THU0249 Functional Clusters of Autoantibodies Targeting TLR and Smad Pathways Define New Subgroups in Systemic Lupus Erythematosus. (15th July 2016)

• Record Type: Journal Article
• Authors: Lewis, M.J.; McAndrew, M.B.; Wheeler, C.; Workman, N.; Agashe, P.; Koopmann, J.; Uddin, E.; Zou, L.; Stark, R.; Anson, J.; Cope, A.P.; Vyse, T.J.
• Is Part Of: Annals of the rheumatic diseases. Volume 75(2016)Supplement 2
• Page Start: 278
• Page End: 278
• Publication Date: 2016-07-15
• Subjects: Rheumatism -- Periodicals; 616.723005
• DOI: 10.1136/annrheumdis-2016-eular.6111
• Languages: English
• ISSNs: 0003-4967

Abstract

Abstract : Background: The molecular targets of the vast majority of autoantibodies in systemic lupus erythematosus (SLE) are unknown. Multiple recent failures of new therapies in SLE clinical trials demonstrate the need for improved classification of SLE based on pathogenesis. Objectives: To identify novel autoantibodies in SLE for the development of an improved assay for diagnosis and identification of autoantibody clusters for classification of SLE subgroups. Methods: We used a baculovirus-insect cell expression system to create an advanced protein microarray with 1545 full-length human proteins. All proteins were expressed with a biotin carboxyl carrier protein (BCCP) folding tag, so that only correctly folded proteins were bound to the array. Sera were assayed from three independent cohorts of SLE individuals (total n=277) and age, gender and ancestry-matched controls (n=280). Results: We confirmed 116 proteins (FDR<0.01) as autoantigens in SLE, of which 13 had been previously reported, hence we identified 103 novel SLE autoantigens. The 116 SLE autoantigens clustered into four distinct groups by unsupervised hierarchical clustering. The presence of these four clusters was confirmed by principal component analysis. Protein-protein interaction network analysis showed that the autoantigens in each of the four clusters were functionally linked. The original SLE autoantigens Ro60, La, and SMN1/Sm complex formed a distinct antigen cluster (SLE1a), extended by a second cluster (SLE1b) of proteins involved in RNA/DNA/chromatin processing. The largest two clusters of novel autoantigens revealed two networks of interconnected proteins: the receptor-regulated SMAD2, SMAD5 and proteins linked to TGF-β signalling (SLE2); and the TLR adaptor MyD88 and multiple key proteins involved in TLR signalling, regulation of NF-κB, and lymphocyte development (SLE3). ROC curve analysis of antibody biomarker panels showed improved diagnostic sensitivity/specificity compared to conventional anti-nuclear (ANA) and anti-dsDNA antibody tests. Of particular clinical importance, autoantibodies from cluster SLE1b and SLE3 enabled detection of SLE individuals who were negative for ANA or anti-dsDNA antibody tests. Conclusions: Novel autoantibodies, especially from cluster SLE1b and SLE3, identified in this study may improve diagnosis of SLE, through identification of SLE patients who are negative for ANA or dsDNA antibodies. SLE patients clustered into four groups with autoantibody responses against networks of proteins with related cellular or immunological functions, suggesting that different pathogenic mechanisms underlie the four SLE subgroups. Thus the autoantigen clusters may be clinically useful for stratifying SLE patients for specific therapies: SLE patients with TLR signaling autoantigens might preferentially benefit from B cell targeting therapies; SLE patients with autoantibodies against SMAD2/5 and related pathway autoantigens might benefit from TGF-β targeting therapies. Disclosure of Interest: M. Lewis: None declared, M. McAndrew Employee of: Oxford Gene Technology, C. Wheeler Employee of: Oxford Gene Technology, N. Workman Employee of: Oxford Gene Technology, P. Agashe: None declared, J. Koopmann: None declared, E. Uddin Employee of: Oxford Gene Technology, L. Zou: None declared, R. Stark: None declared, J. Anson Employee of: Oxford Gene Technology, A. Cope: None declared, T. Vyse: None declared