Felix Hammann, Claudia Suenderhauf, Matthias E. Liechti
Division of Clinical Pharmacology and Toxicology, Department of Clinical Research and Department of Internal Medicine, University Hospital Basel, University of Basel, Basel, Switzerland
Objectives: The semi-synthetic opioid analgesic oxycodone has in some trials exhibited flip-flop pharmacokinetics when administered as an immediate-release formulation [1]. We noticed the same phenomenon in an analysis of a trial of oral oxycodone solution (0.1 and 0.2 mg/kg) in patients 5-16 years old, and explored the underlying reasons using a whole-body physiological pharmacokinetic (PBPK) modeling approach.
Methods: The pharmacokinetic data are from a double-blind, randomized, dose-ranging study in pediatric patients 5 to 16 years old, hospitalized for medical and/or surgical conditions, and receiving morphine as a standard supplemental pain medication to evaluate pharmacokinetics, efficacy and safety of oxycodone 1 mg/mL solution versus placebo for pain. Patients were stratified into two age groups (5 to < 12 years, and 12 to 16 years) and randomized to receive either oxycodone 0.1 mg/kg, 0.2 mg/kg, or placebo every 6 hours for 18 to 24 hours according to a randomization ratio of 3:3:2. A total of 46 patients were enrolled in the active treatment group, with 1-8 samples taken per patient.
Population pharmacokinetic analysis was carried out using NONMEM (Version 7.4.1; Icon Development Solutions, http://www.iconplc.com, Ellicott City, MD, USA). The first order conditional estimation with interaction (FOCE-I) was used throughout all runs. PK-Sim and MoBi (version 7.0; Open Systems Pharmacology Suite Community, http://www.systems-biology.com/products/pk-sim.html) was used for PBPK modeling and simulation.
Results: The final population pharmacokinetic model is a one-compartment model with first-order absorption and elimination. Total body weight was included as a covariate on Vd and Cl. We estimated the elimination rate constant (kel) at 0.55 h-1, and the absorption rate constant (ka) at 0.14 h-1. As kel > ka, we noted the presence of flip-flop pharmacokinetics, i.e. elimination is driven by absorption. Possible reasons for this may be opioid effects on gastrointestinal motility (either from oxycodone or concomitant morphine treatment), and underlying medical and surgical conditions.
To see whether drug- or disease-induced changes in oral absorption can be held accountable for this, we created a PBPK model of oxycodone in adult patients using a middle-out approach from a clinical trial of oral oxycodone solution (0.2 mg/kg) [2], and published data on the physico-chemical and absorption / distribution / metabolism / elimination properties of oxycodone. The model faithfully depicted the historical adult data. We then proceeded to scale the PBPK model to the population of this trial, and noted absorption was predicted to be much faster than observed. Particularly, maximum concentrations (Cmax) and areas under the curve from 0-12h (AUC0-12h) were over-predicted, and time to maximum concentration (Tmax) was underpredicted.
By changing the release model of oxycodone to conform to a Weibull function parameterized from visual inspection, we were able to create a modified PBPK model that matches all three relevant secondary pharmacokinetic parameters. The Weibull function has been previously established as an alternative for modeling time-dependent first-order release [3-5]. A similar approach was taken by Li et al. who used semi-mechanistical modeling using a Weibull function for oral absorption of oxycodone [1].
Conclusions: Oxycodone oral solution (0.1 and 0.2 mg/kg, respectively) exhibited flip-flop pharmacokinetics in pediatric patients aged 5-16 years. This may be due to slowed gastrointestinal transit because of opioid effects and/or underlying medical or surgical conditions. This phenomenon can be modeled with a middle-out approach PBPK model when slowed oral absorption is accounted for with a Weibull function. Future studies of oral oxycodone should be wary of flip-flop behavior when defining sampling points.
References:
[1] Li, Y., et al., Slow drug delivery decreased total body clearance and altered bioavailability of immediate- and controlled-release oxycodone formulations. Pharmacol Res Perspect, 2016. 4(1): p. e00210.
[2] Samer, C.F., et al., The effects of CYP2D6 and CYP3A activities on the pharmacokinetics of immediate release oxycodone. Br J Pharmacol, 2010. 160(4): p. 907-918.
[3] Kosmidis, K., et al., A reappraisal of drug release laws using Monte Carlo simulations: the prevalence of the Weibull function. Pharm Res, 2003. 20(7): p. 988-95.
[4] Macheras, P., Dokoumetzidis A., On the heterogeneity of drug dissolution and release. Pharm Res, 2000. 17(2): p. 108-12.
[5] Papadopoulou, V., et al., On the use of the Weibull function for the discernment of drug release mechanisms. Int J Pharm, 2006. 309(1-2): p. 44-50.
Reference: PAGE 27 (2018) Abstr 8426 [www.page-meeting.org/?abstract=8426]
Poster: Drug/Disease Modelling - Absorption & PBPK