Background Many autoantibodies are known to be associated with SLE, although their role in clinical practice is limited because of low sensitivity and weak associations with clinical manifestations. Therefore, there has been great interest in the discovery of new autoantibodies or autoantibody patterns for clinical practice. In this study, we investigated patterns of new and known antibodies and their possible role in diagnostics or risk stratification.

Methods Between 2014 and 2017, residual sera of all anti-dsDNA tests in the UMC Utrecht were stored in a biobank. Diagnosis and presence of symptoms at each blood draw were retrospectively assessed in the patient records with the Utrecht Patient-Oriented Database (UPOD) using a newly developed text mining algorithm. Sera from a balanced cohort of patients with different diagnoses and patients without an assigned diagnosis were analyzed for the presence of 74 autoantibodies by a custom-made immunofluorescent microarray. Whenever possible, results were compared to corresponding historic in-house tests to assess quality. Differences in autoantibodies between patients with SLE and patients with a low suspicion of SLE were investigated by univariate and machine learning (XGBoost) analyses.

Results Autoantibody profiles of 484 patients with SLE were compared to 218 controls. Results from the microarray corresponded well with those from corresponding validated assays (AUC 0.726–0.902). Both univariate and machine learning analysis showed anti-dsDNA as most distinctive feature between both groups. Moreover, antibodies against Cytosine-phosphate-Guanine (anti-CpG) DNA motifs were found to be strongly associated with SLE (p<0.0001) and lupus nephritis (N=161, p=0.0015). Anti-dsDNA and anti-CpG antibodies correlated moderately with each other.

Conclusions Anti-CpG antibodies are prevalent in patients with SLE and are associated with Lupus Nephritis independent of anti-dsDNA, suggesting an additive diagnostic value of anti-CpG antibodies.

Acknowledgements This project was supported by Thermo Fisher Scientific.