Population pharmacokinetic analysis for nelfinavir and its metabolite M8 in virologically controlled HIV-infected patients on HAART

Abstract

Aims: To describe the pharmacokinetics of nelfinavir and its main metabolite M8 in HIV-infected patients with a sustained virological response, to characterize the effect of covariates and to estimate inter- and intra-individual variability in the pharmacokinetics.

Methods: Three hundred and twenty concentrations of both nelfinavir and M8 were measured in 46 patients enrolled in the COPHAR 1-ANRS 102 study. Blood samples were taken at a first visit (one sample before drug administration and four samples at fixed times after) and at a second visit 1 to 3 months later (one before and one 3 h after drug administration). The data from both visits on nelfinavir and M8 were modelled jointly in all patients using a population approach.

Results: A one-compartment model with first-order absorption and elimination best described nelfinavir data, with an additional compartment incorporating a first order rate-constant describing the metabolism of the drug to M8. For nelfinavir, the apparent volume of distribution (V/F ) (95% confidence interval for the mean), was 309 l (185, 516), the absorption rate constant (k(a)) was 0.4 h(-1) (0.2, 0.8), and the apparent clearance (CL/F ) was 37.3 l h(-1) (32, 44). For M8, V(m) /(Fk(m)) and CL(m)/(Fk(m)) were 866 l h(-1) (351, 2161) and 1670 l (965, 2894), respectively. The interindividual variabilities were 34.9%, 34.3% and 62.2% for V/F, CL/F and CL(m)/(Fk(m)), respectively. The interoccasion variability was 27.8% for CL/F. The mean half-lives were 05.38 h and 00.44 h for nelfinavir and M8, respectively. Significant but opposite effects of comedication with zidovudine were found on nelfinavir CL/F and M8 CL(m)/(Fk(m)), but they were not considered to be clinically relevant.

Conclusions: A joint model was found to describe adequately nelfinavir and M8 concentrations and was used to estimate pharmacokinetic parameters for M8. The model can be used to build reference pharmacokinetic profiles for therapeutic drug monitoring of the drug.

Figure 1

The pharmacokinetic model for nelfinavir and M8. D: the dose of nelfinavir per intake; F: the bioavailability of nelfinavir; k_a: the first-order absorption rate constant for nelfinavir; k_m: the first-order metabolic rate constant; k_e: the elimination rate constant for nelfinavir; V: the volume of distribution of nelfinavir; V_m: the volume of distribution of M8; k_em: the elimination rate constant for M8

Figure 2

Observed plasma concentrations of nelfinavir (top) and M8 (bottom) vs. time at both clinic visits (1 for V1 and 2 for V2)

Figure 3

Goodness of fit plots based on the final model for nelfinavir: population predicted concentrations vs. observed concentrations (A), individual predicted concentrations vs. observed concentrations (B), population weighted residuals vs. observed concentrations (C) and population weighted residuals vs. time (D)

Figure 4

Goodness of fit plots based on the final model for M8: population predicted concentrations vs. observed concentrations (A), individual predicted concentrations vs. observed concentrations (B), population weighted residuals vs. observed concentrations (C) and population weighted residuals vs. time (D)

Figure 5

Evaluation of the final model: comparison between the 10th, 50th (in bold) and 90th percentiles obtained from 1000 simulations for patients receiving 1250 mg twice daily (A: nelfinavir, C: M8) and 750 mg three times daily (B: nelfinavir, D: M8) of nelfinavir, and the observed data at V1 and V2 in patients receiving the corresponding dosing regimen

Figure 6

Box plots of trough nelfinavir (A), M8 (B and D) and the sum of nelfinavir + M8 (C) concentrations simulated from the population model for 1250 mg nelfinavir twice daily according to comedication with zidovudine (A, B and C) or age quartiles (D) (the boxes represent 25th to 75th percentiles, with the median shown within the boxes; capped bars represent 1.5 times the interquartile range)