AB0617 Characterization of SLE patients experiencing flares in epratuzumab clinical trials

Background Epratuzumab is a humanized anti-CD22 monoclonal antibody in development for the treatment of moderate-to-severe systemic lupus erythematosus (SLE). Its efficacy in SLE has been demonstrated in three double-blind, placebo-controlled studies: SL0003/SL0004 (ALLEVIATE 1/2) and SL0007 (EMBLEM). Recurrent flare is a major pattern of SLE disease activity and animportant outcome measure in clinical studies, but its definition varies. The BILAG disease activity index can be adapted to measure the occurrence of flare in SLE. Objectives 1) To assess BILAG incidence of SLE flare in the ALLEVIATE and EMBLEM studies, in body systems that were not flaring at baseline. 2) To identify possible predictors of flare occurrence in patients randomized to placebo in ALLEVIATE. Methods Flare was defined as the development of a BILAG A score from a B, C, D or E score at baseline, or the development of two BILAG B scores from C, D or E scores at baseline in any organ system, at any visit during the study (planned 48 weeks for ALLEVIATE before drug supply interruption; 12 weeks for EMBLEM). ALLEVIATE used the original BILAG index, while EMBLEM used the BILAG-2004 index. Biomarker candidates included in the analysis (CD19+ B cells, CD3+ T cells, Ig isotypes, serum antibodies and C3 complement) were assessed in placebo patients at baseline and either at the last value before flare or, if patients did not flare, at the last available value. Results In ALLEVIATE (n=88), the proportion of patients experiencing a new flare was highest in the placebo group (37.8% vs 30.0% in the epratuzumab 360 mg group and 9.1% in the epratuzumab 720 mg group). In EMBLEM (n=225), the proportion of patients experiencing a flare was low in all groups: 5.3% for placebo, 7.7% for 100 mg epratuzumabevery other week (EOW), 5.3% for 400 mg EOW, 5.7% for 600 mg weekly, 8.1% for 1200 mg EOW and 5.3% for 1800 mg EOW. In ALLEVIATE, baseline B cell numbers, C3, and rheumatoid factor were numerically lower among placebo patients who flared (n=14) vs who did not (n=23), whereas baseline C-reactive protein, anti-dsDNA antibodies, and IgM were numerically higher. The mean changes from baseline in patients who did not flare compared to those who did were: B cells, -101 cells/μL vs +9.4 cells/μL; T cells, unchanged vs +30.3 cells/μL; IgA, +7.4 mg/dL vs -34.5 mg/dL; IgM, +6.5 mg/dL vs -24.8 mg/dL. Of the patients who flared, 70.0% were positive for anti-dsDNA antibodies both at baseline and prior to flare, vs 43.5% of patients who did not flare. Conclusions In this post-hoc exploratory analysis of ALLEVIATE data, the proportion of patients who flared was highest in the placebo group, whereas in EMBLEM, the proportion was low and not different between groups. The lower incidence in EMBLEM may be attributable to the shorter duration of the trial and the fact that there was less time for steroid reduction than in ALLEVIATE. Further analysis is warranted to determine the significance of the observed differences in biomarker levels between flaring and non-flaring patients. Disclosure of Interest C. Gordon Consultant for: UCB Pharma, F. Houssiau Consultant for: UCB Pharma, K. Kalunian Consultant for: UCB Pharma, B. Kilgallen Employee of: UCB Pharma, H. Bartz Employee of: UCB Pharma, C. Galateanu Employee of: UCB Pharma, M. Petri Consultant for: UCB Pharma, W. Wegener Employee of: Immunomedics Inc, V. Strand Consultant for: UCB Pharma