Translating the Investigator’s Static Global Assessment to the Eczema Area and Severity Index in Studies of Crisaborole for Atopic Dermatitis

Abstract

Atopic dermatitis (AD) severity was measured in two phase 3 US studies of crisaborole ointment, 2%, in patients aged\u2009≥\u20092 years using the Investigator’s Static Global Assessment (ISGA), an FDA-recommended scale. Eczema Area and Severity Index (EASI) is a validated scale used globally to assess AD severity in clinical trials. The objective of this study is to aid interpretability of ISGA by translating ISGA scores to EASI scores. ISGA was mapped to EASI using published EASI severity strata by Chopra et al. and Leshem et al. and pooled data from phase 3 trials CrisADe CORE 1 and CORE 2, which evaluated crisaborole in patients aged\u2009≥\u20092 years with mild-to-moderate AD (crisaborole, n\u2009=\u20091016; vehicle, n\u2009=\u2009506). Least squares mean (LSM) percentage change from baseline (%CFB) in EASI and proportion of patients with 50%, 75%, and 90% improvement (EASI-50, EASI-75, and EASI-90, respectively) on day 29 were computed for mapped EASI. The relationship between changes in ISGA and changes in EASI was assessed using data from three abrocitinib trials. ISGA was mapped to EASI using 70,000 random simulations. LSM (standard error) for %CFB in mapped EASI at day 29 (crisaborole versus vehicle) was\u2009−26.3% (17) versus 45.2% (35) (P\u2009=\u20090.0671) using Chopra strata and\u2009−43.1% (4.6) versus\u2009−5.2% (8.4) (P\u2009<\u20090.0001) using Leshem strata. EASI-50, EASI-75, and EASI-90 rates were 72.1% versus 57.6%, 63.0% versus 47.8%, and 55.0% versus 40.1%, respectively, using Chopra strata (P\u2009<\u20090.0001 for each difference). These rates were 68.8% versus 54.0%, 54.8% versus 40.5%, and 38.9% versus 27.2%, respectively (P\u2009<\u20090.0001 for each difference) using Leshem strata. Mean two-point improvement in ISGA was comparable to EASI-90. Mapped EASI results were consistent with ISGA results in crisaborole phase 3 trials. Simulation methodologies yielded consistent results and may aid in interpretability of ISGA across clinical studies. ClinicalTrials.gov identifier: NCT02118766, NCT02118792.