Rebecca D. Kehm, PhD
At the 2024 American Association for Cancer Research (AACR) Conference on the Science of Cancer Health Disparities in Racial/Ethnic Minorities and the Medically Underserved, Rebecca D. Kehm, PhD, of Columbia University Mailman School of Public Health, New York, and colleagues presented some of the first data indicating a potential link between food insecurity and the risk of developing hepatocellular carcinoma.1 Their findings from a county-level cross-sectional analysis, which were simultaneously published as an abstract in the journal Cancer Epidemiology, Biomarkers & Prevention,2 revealed a particularly pronounced association with early-onset disease.
“If confirmed, these findings support the need for policies and interventions that can address food insecurity, particularly in populations that we know are at an increased risk for hepatocellular carcinoma and other types of liver cancer,” Dr. Kehm commented. “We also think that screening for food insecurity should be implemented in clinical settings.”
Per Dr. Kehm, in the context of this analysis, food insecurity may be defined as limited or uncertain access to nutritionally adequate, safe, and acceptable foods because of household-level economic and social conditions.
Study Details
The investigators obtained county-level counts of hepatocellular carcinoma cases, all of which were diagnosed between 2018 and 2021, from the Surveillance, Epidemiology, and End Results database (1,085 counties; approximately 48% of the U.S. population). Stratification factors included age at diagnosis, sex, and race/ethnicity.
County-level food insecurity data for the same years were obtained from Feeding America’s Map the Meal Gap reports. The investigators obtained additional county-level covariate data from the Behavioral Risk Factor Surveillance System.
Multilevel Poisson regression models with robust standard errors were used to estimate the association between county-level food insecurity and hepatocellular carcinoma risk. The models included county population size as the offset term, state-level clustering as the random effect, and county-level food insecurity as the primary predictor. The first model adjusted for the patient-level characteristics of age at diagnosis, sex, and race/ethnicity; an additional model further adjusted for county-level factors, such as binge drinking, smoking, and obesity. Multiplicative effect modification was tested using individual-level factors.
Key Findings
The analysis included more than 38,000 cases of hepatocellular carcinoma. The mean rate of county-level food insecurity was 12.9% (standard deviation [SD] = 3.6%). Referring to a mapped representation of the sample characteristics, Dr. Kehm stated: “You start to see some potential clustering of both food insecurity and hepatocellular carcinoma incidence rates, with the southwestern region of the U.S. being a potential hotspot.”
When measuring food insecurity as a continuous variable (per 1-SD increase), it appeared to be significantly associated with the incidence of hepatocellular carcinoma in both the patient-level (incidence rate ratio [IRR] = 1.13) and fully adjusted (IRR = 1.08) models. According to Dr. Kehm, looking at food insecurity as a categorical variable “further confirmed that there may be this dose-response relationship where counties with higher-level food insecurity also have higher rates of hepatocellular carcinoma.” The IRRs for medium (> 11.1% to < 14.5%) and high (≥ 14.5%) food insecurity were 1.13 and 1.28 in the patient-level model (vs low [≤ 11.1%]; P trend = .01) and 1.07 and 1.12 in the fully adjusted model (vs low; P trend = .15), respectively.
Association Stratified by Patient-Level Characteristics
An analysis investigating whether the patient-level characteristics of race/ethnicity and sex may impact the relationship between county-level food insecurity and hepatocellular carcinoma risk revealed no evidence of effect modification.
“However, an interesting interaction emerged when we looked at the association stratified by age at diagnosis,” Dr. Kehm remarked, highlighting that a correlation between food insecurity and hepatocellular carcinoma risk was seen only in the group of patients who were younger vs older than age 65. “They are exploratory data and an ecologic analysis, but we think this is one finding that needs to be further explored, especially given that we know this younger age group has experienced the greatest increase in hepatocellular carcinoma incidence over time.”
Regarding the association stratified by stage at diagnosis, Dr. Kehm stated: “We found that for localized, regional, and distant-stage disease, there was an association with food insecurity. However, our IRRs were stronger for distant-stage disease. This could possibly be because we know food insecurity is associated with other socioeconomic factors that could result in delayed access to care or [poorer] health-care quality.” She noted that this should be further investigated in future studies.
Dr. Kehm concluded: “This provides some preliminary evidence supporting an association between food insecurity and hepatocellular carcinoma risk, specifically in the U.S. and particularly in younger adults. We fully recognize the caveats and limitations of ecologic studies, so these findings need to be confirmed in studies with individual-level data.”
DISCLOSURE: Dr. Kehm reported no conflicts of interest.
REFERENCES
1. Kehm RD, Vilfranc CL, McDonald JA, et al: County-level food insecurity and hepatocellular carcinoma risk: A cross-sectional analysis. 2024 AACR Conference on the Science of Cancer Health Disparities in Racial/Ethnic Minorities and the Medically Underserved. Abstract PR002. Presented September 22, 2024.
2. Kehm RD, Vilfranc CL, McDonald JA, et al: County-level food insecurity and hepatocellular carcinoma risk: A cross-sectional analysis. Cancer Epidemiol Biomarkers Prev 33(suppl):PR002, 2024.
