Physiologically Based Pharmacokinetic (PBPK) Model for Prediction of Vancomycin Pharmacokinetics in Children

Conference: AAPS
Division: PBPK

Abstract

Aim: Ethical considerations prevent extensive clinical trials in pediatric populations; however, with the use of PBPK modeling, in vivo data from adults can be used to explore the mechanisms of drug disposition and pharmacokinetics (PK) in children following a variety of administration routes. Simulation tools allow exploring the sensitivity of exposure to individual processes involved in drug absorption, distribution, and elimination, and so can help in design of the trials to maximize their efficiency. Several studies were published in the past demonstrating the accuracy of PBPK models in predicting pediatric PK for compounds with simple (perfusionlimited) tissue distribution and elimination mainly by CYP metabolism. For this study, vancomycin (VCN) was selected for its very low membrane permeation that is not captured well by perfusion-limited tissue models, and for its elimination by renal secretion. Rapid changes in glomerular filtration rate (GFR) in the first few weeks after birth and the effect of both gestational age (GA) and postnatal age (PNA) add to the variability in clearance in neonates. Changes in body water content and distribution also affect drug distribution throughout the body and need to be accounted for when trying to predict PK for this age group.

Methods: VCN pharmacokinetics was simulated using the PBPKPlus™ module in GastroPlus™ 9.0 (Simulations Plus, Inc., Lancaster, CA). To account for the low diffusion of VCN through cell membranes, all tissues were treated as permeabilitylimited tissues. Organ weights, volumes, and blood perfusion rates were generated by the program’s internal Population Estimates for Age-Related (PEAR™) Physiology™ module. Renal clearance was estimated from GFR and fraction unbound in plasma (Fup*GFR). Tissue/plasma partition coefficients (Kp’s) were calculated using Poulin’s equation for drug partitioning into extracellular space (Poulin 2002) from in vitro and in silico physicochemical properties (ADMET Predictor™ 7.2, Simulations Plus, Lancaster, CA). The permeability-surface area products (PStcs) for individual tissues were calculated as the product of the Specific PStc (PStc per mL of tissue cell volume) and total cell volume of each tissue. The single value of Specific PStc used for all tissues was fitted against in vivo plasma concentration-time (Cp-time) data after i.v administration of VCN in rats. The model was subsequently used to predict the VCN PK in human adults. After validating against adult data, the model was used to predict VCN PK in different pediatric groups, including neonates and infants. The importance of GA vs PNA on VCN PK was explored.

American Association of Pharmaceutical Scientists (AAPS), October 25-29, 2015, Orlando, FL

By Viera Lukacova, Walter S. Woltosz, Michael B. Bolger