Synovial and serum B cell signature of autoantibody-negative rheumatoid arthritis vs autoantibody-positive rheumatoid arthritis and psoriatic arthritis.

Abstract

Autoantibody-negative RA differs from autoantibody-positive RA in several clinical aspects, possibly underpinned by pathogenetic differences. At present, the role of adaptive immune responses in autoantibody-negative RA remains unclear. Here, we investigated the synovial and serum immunophenotype indicative of B lymphocyte involvement across the spectrum of autoantibody-positive and -negative chronic arthritides. Ultrasound-guided synovial biopsies were retrieved from 131 patients: 43 autoantibody-positive RA, 35 autoantibody-negative RA, 25 polyarticular PsA and 28 oligoarticular PsA. Samples were analysed for the degree of histological inflammation, B lymphocyte infiltration and the distribution of different pathotypes (lympho-myeloid, myeloid, pauci-immune). Serum levels of the B cell chemoattractant CXCL13 were compared among groups. Synovitis scores and CD68+ sublining macrophage infiltration were comparable irrespective of clinical diagnosis and disease subtype. In contrast, the degree of B lymphocyte infiltration and the frequency of lympho-myeloid synovitis in autoantibody-negative RA were lower than those of autoantibody-positive RA (mean [s.d.] 1.8 [1] vs 2.4 [0.6], P = 0.03, and 38.2% vs 62.9%, P = 0.07, respectively), and similar to polyarticular PsA. Oligoarticular PsA had the lowest B cell scores. Serum CXCL13 was associated with lympho-myeloid synovitis and followed a similar gradient, with the highest levels in autoantibody-positive RA, intermediate and comparable levels in autoantibody-negative RA and polyarticular PsA, and low levels in oligoarticular PsA. The synovial and serum immunophenotype indicative of B lymphocyte involvement in autoantibody-negative RA differs from that of autoantibody-positive RA and more closely resembles that observed in polyarticular PsA. The pathobiological stratification of chronic inflammatory arthritides beyond clinical diagnosis may fuel personalized treatment strategies.