Impact of pre- and postnatal chemical mixture exposures on child neurobehavior and neuroimaging

点击翻译按钮获取中文摘要

Project Summary One in six children in the U.S. and Canada have one or more learning or behavioral problems, such as learning disability, anxiety, autism spectrum disorder, conduct disorder, depression, or attention deficit hyperactivity disorder (ADHD). Early brain development is sensitive to toxicant exposures, including heavy metals, persistent organic pollutants, and endocrine disrupting chemicals. Exposure to mixtures of environmental chemicals is a reality in children, and chemical mixtures may have different modes of action affecting neuronal proliferation, migration, differentiations, synaptic formation/trimming/plasticity, myelination, and neurotransmitters, resulting in adverse impact on the central nervous system. Majority of environmental epidemiologic studies have only examined the impact of a single chemical on neurobehavioral outcomes. Recent development and application of mixture statistical methods will provide great potential to reveal the impact of an individual chemical, interactions between chemicals, and cumulative exposure. These methods have only been applied in limited studies of child neurobehavior and none has been used for neuroimaging outcomes. We will use two existing birth cohorts to examine the impact of both pre- and postnatal exposures to chemical mixtures on child neurobehavior. The Health Outcomes and Measures of the Environment (HOME) Study is a Cincinnati-based birth cohort of 400 pregnant women with children followed up to age 12 years, and the Maternal-Infant Research on Environmental Chemicals (MIREC) is a Canadian study of 1983 pregnant women with children followed up to age 9-11 years. The two North American birth cohorts both measured over 60 environmental contaminants, including lead, mercury, cadmium, arsenic, polybrominated diphenyl ethers, polychlorinated biphenyls, perfluoroalkyl substances, organochlorine and organophosphate pesticides, bisphenol A, phthalates, triclosan, and organophosphate flame retardants, as well as child cognitive abilities (n>1000), behavior (n>1000), and neuroimaging (n=390). We will utilize advanced statistical methods for chemical mixtures, including Elastic Net (ENET) for variable selection, Sparse Partial Least Squares (SPLS) regression for individual chemical effect estimation, and Bayesian Kernel Machine Regression (BKMR) for interactions, nonlinearities, and joint effects. This project will be among the first to test and quantify the potential impact of prenatal and postnatal exposures to chemical mixtures on neurobehavioral and neuroimaging outcomes in well-established cohorts. The results have the potential to greatly increase our understanding of developmental neurotoxicity of chemical mixtures in children and affect environmental health policy making.