Comprehensive Transcriptomic Analysis of Atopic Dermatitis Patients Documents the Spectrum of Molecular Abnormalities and the Response to Treatment

ABSTRACT Atopic dermatitis (AD) is a chronic inflammatory skin disease characterized by immune dysregulation and barrier dysfunction. To define the molecular architecture of AD in greater detail, we integrated lesional (LES) and non-lesional (NLS) transcriptomic data from multiple datasets using gene expression data from normal skin and psoriasis (PSO) and nummular eczema (NME) cohorts as reference. Gene set variation analysis revealed that adult AD exhibits broad immune activation and consistent barrier impairment in both LES and NLS skin, whereas pediatric AD is dominated by IL-1–driven inflammation with minimal barrier alteration. Comparative analyses showed stronger Th2 and myeloid activity in AD, metabolic enrichment in PSO, and complement and NK cell activation in NME. Longitudinal profiling identified temporal variation in Th1, Th2 and IFN pathways in AD skin. An eczema immune and cellular score, ECZECIS, was developed to quantify transcriptomic abnormalities and correlated with clinical improvement following dupilumab therapy. Among all treatments analyzed, dupilumab produced the most extensive reduction of immune and cytokine pathway activity in skin and attenuated systemic immune activation in blood. These findings delineate distinct immune and barrier signatures across age groups and disease types and establish ECZECIS as a quantitative biomarker for monitoring molecular treatment response in AD.