Heightened Risk for Gastric Cancer Among Immigrant Populations in New York State from High-Incidence Countries

Emerging global evidence points to country of origin as an important risk factor for developing gastric cancer (GC). Immigrants from high-incidence countries for GC may retain an elevated risk despite living in low-incidence regions, and precursor lesions such as intestinal metaplasia may be more common in foreign-born individuals than individuals born in the United States (U.S.), particularly within high-risk racial/ethnic groups.^1,2 While the U.S. had 51 million immigrants in 2020—the highest number of foreign-born residents globally—data on the association between immigrant status and GC risk within the U.S. remains limited.^3,4 The aim of this study was to examine the association between GC incidence and proportion of foreign-born residents, particularly those from high-risk countries, in New York State (NYS).

This analysis included individual case-level data for all malignant GC cases diagnosed in 2014 to 2018 from the NYS Cancer Registry Surveillance, Epidemiology, and End Results Program Database Management System (SEER*DMS) database. Cases were divided into cardia (primary site: C16.0) or non-cardia (primary sites: C16.1–16.6) gastric cancers. Census tract-level demographic data was obtained from the 2014–2018 U.S. Census Bureau’s American Community Survey (ACS) based on 2010 census tracts. All census tract-level variables were divided into quintiles. Exclusion criteria and variable definitions are detailed in the Supplementary Methods.

We conducted descriptive analyses for both cardia and non-cardia GC (CGC and NCGC, respectively) cases using individual-level data (see Supplementary Methods). After merging case and ACS data by 2010 census tract of residence at diagnosis, we calculated the age-adjusted incidence rates (IRs) and corresponding 95% confidence intervals (CIs) of NCGC and CGC by quintile for each covariate (see Supplementary Methods). We then used multivariable-adjusted Poisson regression models to estimate the age- and sex-adjusted incidence rate ratios (IRRs) and 95% CIs for each quintile compared to the reference group (the lowest risk quintile), using 2014–2018 population data with the log population counts as the offset term.

5,163 cases were included in the analysis, with 2,979 (57.7%) individuals with NCGC and 2,184 (42.3%) individuals with CGC. Supplementary Table 1 shows the distribution of NCGC and CGC cases by sex, race, ethnicity, foreign-born status, high-risk country of origin, region of NYS, and cancer stage. The overall age-adjusted incidence of NCGC was 3.58 per 100,000 population and for CGC was 2.54 per 100,000 population. Figure 1 and Supplementary Figure 1 show the age-adjusted IRs of NCGC and CGC, respectively, by quintile for each census tract-level variable. Notably, rates of NCGC increased as the percentage of foreign-born individuals increased (p<0.01), but this trend was not observed for CGC.

Figure 1.

Age-adjusted IRs of non-cardia gastric cancer per 100,000 population from 2014 to 2018 in New York State and linear regression analysis of the p-value for trend by median of the quintile of census tract-level variables of interest, as well as age- and sex- adjusted IRRs and 95% CIs from Poisson regression analysis for quintiles of each census tract-level variable of interest compared to the reference for non-cardia gastric cancer cases. (IR=incidence rate. IRR=incidence rate ratio. CI=confidence interval. $=U.S. dollar)

Adjusting for age and sex, incidence of NCGC was 4.16 times higher in areas with the highest proportion of foreign-born individuals compared to the lowest (IRR=4.16, 95% CI=3.66–4.72; Figure 1). IRRs for the highest vs. lowest-risk quintiles were 3.60 for the percentage of residents with high-risk race, defined as Black/African American, Asian/Pacific Islander, and American Indian/Alaskan Native (95% CI=3.18–4.08), 3.10 for Hispanic ethnicity (95% CI=2.74–3.50), and 3.42 for percentage of immigrants from a high-risk country for GC (95% CI=3.03–3.85; defined in Supplementary Methods). There were also statistically significant associations with NCGC for the lowest vs. highest quintiles of median household income (IRR=2.09, 95% CI=1.86–2.35) and percentage of residents with a college degree or higher (IRR=2.07, 95% CI=1.84–2.33). While the associations were strongest for comparisons of the extreme quintiles, the IRRs for NCGC were statistically significantly higher for almost all quintiles compared to the reference levels (Figure 1).

In contrast, the age and sex-adjusted IRR for CGC comparing the highest and lowest-risk quintiles for percent foreign-born individuals was 0.69 (95% CI=0.60–0.79), suggesting a statistically significant lower incidence of CGC in areas with a high percentage of foreign-born individuals (Supplementary Figure 1). Similarly, the adjusted incidence of CGC was lower in areas with the highest vs. lowest percentage of residents with high-risk race, Hispanic ethnicity, and percentage of immigrants from a high-risk country (Supplementary Figure 1).

In summary, we observed that incidence rates of NCGC increased with increasing census tract-level percentage of residents of high-risk race and ethnicity, residents who immigrated to the U.S., and residents who were born in a high-risk country for GC. These trends were not observed for CGC incidence (and incidence of CGC was in fact lower in census tracts with higher proportions of these individuals), highlighting the distinct racial and ethnic patterns in incidence among these GC subtypes.

This is the first study to investigate the association between country of origin and GC incidence in NYS, a state with one of the highest proportions of immigrant residents in the U.S. It is also the first study to broaden the scope of potentially impacted immigrant populations to include those from all high-incidence countries for GC rather than specific immigrant groups.^5,6 Therefore, our comprehensive analysis is more representative of the diverse immigrant populations in NYS and the U.S. as a whole. Our results also affirm those of a recent meta-analysis, which found that immigrants from countries with high incidence of GC maintain their risk status even when living in a low incidence country, and are consistent with previous studies on GC risk among specific racial and ethnic subgroups in other U.S. states.^17–9

Our study has several limitations. First, a high percentage of cases reported to the NYS Cancer Registry do not have birthplace information and these data are often obtained from death certificates, which can introduce bias. Therefore, we utilized census tract-level data to explore geographical associations rather than exploring individual-level GC risk. We were also unable to control for other common risk factors for GC, including Helicobacter pylori infection, smoking, and diet, or to determine how long NCGC may remain a risk to immigrant patients.¹⁰ Given that our analysis is limited to residents of NYS, our conclusions may not be generalizable to other regions.

In summary, our findings suggest that immigrants, especially those from countries with a high burden of gastric cancer, remain a vulnerable subgroup for NCGC. Targeted screening and prevention efforts in these communities may be beneficial for early detection and improved GC outcomes.

Abbreviations:

ACS

American Community Survey

CGC

cardia gastric cancer

CI

confidence interval

GC

gastric cancer

IARC

International Agency for Research on Cancer

ICD-O-3

International Classification of Diseases for Oncology, third edition

IR

incidence rate

IRR

incidence rate ratio

NCGC

non-cardia gastric cancer

NYC

New York City

NYS

New York State

SEER*DMS

Surveillance, Epidemiology, and End Results Program Data Management System

U.S.

United States