Population pharmacokinetics of lopinavir in HIV-infected adults receiving lopinavir/ritonavir

Objectives: Lopinavir (LPV) is an inhibitor of the protease of the human immunodeficiency virus (HIV). In order to exploit its pharmacokinetic (PK) profile, LPV is co-formulated with the cytochrome P450 inhibitor ritonavir (rtv). The aims of the present study were to develop a population PK model of LPV in HIV-1 infected adults receiving LPV/rtv, to assess the interindividual variability (IIV) in the PK parameters, and to identify possible covariates that could explain part of such variability.

Methods: The study included HIV-infected adults on stable therapy including LPV/rtv during at least 4 weeks. Blood samples were taken before and after drug administration at steady state to determine both LPV and rtv plasma concentrations. PK analysis was performed with the nonlinear mixed-effect modeling program (NONMEM v V) using FOCE method. The following covariates were evaluated in the analysis: age, sex, body weight, rtv AUC, rtv Cmin, hepatitis C virus (HCV) co-infection, concomitant use of tenofovir (TDF), total plasma proteins, albumin, AST, or ALT. Before inclusion in the model, correlations between each individual parameter and covariates were tested using the stepwise generalized additive models implemented in Xpose.

Results: Fifty HIV-infected adults were enrolled in the study (men 76%, age 43.4± 8.2 years, body weight 67.9± 9.6 Kg, HCV co-infection 46%). LPV/rtv doses were 400/100 mg twice daily in 46 patients, 266/66 mg twice daily in 2 patients, and 800/200 mg once daily in 2 patients. A total of 391 blood samples were analyzed (7.8; range: 4-9 samples per patient). A one compartment model with first order absorption and elimination was the most suitable model to describe LPV plasma concentrations. IIV in PK parameters and the residual error were respectively described by exponential and additive models. LPV clearance was found to be exponentially related to ritonavir AUC: TVCL=6.9*EXP(-0.07*AUCrtv); and volume of distribution was exponentially related to ritonavir minimum plasma concentration: TVV=39.7*EXP(1.23*CMINrtv). IIV estimated for clearance, volume of distribution and absorption rate constant after the inclusion of the covariates was 21%, 23%, and 93%, respectively.

Conclusion: The population model developed with the present data could be used to estimate LPV/rtv appropriate dosage in HIV-infected subjects. The suitability of LPV/rtv dosage individualization in clinical practice needs to be confirmed in prospective clinical trials.