Tamara van Donge

Simplification of methadone dosing in preterm neonates 79 5 Methods Characterization of pharmacokinetic data Data on methadone concentrations used for this analysis originates from a single center open-label prospective PK study conducted at Shahid Akbarabadi Hospital in Tehran, Iran. The study was approved by the local ethics committee. After informed consent was obtained, one dose of standard opioid medication (fentanyl or morphine), which was prescribed for clinical reasons and is not expected to modify the PK of methadone, was replaced by one dose of 0.1 mg∙kg -1 orally administered methadone. The study aimed at a sample size of 60 neonates with a postnatal age (PNA) less than 3 months. Patient demographics such as bodyweight (BW), gestational age (GA), PNA, sex and maternal age at birth were included in the analysis. Plasma (R)- and (S)-methadone levels were quantified in plasma samples (200 µL) after administration of methadone and concentrations were analyzed by high-performance liquid chromatography coupled with mass spectrometry (HPLC-MS/MS). Population pharmacokinetic model Population PK parameter estimates were obtained using nonlinear mixed effects modeling software (NONMEM v7.4.1; ICON Development Solutions, Ellicott City, MD). 23 A first-order estimation method with interaction (FOCE-I) was applied and the subroutine ADVAN13 TOL3 was used to solve the ordinary differential equations. We applied a 3-step approach to identify the model that best describes available methadone data. First, the structural population PK model was developed, second various covariate models were tested, and third the population PK model was evaluated. Numerical and graphical diagnostic information were created using R (version 3.4.4). 24 Plasma concentration measurements, which were below the lower limit of quantification (LLOQ; 0.25 ng∙mL -1 ) were excluded from analysis. Methadone can be transferred to the newborn (via breastmilk or via placental transfer) during pregnancy. As no information was available if the pregnant women were under methadone maintenance treatment, all pre-dose measurements were excluded from analysis. Methadone was only administered orally and therefore estimated parameters represent apparent values. Since methadone is a racemic mixture of two enantiomers (50:50) and both these enantiomers display different pharmacokinetic properties, they were modelled separately. Development of structural population pharmacokinetic model Both 1- and 2- compartment models were assessed as structural model including appropriate absorption and elimination processes for methadone. 25-27 Estimated parameters included absorption rate constant (k a ), volume of distribution (V d ) and clearance (CL). Log-normally distributed inter-individual