Environmental Pollutants and Markers of Autoimmunity as Risk Factors for Autism Spectrum Disorders in California's Central Valley

Background: The etiology of autism spectrum disorders (ASD) appears to be influenced by a complex set of genetic and environmental factors, which may be interacting with each other to determine risk.  While life in the modern industrialized world involves exposure to myriad chemicals and pollutants on a daily basis, few of these have been well-studied with respect to ASD.  Current evidence suggests a relationship between autoimmunity and ASD; chemical exposures that effect autoimmunity may exacerbate any such relationship. 
Objectives: To assess potential interactions between selected chemical exposures and neonatal levels of autoantibodies in association with risk of autism spectrum disorders.
Methods: We measured IgG autoantibodies to nervous system and thyroid antigens in neonatal blood spots from ASD cases (N=384) identified through the California Department of Developmental Services and population controls (N=596) born between 1996 and 1998 in California’s Central Valley.  Estimates of ambient levels of hazardous air pollutants (HAPs) and measures of regional pesticide applications were derived through linkage to existing databases, based on mother’s residence at the time of delivery.  Interactions between exposures and autoantibodies were explored using logistic regression models to calculate odds ratios (ORs) for ASD risk adjusted for demographic and birth characteristics.  A total of 228 autoantibody-chemical combinations were tested in exploratory models including interaction terms; autoantibody-ASD ORs, stratified by exposure level, were computed based on significant p-values for tests of interaction.  Associations of these chemical exposures with ASD risk and associations between exposure and autoantibodies were also explored.
Results: We previously observed inverse associations between autoantibody levels and odds of ASD in this sample.  Consistent with this, GFAP and MBP autoantibody levels above the 95^th percentile were associated with significantly reduced odds of ASD in the lower exposure categories (quartiles 1-3 for HAPs) of chemical groups including immunotoxicants, neurotoxicants, and chlorinated solvents.  However, in the higher exposure categories (quartile 4) of these chemical groups, ORs for ASD and GFAP and MBP autoantibody levels above the 95^th percentile were associated with increased risk for ASD.  For example, where chlorinated solvent levels fell in the lower three quartiles, the OR for ASD-MBP autoantibody association was 0.33 (95% CI (0.14, 0.80)); for the 4^th quartile of exposure, the OR for ASD-MBP autoantibodies was 2.80 (95% CI (0.60, 13.1)).  Other findings included associations of both chlorinated solvents in ambient air and organochlorine pesticides with levels of thyroid peroxidase antibodies above the 95^th percentile (OR (95% CI) = 3.0 (1.02, 8.79) and OR (95% CI) = 7.23 (1.57, 33.2), respectively). 
Conclusions: Exposure to environmental pollutants may modify associations between autoantibodies and ASD.  This could indicate that additional immunologically active stimuli are required to potentiate autoantibody-mediated processes involved in ASD etiology.  However, as these were exploratory models comparing multiple exposure-autoantibody combinations, further research will be required to fully understand these complex relationships. Funding sources: Autism Speaks and the California Department of Public Health