OP025 A randomized, double-blind placebo-controlled phase 2a induction study of MEDI2070 (anti-p19 antibody) in patients with active Crohn's disease who have failed anti-TNF antibody therapy
B.E. Sands*1, J. Chen2, M. Penney2, P. Newbold2, R. Faggioni2, R. van der Merwe2, K. Patra2, P. Klekotka3, E. Pulkstenis2, J. Drappa2, R.A. Gasser, Jr.2
1Icahn School of Medicine at Mount Sinai, Dr. Henry D. Janowitz Division of Gastroenterology, New York, NY, United States, 2MedImmune, Clinical Development, Gaithersburg, MD, United States, 3Amgen, Clinical Development, Thousand Oaks, CA, United States
IL-23 is implicated in the pathogenesis of Crohn's disease (CD). Previous studies of antibodies targeting the p40 subunit common to IL-12 and IL-23 suggested benefit in patients with CD who had failed treatment with anti-tumor necrosis factor (TNF) therapy. We studied the efficacy and safety of MEDI2070, a fully human IgG2 monoclonal antibody that selectively binds the p19 subunit of IL-23, in patients with active CD who had failed or were intolerant to anti-TNF therapy.
Subjects were 18-65 years old with active CD (Crohn's Disease Activity Index [CDAI] score ≥ 220 and ≤ 450) and active inflammation (C-reactive protein [CRP] ≥ 5 mg/L, or fecal calprotectin [FCP] ≥250 mcg/g, or endoscopic evidence of inflammation within 12 weeks of screening). All subjects had received ≥1 anti-TNF agent, with primary nonresponse, loss of response, or intolerance. Stable doses of 5-aminosalicylates, prednisone ≤20 mg/d, budesonide ≤ 6 mg/d, antibiotics for CD for at least 2 weeks, or immunomodulators (IMM) for at least 8 weeks, were permitted. Using concealed allocation, subjects were randomized blindly to MEDI2070 700 mg IV or placebo at Weeks 0 and 4, with stratification by number of prior anti-TNF agents (1 vs. >1), and followed to week 12. Primary analysis was comparison of the proportion of subjects receiving at least one dose of study medication who achieved clinical effect, defined as clinical response (≥100 point drop from baseline CDAI) OR clinical remission (CDAI <150) at week 8 (W8). Assuming clinical effect of 20% with placebo, 54 subjects per arm provided 87% power to detect a 25% treatment difference using a 2-sided significance level of 0.1.
121 subjects (61 placebo, 60 MEDI2070) were randomized, with 1 subject per arm not dosed. 57 placebo and 55 MEDI2070 subjects completed W8. Baseline characteristics were similar, with mean (SD) baseline CDAI 319 (58), 71.4% with CRP ≥5, and 75.9% with FCP ≥250. 31.1% had 1 prior anti-TNF agent; 68.9% had ≥2. 38.7% had primary anti-TNF failure. With MEDI2070, 49.2% had clinical effect at W8 vs. 26.7% with placebo (P=0.010). At W8, clinical remission was noted in 27.1% with MEDI2070 vs. 15.0% with placebo (p=0.102) and clinical response in 45.8% and 25.0%, respectively (p=0.017). A composite outcome of clinical effect AND ≥50% reduction from baseline FCP or CRP was achieved in 42.4% with MEDI2070 vs. 10.0% with placebo (p<0.001). No increased rate of adverse events with active treatment was observed over 12 weeks as compared to placebo.
MEDI20170, a specific anti-IL-23 antibody, demonstrates clinical effect in patients with active CD who have failed anti-TNF therapy, and has a favourable safety profile over 12 weeks.