ACTIVITY OF ABATACEPT IN SLE: RESULTS OF A 12-MONTH PHASE II EXPLORATORY STUDY

Background: The clinical heterogeneity of SLE makes assessment of disease activity and treatment strategies a complex and challenging process. Here we assess the clinical activity and safety of abatacept (ABA) in patients (pts) with SLE and flare of polyarthritis/serositis/discoid lesions. Methods: In a 1-yr, exploratory, Phase II trial, pts with SLE and active polyarthritis/serositis/discoid lesions were randomized 2:1 to ABA ($∼ $10mg/kg) or placebo (PBO). Prednisone (30mg/day or equivalent) was given for 1mth then tapered per protocol. The primary endpoint was the proportion of pts with new SLE flare (adjudicated BILAG A or B) after the start of steroid taper over 1yr. Secondary BILAG endpoints, physician-assessed flare, pt-reported outcomes (PRO) and safety were assessed over 1yr. Results: 118 ABA and 57 PBO pts were evaluated: baseline characteristics were similar. Primary manifestations at entry were discoid rash (34.3%), polyarthritis (54.3%) and serositis (11.4%). The proportion of pts with new BILAG A or B flare over 1yr (95%CI) was 79.7% (72.4, 86.9) for ABA vs 82.5% (72.6, 92.3) for PBO. Results for secondary BILAG endpoints were similar. In pre-specified exploratory analyses, PROs showed statistically persuasive improvements for ABA vs PBO (treatment difference [95% CI]) in the SF-36 physical component summary (3.9 [1.2, 6.6]), fatigue VAS 100mm (-9.5 [-17.7, -1.3]) and MOS-sleep (-7.6 [-12.3, -2.9]). In post-hoc analyses, flare was assessed using different measures. A lack of concordance was observed when new SLE flare was based on 1) BILAG A or B vs BILAG A alone, and 2) BILAG A or B vs physician-assessed flare. When the primary endpoint was re-assessed using BILAG A flare only, 40.7% (95%CI: 31.8, 49.5) in the ABA group vs 54.4% (41.5, 67.3) in the PBO group experienced flare over 1yr. Further evidence of a difference in flares when rated by treating physicians supported this. The proportion of pts rated as flaring by the physicians over 1yr (95%CI) was 63.6% (54.9, 72.2) for ABA vs 82.5% (72.6, 92.3) for PBO; the difference was greatest in the polyarthritis subgroup (57.1% [44.9, 69.4] vs 84.4% [71.8, 97.0]). AEs were comparable (90.9% for ABA vs 91.5% for PBO). SAEs were higher for ABA (19.8%) vs PBO (6.8%); most were single events, appeared related to underlying disease, and occurred during or shortly after steroid taper with no discernible pattern across organ class. Conclusion: In this Phase II study in pts with moderate to severe SLE, primary and secondary endpoints were not met. Other clinical measures and PROs (including the SF-36) were suggestive of ABA activity, supporting further assessment in lupus. The discordance between flare assessed by the BILAG criteria (primary endpoint) vs the treating physician's opinion suggest that greater understanding is required to determine the most appropriate definition of flare for use in SLE clinical trials. Disclosure of Interest: JTM: BMS, Consultant (C)/Ad Board/Investigator RBV: Abbott, BMS, Pfizer, Roche, Schering-Plough (SP), C/Speakers Bureau (SB) RW: BMS, SP, C/SB; UCB, Grant Support(GS) AC: Roche, BMS, GS DDC: Aspreva, GS/C/SB; BMS, Novartis, C DW, SCB, YP: No disclosure LHS: BMS, Employee (E) JCB: BMS, E/Shareholder (S)/Pension (P) SK: BMS, S/E/P KR: BMS, E/S/P; PPD, S TL: BMS, S/E MK: BMS, S/E/P; J&J, S PN: Centocor, GS; SP, SBCitation: Annals of the Rheumatic Diseases, volume 68, supplement 3, year 2009, page 70Session: Abstract Session: SLE treatment (Oral Presentations )