Helicobacter pylori, Homologous-Recombination Genes, and Gastric Cancer PDF

The n e w e ng l a n d j o u r na l of m e dic i n e Original Article Helicobacter pylori, HomologousRecombination Genes, and Gastric Cancer Yoshiaki Usui, M.D., Ph.D., Yukari Taniyama, Ph.D., Mikiko Endo, B.Sc., Yuriko N. Koyanagi, M.D., Ph.D., Yumiko Kasugai, M.M.Sc., Isao Oze, M.D., Ph.D., Hidemi...

The n e w e ng l a n d j o u r na l of m e dic i n e Original Article Helicobacter pylori, HomologousRecombination Genes, and Gastric Cancer Yoshiaki Usui, M.D., Ph.D., Yukari Taniyama, Ph.D., Mikiko Endo, B.Sc., Yuriko N. Koyanagi, M.D., Ph.D., Yumiko Kasugai, M.M.Sc., Isao Oze, M.D., Ph.D., Hidemi Ito, M.D., Ph.D., M.P.H., Issei Imoto, M.D., Ph.D., Tsutomu Tanaka, M.D., Ph.D., Masahiro Tajika, M.D., Ph.D., Yasumasa Niwa, M.D., Ph.D., Yusuke Iwasaki, M.E., Tomomi Aoi, B.Sc., Nozomi Hakozaki, Sadaaki Takata, B.Sc., Kunihiko Suzuki, Chikashi Terao, M.D., Ph.D., Masanori Hatakeyama, M.D., Ph.D., Makoto Hirata, M.D., Ph.D., Kokichi Sugano, M.D., Ph.D., Teruhiko Yoshida, M.D., Ph.D., Yoichiro Kamatani, M.D., Ph.D., Hidewaki Nakagawa, M.D., Ph.D., Koichi Matsuda, M.D., Ph.D., Yoshinori Murakami, M.D., Ph.D., Amanda B. Spurdle, Ph.D., Keitaro Matsuo, M.D., Ph.D., and Yukihide Momozawa, D.V.M., Ph.D.​​ A BS T R AC T BACKGROUND Helicobacter pylori infection is a well-known risk factor for gastric cancer. However, the contribution of germline pathogenic variants in cancer-predisposing genes and their effect, when combined with H. pylori infection, on the risk of gastric cancer has not been widely evaluated. METHODS We evaluated the association between germline pathogenic variants in 27 cancerpredisposing genes and the risk of gastric cancer in a sample of 10,426 patients with gastric cancer and 38,153 controls from BioBank Japan. We also assessed the combined effect of pathogenic variants and H. pylori infection status on the risk of gastric cancer and calculated the cumulative risk in 1433 patients with gastric cancer and 5997 controls from the Hospital-based Epidemiologic Research Program at Aichi Cancer Center (HERPACC). The authors' affiliations are listed in the Appendix. Dr. Momozawa can be contacted at ­momozawa@​­riken​.­jp or at the Laboratory for Genotyping Development, RIKEN Center for Integrative Medical Sciences, 1-7-22 Suehiro-cho, Tsurumi-ku, Yokohama 230-0045, Japan. Drs. Matsuo and Momozawa contributed equally to this article. N Engl J Med 2023;388:1181-90. DOI: 10.1056/NEJMoa2211807 Copyright © 2023 Massachusetts Medical Society. RESULTS Germline pathogenic variants in nine genes (APC, ATM, BRCA1, BRCA2, CDH1, MLH1, MSH2, MSH6, and PALB2) were associated with the risk of gastric cancer. We found an interaction between H. pylori infection and pathogenic variants in homologous-recombination genes with respect to the risk of gastric cancer in the sample from HERPACC (relative excess risk due to the interaction, 16.01; 95% confidence interval [CI], 2.22 to 29.81; P = 0.02). At 85 years of age, persons with H. pylori infection and a pathogenic variant had a higher cumulative risk of gastric cancer than noncarriers infected with H. pylori (45.5% [95% CI, 20.7 to 62.6] vs. 14.4% [95% CI, 12.2 to 16.6]). CONCLUSIONS H. pylori infection modified the risk of gastric cancer associated with germline pathogenic variants in homologous-recombination genes. (Funded by the Japan Agency for Medical Research and Development and others.) n engl j med 388;13 nejm.org March 30, 2023 1181 The New England Journal of Medicine Downloaded from nejm.org at Lane Medical Library, Stanford University Med Center on December 21, 2023. For personal use only. No other uses without permission. Copyright © 2023 Massachusetts Medical Society. All rights reserved. The n e w e ng l a n d j o u r na l G astric cancer is the fifth most common neoplasm and the fourth leading cause of death from cancer worldwide.1 Helicobacter pylori has been classified as a group I carcinogen and is an environmental risk factor for gastric cancer.2 Although H. pylori infection affects more than half the world population,3 the age-standardized global incidence rate of cancer attributable to H. pylori infection is highest in East Asia, at 17.6 cases per 100,000 person-years.4 Tailored surveillance and eradication of H. pylori are recommended for controlling the incidence of gastric cancer.2 Germline pathogenic variants in cancerpredisposing genes are also essential factors in tailored surveillance and prevention of gastric cancer.5,6 For example, CDH1 is a risk gene for hereditary diffuse gastric cancer.5,6 Families with hereditary diffuse gastric cancer meeting certain criteria are recommended for genetic testing, and carriers of pathogenic variants in CDH1 are considered for prophylactic gastrectomy.6 We recently found that germline pathogenic variants in BRCA1 and BRCA2 in the homologous-recombination7 genes substantially increased the risk of gastric cancer.8 It has also been suggested that other cancer-predisposing genes, such as mismatch-repair genes,7 are associated with the risk of gastric cancer.5 However, with the exception of CDH1, the identification of variants in these genes are not included in screening recommendations.9 A randomized, controlled trial showed that eradication of H. pylori reduced the incidence of gastric cancer, with effects seen even in persons with a first-degree family history of gastric cancer.10 However, because a family history generally involves shared genetic and environmental factors,11 it is unclear whether eradication was effective primarily in genetically susceptible persons or was effective regardless of genetic status. Because gene–environment interactions are associated with an excess risk of disease,12 an evaluation of pathogenic-variant carrier status and H. pylori infection status may be informative with respect to risk stratification. We therefore tested for an association between variants in cancer-predisposing genes and the risk of gastric cancer and then evaluated the combined effect of pathogenic variants and H. pylori infection status on risk. 1182 n engl j med 388;13 of m e dic i n e Me thods Study Population We included persons from two independent study cohorts in the analysis: BioBank Japan (BBJ) and the Hospital-based Epidemiologic Research Program at Aichi Cancer Center (HERPACC). Our study design is summarized in Figure S1 in the Supplementary Appendix, available with the full text of this article at NEJM.org. The study sample from BBJ included 10,426 patients with gastric cancer and 38,153 controls with nonmalignant diseases. The study sample from HERPACC included 1433 patients with newly diagnosed gastric cancer and 5997 controls without cancer, all of whom had an H. pylori infection status indicated. We defined a negative result as negative tests for both anti–H. pylori IgG antibodies and atrophic gastritis (as determined with the pepsinogen test according to the “ABC method,”13 a noninvasive population surveillance tool14). We defined a positive result as one in which either test was positive. We used both tests because anti–H. pylori IgG serum antibody is known to spontaneously disappear with atrophic gastritis, which occurs with progression of H. pylori infection in some persons.15 All participants provided written informed consent. The study was approved by the ethics committees of the Institute of Medical Sciences, the University of Tokyo, the RIKEN Center for Integrative Medical Sciences, and Aichi Cancer Center. Details are provided in the Supplementary Methods section of the Supplementary Appendix. The authors vouch for the accuracy and completeness of the data in this report. Sequencing and Bioinformatics Analysis We analyzed 27 cancer-predisposing genes, including CDH1, homologous-recombination genes (ATM, BARD1, BRCA1, BRCA2, BRIP1, CHEK2, PALB2, RAD51C, and RAD51D),7 mismatch-repair genes (MLH1, MSH2, MSH6, and PMS2),7 and other genes (APC, BMPR1A, CDK4, CDKN2A, EPCAM, HOXB13, MUTYH, NBN, NF1, PTEN, SMAD4, STK11, and TP53), as described previously.16 In the BBJ cohort, we used the raw data from sequencing of BRCA1 and BRCA2 in patients with gastric cancer and unaffected persons8 and data from sequencing of the other 25 genes in 23,678 unaffected persons16 from our previous studies. We performed se- nejm.org March 30, 2023 The New England Journal of Medicine Downloaded from nejm.org at Lane Medical Library, Stanford University Med Center on December 21, 2023. For personal use only. No other uses without permission. Copyright © 2023 Massachusetts Medical Society. All rights reserved. H. pylori and Gastric Cancer quencing for participants who had not undergone sequencing in the previous studies. With respect to the sequencing data, we implemented quality-control measures that were based on sequencing read depth, call rates, and deviation from Hardy–Weinberg equilibrium. We annotated variants that had a minor-allele frequency of less than 0.1% in the relevant control group. We defined pathogenic variants as those that resulted in loss of function according to SnpEff, version 4.3t,17 or that were designated as pathogenic or likely pathogenic by ClinVar, version 2021-06-19.18 Details are provided in Supplementary Methods. in Japan,22 and the age distribution of the population in Japan.22 All statistical tests were two-sided, and a P value of less than 0.05 was considered to indicate statistical significance. A Bonferroni correction was applied where appropriate. All statistical analyses were performed with the use of Stata software, version 16.0 (StataCorp), and R software, version 3.5.2 (R Foundation for Statistical Computing). Details are provided in Supplementary Methods. R e sult s Characteristics of Study Participants Statistical Analysis First, we evaluated the association between cancer-predisposing genes and the risk of gastric cancer, using a logistic-regression model with adjustment for age at entry and sex as recorded in BBJ. On the basis of a Bonferroni-corrected threshold of significance, genes with a P value of less than 1.85×10−3 (i.e., 0.05 ÷ 27) were defined as gastric cancer risk genes. Second, we evaluated the clinical characteristics of the variant carriers. We also conducted a survival analysis involving patients with newly diagnosed gastric cancer. Third, we estimated the odds ratios and corresponding 95% confidence intervals to evaluate the combined effect of germline pathogenic variants in the gastric cancer risk genes and H. pylori infection status on the risk of gastric cancer, using a logistic-regression model with adjustment for age at entry and sex as recorded in HERPACC. Interactions between pathogenic-variant carrier status and H. pylori infection status with respect to the effect on the risk of gastric cancer were estimated as an additive interaction, measured as the relative excess risk due to interaction,19 and as a multiplicative interaction, measured with the use of interaction terms. For the evaluation of interactions, we used dichotomous categories for pathogenic-variant carrier status (carrier or noncarrier) and H. pylori infection status (positive or negative). No values were missing in these two data sets. Finally, we calculated the cumulative risk20,21 of gastric cancer by using the odds ratio of the combined pathogenic-variant carrier status and H. pylori infection status, the prevalence of each status subgroup in our study control group, the age-specific cancer incidence rate n engl j med 388;13 The characteristics of the study participants are shown in Table 1 and Table S1. The median age at entry into the registry was 69 years among the BBJ patients with gastric cancer, 62 years among the HERPACC patients with gastric cancer, 64 years among the BBJ controls, and 55 years among the HERPACC controls. The percentage of male participants was 74.5% among the BBJ patients with gastric cancer, 74.6% among the HERPACC patients with gastric cancer, 53.1% among the BBJ controls, and 51.1% among the HERPACC controls. The percentage of patients with intestinal cancer was 61.2% among the BBJ patients and 49.3% among the HERPACC patients; this difference may be due to variations in the characteristics of the recruiting hospitals. Annotation of Germline Variants The mean read depths were 1003× among BBJ participants and 800× among HERPACC participants. After quality-control measures were applied, the genetic data from 10,366 BBJ patients with gastric cancer, 37,592 BBJ controls, 1423 HERPACC patients with gastric cancer, and 5990 HERPACC controls were included in the analysis. The read depth was greater than 20× in more than 99.8% of the target region for all participants. We annotated 459 of the 5676 germline variants in BBJ and 104 of the 1572 germline variants in HERPACC as pathogenic (Table S2). In total, we identified 518 pathogenic variants; the ClinVar accession numbers are provided in Table S2. Identification of Gastric Cancer Risk Genes The associations between variants in cancerpredisposing genes and the risk of gastric cancer from the BBJ data are shown in Table 2. In ad- nejm.org March 30, 2023 1183 The New England Journal of Medicine Downloaded from nejm.org at Lane Medical Library, Stanford University Med Center on December 21, 2023. For personal use only. No other uses without permission. Copyright © 2023 Massachusetts Medical Society. All rights reserved. The n e w e ng l a n d j o u r na l of m e dic i n e Table 1. Characteristics of the Study Population.* Characteristic BBJ HERPACC Controls (N = 38,153) Patients with Gastric Cancer (N = 1433) Controls (N = 5997) 69 (62–75) 64 (54–72) 62 (54–69) 55 (45–64) Male 7770 (74.5) 20,242 (53.1) 1069 (74.6) 3063 (51.1) Female 2656 (25.5) 17,911 (46.9) 364 (25.4) 2934 (48.9) Papillary adenocarcinoma 175 (1.7) — 19 (1.3) — Tubular adenocarcinoma 6204 (59.5) — 688 (48.0) — Poorly differentiated adenocarcinoma 1848 (17.7) — 550 (38.4) — Signet-ring-cell adenocarcinoma 1016 (9.7) — 125 (8.7) — 148 (1.4) — 9 (0.6) — 1035 (9.9) — 42 (2.9) — Antibody positive and atrophic gastritis negative — — 454 (31.7) 1365 (22.8) Antibody positive and atrophic gastritis positive — — 654 (45.6) 1226 (20.4) Antibody negative and atrophic gastritis positive — — 128 (8.9) 177 (3.0) — — 197 (13.7) 3229 (53.8) Median age at entry (IQR) — yr Patients with Gastric Cancer (N = 10,426) Sex — no. (%) Gastric cancer subtype — no. (%) Intestinal Diffuse Mucinous adenocarcinoma Gastric cancer not otherwise specified Helicobacter pylori infection status (%)† Positive Negative Antibody negative and atrophic gastritis negative * Percentages may not total 100 because of rounding. BBJ denotes BioBank Japan, HERPACC Hospital-based Epidemiologic Research Program at Aichi Cancer Center, and IQR interquartile range. † H. pylori infection status was defined as negative for persons who were negative both for anti–H. pylori IgG antibody and for atrophic gastritis as defined by pepsinogen testing. Persons were regarded as positive if they had positive results in either of these two tests. dition to BRCA1 and BRCA2,8 seven genes that, when variant, confer a predisposition to cancer showed significant associations with the risk of gastric cancer (P