Ultrasonographic non-radiographic erosions could predict the efficacy of belimumab in articular systemic lupus erythematosus

The aim of this study is to characterise lupus-related arthritis and assess if the presence of ultrasound-detected erosions could be associated with belimumab in the treatment of systemic lupus erythematosus (SLE) articular manifestations. We performed a spontaneous, monocentric, retrospective, and observational study. We enrolled patients affected by SLE with articular involvement treated with belimumab. We excluded patients with positive rheumatoid factor (RF) or anti-citrullinated peptide antibody (ACPA), Jaccoud's arthropathy, and radiographic erosions. Patients were assessed at baseline, 3, and 6 months. We collected laboratory and clinical data from electronic records. Joint disease activity was assessed using disease activity score on 28 joints based on C-reactive protein (DAS28-CRP), swollen and tender joints count. All patients underwent an ultrasound examination of the wrist, metacarpophalangeal, proximal interphalangeal, and metatarsal-phalangeal joints before the initiation of treatment with belimumab. We performed Student's T-test and Mann-Whitney's U-test to assess the difference between means and Fisher's exact test to assess difference in proportions, and linear univariate regression to investigate predictors of disease activity. We enrolled 23 patients (female 82.6%, mean age of 50.65 ± 14.1 years). Seven patients (30.4%) presented bone erosions at baseline. Patients with bone erosions were generally older (61 ± 16.1 vs 46.13 ± 10.7 years, p = 0.016), more frequently male (42.8 vs 6.2%, p = 0.03), with higher baseline CRP levels (10.29 ± 11.6 vs 2.25 ± 3.1 mg/L, p = 0.015) and C4 levels (0.19 ± 0.17 vs 0.1 ± 0.04 g/L, p = 0.05). After 6 months of treatment with belimumab, patients without erosions improved their DAS28-CRP significantly (2.95 ± 0.89 vs 2.26 ± 0.48, p = 0.01), while patients with erosions did not (3.6 ± 0.79 vs 3.2 ± 0.95, p = 0.413). DAS28-CRP did not differ between the two groups at baseline, while it was significantly lower at the other two time points in patients without erosions. The majority of patients achieved remission at 6 months follow-up based on DAS28-CRP criteria (73.9%), with a significant difference between patients with and without erosions (42.8 vs 87.5%, p = 0.045). The presence of articular ultrasound-detected erosions could be predictive of a decreased efficacy of belimumab in the articular manifestations of SLE. A possible explanation is a rheumatoid-like articular phenotype, despite the lack of ACPA-positivity and radiologic erosions. However, due to the small sample population, larger cohorts are needed to assess the possible predictive role of this finding.