THU0352 Association between memory b-cells and phenotypic features of sjÖgren’s syndrome

Background B-cell disturbances are a hallmark of pSS and play a pivotal role in the disease pathogenesis and clinical evolution, and may as well have a potential role in diagnosis. In pSS, an increase of the naive subset and a decrease of memory B-cells have been reported. A decreased frequency of memory cells has also been identified in patients with Sicca syndrome without criteria for pSS. Objectives Our study aims to evaluate the distribution of B-lymphocyte subpopulations in pSS and Sicca patients and to establish cut-off points for pSS classification in relation to healthy controls. Moreover, we aim to evaluate the relation between lymphocyte subpopulations and phenotypic features in pSS. Methods Fifty-seven pSS patients, 68 non-Sjogren Sicca patients and 24 healthy controls were included. Circulating B-cell frequencies were determined by flow cytometry, and the naive and memory (switched and unswitched) subsets were characterised based on surface marker expression of the following monoclonal antibodies: CD19, CD24, CD27, Anti-IgD and Anti-IgM. Kruskal-Wallis test was applied for groups’ comparison. ROC curves were used to establish cut-off points in the B-cells subset levels and to estimate corresponding sensitivity and specificity. Data analysis was performed with R software. Results Absolute numbers of lymphocytes in pSS were lower compared to controls, with Sicca presenting intermediate levels. Significant differences were found between pSS and controls in absolute counts of all memory populations: total memory (TMem) (CD19+CD27+), switched (SwM) (CD19+IgD-CD27+) and unswitched memory (UnSwM) (p Through ROC curves, the B-cell subsets that better discriminate between pSS and controls were TMem and SwM. A cut-off of equal to 58 TMem cells/µl yelded a specificity of 0.88 and a sensitivity of 0.60 for pSS, and was met by 59.6% of pSS, 12.5% of controls and 38.8% of Sicca, and a cut-off of equal to 23.5 SwM cells/µl yelded a specificity of 0.88 and a sensitivity of 0.54 and was met by 54.4% of pSS, 12.5% of controls and 37.3% of Sicca. pSS patients with lower values than the established cut-off points had longer disease duration, higher disease activity (ESSDAI), and were more likely to present auto-antibodies and positive biopsy. Several Sicca patients also presented memory B-cell subsets counts lower than the pSS cut-off, but no consistent differences in clinical profile were identified. Conclusions Decreased numbers of memory B-cell subsets clearly discriminate pSS from healthy controls. Lower memory B-cells counts are associated with more active pSS disease profile. It remains to be clarified whether Sicca patients with decreased memory B-cells represent pSS and if B-cell profiling could help in the diagnosis of pSS. Disclosure of Interest None declared