Sie können Operatoren mit Ihrer Suchanfrage kombinieren, um diese noch präziser einzugrenzen. Klicken Sie auf den Suchoperator, um eine Erklärung seiner Funktionsweise anzuzeigen.
Findet Dokumente, in denen beide Begriffe in beliebiger Reihenfolge innerhalb von maximal n Worten zueinander stehen. Empfehlung: Wählen Sie zwischen 15 und 30 als maximale Wortanzahl (z.B. NEAR(hybrid, antrieb, 20)).
Findet Dokumente, in denen der Begriff in Wortvarianten vorkommt, wobei diese VOR, HINTER oder VOR und HINTER dem Suchbegriff anschließen können (z.B., leichtbau*, *leichtbau, *leichtbau*).
Baricitinib is approved for the treatment of adults with moderate-to-severe atopic dermatitis (AD) who are candidates for systemic therapy and has received regulatory authorization in Europe for moderate-to-severe AD in patients 2 to <18 years.
Objective
This study aims to optimize dosing for baricitinib in pediatric patients with atopic dermatitis using pharmacokinetic/pharmacodynamic modeling leveraging adult data.
Methods
The phase III, randomized, double-blind, placebo-controlled study, BREEZE-AD-PEDS (NCT03952559, registration date: 2019-05-16), enrolled patients (aged 2 to <18 years) with moderate-to-severe AD. During a pharmacokinetic (PK) lead-in period, baricitinib concentration data from age-based dose cohorts (4 mg once daily [QD]: 10 to <18 years; 2 mg QD: 2 to <10 years) were compared with actual and simulated concentration values from adult patients receiving baricitinib 4 mg QD. A population PK model incorporating allometric scaling was developed to determine weight-based dosing in pediatric patients that matches adult exposures. The exposure–response (E-R) relationships were analyzed for the primary endpoint: a validated Investigator Global Assessment® (vIGA-AD) score of 0 or 1 (clear to almost clear skin) with ≥2-point improvement from baseline at week 16. Baricitinib pharmacokinetics were characterized from 393 pediatric patients using a 2-compartment model with allometric scaling on clearance and volume of distribution.
Results
The age-based and subsequent weight-based dosing (2 mg for patients 10 to <30 kg and 4 mg for patients ≥30 kg) was comparable to the 4-mg adult exposure. A clear E-R relationship was observed for the primary endpoint when sorted for age or weight groups.
Conclusion
The population PK model developed using baricitinib concentrations from adult patients, with allometric scaling for weight on clearance and volume, adequately predicted exposures in the pediatric population. The PK modeling, with E-R analysis, informed an appropriate weight-based dosing regimen.
R.L. Decker and C.S. Ernest II were employees of Eli Lilly and Company at the time of this study.
Key Points
Baricitinib is approved for the treatment of adults with moderate-to-severe atopic dermatitis (AD) who are candidates for systemic therapy, and has received regulatory authorization in Europe for moderate-to-severe AD in patients aged 2 to <18 years.
In BREEZE-AD PEDS, a phase III, double-blind, placebo-controlled study in patients aged 2 to <18 years with moderate-to-severe AD, baricitinib was tested for age-based dose cohorts (4 mg once daily [QD]: 10 to <18 years; 2 mg QD: 2 to <10 years).
A population pharmacokinetic (PK) modeling conducted using the previously developed adult PK model with the incorporation of allometric scaling for weight on clearance and volume adequately described exposures and informed commercial doses for the pediatric population.
1 Introduction
Atopic dermatitis (AD) is a chronic inflammatory skin disease characterized by intense skin itch and eczema lesions that is associated with poor health-related quality of life and affects up to 20% of children and 2–5% of adults worldwide [1]. Baricitinib is a Janus kinase (JAK) inhibitor demonstrating selectivity for and inhibition of JAK1 and JAK2 with lower potency towards inhibition of JAK3 or TYK2 [2]. Inhibition of JAK-Signal Transducer and Activator of Transcription (STAT) signaling by baricitinib can target multiple cytokine pathways implicated in AD pathogenesis and may provide novel therapeutic approaches to disease management [3].
Anzeige
Baricitinib has demonstrated clinical safety and efficacy in adult patients with moderate-to-severe AD throughout the BREEZE-AD clinical trial program and is approved in more than 70 countries to treat moderate-to-severe AD in adults who are candidates for systemic therapy [4‐9]. Throughout the phase III clinical trials in adults, baricitinib 4 mg and 2 mg once daily (QD) was efficacious across domains of efficacy, including measures of disease severity, skin lesions, pruritus, skin pain, sleep disturbance, and health-related quality of life. Similarly, baricitinib 4 mg equivalent dose has demonstrated a favorable benefit–risk profile in pediatric patients with moderate-to-severe AD (BREEZE-AD-PEDS, ClinicalTrials.gov identifier: NCT03952559) and has received regulatory authorization in Europe for the treatment of moderate-to-severe AD in children and adolescents (2 to <18 years old) [10, 11]. Baricitinib treatment displays linear pharmacokinetics with minimal accumulation with repeated dosing in healthy adult volunteers and patients with RA [12, 13].
While the pathophysiology of AD is similar in pediatric patients to that in adults and is likewise treated with topical (e.g., corticosteroids or calcineurin inhibitors) or systemic medicines, newer treatments are needed for patients who cannot tolerate available treatments or may not achieve good disease control.
Modeling-based methodology is frequently used in clinical trial development to predict the dose, dosing frequencies, and performance of investigational products. Optimal dose selection is crucial to identifying an ideal balance between efficacy and safety for a specific patient population [14]. Allometric scaling of adult pharmacokinetic (PK) parameters is useful to predict PK and determine dosage in pediatric populations [15, 16].
The objective of the current analyses was to use an existing PK model for baricitinib in adult patients to identify dosage in pediatric patients with AD that results in similar exposures to that seen in adult patients receiving baricitinib 4 mg QD, and characterize the exposure–response (E-R) relationship for efficacy in informing optimal dose recommendations in pediatric patients with AD.
Anzeige
2 Methods
2.1 Patients and Study Design
This multicenter, double-blind, randomized, placebo-controlled, phase III study enrolled pediatric patients (aged 2 to <18 years) with a prior diagnosis of moderate-to-severe AD (≥12 months before screening if aged ≥6 years or >6 months before screening if aged 2 to <6 years) with an inadequate response to moderate or higher potency topical corticosteroids (TCS) within the preceding 6 months and inadequate response or history of intolerance to topical calcineurin inhibitors (TCNI), or inadequate response to systemic treatments within the preceding 6 months [10]. Moderate-to-severe AD was based on a validated Investigator Global Assessment for Atopic Dermatitis (vIGA-AD® [17]) score of ≥3, Eczema Area and Severity Index (EASI) score ≥16 and body surface area involvement ≥10%. The trial consisted of an open-label lead-in period, in which the PK of baricitinib QD dosing in pediatric patients (i.e., 4 mg [high dose]: 10 to <18 years, and 2 mg [high dose]: 2 to <10 years) was assessed to confirm that these doses in pediatric patients produced an exposure similar to adult patients given baricitinib 4 mg QD. Enrolment for the open-label PK lead-in was staggered by age group (10 to <18 years, 6 to <10 years, and 2 to <6 years), with older groups enrolling before younger groups. Internal safety reviews as well as an external Data Monitoring Committee (DMC) review of safety and PK data occurred after enrolment of each PK lead-in age cohort, and younger age cohorts did not begin enrolling into the open-label lead-in until the internal safety review and DMC review were completed. At each review, the DMC did not identify safety concerns and determined that the study should continue without modifications.
Once the high doses were confirmed for each age group, patients in the respective age group were enrolled directly into a 16-week double-blind treatment period where they were randomized 1:1:1:1 to oral once-daily placebo, low-dose baricitinib (1-mg equivalent), medium-dose baricitinib (2-mg equivalent), or high-dose baricitinib (4-mg equivalent). Efficacy was assessed by the proportion of patients achieving a vIGA-AD® score of 0 or 1 (clear to almost clear skin) with ≥2-point improvement from baseline by week 16.
2.2 Sample Collection and Analysis
The current analyses correspond to blood and plasma samples collected within the open-label PK lead-in and double-blind treatment periods. Whole blood samples were collected from all participants during the open-label PK lead-in period using a microsampling device (Mitra®) [18]. Plasma concordance samples with time-matched whole blood samples were collected from a subset of older age group participants to calculate the blood/plasma (B/P) ratio, which was used to convert the whole blood data to plasma equivalents. During the open-label PK lead-in period, plasma-equivalent concentration–time data were collected from pediatric patients and compared with model-predicted 90% confidence intervals (CI) of adults to confirm that exposure levels were equivalent to those observed in adult patients receiving baricitinib 4 mg QD. Blood samples were collected on days 0, 4, and 11 at multiple time points (at predose, 0.25, 0.5, 1, 2, 4, and 6 hours postdose) during the open-label PK lead-in period (7 per patient), and at weeks 0, 4, 8, and 16 post-randomization (at predose, 0.25, 1, 2–4, and 4–6 h postdose) in the double-blind treatment period (5 per patient). All concentration data available from the open-label PK lead-in and double-blind treatment periods were included in the population PK (PopPK) analysis.
Plasma samples were analyzed for baricitinib using a validated liquid-liquid extraction method at LabCorp Bioanalytical Services, LLC located in Indianapolis, Indiana, USA. The lower limit of quantification (LLOQ) was 0.200 ng/mL and the upper limit of quantification (ULOQ) was 200.000 ng/mL. Dried whole blood samples obtained were analyzed for baricitinib using a validated impact-assisted extraction method at Altasciences Company Inc. (575 Armand-Frappier Blvd., Laval, QC, Canada H7V 4B3). The LLOQ was 0.20 ng/mL and the ULOQ was 200.00 ng/mL.
2.3 Pharmacokinetic (PK) Model Development and Exposure–Response (E-R) Relationship
A population PK model structure, previously developed for adult patients with RA and AD, was the basis for the pediatric PK analysis [13]. The base model was a linear 2-compartment model with zero-order absorption (including lag time) and a semi-mechanistic partitioning of apparent total body clearance (CL/F) into apparent renal clearance (CLr/F) dependent on eGFR (estimated with the Bedside Schwartz method for children) and apparent nonrenal clearance (CLnr/F); Fig. S1 (see electronic supplementary material [ESM]). Exponential between-subject variability (BSV) was included on CLr/F, CLnr/F, central (V1/F), and peripheral (V2/F) volumes of distribution. The absorption duration (D1) parameter included a Box-Cox-transformed BSV. Estimates for PK parameters from the adult AD PopPK analyses were adopted as priors for initiating the current PopPK model. An allometric relationship, which was not included in the original adult model, was used for the effect of baseline body weight on clearance-related parameters (CLr/F, CLnr/F, and intercompartmental clearance [Q]) with the allometric exponent fixed to 0.75, and fixed to 1 for the effect of weight on central and peripheral volume of distribution (V1/F and V2/F). Fixing the exponents is considered appropriate given the accumulated evidence in the literature [19]. The effect of baseline body weight CL/F and V/F are shown in Fig. S2 (see ESM).
After the appropriate base model was confirmed, a stepwise covariate modeling (SCM) assessment of continuous (age) and categorical covariate (gender and race) effects were conducted using Perl-speaks-NONMEM (non-linear mixed effects modeling) (PsN). Key covariate selection criteria for forward inclusion were a p-value no greater than 0.01 (Δ6.635 minimum objective function for inclusion of 1 parameter), and variance estimate for BSV on the relevant parameter had to decrease by ≥5% for the covariate to be retained in the full model. No covariates met this criterion, therefore backward exclusions were not conducted; hence, the base model was the final model. Missing values of independent variables (participant characteristic data) were imputed per participant, using the last observation carried forward method.
A prediction-corrected visual predictive check (VPC) was performed using PsN to ensure the model maintained fidelity with the observed data. A non-parametric bootstrap analysis was performed using PsN to determine the stability and robustness of the final model to produce reliable PK parameter estimates.
The study involved administering baricitinib doses of 1 -mg, 2 mg or 4 mg QD to patients aged 10 to <18 years, and 0.5 mg, 1 mg or 2 mg to patients aged 2 to <10 years. Since weight is a physiologically more relevant patient factor than age to PK, the final PK model included allometric scaling for clearance and volume. The effect of weight on peak concentration at steady state (Cmax,ss) and area under the plasma concentration–time curve at a dosing interval at steady state (AUC τ,ss) was further assessed to identify the optimal weight cut-off value for dosing. Simulations were conducted in the weight range between 10 and 120 kg, which encompasses the range of weights enrolled in the study. Additionally, 10 kg is the 5th percentile weight for a typical 2-year-old [20]. Simulations were conducted by fixing the range of GFR to the study range and varying weight by 10 kg between 10 and 60 kg and then by 20 kg between 60 and 120 kg. The simulated exposures were compared with adult exposures at 4 mg QD.
Anzeige
The E-R relationships were evaluated for the primary efficacy endpoint of vIGA-AD 0 or 1 at 16 weeks using the exposure quartile analysis. Specifically, model-estimated average concentration during a dosing interval at steady state (Cav,ss) values from all patients receiving baricitinib were grouped into four quartiles compared with placebo and the proportion of patients achieving vIGA-AD 0 or 1 at 16 weeks was summarized for each exposure quartile.
The primary dataset contained 2035 baricitinib concentrations from 393 participants from one phase III study in pediatric patients with AD. One patient was excluded from data presented in the final report due to an erroneous weight collected at screening; hence, only data from 392 patients were reported. The demographic and baseline characteristics of the study population are presented in Table S1 (see ESM). Among 392 participants, the majority (72%) of patients were between the ages of 10 and <18 years old, 78% weighed ≥30 kg, 50% were female, and 73% were White. The mean weight was 46.6 kg and mean baseline eGFR was 109 mL/min/1.73 m2.
3.2 PK Lead-in Period Analysis
A total of 15 plasma concordance samples were collected from 15 participants in the open-label PK lead-in period. The blood-to-plasma ratio determined as the slope of the regression line using time-matched blood and plasma samples was 1.32 and was used to convert the blood data to plasma equivalents (Fig. S7, see ESM). A total of 214 whole blood samples collected from 33 participants were converted to plasma equivalents. The plasma equivalents were used to perform the PK analyses for the open-label PK lead-in period.
Anzeige
A comparison of the individual observed plasma equivalent concentrations of baricitinib in pediatric participants with AD for all age groups receiving baricitinib high dose (2 or 4 mg QD) for approximately 2 weeks, and the mean (with 90% CI) and observed plasma concentration of baricitinib in adult participants with AD receiving baricitinib 4 mg QD (efficacious exposure level) are shown in Fig. 1. One individual maximum observed drug concentration (Cmax) value in pediatric participants in the age group of 10 to <18 years was slightly higher than the 90% prediction interval (dashed lines) but was within the range (grey symbols) of the adult concentrations at the 4-mg dose. In addition, plasma concentrations in pediatric participants decreased quickly with time after reaching Cmax and thereafter stayed within the 90% prediction interval of adult concentrations.
Fig. 1
Comparison of observed plasma concentrations of baricitinib in pediatric patients in all age groups with AD from the BREEZE-AD-PEDS study and plasma concentrations in adult patients with AD receiving baricitinib 4 mg once daily. AD atopic dermatitis, QD once daily. A Pediatric participants aged 10 to <18 years (N = 20) with AD receiving baricitinib 4 mg QD. B Pediatric participants aged 6 to <10 years (N = 7) with AD receiving baricitinib 2 mg QD. C Pediatric participants aged 2 to <6 years (N = 6) with AD receiving baricitinib 2 mg QD. Note: Black solid line and dashed lines are population pharmacokinetic model-estimated median with 90% prediction interval of concentrations, and gray circles are observed concentrations at 4 mg QD in adult patients with AD. Colored lines and symbols are for individual pediatric patients from BREEZE-AD-PEDS
The adult AD PK model, with the incorporation of allometric scaling for clearance- and volume-related parameters and a partition of total clearance into renal and non-renal clearance, adequately described the PK of baricitinib in pediatric patients with AD. The goodness-of-fit plots (Fig. S3 in the ESM) and visual predictive check (Fig. S4 in the ESM) demonstrated concordance between the model predictions and observed data. Key PK parameter estimates from PopPK modeling are included in Table 1. The bootstrap analysis showed all parameters were estimated with adequate precision. The total clearance estimate (for a weight of 74 kg) is 10.7 L/h, similar to that estimated for adult patients (11.2 L/h) [22]. The relationship between model-estimated CL/F and V/F and baseline body weight are shown in Fig. S2 (see ESM).
Table 1
Pharmacokinetic and covariate parameters in final population model for baricitinib in patients with AD
Model parameter
(unit)
Population mean
(%SEE)
Shrinkage
(SD)
BSVa (%SEE)
Mean (95% CI) from bootstrap analysis
D1 (h)
0.263 (8.86)
27.2
164 (13.9)
0.253 (0.232–0.295)
Box-Cox transformation parameter for D1
0.311 (6.85)
0.395 (0.167–0.481)
CLnr/F(L/h)b
2.76 (15.4)
5.84
58.4 (4.54)
2.68 (2.51–3.03)
CLr/F (L/h)b
7.9 (12.1)
1.00 E−10
62.3 (16.5)
8.02 (7.44–8.38)
V1/F (L)c
119 (4.12)
57.3
12.7 (19.9)
119 (116–122)
Q (L/h)d
2.4 (9.17)
88.5
15.1 (FIX)
2.52 (2.19–2.63)
V2/F (L)e
46.8 (14.7)
79.5
117 (44.3)
46.5 (34.3–60.5)
LAG (h)
0.144 (1.73)
0.146 (0.135–0.151)
Allometric scaling CLb
0.75 (FIX)
Allometric scaling Vc,e
1 (FIX)
Covariate for change in eGFR on CLr/Ff
0.00778 (42.5)
0.00738 (0.00462–0.0115)
Covariance for CLnr/F and CLr/Fg
0.303 (10.4)
0.295 (0.247–0.359)
Covariance for CLr/F and V1/Fg
−0.0265 (15.8)
−0.0278 (−0.0360 to −0.0182)
Proportional errorh
0.427 (13.9)
0.426 (0.411–0.441)
AD atopic dermatitis, BSV between-subject variability, CI confidence interval, CL total body clearance of drug calculated after intravenous administration, CLnr/F apparent non-renal clearance, CLr/F apparent renal clearance, CV coefficient of variation, D1 absorption duration, ∆eGFR change in eGFR from baseline, eGFR estimated glomerular filtration rate, EXP exponential, FIX fixed, LAG absorption lag, PopPK population pharmacokinetics, Q intercompartmental clearance, SD standard deviation, SEE standard error of estimate, SQRT square root, V volume of distribution, V1/F apparent central volume of distribution, V2/F apparent peripheral volume of distribution, WTE weight at entry
aBSV reported as %CV = (SQRT(EXP(OMEGA(N))-1))*100
bCL/F = (CLnr/F+CLr/F)*((WTE/74)^0.75)
cV1/F = 119*((WTE/74)^1.00)
dQ = 2.4*((WTE/74)^0.75)
eV2/F = 46.8*((WTE/74)^1.00)
feGFR was estimated using the Bedside Schwartz equation. CLr/F = (apparent renal clearance*((baseline eGFR/93) + 0.00778 * (∆eGFR)), where 93 is median eGFR in mL/min/1.73 m2 from the previous PopPK analysis
gCovariance ω2
hStandard deviation
A summary of post-hoc PK parameter estimates for patients by age and dose group compared with adult patients with AD are included in Table 2. As expected, CL/F and V/F decreased with a decrease in body weight and age. The estimated mean CL/F range between age and dose groups was 4.98–10.4 L/h in patients with AD, whereas in adult patients with AD the mean CL/F was 10.5 L/h (5th to 95th percentile: 4.37–16.4 L/h). Half-life estimates across groups are in general comparable to the adult value.
Table 2
Post-hoc population pharmacokinetic parameter estimates in pediatric patients with AD by age group, weight group, and dose compared with adult patients with AD
Parameters
Age groups
Weight groupsa
Adult ADb
Weight category (kg)
<30
≥30
Age (y)
2 to <6
6 to <10
10 to <18
6.4
(2.5–10.9)c
14.0
(10.0–17.9)c
33.0
(17–84)c
Weight (kg)c
17.4
(12.0–25.0)
28.5
(17.0–59.3)
55.0
(24.4–104)
21.5
(13.4–29.7)
56.9
(31.0–104)
74.5
(42.9–151)
Dose (mg)
0.5
1
2
0.5
1
2
1
2
4
2
4
4
N
9
8
15
24
26
29
87
86
108
35d
106d
819
Cmax,ss
(ng/mL)
18.9
(29)
35.1
(21)
64.8
(22)
11.6
(29)
23.1
(23)
45.7
(35)
13.2
(34)
27.8
(34)
50.7
(29)
57.1
(22)
50.3
(28)
45.0e
(20)
AUCτ,ss
(h×ng/mL)
94.3
(108)
200
(63)
298
(51)
74.8
(64)
155
(65)
279
(77)
109
(63)
222
(66)
383
(61)
298
(59)
383
(62)
380e
(41)
CL/Ff (L/h)
5.29
(108)
4.98
(63)
6.69
(51)
6.67
(64)
6.42
(65)
7.14
(77)
9.16
(63)
8.98
(66)
10.4
(61)
6.70
(59)
10.4
(62)
10.5
(41)
CLr (L/h)
4.18
(113)
3.95
(66)
5.25
(51)
5.26
(65)
5.04
(68)
5.62
(81)
6.90
(66)
6.78
(69)
7.85
(64)
5.27
(61)
7.85
(65)
7.50
(45)
CLnr (L/h)
1.09
(90)
1.02
(54)
1.43
(49)
1.39
(62)
1.37
(58)
1.50
(67)
2.22
(59)
2.16
(61)
2.52
(56)
1.42
(53)
2.52
(56)
2.95
(33)
V/Ff (L)
37.4
(23)
38.2
(15)
38.9
(16)
61.9
(23)
61.5
(24)
63.4
(31)
121
(29)
111
(30)
119
(29)
47.1
(25)
120
(28)
124
(16)
t½ (h)
12.6
(29)
11.1
(45)
10.7
(36)
14.2
(21)
14.1
(39)
13.2
(56)
17.9
(38)
16.4
(43)
16.0
(41)
11.7
(45)
16.1
(41)
12.3
(29)
Values are presented as geometric mean (CV%)
AD atopic dermatitis, AUCτ,ss area under the concentration versus time curve during 1 dosing interval at steady state, CL/F apparent total body clearance of drug calculated after extravascular administration, CLnr/F apparent non-renal clearance, CLr/F apparent renal clearance, Cmax,ss maximum observed drug concentration during a dosing interval at steady state, CV coefficient of variation, N number of participants, QD once daily, t½ elimination half-life, V/F apparent volume of distribution, V1/F apparent central volume of distribution, V2/F apparent peripheral volume of distribution
aIndividual post-hoc estimates based on data from the current analysis in pediatric patients with AD at that dose
bIndividual post-hoc estimates based on data combined from phase II and III studies in adult patients with AD
cMean (min–max)
dOnly patients that weighed <30 kg and given baricitinib 2 mg QD, or that weighed ≥30 kg and given 4 mg QD, are included in the data. Patients that received other doses are excluded
eValues were based on all patients in the respective analysis dataset and a dose normalized to 4 mg QD
fEstimates of CL/F and V/F were based on individual post-hoc estimates; CL/F = CLnr/F + CLr/F and V/F = V1/F + V2/F
The resulting exposure estimates for patients aged 2 to <10 years receiving baricitinib 2 mg QD and those aged 10 to <18 years receiving 4 mg QD ranged from 279 to 383 h*ng/mL for AUCτ,ss and 45.7 to 64.8 ng/mL for Cmax,ss. The exposure estimates were overall consistent with the exposure at 4 mg QD in adult patients with AD (AUCτ, ss = 380 ng*h/mL; Cmax,ss = 45.0 ng/mL).
Anzeige
3.4 Weight-Based Simulations
The simulated exposures were compared with adult exposures at 4 mg QD; Fig. 2. The data suggests that giving baricitinib 2 mg to patients 10 to <30 kg and giving 4 mg to patients ≥30 kg would provide the most comparable exposure to adult patients with AD who receive 4 mg QD. Plots of both the simulated and actual post-hoc AUC values at 4 mg QD for ≥30 kg and at 2 mg for 10 to <30 kg indicate comparable exposures for pediatric patients to those for adult patients with AD who are administered 4 mg QD; Fig. 3B. The values of Cmax,ss at 4 mg QD for ≥30 kg and at 2 mg for 10 to <30 kg are slightly higher than the adult value (12% and 27% higher, respectively, for pediatric patients ≥30 kg and <30 kg).
Fig. 2
Boxplots of simulated post-hoc ACmax,ss and B AUCτ,ss at 4 mg QD by weight ranges, in pediatric patients with AD compared with adult patients with AD. AD atopic dermatitis, AUCτ,ss area under the concentration-versus-time curve during 1 dosing interval at steady state, Cmax,ss maximum observed drug concentration during a dosing interval at steady state, PK pharmacokinetics, QD once daily. Note: Boxes for adult AD are based on PK modeling with data combined from three phase II and phase III studies in adult patients with AD
Boxplots of simulated post-hoc ACmax,ss and B AUCτ,ss at 2 mg QD for pediatric patients with AD <30 kg, and 4 mg QD for patients ≥30 kg, compared with adult patients with AD. Observed actual and dose-normalized post-hoc data overlaid. AD atopic dermatitis, AUCτ,ss area under the concentration versus time curve during a dosing interval at steady-state, Cmax,ss maximum observed drug concentration at steady-state, QD once daily. Box for adult AD is based on PK modeling with data combined from three phase II and phase III studies, in adult patients with AD. Grey circles = 4-mg dose normalized adult AD individual post-hoc estimates. Observed data dose normalized to 2 or 4 mg if the patient received a different dose. Black circles = 4 mg, blue triangles = 2 mg, green squares = 1 mg, and red diamonds = 0.5 mg
The post-hoc PK parameter estimates for participants enrolled in BREEZE-AD-PEDS by weight group with a weight cut-off of 30 kg are summarized in Table 2 and compared with those in adult patients with AD. Exposure estimates (AUC and Cmax) using weight-based dosing (4 mg QD for patients ≥30 kg and 2 mg QD for patients <30 kg) are more consistent with exposures at 4 mg QD in adult patients with AD than exposure estimates using age-based dosing.
3.5 Efficacy
Data on the proportion of patients achieving vIGA-AD 0 or 1 at 16 weeks from 514 patients were used for the exposure quartile analyses for this primary efficacy endpoint. A trend towards increasing response rate with higher exposures expressed as Cav,ss was observed at Week 16 for the efficacy endpoint; Fig. 4. Furthermore, when assessing the E-R relationship for vIGA-AD 0 or 1 response using data sorted by the proposed weight ranges (≥30 kg or more or <30 kg) the E-R relationship is similar to that of the overall population; Fig. 5. Prespecified baseline age group analyses (Fig. S5-A in the ESM) showed a similar dose–response relationship in the older (10 to <18 years) and younger (2 to <10 years) age subgroups with no statistical difference between age subgroups (treatment-by-age group interaction term was >0.1). Additional analyses by baseline weight subgroups (Fig. S5-B in the ESM) are consistent with age subgroup analyses and show a similar pattern of dose response with no statistical difference between weight subgroups (treatment-by-weight group interaction term was >0.1).
Fig. 4
Exposure-response analysis: observed vIGA-AD® 0 or 1 at Week 16 by baricitinib average plasma concentration at steady-state (Cav,ss) quartiles for patients receiving placebo or baricitinib 0.5, 1, 2, or 4 mg once daily in BREEZE-AD-PEDS. Cav,ss average concentration during a dosing interval at steady state, IGA Investigator’s Global Assessment, N number of subjects analyzed, NA not applicable, Q quartile. Note: Light blue bars indicate response rate
Exposure-response analysis by A patients <30 kg and B patients ≥30 kg bodyweight: observed IGA 0 or 1 at Week 16 by baricitinib average plasma concentration at steady-state (Cav,ss) quartiles for patients receiving placebo or baricitinib 0.5, 1, 2, or 4 mg once daily in BREEZE-AD-PEDS. Cav,ss average concentration during a dosing interval at steady state, IGA Investigator’s Global Assessment, N number of subjects analyzed, NA not applicable, Q quartile. Note: Light blue bars indicate response rate
The BREEZE-AD-PEDS study is the first baricitinib study conducted in pediatric patients with AD. Initial dose selection for pediatric patients was based on the exposure-matching approach which has been used for pediatric AD development for other JAK inhibitors [23]. An open-label lead-in period was incorporated in BREEZE-AD-PEDS and confirmed exposures comparable to adults at the high doses selected for pediatric patients (4 mg QD or 2 mg QD) in the various age/weight groups. The high doses (adult 4-mg QD equivalent) confirmed during the open-label lead-in period were implemented during the remainder of the phase III trial along with medium (2-mg equivalent) and low (1-mg equivalent) doses, and the baricitinib 4-mg equivalent dose was shown to be efficacious with an acceptable safety profile for the treatment of AD in the pediatric population [10]. The popPK analysis using data from all periods of the study adequately characterized the PK in this population and allowed for optimization of dosing with the appropriate weight cutoff values.
The previously developed PK model for baricitinib in adult patients with AD was a two-compartment model with zero-order absorption and a partitioning of total CL/F into CLr/F and CLnr/F [13]. The CLr/F term correlates with the eGFR estimated with modification of diet in renal disease and represents mainly renal filtration and secretion. While the CLnr/F term largely represents hepatic metabolism, it may also include some renal secretion [24]. The model structure for CL/F was set up in this manner based on the knowledge that renal excretion represents the primary elimination route for baricitinib [24]. The approach is similar to the typical covariate analysis with renal function as a covariate on total CL/F, but with the advantage of having numeric values of the renal and nonrenal component of CL/F readily estimated from the modeling.
Considering the elimination mechanism of baricitinib, and that only patients aged >2 years were enrolled in BREEZE-AD-PEDS, weight was a primary contributing factor to the differences in PK between the adult and pediatric populations. It was assumed a priori that body weight would have a significant effect on the clearance and volume-related PK parameters and these relationships were assumed to follow the typical allometric scaling. The adult PK model incorporating allometric scaling adequately described the PK data in this pediatric study. No additional covariates besides weight and renal function were identified, with both prospectively incorporated into the model structure. Specifically, age was not a significant covariate after the effect of weight was accounted for. VPCs demonstrated reasonable agreement between model-predicted and observed data, indicating accurate parameter estimation across age and weight groups over time. Accordingly, the PK of baricitinib in pediatric AD patients characterized in the current study is consistent with that for adult AD patients [22]; the total clearance, normalized for a bodyweight of 74 kg, was estimated to be 10.7 L/h and 11.2 L/h for pediatric (Table 1) and adult patients [22], respectively. The values are also similar to the total clearance value in pediatric juvenile idiopathic arthritis (JIA) patients (9.70 L/h) [24].
It should be noted that what appears to be some inconsistency in the model fitting likely resulted from a possible data issue due to incorrect recording of time from dose in the phase III study. As shown in the VPC plot (Fig. S4 in the ESM), there is a cluster of data points with high concentration values around 24 h post-dosing (with daily dosing) and a cluster of data points with low concentration values around 4 h post-dosing. These values are physiologically implausible based on known PK properties of baricitinib from multiple clinical studies. A similar phenomenon was observed in the baricitinib adult phase III studies for RA and AD and pediatric phase III studies for JIA. As the issue here lies in the data, rather than a mechanism related to drug disposition, it is not considered resolvable by improving the model structure. Investigation of the phenomenon by modeling a revised dataset from which the values in the clusters were excluded resulted in minimal differences in the estimation of model parameters. As it was infeasible to confirm that these values resulted from inaccurate recording of dosing time, and the PK parameter estimates were not significantly influenced by these values, these data points were retained in the final analysis.
In the allometric function, the exponents were fixed for CL-related (0.75 for CLr/F, CLnr/F, and Q) and volume-related (1 for V1/F and V2/F) parameters. A separate analysis with this PK model with the exponent estimated was also conducted. The exponent was estimated to be 0.59 for CL and 0.66 for V, both close to the fixed values. Fixing the exponents is considered appropriate given accumulated evidence in literature [19] and considering the relatively sparse data available for the younger age group in the clinical study. Additionally, a comparison of the estimated CLr/F and CLnr/F with the fixed-exponent and estimating-exponent approaches indicates that dosage for pediatric patients by various weight groups would not differ based on the approach of estimating the exponents.
The age-based dosing was converted to weight-based dosing given that weight is a more physiologically relevant patient factor affecting PK and was identified as a covariate on clearance and volume of distribution in the PopPK model [19]. Figure S6 demonstrates a high correlation between age and weight (see ESM). The effect of weight on the Cmax and AUC was further considered to identify an optimal cut-off weight value for dosing. In Fig. 3, the boxes for the AUC and Cmax estimates were generated using simulations for 5000 pediatric patients with AD in the various weight groups and adult patients with AD using the respective final pediatric and adult PK model. Individual post-hoc AUC estimates from pediatric and adult patients with AD were also overlaid for comparison.
As shown in Fig. 3B, the AUC data as summarized by the box plot exhibit a high overlap between pediatric patients with AD (2 mg for 10 to <30 kg and 4 mg for ≥30 kg) and adult patients with AD (4 mg). Notably, the median AUC for 2 mg in the 10 to <30-kg weight group is comparable to that of the adult. The Cmax data (Fig. 3A) are largely overlapping between pediatric patients ≥30 kg receiving 4 mg QD and adult patients with AD (4 mg). The Cmax at a dose of 2 mg QD for the 10 to <30-kg weight group is estimated to be higher than that for adults (approximately 24% higher at the mean level). Based on the exposure-efficacy and exposure-safety relationship from the large dataset of adults with RA, Cmax is not expected to have a substantial effect on efficacy and safety. Matching AUC is largely considered as the most important acceptance criterion versus Cmax due to its relevance to efficacy and safety of baricitinib based on the E-R relationship established in adults [13]. Additionally, pediatric patients received the proposed baricitinib doses (4 mg QD and 2 mg QD), and there were no new safety signals observed compared with adult AD studies.
A weight cut-off of 30 kg (approximately the median weight of 10-year-old pediatric patients) was selected based on the simulated relationship between baricitinib exposure and weight for those weighing 10 kg and above. For reference, 10 years of age was also used as the age cut-off for dose assignment in this trial. As expected, response analysis by baseline weight (Fig. S5-B, see ESM) was similar to the prespecified age subgroup response analysis (Fig. S5-A, see ESM) because the majority of patients 10 years old or more at baseline were included in the baseline weight category of 30 kg or more given that the fifth percentile of weight for a 10-year-old child is approximately 25 kg; and the majority of patients <10 years old at baseline were included in the <30-kg baseline weight category given that the 95th percentile of weight for a 7-year-old is approximately 30 kg and the 50th percentile of weight for a 10-year-old is approximately 32 kg [20].
The exposure quartile analyses showed a clear trend of increase in the proportion of patients achieving vIGA-AD® 0 or 1 at 16 weeks with increase in exposure. This is as expected since 4 mg QD, compared with 1 mg and 2 mg QD, is the optimal dose that showed positive benefit/risk profiles in the phase III clinical studies in adults and at which exposure matching in pediatric patients was targeted. Since no apparent plateau is shown in Fig. 4, from an efficacy perspective, it is possible that doses higher than the adult 4-mg equivalent may provide greater efficacy. However, the safety evaluation in pediatric patients also relies on the larger adult safety database that includes data up to 4 mg QD. Therefore, the high end of the dose range being explored for pediatric patients was limited to the highest dose studied in the adult phase III studies to ensure safety in pediatric patients.
Another attribute of this trial was microsampling with the use of the Mitra® VAMS® blood collection device in a small number of pediatric patients. The advantage of microsampling over traditional phlebotomy techniques allows for greater access to PK data in the pediatric population and has been used in various pediatric clinical trials [18].
5 Conclusion
We show an example of successful dose selection strategies for pediatric AD patients by applying an established adult population PK model with the addition of allometric scaling for weight on clearance and volume terms. The work suggested that exposure matching is a valid approach for the JAK inhibitor baricitinib in pediatric AD patients based on the observed PK, efficacy, and safety data in the clinical study. Furthermore, the PK model for pediatric patients with AD allowed the simulation by weight ranges to identify an appropriate weight grouping. The final dosage was established as 2 mg QD for patients 10 to <30 kg and 4 mg QD for patients ≥30 kg.
Acknowledgments
The authors thank the participants, caregivers, and investigators for their participation in this study. This study was sponsored/funded/supported by Eli Lilly and Company. Eric A. Rodriguez, Ph.D., and Rebecca C. Anderson, Ph.D., of Eli Lilly and Company, provided writing and editorial assistance.
Declarations
Funding
This study was funded by Eli Lilly and Company under license from Incyte.
Conflicts of Interest
D.B. Radtke, A. Prakash, and X. Zhang are employees and shareholders of Eli Lilly and Company. R.L. Decker is a former employee of Eli Lilly and Company. C.S. Ernest II is a former employee of Eli Lilly and Company and currently employed with Metrum Research Group.
Availability of Data and Material
Lilly provides access to all individual participant data collected during the trial, after anonymization, with the exception of pharmacokinetic or genetic data. Data are available to request 6 months after the indication studied has received first regulatory authorization and after primary publication acceptance, whichever is later. No expiration date of data requests is currently set once data are made available. Access is provided after a proposal has been approved by an independent review committee identified for this purpose and after receipt of a signed data sharing agreement. Data and documents, including the study protocol, statistical analysis plan, clinical study report, and blank or annotated case report forms, will be provided in a secure data sharing environment. For details on submitting a request, see the instructions provided at www.vivli.org.
Ethics Approval
The protocol was approved by institutional review boards or ethics committees from all participating sites according to local requirements and was conducted according to International Conference on Harmonization Good Clinical Practice guidelines and the Declaration of Helsinki.
Consent to Participate
Informed consent, or assent when appropriate, was provided for participation in the study and prior to any study-specific procedures.
Consent for Publication
Not applicable.
Code Availability
Not applicable.
Author Contributions
R.L.D., C.S.E., A.P., and X.Z. wrote the manuscript; R.L.D., C.S.E., A.P., and X.Z. designed the research; R.L.D., C.S.E., D.B.R., and A.P. performed the research; R.L.D., C.S.E., D.B.R., and X.Z. analyzed the data.
Open Access This article is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License, which permits any non-commercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc/4.0/.
Urban K, Chu S, Giesey RL, et al. The global, regional, and national burden of atopic dermatitis in 195 countries and territories: an ecological study from the Global Burden of Disease Study 2017. JAAD Int. 2021;2:12–8.CrossRefPubMed
2.
Fridman JS, Scherle PA, Collins R, et al. Selective inhibition of JAK1 and JAK2 is efficacious in rodent models of arthritis: preclinical characterization of INCB028050. J Immunol. 2010;184:5298–307.CrossRefPubMed
3.
Huang IH, Chung WH, Wu PC, Chen CB. Jak-STAT signaling pathway in the pathogenesis of atopic dermatitis: an updated review. Front Immunol. 2022;13:1068260.CrossRefPubMedPubMedCentral
Reich K, DeLozier AM, Nunes FP, et al. Baricitinib improves symptoms in patients with moderate-to-severe atopic dermatitis and inadequate response to topical corticosteroids: patient-reported outcomes from two randomized monotherapy phase III trials. J Dermatolog Treat. 2022;33:1521–30.CrossRefPubMed
6.
Reich K, Kabashima K, Peris K, et al. Efficacy and safety of baricitinib combined with topical corticosteroids for treatment of moderate to severe atopic dermatitis: a randomized clinical trial. JAMA Dermatol. 2020;156:1333–43.CrossRefPubMed
7.
Silverberg JI, Simpson EL, Wollenberg A, et al. Long-term efficacy of baricitinib in adults with moderate to severe atopic dermatitis who were treatment responders or partial responders: an extension study of 2 randomized clinical trials. JAMA Dermatol. 2021;157:691–9.CrossRefPubMed
8.
Bieber T, Reich K, Paul C, et al. Efficacy and safety of baricitinib in combination with topical corticosteroids in patients with moderate-to-severe atopic dermatitis with inadequate response, intolerance or contraindication to ciclosporin: results from a randomized, placebo-controlled, phase III clinical trial (BREEZE-AD4). Br J Dermatol. 2022;187:338–52.CrossRefPubMed
9.
Simpson EL, Forman S, Silverberg JI, et al. Baricitinib in patients with moderate-to-severe atopic dermatitis: results from a randomized monotherapy phase 3 trial in the United States and Canada (BREEZE-AD5). J Am Acad Dermatol. 2021;85:62–70.CrossRefPubMed
10.
Torrelo A, Rewerska B, Galimberti M, et al. Efficacy and safety of baricitinib in combination with topical corticosteroids in paediatric patients with moderate-to-severe atopic dermatitis with an inadequate response to topical corticosteroids: results from a phase III, randomized, double-blind, placebo-controlled study (BREEZE-AD PEDS). Br J Dermatol. 2023;189:23–32.CrossRefPubMed
Shi JG, Chen X, Lee F, et al. The pharmacokinetics, pharmacodynamics, and safety of baricitinib, an oral JAK 1/2 inhibitor, in healthy volunteers. J Clin Pharmacol. 2014;54:1354–61.CrossRefPubMed
13.
Zhang X, Chua L, Ernest C 2nd, et al. Dose/Exposure-Response Modeling to Support Dosing Recommendation for Phase III Development of Baricitinib in Patients with Rheumatoid Arthritis. CPT Pharmacometrics Syst Pharmacol. 2017;6:804–13.CrossRefPubMedPubMedCentral
14.
Love SB, Brown S, Weir CJ, et al. Embracing model-based designs for dose-finding trials. Br J Cancer. 2017;117:332–9.CrossRefPubMedPubMedCentral
15.
Wu Q, Peters SA. A retrospective evaluation of allometry, population pharmacokinetics, and physiologically-based pharmacokinetics for pediatric dosing using clearance as a surrogate. CPT Pharmacometrics Syst Pharmacol. 2019;8:220–9.CrossRefPubMedPubMedCentral
16.
Mahmood I, Tegenge MA. A comparative study between allometric scaling and physiologically based pharmacokinetic modeling for the prediction of drug clearance from neonates to adolescents. J Clin Pharmacol. 2019;59:189–97.CrossRefPubMed
17.
Simpson E, Bissonnette R, Eichenfield LF, et al. The validated investigator global assessment for atopic dermatitis (vIGA-AD): the development and reliability testing of a novel clinical outcome measurement instrument for the severity of atopic dermatitis. J Am Acad Dermatol. 2020;83:839–46.CrossRefPubMed
18.
Wickremsinhe ER, Decker RL, Lee LB, et al. Microsampling in pediatric studies: pharmacokinetic sampling for baricitinib (Olumiant™) in global pediatric studies. Bioanalysis. 2023;15:621–36.CrossRefPubMed
19.
Holford N, Heo YA, Anderson B. A pharmacokinetic standard for babies and adults. J Pharm Sci. 2013;102:2941–52.CrossRefPubMed
Purohit VS, Ports WC, Wang C, Riley S. Systemic tofacitinib concentrations in adult patients with atopic dermatitis treated with 2% tofacitinib ointment and application to pediatric study planning. J Clin Pharmacol. 2019;59:811–20.CrossRefPubMed
24.
Decker RL, Steven Ernest C 2nd, Radtke DB, et al. A population pharmacokinetic model using allometric scaling for baricitinib in patients with juvenile idiopathic arthritis. CPT Pharmacometrics Syst Pharmacol. 2024;13:970–81.CrossRefPubMedPubMedCentral
