Multi-omic triangulation identifies molecular candidates of atopic dermatitis severity

Atopic dermatitis (AD) is a common skin disease with most of the health, social and economic impact driven by those with more severe disease. Determining the molecular pathways that influence severity is therefore crucial, offering opportunity to identify novel drug targets, as well as use in risk prediction tools. In this large-scale multi-omics study, we used complementary methods and datasets to identify molecular markers with robust evidence for involvement in AD severity. We undertook a case-only genome-wide association study meta-analysis (N=100,766) and subsequent transcriptome-wide association study (TWAS), differential expression meta-analysis in blood (N=340) and skin (N=185) as well as a differential protein abundance analysis in blood (N=75). A total of 440 genes/proteins showed evidence of association across all the analyses. Of these, four were significant in two or more analyses. For CEP85 (Pexpression=2.8x10-7; PTWAS=8.2x10-13), a gene not previously associated with AD, we also found strong evidence that the genetic variants affect CEP85 mRNA expression in monocytes. Functional in vitro follow-up showed that CEP85 over-expression in monocyte-derived macrophages can disrupt phagocytosis which we hypothesise may contribute to severity by impairing phagocytosis of S aureus. Together this work provides evidence of genetic risk and candidate molecular pathways to severe AD.