Association of the HLA-DRB1 gene locus with gastric adenocarcinoma in Japan

Digestive and Liver Disease 35 (2003) 468–472 www.elsevier.com / locate / dld

Alimentary Tract

Association of the HLA-DRB1 gene locus with gastric adenocarcinoma in Japan M. Ohtani a , T. Azuma a , *, S. Yamazaki a , A. Yamakawa a , Y. Ito a , A. Muramatsu a , M. Dojo b , Y. Yamazaki b , M. Kuriyama a a

Second Department of Internal Medicine, Fukui Medical University, Matsuoka-cho, Yoshida-gun, Fukui 910 -1193, Japan b Department of Endoscopic Medicine, Fukui Medical University, Fukui 910 -1193, Japan Received 18 November 2002; accepted 25 March 2003

Abstract Background and aim. Helicobacter pylori infection is associated with gastric adenocarcinoma, however, the odds ratio is relatively low. The aim of the present study was to investigate host genetic factors that increase the risk of gastric adenocarcinoma among H. pylori-infected individuals. Methods. A total of 70 patients with early gastric adenocarcinoma and 121 unrelated healthy controls were examined for H. pylori infection and HLA-DRB1 genotyping. The frequencies of HLA-DRB1 alleles were compared among groups. Results. The allele frequency of DRB1*04051 was significantly higher in patients with gastric adenocarcinoma (17.9%) than in controls (7.9%) ( pcorrect 50.045). The odds ratio of gastric adenocarcinoma associated with the presence of the HLA-DRB1*04051 allele compared with its absence was 2.55 (95% confidence limits, 1.35–4.83). This genetic risk was not associated with H. pylori infection. There was no significant difference in the HLA-DRB1 allele frequency between H. pylori-positive and H. pylori-negative controls. The frequency of genotypes that possessed the DRB1*04051 allele in gastric adenocarcinoma patients (34.3%) was significantly higher than that in H. pylori-negative controls (11.9%) ( p50.0089) and H. pylori-positive controls (15.2%) ( p50.0066). Conclusion. These findings suggest that immunogenetic factors for susceptibility to gastric adenocarcinoma are present in the host, the HLA-DRB1*04051 allele is a host genetic risk factor for gastric adenocarcinoma, and that this genetic risk is independent of H. pylori infection.  2003 Editrice Gastroenterologica Italiana S.r.l. Published by Elsevier Science Ltd. All rights reserved. Keywords: Gastric adenocarcinoma; Helicobacter pylori infection; HLA-DRB1 genotyping

1. Introduction Gastric adenocarcinoma is one of the most frequent malignant diseases in the world, particularly in Japan [1,2]. Several prospective studies have reported that H. pylori infection is strongly associated with gastric adenocarcinoma [3–5]. Infection with H. pylori was classified as a class 1 carcinogen by the International Agency for Research on Cancer (IARC) [6]. However, the mechanisms of this association are still unclear. Although infection with H. pylori is significantly associated with gastric adenocarcinoma, the odds ratios are relatively low, ranging from 2.7 to 6.0 [3,4]. Most people infected with H. pylori never develop gastric adenocarcinoma. Therefore, the additional *Corresponding author. Fax: 181-776-61-8110. E-mail address: [email protected] (T. Azuma).

factors that increase the risk of gastric adenocarcinoma among individuals infected with H. pylori need to be identified. Investigation of host factors with respect to the pathogenesis of H. pylori has been recommended [7,8]. In animal models, genetic differences in the host response to infection by Helicobacter species result in varying intensities of inflammation and degrees of gastric epithelial erosion [9,10]. Immunogenetic factors might influence the severity of inflammation, and thereby different clinical outcomes. Many immune responses are controlled by genes of the major histocompatibility complex (MHC) which encodes human leukocyte antigens (HLA). HLA class 2 genes are a crucial genetic factor in initiating or regulating immune response by presenting foreign or selfantigens to T lymphocytes [11,12]. The HLA polymorphism and susceptibility or resistance for diseases have been reported in infectious and autoimmune diseases

1590-8658 / 03 / $30  2003 Editrice Gastroenterologica Italiana S.r.l. Published by Elsevier Science Ltd. All rights reserved. doi:10.1016 / S1590-8658(03)00218-4

M. Ohtani et al. / Digestive and Liver Disease 35 (2003) 468–472

[13,14]. Previous studies also showed the association of HLA polymorphism and human cancers, such as malignant melanoma [15], cervical cancer [16], and nasopharyngeal cancer [17]. Class 2 HLA genes may affect the development of gastric adenocarcinoma. It has been reported that HLA-DR antigens are expressed in gastric epithelial cells in H. pylori-positive patients with chronic gastritis [18]. We assume that differences in HLA-DR antigens may influence the development of gastric adenocarcinoma in H. pylori-infected individuals. In contrast, CagA is the product of the cagA gene, which is carried in virulent type I strains of H. pylori [19]. The correlation between expression of CagA and H. pylori virulence has been well documented. cagA-positive H. pylori infection is associated with gastric mucosal atrophy and gastric cancer [20,21]. CagA plays important roles in the pathogenicity of H. pylori infection. Therefore, we examined the HLA-DR genotypes of patients with gastric adenocarcinoma as a host factor, and the cagA status of H. pylori as a bacterial factor, to investigate the risk of gastric adenocarcinoma induced by H. pylori infection in the present study.

2. Materials and methods

2.1. Subjects A total of 70 patients with early gastric adenocarcinoma (40 males and 30 females; mean age, 64.7 years) were studied. All patients were referred for investigation to the Second Department of Internal Medicine, Fukui Medical University, and subjected to the endoscopic mucosal resection or gastrectomy. Unrelated healthy control subjects (121—53 males and 68 females; mean age, 59.5 years) were taken at random from subjects who visited multiphasic health testing services held by the Second Department of Internal Medicine, Fukui Medical University. The services included endoscopic examination to screen for gastric adenocarcinoma. All subjects were Japanese and were living in Fukui Prefecture. These studies were performed according to the principles of the Declaration of Helsinki, and consent was obtained from each individual after a full description of the nature and protocol of the study.

469

leukocytes by the SDS-proteinase K digestion / phenol– chloroform extraction method. HLA-DRB1 genotyping was performed by the polymerase chain reaction (PCR) direct sequencing method with a BigDye terminator sequencing-based typing kit (Applied Biosystems, Foster City, CA, USA). The low resolution sequence-specific primer PCR (SSP-PCR) amplifications were based on 11 allele group-specific motifs in the first hyper variable region of the second exon of the HLA-DRB1 gene. For the high resolution, the positive amplification reactions from the low resolution SSP-PCR were reamplified as sequencing templates and determined the alleles automatically according to the procedure provided by the manufacturer.

2.4. H. pylori culture Gastric biopsy specimens from each patient were inoculated onto a trypticase soy agar (TSA)-II–5% sheep blood plate and cultured under microaerobic conditions (O 2 , 5%; CO 2 , 15%; N 2 , 80%) at 37 o C for 5 days. H. pylori was identified as a Gram-negative spiral bacilli with urease activity. A single colony was picked from each primary culture plate, inoculated onto a fresh TSA-II plate, and cultured under the conditions described above. A few colonies were picked from each plate and transferred into 20 ml of Brucella broth liquid culture medium containing 10% foetal calf serum, and cultured for 3 days under the conditions described above. DNA from each H. pylori isolate was extracted from the pellet of the liquid culture sample by the protease / phenol–chloroform method, suspended in 300 ml of a TE buffer (10 mM Tris–HCl, 1 mM EDTA) and stored at 4 8C until PCR amplification.

2.5. PCR analysis of the cagA gene We examined the cagA gene by carrying out PCR among isolates. The 39 region of cagA was amplified by PCR using the following primers. Forward primer: 59GAATTGTCTGATAAACTTGAAA, reverse primer: 59GCGTATGTGGCTGTTAGTAGCG. PCR conditions were as follows: heating at 94 8C for 5 min, followed by 25 cycles consisting of 94 8C for 30 s, 55 8C for 30 s, and 72 8C for 3 min. The tubes were kept at 72 8C for 10 min before storage at 4 8C. PCR products were examined using 2% agarose gel electrophoresis. The gels were stained with ethidium bromide to detect the cagA gene.

2.2. Diagnosis of H. pylori infection 2.6. Statistical analysis H. pylori infection was determined by culture, histology, and rapid urease test. Subjects with H. pylori identified by more than one test were defined as H. pylori-positive, whereas those who did not have H. pylori in any of the tests were defined as negative.

2.3. HLA-DRB1 typing Genomic DNA was extracted from peripheral blood

The frequencies of the HLA-DRB1 allele and genotype were compared among groups by the chi-square ( x 2 ) test or Fisher’s exact tests. Bonferroni’s correction was applied for multiple comparisons. Corrections for multiple comparisons were made for HLA-DRB1*0101, 15011, 15021, 1602, 04051, 0406, 11011, 1201, 13021, 14011, 1403, 1405, 08021, 08032, and 09012. The remaining alleles that were found to be infrequent (less than four in controls)

M. Ohtani et al. / Digestive and Liver Disease 35 (2003) 468–472

470

were not individually analysed; pcorrect values ,0.05 were considered significant. The logit estimate of the common odds ratio was calculated.

3. Results

3.1. Frequency of HLA-DRB1 genotypes

Table 1 Allele frequencies of HLA-DRB1 in gastric adenocarcinoma patients and controls Controls (%)

Adenocarcinoma (%)

0101 03011 03021 15011 15021 1602 04011 04031 04051 0406 0407 0410 0701 10011 11011 1201 12021 13011 13021 14011 1403 1404 1405 1406 1412 08021 08032 09012

25 (10.3) 0 (0.0) 1 (0.0) 13 (5.4) 23 (9.5) 4 (1.7) 1 (0.0) 3 (1.2) 19 (7.9) 8 (3.3) 1 (0.0) 3 (1.2) 3 (1.2) 2 (0.8) 6 (2.5) 12 (5.0) 1 (0.0) 3 (1.2) 29 (12.0) 6 (2.5) 7 (2.9) 1 (0.0) 5 (2.1) 1 (0.0) 1 (0.0) 8 (3.3) 24 (9.9) 32 (13.2)

11 (7.9) 1 (0.7) 0 (0.0) 7 (5.0) 23 (16.4) 0 (0.0) 0 (0.0) 2 (1.4) 25 (17.9) 3 (2.1) 0 (0.0) 2 (1.4) 0 (0.0) 0 (0.0) 2 (1.4) 4 (2.9) 4 (2.9) 0 (0.0) 12 (8.6) 6 (4.3) 4 (2.9) 0 (0.0) 2 (1.4) 4 (2.9) 0 (0.0) 1 (0.7) 7 (5.0) 20 (14.3)

Total

242

140

Subject

H. pylori status

No.

Genotypes with DRB1*04051 n (%)

Controls

(2) (1)

42 79

5 (11.9) 12 (15.2)

Total

121

17 (14.0)

(1)

70

24 (34.3)[

–

The frequencies of the HLA-DRB1 allele in gastric adenocarcinoma patients and healthy controls are shown in Table 1. The allele frequency of DRB1*04051 was significantly higher in patients with gastric adenocarcinoma (17.9%) than in controls (7.9%) ( pcorrect 50.045 after Bonferroni’s correction). In order to gain a better appreciation of the genotypic risk, the odds ratio of gastric adenocarcinoma associated with each genotype was estimated. The frequency of genotypes that possessed the DRB1*04051 allele (heterozygous or homozygous of DRB1*04051) in gastric adenocarcinoma patients was significantly higher (24 / 70, 34.3%) than in healthy controls (17 / 121, 14.0%) ( p5 0.0012). The odds ratio of gastric adenocarcinoma associated with the presence of the HLA-DRB1*04051 allele

Allele

Table 2 Frequency of genotypes possessing HLA-DRB1*04051 (heterozygous or homozygous of DRB1*04051) in H. pylori-negative controls, H. pyloripositive controls, and gastric adenocarcinoma patients

p

OR

95% CI

0.63

1.87

1.01–3.48

0.04

2.55

1.35–4.83

Gastric adenocarcinoma [

, Significantly different from H. pylori (2) controls ( p50.0089), and H. pylori (1) controls ( p50.0066).

compared with its absence was 2.55 (95% confidence limits, 1.35–4.83).

3.2. Relationship between HLA-DRB1 allele frequency and H. pylori infection All patients with gastric adenocarcinoma were H. pyloripositive. A total of 42 of 121 controls were H. pylorinegative and 79 were H. pylori-positive. Gastric adenocarcinoma was significantly associated with H. pylori infection ( p,0.0001). There was no significant difference in the HLA-DRB1 allele frequency between H. pyloripositive and H. pylori-negative controls. The frequency of genotypes that possessed the DRB1*04051 allele (heterozygous or homozygous of DRB1*04051) in gastric adenocarcinoma patients was significantly higher (24 / 70, 34.3%) than that in H. pylorinegative controls (5 / 42, 11.9%) ( p50.0089) and H. pylori-positive controls (12 / 79, 15.2%) ( p50.0066) (Table 2).

3.3. Frequency of HLA-DRB1 genotypes according to histological classification Patients with gastric adenocarcinoma were classified histopathologically. A total of 45 patients had the intestinal type adenocarcinoma, and 25 patients had the diffuse type adenocarcinoma. The genotype frequency of the DRB1*04051 allele (heterozygous or homozygous of DRB1*04051) was not different between intestinal type and diffuse type gastric adenocarcinoma (Table 3). Table 3 Frequency of genotypes possessing HLA-DRB1*04051 (heterozygous or homozygous of DRB1*04051) in gastric adenocarcinoma patients according to histopathological classification

1.51 1.26

0.21 0.48

0.03–1.70 0.20–1.14 Intestinal type Diffuse type

No.

DRB1*04051 (1) genotype n (%)

45 25

15 (33.3) 9 (36.0)

M. Ohtani et al. / Digestive and Liver Disease 35 (2003) 468–472

3.4. cagA gene status of H. pylori H. pylori isolates were obtained from all H. pyloripositive subjects. Almost all H. pylori isolates were cagApositive (148 / 149, 99.3%). One strain from a control case was cagA-negative.

4. Discussion The present data indicate an association between the HLA-DRB1 genotype and susceptibility to gastric adenocarcinoma. The frequency of the genotype which possessed HLA-DRB1*04051 was significantly higher in patients with gastric adenocarcinoma than in healthy controls. The risk of gastric adenocarcinoma may increase in people who possess the HLA-DRB1*04051 allele. The odds ratio of gastric adenocarcinoma associated with the presence of the HLA-DRB1*04051 allele compared with its absence was 2.55. This association was observed both in intestinal and diffuse type gastric adenocarcinoma. Magnusson et al. examined the association between HLADR and gastric adenocarcinoma in a Swedish population. They indicated that the DRB1*1601 allele was associated with gastric adenocarcinoma with an odds ratio of 8.7, and that the association was stronger with the diffuse, rather than with the intestinal, histological type of gastric cancer [22]. However, the frequency of the DRB1*1601 allele was low; 1.0% in the controls and 7.2% in patients with gastric adenocarcinoma. In this study, the DRB1*1601 allele was not detected in the study population. The frequency of the HLA allele is different among populations [23]. Due to racial differences and genetic background, characteristic genetic risk factors may be present in each race. These findings suggest that the genotype which possesses the DRB*04051 allele is a risk factor for gastric adenocarcinoma in the Japanese population. This genetic risk was not associated with H. pylori infection, although all gastric adenocaricnoma patients were H. pylori-positive in the present study. There was no significant difference in the HLA-DRB1 allele frequency between H. pylori-positive and H. pylori-negative controls. The frequency of genotypes that possessed the DRB1*04051 allele in gastric adenocarcinoma patients was significantly higher than that in H. pylori-negative controls and H. pylori-positive controls. Therefore, the genetic risk which possesses the DRB1*04051 allele is independent from H. pylori infection. The mechanism of association between the HLADRB1*04051 allele and gastric adenocarcinoma was not clear. There are two potential explanations for the role of the HLA allele in the outcome of infection with H. pylori. First, the HLA-DRB1*04051 allele itself may be one of the genes responsible for the association. Recently, polymorphisms in the interleukin 1 gene cluster have been proposed to modulate the risk of gastric cancer, emphasis-

471

ing that inherited variation in components of the immune system may be critical for gastric adenocarcinoma development [24]. HLA class 2 genes are a group of highly polymorphic genes on chromosome 6p that are particularly important in controlling specific immune recognition. HLA antigens bind peptide antigens, which are recognised by T cells through the formation of a trimolecular complex between the HLA antigen, the peptide antigen, and the antigen-specific T-cell receptor on the T cell. The formation of this trimolecular complex is the fundamental antigen specific signaling event in the anti-tumour immune response [25–27]. Individuals with different HLA types differ in their immune response. Allele-specific binding of antigenic peptides may account for associations between HLA class 2 genes and disease susceptibility. Second, there could be linkage disequilibrium; HLA-DRB1*04051 itself is not responsible for the pathophysiologic mechanism, but is linked closely to one of the responsible genes. HLA-DQA1, DQB1, a transporter associated with antigen processing 2 (TAP2), and tumour necrosis factor genes have been reported to be in linkage disequilibrium [28– 30]. Additional molecular biologic analysis using polymorphic markers in this region will be necessary. It has been shown some bacterial factors affect the outcome of H. pylori infection. Blaser et al. reported that infection with a cagA-positive H. pylori strain in comparison with a cagA-negative strain increased the risk of gastric adenocarcinoma development [31]. The strain diversity of H. pylori has been noted worldwide. In Asia, the percentage with a cagA-positive strain is almost 100%, but in Western countries the percentage is only about 60% [20,32]. In this study, 99.3% of H. pylori strains were cagA-positive, regardless of clinical outcome. Therefore, we have not to consider the cagA status of H. pylori to investigate host genetic factors for gastric adenocarcinoma induced by H. pylori infection in our population. In conclusion, we found that a genotype with the HLADRB1*04051 allele is a host genetic risk factor for gastric adenocarcinoma in Japan. However, the odds ratio of this genetic risk remains relatively low (2.55). Further assessments of the genetic risks of gastric adenocarcinoma are necessary.

Conflict of interest statement None declared.

List of abbreviations HLA, human leukocyte antigens; MHC, major histocompatibility complex.

472

M. Ohtani et al. / Digestive and Liver Disease 35 (2003) 468–472

Acknowledgements This work was supported by a grant-in-aid for Scientific Research on Priority Areas (C) from the Ministry of Education, Culture, Sports, Science and Technology, Japan, and a grant-in-aid for Scientific Research (C) from Japan Society for the Promotion of Science.

[17]

[18]

[19]

References [1] Correa P, Haenszel W. Epidemiology of gastric cancer. In: Correa P, Haenszel W, editors, Epidemiology of Cancer of the Digestive Tract, The Hague: M. Nijhoff, 1982, pp. 58–84. [2] Correa P. The epidemiology of gastric cancer. World J Surg 1991;15:228–34. [3] Nomura A, Stemmermann GN, Chyou P, Kato I, Perez-Perez GI, Blasert MJ. Helicobacter pylori infection and gastric carcinoma among Japanese Americans in Hawaii. New Engl J Med 1991;325:1132–6. [4] Parsonnet J, Friedman GD, Vandersteen DP, Chang Y, Vogelman JH, Orentreich N, et al. Helicobacter pylori infection and risk of gastric carcinoma. New Engl J Med 1991;325:1127–31. [5] Uemura N, Okamoto S, Yamamoto S, Matsumura N, Yamaguchi S, Yamakido M, et al. Helicobacter pylori infection and the development of gastric cancer. New Engl J Med 2001;345:784–9. [6] IARC Schistosomes, Liver flukes and Helicobacter pylori. Monographs on the evaluation of carcinogenic risks to humans. IARC Sci Publ, 1994;61:1–241. [7] Lee A. Helicobacter pylori-initiated ulcerogenesis: look to the host. Lancet 1993;341:280–1. [8] Azuma T, Konishi J, Tanaka Y, Hirai M, Ito S, Kato T, et al. Contribution of HLA-DQA gene to host’s response against Helicobacter pylori. Lancet 1994;343:542–3. [9] Mohammadi M, Redline R, Nedrud J, Czinn S. Role of the host in pathogenesis of Helicobacter-associated gastritis: H. felis infection of inbred and congenic mouse strains. Infect Immun 1996;64:238– 45. [10] Sakagami T, Dixon M, O’Rourke J, Howlett R, Alderuccio F, Vella J, et al. Atrophic gastric changes in both Helicobacter felis and Helicobacter pylori infected mice are host dependent and separate from antral gastritis. Gut 1996;39:639–48. [11] Ceppellini R, Frumento G, Ferrara GB, Tosi R, Chersi A, Pernis B. Binding of labelled influenza matrix peptide to HLADR in living B lymphoid cells. Nature 1989;339:392–4. [12] Topalian SL, Rivoltini L, Mancini M, Markus NR, Robbins PF, Kawakami Y, et al. Human CD41 T cells specifically recognise a shared melanoma-associated antigen encoded by the tyrosinase gene. Proc Natl Acad Sci USA 1994;91:9461–5. [13] Todd JA, Bell JI, McDevitt HO. HLA-DQ b gene contributes to susceptibility and resistance to insulin-dependent diabetes mellitus. Nature 1987;329:599–604. [14] Todd JA, Acha-Orbea H, Bell JI, Chao N, Fronek Z, Jacob CO, et al. A molecular basis for MHC class II-associated autoimmunity. Science 1988;240:1003–9. [15] Lee JE, Reveille JD, Ross MI, Platsoucas CD. HLA-DQB1*0301 association with increased cutaneous melanoma risk. Int J Cancer 1994;59:510–3. [16] Apple RJ, Erlich HA, Klitz W, Manos MM, Becker TM, Wheeler

[20]

[21]

[22]

[23]

[24]

[25]

[26]

[27]

[28]

[29]

[30]

[31]

[32]

CMHLA. DR-DQ associations with cervical carcinoma show papillomavirus-type specificity. Nature Genet 1994;6:157–62. Burt RD, Vaughan TL, McKnight B, Davis S, Beckmann AM, Smith AG, et al. Associations between human leukocyte antigen type and nasopharyngeal carcinoma in Caucasians in the United States. Cancer Epidemiol Biomarkers Prev 1996;5:879–87. Wee A, Teh M, Kang JY. Association of Helicobacter pylori with HLA-DR antigen expression in gastritis. J Clin Pathol 1992;45:30– 3. Censini S, Lange C, Xiang Z, Crabtree JE, Ghiara P, Borodovsky M, et al. Cag, a pathogenicity island of Helicobacter pylori, encodes type I-specific and disease-associated virulence factors. Proc Natl Acad Sci USA 1996;93:14648–53. Blaser MJ, Perez-Perez GI, Kleanthous H, Cover TL, Peek RM, Chyou PH, et al. Infection with Helicobcter pylori strains possessing cagA is associated with an increased risk of developing adenocarcinoma of the stomach. Cancer Res 1995;55:2111–5. Parsonnet J, Friedman GD, Orentreich N, Vogelman H. Risk for gastric cancer in people with CagA positive or CagA negative Helicobacter pylori infection. Gut 1997;40:297–301. Magnusson PKE, Enroth H, Eriksson I, Held M, Nyren O, Engstrand L, et al. Gastric cancer and human leukocyte antigen: distinct DQ and DR alleles are associated with development of gastric cancer and infection by Helicobacter pylori. Cancer Res 2001;61:2684–9. Saito S, Ota S, Yamada E, Inoko H, Ota M. Allele frequencies and haplotypic associations defined by allelic DNA typing at HLA class I and class II loci in the Japanese population. Tissue Antigens 2000;56:522–9. El-Omar EM, Carrington M, Chow WH, McColl KE, Bream JH, Young HA, et al. Interleukin-1 polymorphisms associated with increased risk of gastric cancer. Nature 2000;404:398–402. Pullen AM, Marrack P, Kappler JW. The T-cell repertoire is heavily influenced by tolerance to polymorphic self-antigens. Nature 1988;335:796–801. Brown JH, Jardetzky TS, Gorga JC, Stern LJ, Urban RG, Strominger JL, et al. Three-dimensional structure of the human class II histocompatibility antigen HLA-DR1. Nature 1993;364:33–9. Silver ML, Guo HC, Strominger JL, Wiley DC. Atomic structure of a human MHC molecule presenting an influenza virus peptide. Nature 1992;360:367–9. Begovich AB, McClure GR, Suraj VC, Helmuth RC, Fildes N, Bugawan TL, et al. Polymorphism, recombination, and linkage disequilibrium within the HLA class II region. J Immunol 1992;148:249–58. Carrington M, Colonna M, Spies T, Stephens JC, Mann DL. Haplotypic variation of the transporter associated with antigen processing (TAP) genes and their extension of HLA class II region haplotypes. Immunogenetics 1993;37:266–73. McGuire W, Hill AV, Allsopp CE, Greenwood BM, Kwiatkowski D. Variation in the TNF-a promoter region associated with susceptibility to cerebral malaria. Nature 1994;371:508–10. Blaser MJ, Perez-Perez GI, Kleanthous H, Cover TL, Peek RM, Chyou PH, et al. Infection with Helicobacter pylori strains possessing cagA is associated with an increased risk of developing adenocarcinoma of the stomach. Cancer Res 1995;55:2111–5. Ito Y, Azuma T, Ito S, Miyaji H, Hirai M, Yamazaki Y, et al. Analysis and typing of the vacA gene from cagA-positive strains of Helicobacter pylori isolated in Japan. J Clin Microbiol 1997;35:1710–4.