Christopher Walsh (1), Jennifer Bonner (1), Alan Boddy (2), Trevor Johnson (3), Sibylle Neuhoff (3), Gareth Veal (1)
(1) Northern Institute for Cancer Research, Newcastle University, UK; (2) Faculty of Pharmacy, University of Sydney, Australia; (3) SimCyp Limited, UK.
Purpose: To use physiologically-based pharmacokinetic (PBPK) modelling and simulation in the prediction of the pharmacokinetics of actinomycin D in children, especially those less than one year of age who are at most risk of toxicity from the drug. Actinomycin D is an antibiotic used in the treatment of Wilms tumour and rhabdomyosarcoma in children. Despite being in use for over 40 years, very few studies have been conducted to characterise its pharmacokinetics. In particular the paucity of data available for very young children makes it hard to develop a sound rationale for dose selection in these patients.
Methods: The project is being carried out using the PBPK modelling and simulation software Simcyp version 13.1 (Simcyp Ltd., Sheffield, UK). Log P, pKa, blood to plasma ratio, cell monolayer permeability and p-glycoprotein transport data were determined in house and combined with data sourced from the available literature, including information on renal and biliary clearance, to generate an actinomycin D compound file. For the simulation of clinical trial data, information was taken from previous work by Hill et al. (2014)1 who studied 117 patients 2 actinomycin D intravenously. The population was split into three age categories: 1-6, 6-10 and 10-20 years. These age bins were further subdivided based on the dose received. Simulations were run for 26 hours following an IV bolus dose (relative to body surface area) over 3 minutes. A subset of the 10-20 year age group receiving 1.24 mg/kg actinomycin D was used to develop the model which was then tested against other age groups and adult data.
Results and conclusions: Simulations using p-glycoprotein transport information resulted in an over-prediction of drug clearance, potentially suggesting the involvement of additional transporters which were not accounted for in this model. The observed biliary and renal clearance values were not sufficient to account for total body clearance of actinomycin D and so an additional systemic clearance value was added based on fitting to observed data. Preliminary visual checks suggest a reasonable fit of the model to observed data, although the model fails to fully capture the plasma concentration variability seen at some early time points. The mean AUC0-26 of simulated subjects is within 1.5-fold of the observed AUC0-26 (84 ng/mL.h simulated vs. 93 ng/mL.h observed).
References:
[1] Hill CR, Cole M, Errington J, Malik G, Boddy AV, Veal GJ. Characterisation of the clinical pharmacokinetics of actinomycin D and the influence of ABCB1 pharmacogenetic variation on actinomycin D disposition in children with cancer. Clinical Pharmacokinetics 2014 53: 741-751
Reference: PAGE 24 (2015) Abstr 3330 [www.page-meeting.org/?abstract=3330]
Poster: Drug/Disease modeling - Absorption & PBPK