Interleukin 1B and Interleukin 1RN Polymorphisms Are Associated With Increased Risk of Gastric Carcinoma

Transcription

1 GASTROENTEROLOGY 2001;121: Interleukin 1B and Interleukin 1RN Polymorphisms Are Associated With Increased Risk of Gastric Carcinoma JOSÉ CARLOS MACHADO,* PAUL PHAROAH,, SÓNIA SOUSA,* RALPH CARVALHO,* CARLA OLIVEIRA,*, CÉU FIGUEIREDO,* ANTÓNIO AMIM,*, RAQUEL SERUCA,* CARLOS CALDAS, FÁTIMA CARNEIRO,*, and MANUEL SOBRINHO SIMÕES*, *Institute of Molecular Pathology and Immunology of the University of Porto, Porto, Portugal; Departments of Oncology and Public Health, University of Cambridge, Cambridge, England; Faculty of Sciences, Porto, Portugal; and Faculty of Medicine, Hospital S. João, Porto, Portugal See editorial on page Background & Aims: Interleukin (IL)-1 gene cluster proinflammatory polymorphisms have been associated with development of gastric atrophy and with increased risk of gastric carcinoma. We aimed to determine the association between IL-1 loci polymorphisms and increased risk of gastric carcinoma in samples from a Portuguese population, and to find whether there was any relationship with the histologic types of gastric carcinoma. Methods: In a case-control study including 220 controls and 152 patients with gastric carcinoma (intestinal, 76; diffuse, 37; and atypical, 39), both the IL-1B 511 biallelic polymorphism and the IL-1RN penta-allelic variable number of tandem repeats were genotyped. Results: We found a significant association between the IL-1 polymorphisms and increased risk for tumor development in patients with intestinal-type gastric carcinoma. A trend towards an increased risk of tumor development was also observed in patients with diffusetype gastric carcinoma. No significant relationship was observed in patients with atypical carcinoma. Carriers of IL-1B-511T and IL-1RN*2 homozygotes had increased risk for developing intestinal-type gastric carcinoma with odds ratios of 2.7 (95% confidence interval [CI ], ) and 3.1 (95% CI, ), respectively. Statistical analysis showed an interaction between the 2 loci with the risk conferred by the IL-1B-511T allele substantially increased (odds ratio, 9.0; 95% CI, ) in individuals homozygous for the IL-1RN*2 allele. Conclusions: Our results provide further support to the association between IL-1 gene cluster proinflammatory polymorphisms and increased risk of gastric carcinoma. Furthermore, we found evidence pointing to the existence of a synergistic interaction between the IL-1B and IL-1RN polymorphisms. Patients infected with Helicobacter pylori, a bacteria that colonizes the human stomach, are at increased risk for developing gastric carcinoma. 1 4 The first consequence of H. pylori infection is the development of neutrophilic gastritis, which progresses to active chronic gastritis in most people. 5 Persistent inflammation, possibly intensified via the inflammatory cytokine cascade and the generation of H. pylori specific T- and B-cell immune responses, 6 eventually leads to gastric atrophy, hypochlorhydria, and increased risk for developing gastric carcinoma. 2,3,7 9 Interleukin (IL)-1 is a potent proinflammatory cytokine 10 that is up-regulated in the presence of H. pylori and is important in initiating and amplifying the inflammatory response to this infection. 11,12 The IL-1 receptor antagonist (IL-1ra) is a naturally occurring anti-inflammatory cytokine that competitively binds to IL-1 receptors, thus modulating the potentially injurious effects of IL Recently, El-Omar et al. 14 reported that IL-1B (encoding IL-1 ) and IL-1RN (encoding IL-1ra) gene polymorphisms (IL-1B-31C, IL-1B-511T, and IL-1RN*2 alleles) are associated with an increased risk of both hypochlorhydria, in H. pylori infected first-degree relatives of gastric carcinoma cases, and gastric carcinoma. Although the molecular mechanisms involved in the inter-regulation of the IL-1 loci are still unclear, the aforementioned alleles are associated with increased IL-1 production According to this model, individuals with proinflammatory genotypes overexpress gastric IL-1 in response to H. pylori infection, leading to increased inflammation, gastric atrophy, and hypochlorhydria, and possibly to the development of gastric carcinoma. 14 Abbreviations used in this paper: CI, confidence interval; IL, interleukin; IL-1ra, interleukin 1 receptor antagonist; LD, linkage disequilibrium;, odds ratio; PCR, polymerase chain reaction; SSCP, singlestrand conformation polymorphism; VNTR, variable number of tandem repeats by the American Gastroenterological Association /01/$35.00 doi: /gast

2 824 MACHADO ET AL. GASTROENTEROLOGY Vol. 121, No. 4 It is generally acknowledged that 2 main pathways operate in gastric carcinogenesis, 1 leading to intestinaltype gastric carcinoma and the other leading to diffusetype gastric carcinoma. 7 9,20,21 Although both pathways seem to start from H. pylori derived chronic gastritis, epidemiologic and histopathological evidence have shown that gastric atrophy is mainly associated with intestinal-type carcinoma. 7 9 Hence we hypothesize that the observed association between IL-1 loci polymorphisms and increased risk for gastric carcinoma may be influenced by the histologic type of the cases. The present study had two main goals. The first was to determine the association between IL-1 loci polymorphisms and risk for gastric carcinoma in a population of different geographical backgrounds. El-Omar et al. 14 studied a Polish population, whereas in the present study samples from a Portuguese population were analyzed. The second goal was to elucidate whether or not the putative association between IL-1 loci polymorphisms and increased risk of gastric carcinoma varies according to the histologic type of the tumors. Materials and Methods Tissue Material We evaluated, in a case-control study, the IL-1B-511 biallelic and IL-1RN penta-allelic variable number of tandem repeats (VNTR) polymorphisms. The study was performed in a control group of blood donors (n 220) and in a series of patients with gastric carcinoma (n 152). Both sample groups were collected in northern Portugal. In the control group, the average age was years (range, years) and the male:female ratio was 1.6:1. In the gastric carcinoma group, the average age was years (range, years) and the male:female ratio was 1.7:1. DNA from the control group was isolated from blood samples and collected after obtaining informed consent. DNA from patients with gastric carcinoma was isolated from frozen samples of non-neoplastic gastric mucosa. In a subset of cases, IL-1 genotypes were confirmed in blood samples. Gastric carcinomas were resected and diagnosed at Hospital S. João, Medical Faculty and IPATIMUP, Porto, Portugal. According to Laurén s classification, 20 cases were classified as intestinal (n 76), diffuse (n 37), and atypical (n 39) carcinomas. All of the histopathological slides were reviewed by 2 pathologists (F.C. and M.S.-S.). IL-1B-511 and IL-1RN VNTR Genotyping Genomic DNA was retrieved from the available material using standard phenol/chloroform extraction. The IL- 1B-511C/T biallelic polymorphism was genotyped by cold polymerase chain reaction single-strand conformation polymorphism (PCR-SSCP) analysis and the IL-1RN penta-allelic VNTR by PCR-standard agarose gel electrophoresis. Primer sequences and PCR conditions were based on those reported by El-Omar et al. 14 PCR-negative controls were performed by replacing the template DNA with water. For IL-1B-511 SSCP analysis, PCR reaction products were diluted 1:1 with loading buffer (95% formamide, 0.05% bromophenol blue, and 0.05% xylene cyanol), denatured at 99 C for 2 minutes, and cooled on ice for 5 minutes. Electrophoresis of the denatured PCR products was carried out in nondenaturing 0.8 mutation detection enhancement gels (BMA, Rockland, ME) and run at 2W, 20 C for 15 hours. PCR-SSCP products were visualized by standard DNA silver staining. To interpret the SSCP banding pattern, the bands were recovered from the MDE gels and submitted to cold PCR reamplification with the original set of primers. Reamplification products were purified and sequenced using the ABI Prism Dye Terminator Cycle Sequencing kit (Perkin-Elmer, Foster City, CA) and an ABI Prism 377 DNA Sequencer (Perkin-Elmer). Sequencing was performed on both strands using the original primers. For IL-1RN VNTR analysis, the PCR products were separated by electrophoresis on 2% agarose gels and stained with ethidium bromide. The IL-1RN gene has a penta-allelic 86 base pair tandem repeat in the second intron, 22 and the alleles were coded conventionally as follows: allele 1, 4 repeats; allele 2, 2 repeats; allele 3, 5 repeats; allele 4, 3 repeats; and allele 5, 6 repeats. For the purpose of statistical analysis and because of the rarity of alleles 3, 4, and 5, this polymorphism was treated as biallelic by dividing alleles into short and long categories, in which the short allele has 2 repeats (allele 2) and the long allele has 3 or more repeats (alleles 1, 3, 4, and 5). Statistical Analysis Hardy Weinberg equilibrium of alleles at individual loci was assessed by exact tests using the program GENEPOP (available from ftp://ftp.cefe.cnrs-mop.fr/pub/pc/msdos/genepop). Comparison of genotype frequencies between cases and controls was assessed by 2 statistics. The observed haplotype frequencies were estimated and compared with those expected under no association by maximum likelihood using the Estimating Haplotype frequencies (EH) software program (available from ftp://linkage.rockefeller.edu/software/eh/). An overall linkage disequilibrium coefficient was calculated from the formula: i D j i 1 j 1 p i q j D ij where p i is the ith allele frequency at locus p and q j the jth locus frequency at locus j, and D ij D D max

3 October 2001 IL-1 POLYMPHISMS AND GASTRIC CARCINOMA 825 formula p e (-1)/(p e (-1) 1), where p e stands for proportion of exposed individuals among the controls. Differences were considered to be significant at P Figure 1. Genotyping of the IL-1B-511C/T and IL-1RN VNTR polymorphisms. (A) PCR/SSCP analysis of the IL-1B-511C/T polymorphism showing the C/T, C/C, and T/T genotypes. (B) PCR/agarose gel electrophoresis analysis of the IL-1RN VNTR polymorphism illustrating the most common genotypes in our series (M, size marker; C, negative control). where D x ij p ij where x ij is the observed haplotype frequency, p i is the expected haplotype frequency, and D max minimum[p i q i, (1-p i )(1-q i )] if D 0 and D max minimum[p i (1-q i ), p i (1-q i )] if D 0. Odds ratios (s) with 95% confidence intervals (CIs) and unconditional logistic regression models were computed using the SPSS software program (SPSS Science, Chicago, IL). Population attributable fractions were calculated according to the Results IL-1B-511C/T In the control population, the IL-1B-511C/T genotypes (Figure 1A) showed no evidence of deviation from Hardy Weinberg equilibrium (P 0.1; Table 1). Comparison of genotype frequencies showed a significant difference between gastric carcinoma cases and controls ( , P 0.01; Table 1). IL-1B-511 heterozygotes had a significantly increased gastric cancer risk (, 2.0; Table 1). In contrast, no significant increase in gastric carcinoma risk was observed for IL-1B-511T homozygotes (Table 1). However, the number of cases with this genotype was small, and the homozygote risk estimate has wide confidence limits. The pattern of risks suggests a dominant effect, and the risk of gastric carcinoma for IL-1B-511T carriers (both IL-1B-511T homozygotes and IL-1B-511 heterozygotes) was significantly increased with an of 1.7 (Table 1). When cases were divided according to the histologic type of the tumors, a significant difference in genotype frequency was observed only between intestinal-type cases and controls ( , P 0.002; Table 1). In this subgroup of patients, risk was increased for both heterozygotes and IL-1B-511T homozygotes, although the homozygote risk was nonsignificant (Table 1). The risk of tumor development for IL-1B-511T carriers was increased with an of 2.7 (Table 1). A modest increase in risk of diffuse type carcinoma was observed for IL-1B- 511T allele carriers, but this was not statistically significant (Table 1). Genotype frequencies in atypical cases were similar to controls, and genotypic-specific risks were not elevated in this subgroup of patients (Table 1). Table 1. Comparison Between IL-1B-511 Genotypic Frequencies in Controls and Gastric Carcinoma According to the Histologic Type of the Tumors Genotype Controls (N 218) (N 152) Gastric carcinoma All Intestinal Diffuse Atypical (N 76) (N 37) (N 39) C/C C/T ( ) ( ) ( ) ( ) T/T ( ) ( ) ( ) ( ) T carriers ( ) ( ) ( ) ( )

4 826 MACHADO ET AL. GASTROENTEROLOGY Vol. 121, No. 4 Table 2. IL-1RN VNTR Genotypic Frequencies in Controls and Gastric Carcinoma Genotype Controls (N 220) (N 152) 1/ / / / / / / /4 0 1 IL-1RN VNTR Genotype frequencies of the IL-1RN polymorphism (Figure 1B) in the control group did not deviate significantly from those expected under Hardy Weinberg equilibrium (P 0.4; Table 2). There was no significant difference in genotype frequency between cases and controls ( , P 0.1; Table 2). After reclassifying the alleles into long and short, no significant associations were observed between IL-1RN genotype and the risk of carcinoma development (Table 3). The for 2/L heterozygotes was 0.8 and for 2/2 homozygotes 1.7 (Table 3). However, when the cases were divided according to the histologic type of carcinomas, there was a significant difference in genotype frequencies that was restricted to intestinal-type cases ( , P 0.005; Table 3). Individuals homozygous for the IL-1RN*2 allele were found to have an increased risk of tumor development (, 3.1; Table 3). No significant increase in IL-1RN genotypic-specific risks were seen in the subgroups of patients with diffuse and atypical gastric carcinoma (Table 3). Table 4. Observed and Expected IL-1B-511/IL-1RN Haplotype Frequencies in the Control Population Haplotype Frequency IL-1B-511 IL-1RN Observed Expected C C C C T T T T IL-1B-511C/T and IL-1RN VNTR Interaction In the control population, estimation of observed haplotype frequencies and comparison with those expected under no association (Table 4) showed linkage disequilibrium (LD) between the IL-1B-511 and IL-1RN VNTR loci (D 0.17, , P 0.04). In patients with intestinal-type gastric carcinoma, each polymorphism was individually associated with an increased risk for tumor development. To further investigate whether these risks were independent or a consequence of LD, unconditional logistic regression was performed with both genotypes being tested in the model. The possibility of an interaction between the 2 polymorphisms was assessed by including an interaction term. The IL-1B 511 polymorphism was treated as dominant with 2 genotype categories (C/C homozygotes and T- carriers). IL-1RN was treated as recessive with 2 genotype categories (long allele carriers and allele 2 homozygotes). In the final logistic regression model, only the terms for the IL-1B 511 genotype and the interaction term were statistically significant ( , P and , P 0.001, respectively). This suggests that the risk conferred by the IL-1RN polymorphism is dependent on an interaction with IL-1B-511. Furthermore, the IL-1B-511 polymorphism confers an increased risk of intestinal gastric carcinoma, but that risk is increased in individuals who are also homozygous for the IL-1RN*2 allele. Table 3. Comparison Between IL-1RN VNTR Genotypic Frequencies in Controls and Gastric Carcinoma According to the Histologic Type of the Tumors Genotype Controls (N 220) (N 152) Gastric carcinoma All Intestinal Diffuse Atypical (N 76) (N 37) (N 39) L/L L/ ( ) ( ) ( ) ( ) 2/ ( ) ( ) ( ) ( )

5 October 2001 IL-1 POLYMPHISMS AND GASTRIC CARCINOMA 827 Table 5. IL-1B-511 and IL-1RN VNTR Combined Genotypes in Controls and Intestinal-Type Gastric Carcinoma IL-1B-511 Genotypes IL-1RN Controls C/C L carriers C/C 2/ ( ) T carriers L carriers ( ) T carriers 2/ ( ) Thus, 22% of cases were found to be both IL-1B-511T allele carriers and IL-1RN*2 allele homozygotes, compared with 4.6% of controls; this corresponds to a 9-fold increase in risk of intestinal-type gastric carcinoma for individuals with these genotypes compared with those who are IL-1B-511C homozygotes and IL-1RN long allele carriers (Table 5). Carriers of the IL-1B-511T allele with at least 1 copy of an IL-1RN long allele also had an increased risk of intestinal-type gastric carcinoma with an of 2.0 (Table 5). Calculation of population-attributable fractions showed that the fraction of intestinal-type gastric carcinomas, which is attributable (assuming a causal effect) to the IL-1B-511T/ IL-1RN*2 haplotype, either in its homozygous form or with the IL-1B-511C/IL-1RN*2 haplotype, was 27%. For carriers of the IL-1B-511T allele with at least 1 copy of the IL-1RN*1 allele, the population attributable fraction was calculated to represent 33% of the intestinaltype gastric carcinoma cases. Their combined population attributable fraction was estimated to be 46%. Discussion We assessed the association between the IL-1B- 511C/T and IL-1RN VNTR polymorphisms and the risk of gastric carcinoma in a population from northern Portugal. In Portugal, the incidence and mortality rates of gastric carcinoma are among the highest in the world 23 and in the north of the country it has been shown that H. pylori infects more than 80% of the asymptomatic adult population 24 and more than 95% of relatives of patients with gastric carcinoma. 25 In a population of distinct geographical backgrounds, we confirmed that the proinflammatory alleles IL-1B-511T and IL-1RN*2 are associated with an increased risk of gastric carcinoma. 14 We confirmed also that the IL-1B-511T allele seems to act in a dominant manner, whereas the IL-1RN*2 allele seems to have a recessive effect. According to El-Omar et al., 14 carriers of IL-1B-31C (the results were identical for the IL-1B-511T allele) as well as IL-1RN*2 homozygotes had an increased risk for developing gastric carcinoma at s of 1.9 (95% CI, ) and 3.7 (95% CI, ), respectively. We found a significant association between both IL-1 polymorphisms and increased risk of intestinal-type gastric carcinoma. These results concur with the generally accepted model of gastric carcinogenesis. According to this model, there are 2 main pathways of malignant transformation of gastric mucosa, both starting from H. pylori derived chronic gastritis. One pathway is via gastric atrophy, intestinal metaplasia, and adenomatous dysplasia, leading to intestinal-type carcinomas; the other is via hyperplastic or de novo changes, with or without concurrent nonmetaplastic changes, leading to diffusetype carcinomas. 7 9 Gastric atrophy, the main consequence of the amplified inflammatory response associated with the IL-1 proinflammatory genotypes, is mainly associated with the pathway leading to intestinal-type carcinoma. 7 9 The association between the IL-1B-511T allele and diffuse-type carcinoma did not attain the threshold of statistical significance, and no significant relationship was observed regarding patients with atypical carcinomas. These observations suggest that the association between IL-1 polymorphisms and increased risk of gastric carcinoma may depend on the histologic type of the cases. However, the difference in IL-1B-511T allele carrier frequency between intestinal-type and diffuse-type cases was not statistically significant ( , P 0.12). The lack of a significant association with diffusetype carcinoma may therefore be a type 2 statistical error, which reflects the lack of statistical power to detect modest risks. This possibility has to be considered because H. pylori infection is equally associated with intestinal- and diffuse-type cases, 2 and chronic inflammation is also a feature of diffuse-type tumors. 8,9 We found a significant LD between the 2 IL-1 loci. The association between the alleles at these 2 loci is not specified by El-Omar et al., 14 but for IL-1B-31 and IL-1RN, the investigators found D 0.52, with IL- 1B-31 being very tightly linked to IL-1B-511. We found that LD between IL-1B-511 and IL-1RN was weaker (D 0.17). This difference may be caused by differences in LD between populations. However, substantial differences in LD relationships are not found between European populations, 26 so chance is a more likely explanation. It is unclear whether LD simply reflects genomic clustering of the 2 genes (IL-1B and IL-1RN are located on chromosome 2q14 within a 360- kilobase region 27 ) or, as suggested by El-Omar et al., 14 reflects past selective pressures for or against enhancement of the IL-1 response to environmental challenges.

6 828 MACHADO ET AL. GASTROENTEROLOGY Vol. 121, No. 4 We found that IL-1RN was not an independent predictor of risk in a logistic regression model, which was supported by the finding that IL-1RN*2 homozygotes/ IL-1B-511C homozygotes do not have an increased risk of intestinal-type gastric carcinoma. This differs from the findings of El-Omar et al., 14 who reported an increased risk in IL-1RN*2 homozygotes even if they were also IL-1B-31T or IL-1B-511C homozygotes. However, the number of cases in our series was small, and the CI was wide. Thus we cannot exclude the possibility that the IL-1RN polymorphism has an independent effect, albeit smaller than that of IL-1B-511. We found evidence pointing to an interaction between the 2 aforementioned loci because the risk conferred by the IL-1B-511T allele was substantially increased (, 9.0) in individuals who were also homozygous for the IL-1RN*2 allele. These results suggest a synergistic effect between the IL-1B-511T and IL-1RN*2 alleles, a biologically plausible result given the tight relationship between IL-1 and IL-1ra in the modulation of the inflammatory response. 11,12 The proinflammatory effect of these alleles might be explained by their DNA sequence properties. IL-1B- 31C/T is a TATA-box polymorphism that markedly affects DNA-protein interactions in vitro, hence modulating the expression of IL The IL-1B-511T allele is in near complete LD with IL-1B-31C, 14 and therefore is probably a marker for IL-1B-31C with no direct effect on IL-1B expression. The mechanism underlying the association between the IL-1RN*2 allele and enhanced IL-1 expression is not known, but the IL-1RN VNTR region contains 3 potential protein binding sites suggesting possible functional significance. 22 Although there are conflicting data on this issue, proinflammatory genotypes of the IL-1 loci have been associated with a wide range of chronic inflammatory and autoimmune conditions. 13,18 In summary, our results confirm that IL-1B and IL- 1RN polymorphisms, which enhance IL-1 expression, are associated with an increased risk of gastric carcinoma. The importance of these polymorphisms may vary depending on the histotype of the cases, but larger studies are needed to confirm this. The putative underlying mechanism involves amplification of the host inflammatory response to H. pylori infection leading eventually to gastric atrophy. Synergistic interaction between IL-1B and IL-1RN proinflammatory polymorphisms seems to play a major role in this process as suggested by the greatly increased risk associated with IL-1B-511T carriers/il-1rn*2 homozygotes. The potential impact of these susceptibility polymorphisms in intestinal-type gastric carcinoma development is not neglectable; the population attributable fraction for the risk-associated IL-1B-511/IL-1RN genotypic combinations was estimated to be 46%, which (assuming a causal effect) represents the fraction of intestinal-type gastric carcinoma cases that are caused by these IL-1 alleles. In agreement with El-Omar et al., 14 the present results support the hypothesis that host genetic factors that affect IL-1 may determine why some individuals infected with H. pylori develop gastric carcinoma while others do not. Still, the putative role played by the interaction of host genotypic constitution with different H. pylori strains in gastric carcinogenesis remains to be elucidated. References 1. Blaser MJ. Hypothesis: the changing relationships of Helicobacter pylori and humans: implications for health and disease. J Infect Dis 1999;179: Parsonnet J, Friedman GD, Vandersteen DP, Chang Y, Vogelman JH, Orentreich N, Sibley RK. Helicobacter pylori infection and the risk of gastric carcinoma. N Engl J Med 1991;325: Sipponen P. Gastric cancer: a long-term consequence of Helicobacter pylori infection? Scand J Gastroenterol Suppl 1994;201: Blaser MJ. Helicobacter pylori and gastric diseases. BMJ 1998; 316: Dixon MF. Pathophysiology of Helicobacter pylori infection. Scand J Gastroenterol Suppl 1994;201: Shimoyama T, Crabtree JE. Bacterial factors and immune pathogenesis in Helicobacter pylori infection. Gut 1998;43(suppl 1): S2 S5. 7. Correa P. Human gastric carcinogenesis: a multistep and multifactorial process. First American Cancer Society Award Lecture on Cancer Epidemiology and Prevention. Cancer Res 1992;52: Solcia E, Fiocca R, Luinetti O, Villani L, Padovan L, Calistri D, Ranzani GN, Chiaravalli A, Capella C. Intestinal and diffuse gastric cancers arise in a different background of Helicobacter pylori gastritis through different gene involvement. Am J Surg Pathol 1996;20(suppl 1):S8 S Carneiro F, Seixas M, Sobrinho-Simões M. New elements for an updated classification of the carcinomas of the stomach. Pathol Res Pract 1995;191: Dinarello CA. Biologic basis for interleukin-1 in disease. Blood 1996;87: Noach LA, Bosma NB, Jansen J, Hoek FJ, van Deventer SJ, Tytgat GN. Mucosal tumor necrosis factor-alpha, interleukin-1 beta, and interleukin-8 production in patients with Helicobacter pylori infection. Scand J Gastroenterol 1994;29: Basso D, Scrigner M, Toma A, Navaglia F, Di Mario F, Rugge M, Plebani M. Helicobacter pylori infection enhances mucosal interleukin-1 beta, interleukin-6, and the soluble receptor of interleukin-2. Int J Clin Lab Res 1996;26: Arend WP, Malyak M, Guthridge CJ, Gabay C. Interleukin-1 receptor antagonist: role in biology. Annu Rev Immunol 1998;16: El-Omar EM, Carrington M, Chow WH, McColl KE, Bream JH, Young HA, Herrera J, Lissowska J, Yuan CC, Rothman N, Lanyon G, Martin M, Fraumeni JFJ, Rabkin CS. Interleukin-1 polymor-

7 October 2001 IL-1 POLYMPHISMS AND GASTRIC CARCINOMA 829 phisms associated with increased risk of gastric cancer. Nature 2000;404: (correction published in Nature 2001;412: 99). 15. Pociot F, Molvig J, Wogensen L, Worsaae H, Nerup J. A TaqI polymorphism in the human interleukin-1 beta (IL-1 beta) gene correlates with IL-1 beta secretion in vitro. Eur J Clin Invest 1992;22: Hurme M, Santtila S. IL-1 receptor antagonist (IL-1Ra) plasma levels are co-ordinately regulated by both IL-1Ra and IL-1beta genes. Eur J Immunol 1998;28: Santtila S, Savinainen K, Hurme M. Presence of the IL-1RA allele 2 (IL1RN*2) is associated with enhanced IL-1beta production in vitro. Scand J Immunol 1998;47: Hurme M, Lahdenpohja N, Santtila S. Gene polymorphisms of interleukins 1 and 10 in infectious and autoimmune diseases. Ann Med 1998;30: Danis VA, Millington M, Hyland VJ, Grennan D. Cytokine production by normal human monocytes: inter-subject variation and relationship to an IL-1 receptor antagonist (IL-1Ra) gene polymorphism. Clin Exp Immunol 1995;99: Laurén P. The two histological main types of gastric carcinoma: diffuse and so-called intestinal-type carcinoma. An attempt at a histoclinical classification. Acta Pathol Microbiol Scand 1965;64: Machado JC, Nogueira AMMF, Carneiro F, Reis CA, Sobrinho- Simões M. Gastric carcinoma exhibits distinct types of cell differentiation: an immunohistochemical study of trefoil peptides (TFF1 and TFF2) and mucins (MUC1, MUC2, MUC5AC and MUC6). J Pathol 2000;190: Tarlow JK, Blakemore AI, Lennard A, Solari R, Hughes HN, Steinkasserer A, Duff GW. Polymorphism in human IL-1 receptor antagonist gene intron 2 is caused by variable numbers of an 86-bp tandem repeat. Hum Genet 1993;91: Parkin DM, Laara E, Muir CS. Estimates of the worldwide frequency of sixteen major cancers in Int J Cancer 1988;41: Quina MG. Helicobacter pylori: the Portuguese scene. Grupo de Estudo Portugues do Helicobacter pylori (GEPHP). Eur J Cancer Prev 1994;3(suppl 2): Carneiro F, Taveira-Gomes A, Cabral-Correia A, Vasconelos-Teixeira A, Barreira R, Cardoso-Oliveira M, Sobrinho-Simões M. Characteristics of the gastric mucosa of direct relatives of patients with sporadic gastric carcinoma. Eur J Cancer Prev 1993;2: Dunning AM, Durocher F, Healey CS, Teare MD, McBride SE, Carlomagno F, Xu CF, Dawson E, Rhodes S, Ueda S, Lai E, Luben RN, Van Rensburg EJ, Mannermaa A, Kataja V, Rennart G, Dunham I, Purvis I, Easton D, Ponder BA. The extent of linkage disequilibrium in four populations with distinct demographic histories. Am J Hum Genet 2000;67: Nicklin MJ, Weith A, Duff GW. A physical map of the region encompassing the human interleukin-1 alpha, interleukin-1 beta, and interleukin-1 receptor antagonist genes. Genomics 1994;19: Received February 20, Accepted June 6, Address requests for reprints to: José Carlos Machado, IPATIMUP, Rua Roberto Frias, s/n, 4200 Porto, Portugal. josem@ ipatimup.pt; fax: (351) Supported by Fundação para a Ciência e a Tecnologia (FCT) (grant 35374/99 Sapiens Proj99) and by Programa Operacional Ciência, Tecnologia e Inovação (POCTI) do Quadro Comunitário de Apoio (QCA) II.