Exposure to maternal human milk components interacts with infant polygenic risk to predict childhood atopy

Although human milk (HM) is known to benefit infant health, how its many bioactive components (e.g., microbiota (HMM), oligosaccharides (HMOs), and fatty acids (HMFAs)) influence the risk of childhood atopy remains poorly understood. Our study integrated polygenic risk scores (PRS) of nursing infants with human milk components (HMCs) from their mothers in the CHILD Cohort Study to explore potential gene-milk interactions associated with childhood atopy at ages 1-5. For example, among infants with high PRS, exposure to increased abundance of the microbe Abiotrophia was associated with increased childhood atopy prevalence. Moreover, network analyses combining all HMCs revealed that a module of correlated HMM and HMOs including Abiotrophia and 3-fucosyllactose, respectively, was associated with reduced childhood atopy prevalence. Another network module of correlated HMCs (e.g., Bifidobacterium longum, 2'-fucosyllactose, and eicosapentaenoic acid) was associated with reduced childhood atopy prevalence among infants with high PRS. Finally, a gradient-boosting model (GBM) integrating HMCs and infant PRS predicted childhood atopy with an area under the curve (AUC) of 0.78, outperforming GBMs using any individual HMC or infant PRS alone (AUC range: 0.48-0.63). Our study reports both main and interaction effects of HMCs and infant PRS on childhood atopy. An improved understanding of how early-life exposures such as HMCs impact the health of children differently depending on their genomic profiles may facilitate the development of personalized intervention strategies to reduce the burden of these health outcomes during childhood.