ARTICLE
Auteur(s) : Patricia Entz1,2, Bettina
Blaumeiser3, Regina C Betz3,4, Julien
Lambert5, Katia Seymons5, Sibylle
Eigelshoven6, Sandra HAnneken6, Roland
Kruse6, Peter Nürnberg2,7, Marion
Nagy1, Markus M Nöthen4,8
1Institute for Legal Medicine, Charité University
Hospital, Berlin, Germany
2Gene Mapping Center and Department of Molecular
Genetics, Max Delbrück Center for Molecular Medicine, Berlin,
Germany
3Department of Medical Genetics, University Hospital of
Antwerp, Antwerp, Belgium
4Institute of Human Genetics, University of Bonn, Bonn,
Germany
5Department of Dermatology, University Hospital of
Antwerp, Antwerp, Belgium
6Department of Dermatology, University Hospital of
Düsseldorf, Düsseldorf, Germany
7Cologne Center for Genomics, University of Cologne,
Köln, Germany
8Department of Genomics, Life & Brain Center,
University of Bonn, Bonn, Germany
accepté le 1 Mars 2006
Alopecia areata (AA) is a common form of hair loss affecting
approximately 1-2% of the general population [1]. AA can manifest
in both sexes of all age groups and the progression of hair loss is
extremely variable. Classification is divided into three subtypes,
depending on the degree of hair loss: the mildest form of AA
involves patchy hair loss on the scalp or other areas of the body,
and is also named patchy alopecia areata (patchy AA). This can
either progress to alopecia totalis (AT), involving a total loss of
scalp hair or to alopecia universalis (AU), involving a complete
loss of scalp and body hair.The etiopathogenesis of AA is not
completely understood. Alopecia areata (AA) is thought to be a
tissue-specific autoimmune disease directed against the hair
follicle, and may be associated with other autoimmune diseases. The
mechanism of hair follicle dysfunction is immunological and is
mediated by activated T-cells [2].Familial aggregation of AA has
been reported by numerous studies [3-6]. In the first family study
of AA to obtain information directly from the relatives of AA
patients, we recently found that 21.8% of 206 AA patients have at
least one first-degree relative with AA and 34.0% have at least one
first- or second-degree relative with AA [6]. Among first-degree
relatives, the lifetime risk is estimated at 7.8% in parents, 7.1%
in sibs and 5.7% in children. The sibling risk ratio was calculated
to be 4.2, which is in the range found for other common disorders
with a multifactorial etiology.Various functional candidates have
been tested as possible disease susceptibility genes [7-10] but it
is only in the case of the major histocompatibility complex (HLA)
region on chromosome 6p21, that evidence has been obtained from a
large number of independent studies. The HLA alleles found most
consistently to be involved include alleles of the HLA class II
DRB1 system [11-23]. Despite these studies, however, a specific
gene defect for AA has not yet been identified in the HLA region.
Further studies of AA patients that define the HLA region in more
detail are needed, an approach that might finally lead to the
identification of a causal genetic variant. In this study we aimed
to investigate the relation between AA and the HLA-DRB1 locus in a
Belgian-German sample of AA patients and various subgroups.
Materials and methods
Patients
161 unrelated patients with AA (100 women and 61 men) aged 5-78
years (mean age 53) were included in this study. The patients were
recruited from the outpatient hair clinics of two Departments of
Dermatology, the University Hospitals at Antwerp (Belgium) and
Düsseldorf (Germany). Patients from Düsseldorf (n = 20) represent a
group of newly diagnosed patients, while the sample from Antwerp (n
= 141) was collected retrospectively. Clinical data of all patients
were obtained, including age at onset and familial occurrence. The
severity of alopecia was assessed according to the alopecia areata
investigational assessment guidelines [24] and patients were
categorized as having either patchy alopecia
(S1-S4), alopecia totalis (AT), alopecia
totalis/universalis (AT/AU), or alopecia universalis (AU). Patchy
alopecia includes the stages S1 (less than 25% hair
loss) to S4 (75-99% hair loss), AT was defined as 100%
scalp hair loss without loss of body hair, AT/AU was defined as
100% scalp hair loss with variable loss of body hair, AU was
defined as 100% loss of both scalp and body hair. The assessment of
severity is based on a lifetime perspective and is according to the
most severe episode ever experienced. Inherent to this severity
assignment is that assignment of patients to the less severely
affected subgroups is not certain since the patients may develop
more severe forms of the disease later in life.
74 patients had patchy AA, 12 AT, 2 AT/AU and 73 had AU. The
patients were grouped into two categories, i.e. mild AA (all
patients with patchy AA) and severe AA (combining the patients with
AT, AT/AU and AU). There was a family history of AA in 44/161
patients (27.3%), defined as having at least one first or second
degree relative with AA. Age at onset was defined as the age when
patchy hair loss was first noticed. Early onset alopecia (age at
onset ≤ 20 years) occurred in 79/161 patients (49.1%), late onset
alopecia (age at onset > 20 years) was observed in 82/161
patients (50.9%). All patients gave written informed consent for
the genetic studies, and the ethics committees of both institutions
approved the study.
The control group was comprised of 165 healthy unrelated sex and
age matched blood donors of German descent. Blood donors were not
specifically screened for the absence of AA as this would have
little impact on the power of a case-control study when the disease
studied has a population prevalence of approximately 1-2% as
reported for AA [25].
Typing methods and statistical analysis
HLA-DRB1 typing was performed using a combination of SSP
(sequence-specific-priming) and a HLA-DRB1 specific
PyrosequencingTM approach as described elsewhere [26].
In a first step, the PCR-SSP typing allows determination of DRB1
group specificity in patients and in controls. In a second step,
samples with genotypes belonging to those allele groups in which
the allele frequencies in patients clearly differed from controls
were analyzed at a high resolution level by pyrosequencing.
Pyrosequencing was carried out on a PSQ96 (Pyrosequencing, Uppsala,
Sweden) according to the protocols of the supplier and using the
before mentioned sequencing strategy. The pyrograms were evaluated
“by hand” and the resulting sequences were compared with the
HLA-DRB1 allele sequences in the database (IMGT/HLA data base,
http://www.ebi.ac.uk/imgt/hla/). Statistical analysis was performed
by two-by-two contingency tables. Fisher’s exact two-tailed test
was calculated. When significant p-values were achieved the odds
ratio and 95% confidence intervals were calculated. In calculating
the frequencies of HLA alleles, patients with only a single
detected allele were assumed to be homozygous for that allele. A
significance level of p > 0.05 was assumed for all statistical
tests.
Results
In our study, the DRB1 locus was examined in a sample of 161 AA
patients as compared to a sex- and age matched control sample using
a combination of PCR-SSP and PyrosequencingTM.
Table 1( Table 1 ) summarizes the
results of the HLA-DRB1 allele frequencies obtained in AA patients
and controls. The frequency of the HLA-DRB1*03 allele was
significantly decreased in patients (7.5%) as compared to controls
(13.6%) (p = 0.011, odds ratio 0.510, 95% CI 0.293-0.884). Further
subgrouping of DRB1*03 alleles demonstrated that *0301 appears to
be the mainly responsible allele (6.8% versus 11.2%, p = 0.048,
odds ratio 0.581, 95% CI 0.323-1.041). The frequency of the DRB1*04
allele was significantly increased in AA patients (20.8%) when
compared to controls (13.3%) (p = 0.012, odds ratio 1.708, 95% CI
1.104-2.645). Subdivision of the *04 group revealed that the *0401
allele was the only allele to achieve significance (13.4% versus
7.3%, p = 0.014, odds ratio 1.965, 95% CI 1.129-3.435) and
accounted for the greatest proportion of the effect.
The other DRB1 allele groups that have been tested (*01, *07,
*08, *09, *10, *11, *13, *14, *15,*16) failed to achieve
statistical significance.
Table 2( Table 2 ) divides the DRB1
allele frequencies according to severity, age at onset and
familiarity. In terms of severity, we found no significant
differences between patients and controls for the HLA-DRB1 alleles
tested. However, when we also take non-significant trends into
account, the risk allele *0401 seems to have a more pronounced
effect in severe cases than in mild cases of AA (16.9% versus 9.5%,
p = 0.071). When we compared patients with an early onset of
alopecia (≤ 20 years) to patients with a later age at onset (>
20 years), only one comparison reached statistical significance
(group *03, *11, *13, *14: 30.0% versus 41.3%, p = 0.037, odds
ratio 1.640, 95% CI 1.006-2.677). Further subdivision of alleles,
however, failed to implicate individual alleles.
In patients with a positive family history of AA, we observed a
significant difference in HLA-DRB1 allele frequencies for *03 and
*0301 when compared to patients without a family history (10.3% vs.
2.3%, p = 0.020, odds ratio 4.914, 95% CI 1.092-30.778 and 8.6% vs.
2.3%, p=0.049, odds ratio 4.019, 95% CI 0.879-25.489). A similar
trend for a stronger effect in familial cases was observed with the
*0401 allele (18.2% in familial cases versus 11.5% in sporadic
cases of AA). This comparison, however, did not reach statistical
significance (p = 0.141).
Table 1 Frequency of HLA-DRB1 alleles in all patients
with alopecia areata compared with controls
|
HLA-DRB1
|
Alleles in Controls
|
%
|
Alleles in Patients
|
%
|
p-value
|
|
*01
|
39
|
11.8
|
41
|
12.7
|
0.744
|
|
*15, *16
|
58
|
17.6
|
58
|
18.0
|
0.913
|
|
*08, *12
|
20
|
6.1
|
16
|
5.0
|
0.529
|
|
*03, *11,*13, *14
|
140
|
42.4
|
116
|
36.0
|
0.082
|
|
thereof *03
|
45
|
13.6
|
24
|
7.5
|
0.011
|
|
*0301
|
37
|
11.2
|
22
|
6.8
|
0.048
|
|
*0302
|
1
|
0.3
|
0
|
0
|
1.000
|
|
*03XXa
|
7
|
|
2
|
|
|
|
thereof *11
|
45
|
13.6
|
38
|
11.8
|
0.464
|
|
*04
|
44
|
13.3
|
67
|
20.8
|
0.012
|
|
*0401
|
24
|
7.3
|
43
|
13.4
|
0.014
|
|
*0402
|
1
|
0.3
|
1
|
0.3
|
1.000
|
|
*0403
|
5
|
1.5
|
0
|
0
|
0.076
|
|
*0404
|
6
|
1.8
|
3
|
0.9
|
0.520
|
|
*0405
|
1
|
0.3
|
2
|
0,6
|
0.987
|
|
*0407
|
1
|
0.3
|
3
|
0.9
|
0.603
|
|
*0408
|
1
|
0.3
|
3
|
0.9
|
0.603
|
|
*04XXa
|
5
|
|
12
|
|
|
|
*07
|
27
|
8.2
|
21
|
6.5
|
0.405
|
|
*09
|
0
|
0
|
1
|
0.3
|
0.993
|
|
*10
|
2
|
0.6
|
2
|
0.6
|
1.000
|
|
alleles (2n)
|
330
|
|
322
|
|
|
Table 2 Frequency of HLA-DRB1 alleles according to
severity, age at onset and familiarity (%)
|
HLA-DRB1
|
Mild AA
|
Severe AA
|
P-value
|
Early-onset AA
|
Late-onset AA
|
P-value
|
Familial AA
|
Sporadic AA
|
P-value
|
|
alleles (2n)
|
148
|
174
|
|
158
|
164
|
|
88
|
234
|
|
|
*01
|
14.9
|
11.1
|
0.317
|
14.7
|
11.1
|
0.403
|
10.2
|
13.7
|
0.459
|
|
*15, *16
|
21.6
|
15.1
|
0.146
|
19.3
|
16.9
|
0.565
|
18.2
|
18.0
|
1.000
|
|
*08, *12
|
6.8
|
3.5
|
0.204
|
4.7
|
5.2
|
1.000
|
2.3
|
6.0
|
0.252
|
|
*03, *11,*13, *14
|
32.4
|
39.0
|
0.294
|
30.0
|
41.3
|
0.037
|
34.1
|
36.8
|
0.697
|
|
thereof *03
|
8.1
|
7.0
|
0.678
|
6.0
|
8.7
|
0.400
|
2.3
|
10.3
|
0.020
|
|
*0301
|
6.8
|
7.0
|
1.000
|
6.0
|
7.6
|
0.661
|
2.3
|
8.6
|
0.049
|
|
*0302
|
0.0
|
0.0
|
1.000
|
0.0
|
0.0
|
1.000
|
0.0
|
0.0
|
1.000
|
|
thereof *11
|
8.8
|
14.5
|
0.165
|
10.7
|
12.8
|
0.606
|
10.2
|
12.4
|
0.700
|
|
*04
|
16.9
|
24.4
|
0.130
|
22.7
|
19.2
|
0.492
|
26.1
|
18.8
|
0.166
|
|
*0401
|
9.5
|
16.9
|
0.071
|
16.0
|
11.1
|
0.250
|
18.2
|
11.5
|
0.141
|
|
*0402
|
0.0
|
0.6
|
1.000
|
0.0
|
0.6
|
1.000
|
1.1
|
0.0
|
0.273
|
|
*0403
|
0.0
|
0.0
|
1.000
|
0.0
|
0.0
|
1.000
|
0.0
|
0.0
|
1.000
|
|
*0404
|
0.7
|
1.7
|
0.628
|
0.7
|
1.7
|
0.626
|
1.1
|
1.3
|
1.000
|
|
*0405
|
0.7
|
0.6
|
1.000
|
0.7
|
0.6
|
1.000
|
0.0
|
0.9
|
1.000
|
|
*0407
|
0.7
|
1.2
|
1.000
|
1.3
|
0.6
|
0.600
|
1.1
|
0.9
|
1.000
|
|
*0408
|
2.0
|
0.0
|
0.096
|
0.7
|
1.2
|
1.000
|
0.0
|
1.3
|
0.565
|
|
*07
|
6.8
|
6.4
|
1.000
|
7.3
|
5.8
|
0.654
|
9.1
|
5.6
|
0.310
|
|
*09
|
0.0
|
0.6
|
1.000
|
0.7
|
0.0
|
0.466
|
0.0
|
0.4
|
1.000
|
|
*10
|
0.7
|
0.6
|
1.000
|
0.7
|
0.6
|
1.000
|
0.0
|
0.9
|
1.000
|
Discussion
Due to the pattern of familial occurrence in AA, there is a strong
assumption that the inheritance pattern of AA is that of a complex
genetic trait. Although an autoimmune pathomechanism for AA has
been suggested, the precise etiology is unknown, and no autoantigen
or causative gene has been identified so far. However, an
association between alopecia areata and HLA alleles, especially
HLA-DRB1, has been described in the literature [11-23]. A
comparison of several studies with samples of diverse ethnic
backgrounds reveals differences in the critical allele as well as
varying allele frequencies between populations. Since no specific
gene variant has yet been identified as a cause of AA despite
positive associations, there is a need to identify associated
alleles in each particular population.
Our study, which concentrated on the HLA-DRB1 locus in a
Belgian-German sample, showed a significant increase in the
appearance of DRB1*04 alleles and a significant decrease in the
appearance of DRB1*03 alleles in patients with AA, as discussed in
more detail below.
Our study confirms that DRB1*03 is a protective factor against
the development of AA and suggests that it is the DRB1*0301 allele
that confers the major portion of this effect. A protective effect
of DRB1*03 has been suggested previously, based on the
investigation of Polish and Turkish patients [23, 22]. Due to the
small sample sizes of these previous studies, with 52 patients
having been included in the Polish study and 65 patients in the
Turkish study, the results from their further subgrouping of
patients were not conclusive. In our study, where we investigated a
substantially larger sample of patients, we observed no significant
frequency differences in patients subgrouped according to severity
of disease and age of onset. Smaller effects still remain possible,
however, and require enlargement of samples. Interestingly, we
observed a significant difference (P = 0.02) in DRB1*03 frequency
between familial (2.3%) and sporadic cases (10.3%), suggesting that
the protective effect of DRB1*03 is mainly present in patients with
a genetic background of the disease while it is very small in
non-familial cases. We regard our method of assessing familiarity
in our patients as relatively sensitive since we directly
interviewed all available first-degree relatives and obtained
information about second-degree relatives through interviews with
patients and first-degree relatives. The high sensitivity in
identifying familial cases is illustrated by the percentage of
familial cases (27.3%) found in our sample which is in the upper
range of rates reported in the literature [3, 5].
Our study also provides strong support for a risk conferring
role of DRB1*04 in the European population. Similar results have
been reported for North American white, UK white, Danish and
Turkish patients [12, 14-16, 18-20, 22]. It has been previously
suggested that the effect of DRB1*04 is restricted to long-standing
disease [18, 19]. We did not test this since we had not obtained
this information from our patients in a standardized manner. A
long-standing course of disease is, however, associated with
early-onset and a severe clinical expression of the phenotype.
Subdividing our patients, we observed differences that were in the
expected direction. The observed tendencies, however, did not reach
thresholds for significance, highlighting the importance of a large
sample size in addressing differential effects between subgroups of
patients with sufficient power. In our study, we further
discriminated between subclasses of DRB1*04. The most prevalent
allele, DRB1*0401, seems to be responsible for the effect of
DRB1*04, confirming results obtained in a North American White
population [18, 19].
To our knowledge, this is the first study investigating HLA
class II alleles in Belgian-German patients with AA. We have been
able to confirm the protective effect of DRB1*03 and the
predisposing effect of DRB1*04 in the development of AA. Additional
genotyping at the allelic level has attributed the effect of
DRB1*03 to DRB1*0301 and confirmed the effect of DRB1*0401 in the
DRB1*04 group. On the basis of results from the subgrouping of
patients, we suggest that the protective effect of DRB1*03 is
mainly conferred by patients with a positive family history of the
disease. This finding underscores the importance of documenting
family history in genetic research into AA.
Acknowledgments
The authors thank the patients for their participation in the
study. We thank Dr. Christine Schmael for the help in preparing the
manuscript. Regina C. Betz is a recipient of an Emmy Noether grant
from the German research foundation (DFG) and of a grant of the
BONFOR programme of the Medical Faculty of the University of Bonn.
Markus M. Nöthen is recipient of a grant of the Alfried Krupp von
Bohlen und Halbach-Stiftung.
References
1 Safavi KH, Muller SA, Suman VJ, Moshell AN,
Melton LJ. Incidence of alopecia areata in Olmsted County,
Minnesota, 1975 through 1989. Mayo Clin Proc 1995; 70: 628-33.
2 Alexis AF, Dudda-Subramanya R, Sinha AA.
Alopecia areata: autoimmune basis of hair loss. Eur J Dermatol
2004; 14: 364-70.
3 Shellow W, Edwards J, Koo J. Profile of
alopecia areata: a questionnaire analysis of patients and family.
Int J Dermatol 1992; 31: 186-9.
4 van der Steen P, Traupe H, Happle R,
Boezeman J, Strater R, Hamm H. The genetic risk for
alopecia areata in first degree relatives of severely affected
patients. Acta Dermatol Venerol 1992; 72: 373-5.
5 Yang S, Yang J, Liu JB, Wang HY,
Yang Q, Gao M, Liang YH, Lin GS, Lin D,
Hu XL, Fan L, Zhang XJ. The genetic epidemiology of
alopecia areata in China. Br J Dermatol 2004; 151: 16-23.
6 Blaumeiser B, van der Goot I, Fimmers R,
Hanneken S, Ritzmann S, Seymons K, Betz RC,
Ruzicka T, Wienker TF, De Weert J, Lambert J,
Kruse R, Nöthen MM. Familial aggregation of alopecia
areata. J Am Acad Dermatol 2006; 54(4): 627-32.
7 Galbraith GM, Pandey JP. Tumor necrosis factor alpha
(TNF-alpha) gene polymorphism in alopecia areata. Hum Genet 1995;
96: 433-6.
8 Barahamani N, de Anrade M, Slusser J,
Zhang Q, Duvic M. Interleukin-1 receptor antagonist
allele 2 and familial alopecia areata. J Invest Dermatol 2002; 118:
335-7.
9 Tazi-Ahnini R, di Giovine FS, McDonagh AJ,
Messenger AG, Amadou C, Cox A, Duff GW,
Cork MJ. Structure and polymorphism of the human gene for the
interferon-induced p78 protein (MX1): evidence of association with
alopecia areata in the Down syndrome region. Hum Genet 2000; 106:
639-45.
10 Tazi-Ahnini R, Cork MJ, Gawkrodger DJ,
Birch MP, Wengraf D, McDonagh AJ, Messenger AG.
Role of the autoimmune regulator (AIRE) gene in alopecia areata:
strong association of a potentially functional AIRE polymorphism
with alopecia universalis. Tissue Antigens 2002; 60: 489-95.
11 Averbakh EV, Pospelov LE. HLA antigens of patients
with alopecia areata. Vestn Dermatol Venerol 1986; 1: 24-6.
12 Frentz G, Thomsen K, Jakobsen BK,
Svejgaard A. HLA-DR4 in alopecia areata. J Am Acad Dermatol
1986; 14: 129-30.
13 Orecchia G, Belvedere MC, Martinetti M,
Capelli E, Rabbiosi G. Human leukocyte antigen region
involvement in the genetic predisposition to alopecia areata.
Dermatologica 1987; 175: 10-4.
14 Odum N, Morling N, Georgsen J,
Jakobsen BK, Frentz G, Jensen GF, Fugger L,
Svejgaard A. HLA-DP antigens in patients with alopecia areata.
Tissue Antigens 1990; 35: 114-7.
15 Duvic M, Hordinsky MK, Fiedler VC,
O’Brien WR, Young R, Reveille JD. HLA-D locus
associations in alopecia areata. DRw52a may confer disease
resistance. Arch Dermatol 1991; 127: 64-8.
16 Zhang L, Weetman AP, Friedmann PS,
Oliveira DB. HLA associations with alopecia areata. Tissue
Antigens 1991; 38: 89-91.
17 Welsh EA, Clark HH, Epstein SZ,
Reveille JD, Duvic M. Human leukocyte antigen-DQB1*03
alleles are associated with alopecia areata. J Invest Dermatol
1994; 103: 758-63.
18 Colombe BW, Price VH, Khoury EL,
Garovoy MR, Lou CD. HLA class II antigen associations
help to define two types of alopecia areata. J Am Acad Dermatol
1995; 33: 757-64.
19 Colombe BW, Lou CD, Price VH. The genetic
basis of alopecia areata: HLA associations with patchy alopecia
areata versus alopecia totalis and alopecia universalis. J Investig
Dermatol Symp Proc 1999; 4: 216-9.
20 De Anrade M, Jackow CM, Dahm N,
Hordinsky M, Reveille JD, Duvic M. Alopecia areata
in families: associations with the HLA locus. J Investig Dermatol
Symp Proc 1999; 4: 220-3.
21 Kavak A, Baykal C, Özarmagan G, Akar U.
HLA in alopecia areata. Int J Dermatol 2000; 39: 589-92.
22 Akar A, Orkunoglu E, Sengül A, Ozata M,
Gür AR. HLA class II in patients with alopecia areata. Eur J
Dermatol 2002; 12: 236-9.
23 Broniarczyk-Dyla G, Prusinska-Bratos M,
Dubla-Berner M, et al. The protective role of the HLA-DR
locus in patients with various clinical types of alopecia areata.
Arch Immunol Ther Exp (Warsz) 2002; 50: 333-6.
24 Olsen E, Hordinsky M, Price V, Roberts J,
Shapiro J, Canfield D, Duvic M,
King Jr. L, McMichael A, Randall V,
Turner M, Sperling L, Whiting D, Norris D.
Alopecia areata investigational assessment guidelines-Part II. J Am
Acad Dermatol 2004; 51: 440-7.
25 Moskvina V, Holmans P, Schmidt KM,
Craddock N. Design of case-controls studies with unscreened
controls. Ann Hum Genet 2005; 69: 566-76.
26 Entz P, Toliat MR, Hampe J,
Valentonyte R, Jenisch S, Nürnberg P, Nagy M.
New strategies for efficient typing of HLA class-II loci DQB1 and
DRB1 by using Pyrosequencing. Tissue Antigens 2005; 65: 67-80.
|
Version imprimable
Version PDF