Objectives We aimed at assessing serological predictors of flare in SLE patients who discontinued immunosuppressive therapies (IS) due to remission achievement.

Methods SLE patients (ACR criteria) diagnosed between 1990 and 2018, currently in follow-up and who discontinued IS after remission achievement were considered. Remission was defined as clinical SLE Disease Activity Index (SLEDAI)-2K=0 on a stable immunosuppressive and/or antimalarial therapy and/or prednisone ≤5 mg/day. Flares were defined according to SLEDAI-Flare-Index. Serum levels of C3c, C3a, BAFF/BLyS, APRIL, anti-C1q, and anti-Pentraxin 3 (PTX3) antibodies were measured within 6 months after IS discontinuation by commercially available ELISA kits. Mann-Whitney test was used, and ROC curves with AUC were generated.

Results Fifty-seven sera were evaluated.

Fifteen patients (26%) experienced a flare, after a median (range) follow-up of 53 (9–132) months. No difference in serum levels of C3c, C3a, APRIL, anti-PTX3, and anti-C1q at the time of IS discontinuation were found between patients who did or did not develop a flare. On the other hand, BAFF/BLyS serum levels at the time of IS discontinuation were higher in patients who then developed a flare (1.12, IQR 0.86–4.55) than in patients without a flare (0.75, IQR 0.64–1.03, p=0.009), although serum concentrations were generally low. AUC for BAFF/BLyS was 0.72. The cut-off with better accuracy was 0,90 ng/mL, showing a sensibility of 0.73 and a specificity of 0.64.

Conclusions In remitted patients, higher BAFF/BLyS serum levels at the time of IS discontinuation are associated with an increased risk of subsequent flares. Nevertheless, this biomarker showed low accuracy in predicting flares.