Voriconazole Population Pharmacokinetics and Pharmacodynamics in Children
M. Neely (1,2,3), T. Rushing (4), A. Kovacs (3), R. Jelliffe (2), J. Hoffman (1)
Keck School of Medicine, University of Southern California, Los Angeles, CA: Divisions of (1) Infectious Diseases, Childrens Hospital Los Angeles; (2) USC Laboratory of Applied Pharmacokinetics; (3) Pediatric Infectious Diseases LAC+USC Medical Center; and (4) Pharmacy, Childrens Hospital Los Angeles
Objectives: Published voriconazole pharmacokinetic (PK) and pharmacodynamic (PD) data are limited in children and adolescents and differ from adults. We reviewed our experience with voriconazole therapeutic drug management (TDM) in this population.
Methods: Records at the Childrens Hospital Los Angeles were reviewed for children with ≥1 serum voriconazole concentration measured (at a commercial laboratory) from May 1, 2006 to June 1, 2007. Demographics, dosing histories, serum concentrations, and toxicity/survival data were obtained. The non-parametric population modeling and simulation MM-USCPACK software was used to compare distributions of concentrations simulated from published PK parameters1 as well as to fit the observed data to additional candidate PK models, which were evaluated on the basis of log-likelihood and observed vs. predicted plots. Statistical analyses were done with R 2.8.0.
Results: There were 207 voriconazole levels obtained from 46 patients (0.8 – 20.5 years). The median (range) dose was 5.8 mg/kg (2.0 – 12.9); 90% were oral. There were 3 (1 – 21) levels per patient. In each patient, levels were measured after different doses 99% of the time. The max concentration was 21.3 mg/L; min was <0.2 mg/L. The distribution of observed concentrations was similar to that predicted by the published model, except in those >12y, whose geometric mean concentration was 1.5 mg/L vs. 1.0 mg/L in those <12y, P=0.05. A 2-compartment Michaelis-Menten PK model fit the data best, but only explained 64% of the observed variability. Mean (CV%) bioavailability was 0.81 (20%), higher than previously reported. ALT increased over baseline in 33 (73%) of 45, and resolved in 23 (70%) of the 33. Crude mortality was 13 (28%). In a Cox model, there was a 2.6-fold increased odds of death (95% CI 1.6 – 4.8, P=0.002) per annualized trough serum voriconazole concentration <1 mg/L. Simulations predict that the EMEA-approved IV dose of 7 mg/kg will achieve a trough >1 mg/L in >50% of children, but that the EMEA-approved oral dose of 200 mg twice daily for all ages will achieve this trough in only 25-50% of children.
Conclusions: We found marked between-individual voriconazole PK variability, and additional within-individual variability after enteral dosing, likely due to random temporal changes in absorption. Nonetheless, we found a PD association between a voriconazole trough >1 mg/L and survival. Voriconazole TDM with a target trough of >1 mg/L appears to be indicated.
References:
[1] Karlsson MO, Lutsar I, Milligan PA. Population pharmacokinetic analysis of voriconazole plasma concentration data from pediatric studies. Antimicrob Agents Chemother. 2009;53(3):935-44.