Seizure 20 (2011) 446–448

Contents lists available at ScienceDirect

Seizure
journal homepage: www.elsevier.com/locate/yseiz

Association between HLA-B*1502 allele and carbamazepine-induced severe

cutaneous adverse reactions in Han people of southern China mainland
Qian Wang, Jue-qian Zhou *, Lie-min Zhou, Zi-yi Chen, Zi-yan Fang, Shu-da Chen, Li-bai Yang,
Xiao-dong Cai, Qi-lin Dai, Hua Hong, Hong-xuan Wang
Department of Neurology, The First Affiliated Hospital of Sun Yet-sen University, Guangzhou, China

A R T I C L E I N F O A B S T R A C T

Article history: Previous studies have found a strong association between HLA-B*1502 and carbamazepine-induced
Received 19 October 2010 Stevens–Johnson syndrome in Asian areas including Taiwan, Hongkong and Thailand. This study
Received in revised form 27 January 2011 explores the association between HLA-B*1502 allele and carbamazepine-induced cutaneous adverse
Accepted 7 February 2011
reactions in Han Chinese of southern China mainland, and find the genetic marker that can predict
carbamazepine-induced cutaneous adverse reactions.
Keywords: HLA-B*1502 allele genotyping was performed by a polymerase chain reaction–sequence specific
HLA-B*1502
primers (PCR–SSP) method in 48 Han Chinese subjects who had carbamazepine-induced cutaneous
Carbamazepine
Stevens–Johnson syndrome
adverse reactions, including 9 severe cutaneous adverse reaction patients with Stevens–Johnson
Toxic epidermal necrolysis syndrome (SJS) or toxic epidermal necrolysis (TEN) and 39 cutaneous adverse reaction patients with
Maculopapular eruption maculopapular eruption (MPE). Meanwhile 80 carbamazepine-tolerant controls and 62 healthy
individuals were also tested.
The frequency of HLA-B*1502 allele among SJS/TEN patients (100%) is significantly higher than
carbamazepine-tolerant controls (13.75%, P < 0.001) and healthy individuals (17.74%, P < 0.001). But the
frequency between MPE patients and carbamazepine-tolerant controls (25.64% vs.13.75%, P = 0.110) did
not have any significant difference.
The data showed that HLA-B*1502 allele is strongly associated with carbamazepine-induced SJS/TEN
but not MPE in Han Chinese of southern China mainland.
ß 2011 British Epilepsy Association. Published by Elsevier Ltd. All rights reserved.

1. Introduction 2. Subjects and methods

Carbamazepine (CBZ) is widely used to treat epilepsy, bipolar 2.1. Subjects

disorder, trigeminal neuralgia and chronic pain. However,
carbamazepine can cause many forms of adverse reactions,
from mild skin rash to serious blistering conditions such as
Stevens–Johnson syndrome (SJS) and toxic epidermal necrolysis
(TEN). The latter two disorders carry high mortality up to 30%.1
Recently, the associations between HLA-B*1502 and carbamaz-
epine-induced Stevens–Johnson syndrome or toxic epidermal
necrolysis (CBZ-SJS/TEN) had been detected by more and more
studies. In addition, the US FDA recommended to do genetic
screening of HLA-B*1502 in patients from Asian ancestry before
initiation of carbamazepine therapy.2 Nevertheless, up to now
there are few reports in southern China mainland possessing the
largest population of Asian. Hence, we report preliminary results
about the study in southern China mainland.
* Corresponding author. Tel.: +86 13632378811; fax: +86 20 37864877.
E-mail addresses: zhoujq0215@163.com, 21324667@qq.com (J.-q. Zhou).
All subjects in the study were recruited from outpatients and
inpatients in the First Affiliated Hospital of Sun Yat-sen University
between 1995 and 2010. A total of 48 cases from unrelated families
were patients who developed cutaneous adverse reaction within
12 weeks after taking carbamazepine and for which no other
causes were found. They were classified into two groups according
to Roujeau’s diagnostic criteria.3 One group was diagnosed
Stevens–Johnson syndrome or toxic epidermal necrolysis, includ-
ing 9 patients (3 men and 6 women). Eight cases were diagnosed
SJS and one case was diagnosed TEN; the mean recipient age were
32.00  16.81. The other group was maculopapular eruption (MPE)
defined as erythematous exanthem without blistering, pustulation,
mucosal or systemic involvement, including 39 patients (25 men and
14 women); the mean recipient age were 32.23  19.85. Meanwhile,
80 patients (46 men and 34 women) of carbamazepine-tolerance
were invested, who taked carbamazepine at least 3 months without
adverse effects; the mean age were 29. 57  16.22. Healthy
individuals comprised subjects without taked carbamazepine and

1059-1311/$ – see front matter ß 2011 British Epilepsy Association. Published by Elsevier Ltd. All rights reserved.
doi:10.1016/j.seizure.2011.02.003
 Q. Wang et al. / Seizure 20 (2011) 446–448
had no history of drug induced cutaneous adverse reactions. In
healthy individuals group (27 men and 35 women), the mean age
were 38.08  13.60. The study was performed in accordance with the
Declaration of Helsinki and its amendments and the document of
informed consent was obtained from all subjects.
2.2. Genotyping
Genomic deoxyribonucleic acid (DNA) was isolated from
peripheral blood using the Genomic DNA Isolation Kit (Shanghai
Watsonbiot Technology Co., China). HLA-B*1502 genotype was
determined by polymerase chain reaction using sequence specific
primers (PCR-SSP).4 Four sets of HLA-specific primers (Pair1:
Forward primer [F1] 50 -CGAGAGAGCCTGCGGAAC-30 , reverse
primer [R1] 50 -GCCCACTTCTGGAAGGTTCT-30 ; Pair2: [F2] 50 -
CGCGAGTCCGAGGATGGC-30 , [R2] 50 -GCAGGTTCCGCAGGCTCT-30 ;
Pair3: [F3] 50 -ACCGGAACACACAGATCTC-30 , [R3] 50 -GAGCCACTC-
CACGCACAGT-30 ; Pair4: [F4] 50 -GGAGTATTGGGACCGGAAC-30 , [R4]
50 -GCCATACATCCTCTGGATGA -30 )and 2 sets of internal control
primers (PairC1: [FC1] 50 -TGCCAAGTGGAGCACCCAA-30 , [RC1] 50 -
GCATCTTGCTCTGTGCAGAT-30 ; PairC2: [FC2] 50 -ATGATGTTGACCT
TTCCAGGG-30 , [RC2] 50 -TTCTGTAACTTTTCATCAGTTFC-30 ;) were
used to perform PCR.4,9 The amplified fragment length of Pair1 to
Pair4 PCR products were 1340 base pairs (bp), 125 bp, 562 bp,
369 bp respectively, while the PairC1 and PairC2 PCR products
were 796 bp and 256 bp. The PCR conditions consisted of a
denaturation step at 968 C for 60 s; 5 cycles of 968 C for 20 s, 708 C
for 45 s, and 728 C for 25 s; 21 cycles of 968 C for 25 s, 658 C for 50 s,
728 C for 30 s; 4 cycles of 968 C for 30 s, 558 C for 60 s, and 728 C for
90 s. Cool by ramping to 208 C for 30 s prior to termination of the
program. PCR products were analyzed by electrophoretic separa-
tion on 2% agarose gels, followed by direct visualization over an
ultraviolet transilluminator after ethidium bromide staining.
Presence of all 4 HLA-specific PCR products and 2 internal control
PCR products denotes HLA-B*1502.
2.3. Statistics
Data were counted by positive or negative for HLA-B*1502. Chi-
square test and Fisher’s exact test were used to analyze the
association between carbamazepine-induced cutaneous adverse
reactions and HLA-B*1502. The strength of association was
estimated by calculating the odds ratio.5 P-values  0.05 (two-
sided) were considered to indicate statistical significance. Positive
predictive value, negative predictive value, sensitivity and
specificity were analyzed by fixed formula.
3. Results
Frequency of HLA-B*1502 genotype in three groups was shown
in Table 1. All of the 9 carbamazepine-induced severe cutaneous
adverse reaction patients carried HLA-B*1502. It showed signifi-
cant increase in HLA-B*1502 allele frequency when compared with
two control groups, carbamazepine-tolerant group (100% vs.
13.75%, x2 = 34.543, P < 0.001; OR = 114.826, 95% CI 6.25–
2,111.03) and healthy individuals group (100% vs.17.74%,
x2 = 26.281, P < 0.001; OR = 85.087, 95% CI 4.61–1569.48). But
there was no difference in the frequency of HLA-B*1502 allele
between the CBZ-MPE group and the control group (25.64% vs.
13.75%, x2 = 2.551, P = 0.110; OR = 2.163, 95% CI 0.828–5.649;
25.64% vs. 17.74%, x2 = 0.907, P = 0.341; OR = 1.599, 95% CI 0.606–
4.218). As the same, no significant association was found between
carbamazepine-tolerant group and healthy individuals group
(13.75% vs. 17.74%, x2 = 0.425, P = 0.514).
When carbamazepine-tolerant group was used as control, the
presence of HLA-B*1502 had a 45% positive predictive value for
447
Table 1
Frequency of HLA-B*1502 genotype in three groups [n (%)].
Group n HLA-B*1502 positive HLA-B*1502 negative
CBZ-SJS/TEN 9 9(100)a,b,c
0(0)
CBZ-MPE 39 10(25.64) 29(74.36)
CBZ-tolerant 80 11(13.75) 69(86.25)
Healthy individuals 62 11(17.74) 51(82.26)
a
Compared with CBZ-MPE group, x2 = 16.907, P < 0.001.
b
Compared with CBZ-tolerant group x2 = 34.543, P < 0.001.
c
Compared with normal individuals group x2 = 26.281, P < 0.001.
CBZ-SJS/TEN, whereas its absence had a negative predictive value
of 100%. As the test for CBZ-SJS/TEN, HLA-B*1502 allele had 100%
sensitivity and 86.25% specificity.
4. Discussion
In healthy individuals, the frequency of HLA-B*1502 allele
observed in our study was 17.74%, which was a little higher than
the study in healthy Han Chinese by Li Yan (11.5%).6 We presumed
our small sample make the difference. However the allele
frequency of HLA-B*1502 was only 1–2% in prior studies of
European populations.7 The obviously low frequency of HLA-
B*1502 in European might explain the reason of rare morbility of
SJS/TEN in Caucasian compared with Chinese.
Our study explores the association between HLA-B*1502 and
carbamazepine-induced cutaneous adverse reaction in Han
Chinese living in southern China mainland possessing the most
Han Chinese in the world. Although our study obtained a limited
number of patients, all of 9 cases with CBZ-SJS/TEN carried HLA-
B*1502 allele. The difference of association with HLA-B*1502
allele was markedly significant comparing with the controls.
Similar association was reported in other areas. Chung et al.
found that HLA-B*1502 allele was present in 100% of 44 CBZ–SJS
Han Chinese patients living in Taiwan but in only 3% of CBZ-
tolerant patients and in 9% of the control population.8 Since HLA-
B*1502 was showed strong association with CBZ–SJS/TEN in
Taiwan, similar result was also detected in Hongkong.9 A follow-
up study in Thai people, another Asian population who are non-
Han Chinese, got result in coincide with previous study in
Taiwan.10 But the association between HLA-B*1502 and CBZ-SJS/
TEN could not be confirmed in Japanese,11 HLA-B*1502 was even
not detected in any of 486 healthy Japanese.12 Studies in
Caucasians found no association between HLA-B*1502 and CBZ-
SJS patients who were European ancestry.13 Thus, HLA-B*1502
allele did not exist in CBZ-SJS/TEN patients of all ethnicities. We
postulated the failure might be explained by differences in
linkage disequilibrium patterns between ethnic groups. It
prompted that genetic association of HLA-B*1502 and CBZ-
SJS/TEN is ethnicity specific.
Unlike CBZ-SJS/TEN, we genotyped HLA-B*1502 allele in CBZ-
MPE in Han Chinese of southern China mainland, no significant
association was found when compared with frequency in the CBZ-
tolerant group or healthy individuals. The finding was in accordance
with reports in Taiwan and Hongkong.9,14 SJS/TEN and MPE might
belong to different types of cutaneous adverse reactions, SJS/TEN
are characterized by bullous lesion, while MPE is non-bullous
reaction. So we considered that genetic association of carbamaze-
pine induced cutaneous adverse reactions is phenotype-specific.
In conclusion, we confirmed the strong association between
HLA-B*1502 and CBZ-SJS/TEN in Han Chinese of southern China
mainland. On the contrary, no association between HLA-B*1502
and CBZ-MPE was observed. As a test for CBZ-SJS/TEN, it revealed
100% sensitivity, 86.25% specificity, 45% positive predictive value
and 100% negative predictive value of HLA-B*1502. Owing to data
in our study, we suggested HLA-B*1502 allele as a useful prediction
448 Q. Wang et al. / Seizure 20 (2011) 446–448
marker for carbamazepine-induced Stevens–Johnson syndrome
and toxic epidermal necrolysis in Han Chinese living in southern
China mainland.
Acknowledgements
This study is supported by Guangdong Technology Project
Foundation of China (2007B0315102004, 2009B060700108) and
the National Natural Science Foundation of China (10671213*). We
would like to thank Pro. Ding-lie Shen, Prof. Jing-hua Zhou, Dr. Pei-
qi Zhang of Guangdong 999 Brain Hospital for their assistance with
participant recruitment.
References
1. Zaccara G, Franciotta D, Perucca E. Idiosyncratic adverse reactions to anti-
epileptic drugs. Epilepsia 2007;48:1223–44.
2. Ferrell Jr PB, McLeod HL. Carbamazepine HLA-B*1502 and risk of Stevens–
Johnson syndrome and toxic epidermal necrolysis: US FDA recommendations.
Pharmacogenomics 2008;9:1543–6.
3. Roujeau JC, Stern RS. Severe adverse cutaneous reactions to drugs. N Engl J Med
1994;331:1272–85.
4. Bunce M, O’Neill CM. Phototyping: comprehensive DNA typing for HLA-A, B, C,
DRB1, DRB3, DRB4 DRB5 & DQB1 by PCR with 144 primer mixes utilizing
sequence-specific primers (PCR-SSP). Tissue Antigens 1995;46:355–67.
5. Jewell NP. On the bias of commonly used measures of association for 2  2
tables. Biometrics 1986;42:351–8.
6. Li Y, Tang XW, Wu JY, Yuan SM, Yang X, Huang YN, et al. HLA-B gene
polymorphism detected by high-resolution sequence-based typing in
Guangdong Han populations. Zhonghua Yi Xue Yi Chuan Xue Za Zhi
2006;23:50–4.
7. Geer L, Terasaki PI, Gjertson DW. In: Gjertson DW, Terasaki PI, editors. HLA 1998.
Kansas: Am. Soc. Histocompat. Immunogenet.; 1998. p. 327–63.
8. Chung WH, Hung SI, Hong HS, Hsih MS, Yang LC, Ho HC, et al. Medical genetics:
a marker for Stevens–Johnson syndrome. Nature 2004;428:486.
9. Man CB, Kwan P, Baum L, Yu E, Lau KM, Cheng AS, Ng MH. Association between
HLA-B*1502 allele and antiepileptic druginduced cutaneous reactions in Han
Chinese. Epilepsia 2007;48:1015–8.
10. Locharernkul C, Loplumlert J, Limotai C, Korkij W, Desudchit T, Tongkobpetch S,
et al. Carbamazepine and phenytoin induced Stevens–Johnson syndrome is
associated with HLA-B*1502 allele in Thai population. Epilepsia 2008;49:
2087–91.
11. Kaniwa N, Saito Y, Aihara M, Matsunaga K, Tohkin M, Kurose K, et al. HLA-B
locus in Japanese patients with anti-epileptics and allopurinol-related Stevens–
Johnson syndrome and toxic epidermal necrolysis. Pharmacogenomics 2008;9:
1617–22.
12. Tanaka H, Akaza T, Juji T. Report of the Japanese central bone marrow data
center. Clin Transpl 1996:139–44.
13. Alfirevic A, Jorgensen AL, Williamson PR, Chadwick DW, Park BK, Pirmohamed
M. HLA-B locus in Caucasian patients with carbamazepine hypersensitivity.
Pharmacogenomics 2006;7:813–8.
14. Hung SI, Chung WH, Jee SH, Chen WC, Chang YT, Lee WR, et al. Genetic
susceptibility to carbamazepine-induced cutaneous adverse drug reactions.
Pharmacogenet Genomics 2006;16:297–306.