Multiple clinical trials have evaluated the safety and efficacy of crisaborole in patients with mild-to-moderate AD.10-14 Crisaborole received its first approval in December 2016 in the United States for the treatment of mild-to-moderate AD in patients aged 2 years and older.
Although results of current head-to-head trials to compare crisaborole to other available topical therapies are not available (ClinicalTrials.gov, NCT03539601), a review of efficacy and safety data can contribute to a better understanding of the potential role of crisaborole in treating mild-to-moderate AD. Herein, the data published between 1997 and 2018 supporting the effectiveness and safety of available topical therapies (TCSs, TCIs, and crisaborole) will be reviewed.
Although results of current head-to-head trials to compare crisaborole to other available topical therapies are not available (ClinicalTrials.gov, NCT03539601), a review of efficacy and safety data can contribute to a better understanding of the potential role of crisaborole in treating mild-to-moderate AD. Herein, the data published between 1997 and 2018 supporting the effectiveness and safety of available topical therapies (TCSs, TCIs, and crisaborole) will be reviewed.
LITERATURE ANALYSIS
Our strategy consisted of searching Medline (via PubMed; Table
S1) for blinded, vehicle-controlled trials of topical agents published
from January 1, 1997 to April 30, 2018. Identified citations
were then selected as shown in Figure S1. In total, 134 publications
were identified through the search; publications were hand
screened for inclusion. 85 publications were excluded based on
the abstract; 21 were excluded based on full text. A total of 28
publications are summarized herein (Table 1).10,11,15-41 The P values
represent 2 comparator tests unless otherwise specified.
CONSIDERATIONS FOR COMPARISONS
Variety of Assessment Tools
There is no standardized single assessment tool for measuring the severity of AD and therapy response. A systematic review published in 2003 assessed the heterogeneity of AD severity measures, and it was found that only 27% (23 of 85) of randomized controlled trials (RCTs) used a previously published scoring system.42 The most commonly used disease severity scales include different aspects of AD.43 The SCORing of Atopic Dermatitis (SCORAD) assesses representative signs of AD and symptoms of pruritus and sleep loss. The Eczema Area and Severity Index (EASI) assesses lesional extent and intensity for each of 4 body regions, and the Investigator’s Static Global Assessment (ISGA) and the Investigator’s Global Assessment (IGA) assess overall severity by clinical signs.43 Heterogeneity of severity assessment instruments was also observed for the trials identified in the current search (Table 1): investigators variously used SCORAD, ISGA/IGA, total severity score (TSS), dermatologic sum score, and days on trial without flare (defined by use of TCS). The variety of instruments used to assess the severity of AD, different interpretation of scores, and lack of precise definitions makes comparison of efficacy between clinical trials challenging. The remitting-relapsing course of AD further complicates the interpretation of RCT results because waning disease severity can be misinterpreted as efficacy, especially in clinical trials with small sample sizes.
Vehicle Effects
“Vehicle effects†are an additional challenge, in that vehicle emollient effects may improve barrier function and reduce TEWL leading to improved outcomes in AD, which are observed frequently in vehicle-controlled RCTs.10,19,26,44 Vehicle composition, which commonly varies from product to product, can also contribute to local application site AEs such as stinging and erythema.45 For example, a number of trials included in this analysis report more AEs and treatment-related AEs (TRAEs) in the vehicle arm than in the active treatment arm (Table 1).
There is no standardized single assessment tool for measuring the severity of AD and therapy response. A systematic review published in 2003 assessed the heterogeneity of AD severity measures, and it was found that only 27% (23 of 85) of randomized controlled trials (RCTs) used a previously published scoring system.42 The most commonly used disease severity scales include different aspects of AD.43 The SCORing of Atopic Dermatitis (SCORAD) assesses representative signs of AD and symptoms of pruritus and sleep loss. The Eczema Area and Severity Index (EASI) assesses lesional extent and intensity for each of 4 body regions, and the Investigator’s Static Global Assessment (ISGA) and the Investigator’s Global Assessment (IGA) assess overall severity by clinical signs.43 Heterogeneity of severity assessment instruments was also observed for the trials identified in the current search (Table 1): investigators variously used SCORAD, ISGA/IGA, total severity score (TSS), dermatologic sum score, and days on trial without flare (defined by use of TCS). The variety of instruments used to assess the severity of AD, different interpretation of scores, and lack of precise definitions makes comparison of efficacy between clinical trials challenging. The remitting-relapsing course of AD further complicates the interpretation of RCT results because waning disease severity can be misinterpreted as efficacy, especially in clinical trials with small sample sizes.
Vehicle Effects
“Vehicle effects†are an additional challenge, in that vehicle emollient effects may improve barrier function and reduce TEWL leading to improved outcomes in AD, which are observed frequently in vehicle-controlled RCTs.10,19,26,44 Vehicle composition, which commonly varies from product to product, can also contribute to local application site AEs such as stinging and erythema.45 For example, a number of trials included in this analysis report more AEs and treatment-related AEs (TRAEs) in the vehicle arm than in the active treatment arm (Table 1).
EFFICACY OF TOPICAL TREATMENTS
Phosphodiesterase 4 Inhibitors
Three vehicle-controlled trials have been conducted to investigate the efficacy of crisaborole ointment, 2%, including 2 identically designed 4 week phase 3 trials10 and a 6-week phase 2a proof-of-concept trial.11 In the phase 3 trials, higher proportions of crisaborole-treated patients than vehicle-treated patients achieved the primary endpoint of ISGA of clear (0) or almost clear with ≥2-grade improvement from baseline at day 29 (Study 1: P=0.038; Study 2: P<0.001) (Table 1, Figure 1).10 Higher proportions of crisaborole-treated patients also achieved the type 1 error-controlled key secondary endpoint of ISGA clear (0) or almost clear (1) versus vehicle (Study 1: 51.7% vs 40.6%, P=0.005; Study 2: 48.5% vs 29.7%, P<0.001).10 In a separate analysis of secondary endpoints based on Severity of Pruritus Scale (SPS; 4 point rating scale adapted from the Atopic Dermatitis Severity Index (ADSI) to assess pruritus severity with a 24 hour recall46), higher proportions of crisaborole-treated patients achieved improvement in pruritus (SPS score ≤1 with ≥1 point improvement from baseline) compared with vehicle (Study 1: 37% vs 21%, P<0.0001; Study 2: 34% vs 21%, P=0.0006), and that crisaborole-treated patients achieved pruritus improvement earlier than vehicle-treated patients (median days, Study 1: 4.0 vs 9.0, P=0.0008; Study 2: 5.0 vs 9.0, P=0.0042).47 In the phase 2a intrapatient comparison trial, lesion-specific efficacy was assessed by change from baseline in ADSI score (none [0] to most severe [15]).11 More patients experienced a greater decrease in ADSI score for the crisaborole-treated lesion than the vehicletreated lesion at day 28 (primary end point; P=0.017) (Table 1, Figure 1) as well as at each of the other time points assessed (days 14 and 42).11
Topical Corticosteroids
In 8 articles summarizing 9 trials, the efficacy of low potency (fluocinolone acetonide in peanut oil, 0.01%; desonide hydrogel, 0.05%),16,17 lower-medium potency (hydrocortisone butyrate ointment and cream, 0.1%),18,19 medium potency (betamethasone valerate cream, 0.1%; triamcinolone acetonide cream, 0.05%),20,21 and very high potency (clobetasol propionate cream and lotion, 0.05%)22,23 TCSs were assessed. Reiterating the importance of vehicle, 7 of these trials explored various vehicles
Three vehicle-controlled trials have been conducted to investigate the efficacy of crisaborole ointment, 2%, including 2 identically designed 4 week phase 3 trials10 and a 6-week phase 2a proof-of-concept trial.11 In the phase 3 trials, higher proportions of crisaborole-treated patients than vehicle-treated patients achieved the primary endpoint of ISGA of clear (0) or almost clear with ≥2-grade improvement from baseline at day 29 (Study 1: P=0.038; Study 2: P<0.001) (Table 1, Figure 1).10 Higher proportions of crisaborole-treated patients also achieved the type 1 error-controlled key secondary endpoint of ISGA clear (0) or almost clear (1) versus vehicle (Study 1: 51.7% vs 40.6%, P=0.005; Study 2: 48.5% vs 29.7%, P<0.001).10 In a separate analysis of secondary endpoints based on Severity of Pruritus Scale (SPS; 4 point rating scale adapted from the Atopic Dermatitis Severity Index (ADSI) to assess pruritus severity with a 24 hour recall46), higher proportions of crisaborole-treated patients achieved improvement in pruritus (SPS score ≤1 with ≥1 point improvement from baseline) compared with vehicle (Study 1: 37% vs 21%, P<0.0001; Study 2: 34% vs 21%, P=0.0006), and that crisaborole-treated patients achieved pruritus improvement earlier than vehicle-treated patients (median days, Study 1: 4.0 vs 9.0, P=0.0008; Study 2: 5.0 vs 9.0, P=0.0042).47 In the phase 2a intrapatient comparison trial, lesion-specific efficacy was assessed by change from baseline in ADSI score (none [0] to most severe [15]).11 More patients experienced a greater decrease in ADSI score for the crisaborole-treated lesion than the vehicletreated lesion at day 28 (primary end point; P=0.017) (Table 1, Figure 1) as well as at each of the other time points assessed (days 14 and 42).11
Topical Corticosteroids
In 8 articles summarizing 9 trials, the efficacy of low potency (fluocinolone acetonide in peanut oil, 0.01%; desonide hydrogel, 0.05%),16,17 lower-medium potency (hydrocortisone butyrate ointment and cream, 0.1%),18,19 medium potency (betamethasone valerate cream, 0.1%; triamcinolone acetonide cream, 0.05%),20,21 and very high potency (clobetasol propionate cream and lotion, 0.05%)22,23 TCSs were assessed. Reiterating the importance of vehicle, 7 of these trials explored various vehicles
