Richard Furie1, Eric Morand2, Kenneth Kalunian3, Konstantina Psachoulia4, Emmanuelle Maho5, Catharina Lindholm6 and Raj Tummala4, 1Zucker School of Medicine at Hofstra/Northwell Health, Great Neck, NY, 2Monash University, Melbourne, Australia, 3University of California, La Jolla, CA, 4BioPharmaceuticals R&D, AstraZeneca, Gaithersburg, MD, 5BioPharmaceuticals R&D, AstraZeneca, Cambridge, United Kingdom, 6BioPharmaceuticals R&D, AstraZeneca, Gothenburg, Sweden
Background/Purpose: In the phase 3 TULIP-1 and TULIP-2 trials, anifrolumab, a type I IFN receptor mAb, improved disease activity in patients with SLE.1,2 We investigated whether prior biologic exposure impacted anifrolumab efficacy and safety in TULIP-1 and TULIP-2 pooled data.
Methods: This analysis included patients who received intravenous anifrolumab 300 mg or placebo every 4 weeks for 48 weeks in the 52-week TULIP-1 (NCT02446912) and TULIP-2 (NCT02446899) trials, for which eligible patients met the ACR 1997 SLE criteria, had moderate to severe SLE, and were permitted prior biologic use with a 3−6 month washout period, regardless of the reason for cessation. Patients were split into biologic-experienced or biologic-naïve subgroups (≥1 or 0 previous biologic immunomodulators, respectively). Baseline SLE disease characteristics, efficacy, and safety were compared across subgroups. Efficacy measures included BILAG–based Combined Lupus Assessment (BICLA) response at Week (W) 52; SLE Responder Index of ≥4 (SRI) response at W52; sustained oral glucocorticoid (GC) taper (≤7.5 mg/day prednisone equivalent, W40−52, if ≥10 mg/day at baseline); and annualized flare rate through W52. Binary endpoints and safety were analyzed with a Cochran–Mantel–Haenszel approach controlling for randomization stratification factors and study. Annualized flare rate was analyzed with a negative binomial regression model with treatment, randomization stratification factors, and study as covariates.
Results: There were 145 biologic-experienced patients (anifrolumab [n=75]; placebo [n=70]), and 581 biologic-naïve patients (anifrolumab [n=285]; placebo [n=296]). Most previous biologic use was with belimumab (n=70), epratuzumab (n=49), tabalumab (n=18), or rituximab (n=14). Baseline demographics, disease characteristics, and non-biologic SLE treatments were generally similar between groups. However, compared with biologic-naïve patients, biologic‑experienced patients had longer times from SLE diagnosis, were more likely to be from USA/Canada, have SLICC/ACR Damage Index score ≥1, anti‑dsDNA antibodies, and IFN gene signatures, and less likely to have swollen/tender joints (Table 1). There were lower placebo responses (potentially more refractory disease) in biologic‑experienced vs biologic-naïve patients (Figure). Anifrolumab was associated with comparable/greater treatment differences over placebo (∆) in biologic‑experienced vs biologic-naïve patients across endpoints, including BICLA (∆=19.4 vs ∆=16.6), SRI(4) (∆=25.3 vs ∆=9.1), and GC tapers (∆=24.7 vs ∆=17.5). Incidence of serious adverse events was higher in biologic‑experienced vs biologic-naïve patients with anifrolumab and placebo (Table 2). Herpes zoster incidence was higher with anifrolumab vs placebo in both biologic‑experienced and biologic-naïve patients.
Conclusion: Regardless of whether patients with SLE had previously received biologics, anifrolumab 300 mg provided clinically meaningful benefit over placebo across efficacy endpoints and was generally well tolerated.