DNA methylation-based analysis of leukocyte landscape in children with atopic dermatitis

Atopic dermatitis (AD) is the most common chronic, inflammatory skin condition among the pediatric population. There is a need to better characterize the leukocytic landscape of AD for disease stratification, surveillance, and to identify therapeutic targets. We utilized a novel DNA methylation (DNAm)-based algorithm to study a publicly available, epigenome-wide DNAm dataset of pediatric AD patients and healthy subjects. In this pediatric cohort, AD was associated with significantly lower estimated proportions of memory B, memory CD4 T, naïve CD4 T, naïve CD8 T, and NK cells, but higher proportions of eosinophils and Tregs. Patients with lower IgE (< 500 IU/mL) had lower abundance of naïve B cells and higher abundance of eosinophils. Males had higher eosinophil abundance than female patients. This study corroborates existing on data on leukocytic populations in AD and demonstrates a scalable method for leukocyte quantification, that may be expanded to allow for better correlation to clinical presentation.