1503: Disease Course and Complement as Predictors of Response to Standard of Care Plus Placebo in Patients with SLE: A Post Hoc Analysis of Dapirolizumab Pegol and Epratuzumab Clinical Trial Data
Oklahoma Medical Research Foundation Oklahoma City 73104, OK, United States
Disclosure information not submitted.
Christian Stach1, Caroline Gordon2, Vanessa Taieb3, George Stojan4 and Joan Merrill5, 1UCB Pharma, Monheim am Rheim, Germany, 2Institute of Inflammation and Ageing, University of Birmingham, Birmingham, United Kingdom, 3UCB Pharma, Colombes, France, 4UCB, Baltimore, MD, 5Oklahoma Medical Research Foundation, Oklahoma City, OK
Background/Purpose: High placebo responses pose challenges to systemic lupus erythematosus (SLE) clinical trials.1This analysis aims to identify predictors of standard of care plus placebo (SOC+PBO) response using data from epratuzumab (Emab; a humanized anti-CD22 monoclonal antibody) and dapirolizumab pegol (DZP; a polyethylene glycol-conjugated antigen-binding fragment lacking a functional Fc domain that inhibits CD40L) trials.2,3
Methods: Analyses were performed on the full analysis sets of the SOC+PBO arms of two phase 3 trials of Emab (EMBODY 1 [NCT01262365]; n=249, and EMBODY 2 [NCT01261793]; n=263) and the phase 2b trial of DZP (RISE [NCT02804763]; n=43) in SLE.2,3 Response was assessed by British Isles Lupus Assessment Group (BILAG)-based Composite Lupus Assessment (BICLA) response at Week (Wk) 24.4 Univariate/multivariate analyses were performed, with 18 and 22 potential predictors of SOC+PBO response evaluated in EMBODY 1/2 and RISE, respectively. Stepwise multivariate analysis included univariate predictors with p< 0.25, using p≤0.1 as entry/stay criteria. In subgroup analyses, disease course at screening was defined using BILAG 2004 item level scores as acute flare (worsening/new symptoms) or persistent disease (symptoms rated as the same) based on the past 4 wks compared with the 4 wks prior to those; low C3/C4 was defined as below the lower limit of normal at screening.
Results: Univariate and subsequent stepwise multivariate analysis found the following significant (p < 0.05) predictors of BICLA response at Wk 24 in EMBODY 1/2: acute flare (vs persistent disease activity; p< 0.001), Latin America/Eastern European (vs North America; p=0.004), and Hispanic or Latino (vs not; p=0.031; Figure 1). Significant (p < 0.05) predictors of BICLA response at Wk 24 in RISE following stepwise multivariate analysis were: acute flare (vs persistent disease activity; p=0.011) and normal C4 (vs low C4; p=0.036; Figure 2).
Focusing on disease characteristics, recent acute flares and normal complement were identified as predictors of SOC+PBO response. In all trials, a significantly higher proportion of patients (pts) with acute flare (vs persistent disease activity) achieved BICLA response (Figure 3a& 3d). In RISE, a significantly higher proportion of pts with normal C3/C4 (vs low C3/C4) achieved BICLA response (Figure 3e); this pattern was not observed in EMBODY 1/2 (Figure 3b). In all trials, when the subgroups were combined, a significantly higher proportion of pts with acute flare and normal C3/C4 (vs acute flare with low C3/C4 or persistent disease activity) achieved BICLA response (Figure 3c & 3f).
Conclusion: In both EMBODY 1/2 and RISE, acute flare predicted SOC+PBO response. In RISE, normal complement levels also predicted SOC+PBO response; low complement has been described as a predictor of increased treatment effect in previous studies.5 Recent medical history of pts may have to be considered when defining SLE study populations.
References:1. Durcan L. Lancet. 2023;S0140-6736(23)00342-2.2. Clowse ME. Arthritis Rheumatol. 2017;69(2):362–75. 3. Furie RA. Rheumatology (Oxford). 2021;60:5397–407. 4. Wallace D. Arthritis Rheum. 2011;63(Suppl 10):S885. 5. van Vollenhoven RF. Ann Rheum Dis. 2012;71(8):1343–9.
.jpg)
.jpg)
.jpg)
