Takayuki Katsube (1), Toru Ishibashi (1), Takeshi Kano (2), Toshihiro Wajima (1)
(1) Clinical Research Department, Shionogi & Co., Ltd., Japan, (2) Global Project Management, Shionogi & Co., Ltd., Japan
Objectives: Lusutrombopag (also known as S-888711) is a small-molecule orally active thrombopoietin (TPO) receptor agonist. Lusutrombopag acts on human TPO receptors expressing in megakaryocytes and leads to thrombocytopoiesis. Lusutrombopag may have potency for the treatment of thrombocytopenia. Phase 1 studies in Japan and US were conducted in the development program. The aim of this study is to develop a population pharmacokinetic (PK)/pharmacodynamic (PD) model to describe plasma lusutrombopag concentrations and platelet counts following oral lusutrombopag doses and also to explore PK/PD covariates.
Methods: A 3-compartment model with first-order rate and lag time for absorption was used as a PK model. A PD model was developed based on semi-physiological model with production, maturation and elimination processes of blood platelet [1]. One platelet compartment with 3 transit compartments, a sigmoid Emax model and a feedback for production was used for describing platelet change with drug effect. A total of 2539 plasma concentration data and 1408 platelet count data in 78 healthy adult subjects following single or multiple (once daily for 14 days) doses of lusutrombopag were applied to the PK/PD model using non-linear mixed effects model approach by NONMEM [2]. PK and PD covariates were selected by screening using univariate regression to construct a full model, followed by backward deletion from the full model.
Results: The population PK/PD model well described time courses for plasma lusutrombopag concentrations and platelet counts. The model indicated good prediction and a lack of bias in goodness-of-fit plots and visual predictive check [3]. Body weight (WT) and ethnicity (Japanese or non-Japanese subjects) were significant as PK potential covariates. The empirical allometric relations for the effects of WT on clearance and volume of distribution were applied since the effect of ethnicity may be confounded with the effect of WT. WT was suggested to be the most influential PK covariate when assuming the allometric relations, while the ethnicity was not clinically significant. Any significant PD covariates were not found.
Conclusions: The difference of body weight between populations was suggested to result in differences in lusutrombopag PK profiles between the populations. The PD sensitivity for platelet response with the lusutrombopag effect was suggested to be similar between Japanese and non-Japanese subjects.
References:
[1] Hayes S, Ouellet D, Zhang J, Wure MB, Gibiansky E. Population PK/PD modeling of eltrombopag in healthy volunteers and patients with immune thrombocytopenic purpura and optimization of response-guided dosing. J Clin Pharmacol. 2011. 51: 1403-1417.
[2] Beal SL, Sheiner LB, Boeckmann AJ. NONMEM users guide. Icon Development Solutions, Ellicott City, MD. 1989-2006.
[3] Holford NH. The Visual Predictive Check – Superiority to Standard Diagnostic (Rorschach) Plots. PAGE 14. 2005. Abstr 738. URL: www.page-meeting.org/?abstract=738.
Reference: PAGE 23 (2014) Abstr 3059 [www.page-meeting.org/?abstract=3059]
Poster: Drug/Disease modeling - Other topics