1696: Multi-Omic Profiling Reveals Immune Cell Priming Signature Linked to Systemic Lupus Erythematosus Prognosis

Michael A. Smith¹, Dominic Sinibaldi², Saifur Rahman¹, Chia-Chien Chiang², Anna M. Hansen¹, Jill Henault¹, Carlos P. Roca³, Shu Wang¹, Kamelia Zerrouki¹, Rebecca Filippi¹, Christopher Groves¹, Zerai Manna⁴, Jun Chu⁴, Michael Davis⁴, sarthak gupta⁴, Christopher Morehouse¹, Melissa De los Reyes¹, Rachel Ettinger¹, Roland Kolbeck¹, Mariana Kaplan⁵, Miguel A. Sanjuan¹, Richard M. Siegel⁶, sarfaraz Hasni⁴ and Kerry A. Casey¹, ¹BioPharmaceuticals R&D, AstraZeneca, Gaithersburg, MD, ²Data Science and AI, AstraZeneca, Gaithersburg, MD, ³Data Science and AI, AstraZeneca, Cambridge, United Kingdom, ⁴Lupus Clinical Trials Unit, National Institute of Arthritis and Musculoskeletal and Skin Diseases, National Institutes of Health (NIH), Bethesda, MD, ⁵Lupus Clinical Trials Unit, National Institute of Arthritis and Musculoskeletal and Skin Diseases, National Institutes of Health (NIH); Systemic Autoimmunity Branch, National Institute of Arthritis and Musculoskeletal and Skin Diseases, National Institutes of Health (NIH), Bethesda, MD, ⁶Office of the Clinical Director, National Institute of Arthritis and Musculoskeletal and Skin Diseases, National Institutes of Health (NIH), Bethesda, MD

Background/Purpose: Much of our understanding of systemic lupus erythematosus (SLE) immunopathogenesis is derived from gene profiling studies, where core pathways such as neutrophil dysregulation and uncontrolled type I interferon (IFN) production by plasmacytoid dendritic cells (pDCs) have been identified. However, gene signatures found in whole blood may not reflect those in the tissues, thus potentially overlooking key drivers of disease activity and pathogenesis. While gene expression is commonly used as a proxy for protein abundance, significant discordance can exist between transcript and protein levels. We hypothesized that multidimensional blood profiling could identify new biology and reveal key molecular signatures that underlie distinct SLE pathologies.

Methods: Whole blood samples from 87 patients with mild to moderate SLE and 48 matched healthy controls were analyzed (discovery cohort). To ensure robustness of data, samples from an independent cohort of 43 patients and 52 healthy cohorts were also analyzed (validation cohort). All samples and data were collected after obtaining informed consent. Whole blood gene expression, serum proteins, autoantibodies, and immune cell assessments were collected with longitudinal clinical data. Statistical analyses were carried out in R v. 3.5.1. All data were made available in an interactive SLE Immune Atlas.

Results: We identified three protein signatures relating to interferon (IFN) signaling, granulocyte activation, and immune cell priming (IAI, GRN, and ICP) that significantly associated with disease status and unique organ-specific manifestations in SLE patients. Type I IFN activity (21-IFNGS) was present in 69.0% of patients, the IAI, GRN, and ICP signatures were present in 52.9%, 35.6%, and 42.5% of patients, respectively. While the IAI and GRN signatures correlated with transcripts, no strong gene correlates of the ICP signature were found. The ICP signature featured proteins associated with antigen presenting cells and kidney injury and correlated with proteinuria (p=1.1E^–3), decreased GFR (p=5.0E–7), and active nephritis (p=0.044). The ICP signature was predictive of damage accrual as measured by the Systemic Lupus International Collaborating Clinics/American College of Rheumatology Damage Index (SDI) over a 6-year follow-up (RR= 1.92, 95% CI 1.41–2.67). The ICP-positive patients were more likely to exhibit worsening in the renal, pulmonary, ocular and skin domains of SDI.

Conclusion: Our analyses identified a previously uncharacterized ICP protein signature enriched in patients with renal involvement, predictive of disease worsening and organ damage over time. Prospective studies would be useful to confirm the ability of the ICP signature to identify patients in need of therapeutic intervention to prevent increased risk for damage accrual.


M. Smith: AstraZeneca, 3, 11, Horizon Therapeutics, 3, 11; D. Sinibaldi: AstraZeneca, 3, Neuraly, 3; S. Rahman: Sanofi, 3; C. Chiang: AstraZeneca, 3; A. Hansen: AstraZeneca, 3; J. Henault: None; C. Roca: AstraZeneca, 3, CSL Behring, 3; S. Wang: Horizon Therapeutics, 3; K. Zerrouki: None; R. Filippi: AstraZeneca, 3; C. Groves: Q Squared Solutions, 3; Z. Manna: None; J. Chu: None; M. Davis: None; s. gupta: None; C. Morehouse: None; M. De los Reyes: AstraZeneca, 11; R. Ettinger: AstraZeneca, 3, 11, Horizon Therapeutics, 3, 11, VielaBio, 3, 11; R. Kolbeck: None; M. Kaplan: AstraZeneca, 5, Bristol Myers Squibb, 5, Cytrill, 2, Neutrolis, 2; M. Sanjuan: AstraZeneca, 3; R. Siegel: Novartis, 3, 11; s. Hasni: AstraZeneca, 5; K. Casey: AstraZeneca, 3, 11, Regeneron, 3, 11.