Comparative efficacy of upadacitinib and nemolizumab in combination with background topical therapy in adults and adolescents with moderate-to-severe atopic dermatitis: a placebo-anchored matching-adjusted indirect comparison

Background: Upadacitinib (UPA), a Janus kinase 1 inhibitor, and nemolizumab (NEMO), an interleukin-31 receptor alpha antagonist, are both approved for the treatment of moderate-to-severe atopic dermatitis (AD). A placebo-anchored matching-adjusted indirect comparison (MAIC) comparde efficacy of UPA versus NEMO in adults and adolescents with moderate-to-severe AD. Methods: In the AD Up Ph3 trial , adults and adolescents with moderate-to-severe AD received UPA 15mg (UPA15), UPA 30mg (UPA30), or placebo in combination with topical corticosteroids (TCS) for 16-weeks (W). Aggregate-level patient data for NEMO were collected from ARCADIA 1& 2 Ph3 trials, in which AD patients received NEMO 30mg or placebo in combination with TCS for 16Ws. Selection criteria included Eczema Area and Severity Index (EASI) ≥ 16, Investigator Global Assessment (IGA) of moderate/severe, Worst Pruritus Numerical Rating Scale (WP-NRS) ≥ 4, and body surface area (BSA) ≥ 10%. Upadacitinib patients were weighted to match NEMO baseline characteristics. Outcome measures included the Aiming High in Eczema/Atopic Dermatitis (AHEAD) recommendations1 for moderate treatment targets (≥ 75% improvement in EASI [EASI 75], WP-NRS improvement ≥ 4 [∆WP-NRS ≥ 4], EASI 75 + ∆WP-NRS ≥ 4) and optimal treatment targets (≥ 90% improvement in EASI [EASI 90], WP-NRS < 2). Response rates for UPA vs NEMO used placebo-adjusted rates, statistical comparisons based on the risk difference (RD). Results: UPA demonstrated statistically significant superiority over NEMO across all outcomes at Week (W) 16, with benefits emerging by W2. At W16, matching-adjusted response rates for UPA were numerically higher than those reported for NEMO for EASI 75 (NEMO: 42.9%; UPA15: 68.3% | UPA30: 79.2%), EASI 90 (25.6%; 48.4% | 67.6%), ∆WP-NRS ≥ 4 (41.9%; 54.2% | 62.4%), WP-NRS < 2 (29.6%; 40.1% | 50.8%), and EASI 75 + ∆WP-NRS ≥ 4 (23.4%; 45.5% | 57.8%). Placebo-adjusted response rates were higher for UPA15 and UPA30 versus NEMO for EASI 75 (UPA15: +40.8% | UPA30: +51.8% vs NEMO: +13.4%), EASI 90 (+35.7% | +54.9% vs +8.0%), ∆WP-NRS ≥ 4 (+40.4% | +48.6% vs +24.0%), WP-NRS < 2 (+31.4% | +42.2% vs +18.3%), and EASI 75 + ∆WP-NRS ≥ 4 (+36.2% | +48.5% vs +12.8%) (Figure); RDs were statistically significant (p < 0.001) for these comparisons. Statistically significant RDs favoring UPA were observed at W2 for ∆WP-NRS ≥ 4 (UPA15 vs NEMO: RD = 10.7% [95% CI: 2.9, 18.4]; UPA30 vs NEMO: RD = 20.1% [95% CI: 12.0, 28.1]) and WP-NRS < 2 (9.9% [3.7, 16.2]; 15.4% [8.9, 22.0]), and at W4 for EASI 75 (39.2% [30.4, 48.0]; 51.3% [42.9, 59.6]), EASI 90 (23.4% [16.3, 30.4]; 42.2% [34.6, 49.8]), and EASI 75 + ∆WP-NRS ≥ 4 (27.1% [19.3, 34.9]; 38.8% [30.8, 46.8]). Statistically significant differences for UPA15 and UPA30 versus NEMO were sustained through W16 for all outcomes assessed. Conclusions: Based on trials that included background topical therapy, adults and adolescents with moderate-to-severe AD achieved higher rates of moderate and optimal treatment targets with UPA15 and UPA30 compared with NEMO as early as W2 and through W16 in this placebo-anchored MAIC.