Population analysis of plasma and intracellular pharmacokinetics of indinavir in HIV-1 infected patients with a stable antiretroviral therapy
Dubois A (1), Duval X (1,2), Peytavin G (2), Taburet AM (3), Goujard C (3), Mentré F (1,2) and the Cophar1-ANRS 102 Trial Group
(1) INSERM, U738, Paris, France; University Denis Diderot, Paris, France. (2) AP-HP, Bichat University Hospital, Paris, France. (3) AP-HP, Bicêtre University Hospital, Paris, France.
Objectives: Indinavir is a HIV protease inhibitor whom activity is intracellular. However, only few studies with measurements of intracellular indinavir concentrations have been performed [1, 2, 3]. The objective of the present study was to characterize the intracellular pharmacokinetics (PK) of indinavir in connection with its plasma PK in HIV infected patients with a stable antiretroviral therapy.
Methods: Data came from Cophar1-ANRS 102 Trial . Patients were required to have unchanged antiretroviral treatment for 6 months with a sustained virological response defined by plasma HIV RNA level <200 copies/mL for at least 4 months. Plasma concentrations were measured at 5 different sampling times (one before indinavir administration and four at fixed times after) from 42 patients who received different dosages of indinavir either alone or with a booster dose of ritonavir (13 patients). Among the 42 patients, 8 had also measurements of intracellular concentrations at 4 of the sampling times. Plasma alone and then plasma with intracellular data were modelled in all patients using a population approach. No model for joint analysis of plasma and intracellular concentrations of protease inhibitors has been published and we tested several models. Population Parameters were estimated using the SAEM algorithm  in monolix v.2.1 .
Results: A two-compartment model with first order absorption with a lag time and first-order elimination best described indinavir plasma PK. The best joint model had the same model for plasma concentrations and intracellular concentrations were proportional to plasma concentrations. It should be recalled that concentrations were measured at steady state. For plasma PK, the lag time Tlag was 0.38 h (44.5% inter-individual variability, IIV), the absorption rate constant ka was 2.05 h-1 (72% IIV), the apparent volume of distribution V/F was 62.3 L and the apparent clearance Cl/F was 45.4 Lh-1 (17.1% IIV) for patients treated with indinavir alone. The administration of indinavir plus ritonavir decreased Cl/F by 49%. Proportionality coefficient between plasma and intracellular concentrations δ was 1.84 d.l. (15.8% IIV).
Conclusions: A joint proportional model was found to describe adequately plasma and intracellular concentrations of indinavir at steady state. This is the first model of plasma and intracellular PK of indinavir in patients.
 Hennessy M, Clarke S, Spiers JP, Mulcahy F, Kellher D, Meadon E, Maher B, Bergin C, Khoo S, Tjia J, Hoggard P, Back D, Barry M. Intracellular indinavir pharmacokinetics in HIV-infected patients: comparison with plasma pharmacokinetics. Antivir Ther. 2003; 8: 191-198.
 Nascimbeni M, Lamotte C, Peytavin G, Farinotti R, Clavel F. Kinetics of antiviral activity and intracellular pharmacokinetics of HIV 1 protease inhibitors in tissue culture. Antimicrob Agents Chemother. 2000; 43: 2629-2634.
 Ford J, Khoo SH, Back DJ. The intracellular pharmacology of antiretroviral protease inhibitors. Antimicrob Agents Chemother. 2004; 54: 982-990.
 Goujard C, Legrand M, Panhard X, Diquet B, Duval X, Peytavin G, Vincent I, Katlama C, Leport C, Bonnet B, Salmon-Céron D, Mentré F, Taburet AM, the COPHAR1-ANRS 102 Study Group. High variability of indinavir and nelfinavir pharmacokinetics in HIV-infected patients with a sustained virological response on highly active antiretroviral therapy. Clin Pharmacokinet. 2005; 44: 1267-1278.
 Kuhn E, Lavielle M. Maximum likelihood estimation in nonlinear mixed effects model, Comput Stat Data Anal. 2005; 49:1020-1038.