Population pharmacokinetic model development and evaluation after nevirapine administration to mothers and newborns
M. Frank (1), A. Kunz (2), G. Harms (2), C. Kloft (1)
(1) Martin-Luther-Universitaet Halle-Wittenberg, Dept. Clinical Pharmacy, Halle, Germany; (2) Institute of Tropical Medicine, Charité – University Medicine Berlin, Germany.
Objectives: To reduce the risk of HIV transmission in resourced-limited areas a single oral dose of nevirapine (NVP), a non-nucleoside reverse transcriptase inhibitor, was administered to 62 HIV-1 positive pregnant women and their newborns. Due to the sparse data situation population pharmacokinetic (PK) analysis was performed to characterise the PK of three different matrices (mother and newborn plasma as well as breast milk).
Methods: Study medication were a 200 mg NVP tablet for pregnant women during labour and a 2 mg/kg NVP syrup for newborns within 3 d after birth, resulting in 113 mother plasma, 95 breast milk and 113 newborn plasma samples over 3 weeks. Population PK analysis for mother and newborn data were performed using the nonlinear-mixed-effect modelling approach implemented in NONMEMTM (ADVAN6, TRANS1, TOL5; FOCE INTERACTION estimation method). The process of model refinement of previous models for mother and newborn data [1] was guided by standard diagnostic tools and by visual predictive checks (VPC) to investigate model performance.
Results: A 2 compartment PK model was developed for mother data and a 1 compartment model for newborn data. Due to sparse data, absorption rate constant was fixed to 1.66 h-1 [2]. Model simulated maximum concentrations for mother plasma, breast milk and newborn plasma were approximately 2300 ng/mL, 1800 ng/mL and 1500 ng/mL, respectively, indicating high drug transfer into milk and into foetus/newborn. Refined models comprised: i) different types of transfer for NVP from plasma to breast milk, e.g. uni- vs bidirectional, ii) time-dependence of switching on the NVP transfer into breast milk and iii) varied residual error structures. First VPC revealed an adequate central tendency although late time points (2 weeks after birth) were generally underestimated. However, the variability of the model was sufficiently reflected.
Conclusions: Population PK models for mother data and newborn data were developed and evaluated by different diagnostic tools. Preliminary results suggest adequate model performance for both patient groups. Final PK models could assist single dose NVP prevention strategies of HIV transmission from mother-to-child.
References:
[1] Frank M, Kunz A, Harms G, Kloft C, Nevirapine - Population pharmacokinetic model building and simulation for mothers and newborns, PAGE 17 (2008) Abstr 1249 [www.page-meeting.org/?abstract=1249].
[2] Kappelhoff BS, van Leth F, MacGregor TR, et al. Nevirapine and efavirenz pharmacokinetics and covariate analysis in the 2NN study. Antivir. Ther., 10: 145-155 (2005)
