(W-065) Extrapolation of Efficacy and Simplification and Optimization of Pediatric Clinical Trial Design for Azilsartan Medoxomil (Edarbi) through Modeling and Simulation

Ivy Song, Ph.D. – Executive Director, Takeda Development Center Americas Inc.; Taisuke Kondo, MD, Ph.D. – Medical Director, Takeda Pharmaceutical Company Ltd; David Sequeira, Ph.D. – Vice President, Azurity Pharmaceuticals; Axel Facius, Ph.D. – Executive Director, ThinkQ2; Gezim Lahu, Ph.D. – Executive Director, ThinkQ2

Objectives: This study elucidates the justification for efficacy extrapolation from adults to the pediatric population and simplifies the design of clinical trials for Azilsartan Medoxomil (AZM), an angiotensin II type-1 receptor blocker (ARB), in children aged 2 to less than 6 years. The application of modeling and simulation was critical in shaping the study design, including dose selection, the number of pharmacokinetic (PK) subjects required for PK evaluation, and the determination of optimal PK sampling times.

Methods: A population pharmacokinetic (popPK) model and a pharmacokinetic/pharmacodynamic (PK/PD) model were developed using a nonlinear mixed-effect modeling approach (NONMEM 7.4.1), incorporating data from previous trials in adults and older children treated with AZM. To assess the PK/PD relationship across different age groups, we conducted an extensive literature review of ARBs and performed a PK/PD analysis to confirm the consistency of the exposure-response relationship for AZM. The established popPK model was employed to predict the PK exposure of azilsartan in young pediatric patients, aiming to match the exposure seen in adults and older children. The model also guided the determination of the optimal number of pediatric PK subjects and PK sampling times required for PK evaluation^[1]. Simulations using the PK/PD model were additionally performed to support the dose selection and predict blood pressure effects in the trial of the target age group aged 2 to less than 6 years.

Results: The PK/PD analysis revealed a comparable PK/PD relationship between adults and older children, which supports the extrapolation of efficacy to the pediatric population for AZM. Literature data on other ARBs further corroborated this finding and affirmed the feasibility of extending the efficacy extrapolation strategy to the young pediatrics^[2,3]. The PopPK simulations suggested pediatric doses (0.34, 0.68, and 1.36 mg/kg) that would achieve exposure similar to adult reference doses (20, 40, and 80 mg). The PK/PD model simulations also supported the selection of the proposed pediatric doses by showing that the anticipated mean blood pressure reduction was consistent with outcomes from adult and adolescent studies. The PopPK simulations recommended obtaining a minimum of four PK samples within specific windows from at least 10 pediatric subjects.

Conclusions: The study establishes the extrapolatability of efficacy for AZM through PK/PD modeling of AZM and literature review. The strategic use of modeling and simulation was pivotal in optimizing the clinical trial design for the pediatric AZM study, ensuring an efficient and effective approach to dose selection, the number of PK subjects required for PK evaluation, and PK sampling strategies.

Citations: [1]Wang et al. J Clin Pharmacol 2012; 52: 1601-1606
[2]FDA Candesartan Cilexetil SBA of Clinical Pharmacology Review NDA 20-838/s031
[3]FDA Olmesartan Medoxomil SBA of Clinical Pharmacology Review NDA 21-286