Abstract B008: Neighborhood fast food, alcohol, and fruit/vegetable expenditure and early-onset colorectal cancer: A statewide spatio-ecological analysis in Alabama

Introduction: U.S. incidence of early-onset colorectal cancer (eoCRC) has increased by 50% since the mid-1990s, yet underlying causes remain poorly understood. Shifts in food consumption patterns and westernized diets have been postulated as contributors. This study utilizes unique neighborhood-level food expenditure data and cancer registry data to examine relationships between fast food, alcohol, and fruit/vegetable expenditures, and eoCRC incidence. Methods: Colorectal cancer incidence data from 2010-2019 was obtained from the Alabama Statewide Cancer Registry. Annual average fast food, alcohol, fruit/vegetable, and total food/beverage expenditure data by block group were sourced from 2025 Esri data, based on U.S. Bureau of Labor Statistics consumer expenditure surveys. Model covariates included area deprivation index, percentages of non-Hispanic Black, Hispanic, female, and uninsured residents (2019 American Community Survey 5-year estimates), and 2010 USDA rural-urban commuting area codes. The primary outcome was the ratio of eoCRC cases (<50 years at diagnosis) to average-onset colorectal cancer (aoCRC) cases (50+ years at diagnosis) across 3,357 block groups in Alabama. Primary predictors (fast food, alcohol, fruit/vegetable expenditure) were converted to proportions of the total food/beverage expenditure per block group. A hierarchical Bayesian spatial hurdle model was fit, where inverse-variance weighting was employed to synthesize fixed effects. The same analyses were performed for late-stage CRC (regional/distant SEER Summary Stages). Results: Of 24,926 new CRC cases in Alabama from 2010-2019, 12,727 were late-stage cases, 2,607 were eoCRC, and 1,596 were late-stage eoCRC. 61.2% of eoCRC cases were diagnosed at late stages, while 49.8% of aoCRC cases were late stage. Across Alabama block groups, an annual mean of 17.3% of food/beverage expenditures were allocated to fast food, 10.8% to fruits/vegetables, and 5.2% to alcohol. A 1% increase in mean fast food expenditure per block group was associated with a 14.25% increase (all stage; 95% Credible Interval: 9.72% to 18.75%) and 5.23% increase (late stage; 95% CrI: 3.03% to 11.16), respectively, in the proportion of cases that were eoCRC rather than aoCRC. In contrast, a 1% increase in mean fruit/vegetable expenditure corresponded to a 17.72% decrease (all stage; 95% CrI: -31.20% to -4.33%) and 33.77% decrease (late stage; 95% CrI: -50.95% to -16.58%), respectively, in eoCRC cases. Alcohol expenditure was not significantly associated with an increased or decreased share of all- or late-stage eoCRC cases. Conclusions: Higher fast food expenditure at the neighborhood level was strongly associated with a greater proportion of both all-stage and late-stage eoCRC in Alabama, while higher fruit/vegetable spending appeared strongly protective. These findings provide preliminary ecological evidence that westernized dietary patterns may contribute to increased eoCRC risk. Further individual-level research is warranted to better understand how specific food exposures influence eoCRC risk. Citation Format: R. Blake. Buchalter, S.M. Qasim. Hussaini, Mahak Bhargava, Geetanjali Saini, Nashira I. Brown, Mackenzie E. Fowler, Ritu Aneja. Neighborhood fast food, alcohol, and fruit/vegetable expenditure and early-onset colorectal cancer: A statewide spatio-ecological analysis in Alabama [abstract]. In: Proceedings of the AACR Special Conference in Cancer Research: The Rise in Early-Onset Cancers—Knowledge Gaps and Research Opportunities; 2025 Dec 10-13; Montreal, QC, Canada. Philadelphia (PA): AACR; Clin Cancer Res 2025;31(23_Suppl):Abstract nr B008.