregulates the expression of vascular endothelial growth factor glands of scalp in male-pattern baldness. J. Invest. Dermatol. 91,
in human hair dermal papilla cells. Br. J. Dermatol. 138, 101 –105.
407–411. Sawaya, M.E., Price, V.H., 1997. Different levels of 5alpha-
Lavker, R.M., Miller, S., Wilson, C., Cotsarelis, G., Wei, Z.G., reductase type I and II, aromatase, and androgen receptor in hair
Yang, J.S., Sun, T.T., 1993. Hair follicle stem cells: their follicles of women and men with androgenetic alopecia.
location, role in hair cycle, and involvement in skin tumor J. Invest. Dermatol. 109, 296 –300.
formation. J. Invest. Dermatol. 101, 16S– 26S. Sawaya, M.E., Shapiro, J., 2000. Androgenetic alopecia. New
Li, L., Hoffman, R.M., 1995. The feasibility of targeted selective approved and unapproved treatments. Dermatol. Clin. 18,
gene therapy of the hair follicle. Nat. Med. 1, 705–706. 47 –61.
Li, M., Marubayashi, A., Nakaya, Y., Fukui, K., Arase, S., 2001. Stenn, K.S., Combates, N.J., Eilertsen, K.H., Gordon, J.S., Pardinas,
Minoxidil-induced hair growth is mediated by adenosine in J.R., Parimoo, S., Prouty, S.M., 1996. Hair follicle growth
cultured dermal papilla cells: possible involvement of sulfony- controls. Dermatol. Clin. 14, 543–558.
990 R.M. Trüeb / Experimental Gerontology 37 (2002) 981–990

Takashima, I., Adachi, K., Montagna, W., 1970. Studies of common sections of scalp biopsy specimens in male pattern androgenetic
baldness in the stumptailed macaque IV. In vitro metabolism of alopecia. J. Am. Acad. Dermatol. 28, 755 –763.
testosterone in the hair follicles. J. Invest. Dermatol. 55, Zinkernagel, M.S., Trüeb, R.M., 2000. Fibrosing alopecia in a
329–334. pattern distribution. Patterned lichen planopilaris or androge-
Trüeb, R.M., Meyer, J.C., 2000. Male-pattern baldness in men with netic alopecia with a lichenoid tissue reaction pattern? Arch.
X-linked recessive ichthyosis. Dermatology 200, 247 –249. Dermatol. 136, 205–211.
Whiting, D.A., 1993. Diagnostic and predictive value of horizontal