TMC114 is a HIV protease inhibitor (PI) that is potent against wild-type and PI-resistant human immunodeficiency virus. TMC114 is administered with low-dose ritonavir (TMC114/r). Phase IIb studies (POWER 1 and 2) are ongoing randomized, placebo-controlled studies that evaluate dose-response after 24 weeks in 3-class-experienced patients with ³1 primary PI mutation. Four TMC114/r doses in an experimental formulation were compared with control PI regimens. All patients received an optimized background regimen (OBR), composed of nucleoside reverse transcriptase inhibitors with or without enfuvirtide.

A population model was developed to describe the pharmacokinetics of TMC114 in healthy volunteers and HIV-1 infected subjects who received TMC114/r, and a covariate analysis was performed. Richly sampled TMC114 plasma concentration time profiles were obtained from Phase I studies as well as from sub-groups of HIV-1 infected subjects in the POWER 1 and 2 trials and combined with sparse data obtained in all subjects in the Phase IIb trials. TMC114 was always administered under fed conditions with low dose ritonavir (100 mg). The analysis included a total of 580 subjects with TMC114 doses ranging from 400 mg qd up to 800 mg bid. The effects of subject covariates (age, body weight, sex, race, creatinine clearance, HIV status, hepatitis co-infection) on the pharmacokinetics of TMC114 were evaluated.

The final population pharmacokinetic model was a two compartmental model with a first order absorption. CL/F was inversely dependent on alpha-1 acid glycoprotein (AAG) concentrations. Additionally, there was a residual dose-disproportionality in the pharmacokinetics of TMC114 which was modelled descriptively as CL/F increasing with increasing total daily dose. This provided a significant improvement in the model fit and removed any further dose non-linearity. A predictive check was performed indicating that the model was satisfactory.

The model was then applied to obtain empirical Bayes’ estimates of the AUC and Cmin of TMC114 in 292 subjects in the POWER 3 study which examined the efficacy and safety of TMC114/r at the recommended therapeutic dose of 600/100mg bid (commercial formulation) in treatment-experienced HIV-infected patients receiving an OBR.

Description of the TMC114 pharmacokinetics based on the model resulted in a bias in population predictions for concentration and a consistent underestimation of the individual apparent CL/F, likely due to a different bioavailability of the commercial formulation as compared to the experimental formulation used in model development. Updating the model by inclusion of a relative bioavailability of the commercial tablet removed most of the bias in the predicted values. The estimated correction factor suggested a modest increase of 18% in the bioavailability of the commercial formulation. In addition, evaluation of the exposure estimates in subjects for whom data was available on both formulations suggested that in the clinical setting there was little difference in exposures when switching to the commercial tablet formulation.

In conclusion, the population model for TMC114 described the data well and was able to explain some of the variability in CL/F on the basis of the influence of AAG concentrations. Furthermore, CL/F was shown to increase with total daily doses of TMC114 resulting in a less than dose-proportional increase in total plasma concentrations of TMC114. None of the other tested covariates were considered to provide a clinically meaningful effect on the pharmacokinetics of TMC114. The model appears to have adequate predictive performance and was able to describe the pharmacokinetics of TMC114 following administration of a different formulation when a correction factor for bioavailability was included