Assessing residential neighborhood exposome exposures and the associations with cancer incidence

点击翻译按钮获取中文摘要

Colorectal, lung, and female breast cancers are major public health burdens and there are clear geographic and racial/ethnic disparities in their incidence. Each of these cancers has been linked to neighborhood factors including socioeconomic deprivation, the built environment, and environmental pollutants in attempts to explain existing disparities in risk. While associations exist, most previous studies considered neighborhood exposures at only one time (e.g., time of diagnosis) and used only single measures of neighborhood exposures (e.g., neighborhood deprivation or singular environmental pollutants), which is a simplification of the multifactorial nature of cancer. As a result, the relative importance of exposure domains is unknown, effects may be underestimated, and a cumulative assessment of risk factors is lacking. Increasing interest in the exposome calls for a more thorough assessment of neighborhood exposures over time that could better explain the factors leading to disparities in cancers. Therefore, we propose to study comprehensive neighborhood disadvantage (ND), a combination of socioeconomic deprivation, racial segregation, environmental pollutant, and built environment domains, to provide stronger evidence of neighborhood associations and identify risk factors that could be modified to eliminate geographic and racial cancer disparities. The important limitations of existing work that this proposal overcomes are 1) the temporality and extent of exposures (i.e., earlier life exposure, later life exposure, or cumulative lifetime exposure), 2) the identification of key neighborhood exposure variables for cancer incidence, and 3) consideration of different domains of the neighborhood exposome over time both independently and in combination to more comprehensively consider relationships with cancer. Our specific aims are to estimate exposure effects for ND domains over time and study the trajectories of these domain exposures and effects by cancer site, race, and sex. We will bring together data from the Virginia and Pennsylvania state cancer registries, population-based controls, residential histories, and exposure data documenting historical disadvantage indicators. Highly innovative aspects of our approach include: 1) examination of historical levels of exposure to several important neighborhood disadvantage domains and 2) the application of novel Bayesian statistical methods that our team has been refining for estimating neighborhood disadvantage and its effects. We hypothesize that our novel approaches to estimating neighborhood disadvantage will better explain variation in cancer incidence than existing methods and will identify the most influential exposure variables over time for each cancer. This study is highly significant as it is the first study to estimate neighborhood disadvantage effects for multiple cancers that considers cumulative risk from several historic exposure domains using residential histories. The expected outcomes of this research will be the identification of historic neighborhood disadvantage exposures associated with significant cancer risk to target for policy development and interventions and to help reduce disparities in cancer.