Frequency of HLA-A alleles in Japanese patients with head and neck cancer

doi:10.1093/jjco/29.11.535

This Article

	Abstract
	Alert me when this article is cited
	Alert me if a correction is posted

Services

	Email this article to a friend
	Similar articles in this journal
	Similar articles in ISI Web of Science
	Similar articles in PubMed
	Alert me to new issues of the journal
	Add to My Personal Archive
	Download to citation manager
	Search for citing articles in: ISI Web of Science (4)
	Request Permissions

Google Scholar

	Articles by Eura, M
	Articles by Ishikawa, T
	Search for Related Content

PubMed

	PubMed Citation
	Articles by Eura, M
	Articles by Ishikawa, T

Social Bookmarking

	What's this?

Frequency of HLA-A Alleles in Japanese Patients with Head and Neck Cancer

Japanese Journal of Clinical Oncology Pages 535-540

Frequency of HLA-A Alleles in Japanese Patients with Head and Neck Cancer
Introduction
Materials And Methods
   Subjects
   DNA Extraction and PCR Amplification of the HLA-A Gene
   Dot-blot Hybridization
   Statistical Analysis
Results
   Specific Amplification of the HLA-A Locus and Dot-blot Hybridization with SSOPs
   Frequencies of HLA-A Alleles in SCCHN Patients and Control Subjects
Discussion
Acknowledgments
References

Frequency of HLA-A Alleles in Japanese Patients with Head and Neck Cancer

Masao Eura¹, Fumihiro Katsura¹, Masatake Oiso¹, Atsushi Obata¹, Koji Nakano¹, Keisuke Masuyama¹, Takeru Ishikawa²

¹Department of Otolaryngology, Kumamoto University School of Medicine, Kumamoto and ²Allergy and Immunology Center of Kyushu, Kumamoto, Japan

Background: Association between certain human leukocyte antigen (HLA) types such as HLA-A1 and -A3 and squamous cell carcinoma of the head and neck (SCCHN) has been demonstrated in the Caucasian population. HLA typings in these studies were performed by conventional serological methods. However, recent comparison studies between serological and molecular typings have revealed that the former are often inaccurate.

Methods: The frequency of HLA-A alleles in 100 Japanese patients with SCCHN and 100 control subjects was determined by the polymerase chain reaction, with primers specific for the HLA-A locus, in combination with dot-blot hybridization with 31 sequence-specific oligonucleotides.

Results: The frequencies of HLA-A*2602 and HLA-A*3303 were higher and those of HLA-A*2603 and HLA-A*3101 were lower in the patients with SCCHN than in healthy controls, but these differences were not statistically significant. In the 39 male patients with laryngeal carcinoma, the most common malignancies in Japanese patients with SCCHN, the frequency of HLA-A*2402 was significantly lower than that in the 80 male controls; however, after correction of the P value, statistical significance was not confirmed. In oral carcinoma patients, the frequency of HLA-A*2402 was significantly higher than that in healthy controls.

Conclusions: These results suggest that the contribution of certain HLA-A alleles to susceptibility to SCCHN may differ between sites in the head and neck regions, despite these cancers being of an identical histological type, and that HLA-A*2402 may influence the development of oral carcinoma in Japanese patients.

Key words: HLA-A alleles - PCR-SSOP - Japanese - head and neck cancer

INTRODUCTION

HLA class I glycoproteins are encoded by three separate genetic loci (HLA-A, HLA-B, HLA-C) and are highly polymorphic. In general, HLA class I phenotype is defined by serological assays such as the microcytotoxic test. However, serological typing cannot distinguish many class I subtypes; for example, HLA-A2 has been shown to comprise at least 18 subtypes by polymerase chain reaction (PCR)-based DNA typing (1). Studies of the peptide repertoire of HLA-A2 subtypes have revealed that they are functionally distinct in terms of peptide binding and presentation (2-4). Moreover, serological HLA-A class I typing gives typing errors in comparison with DNA typing (5). Thus, PCR-based DNA typing is required for the precise investigation of HLA-linked predisposition to disease.

Recently, genotype analysis has identified more than 50 HLA-A alleles that differ in frequency among ethnic groups (6,7). With the use of 91 sequence-specific oligonucleotide probes (SSOPs) to determine the genotype for 60 HLA-A alleles, Date et al. (8) examined 553 unrelated healthy Japanese subjects and detected 21 of these 60 alleles. We have now selected 31 of these 91 SSOPs to determine the genotype for these 21 HLA-A alleles in Japanese subjects and have compared allele frequencies between individuals with squamous cell carcinoma of the head and neck (SCCHN) and age-matched controls.

MATERIALS AND METHODS

Subjects

One hundred Japanese patients with SCCHN (89 men and 11 women) were selected in order of their appearance at the inpatient and outpatient clinics of Kumamoto University Hospital between April 1995 and March 1997. These patients were from Kumamoto Prefecture and their ages ranged from 45 to 94 years (mean 67.2 years) at the time of diagnosis. They consisted of 40 individuals with laryngeal, 22 with oral, 14 with oropharyngeal, 13 with hypopharyngeal and 11 with maxillary sinus carcinoma. The control subjects (80 men and 20 women; mean age 62.3 years) were volunteers aged 43-82 years with no history of cancer, recruited from Kumamoto Prefecture. The control group was selected to exhibit similar gender and age distributions to those of the patients.

DNA Extraction and PCR Amplification of the HLA-A Gene

Granulocytes were isolated from heparinized peripheral blood (20 ml) of each subject by Kimura and Sasazuki's method (9). DNA was extracted from the granulocytes, after proteinase K, with the use of Kimura and Sasazuki's phenol-chloroform method. HLA-A DNA was amplified with the use of the HLA-A gene-specific primers CGA01 [5´-CCGAACCCTC(C/G)TCCTGCTA-3´] and CGA33 [5´-TGGCCCCTGGTACCCGT-3´] as described (8). Each PCR reaction mixture (50 µl) contained 2 µl of genomic DNA, 1.0 U of Taq polymerase (GIBCO-BRL), 5 µl of 10× PCR buffer [100 mM Tris-HCl (pH 8.4), 15 mM MgCl₂, 500 mM KCl], 5 µl of 0.5 mM deoxynucleotide triphosphates and 1 µl (25 pmol) of each primer. The PCR protocol consisted of an initial denaturation at 95°C for 5 min, 30 cycles of denaturation (95°C, 1 min), annealing (54°C, 1 min) and extension (72°C, 90 s) and final extension at 72°C for 10 min.

Dot-blot Hybridization

Thirty-one SSOPs were selected from 91 designed by Date et al. (8) and were targeted to highly polymorphic regions of exons 2 and 3 of the HLA-A gene. The SSOPs were labeled with ³²P with the use of [[gamma]-³²P]ATP and T4 polynucleotide kinase (GIBCO-BRL). PCR products were spotted onto nylon membranes (Hybond-N; Amersham), immobilized by alkaline denaturation with 0.4 M NaOH and neutralized with 10× SSPE [1.5 M NaCl, 0.1 M sodium phosphate (pH 7.4) and 10 mM EDTA]. The membranes were exposed for 1 h at 54°C to a hybridization solution containing 50 mM Tris-HCl (pH 8.0), 3 M tetramethylammonium chloride (TMAC), 2 mM EDTA, 5× Denhardt's solution, 0.1% SDS and heat-denatured herring sperm DNA (100 µg/ml) and then for 1 h at 54°C to the same solution containing ³²P-endlabeled SSOPs. The filters were then washed twice for 10 min at room temperature in 2× SSPE containing 0.1% SDS, once for 10 min at room temperature in TMAC solution [50 mM Tris-HCl (pH 8.0), 3 M TMAC, 2 mM EDTA, 0.1% SDS] and twice for 10 min at 58°C in TMAC solution. Hybridization signals were detected by exposure of the filters to X-ray film for 4-12 h at room temperature.

Statistical Analysis

Fisher's exact probability test or the [chi]² test was used to analyze all the 2×2 tables. A P value of <0.05 was considered statistically significant. Relative risk was calculated as (a × d)/(b × c), where a, b, c and d are the number of markers + patients, markers - patients, markers + controls and markers - control, respectively.

RESULTS

Specific Amplification of the HLA-A Locus and Dot-blot Hybridization with SSOPs

Specific amplification of the HLA-A gene was successfully achieved by PCR with primers CGA01 and CGA33 (data not shown). We selected a panel of 31 SSOPs (Table 1) to detect the 21 HLA-A alleles that have been previously identified in the Japanese population (8) and their expected hybridization patterns are shown in Table 2. As expected, SSOP 58, which recognizes all HLA-A alleles, hybridized to all samples tested (Fig. 1A), confirming successful amplification of HLA-A gene. A representative dot-blot hybridization with SSOP 67, which is specific for HLA-A*0207, is shown in Fig. 1B. SSOP 69 (5´-TTCTACACCTCCGTGTC-3´), which is required for distinguishing HLA-A*0206 from A*0201, yielded ambiguous hybridization signals. We therefore compared SSOP 69 with SSOP 9Y designed by Fleischhauer et al. (1) for the same purpose. Because 9Y yielded clearer results than 69, we used the former SSOP for subsequent analysis.

Figure 1. Detection of PCR-amplified products of HLA-A genomic DNA by dot-blot hybridization. Filters containing PCR products from 50 DNA samples were hybridized with ³²P-labeled SSOP 58. (A) Specific for all HLA-A alleles or SSOP 67; (B) specific for HLA-A*0207.

Table 1. SSOPs used in this study

Region Probe DNA Sequence (5´->3´) Amino acid

Sequence Position

Exon 2

A 68 TATTTCTTCACATCCGTG YFFTSV 07-12

72 TATTTCACCACATCCGTG YFTTSV 07-12

9Y GGTATTTCTACACCTCCG RYFYTS 06-11

B 74 AAGCGGGGCTCCCCGCGG RGEPRF 17-22

75 TGAAGCGGGGCTTCCCGC RGKPRF 17-22

76 AAGCGGGGCTCTCCACTG SGEPRF 17-22

C 01 GAGCCAGAAGATGGAGCC SQKMEP 42-47

D 06 CAGGAGAGGCCTGAGTAT QERPEY 54-59

E 08 TATTGGGACGGGGAGACA YWDGET 59-64

10 GGACGAGGAGACAGGGAA WDDETG 60-65

11 GGGACCGGAACACACGGA WDRNTR 60-65

F 07 TGACTGGGCCTTCACTTT KVKAQS 66-71

14 TGAGTGGGCCTTCACTTT KVKAHS 66-71

15 TGACTGGGCCTTCACATT NVKAQS 66-71

G 17 ACCGAGCGAACCTGGGGA QRANLG 74-79

18 ACCGAGTGGACCTGGGGA DRVDLG 74-79

21 ACCGAGAGAACCTGGGGA DRENLG 74-79

Exon 3

H 67 GCAGCCACACATCCTCTG QRMCGC 96-101

25 GCAGCCAAACATCCTCTG QRMFGC 96-101

26 GCAGCCATACATCCTCTG QRMYGC 96-101

I 55 CGTAAGCGTCCTGCTGGT YQQDAY 113-118

56 CGTAAGCGTTCTGCTGGT YQQNAY 113-118

78 CGTAAGCGTCCCGCTGGT YQRDAY 113-118

K 89 GACCACCAAGCACAAGTG TTKHKW 142-147

35 GAGGCGGCCCATGTGGCG EAAHVA 148-153

36 GAGACGGCCCATGAGGCG ETAHEA 148-153

37 GAGGCGGCCCGTGTGGCG EAARVA 148-153

40 GAGGCGGCCCATGCGGCG EAAHAA 148-153

43 GTGGCGGAGCAGCAGAGA VAEQQR 152-157

L 51 GTGCGTGGAGTGGCTCCG CVEWLR 164-169

58 GCTCCGCAGATACCTGGA LRRYLE 168-173

Frequencies of HLA-A Alleles in SCCHN Patients and Control Subjects

We determined the HLA-A genotypes of 100 Japanese patients with SCCHN and 100 normal controls. The HLA-A allele frequencies of these two groups are shown in Table 3. The frequencies of HLA-A*2602 and HLA-A*3303 in SCCHN patients were 1.7 and 1.9 times, respectively, those of control subjects; however, control subjects were not statistically significant. In contrast, the frequencies of HLA-A*2603 and HLA-A*3101 in the control subjects were 2.3 and 1.7 times, respectively, those for SCCHN patients but, again, the differences were not significant.

Next, we determined the frequency of HLA-A alleles based on the primary tumor sites in the head and neck. As shown in Table 3, a significant association was observed only in patients with carcinoma of the oral cavity. Namely, the frequency of HLA-A*2402 in the 22 oral carcinoma patients (age range 55-80 years, mean 68.4 years ) was significantly higher than that in the control (p = 0.030, RR = 3.833).

Laryngeal carcinoma is the most common malignancy among SCCHN patients in Japan; in the present study, 40 (39 men and 1 woman) of the 100 SCCHN patients had laryngeal carcinoma. We compared HLA-A allele frequencies between the male patients with laryngeal carcinoma (age range 45-95, mean 67.1 years) and the 80 male healthy controls (age range 43-82, mean 64.7 years) (Table 4). The frequency of HLA-A*2402 in the male laryngeal carcinoma patients was significantly smaller than that in the male healthy controls (P = 0.037, RR = 0.436). However, after correction for the number of alleles, the difference did not reach statistical significance (p = 0.059). In addition, the frequency of HLA-A*3101 in the male laryngeal cancer patients was one-fifth of that in the male controls, although this difference was not statistically significant (P = 0.099, RR = 0.184).

Table 4. Frequency of HLA-A alleles in 39 male patients with laryngeal carcinoma and 80 male healthy controls

Allele Allele frequency (%)

Patients Controls

A*0101 0 0

A*0201 12.8 16.3

A*0203 0 0

A*0206 28.2 20.0

A*0207 5.1 10.0

A*0210 2.6 0

A*0301 0 0

A*1101 28.2 21.1

A*1102 0 0

A*2301 0 0

A*2402 35.9* 56.3

A*2407 0 0

A*2408 0 0

A*2601 20.5 12.5

A*2602 7.7 3.8

A*2603 0 5.0

A*2605 0 0

A*2606 0 0

A*3001 0 1.3

A*3101 2.6 12.5

A*3303 15.4 11.3

*P = 0.059 compared with controls.

DISCUSSION

Associations between HLA type and various malignant diseases have been demonstrated. Thus, HLA-A3 has been shown to be significantly associated with an increased incidence of SCCHN (10,11), whereas HLA-A1 showed a significant negative relation with disease incidence (10). These studies, based on conventional serological HLA typing, were performed with populations in which most individuals were Caucasian.

The frequencies of HLA types A1 and A3 in Japanese are <1% and none of the subjects in the present study exhibited any of these phenotypes. Hence we could not confirm the associations between the incidence of SCCHN and HLA-A1 and A3. We were able to determine the HLA-A genotype of all the Japanese subjects in the present study with the use of only 31 SSOPs and this is the first study to determine the frequencies of HLA-A alleles using DNA typing method in Japanese patients with SCCHN.

Although the frequencies of HLA-A*2602 and HLA-A*3303 were higher and those of HLA-A*2603 and HLA-A*3101 were lower in the SCCHN patients than in normal controls, these differences were not significant. Moreover, no significant association between the various HLA-A alleles and clinical parameters such as stage, lymph node metastasis or histological grade was apparent (data not shown). Additionally, to examine the influence of smoking, we divided the patients into two groups: 45 heavy smokers (Brinkmann index >600; Brinkman index = number of cigarettes smoked/day × smoking years) and 55 non-heavy smokers including 19 non-smokers (Brinkmann index <600)

Although the frequency of HLA-A*2402 was higher in the heavy smokers (65.5%) than in non-heavy smokers (54.2%), the difference was not significant.

Although SCCHN exhibits a distinct histology, the clinical characteristics vary depending on the site of the carcinoma in the head and neck regions. For example, laryngeal and maxillary sinus carcinomas show a lower incidence of neck metastasis (<30%) than does pharyngeal carcinoma (>50%). It was therefore of interest to investigate the possible association of HLA-A alleles with SCCHN at specific sites. Despite our limited number of 22 oral carcinoma patients, the frequency of HLA-A*2402 was significantly higher among patients than among healthy controls. In contrast, the frequency of HLA-A*2402 in patients with laryngeal carcinoma was significantly smaller than that in healthy control; however, after correction for the number of alleles, the difference did not reach statistical significance.

Thus, the frequency of HLA-A alleles differed according to the primary tumor site in the head and neck region. These results suggest that the contribution of HLA-A related factors to tumorigenicity of SCCHN may differ according to the SCCHN site.

Acknowledgments

We thank Y. Date for technical advice on dot-blot hybridization. This work was supported in part by a Grant-in-Aid for Scientific Research (B) and Priority Areas (Cancer) from the Ministry of Education, Science, Sports and Culture of Japan.

References

1. Fleischhauer K, Zino E, Mazzi B, Severini, GM, Benazzi E, Bordignon C. HLA-A*02 subtype distribution in Caucasians from northern Italy: identification of A*0220. Tissue Antigens 1996;48:673-9. MEDLINE Abstract

2. Rivoltini L, Loftus DJ, Barracchini K, Arienti F, Mazocchi A, Biddison WE, et al. Binding and presentation of peptides derived from melanoma antigens MART-1 and glycoprotein-100 by HLA-A2 subtypes. J Immunol 1996;156:3882-91. MEDLINE Abstract

3. Barouch D, Friede T, Stevanovic S, Tussey L, Smith K, Jones SR, et al. HLA-A2 subtypes are functionally distinct in peptide binding and presentation. J Exp Med 1995;182:1847-56. MEDLINE Abstract

4. Sudo T, Kamikawaji N, Kimura A, Date Y, Savoie J C, Nakashima H, et al.Differences in MHC class I self peptide repertoires among HLA-A2. subtypes. J Immunol 1995;155:4749-56. MEDLINE Abstract

5. Ohashi YM, Alosco S, Granja C, Salazar M, Hegland J, Yunis E. Accurate typing of HLA-A antigens and analysis of serological deficiencies. Tissue Antigens 1997;50:380-6. MEDLINE Abstract

6. Lorentzen DF, Iwanaga KK, Meuer KJ, Moritz TL, Watkins DI. A 25% error rate in serologic typing of HLA-B homozygotes. Tissue Antigens 1997;50:359-65. MEDLINE Abstract

7. Krausa P, Brywka M III, Savage D, Hui KM, Bunce M, Ngai JLF, et al. Genetic polymorphism within HLA-A*02: significant allelic variation revealed in different populations. Tissue Antigens 1995;45:223-31. MEDLINE Abstract

8. Date Y, Kimur A, Kato H, Sasazuki T. DNA typing of the HLA-A gene: population study and identification of four new alleles in Japanese. Tissue Antigens 1996;47:93-101. MEDLINE Abstract

9. Kimura A, Sasazuki T. Eleventh International Histocompatibility Workshop reference protocol for the HLA DNA-typing technique. In: Tsuji K, Aizawa M, Sasazuki T, editors. HLA 1991 Vol. 1. Oxford: Oxford University Press 1992;397-419.

10. Tarpley JL, Chretien PB, Rogentine GN, Twomey PL, Dellon AL. Histocompatibility antigens and solid malignant neoplasms. Arch Surg 1975;110:269-71. MEDLINE Abstract

11. Houck JR, Romano PJ, Bartholomew M, Smith PJ, Kloszewski F, Velsell ES. Do histocompatibility antigens influence the risk of head and neck carcinoma? Cancer 1997;69:2327-32.

Received February 5, 1999; accepted July 21, 1999
For reprints and all correspondence: Masao Eura, Department of Otolaryngology, Kumamoto University School of Medicine, 1-1-1 Honjo, Kumamoto 860-8556, Japan. E-mail: eura{at}gpo.kumamoto-u.ac.jp
Abbreviations: HLA, human leukocyte antigen; SCCHN, squamous cell carcinoma of the head and neck; PCR, polymerase chain reaction; SSOP, sequence-specific oligonucleotide probe; TMAC, tetramethylammonium chloride

This page is run by Oxford University Press, Great Clarendon Street, Oxford OX2 6DP, as part of the OUP Journals
Comments and feedback: jnl.info{at}oup.co.uk
Last modification:
Copyright© 1999 Foundation for the Promotion of Cancer Research.

CiteULike

Connotea

Del.icio.us What's this?

This Article

	Abstract
	Alert me when this article is cited
	Alert me if a correction is posted

Services

	Email this article to a friend
	Similar articles in this journal
	Similar articles in ISI Web of Science
	Similar articles in PubMed
	Alert me to new issues of the journal
	Add to My Personal Archive
	Download to citation manager
	Search for citing articles in: ISI Web of Science (4)
	Request Permissions

Google Scholar

	Articles by Eura, M
	Articles by Ishikawa, T
	Search for Related Content

PubMed

	PubMed Citation
	Articles by Eura, M
	Articles by Ishikawa, T

Social Bookmarking

	What's this?

Region	Probe	DNA Sequence (5´->3´)	Amino acid
Region	Probe	DNA Sequence (5´->3´)	Sequence	Position
		Exon 2
A	68	TATTTCTTCACATCCGTG	YFFTSV	07-12
	72	TATTTCACCACATCCGTG	YFTTSV	07-12
	9Y	GGTATTTCTACACCTCCG	RYFYTS	06-11
B	74	AAGCGGGGCTCCCCGCGG	RGEPRF	17-22
	75	TGAAGCGGGGCTTCCCGC	RGKPRF	17-22
	76	AAGCGGGGCTCTCCACTG	SGEPRF	17-22
C	01	GAGCCAGAAGATGGAGCC	SQKMEP	42-47
D	06	CAGGAGAGGCCTGAGTAT	QERPEY	54-59
E	08	TATTGGGACGGGGAGACA	YWDGET	59-64
	10	GGACGAGGAGACAGGGAA	WDDETG	60-65
	11	GGGACCGGAACACACGGA	WDRNTR	60-65
F	07	TGACTGGGCCTTCACTTT	KVKAQS	66-71
	14	TGAGTGGGCCTTCACTTT	KVKAHS	66-71
	15	TGACTGGGCCTTCACATT	NVKAQS	66-71
G	17	ACCGAGCGAACCTGGGGA	QRANLG	74-79
	18	ACCGAGTGGACCTGGGGA	DRVDLG	74-79
	21	ACCGAGAGAACCTGGGGA	DRENLG	74-79
		Exon 3
H	67	GCAGCCACACATCCTCTG	QRMCGC	96-101
	25	GCAGCCAAACATCCTCTG	QRMFGC	96-101
	26	GCAGCCATACATCCTCTG	QRMYGC	96-101
I	55	CGTAAGCGTCCTGCTGGT	YQQDAY	113-118
	56	CGTAAGCGTTCTGCTGGT	YQQNAY	113-118
	78	CGTAAGCGTCCCGCTGGT	YQRDAY	113-118
K	89	GACCACCAAGCACAAGTG	TTKHKW	142-147
	35	GAGGCGGCCCATGTGGCG	EAAHVA	148-153
	36	GAGACGGCCCATGAGGCG	ETAHEA	148-153
	37	GAGGCGGCCCGTGTGGCG	EAARVA	148-153
	40	GAGGCGGCCCATGCGGCG	EAAHAA	148-153
	43	GTGGCGGAGCAGCAGAGA	VAEQQR	152-157
L	51	GTGCGTGGAGTGGCTCCG	CVEWLR	164-169
	58	GCTCCGCAGATACCTGGA	LRRYLE	168-173

HLA-Aallele	SSOP
	A			B			C	D	E			F			G			H			I			K						L
	68	72	9Y	74	75	76	01	06	08	10	11	07	14	15	17	18	21	67	25	26	55	56	78	89	35	36	37	40	43	51	58
*0101	+	-	-	+	-	-	+	-	-	-	-	-	-	-	+	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	+
*0201	+	-	-	+	-	-	-	-	+	-	-	-	+	-	-	+	-	-	-	+	-	-	-	+	+	-	-	-	-	+	+
*0203	+	-	-	+	-	-	-	-	+	-	-	-	+	-	-	+	-	-	-	+	-	-	-	+	-	+	-	-	-	+	+
*0206	-	-	+	+	-	-	-	-	+	-	-	-	+	-	-	+	-	-	-	+	-	-	-	+	+	-	-	-	-	+	+
*0207	+	-	-	+	-	-	-	-	+	-	-	-	+	-	-	+	-	+	-	-	-	-	-	+	+	-	-	-	-	+	+
*0210	-	-	+	+	-	-	-	-	+	-	-	-	+	-	-	+	-	-	+	-	-	-	-	+	+	-	-	-	-	+	+
*0301	+	-	-	+	-	-	-	-	-	-	-	-	-	+	-	+	-	-	-	-	-	-	-	-	-	-	-	-	-	+	+
*1101	-	-	+	+	-	-	-	-	-	-	-	-	-	+	-	+	-	-	-	-	-	-	-	-	-	-	-	+	-	+	+
*1102	-	-	+	-	+	-	-	-	-	-	-	-	-	+	-	+	-	-	-	-	-	-	-	-	-	-	-	+	-	+	+
*2301	-	-	-	+	-	-	-	-	-	+	-	-	+	-	-	-	-	-	-	-	-	-	-	-	-	-	+	-	-	-	+
*2402	-	-	-	+	-	-	-	-	-	+	-	-	+	-	-	-	-	-	-	-	-	-	-	-	+	-	-	-	+	-	+
*2407	-	-	-	+	-	-	-	-	-	+	-	+	-	-	-	-	-	-	-	-	-	-	-	-	+	-	-	-	+	-	+
*2408	-	-	-	+	-	-	-	-	+	-	-	-	+	-	-	-	-	-	-	-	-	-	-	-	+	-	-	-	+	-	+
*2601	-	-	+	+	-	-	-	-	-	-	+	-	-	-	+	-	-	-	-	+	+	-	-	-	-	+	-	-	-	+	+
*2602	-	-	+	+	-	-	-	-	-	-	+	-	-	-	+	-	-	-	-	+	-	+	-	-	-	+	-	-	-	+	+
*2603	-	-	+	+	-	-	-	-	-	-	+	-	-	-	-	+	-	-	-	+	-	-	-	-	-	+	-	-	-	+	+
*2605	-	-	+	+	-	-	-	-	-	-	-	-	-	-	-	-	+	-	-	+	+	-	-	-	-	+	-	-	-	+	+
*2606	-	-	+	+	-	-	-	-	-	-	+	-	-	-	-	+	-	-	-	+	-	-	+	-	-	+	-	-	-	+	+
*3001	-	-	-	-	-	+	-	+	-	-	-	-	-	-	-	+	-	-	-	-	-	-	-	-	-	-	-	-	-	+	+
*3101	-	+	-	+	-	-	-	+	-	-	-	-	-	-	-	+	-	-	-	-	-	-	-	-	-	-	+	-	-	+	+
*3303	-	+	+	+	-	-	-	-	-	-	+	-	-	-	-	+	-	-	-	-	-	-	-	-	-	-	+	-	-	+	+

Allele	Allele frequency
	Patients: % (No.)
	Maxillary sinus (11)	Oral cavity (22)	Oro-pharynx (14)	Hypo-pharynx (13)	Larynx (40)	Total (100)	Controls (100)
A*0101	0 (0)	0 (0)	0 (0)	0 (0)	0 (0)	0	0
A*0201	45.5 (5)	18.2 (4)	21.4 (4)	7.7 (1)	12.5 (5)	18.0	21.0
A*0203	0 (0)	0 (0)	0 (0)	0 (0)	0 (0)	0	0
A*0206	0 (0)	27.3 (6)	21.4 (3)	7.7 (1)	27.5 (11)	21.0	20.0
A*0207	0 (0)	0 (0)	14.3 (2)	7.7 (1)	5.0 (2)	5.0	8.0
A*0210	9.1 (1)	0 (0)	0 (0)	0 (0)	2.5 (1)	2.0	0
A*0301	0 (0)	0 (0)	0 (0)	0 (0)	0 (0)	0	1.0
A*1101	0 (0)	4.5 (1)	14.3 (2)	7.7 (1)	30.0 (12)	16.0	20.0
A*1102	0 (0)	0 (0)	0 (0)	0 (0)	0 (0)	0	0
A*2301	0 (0)	0 (0)	0 (0)	0 (0)	0 (0)	0	0
A*2402	72.7 (8)	81.8 (18)*	57.1 (8)	53.8 (7)	35.0 (14)	55.0	54.0
A*2407	0 (0)	0 (0)	0 (0)	0 (0)	0 (0)	0	0
A*2408	0 (0)	0 (0)	0 (0)	0 (0)	0 (0)	0	0
A*2601	0 (0)	22.7 (5)	21.4 (3)	7.7 (1)	20.0 (8)	17.0	12.0
A*2602	9.1 (1)	9.1 (2)	7.1 (1)	0 (0)	7.5 (3)	7.0	4.0
A*2603	0 (0)	0 (0)	7.1 (1)	7.7 (1)	2.5 (1)	3.0	7.0
A*2605	0 (0)	0 (0)	0 (0)	0 (0)	0 (0)	0	0
A*2606	0 (0)	0 (0)	0 (0)	0 (0)	0 (0)	0	0
A*3001	0 (0)	0 (0)	0 (0)	0 (0)	0 (0)	0	1.0
A*3101	18.2 (2)	4.5 (1)	0 (0)	23.1 (3)	2.5 (1)	7.0	12.0
A*3303	27.3 (3)	13.6 (3)	14.3 (2)	15.4 (2)	17.5 (7)	17.0	9.0

Allele	Allele frequency (%)
Allele	Patients	Controls
A*0101	0	0
A*0201	12.8	16.3
A*0203	0	0
A*0206	28.2	20.0
A*0207	5.1	10.0
A*0210	2.6	0
A*0301	0	0
A*1101	28.2	21.1
A*1102	0	0
A*2301	0	0
A*2402	35.9*	56.3
A*2407	0	0
A*2408	0	0
A*2601	20.5	12.5
A*2602	7.7	3.8
A*2603	0	5.0
A*2605	0	0
A*2606	0	0
A*3001	0	1.3
A*3101	2.6	12.5
A*3303	15.4	11.3