Secondary Logo

Journal Logo

Institutional members access full text with Ovid®

Population Pharmacokinetics of Atazanavir in Human Immunodeficiency Virus-Infected Patients

Solas, Caroline PhD*; Gagnieu, Marie-Claude PhD; Ravaux, Isabelle MD; Drogoul, Marie-Pierre MD§; Lafeuillade, Alain MD; Mokhtari, Saadia MD; Lacarelle, Bruno PhD*; Simon, Nicolas PhD, MD*

doi: 10.1097/FTD.0b013e3181897bff
Original Article

The aim of this study was to determine the population pharmacokinetic (PK) parameters of atazanavir in adult human immunodeficiency virus-infected patients to build up a Bayesian strategy for dosage regimen individualization. This was an observational study of patients treated with the once-daily regimen atazanavir associated with 100 mg of ritonavir. Blood samples were drawn at steady state at various times ranging from 1 to 26 hours postdose. Atazanavir plasma concentrations were determined by a validated reverse-phase high-performance liquid chromatography method. PK analysis of the atazanavir population was performed using a nonlinear mixed-effects model (NONMEM version 6). One hundred eighty-seven patients were included in the study. The atazanavir doses prescribed were 300 mg (n = 169), 400 mg (n = 12), 200 mg (n = 1), and 150 mg (n = 5). The atazanavir population PK was described using a 1-compartment model with first-order absorption. Mean PK parameter estimations (95% confidence interval, coefficients of variation %) were as follows: oral clearance (CL) = 7.6 L/h (6.9-8.3; 34%), volume of distribution (V) = 80.8 L (67.4-94; 37%), and absorption constant rate (Ka) = 1.05 hours (0.01-2.09; 156%). The mean estimated half-life (T-half) was 7.5 hours (95% confidence interval: 7.2 -7.8 hours). The estimated T-half of atazanavir was in agreement with that previously reported of 8.6 and 8.8 hours. We observed a wide interpatient variability for the PK parameters, especially for Ka. This population approach allowed us to determine atazanavir PK parameters in human immunodeficiency virus-infected patients in a real-life context and to perform Bayesian analysis to predict Ctrough from samples collected at any moment during the dosing interval. This could therefore improve therapeutic drug monitoring interpretations and provide an interesting tool for correlation with virologic data.

From the *CHU La Timone, Fédération de Pharmacologie et de Toxicologie, Marseille; †Hôpital Edouart-Herriot, Fédération de Biologie, Lyon; ‡CHU La Conception, Service des Maladies Infectieuses, Marseille; §CHU Sainte-Marguerite, CISIH-Sud-Hématologie, Marseille; ‖Hôpital Chalucet, Service des Maladies Infectieuses, Toulon; and ¶CHU Nord, Service des Maladies Infectieuses, Marseille, France.

Received for publication June 25, 2008; accepted August 7, 2008.

Correspondence: Caroline Solas, PhD, Fédération de Pharmacologie et de Toxicologie, Hôpital de La Timone, 264 rue Saint-Pierre, 13385 Marseille, France (e-mail:

© 2008 Lippincott Williams & Wilkins, Inc.