OP0096 IMMUNOPHENOTYPING-BASED TRANSCRIPTOME ANALYSIS ON SYSTEMIC IMMUNE-MEDIATED DISEASES

Background Systemic immune-mediated diseases vary in manifestations and pathogenesis. Currently, disease diagnoses treatments are generally based on clinical symptoms some specific serological features. Because the share pathological features, it is difficult to comprehend pathogenesis manage them appropriately. To date, a limited number of studies tried stratify such patients into subgroups using transcriptomic data molecular patterns [1]; however, included diseases, patients, transcriptome were restricted. Objectives identify significant immune cells forming then all accordingly, immunophenotyping data. Methods We performed an integrative analysis bulk RNA-sequence (RNA-seq) peripheral mononuclear blood (PBMC) from diagnosed with systemic lupus erythematosus (SLE), mixed connective tissue (MCTD), idiopathic inflammatory myopathy (IIM), sclerosis (SSc), rheumatoid arthritis (RA), large vessel vasculitis (LVV) Immune Cell Gene Expression Atlas University Tokyo (ImmuNexUT) [2]. Immunophenotyping (29 cell subsets) applied machine learning approaches, random forest method, stratified by k-means clustering. Then, we conducted analysis, including differentially expressed gene (DEG) enrichment (5,484 RNA-sequencing data). Finally, compared features among clusters ( Figure 1A ). Results A total 254 -78 SLE, 22 MCTD, 63 IIM, 52 SSc, 20 RA, 19 LVV- included. important distinguish algorithm was applying Using top 15 cells, clustering performed, three 1B Naïve CD8 + T memory CD4 non-classical monocytes predominantly 1, 2, 3, respectively 1C Each showed skewed distribution each cluster, that SLE cluster 2 most SSc 3 Next, DEG 5,484 found DEGs increased when other clusters, which consistent proportions 1D The ontology genes associated adaptive response naïve activation innate vascular process circulatory system suggesting even same can be Uniquely, common as arthritis, positivity anti-SS-A antibody, spectrums MCTD accumulated different, cluster. Conclusion This large-scale integration six revealed classified one has It may enhance understanding heterogeneous diseases. References [1]Barturen G, et al. Integrative Analysis Reveals Molecular Stratification Autoimmune Diseases. Arthritis Rheumatol. 2021;73:1073-1085. [2]Ota M, Dynamic landscape cell-specific regulation Cell. 2021;184:3006-3021.e17. 1. Acknowledgements work supported Grant-in-Aid for JSPS Fellows. Disclosure Interests Shinji Izuka: None declared, Toshihiko Komai Speakers bureau: Receives honoraria Chugai Pharmaceutical Co., Ltd., Tokyo, Japan., Takahiro Itamiya: Mineto Ota Employee of: Belong Social Cooperation Program, Department Functional Genomics Immunological Diseases, Pharmaceutical., Saeko Yamada: Yasuo Nagafuchi Hirofumi Shoda Kosuke Matsuki Employees Kazuhiko Yamamoto: Tomohisa Okamura Keishi Fujio Consultant consulting Grant/research support from: research Pharmaceutical.