AB0180 VALIDATION OF COMPLEX IMMUNOPHENOTYPING STRATIFICATION OF PATIENTSWITH LUPUS AND SJÖGREN'S SYNDROME WITH THERAPEUTIC POTENTIAL. (June 2019)

• Record Type: Journal Article
• Authors: Martin-Gutierrez, Lucia; Peckham, Hannah; Robinson, George; Varnier, Giulia; Jury, Elizabeth; Ciurtin, Coziana
• Is Part Of: Annals of the rheumatic diseases. Volume 78(2019)Supplement 2
• Page Start: 1548
• Page End: 1548
• Publication Date: 2019-06
• Subjects: Rheumatism -- Periodicals; 616.723005
• DOI: 10.1136/annrheumdis-2019-eular.5696
• Languages: English
• ISSNs: 0003-4967

Abstract

Abstract : Background: There is a current paucity in effective treatments for patients with primary Sjögren's syndrome (pSS) and systemic lupus erythematosus (SLE). Their overlapping symptoms mean that they are often treated with similar biologic therapeutics, despite being distinct diagnoses. Our pilot research sought to stratify patients by novel, complex immune signatures, to potentially facilitate a more personalised approach to prescribing medication, irrespective of clinical diagnosis. From a cohort of 40 patient patients with SLE, pSS or combined SS/SLE, and 33 healthy controls; five distinct clusters were originally identified. Objectives: Here we aimed to validate the original patient endotypes from our pilot data with a new patient cohort, and refine the identified clusters to be of more realistic clinical benefit. Methods: A new cohort of patients with SLE (n=32), pSS (n=31) and SS/SLE (n=15) donated peripheral blood samples while attending appointments at UCLH clinics. None had received B-cell depletion therapy in the preceding 48 months. Complex phenotyping of peripheral T and B-cell subsets was performed by flow cytometry. Unsupervised clustering, statistical comparison to healthies and Receiver Operator Characteristic (ROC) analysis, as per our previous work, were used to stratify patients using immune signatures into distinct endotypes considering only those markers where sensitivity and specificity were above 80% (AUC> 0.80). Characteristics of new clusters were compared with previous findings to refine the number of markers and unique groups. Results: Five distinct endotypes across the three disease groups were identified after unsupervised clustering of CD19+ B-cell, CD4+ and CD8+ T-cell subsets. Two of these clusters serve to validate groupings from our original data- one with a distinct B cell fingerprint (Bm1, lBm5 and Total CD19+) with AUC=0.87, AUC=0.81 and AUC=0.83 respectively, and another with a T cell fingerprint (Total CD4+, Total CD8+, CD8+ naïve, CD8+ effector memory T cell subsets) with AUC>0.80 for all of them. Clinical diagnoses and symptoms treatment did not differ significantly between clusters. Conclusion: Our results demonstrate a novel method to re-classify patients using clustered immune phenotype abnormalities. Two distinct groups demonstrated either B or T-cell immunophenotypes, likely reflective of their underlying immunopathogenesis. Stratification by patient endotype has the potential to facilitate a better-informed selection of biologic therapeutics for individual patients; and highlights pathways for further mechanistic research. Acknowledgement: British Sjögren's syndrome Association and Versus Arthritis Disclosure of Interests: None declared