Effect of Environmental PFAS Exposure on Pediatric Neuromorphometry: The HOME Study

BACKGROUND Per- and poly-fluoroalkyl substances (PFAS), also known as “forever chemicals”, is a class of poorly biodegradable chemicals widely used in commercial and industrial applications. The neurodevelopmental effects of PFAS have not been previously studied, warranting investigation of this prevalent and persistent chemical class.

The Health Outcomes and Measures of the Environment (HOME) Study is an ongoing pregnancy and birth cohort study (enrollment from 2003-2006) designed to evaluate known and potentially adverse environmental exposures affecting infant and child health. Thus, we examined the impact of PFAS exposure on pediatric structural MRI (sMRI) imaging acquired in HOME Study participants.

METHODS 468 pregnant individuals were enrolled in the HOME Study. Offspring serum concentrations (ng/mL) of five PFAS (PFOA, PFOS, PFHxS, PFNA, and PFDEA) were measured at multiple time points. High-resolution T1-weighted MRI was performed at 12 years. Post-processing of data was performed in CAT12 software, allowing for voxel-based and surface-based neuromorphometrical analysis. 215 children had complete serum PFAS, sMRI, and covariate information.

We created a general linear model to test for changes in neuromorphometry associated with exposure to log2-transformed PFAS compounds at various time points, while controlling for gender, race, birth weight, and maternal IQ.

RESULTS We found multiple significant (p-value < 0.05) associations with a small Cohen’s effect size (0.2 < d < 0.5) between PFAS exposure and neuromorphometry (at 12 years old). PFOA had a significant effect on GM volume at P3 and P4, while PFNA had a significant effect on GM volume at 11-14 years. PFDEA had a significant effect on cortical surface area at 7-9 years. PFNA had the most significant (p = 0.009) and largest effect size (d = 0.44) on both voxel-based and surface-based morphometry at the 11-14 years.

CONCLUSION PFOA, PFNA, and PFDEA play a significant role in brain-wide neurodevelopment, although there was not sufficient evidence to state that PFOS or PFHxS were influential in our model. There were varied effects depending on the particular exposure time point. Ultimately, this study provides novel evidence regarding the neurodevelopmental impacts of a highly prevalent and persistent chemical class, justifying extensive further investigation into its safety. Limitations of this study included somewhat small sample size and only one imaging time point.

Future analysis will focus on the relationship between PFAS, imaging, and behavioral outcomes. New imaging data at 18 years is also being gathered.