Background: Stomach cancer is a major public health burden worldwide, with 5-year survival rates ranging from 20% to 30% in most countries. However, studies have shown early diagnosis and prompt treatment leading to better survival. Atrophic gastritis is a precancerous lesion of stomach cancer, which is thought to be a reversible stage. Few studies have examined environmental risk factors and the role of genetic susceptibility of atrophic gastritis, compared to the stomach cancer endpoint. The overall objective is to evaluate potential risk factors, including Helicobacter pylori infection, tobacco smoking, alcohol drinking, dietary habits of excessive salt intake, dietary nitrate and nitrites, and observe their association with atrophic gastritis and stomach cancer individually. In addition, we examine the relationship between candidate single nucleotide polymorphisms (SNPs) from the microRNA pathway, stem cell pathway, and genomic wide association studies (GWAS) on atrophic gastritis and stomach cancer respectively.
Methods: A population based case-control study was conducted in Jiangsu Province, China. In our study, 1,617 stomach cancer patients and 6,369 cancer-free controls were included for analysis. For specific aim 1, the evaluation of risk factors of atrophic gastritis, only controls (cancer-free participants) are examined. For Specific Aims 2 and 3, case-control study design was employed including both stomach cancer cases and cancer-free controls for the development of stomach cancer. Epidemiologic data were collected by face-to-face interview using a standardized questionnaire, and a 5 ml blood sample was collected at the time of the interview. Atrophic gastritis status was chemically defined using serum pepsinogen (PG) cutoffs of PG I 70ng/mL and PG I/PG II 6. Unconditional logistic regression models have been used to estimate adjusted odds ratios (ORs) and 95% confidence intervals (CIs). Potential confounding factors have been adjusted in the analyses, including age, gender, study site, average family income, education level, family history of stomach cancer, pack-years of smoking (except smoking related factors), alcohol consumption (except alcohol drinking related factors), total caloric intake (for dietary intake variables only), and H. pylori status. Additive and multiplicative interactions have been evaluated, using relative excess risk due to interaction (RERI) and ratio of odds ratios (ROR) respectively.
Results: In our study, individuals with tonic supplement, having a history of drinking non-boiled water at childhood, and a moderate consumption of red meat were at higher risk of atrophic gastritis respectively, while adjusting for potential confounding factors. However, tobacco smoking, alcohol drinking, tea drinking, and dietary intake of micronutrients (vitamin A, thiamin, riboflavin, niacin, vitamin C, vitamin E, zinc, selenium) and macronutrients (protein, fat, fiber, carbohydrates) were not significantly associated with atrophic gastritis. A protective effect of frequent consumption of BBQ meat, salted meat or fish, and higher waist-to-hip ratio was observed on the odds of atrophic gastritis. For gastric cancer, tobacco smoking, alcohol drinking, a preference for salty foods, spicy foods, and high temperature foods, high vegetable intake, and high intake of certain vitamins and minerals (vitamin A, thiamin, riboflavin, vitamin C, vitamin E, zinc, selenium) were associated with higher odds of stomach cancer. H. pylori IgG seropositivity and atrophic gastritis, expressed by lower serum PG levels, were also positively associated with stomach cancer respectively. Inverse associations of light tea drinking, consumption of raw garlic and ginger, and higher intake of salted or preserved meat/fish (third quartile versus lowest quartile) were found on stomach cancer respectively. When stratifying by atrophic gastritis status, similar associations are observed for the subpopulation without atrophic gastritis, but not in those with atrophic gastritis. For genetic susceptibility markers, significant inverse associations were observed with atrophic gastritis in rs3130932 (Oct4) and rs3729629 (WNT2) from the stem cell pathway, even after semi-bayes adjustment. For the stomach cancer endpoint, rs11077 (XPO5), rs12828 (WWOX), rs4072391 (IL6R), rs11364 (HES2), and rs738722 (CHEK2) are found to be positively associated and rs2075993 (E2F2), rs2273368 (Wnt2B), rs4961280 (Ago2), rs7372209 (miR-26a1), rs1033583 (DLL1), rs1981492 (AXIN1), rs3130932 (Oct4), rs3729629 (WNT2), rs3734637 (HEY2), and rs4835761 (WNT8A) were inversely associated with stomach cancer risk overall, even after semi-bayes shrinkage. When stratifying by atrophic gastritis status, rs12828 (WWOX), rs2273368 (Wnt2B), rs9266 (KRAS) from the miRNA pathway, and rs11364 (HES2), rs2240308 (AXIN2), rs1033583 (DLL1), rs1981492 (AXIN1), rs3130932 (Oct4), rs3729629 (WNT2), rs4835761 (WNT8A), rs915894 (Notch4) from the stem cell pathway were significantly associated with stomach cancer in those without atrophic gastritis. In those with atrophic gastritis, only the rs2273368 (Wnt2B), rs3734637 (HEY2), and rs738722 (CHEK2) were associated with stomach cancer after semi-bayes adjustment. Additive interactions were observed between atrophic gastritis and rs2273368 (Wnt2B), and multiplicative interactions were observed between atrophic gastritis and rs2273368 (Wnt2B), rs11077 (XPO5), rs3130932 (Oct4), and rs738722 (CHEK2) on stomach cancer. Also interactions were found between H. pylori infection and polymorphisms of Ran (rs14035), Gemin4 (rs2740348), HEY1 (rs1046472), Ctbp2 (rs3740535), and between smoking and polymorphisms of Rbl2 (rs3929) and miR-26a1 (rs7372209), and between alcohol drinking and polymorphisms of DOCK4 (rs3801790) on the odds of stomach cancer.
Conclusion: This study confirms the association between established risk factors of stomach cancer such as smoking, H. pylori infection, and atrophic gastritis, and adds evidence on the protective effects of tea drinking and raw garlic consumption. To our knowledge, it is the first study to report inverse associations between ginger consumption and stomach cancer. Furthermore, our study is one of the first studies to show association in polymorphisms in the stem cell pathway and atrophic gastritis, and interactions in H. pylori infection, smoking, and alcohol drinking with atrophic gastritis on stomach cancer respectively. Further studies on different ethnic groups and atrophic gastritis determined by the use of tissue samples should be conducted for a fuller understanding of atrophic gastritis and stomach carcinogenesis.