Racial/ethnic differences in risk factors for non-cardia gastric cancer: an analysis of the Multiethnic Cohort (MEC) Study.

PURPOSE: Gastric cancer (GC) incidence rates show notable differences by racial/ethnic groups in the US. We sought to determine whether stratification by race/ethnicity would reveal unique risk factors for development of non-cardia gastric cancer (NCGC) for US population. METHODS: Analysis included 1,112 incident cases of NCGC and 190,883 controls from the Multiethnic Cohort Study, a prospective US cohort study that recruited individuals living in Hawaii and California, aged 45-75 years from 5 races/ethnicities. Descriptive analysis and Cox regression models examined the association of risk factors for GC and calculate hazard ratios for each race/ethnicity, adjusting for sociodemographic and dietary variables. RESULTS: Increasing age and male sex were risk factors for NCGC for most race/ethnicities. Higher risk was associated with: GC family history for Latino and Japanese American individuals [HRs range from 1.75 to 1.98]; foreign-born for Japanese American individuals [HR: 1.52, 95% CI 1.11-2.09]; lower education for African American, Japanese American, and Native Hawaiian individuals [HRs range from 1.30 to 1.74]; daily alcohol consumption for African American individuals[HR: 1.56, 95% CI 1.04-2.35]; current smoking for Latino and Japanese American individuals [HRs range from 1.89 to 1.94]; sodium consumption in the highest quartile for White individuals [HR: 2.55, 95% CI 1.23-5.26] compared to the lowest quartile; fruit consumption in the 2nd, 3rd, and 4th highest quartile for Native Hawaiian individuals [HRs range from 2.19 to 2.60] compared to the lowest quartile; diabetes for African American individuals [HR: 1.79, 95% CI 1.21-2.64]; and gastric/duodenal ulcers for Native Hawaiian individuals [HR: 1.82, 95% CI 1.04-3.18]. CONCLUSION: Analyses by racial/ethnic group revealed differing risk factors for NCGC. Increased knowledge of the varying pathways to GC can support personalized GC prevention strategies and risk stratification tools for early detection.

Duke Scholars

Author Meira Epplein Population Health Sciences