A prospective test of biological embedding within a randomized trial of the Nurse Family Partnership intervention: DNA methylome change over infancy and relevance for child development

Children born in the context of poverty have 10 years lower life expectancy and an increased likelihood of having compromised health and development, highlighting the urgent need for effective interventions to reduce health disparities. Supportive perinatal interventions reduce health risk disparities rooted in early experience, but to evaluate interventions’ effects on offspring development, researchers must either wait years for health outcomes to emerge in offspring, or use high-cost, labor intensive assessments such as MRI. Causal evidence of perinatal interventions’ effects on a biological level could provide mechanistic insight into interventions’ influence, contributing non-invasive, proximal indicators to evaluate, compare, and refine interventions for specific populations. However, existing studies on perinatal biological embedding are based primarily on observational data and do not provide the requisite experimental leverage for causal inference. The proposed project capitalizes on a randomized trial of an evidence-based perinatal intervention, the Nurse Family Partnership (NFP) to a) experimentally test the hypothesis of perinatal biological embedding via the mechanism of DNA methylation (DNAm) in offspring and b) identify non-invasive biological indicators that could be used to evaluate perinatal interventions’ early effects on child development. Studies in rodents causally implicate DNAm in in the biological embedding of perinatal adversity. Novel potential biomarkers of perinatal biological embedding in humans include 1) a cross-tissue DNAm signature of glucocorticoid (GC) exposure, developed using hippocampal progenitor cells, and validated using neonatal blood samples and 2) the Pediatric Buccal Epigenetic clock (PedBE), derived from buccal cells, which provides a measure of epigenetic aging, which is associated with compromised infant brain growth and neurodevelopment. Leveraging these two methodological advances and a new funded trial of the NFP, we will implement minimally-invasive buccal cell biosampling from 446 infants born to pregnant individuals (>50% Black), participating in the NFP trial (n=223 randomized to receive the NFP, n=223 control) at baseline and 24 months. We will generate DNAm biomarkers across infancy and paired genetic data using low pass whole genome sequencing. We propose that maternal NFP participation, versus care as usual, will alter the biological embedding of perinatal adversity, as reflected by DNAm biomarkers of GC exposure and pediatric epigenetic aging over infancy, which will correlate with infant development at 24 months. Aim 1: Test whether the NFP causes variation in DNAm at GC-sensitive sites in infants over the first two years of life. Aim 2: Determine whether NFP causes differences in epigenetic age in infants over the first two years of life. Aim 3: Identify DNAm signatures as predictors of infant development at 24 months of age. The results would provide causal evidence for perinatal biological embedding of supportive intervention via DNAm, and contribute to direct, proximal indicators of perinatal interventions’ salutary effects against developmental risk in the context of adversity. This would refine intervention evaluation and inform policy decisions to redress health inequities.