Assessment of effect modification in associations between biomarkers of prenatal glyphosate exposure and child neurodevelopment in Puerto Rico
Background: Glyphosate (GLY)-based herbicides are widely used globally, and GLY and its degradant aminomethylphosphonic acid (AMPA) have been associated with impaired neurogenesis and synaptogenesis in toxicological studies. Additionally, breastfeeding supports neurodevelopment, but limited research on GLY in human milk makes its role as an infant exposure source unclear. Methods: The study population comprised mother-child pairs (n = 284) enrolled between 2011 and 2017 in the Puerto Rico Testsite for Exploring Contamination Threats birth cohort in Puerto Rico and its follow-up studies. Prenatal urinary GLY and AMPA were measured up to three times per individual, adjusted for specific gravity, and summarized as geometric means. Neurodevelopment was assessed using the Battelle Developmental Inventory, 2nd edition in Spanish, at multiple points through age four. Linear mixed-effects and generalized least squares linear models examined associations with GLY/AMPA, including an interaction term for exclusive breastfeeding duration (<90 vs. ≥90 days; n = 109); results are expressed as percent change per interquartile range increase in GLY/AMPA. Results: GLY was associated with cognitive (%change: −2.6; 95% confidence interval [CI] = −4.3, −0.9) and communication (%change: −2.5; 95% CI = −5.0, −0.1) Battelle Developmental Inventory, 2nd edition scores. AMPA was associated with cognitive (%change: −2.8; 95% CI = −4.7, −0.9), communication (%change: −3.1; 95% CI = −5.7, −0.4), motor (%change: −1.8; 95% CI = −3.1, −0.4), and personal-social (%change: −2.0; 95% CI = −3.7, −0.3) scores. We found no significant effect modification by breastfeeding duration, though divergent patterns by analyte were observed. Conclusions: These exploratory findings highlight the need for further research into prenatal GLY exposure, its impact on neurodevelopment, and potential exposure through breastmilk.