In addition to bacterial virulence factors, host immune response plays a pivotal role in the outcome of chronic Helicobacter pylori infection. The capacity of the host to mount an interleukin-1β (IL-1β)-driven response is influenced by sequence variants in the IL-1/IL-1RN cluster.1
El-Omar et al2 were the first to report an association between IL-1B -31 and IL-1RN 2/2 of IL-1RN 86 variable number of tandem repeats (VNTRs) and the development of chronic hypochlorhydric response to H pylori and the risk of gastric cancer. In contrast, very little is known about germ-line mutations predisposing patients with chronic H pylori infection to develop gastric mucosa-associated lymphoid tissue (MALT) lymphoma.
Therefore, we read with great interest the report by Rollison et al3 in Blood about an association between the proinflammatory genotype IL-1RN 2/2 with gastric marginal zone lymphoma in a retrospective series of 66 cases from northern England.
We investigated the functional variants in the IL-1 cluster and their influence on the development of primary gastric B-cell lymphoma in 153 patients participating in an intention-to-treat prospective multicenter study of the German-Austrian-Lymphoma Study Group.4 Included as controls were 344 patients with H pylori infection undergoing upper gastrointestinal (GI) endoscopy when histology of 2 biopsies taken from the antrum and the corpus of the stomach excluded gastric lymphoma.
Of 153 patients with primary gastric B-cell lymphoma, 88 presented with low-grade (MALT) and 65 patients with high-grade lymphoma (Table 1). The allele frequencies in the control group match well with those previously reported in the literature.2 There were no significant associations found with the histological grade or stage of disease in single marker analysis. Of patients with disease stages E II to E IV, 20.6% were homozygous for IL-1B -31 allele C, compared to only 10.2% of patients with disease stage E I (Pearson χ2P = .112, OR 2.27, CI 95% 0.81-6.46). IL-1β +3954 CC was found in 9.4% of patients with disease stages E II to E IV, compared to 2.8% of patients with stage E I (Pearson χ2P = .109, OR 3.57, CI 95% 0.69-18.72) (Table 2). Haplotype analysis of the IL-1 cluster and especially the proinflammatory haplotype IL-1β -31 C/IL-1RN 2 did not show any significant associations with histological grade or disease progression (Table 3).
Disease stage . | No. . | Low grade . | High grade . | Sex, no. male/no. female . | Median age, y . |
---|---|---|---|---|---|
E I | 39 | 28 | 11 | 22/17 | 64 |
E I1 | 51 | 40 | 11 | 23/28 | 60 |
E I2 | 27 | 9 | 18 | 17/10 | 59 |
E II | 4 | 1 | 3 | 3/1 | 67 |
E II.1 | 26 | 7 | 19 | 17/9 | 66 |
E III | 2 | 1 | 1 | 2/0 | 63 |
E IV | 4 | 2 | 2 | 2/2 | 57 |
Disease stage . | No. . | Low grade . | High grade . | Sex, no. male/no. female . | Median age, y . |
---|---|---|---|---|---|
E I | 39 | 28 | 11 | 22/17 | 64 |
E I1 | 51 | 40 | 11 | 23/28 | 60 |
E I2 | 27 | 9 | 18 | 17/10 | 59 |
E II | 4 | 1 | 3 | 3/1 | 67 |
E II.1 | 26 | 7 | 19 | 17/9 | 66 |
E III | 2 | 1 | 1 | 2/0 | 63 |
E IV | 4 | 2 | 2 | 2/2 | 57 |
Locus and genotype . | Low grade, range, % . | High grade, range, % . | All, range, % . | Controls, range, % . | E I, range, % . | E II-IV, range, % . |
---|---|---|---|---|---|---|
IL-1β -31 | ||||||
C/C | 8-13.1 | 10-12.1 | 18-12.5 | 45-13.1 | 11-10.2 | 7-20.6 |
C/T | 30-49.2 | 35-42.2 | 65-45.1 | 146-42.4 | 52-48.2 | 13-38.2 |
T/T | 23-37.7 | 38-45.8 | 61-42.4 | 153-44.8 | 45-41.7 | 14-41.2 |
IL-1β + 3954 | ||||||
C/C | 4-6.7 | 3-3.7 | 7-5.0 | 24-7.2 | 3-2.8 | 3-9.4 |
C/T | 20-33.3 | 26-32.1 | 46-32.6 | 121-36.3 | 37-34.6 | 9-28.1 |
T/T | 36-60.0 | 52-64.2 | 88-62.4 | 188-56.5 | 67-62.6 | 20-62.5 |
IL-1RN 86VNTR | ||||||
1/1 | 30-46.9 | 46-52.9 | 76-50.3 | 185-53.8 | 58-51.3 | 17-47.2 |
1/2 | 27-42.2 | 31-35.6 | 58-38.4 | 118-34.3 | 43-38.1 | 14-38.9 |
1/3 | 2-3.1 | 4-4.6 | 6-4.0 | 11-3.2 | 5-4.4 | 1-2.8 |
2/2 | 4-6.3 | 5-5.8 | 9-6.0 | 27-7.9 | 6-5.3 | 3-8.3 |
2/3 | 1-1.6 | 1-1.1 | 2-1.3 | 2-0.6 | 1-0.9 | 1-2.8 |
Locus and genotype . | Low grade, range, % . | High grade, range, % . | All, range, % . | Controls, range, % . | E I, range, % . | E II-IV, range, % . |
---|---|---|---|---|---|---|
IL-1β -31 | ||||||
C/C | 8-13.1 | 10-12.1 | 18-12.5 | 45-13.1 | 11-10.2 | 7-20.6 |
C/T | 30-49.2 | 35-42.2 | 65-45.1 | 146-42.4 | 52-48.2 | 13-38.2 |
T/T | 23-37.7 | 38-45.8 | 61-42.4 | 153-44.8 | 45-41.7 | 14-41.2 |
IL-1β + 3954 | ||||||
C/C | 4-6.7 | 3-3.7 | 7-5.0 | 24-7.2 | 3-2.8 | 3-9.4 |
C/T | 20-33.3 | 26-32.1 | 46-32.6 | 121-36.3 | 37-34.6 | 9-28.1 |
T/T | 36-60.0 | 52-64.2 | 88-62.4 | 188-56.5 | 67-62.6 | 20-62.5 |
IL-1RN 86VNTR | ||||||
1/1 | 30-46.9 | 46-52.9 | 76-50.3 | 185-53.8 | 58-51.3 | 17-47.2 |
1/2 | 27-42.2 | 31-35.6 | 58-38.4 | 118-34.3 | 43-38.1 | 14-38.9 |
1/3 | 2-3.1 | 4-4.6 | 6-4.0 | 11-3.2 | 5-4.4 | 1-2.8 |
2/2 | 4-6.3 | 5-5.8 | 9-6.0 | 27-7.9 | 6-5.3 | 3-8.3 |
2/3 | 1-1.6 | 1-1.1 | 2-1.3 | 2-0.6 | 1-0.9 | 1-2.8 |
Single-marker analysis of the proinflammatory haplotype IL-1β -31/IL-1 RN 86 VNTR. SNPs at IL-1β -31 and + 3954 were genotyped by allelic discrimination (TaqMan technology, ABI 7700, Aplera, Foster City, CA). IL-1RN 86 VNTR was genotyped by Southern blot after amplification. Alleles were sized relative to a 100-bp ladder (allele 1 = 4 repeats, allele 2 = 2 repeats, allele 3 = 5 repeats, allele 4 = 3 repeats, allele 5 = 6 repeats). Haplotype case-control analysis was performed using HAPMAX.9 Hardy-Weinberg equilibrium was confirmed for all polymorphisms tested in gastric lymphoma group and controls. Three types of analysis were performed: (1) all patients with primary gastric B-cell lymphoma were compared against controls, (2) patients separated in low-grade and high-grade lymphoma were compared against controls, and (3) patients with disease stage E I were compared against patients with disease stages E II to E IV. Statistical analysis was performed using SISA Binominal program (Uitenbroek, Daan G, Binomial, SISA, http://home.clara.net/sisa/binomial.htm).
IL-1β -31 . | IL-1RN 86VNTR . | Low grade, %e . | High grade, % . | All, % . | Controls, % . | E I, % . | E II-IV, % . |
---|---|---|---|---|---|---|---|
C | 1 | 20.4 | 15.7 | 17.5 | 17.6 | 19.5 | 17.0 |
C | 2 | 18.1 | 15.8 | 16.5 | 16.5 | 20.2 | 15.9 |
C | 3 | 0.0 | 1.6 | 1.3 | 0.0 | 0.0 | 1.7 |
T | 1 | 48.9 | 58.4 | 55.2 | 55.0 | 48.2 | 57.3 |
T | 2 | 10.8 | 7.1 | 8.0 | 8.8 | 9.2 | 7.0 |
T | 3 | 1.9 | 1.4 | 1.5 | 1.9 | 2.9 | 1.1 |
IL-1β -31 . | IL-1RN 86VNTR . | Low grade, %e . | High grade, % . | All, % . | Controls, % . | E I, % . | E II-IV, % . |
---|---|---|---|---|---|---|---|
C | 1 | 20.4 | 15.7 | 17.5 | 17.6 | 19.5 | 17.0 |
C | 2 | 18.1 | 15.8 | 16.5 | 16.5 | 20.2 | 15.9 |
C | 3 | 0.0 | 1.6 | 1.3 | 0.0 | 0.0 | 1.7 |
T | 1 | 48.9 | 58.4 | 55.2 | 55.0 | 48.2 | 57.3 |
T | 2 | 10.8 | 7.1 | 8.0 | 8.8 | 9.2 | 7.0 |
T | 3 | 1.9 | 1.4 | 1.5 | 1.9 | 2.9 | 1.1 |
Haplotype analysis of the proinflammatory haplotype IL-1β -31/IL-1RN 86 VNTR. Analysis was conducted as described in the notes for Table 2.
In conclusion we could not confirm the results of Rollinson et al.3 One reason may be that Rollinson et al extracted DNA to investigate germ-line mutations from biopsy specimen and surgical blocks of lymphoma tissue and not from peripheral blood. A contamination with tumor material cannot be excluded. The admixture of somatic DNA may have obscured their analysis. To our knowledge, there is no mechanistic evidence so far that the proinflammatory effect of IL-1β contributes to development of primary gastric B-cell lymphoma. In contrast, recombinant IL-1β exerted a marked antilymphoma activity, reflected by significantly improved survival of treated mice after inoculation of BCL-1 cells.5 The genetic susceptibility of patients with chronic H pylori infection to develop primary gastric B-cell lymphoma, especially of the MALT-type, appears to remain unclear.
This feature is available to Subscribers Only
Sign In or Create an Account Close Modal