Y Yamamoto (1), PAJ Valitalo (1), M Danhof (1), DJ van den Berg (1), R Hartman (1), W van den Brink (1), E Wong (1), D Huntjens (2), JH Proost (3), A Vermeulen (2), V Aranzana-Climent (4), C Dahyot-Fizelier (5), S Marchand (4), W Couet (4), U Rohlwink (6), R Mathôt (7), E Wildschut (8), N Ketharanathan (8), D Tibboel (8), A Figaij (6), JGC van Hasselt (1), ECM de Lange (1)
(1) Division of Pharmacology, LACDR, Leiden University, the Netherlands, (2) Quantitative Sciences, Janssen Research & Development, Belgium, (3) Division of Pharmacokinetics, Toxicology and Targeting, University of Groningen, the Netherlands, (4) Department of Medicine and Pharmacy, University of Poitiers, France, (5) Department of Anaesthesiology and Intensive Care Medicine, University Hospital Center of Poitiers, France, (6) Division of Neurosurgery, School of Child and Adolescent Health, Red Cross Children’s Hospital, University of Cape Town, South Africa, (7) Department of Clinical Pharmacy, Faculty of Medicine, University of Amsterdam, the Netherlands, (8) Department of Pediatrics, Division of Intensive Care and Pediatric Surgery, Erasmus MC – Sophia Children’s Hospital, the Netherlands
Objectives: To develop a generic brain distribution model using rat multilevel brain and plasma data and to translate the model to predict drug target site concentrations in human brain.
Methods: Densely sampled concentration-time profiles after administration of 9 compounds (acetaminophen, atenolol, methotrexate, morphine, paliperidone, phenytoin, quinidine, remoxipride and risperidone) to rats were collected for plasma, brain extracellular fluid (ECF), cerebrospinal fluid (CSF) from lateral ventricle (CSFLV) and cisterna magna (CSFCM) using microdialysis sampling. The brain distribution model structure was adapted from a previously published model [1]. A naïve pooling approach was used to fit the rat pharmacokinetic (PK) profiles. Subsequently, brain-distribution parameters were scaled to predict of human ECF and CSF data. Clinical data was available for: 1) acetaminophen plasma and CSFLV concentrations obtained using external ventricular drainage in patients with traumatic brain injury (TBI), 2) acetaminophen plasma and CSF subarachnoid space (CSFSAS) concentrations from patients with nerve-root compression pain, and 3) morphine plasma and ECF concentrations obtained using microdialysis from pediatric patients with TBI.
Results: The model described the PK profiles for the 9 compounds with different physicochemical properties by estimating only two parameters (clearance for drug transport at the blood-brain barrier (CLPL_ECF) and brain-CSF diffusion (QDIFF)). Parameters could be estimated with reasonable precision (relative standard error < 25%). CLPL_ECF was different for each drug whereas QDIFF was similar for the compounds (mean±SD : 0.027 ±0.019 mL/min). The model predicted human acetaminophen CSFLV and CSFSAS concentrations well. The model under-predicted morphine ECF concentrations that were obtained from a patient with diffuse brain injury (normalized root-mean-square deviation (NRMSD) 48%), whereas it predicted morphine ECF concentrations adequately when the ECF data was taken from the “healthy” brain side in focally injured brain (NRMSD 25%).
Conclusions: A generic model structure was developed to capture the PK across key areas of the brain and CSF. Moreover, our model generally allowed for adequate prediction of human acetaminophen and morphine ECF and CSF concentrations. The next step will be to extend this model structure with additional physiological components to predict drug concentrations under injured/diseased brain conditions.
Reference:
[1] Westerhout J, Ploeger B, Smeets J, Danhof M, De Lange ECM. Physiologically based pharmacokinetic modeling to investigate regional brain distribution kinetics in rats. The AAPS Journal (2012)14:543-53.
Reference: PAGE 25 (2016) Abstr 5854 [www.page-meeting.org/?abstract=5854]
Poster: Drug/Disease modeling - CNS