Lumenfantrine Exposure in Malnourished Children: PBPK modeling applied for Predictions and Dose Adjustments
Erik Sjögren (1), Joel Tarning (2,3) and E. Niclas Jonsson (1)
(1) Pharmetheus, Sweden (2) Mahidol-Oxford Tropical Medicine Research Unit, Mahidol University, Bangkok, Thailand; (3) Centre for Tropical Medicine and Global Health, Nuffield Department of Medicine, University of Oxford, Oxford, UK
Introduction: Artemether-lumefantrine combination therapy accounted for 73 % of artemisinin-based combination therapies (ACT) for uncomplicated Plasmodium falciparum malaria in 2013 [1]. In this ACT, the early parasitological response is largely determined by artemether while lumefantrine, which is the slowly eliminated, is used to prevent recrudescence from remaining parasites [2]. Even though the precise pharmacokinetic (PK) determinants of treatment outcome in uncomplicated malaria remain uncertain the area under the blood or plasma concentration-time curve (AUC) and the concentration on day 7 of the slowly eliminated ACT components are considered important predictors [3]. Lumefantrine disposition is characterized by CYP3A4 metabolism, high plasma protein binding, a large volume of distribution and long half-life (>30 hr). Lumefantrine bioavailability is low and variable which can be attributed to first pass metabolism but foremost to solubility limited absorption [4]. Total lumefantrine solubility is considerably increased by colloidal structures and food constituents resulting in significant clinical food effects (16-fold increase in AUC) [4]. Young children (<5 years of age) are especially vulnerable to malaria [5]. Moreover, lower than therapeutic target day 7 lumefantrine concentrations and exposure were recently reported for malnourished children potentially decreasing the therapeutic effect in this already vulnerable population [3,6]. A physiologically based pharmacokinetic (PBPK) approach for prediction of drug exposure in malnourished children enabling physiologically based translation from healthy adults to a malnourished pediatric population has previously been presented [7]. This was accomplished by combining information on a) the differences in body composition between healthy and malnourished adults and b) the differences in physiology between healthy adults and children.
Objectives: To develop a whole body PBPK model describing disposition of lumefantrine for predictions and assessments of drug exposure in malnourished children including retrospective risk-assessments and suggestion of adjustments to the dosing.
Methods: A PBPK model for lumefantrine was developed in PK-Sim® (v8) adopting a middle out modelling approach [4, 5, 8] . Intra luminal solubility in fasted and fed state were optimized towards clinical observation in adults at respective condition. Confirmation of CYP3A4 mediated elimination was performed by comparing predictions of concomitant administration of ketoconazole towards clinical observations. Evaluation of the model was performed towards observations in a sub-Sahara population [6]. Healthy pediatric virtual populations were generated using the population algorithm in PK-Sim® including maturation of biological systems, e.g., metabolic enzymes and plasma proteins [8]. Malnourished pediatric populations were then created, for different levels of protein energy malnutrition, by applying a set of physiological scaling parameters that accommodates for changes to body composition and plasma protein concentrations at the different nutritional states [7]. Performance of the lumefantrine model and the modelling approach was performed towards observations by visual predictive checks and PK parameters fold errors.
Results: The established lumefantrine model accurately described concentration-time profiles and exposure in healthy adults including prandial state dependencies and CYP3A4 interactions. Furthermore, the model was also able to describe PK in healthy adult and pediatric sub-Saharan populations. Nutritional indicators (WFA, HFA, WFH, BMIFA) for generated virtual malnourished pediatric populations agreed well with the study population, indicating the appropriateness of the physiological scaling approach. For the most severely malnourished children a 26% decrease in AUC was predicted which was in line with observation (19%). To reach equivalent exposure applying standard dosing schedule, a 7-fold higher dose was predicted due to solubility limited bioavailability. However, therapeutic targets could also be achieved, according to the model, by adding one additional day of administration to the dosing regimen.
Conclusions: The PBPK approach was able to describe and predict lumefantrine PK in malnourished children. Results demonstrate that additional dosing occasions is needed to the traditional three-days dosing regimen to reach therapeutic targets exposure in malnourished children.
References:
[1] World Health Organization (WHO). Guidelines for the treatment of malaria. 3rd ed. Geneva: WHO; 2015. http://www.who.int/malaria/publications/atoz/9789241549127/en/
[2] World Health Organization (WHO). World malaria report 2014. Geneva:WHO; 2014. http://www.who.int/malaria/publications/world_malaria_report_2014/report/en/
[3] Barnes KI et al. Malar J. 2007;6:122.
[4] Lefèvre G et al. Clin Drug Invest (1999) Dec 18(6): 467-480
[5] World Health Organization (WHO) World Malaria Report 2018
[6] Chotsiri et al. Clin Pharmacol Ther. (2019) Dec;106(6):1299-1309
[7] Sjögren et al. Pharmaceutics. (2021) Feb 2;13(2):204
[8] https://github.com/Open-Systems-Pharmacology
