Crisaborole: A Novel Nonsteroidal Topical Treatment for Atopic Dermatitis

Abstract

Objective: To review the efficacy and safety of crisaborole and its place in therapy for the management of mild to moderate atopic dermatitis (AD). Data Sources: A literature search of PubMed (data inception to February 2019) was performed using the search terms crisaborole and atopic dermatitis. Supplementary sources included the Journal of the American Academy of Dermatology Guidelines of Care for the Management of Atopic Dermatitis, clinicaltrials.gov data, manufacturer prescribing information, and article bibliographies. Study Selection and Data Extraction: Relevant English-language studies and those conducted in humans were considered and reviewed. Abstracts from clinical trials and drug reviews were reviewed. Phases I, II, III, and long-term safety studies were included. Data Synthesis: Data from multiple clinical trials have demonstrated the effectiveness and safety of crisaborole topical ointment. Patients treated with crisaborole experienced improvement in AD symptoms based on improvement in the Investigator’s Static Global Assessment and the AD Severity Index scores. Crisaborole has a limited adverse event profile and low systemic absorption. Relevance to Patient Care and Clinical Practice: Crisaborole is the first topical phosphodiesterase 4 inhibitor indicated for the treatment of mild to moderate AD. Its place in therapy is along with topical calcineurin inhibitors as a second-line option for patients who are recalcitrant to or unable to use topical corticosteroids. Conclusions: Crisaborole is a safe and efficacious second-line option for the treatment of mild to moderate AD in patients 2 years of age and older.

Introduction

Atopic dermatitis (AD; also known as eczema and atopic eczema) has been known since antiquity and today is one of the most common chronic diseases in developed countries. Its clinical features must include pruritus and eczematous lesions; it also typically follows a relapsing and remitting course.^1-4 AD develops in approximately 10% to 20% of the population in the developed world, and of those affected, approximately 80% are children; overall AD affects about 25% of children and 2% to 3% of adults.^2,3

The pathophysiology of AD is complex.⁵ Two primary mechanisms of disease are thought to be (1) abnormalities of the epidermal structure and function (aka, the “outside-in” theory) and (2) cutaneous inflammation due to adverse immune responses (aka, the “inside-out” theory).^3,6 In addition, other theories include genetic dysfunction, environmental factors, and impairment of skin barrier integrity. There are several immune-based targets under investigation. These include interleukins (IL), specifically IL-4, IL-5, and IL-13; immuno globulin E; and histamine receptor type 4.^5,7 Finally, crisaborole’s major action involves inhibition of phosphodiesterase 4 (PDE4) enzyme, having a favorable effect on the immune response of regulatory T cells.^5,7

The management of patients with AD includes skin care (warm baths and moisturizers), basic antisepsis (bleach baths) particularly for those with recurrent skin infections, and trigger avoidance (such as known irritants and allergens).^6,8 Thereafter, a mainstay of therapy is the topical corticosteroids. Low- to medium-potency corticosteroids are preferred for acute flares and maintenance therapy. Second-line therapy includes the topical calcineurin inhibitors (pimecrolimus cream 1% [Elidel] and tacrolimus ointment 0.1% or 0.03% [Protopic]). These are considered when topical corticosteroids are problematic and when AD affects the thinned skin areas such as the face and intertriginous regions.⁸

Crisaborole (Eucrisa) is a new drug ointment first approved by the US Food and Drug Administration on December 14, 2016, with the labeled indication for topical treatment of mild to moderate AD in patients 2 years of age and older.⁹ It is a new topical approach to AD (mechanism of action) that is not related to corticosteroids. This article will provide a review of the efficacy and safety of crisaborole and its place in therapy for the management of mild to moderate AD.

Data Sources

A literature search of PubMed (data inception to February 2019) was performed using the search terms crisaborole and atopic dermatitis. Supplementary sources included the Journal of the American Academy of Dermatology Guidelines of Care for the Management of Atopic Dermatitis, clinicaltrials.gov data, manufacturer prescribing information, and article bibliographies.

Pharmacology

Crisaborole is a boron-containing PDE4 inhibitor. The boron atom binds to the bimetal center of the PDE4 enzyme inhibiting its activity in a competitive, reversible way.^10,11 PDE4 inhibition results in increased intracellular cyclic adenosine monophosphate (cAMP) levels. Increased cAMP is a negative regulator of pro-inflammatory cytokines.^10-12 This reduces the production or effects of inflammatory mediators such as IL-4 and IL-13. Specific mechanism(s) are not well defined⁹ (see Figure 1).

Figure 1.

Crisaborole mechanism.^3,7,12

Abbreviations: CB, crisaborole; cAMP, cyclic adenosine monophosphate; PDE4, phosphodiesterase 4.

Simplified version of crisaborole’s mechanism of action. Once atopic dermatitis has occurred and the skin barrier breached, inflammation occurs. cAMP is active to mitigate the effects of various inflammatory cytokines and suppress their action. PDE4 acts to breakdown cAMP to AMP, contributing to the inflammatory response. Crisaborole inhibits the action on PDE4, alleviating inflammation.

Pharmacokinetics

Absorption

The pharmacokinetics of crisaborole were investigated in 33 (29-34 reporting) pediatric subjects 2 to 17 years of age with mild to moderate AD and a mean ± SD body surface area involvement of 49 ± 20% (range 27% to 92%). Subjects applied approximately 3 mg/cm² of crisaborole ointment (dose range 6 g to 30 g per application) twice daily for 8 days.^9,13

Time to maximum plasma concentrations (T_max) after application was consistent over the 8 days at 3 hours (range = 3-24 hours). Two inactive metabolites were also measured, one of which showed accumulation over the 8-day period of approximately 3-fold.¹³

The mean ± SD maximum plasma concentration (C_max) and area under the concentration time curve from 0 to 12 hours post dose (AUC_0-12) for crisaborole on day 8 were 127 ± 196 ng/mL and 949 ± 1240 ng*h/mL, respectively. Steady state of systemic concentrations of crisaborole were achieved by day 8. Based on the ratios of AUC_0-12 between day 8 and day 1, the mean accumulation factor for crisaborole was 1.9.^9,13

Distribution

Distribution studies for crisaborole showed crisaborole is 97% bound to human plasma proteins.⁹

Metabolism

Crisaborole is substantially metabolized into 2 inactive metabolites as determined in a study of 23 adolescent patients (12-17 years) with mild to moderate AD and a mean treatable ± SD body surface area of 17.6 ± 5.7% using crisaborole topical ointment 2% twice daily for 8 days.^5,14 The major metabolite 5-(4-cyanophenoxy)-2-hydroxyl benzylalcohol (metabolite 1) is formed via hydrolysis; this metabolite is further metabolized into downstream metabolites, among which 5-(4-cyanophenoxy)-2-hydroxyl benzoic acid (metabolite 2), formed via oxidation, is also a major metabolite. The mean apparent half-life on day 8 of the parent drug was 11.9 ± 8.28 hours.¹⁴ The major metabolite showed a similar half-life at day 8 (10.5 ± 6.38 hours) and a second metabolite a longer half-life at day 8 (33.5 ± 10.1 hours) perhaps accounting for accumulation.¹⁴

The 2 metabolites’ systemic concentrations were at or near steady state by day 8. Based on the ratios of AUC_0-12 between day 8 and day 1, the mean accumulation factors for metabolites 1 and 2 were 1.7 and 6.3, respectively.^9,13

Excretion

Renal excretion of metabolites is the major route of elimination.⁹

Clinical Trials

The efficacy of crisaborole has been evaluated in multiple clinical trials. Table 1 contains a list of all phase I, II, and III trials. A phase I trial assessed the tolerability of using crisaborole topical ointment, 2%, on healthy adult volunteers’ sensitive and thin-skinned areas. Thirty-two patients ages 18 to 55 years were randomized in double-blind fashion to use crisaborole or vehicle to 13 skin areas twice daily for 21 days. Using the Local Tolerability Scale (scale range from 0 [none] to 3 [severe]), frequency and severity of local tolerability symptoms (burning/stinging, erythema, and pruritus) were noted at various times throughout the study; 98.8% of all of the assessments for both groups had a grade of 0 or no evidence of signs/symptoms of irritation.¹⁵

Table 1.

Clinical Trials of Crisaborole.^13-15,17-20

Reference	Study Design	Patient Characteristics	Number of Participants	Treatment and Duration	Primary Outcomes Measures	Results
Zane et al15	Randomized, double-blind, vehicle-controlled, single-center phase I study	18-55 years, healthy	32	3:1 crisaborole ointment, 2%:vehicle BID to 13 anatomic sensitive, and nonsensitive skin areas for 21 days	Local Tolerability Scale	98.8% of all tolerability assessments had a grade of 0 (no signs/symptoms of irritation); 0.85% had a grade of 1 (mild); 0.1% had a grade of >1
Zane et al13	Multicenter, open-label, maximal use phase 1b study	2-17 years, diagnosis of AD, minimum AD involvement for treatable percentage of BSA depending on age, and ISGA score of 2 (mild) or 3 (moderate)	34	Crisaborole ointment, 2%, BID to affected areas for 28 days	Pharmacokinetics and safety assessments	Median Tmax for crisaborole was 3 hours of day 1 and day 8. Systemic exposure to crisaborole and metabolite (AN7602) and accumulation was minimal. Twenty-three patients reported one or more TEAEs, 95% of which were mild or moderate.
Gold et al17	Multicenter, double-blind, bilateral-comparison, dose-ranging, phase II study	12-17 years, clinical diagnosis of mild to moderate AD, ≤35% BSA involved, 2 lesions with similar severity	86	1:1 crisaborole ointment, QDay:BID for 28 days. For all patients, one lesion received crisaborole ointment, 0.5% and the other received crisaborole ointment, 2%.	Change from baseline in ADSI score for each lesion	All dosing regimens resulted in improved ADSI scores with the greatest improvement observed with crisaborole ointment, 2%, applied BID.
Tom et al14	Multicenter, open-label, phase IIa study	12-17 years, treatable AD lesions involving 10% to 35% BSA (excluding scalp and venous access area) and ISGA score of 2 (mild) or 3 (moderate)	23	Crisaborole ointment, 2%, BID (except for days 1 and 8 when only one single morning dose was applied) for 28 days	PK, systemic exposure, safety, and tolerability	Median Tmax on day 1 was 2.37 hours for crisaborole, 2.08 hours for AN7602, and 6.25 hours for AN 8323. Accumulation after 8 days was limited. Ten patients reported 19 AEs, with application site pain and nasopharyngitis the most common.
Murrell at al18	Multicenter, randomized, double-blind, vehicle-controlled, bilateral evaluation, phase IIa study	18-75 years, diagnosis of mild to moderate AD, 2 comparable lesions	25	BID crisaborole ointment, 2%, or vehicle, each to 1 of the 2 target lesions	Change in ADSI score from baseline to day 28	At day 28, 17 (68%) patients experienced a greater decrease in ADSI scores for the crisaborole-treated lesion compared to the vehicle-treated lesion
Paller et al19	Two multicenter, randomized, double-blind, vehicle-controlled phase III studies (AD-301 and AD-302)	2-79 years, clinical diagnosis of AD and baseline ISGA score of 2 (mild) or 3 (moderate)	AD-301: 763; AD-302: 764	2:1 crisaborole ointment, 2%: vehicle BID to affected areas (except scalp) for 28 days	ISGA score at day 29 of clear (0) or almost clear (1) with a 2-grade or greater improvement from baseline	Greater number of patients treated with crisaborole 2% ointment achieved ISGA scores success (clear [0] or almost clear [1] with a 2-grade or greater improvement from baseline) compared with patients treated with vehicle.
Eichenfield et al20	Multicenter, open-label, long-term, extension, safety study (AD-303)	2-72 years, patients who completed AD-301 or AD-302 without experiencing a treatment-related AE or serious AE related to crisaborole	517	Crisaborole ointment, 2%, BID to affected areas (except scalp) for up to 52 weeks (4-week pivotal study and 48-week long-term safety study)	ISGA score every 4 weeks	During AD-301, AD-302, and AD-303, 65% of patients reported ≥1 TEAE with 51.2% mild and 44.6% moderate. A total of 93.1% of TEAEs reported was considered unrelated to crisaborole. Most frequently reported treatment-related AEs included dermatitis atopic (3.1%), application-site pain (2.3%), and application-site infection (1.2%).

Abbreviations: AD, atopic dermatitis; BSA, body surface area; ISGA, investigator static global assessment; T_max, time to maximum plasma concentrations; TEAE, treatment emergent adverse event; ADSI, AD severity index; PK, pharmacokinetic.

In a phase Ib open-label trial, the Investigator Static Global Assessment (ISGA; scale range from 0 [clear] to 3 [moderate]) was used to assess efficacy of crisaborole topical ointment, 2%, for the treatment of mild to moderate AD in 34 patients ages 2 to 17 years.^13,16 By the completion of the trial, 47.1% of all patients achieved treatment success (defined as an ISGA score of ≤1 with a ≥2-grade improvement from baseline), and 64.7% of all patients achieved a score of 0 (clear) or 1 (almost clear). Improvement in 5 AD symptoms were assessed using the AD severity index (ADSI; scale range from 0 [none] to 3 [severe]). By the end of the trial, there was improvement in all patients’ symptom severity scores (erythema = −64.9%; excoriation = −58.2%; exudation = −64.3%; lichenification = −61.3%; and pruritus = −63.3%).¹³

A double-blind phase II trial evaluated the use of crisaborole topical ointment, 0.5% and 2%, in 86 adolescents ages 12 to 17 years with mild to moderate AD. Eligible patients were required to have 2 AD lesions. Patients were randomized to 2 groups and used crisaborole either once or twice daily. For each patient, 1 lesion was randomly assigned to treatment with crisaborole, 0.5%, and the other lesion was assigned to crisaborole, 2%. Efficacy was assessed using ADSI scores, and all 4 dosing regimens resulted in improvement in mean ADSI scores. The greatest improvement was with lesions treated with crisaborole topical ointment, 2%, twice daily.¹⁷

An open-label phase II trial assessed efficacy of crisaborole topic ointment, 2%, in 23 adolescents ages 12 to 17 years with AD by using mean ISGA and ADSI scores. Treatment success was defined by an ISGA score of ≤1 with 2-grade or greater improvement from baseline. Throughout the study, 39% of patients achieved treatment success. Mean scores for all clinical signs of AD decreased from baseline. Notably, the mean pruritus score decreased from 1.87 at baseline to 0.57 at the end of the trial.¹⁴

One final double-blind phase II trial utilized ADSI scores to assess the efficacy of crisaborole topical ointment, 2%, for the treatment of AD in 25 adults ages 18 to 75 years with mild to moderate AD. Study participants were required to have 2 lesions and treated 1 with crisaborole twice daily and 1 with vehicle twice daily. At the end of the trial, 68% of study participants experienced a greater decrease in ADSI scores for the lesion treated with crisaborole versus the lesion treated with (P = .017).¹⁸

Two identically designed, double-blind, vehicle-controlled phase III trials (AD-301 and AD-302) assessed the efficacy of crisaborole in 1527 patients ages 2 to79 years with mild to moderate AD. Treatment success was defined by an ISGA score of 0 or 1 with a 2-grade improvement from baseline. Other efficacy analyses included the proportion of patients with an ISGA score of 0 or 1 at the end of the trial and time to success in ISGA score. Patients were randomized to either treatment with crisaborole (n = 763) or vehicle (n = 764) twice daily for 28 days. More patients in the crisaborole-treated group achieved treatment success at day 29 in both trials (Trial 1: 32.8% vs 25.4%, P = .038; Trial 2: 31.4% vs 18.0%, P < .001). In both trials, more patients in the crisaborole-treated group achieved ISGA scores of 0 or 1 at day 29 of the trial than in the vehicle-treated group (Trial 1: 51.7% vs 40.6%, P = .005; Trial 2: 48.5% vs 29.7%, P < .001). A greater proportion of patients in the crisaborole-treated groups achieved improvement in all clinical signs of AD and a greater reduction in mean severity (erythema: P < .001; exudation: P = .001; excoriation: P < .001; induration/papulation: P = .002; lichenification: P < .001). Also, a greater proportion of patients in the crisaborole-treated group achieved improvement in pruritus compared with the vehicle-treated group (days 8, 15, and 22: P < .001; day 29: P = .002).¹⁹

Adverse Events

Patients in the 2 identically designed phase III trials tolerated treatment with crisaborole well and reported rates of treatment emergent adverse events (TEAEs) that were similar to patients treated with the vehicle. The most commonly reported adverse event in the trials was application site pain (eg, stinging, burning), which occurred in 4% of patients treated with crisaborole. Application site pain was the only TEAE that 1% or more patients experienced. Of the patients that reported application site pain, 78% reported that it had resolved within 1 day of onset. Of the adverse events reported, 94% were classified as mild to moderate in severity, and 79% were considered unrelated or unlikely to be related to the study drug. The crisaborole treatment group and the vehicle treatment group had the same rate (1.2%) of study discontinuation due to adverse events. Regarding vital signs, electrocardiograms, and laboratory parameters, no clinically significant differences were found between the 2 groups.¹⁹

A follow-up, 48-week safety study (AD-303) was conducted to assess the long-term safety results of using crisaborole in 517 patients ages 2 to 72 years who continued crisaborole treatment after completing AD-301 or AD-302. TEAEs occurred in 10.3% of patients in the phase III trials and the long-term extension study with 85.9% of them being reported as mild or moderate in severity. Table 2 includes the most commonly reported TEAEs.²⁰ These TEAEs are similar to those reported in earlier phase I and II trials.^13-15,17,18

Table 2.

Incidence of Treatment Emergent Adverse Events Reported During Phase III and Long-Term Safety Studies of Crisaborole Topical Ointment.²⁰

Treatment Emergent Adverse Event, n (%)	Incidence (n = 517)
Dermatitis atopica	16 (3.1)
Application-site painb	12 (2.3)
Application-site infection	6 (1.2)

^aIncludes worsening, exacerbation, flare, or flare-up of an existing condition.

^bIncludes burning or stinging.

Based on the information from clinical trials, treatment with crisaborole ointment 2% is a safe option for the treatment of patients 2 years of age and older with mild to moderate AD. The most commonly reported TEAE was application site burning or stinging. The incidence reported with crisaborole (4%) is similar to or lower than the incidence reported with other topical treatments for the treatment of mild to moderate AD. For comparison, the incidence reported with topical corticosteroids is less than 1% to 6% and with topical calcineurin inhibitors is 20% to 58% for tacrolimus and 8% to 26% for pimecrolimus.^21-23

Dosage and Administration

Crisaborole is available as an ointment containing 20 mg of crisaborole per gram (2%) and is indicated for use in patients 2 years of age and older. Patients should apply a thin layer of crisaborole ointment 2% twice daily to the affected area(s). There are no dosage adjustment recommendations for special populations, including patients with renal or hepatic impairment.⁹ Due to a limited number of patients 65 years of age and older in clinical trials, there is insufficient data to determine whether geriatric patients respond differently from younger patients. Regarding the use of crisaborole during pregnancy and breastfeeding, there are no available human data. Animal data has not indicated cause for concern. This medication is only approved for topical use and should not be used for ophthalmic, oral, or intravaginal use.⁹

Drug-Drug Interactions

Based on in vitro studies only, a crisaborole metabolite was noted to be a weak inhibitor of cytochrome (CYP) P450 enzymes 1A2 and 2B6 and a moderate inhibitor of CYP 2C8 and 2C9.⁹ There are many drugs affected by these isoenzymes. Given that crisaborole is a weak to moderate inhibitor (based on in vitro studies) and its limited absorption, speculation of clinically significant drug-drug interactions is premature.

Riociguat, a soluble guanylate cyclase stimulator, is contraindicated with both nonselective PDE and PDE5 inhibitors due to the potential of these medications to enhance the hypotensive effect of riociguat.^24,25

Place in Therapy and Future Directions

Crisaborole is the first topical PDE4 inhibitor indicated for the treatment of mild to moderate AD in patients 2 years of age and older. Results from multiple clinical trials showed that approximately 31% to 47% of patients achieved treatment success while approximately 49% to 65% of patients achieved an ISGA score of clear or almost clear.¹⁹

Crisaborole is an appealing option due to its limited adverse event profile and low systemic absorption. Topical corticosteroids are associated with adverse events including skin atrophy, telangiectasia, and striae as well as systemic absorption and the risk of hypothalamic-pituitary-adrenal axis suppression.⁹ Although this risk is minimal, it increases with prolonged exposure and is often a concern for use in children, the most susceptible population and the population most affected by this chronic skin disorder.⁹ In a study with healthy volunteers, crisaborole was shown to be well tolerated when applied to sensitive skin areas, areas in which topical corticosteroids are often restricted.¹⁵

Topical calcineurin inhibitors are associated with local application site reactions such as stinging and burning skin.⁹ These adverse events are seen more commonly with topical calcineurin inhibitors than with topical corticosteroids but have a tendency to diminish over time. The 2 topical calcineurin inhibitors, tacrolimus and pimecrolimus, have a US Food and Drug Administration black box warning as rare cases of malignancy (eg, skin and lymphoma) have been reported with their use.^22,23 Since the warning was established, clinical trial data have not shown a correlation between topical calcineurin inhibitor use and increased rates of malignancies.⁹

One drawback of crisaborole is its price compared with topical corticosteroids. Current average wholesale prices (AWP) for crisaborole, tacrolimus, and pimecrolimus are comparable at approximately $10 per gram, whereas the costs of topical corticosteroids are significantly less with AWPs as low as several cents per gram.^25-28

Although approximately 31% to 47% of patients achieved treatment success and approximately 49% to 65% of patients achieved an ISGA score of clear or almost clear in clinical trials, the results are not convincing enough to recommend crisaborole as a first-line therapy over or alongside topical corticosteroids. Crisaborole remains second-line therapy behind topical corticosteroids and should be reserved for situations in which patients are recalcitrant to or are unable to use corticosteroids.

A phase IV head-to-head study in the recruitment phase plans to evaluate the safety and efficacy of several options in patients at least 2 years of age for the treatment of mild to moderate AD, including crisaborole ointment 2%, hydrocortisone butyrate cream 0.1%, pimecrolimus cream 1%, and crisaborole vehicle.²⁹ In the first of its kind, this study will compare crisaborole with other approved treatment options. Another phase IV study in the recruitment phase will evaluate the safety, pharmacokinetics, and efficacy of crisaborole ointment 2% when used in patients ages 3 months to less than 24 months with mild to moderate AD.³⁰

Conclusion

Crisaborole has been shown to be a safe and efficacious option for the treatment of mild to moderate AD in patients ages 2 years of age and older. In clinical trials, crisaborole has improved AD symptoms including pruritus, erythema, exudation, excoriation, induration/papulation, and lichenification. Results from 2 large phase III trials showed that approximately 32% of patients achieved treatment success. Although it is a favorable treatment option along with topical corticosteroids and topical calcineurin inhibitors due to its efficacy and low incidence of adverse effects, crisaborole is considerably more expensive than the current first-line therapy. Its place in therapy is along with topical calcineurin inhibitors as a second-line option for patients who are recalcitrant to or unable to use topical corticosteroids. Future trials will compare crisaborole with first- and second-line treatment options and may expand the indication to include patients less than 2 years of age. This evidence may expand guideline recommendations for the treatment of AD.