Cord Blood DNA Methylation and Large-for-Gestational-Age Birth: A Pilot Epigenome-Wide Study in the GROW Cohort

Background/Objectives: Large-for-gestational-age (LGA) birth is associated with adverse perinatal outcomes and increased risk of metabolic disease later in life. Despite these risks, epigenetic studies of LGA remain limited, particularly those using umbilical cord blood DNA methylation (DNAm) as the tissue of interest. Methods: We conducted an epigenome-wide association study of cord blood DNAm in a nested case-control sample from the Gestational Regulators of Weight (GROW) cohort. The analysis included 28 term LGA infants and 63 term appropriate-for-gestational-age (AGA) controls. DNAm was measured using the Illumina HumanMethylation450 BeadChip. Epigenome-wide association analyses were performed with adjustment for residual principal components and estimated cord blood cell-type proportions. Results: No CpG sites reached statistical significance after false discovery rate correction. Twenty-four CpGs showed nominal associations with LGA status at p-value < 1.00 × 10^-4, including five CpGs with absolute methylation differences greater than 0.05. Among annotated loci, the strongest interpretable signals included cg06750897 in PBX1 (Δβ = 0.110, p-value = 1.70 × 10^-5) and two nearby CpGs in SMAD3, cg23731272 (Δβ = 0.111, p-value = 4.31 × 10^-5) and cg02486855 (Δβ = 0.109, p-value = 9.45 × 10^-5). These two SMAD3 CpGs showed concordant hypermethylation in LGA infants and formed a candidate regional methylation signal in exploratory targeted regional analysis. Conclusions: In this pilot study, LGA was associated with modest cord blood DNAm differences, although no CpG sites reached genome-wide significance. The strongest interpretable signals involved PBX1 and SMAD3, suggesting candidate loci for further evaluation in larger studies of fetal overgrowth.