OP0291 ASSOCIATION OF BELIMUMAB USE WITH ORAL CORTICOSTEROID DOSE AND DISEASE ACTIVITY IN CHILDHOOD-ONSET LUPUS: OUTCOMES FROM A NESTED-CASE CONTROL ANALYSIS IN THE CHILDHOOD ARTHRITIS AND RHEUMATOLOGY RESEARCH ALLIANCE REGISTRY

Background: Belimumab, a biologic medication targeting B-cell activating factor, was recently approved for childhood-onset systemic lupus erythematosus (cSLE). Reduction in oral glucocorticoid doses and improvement in SLEDAI-2K scores have been demonstrated in patients who received belimumab, but the extent to which this is due to belimumab versus other factors including background immunosuppressant medications for cSLE is unknown. Objectives: We aimed to 1) describe characteristics of patients in the Childhood Arthritis and Rheumatology Research Alliance (CARRA) Registry who received belimumab for cSLE compared to all Registry participants with cSLE, and 2) compare changes in glucocorticoid doses and disease activity among youth who did and did not receive belimumab using a nested case-control design. Methods: We included youth with cSLE enrolled in the CARRA Registry. We compared demographic and disease characteristics of those exposed to ≥ 1 dose of belimumab with all cSLE patients enrolled in the Registry. Among all belimumab users, linear mixed effects models were used to assess trajectories of SLEDAI-2K scores and prednisone-equivalent daily oral glucocorticoid doses over the 12 months following belimumab initiation. Youth who received ≥ 1 dose of belimumab were then matched 1:2 with controls who had not received belimumab based on: age at cSLE diagnosis, sex, lupus nephritis, SLEDAI-2k score, and disease duration at time of belimumab initiation (cases) or equivalent disease duration (controls). SLEDAI-2K scores and oral glucocorticoid dosing in mg of prednisone equivalent were compared at index visit, defined as start of belimumab for cases, and first matched visit for controls. Trajectories of oral glucocorticoid dosing and SLEDAI-2K scores over 12 months were compared between belimumab users and non-users using conditional logistic regression models. Results: Of 994 patients with a baseline visit for cSLE, we identified 63 patients who received ≥ 1 dose of belimumab. Patient and disease characteristics of belimumab users and the entire CARRA Registry cSLE cohort were similar, although most belimumab users had longstanding cSLE at time of initiation (Table 1). Forty-four percent of belimumab users had a history of lupus nephritis, equivalent to the entire Registry. At time of belimumab initiation, 83% of patients were prescribed an antimalarial, 70% were on oral glucocorticoids, and 59% on mycophenolate mofetil; 24% had received rituximab, and 16% had received cyclophosphamide. The median duration of belimumab treatment was 15.2 months (IQR 1 month). SLEDAI-2K scores and oral prednisone doses decreased over the 12 months following belimumab initiation, with a mean (95% CI) decrease of 0.29 (0.46, 0.13) SLEDAI-2K score points per month, and decrease of 0.49 (0.87, 0.10) mg prednisone equivalent per month. Fifty-one belimumab users were matched to 102 controls with no belimumab exposure. The mean (SD) SLEDAI-2K score at time of belimumab initiation was 9.6 (6.3) for belimumab users and 9.1 (6.5) for matched controls. Belimumab use was associated with a smaller decrease in SLEDAI-2K score from index visit to 6 months compared to the control group, but there was no difference between groups at 12 months (Table 2). The mean (SD) prednisone equivalent daily dose at time of belimumab initiation was 13.5 mg/ day for belimumab users and 8.6 mg/day for matched controls. Glucocorticoid doses remained higher in the belimumab group at 6 and 12 months after index visit, and there was no association between belimumab use and magnitude of the change in oral glucocorticoid use over time. Conclusion: Youth with cSLE who initiated belimumab had decreased disease activity and prednisone doses in the first year after starting belimumab. However, when compared to matched controls without belimumab exposure, there were no differences in glucocorticoid dose trajectories between belimumab users and non-users. Belimumab users had less improvement in SLEDAI-2K scores over the first 6 months, possibly reflecting slower onset than other treatment strategies, but had equivalent reduction in SLEDAI-2K scores at 12 months compared to those on other medication regimens. REFERENCES: NIL. Acknowledgements: This abstract is submitted on behalf of the CARRA Lupus Nephritis Workgroup, for the CARRA Registry Investigators. Disclosure of Interests: None declared.