Prenatal Exposure to Per- and Polyfluoroalkyl Substances, Metabolic Signatures, and Intellectual Development in Preschoolers: A Birth Cohort Study

Per- and polyfluoroalkyl substances (PFAS) are ubiquitous pollutants with potential neurodevelopmental toxicity. However, epidemiological evidence on prenatal exposure to PFAS particularly emerging PFAS and PFAS isomers in relation to children’s intellectual development remain limited. This study aimed to examine associations of prenatal exposure to a variety of PFAS, including legacy, emerging, and branched PFAS, with children’s intellectual development, and to evaluate early metabolic pathway changes potentially linking prenatal PFAS exposure to these outcomes. In 1,062 mother–child pairs from a prospective cohort in Wuhan, we quantified 32 PFAS in cord sera and conducted metabolomic profiling of cord plasma. At 6 years of age, children’s intellectual development was assessed using the Wechsler Preschool and Primary Scale of Intelligence–Fourth Edition, a standardized and clinically validated instrument capturing multiple intellectual domains. Generalized linear models (GLM) and weighted quantile sum regression models were used to assess the association of individual PFAS and PFAS mixtures with children’s intellectual development. A metabolome-wide association study (MWAS) using GLM was conducted to examine associations between PFAS exposure, individual metabolites, and children’s intellectual development. Then, the meet-in-the-middle framework was applied to evaluate metabolic signatures as potential intermediary biomarkers. After adjusting for confounders, each doubling in 3,4,5m-PFOS concentration was associated with a decreased Full Scale Intelligence Quotient (FSIQ) of 0.92 (95% CI = −1.68, −0.17) points, with similar associations observed for the Cognitive Proficiency Index (CPI), Processing Speed Index (PSI), and Working Memory Index (WMI). Similarly, each doubling in 6m-PFOS was associated with decreased scores of 1.22 (95% CI = −2.02, −0.42) for CPI, 1.05 (95% CI = −1.82, −0.27) for PSI, and 0.89 (95% CI = −1.76, −0.01) for WMI. Furthermore, significant mixture effects of multiple PFAS on CPI and WMI were observed, with 6m-PFOS and 3,4,5m-PFOS identified as the primary contributors, respectively. Metabolomic analyses revealed that prenatal PFAS exposure was linked to changes in lipid metabolism pathways, potential serving as candidate intermediary biomarkers in linkages between prenatal PFAS and childhood intellectual development. In conclusion, prenatal PFAS exposure, particularly to branched PFOS isomers, was associated with lower domain-specific intellectual performance in childhood. Integrative analyses combining exposure and metabolomic data may advance the understanding of PFAS-related neurodevelopmental risk.