Background Neuropsychiatric complaints in systemic lupus erythematosus (SLE) patients present a challenge to the clinician due to the lack of specific diagnostic tests. To date, reports on the associations between specific autoantibodies and distinct NPSLE syndromes are conflicting. New multiplex technologies for the detection of autoantibodies have emerged in the last years and it has been noted that they might be helpful diagnosing SLE. We hypothesized that a cluster of autoantibodies could be associated with a specific NPSLE syndrome or with focal or diffuse NPSLE manifestations.
Objectives Our aim was to analyse a microarray kit of autoantibodies and the antiphospholipd (aPL) autoantibodies status in a prospective well-defined cohort of NPSLE patients. Furthermore, we aimed to identify the relationship between clusters of these autoantibodies and NPSLE syndromes.
Methods We included 133 SLE patients with neuropsychiatric symptoms attending the Leiden NPSLE clinic. The serum samples of all patients were analysed using the FIDIS connective profile kit (Theradiag), a semi-quantitative homogeneous fluorescent based microparticles immunoassay for the simultaneous detection of anti-SSA, anti-SSB, anti-TRIM21, anti-Sm, anti-Sm/RNP, anti-Jo1, anticentromere, anti-Ribosomal-P, anti-dsDNA, anti-Histone, anti-PmScl and anti-PCNA. Furthermore, anticardiolipin (aCL) IgG and IgM antibodies and lupus anticoagulant (LAC) status were available from the clinical evaluation. We performed hierarchical cluster analyses using R software (version 3.0.2) on 1. all autoantibodies from the microarray kit and 2. all autoantibodies from the microarray kit plus the aPL antibodies, and we analysed their associations with NPSLE diagnosis, the American College of Rheumatology (ACR) NPSLE syndromes and groups of focal/diffuse NPSLE manifestations. Statistical significance was defined as p<0.05.
Results In 81 (61%) patients a diagnosis of NPSLE was established. In the first cluster analysis we identified three clusters of autoantibody profiles (No specific autoantibodies, DNA/Ro/La and Sm/RNP), however an association with NPSLE diagnosis or with NPSLE syndromes was not found. In the second cluster analysis, after inclusion of aPL antibodies, we identified four clusters of autoantibody profiles (1 - No specific autoantibodies, 2 - DNA/Ro/La, 3 - Sm/RNP, 4 - DNA/LAC/aCL). In this case the frequency of major focal syndromes in cluster 4 (DNA/LAC/aCL) was significantly greater than in other clusters (p=0,008). With respect to the individual NPSLE syndromes we found an association for cerebrovascular disease (p=0,002), seizure (p=0,016) and myelopathy (p=0,019) with cluster 4 (DNA/LAC/aCL).
Conclusions Our study failed to show an association between autoantibodies retrieved by multiplex testing or clusters of these autoantibodies and NPSLE. We have found an association between a cluster of autoantibodies (DNA/LAC/aCL) and major focal NPSLE syndromes. From all the autoantibodies used in daily practice, only LAC and aCL are indispensable in the NPSLE diagnostic work-up.
To CH, Petri M. Is antibody clustering predictive of clinical subsets and damage in systemic lupus erythematosus?. Arthritis Rheum. 2005;52:4003-10.
Disclosure of Interest None declared