BACKGROUND AND AIM: The relationship between environmental exposures and neurodevelopmental outcomes has been characterized, but the modifying role of microbiome is unknown. We probed prior evidence in the New Hampshire Birth Cohort Study (NHBCS) suggesting relationships between arsenic and neurobehavioral development, which were both associated with the infant gut microbiome in separate studies. METHODS: Total arsenic was quantified in six-week-old toenail clippings of NHBCS participants. Three years postpartum, parents completed the Social Responsiveness Scale (SRS-2), which measures autism-related social behaviors on a continuous scale standardized to a mean of 50 (standard deviation 10) with higher scores indicating more maladaptive behavior. Metagenomic sequencing was performed on infant stools collected at six weeks and one year of age. The ten bacterial species and functions with the highest relative abundances at each time were dichotomized at the median. Separate linear models including interaction terms between exposure and each dichotomous microbiome feature were constructed to examine the modifying role of the microbiome on the arsenic-SRS-2 relationship, adjusting for five other trace elements (copper, manganese, lead, selenium, zinc) and sociodemographic characteristics. RESULTS:The 78 participants with exposure, outcome, and microbiome data, rated better on the SRS-2 than the normative population (mean=44, SD=5).The relative abundances of several functional pathways identified in six-week stool samples appeared to modify the arsenic-SRS-2 association, including the S-adenosyl-L-methionine (SAM) cycle; among infants with high relative abundance of SAM genes there was no association [β=-0.66, 95%CI (-2.41,1.08)], whereas among infants with low relative abundance there was suggestive evidence of an adverse association [β=2.24, 95%CI (-0.04,4.51)], pinteraction=0.046]. Effect modification by individual species in six-week stool samples and features of the one-year-old microbiome was also identified in preliminary analyses. CONCLUSIONS:Our findings indicate that the infant gut microbiome may alter neurodevelopmental susceptibility to environmental exposures. Potential effect modification by SAM genes suggests the gut microbiome may contribute to arsenic methylation/excretion. KEYWORDS: Microbiome, Children's environmental health, Neurodevelopmental outcomes, Chemical exposures