Factors Associated with Persistent Efficacy of Abrocitinib without Flare: A Multivariable Analysis of the JADE-REGIMEN Study

Introduction: JADE REGIMEN (NCT03627767) was conducted to evaluate the feasibility of continual, reduced dose or withdrawal of abrocitinib after induction of response in patients with moderate-to-severe atopic dermatitis (AD). This post hoc analysis evaluated patient factors associated with a higher probability of persistent clinical response with abrocitinib, without flare, during a 40-week maintenance period.Methods: Data were analyzed from patients who responded to abrocitinib 200 mg induction therapy (12 weeks) and were randomly assigned to receive abrocitinib (200 or 100 mg) or placebo in the maintenance period (40 weeks). A multivariable logistic regression model with fixed and random effects was fit to determine factors associated with not experiencing flare by week 52. Fixed effects (factors) considered were randomly allocated treatment, age (<18 vs ≥18 years), race, weight, prior use of systemic agents, AD duration, onset of response in induction (early vs late), Investigator’s Global Assessment score at randomization, body surface area (BSA, ≤50% vs >50%) at baseline, and percentage change in Eczema Area and Severity Index (EASI) at randomization. Backward elimination and stepwise model selection procedures were applied with entry and exit criteria based on a P value threshold of 5%.Results:1233 patients received abrocitinib 200 mg in the induction period. By week 12, 798 (64.7%) achieved a response and were randomly assigned to receive maintenance treatment for 40 weeks. In total, 356 patients (44.6%) experienced a protocol-defined flare: 16.5% (200 mg), 39.6% (100 mg), and 77.5% (placebo). After model selection, the analysis identified continuation of treatment with abrocitinib 200 mg (odds ratio [OR], 19.51; 95% CI, 11.28-33.75) or reduced-dose abrocitinib 100 mg (5.27; 3.29-8.46) as the greatest predictors of not experiencing flare during maintenance. Not having previously used systemic agents (OR, 1.53; 95% CI, 1.09-2.14), lower baseline BSA (1.55; 1.10-2.17), and greater percentage reduction of EASI during induction (1.35; 1.15-1.59) were also associated with not experiencing flare.Conclusions: Multivariable analysis of JADE REGIMEN indicated that maintenance treatment with abrocitinib in patients with AD reduced the risk for flare in a dose-dependent manner. Other factors associated with maintaining response to treatment without flare included no previous exposure to systemic agents, lower extent of BSA involvement at baseline, and greater percent change in EASI during induction. These findings may assist clinicians with abrocitinib dosing decisions in the future.