The incidence of gastric cancer can vary 5-fold to 10-fold between high-risk and low-risk regions. Golestan Province in northeast of Iran is a high-risk area, where gastric cancer is the second most common cancer in both sexes.

This is a nested case-control study within the Golestan Cohort Study (GCS), including all confirmed cases of gastric cancer accrued in GCS until January 1, 2018 (n=288) and a 4 to 1 set of controls (n=1148), matched on age, sex, and place of residence, who were alive and cancer-free at the time of the case diagnosis. We used Multiplex serology to evaluate 17 H. pylori proteins in plasma samples from cases and controls. We used conditional logistic regression (conditioned on the matching factors), to analyze the association between potential risk factors and incident gastric cancer adjusted for H. pylori antigens.

Among 288 confirmed cases, 175 were in the cardia region, 83 were non-cardia cancers, and 30 were undetermined. In addition to the matching factors (age, sex, and place of residence), cases and controls were similar in most baseline characteristics. Compared with controls, the cases were significantly less likely to have had received any formal education (18.8% vs. 27.0%, p<0.01) and brush their teeth (21.7% vs. 30%, p<0.05). Cases also had lower daily intake of fruits and fish compared with controls. Non-O blood groups, especially A, were more common in cases than controls (76.4% vs. 68.9%, p<0.01). The prevalence of H. pylori positivity (presence of 3 or more antigens) was similar between cases and controls (74.5% vs. 74.9%, respectively). However, cases were significantly more likely to be positive for CagA (85.1% vs. 78.0%, p<0.01), VacA (45.5% vs. 37.1%, p<0.01), while NapA was more common in controls (32.4% in cases vs. 45.3% in controls, p<0.01). The level of p53 antigen was also higher in cases, but the difference did not reach statistical significance (6.9% vs. 4.3%, p=0.06). In multivariate adjusted models, daily tooth brushing was protective for cardia cancer (OR=0.49; 95%CI: 0.26-0.94, p<0.05), while a history of diabetes was associated with increased risk (OR=2.03; 95%CI: 1.02-4.05). Drinking hot tea was also associated with higher risk of cardia cancer, but the association did not reach statistical significance (OR=1.48; 95%CI: 0.97-2.25, p=0.06). High intake of red mean was the strongest risk factor associated with the risk of non-cardia cancer (OR=3.39; 95%CI: 1.4-8.24, p<0.01), followed by having a non-O blood group (OR=2.04; 95%CI: 1.01-4.17, p<0.05).

In this high-risk region for gastric cancer, cardia cancer was the predominant type, and similar to many other high-risk populations, a high proportion of both cases and controls were infected with H. pylori. We found significant associations between two known H. pylori virulence factors (CagA and VacA) and the risk of gastric cancer, and confirmed a novel inverse association previously observed in this population for NapA, a neutrophil-activating protein. We also observed significant associations for risk factors such as non-O blood group, diabetes, lack of formal education, poor oral hygiene, and high red meat intake.