2008 - Marseille - France

PAGE 2008: Applications- Anti-infectives
Michael Neely

Population Kinetics and Dynamics of Lopinavir in HIV-Infected Children

M. Neely (1), J. van den Anker (2), S. Soldin (2), K. Williams (2), A. Baghdassarian (2), N. Rakhmanina (2)

(1) Laboratory of Applied Pharmacokinetics, University of Southern California, Los Angeles; (2) Children's National Medical Center, Washington DC

Objectives: To develop a population pharmacokinetic (PK) and pharmacodynamic (PD) model of lopinavir (LPV) in HIV-infected children.

Methods: Blood samples from 52 HIV-infected children (4-17y) receiving >1m LPV-based therapy (mean 264 mg/m2 twice daily) were taken just prior to observed dosing, and at 0.5, 1, 2, 4, 8 and 12 hours after. Viral LPV 50% inhibitory concentration (IC50) was taken from the Phenosense (Virologic, Inc.) assay. LPV concentrations1 were fitted to candidate PK models2, with pre-observed dose concentrations used as initial conditions, thereby controlling for adherence.  The distribution of concentrations in 1000 children simulated from the model was compared to the distribution in real participants (visual predictive check).  A published model3 of probability of viral suppression (<400 copies/mL) as an Emax function of LPV inhibitory quotient (IQ), computed as pre-dose LPV concentration divided by protein-adjusted viral IC50, was transformed to LPV activity ranging from 0 to 100%.  LPV activity in children was estimated at 3 LPV concentrations and the 3 Phenosense susceptibility cutoffs.   

Results:  The final model (median, quartile range) had delayed (0.3 h, 0.2-0.8) linear absorption (0.2 h-1, 0.1-2.0) into a compartment whose volume (13.0 L, 9.1-29.0) was linearly dependent on weight and inversely proportional to age.  Elimination (0.16 h-1, 0.06-0.29) was inversely proportional to weight0.25.  AUC0-12 was 95.9 mg*h/L (62.8-115.3) and half-life 4.2 h (2.4-10.7).  Median predicted and observed 12-h LPV were 4.4 and 5.4 mg/L in children who were not at steady state, and 7.5 and 7.4 mg/L in children at steady state.   Obs vs. pred R2 was 0.96.  Predicted LPV efficacy is shown in the table.  At the mean LPV pre-dose concentration of 8.4 mg/L and mean 26-fold increased IC50 in the real children, the model predicted 27% LPV activity; 33% of the children achieved viral suppression. 



IQ / Estimated LPV activity by fold IC50 increase

Pre-dose LPV

Pop %ile

1 (susceptible)


9 (intermediate)


55 (resistant)

2.9 mg/L


41 / 93%

5 / 27%

 1 / 2%

7.5 mg/L


107 / 98%

12 / 63%

2 / 9%

16.1 mg/L


230 / 99%

 26 / 85%

 4 / 24%

Conclusions: PK model parameters are similar to previous reports and accurately predicted pediatric LPV concentrations. Overall LPV activity, which declines sharply with increasing resistance, was similar to rate of viral suppression.  Given LPV variability, ensuring adequate drug concentrations for intermediate virus could be beneficial. 

[1] Volosov A, Alexander C, Ting L et al. Simple rapid method for quantification of antiretrovirals by liquid chromatography-tandem mass-spectrometry. Clin Biochem. 2002; 35(2):99-103.
[2] USCPACK Software for Parametric and Non-Parametric Population Modeling. Available by licence from the Laboratory of Applied Pharmacokinetics and the University of Southern California, http://www.lapk.org/. 2008.
[3] Hsu A, Isaacson J, Brun S et al. Pharmacokinetic-pharmacodynamic analysis of lopinavir-ritonavir in combination with efavirenz and two nucleoside reverse transcriptase inhibitors in extensively pretreated human immunodeficiency virus-infected patients. Antimicrob Agents Chemother. 2003; 47(1):350-359.

Reference: PAGE 17 (2008) Abstr 1340 [www.page-meeting.org/?abstract=1340]
Poster: Applications- Anti-infectives