INTRODUCTION
Atopic dermatitis (AD) is a common inflammatory skin disease affecting up to 10% of adults and 20% of children worldwide.1,2 Our knowledge of AD pathomechanisms is largely derived from molecular studies in skin biopsies that demonstrate primarily Th2/Th22 inflammation with variable Th1/Th17 skewing, as well as epidermal barrier dysregulation, including epidermal hyperplasia and abnormalities in terminal differentiation, tight junctions, and lipid biosynthesis/metabolism.3-12 While skin biopsies were instrumental to deciphering disease characteristics, they are associated with pain, scarring, and infection risk, restricting their use in large-scale clinical trials and longitudinal studies, which are critical for adults and even more in children.13-15 Tape-strips are a minimally invasive, nonscarring approach to collect skin samples to the level of the upper stratum granulosum.16 Recent studies have demonstrated significant utility of tape-strips in profiling AD skin in both adults and children,6,16-30and in monitoring disease activity in response to treatment.31-34
Comprehensive molecular intra-patient comparison of tape-strips to full-thickness biopsies is currently lacking. Moreover, there is no detailed comparison of immune, barrier, and itch-related markers between the two techniques. 16,35,36 Several studies described differences between tape-strips and biopsies using a limited panel of markers. Andersson et al. analyzed only 17 immune markers in AD skin and noted that CCL17 and markers of innate immunity were significantly up-regulated and correlated with disease activity in tape-strips but not in same-cohort biopsies.36 Simonsen et al. assessed 10 cytokines in AD lesions and noted greater up-regulation in the innate immunity marker IL-1B in tape-strips, while dysregulation of Th2 markers and IFN-γ were only detected in biopsies.35 Of note, this study measured mRNA expression in biopsies while proteomics was performed in tape-strips. Kim et al. measured mRNA of 5 epidermal differentiation markers in tape-strips, reporting positive correlations with protein levels in biopsies.16 Utilizing RNAseq, Sølberg et al. reported that structural genes like keratins were higher in biopsies while immune markers IL-8 and IL-36A were higher in tape-strips.18 Dyjack et al. analyzed tape-strips from non-lesional AD skin, noting better representation of cornification genes in tape-strips compared to biopsies.37 These studies enzymatically separated the epidermis and dermis in biopsies, potentially overlooking important pathogenic features within the entire AD skin transcriptome.
Herein, we present a global RNA-seq intra-patient profiling of tape-strips and whole-skin biopsies in moderate-to-severe AD as compared to healthy subjects. Our data highlight features that are better captured by tape-strips, including Th2 polarization, itch pathway activation, and terminal differentiation impairment, whereas Th22 changes and epidermal hyperplasia are more pronounced in biopsies. These findings suggest that tape-strips and biopsies each preferentially detect unique disease characteristics, which should be considered when selecting skin sampling approaches in clinical trials or longitudinal studies.