Direct and Indirect Effects of Crisaborole Ointment on Quality of Life in 

Patients With Atopic Dermatitis: A Mediation Analysis
Eric Simpson,1 Gil Yosipovitch,2 Andrew G. Bushmakin,3 Joseph C. Cappelleri,3 Thomas Luger,4 Sonja Ständer,5  

Wynnis Tom,6 William C. Ports,3 Anna M. Tallman,7 Huaming Tan,3 Robert A. Gerber3

1Oregon Health and Science University, Portland, OR, USA; 2University of Miami, Miller School of Medicine, Miami, FL, USA; 3Pfizer Inc., Groton, CT, USA; 4Department of Dermatology, University Hospital Münster,  
Münster, Germany; 5Center for Chronic Pruritus, Department of Dermatology, University Hospital Münster, Münster, Germany; 6Rady Children’s Hospital-San Diego, San Diego, CA, USA; 7Pfizer Inc., New York, NY, USA

BACKGROUND
• Atopic dermatitis (AD) is a chronic, inflammatory skin disease characterized by 

intensely pruritic eczematous lesions1,2

• Itch has a significant impact on quality of life (QoL) in children and adults, and it is 
one of the most important aspects of the disease that patients use to judge treatment 
response3,4

• Corticosteroids and calcineurin inhibitors are recommended for topical treatment 
of AD5-7; however, there is a need for new, effective, nonsteroidal treatments that 
address inflammation and itch without the potential limitations associated with current 
topical agents

• Crisaborole ointment is a nonsteroidal phosphodiesterase 4 inhibitor for the treatment 
of mild to moderate AD8

 – In 2 identically designed Phase 3 clinical studies (AD-301: NCT02118766; AD-302: 
NCT02118792), crisaborole ointment, 2%, significantly improved global disease 
severity and all measured signs and symptoms of AD, and did not result in any 
treatment-related serious treatment-emergent adverse events8

• The most common treatment-related adverse event was application site pain 
(pooled AD-301 and AD-302 population; crisaborole: 4.4%, vehicle: 1.2%)

• A qualitative and psychometric analysis of the Severity of Pruritus Scale (SPS), a 
4-point rating scale ranging from 0 (“no itching”) to 3 (“bothersome itching/scratching 
which is disturbing sleep”), used in the Phase 3 studies, was recently completed, 
supporting the use of SPS as a valid measure of pruritus in AD (see posters on 
display by Yosipovitch G et al)9,10

• Mediation modeling has been used to establish the contributions of direct and indirect 
effects of a treatment on an outcome11,12

OBJECTIVES
• Through mediation modeling, determine the interrelationship among patient-reported 

pruritus (as measured by SPS), QoL (as measured by the Dermatology Life Quality 
Index [DLQI] or the Children’s Dermatology Life Quality Index [CDLQI]), and treatment 
using pooled data from AD-301 and AD-302

METHODS
Study Treatment

• In the Phase 3 studies, patients aged ≥2 years were randomly assigned in a 2:1 ratio 
to receive crisaborole or vehicle ointment

• Treatment was applied twice daily for 28 days
• QoL was measured using the DLQI in patients aged ≥16 years and the CDLQI in 

patients aged 2-15 years (Table 1)13,14

Table 1. QoL Assessment Scales and Subscales: CDLQI and DLQI

Category Assessment

CDLQI
Patients Aged 

2-15 Years

DLQI
Patients Aged 

≥16 Years

Symptoms & Feelings

Severity of symptoms  
(itch, soreness, pain, stinging)

0-3 pts 0-3 pts

Embarrassment or  
self-consciousness

0-3 pts 0-3 pts

Personal Relationships

Effect on friendships and  
social interactions (eg, teasing, 

bullying avoidance)
0-6 pts NA

Effect on friendships, relatives,  
and/or partner, and sex life

NA 0-6 pts

School/Work & Holidays
Effect of skin on work/school or 

vacation time
0-3 pts 0-3 pts

Leisure
Effect on playing sports and  

leisure activities
0-6 pts 0-6 pts

Wearing different clothes/shoes 0-3 pts NA

Burden of Treatment Treatment burden on daily life 0-3 pts 0-3 pts

Sleep Effect of skin on sleep 0-3 pts NA

Daily Activities
Influence on clothes worn and  

daily tasks
NA 0-6 pts

Total
Comprehensive assessment  

of patient QoL
0-30 pts 0-30 pts

CDLQI, Children’s Dermatology Life Quality Index; DLQI, Dermatology Life Quality Index; NA, not applicable; pts, points; QoL, quality of life.

Pruritus Scale and Mediation Modeling

• Mediation in its simplest form is represented by a third variable (M, the mediator),  
so that the predictor X influences the mediator M, which, in turn, influences the 
outcome Y (X affects M and then M affects Y) (Figure 1)15

Figure 1. Basic mediation model.

Mediator variable (M)

Direct effect

a

b

c

e2

Indirect effect

Predictor variable (X) Outcome variable (Y)

e1

a represents the effect of the predictor (independent) variable X on the mediator variable M; b represents the direct effect of the predictor variable 
X on the outcome variable Y; c represents the effect of the mediator variable M on the outcome variable Y; e1 and e2 are the error terms.

• Severity of pruritus was assessed using the SPS (Table 2)
 – SPS was administered via electronic diary twice a day (morning and evening,  
with a recall period of 24 hours)

Table 2. Severity of Pruritus Scale (SPS)
Score Grade Definition

0 None No itching
1 Mild Occasional, slight itching/scratching

2 Moderate
Constant or intermittent itching/scratching which is  

not disturbing sleep
3 Severe Bothersome itching/scratching which is disturbing sleep

• Mediation model consisted of the following variables:
 – Independent variable—treatment (crisaborole vs vehicle)
 – Mediator variable—SPS score (averaged SPS scores over week 4 [days 23-29]  
for every patient to be consistent with 1-week recall period of the DLQI and CDLQI)
 – Outcome variable—DLQI or CDLQI (at day 29; 1-week recall)

• All available data were used, and no imputations of missing data were performed

RESULTS
Patients Demographics and Disposition

• Between both studies, 1016 patients were randomly assigned to receive crisaborole 
and 506 patients were randomly assigned to receive vehicle (intent-to-treat 
population)

• Baseline demographics and disease characteristics were balanced between the 
treatment arms

 – The mean age between both groups was approximately 12.2 years; most patients 
(>86%) were 2-17 years of age
 – Approximately 55.6% were female; most (80%) were non-Hispanic
 – Between both groups, distribution by race was approximately 61% white,  
28% black, 5% Asian, and 6% other
 – Baseline disease characteristics are summarized in Table 3

Table 3: Baseline Disease Characteristics
Crisaborole

n = 1016
Vehicle
n = 506

ISGA, n (%)
Mild – 2
Moderate – 3

393 (38.7)
623 (61.3)

193 (38.1)
313 (61.9)

Severity of pruritus,a %
None – 0 
Mild – 1 
Moderate – 2 
Severe – 3 

35 (3.9)
229 (25.4)
331 (36.7)
308 (34.1)

19 (4.3)
119 (27.0)
167 (37.9)
136 (30.8)

Treatable % BSA
Mean (SD)
Range

18.3 (18.02)
5-95

18.1 (17.33)
5-90

CDLQI
N
Mean (SD)

797
9.3 (5.99)

403
9.0 (6.02)

DLQI
N
Mean (SD)

192
9.7 (6.29)

92
9.3 (6.55)

BSA, body surface area; CDLQI, Children’s Dermatology Life Quality Index; DLQI, Dermatology Life Quality Index; ISGA, Investigator’s Static 
Global Assessment; SD, standard deviation; SPS, Severity of Pruritus Scale.
aSeverity of pruritus was patient- or parent/caregiver-reported and measured using the SPS.

Presented at the 2017 Fall Clinical Dermatology Conference; October 12-15, 2017; Las Vegas, NV

Mediation Models

• 226 patients were included in the DLQI analysis, and 1112 patients were included in 
the CDLQI analysis

• The indirect effect of crisaborole on QoL via pruritus constituted 51% of the overall 
effect of active treatment (P = 0.0272) in the DLQI-based model and 72% in the 
CDLQI-based model (P < 0.0001) (Figures 2, 3)

• This suggested that the effects of crisaborole on QoL were mostly mediated by 
improvement in severity of pruritus 

 – The direct effect (representing all other effects) of crisaborole on QoL was less 
than half (49% [DLQI-based model; P = 0.0365] and 28% [CDLQI-based model; 
P = 0.0701]) the total, or overall, effect of the active treatment on QoL (Figures 2, 3)

Figure 2. DLQI-based mediation model.

Pruritus (SPS)

Direct path:
48.6% (P < 0.05)

Indirect path:
51.4% (P < 0.05)

Treatment Quality of life (DLQI)

SPS, Severity of Pruritus Scale; DLQI, Dermatology Life Quality Index.

Figure 3. CDLQI-based mediation model.

Pruritus (SPS)

Direct path:
27.6% (P > 0.05)

Indirect path:
72.4% (P < 0.05)

Treatment Quality of life (CDLQI)

SPS, Severity of Pruritus Scale; DLQI, Dermatology Life Quality Index.

CONCLUSIONS
• Mediation modeling can be used to help explain the effect of a treatment on 

an outcome

• The presented mediation models indicate that crisaborole affects QoL 

mostly indirectly through improvement in the severity of pruritus

• Indirect effects in the CDLQI-based model were more pronounced, possibly 

because of differences in item composition of the questionnaires; for 

example, CDLQI includes sleep, which is highly affected by pruritus

REFERENCES
1. Hong J et al. Semin Cutan Med Surg. 2011;30(2):71-86.
2. Bieber T. N Engl J Med. 2008;358:1483-1494.
3. Drucker AM et al. J Invest Dermatol. 2017;137(1):26-30.
4. von Kobyletzki LB et al. Acta Derm Venereol. 2017;97(1):86-90.
5. Eichenfield LF et al. J Am Acad Dermatol. 2014;71:116-132.
6. Schneider L et al. J Allergy Clin Immunol. 2013;131(2):295-299.e1-27.
7. Wollenberg A et al. J Eur Acad Dermatol Venereol. 2016;30(5):729-747.
8. Paller AS et al. J Am Acad Dermatol. 2016;75(3):494-503.e6.
9. Yosipovitch G et al. Validation of the Severity of Pruritus Scale 

(SPS) for the assessment of pruritus in atopic dermatitis (AD). 
Presented at: 2017 Fall Clinical Dermatology Conference;  
October 12-15, 2017; Las Vegas, NV.

10.  Yosipovitch G et al. The effect of crisaborole ointment, 2%, on 
pruritus in patients with atopic dermatitis (AD): an extended 
analysis. Presented at: 2017 Fall Clinical Dermatology 
Conference; October 12-15, 2017; Las Vegas, NV.

11. Bushmakin AG et al. J Dermatolog Treat. 2015;26(1):19-22.
12. Panés J et al. J Crohns Colitis. 2016;10(11):1310-1315.
13. Lewis-Jones MS, Finlay AY. Br J Dermatol. 1995;132(6):942-949.
14. Finlay AY, Khan GK. Clin Exp Dermatol. 1994;19(3):210-216.
15.  Cappelleri JC et al. Mediation models. In: Patient-Reported 

Outcomes: Measurement, Implementation and Interpretation. 
Boca Raton, FL: Chapman & Hall/CRC Press; 2013:221-259.

ACKNOWLEDGMENTS
Editorial/medical writing support under the guidance of the authors was provided by Corey Mandel, PhD, and  
Robert Schoen, PharmD, at ApotheCom, San Francisco, CA, USA, and was funded by Pfizer Inc., New York, NY, USA, 
in accordance with Good Publication Practice (GPP3) guidelines (Ann Intern Med. 2015;163:461-464). 


	FC17PosterPfizerSimpsonDirectandIndirectEffectsQualityofLife.pdf