Efficacy and safety of tralokinumab with concomitant topical corticosteroids in  
North American adults with moderate-to-severe atopic dermatitis: a subanalysis of the ECZTRA 3 trial 

Boni E. Elewski,1 Matthew J. Zirwas,2 Richard G. Langley,3 Andrew F. Alexis,4 Karen A. Veverka,5 John Zoidis,5 Azra Kurbasic,6 Jonathan I. Silverberg7

1University of Alabama, Birmingham, AL, USA; 2Probity Medical Research, Columbus, OH, USA; 3Dalhousie University, Halifax, Nova Scotia, Canada; 4Icahn School of Medicine at Mount Sinai, New York, NY, USA;  
5LEO Pharma, Madison, NJ, USA; 6LEO Pharma A/S, Ballerup, Denmark; 7The George Washington University School of Medicine and Health Sciences, Washington, DC, USA

LEO 6

Fall Clinical Dermatology Conference, Las Vegas, NV, USA, October 29−November 1, 2020

Introduction
• Atopic dermatitis (AD) is a common, chronic inflammatory skin disease, characterized 

by excessive pruritus and sleep disturbance, among other symptoms1-3

• Tralokinumab is a fully human monoclonal antibody that specifically neutralizes 
interleukin-13 (IL-13), a key cytokine of the chronic type 2 inflammation underlying AD;  

IL-13 is overexpressed in lesional and non-lesional AD skin4-6

• ECZTRA 3 (NCT03363854) was a Phase 3, randomized, double-blind, placebo-
controlled trial that evaluated the efficacy and safety of subcutaneous tralokinumab 

300 mg every 2 weeks (q2w) vs. placebo (after a loading dose of 600 mg), in 

combination with topical corticosteroids (TCS) as needed, for an initial treatment 

period of 16 weeks in adults with moderate-to-severe AD across Europe and  

North America

 — Significantly more patients achieved the primary endpoints of Investigator’s Global 

Assessment (IGA) score of 0/1 (clear/almost clear) and/or a 75% improvement in 

Eczema Area and Severity Index (EASI-75) at week 16 with tralokinumab plus TCS 

compared with placebo plus TCS

 — Tralokinumab demonstrated improvements vs. placebo across key secondary 

endpoints in patient-reported outcomes (Dermatology Life Quality Index [DLQI], 

pruritus Numeric Rating Scale [NRS], and SCORing Atopic Dermatitis [SCORAD]) at 

week 16 

 — Cumulative TCS use in tralokinumab-treated patients was lower than that of those 

who received placebo at week 16, suggesting achievement of endpoints was not 

likely attributable to TCS use alone

Objective
• To evaluate the efficacy and safety of tralokinumab 300 mg q2w in combination with 

TCS in the ECZTRA 3 North American subpopulation at week 16

Methods
Study design and patients

• ECZTRA 3 was a randomized, double-blind, placebo-controlled, 32-week trial in adult 
patients with moderate-to-severe AD (Figure 1)

• Patients were enrolled from Europe (Belgium, Germany, The Netherlands, Poland, Spain, 
and UK) and North America (USA and Canada)

• Eligible patients were 18 years of age, with a confirmed diagnosis of AD for 1 year and 
AD involvement of  10% of body surface area, EASI score of  12 at screening and  

16 at baseline, IGA score of 3, pruritus NRS score of 4, and were candidates for 

systemic therapy due to a recent (within 1 year) history of inadequate response or 

intolerance to topical treatment

• Patients were stratified by region and baseline disease severity (IGA-3 [moderate] or  
IGA-4 [severe]) and were randomized 2:1 to receive subcutaneous tralokinumab  

300 mg or placebo q2w (after a loading dose of 600 mg), plus TCS as needed, for an 

initial treatment period of 16 weeks

• Use of TCS (mometasone furoate: US Class 4 [midstrength]) was permitted as early as 
day 0, after a washout period of 2 weeks for TCS 

• Rescue treatment, which included higher-potency TCS (e.g. clobetasol), was permitted in the 
form of topical and systemic medications to control intolerable AD symptoms 

Endpoints

• Primary endpoints were defined as IGA-0/1 and/or EASI-75 at week 16

• Key secondary endpoints included reduction of worst daily pruritus NRS (weekly average) 
of at least 4 from baseline to week 16 and change from baseline to week 16 in SCORAD 

and DLQI

Safety assessments

• Adverse events were collected from the first trial-related activity after patients provided 
informed consent until completion of the clinical trial

Statistical analysis 

• For binary endpoints, the difference in response rates between treatment groups was analyzed 
using the Cochran-Mantel-Haenszel test, stratified by baseline IGA score; patients receiving 

rescue medication prior to week 16 or with missing data were considered non-responders

• Continuous endpoints were assessed using a mixed-effect model for repeated 
measurements, with an unstructured covariance matrix to model within-patient 

variation and the mean change modeled as: change from baseline = treatment*week 

+ baseline*week + baseline IGA; denominator degrees of freedom were estimated using 

Kenward-Roger approximation

 — Data collected after permanent discontinuation of investigational medicinal product 

or after initiation of rescue medication were excluded from the analysis

• Descriptive statistics were used to present baseline demographics, baseline disease 
characteristics, and safety assessments

Results
Patient characteristics

• In total, 380 patients were randomized in ECZTRA 3, with 160 patients (42.1%) from  
North America (Table 1)

• Overall, the North American and primary study populations had similar baseline 
demographics, although there was slight variation in the baseline disease characteristics:7

 — The percentage of severe AD (IGA-4) was slightly lower in the North American 

population, although mean EASI scores did not objectively differ

 — Mean body surface area involvement with AD was slightly lower in the North American 

population

IGA and EASI-75 at week 16

• At week 16, a numerically higher proportion of tralokinumab-treated patients in the  
North American population achieved IGA-0/1 compared with placebo (40.0% vs. 25.9%) 

[Figure 2]

• A higher proportion of tralokinumab-treated patients in the North American population 
achieved EASI-75, compared with placebo (58.1% vs. 37.0%) at week 16

Use of rescue medication

• Use of rescue medication, which included higher-potency TCS or systemic treatment for 
AD, was low in the North American population during the initial treatment period (Figure 3)

 — When compared with placebo, rescue medication use was lower in tralokinumab-

treated patients (4.7% vs. 9.3%)

Change in SCORAD, DLQI, and pruritus NRS 

• Reduction in SCORAD (−38.6 vs. −23.0) and DLQI (−11.5 vs. −7.6) were greater with 
tralokinumab compared with placebo in the North American population from baseline  

to week 16 (Figure 4) 

• A numerically greater proportion of the North American population treated with 
tralokinumab achieved a worst daily pruritus NRS reduction of 4 at week 16 compared 

with placebo (42.3% vs. 31.5%) [Figure 5]

Safety

• The overall rate of adverse events was similar between tralokinumab and placebo 
groups in the North American population (Table 2)

• Most adverse events were mild to moderate in severity

Figure 1. ECZTRA 3 trial design

Placebo q2w + TCS

Tralokinumab q2w + TCS

Tralokinumab q4w + TCS (n=69)

Tralokinumab q2w + TCS (n=69)

Screening Continuation treatment

Clinical response defined as 
IGA-0/1 or EASI-75 O�-treatment

period

Safety
follow-up

0 32 weeks 46weeks16 weeks-6 weeks

n=253

This ECZTRA 3 sub-analysis focused on the initial treatment period

n=127

2:1 randomization

1:1  re-randomization

Washout of 
TCS and 
other AD

medication

Initial treatment

300 mg q2w after initial
 loading dose (600 mg)

16-week responders

16-week non-responders

16-week responders

Primary endpoints

•  IGA-0/1 and/or EASI-75 at week 16

Secondary endpoints

•  Change in SCORAD from baseline to week 16

•  Change in DLQI score from baseline to week 16

•  Reduction of worst daily pruritus NRS (weekly average) 
   �4 from baseline to week 16

•  Adverse events/serious adverse events by preferred term

16-week non-responders

Tralokinumab q2w + TCS (n=95)

Placebo q2w + TCS (n=41)

Tralokinumab q2w + TCS (n=79)

Figure 3. Rescue medication use in the ECZTRA 3 North American population during the  
initial 16-week treatment period

P
a

ti
e

n
ts

, %

12

8

10

6

4

2

0

9.3%

4.7%

Tralokinumab q2w
+ TCS (n=104)

Placebo + TCS
(n=54)

Treatment

Figure 4. Change in (A) SCORAD and (B) DLQI in the ECZTRA 3 North American population  
at week 16

C
h

a
n

g
e

 in
 S

C
O

R
A

D

A

0

−50

−40

−30

−20

−10

C
h

a
n

g
e

 in
 D

LQ
I

B

0

−14

−8

−6

−12

−10

−4

−2

−38.6
***

−11.5
***

−7.6
−23.0

Tralokinumab q2w
+ TCS (n=105)

Placebo + TCS
(n=54)

TreatmentTreatment

Tralokinumab q2w
+ TCS (n=105)

Placebo + TCS
(n=54)

***P0.001 vs. placebo + TCS.
Data collected after permanent discontinuation of investigational medicinal product or initiation of rescue medication not included. 
Repeated measurements model: change from baseline = treatment*week + baseline*week + baseline IGA.

Figure 5. Proportion of patients achieving a reduction in pruritus NRS 4 in the ECZTRA 3  
North American population at week 16

P
a

ti
e

n
ts

, %

50

40

30

20

10

0

31.5%

42.3%

Tralokinumab q2w
+ TCS (n=104)

Placebo + TCS
(n=54)

Treatment
Patients receiving rescue medication prior to week 16 or with missing data were considered non-responders.

Figure 2. (A) IGA-0/1 and (B) EASI-75 in the ECZTRA 3 North American population at week 16

P
a

ti
e

n
ts

, %

A

0

P
a

ti
e

n
ts

, %

B

70

30

20

10

0

40

50

60

70

30

20

10

40

50

60

25.9%

40.0%

Tralokinumab q2w
+ TCS (n=105)

Placebo + TCS
(n=54)

Tralokinumab q2w
+ TCS (n=105)

Placebo + TCS
(n=54)

TreatmentTreatment

37.0%

58.1%
*

*P0.05 vs. placebo + TCS.
Patients receiving rescue medication prior to week 16 or with missing data were considered non-responders.

Table 1. Patient demographics and disease characteristics at baseline

North America
(n=160)

Tralokinumab q2w + TCS
(n=106)

Placebo + TCS
(n=54)

Mean age (SD) 42.4 (17.6) 39.5 (16.2)

Male, % 46.2 53.7

Mean duration of AD, years (SD) 27.1 (18.3) 27.7 (15.7)

IGA score of 4, % 37.1 38.9

Mean EASI score (SD) 26.2 (11.4) 28.3 (11.4)

Mean SCORAD (SD) 65.2 (13.4) 67.5 (13.2)

Mean DLQI (SD) 16.5 (7.3) 18.0 (7.1)

Mean weekly average worst daily pruritus NRS score (SD) 7.7 (1.6) 8.2 (1.6)

Mean BSA involvement with AD, % (SD) 41.0 (20.6) 41.2 (23.6)

BSA, body surface area; SD, standard deviation.

Table 2. Adverse events in the initial 16-week treatment period

n (%)

North America (n=160)

Tralokinumab q2w + TCS
(n=105)

Placebo + TCS
(n=54)

At least one adverse event 60 (57.1) 28 (51.9)

At least one serious adverse event 0 1 (1.9)

Adverse event leading to withdrawal from trial 2 (1.9) 0

Severity
Mild 
Moderate
Severe

50 (47.6)
18 (17.1)

0

20 (37.0)
12 (22.2)
3 (5.6)

Outcome
Not recovered/not resolved
Recovering/resolving
Recovered/resolved
Recovered/resolved with sequelae

18 (17.1)
2 (1.9)

55 (52.4)
1 (1.0)

5 (9.3)
2 (3.7)

25 (46.3)
N/A

N/A, not available.

References 
1. Weidinger S, Novak N. Lancet 2016; 387: 1109–1122.

2. Silverberg JI et al. Ann Allergy Asthma Immunol 2018;  
121: 340–347.

3. Dalgard FJ et al. J Invest Dermatol 2015; 135: 984–991. 

4. Bieber T. Allergy 2020; 75: 54–62.

5. Tsoi LC et al. J Invest Dermatol 2019; 139: 1480–1489.

6. Popovic B et al. J Mol Biol 2017; 429: 208–219.

7. Silverberg JI et al. Br J Dermatol 2020; in press.

Disclosures
• Boni E. Elewski has received honoraria as a consultant from Boehringer Ingelheim, Bristol-Myers Squibb, Celgene, LEO Pharma, 

Lilly, Menlo Therapeutics, Novartis, Pfizer, Sun, Valeant (Ortho Dermatologics), and Verrica and received research funding from 
AbbVie, AnaptysBio, Boehringer Ingelheim, Bristol-Myers Squibb, Celgene, Incyte, LEO Pharma, Lilly, Menlo Therapeutics, Merck, 
Novartis, Pfizer, Regeneron, Sun, Valeant (Ortho Dermatologics), and Vand

• Matthew J. Zirwas has acted as a consultant for AbbVie, Aclaris, Arcutis, Asana, Aseptic MD, Avillion, DS Biopharma, Fitbit, 
Foamix, Genentech, Incyte, Janssen, Leo Pharma, Lilly, L’Oreal, Menlo, Novartis, Ortho Dermatologics, Pfizer, Regeneron, Sanofi, 
and UCB

• Richard G. Langley has received honoraria as an advisory board member from AbbVie, Amgen, Boehringer Ingelheim, Celgene, 
Leo Pharma, Lilly, Merck, Novartis, Pfizer, and UCB, and as a speaker from AbbVie, Amgen, Celgene, Leo Pharma, Merck, 
Novartis, and Pfizer 

• Andrew F. Alexis has acted as a consultant for Beiersdorf, Bristol-Myers Squibb, Celgene, Dermavant, Foamix, Galderma, LEO 
Pharma, L’Oreal, Menlo Therapeutics, Novartis, Pfizer, Sanofi/Regeneron, Scientis, UCB, Unilever, and Valeant (Bausch Health) 
and received grants/research support from Almirall, Bristol-Myers Squibb, Cara, Celgene, Galderma, LEO Pharma, Menlo 
Therapeutics, Novartis, and Valeant (Bausch Health)

• Karen A. Veverka, John Zoidis, and Azra Kurbasic are employees of LEO Pharma
• Jonathan I. Silverberg has received honoraria as a consultant/advisory board member from LEO Pharma and acted as a 

consultant for, and/or received grants/honoraria from, AbbVie, AnaptysBio, Asana Biosciences, Galderma Research and 
Development, GlaxoSmithKline, Glenmark Generics, Kiniksa, LEO Pharma, Lilly, MedImmune, Menlo Therapeutics, Pfizer, PuriCore, 
Regeneron, and Sanofi

Acknowledgments
• The ECZTRA 3 study was sponsored by LEO Pharma
• Medical writing and editorial assistance were provided by Henna Potigadoo, MSc, and Lauren Smith, BA (Hons), from McCann 

Health, funded by LEO Pharma

• The North American population represented 42.1% of the overall ECZTRA 3 
study population 

• In this subanalysis of the ECZTRA 3 trial, tralokinumab 300 mg q2w plus TCS 
was well tolerated and displayed superior efficacy in patients with moderate-
to-severe AD in the North American population compared with placebo

 — Tralokinumab plus TCS demonstrated improvements in AD symptoms and 
patient quality of life

• Tralokinumab plus TCS was well tolerated in the North American population, 
suggesting no special considerations in safety for this trial subpopulation 
are required 

• Overall, tralokinumab plus TCS displayed similar efficacy and safety across 
the North American population comparable to that of the primary study 
population7

Conclusions