Résumé
Primary Sjögren's syndrome (pSS) is a heterogeneous autoimmune disease (AID) that can be difficult to differentiate from rheumatoid arthritis (RA) or systemic lupus erythematosus (SLE). New diagnostic and prognostic biomarkers are therefore needed. The objective of this work was to identify biomarkers that discriminate pSS from RA and SLE, or correlate with the pSS disease activity score (ESSDAI). Serum concentrations of 63 biomarkers were measured in 42 patients followed for pSS, 28 for RA and 25 for SLE. The probability of pSS compared to RA was associated with increased concentrations of BDNF (OR=0.493 [0.273-0.891]) and I-TAC/CXCL11 (OR=1.344 [1.027-1.76]). The probability of pSS compared to SSc was associated with increased concentrations of sCD163 (OR=0.803 [0.649-0.994]), Fractalkine/CX3CL1 (OR=0.534 [0.287-0.991]), MCP-1/CCL2 (OR=0.839 [0.732-0.962]) and TNF-a (OR=0.479 [0.247-0.928]). The association of low concentrations of BDNF and Fractalkine/CX3CL1 was specific to pSS (96.2%; PPV 80%) but not very sensitive (19%). A significant correlation was observed between pSS activity and the concentrations of sCD163 (r=-0.3386), sgp130 (r=-0.3792), MPIF-1/CCL23 (r=-0.3335) and SCYB16/CXCL16 (r=-0.3072). PSS therefore has a different serum protein signature to RA and SLE, and some of the biomarkers studied were associated with its activity according to the ESSDAI score. However, further studies are needed to validate the use of these biomarkers in clinical practice.