Corneocyte topography/sweating axes–based clustering of atopic dermatitis reveals a distinct immune signature and parallels the treatment response to dupilumab
Need to claim your poster? Find the KiKo table at the conference and they'll help
you get set up.
Presented at: Society for Investigative Dermatology 2025
Date: 2025-05-07 00:00:00
Views: 2
Summary: Abstract Body: We recently demonstrated that corneocyte surface topography/sweating-response axes have potential as candidate biomarkers for the assessment of physical barrier defects in skin with moderate-to-severe atopic dermatitis (AD). Using a machine learning approach, we identified three characteristic phenotypes of AD with distinctive patterns of impaired sweating responses and disrupted skin surface microstructure. In the present study, we investigated whether the immune signature expressed in non-lesional tape-stripped AD skin could be linked to a distinct cluster. Cluster I was a non-inflammatory phenotype, whereas clusters II and III were inflammatory phenotypes. Cluster II was more inflammatory, with greater transepidermal water loss. Cluster III had the lowest sweating responses and the highest skin surface microstructure disruption score. Discrete clusters of AD corresponded well to their specific immunologic gene profiling in the tape strips: filaggrin was downregulated in cluster I, while greater IL-13 skewing was observed in cluster III and marked upregulation of the Th 1, Th 2, and Th 22 axes was observed in cluster II. A total of 22 patients in each cluster were treated with dupilumab for 24 weeks. The response to dupilumab was maintained at 1 year after stopping dupilumab in half of patients in cluster I, whereas no patients in cluster II and III maintained treatment response. After dupilumab treatment, cluster I showed normalization of downregulated filaggrin expression, whereas cluster III showed no significant reduction in IL-13 despite clinical improvement and AD relapsed 4 weeks after stopping dupilumab. Our clusters represent distinct endotypes of AD, which may be linked to their unique inflammatory profiles. Thus, we can select the most optimal treatment among the different targeted therapies according to the cluster. Kenta Nakamoto<sup>1</sup>, Tetsushi Koide<sup>2</sup>, Yumi Aoyama<sup>1</sup> 1. Dermatology, Kawasaki Ika Daigaku, Kurashiki, Okayama Prefecture, Japan. 2. Integrated Systems Research Division, Hiroshima Daigaku, Higashihiroshima, Hiroshima Prefecture, Japan. Epidermal Structure and Barrier Function