Evidence for differential thyroid sensitivity from prenatal exposure to polychlorinated biphenyl chemicals in children with autism

Abstract

Background/Aim: Prenatal exposure to polychlorinated biphenyls (PCBs), persistent organic pollutants, is linked to altered thyroid function and adverse neurodevelopmental outcomes, including autism spectrum disorder (ASD). We examined whether neonatal thyroid-stimulating hormone (TSH) mediates the relationship between prenatal PCB exposure and risk of ASD. Methods: We analyzed data from the Early Markers for Autism (EMA) Study, a case-control study of children born during 2000-2003 in Southern California. Children diagnosed with ASD (n=523), Intellectual disability (ID; n=147), and general population controls (n=405) were identified from birth and Department of Developmental Services records. Eleven PCB congeners were measured above the limit of detection in maternal serum collected during the second trimester and modeled as quartiles. Neonatal TSH levels were obtained from state newborn screening records and ln-transformed. We assessed relationships between PCBs and neonatal TSH in models stratified by neurodevelopmental outcome and TSH mediation of the significant relationships we previously observed in EMA between several PCBs and ASD. Results: In the full sample, correlations between PCB concentrations and neonatal TSH were weak and not significant (ranging from 0 to 0.06). However, among children with ASD, we observed significant increases in neonatal TSH associated with an interquartile increase in prenatal levels of four PCBs, including PCB153 (adj-β: 0.21, 95%CI: 0.02-0.39) and PCB138/158 (adj-β: 0.19, 95%CI: 0.01-0.37). We did not observe associations between neonatal TSH and PCBs among children with ID only or controls. Mediation analyses did not suggest that relationships between these PCBs and ASD were mediated by neonatal TSH. Conclusions: Prenatal exposure to PCBs, particularly congeners previously found to be associated with ASD in EMA, were positively associated with neonatal TSH, but only among children subsequently diagnosed with ASD. These findings, needing replication in larger cohorts, may indicate differential fetal sensitivity to thyroid-disrupting chemicals among individuals born, and later-diagnosed, with ASD.