O24 Multiomic triangulation of biomarkers for atopic dermatitis severity

Abstract Introduction and aims Atopic dermatitis (AD) is a common inflammatory skin disease with varying severity. In this large-scale multiomics study, we used a set of complementary methods and datasets to identify molecular markers with the most robust evidence for involvement in AD severity. Methods We harmonized 11 cohorts with AD severity scores or systemic therapy use/hospitalization as proxies for severity and undertook a case-only genome-wide association study (GWAS) meta-analysis (N = 100 766) and subsequent transcriptome-wide association study (TWAS) with expression quantitative trait loci data, differential expression meta-analysis in blood (n = 340) and skin (n = 185) as well as a differential proteomic analysis in blood (n = 75). Results A total of 440 genes/proteins showed evidence of association with AD severity in at least one of the analyses. Four blood biomarkers demonstrated associations with AD severity in two or more analyses. All four were significant in the gene expression analysis and had consistent genetic evidence from the TWAS analysis, suggesting a causal relationship. AD severity was associated with increased CEP85 expression (Pexp = 2.8 × 10−7; Ptwas= 8.2 × 10−13) and decreased expression of IL18RAP (Pexp= 2.0 × 10−5; Ptwas= 2.6 × 10−11), PTGDR (Pexp = 3.6 × 10−8; Ptwas = 6.2 × 10−7) and TRIM26 (Pexp= 3.1 × 10−6; Ptwas= 9.2 × 10−31). All four genes have prior evidence for involvement in atopic disease, with IL18RAP and PTGDR specifically implicated in AD. However, we show for the first time their importance in AD severity. IL18RAP and TRIM26 are involved in inflammatory pathways, and PTGDR has been investigated as a potential drug target for AD due to its role in prostaglandin D2 signalling. CEP85 represents a more novel AD pathway. We also found strong evidence (posterior probability > 80%) that the genetic variants affecting AD severity at this locus also affect CEP85 expression in monocytes. Conclusions We have identified four candidate biomarkers (IL18RAP, CEP85, PTGDR and TRIM26) with robust evidence for a role in AD severity. These warrant further experimental validation to explore opportunities for early proactive intervention in the trajectory of severe AD.