Pharmacokinetics of antiretroviral drug varies with formulation in the target population of children with HIV-1

The bioequivalence of formulations is usually evaluated in healthy adult volunteers. In our study in 19 HIV-1-infected Ugandan children (1.8-4 years of age, weight 12 to <15 kg) receiving zidovudine, lamivudine, and abacavir solutions twice a day for ≥24 weeks, the use of scored tablets allowed comparison of plasma pharmacokinetics of oral solutions vs. tablets. Samples were collected 0, 1, 2, 4, 6, 8, and 12 h after each child's last morning dose of oral solution before changing to scored tablets of Combivir (coformulated zidovudine + lamivudine) and abacavir; this was repeated 4 weeks later. Dose-normalized area under curve (AUC)(0-12) and peak concentration (C(max)) for the tablet formulation were bioequivalent with those of the oral solution with respect to zidovudine and abacavir (e.g., dose-normalized geometric mean ratio (dnGMR) (tablet:solution) for zidovudine and abacavir AUC(0-12) were 1.01 (90% confidence interval (CI) 0.87-1.18) and 0.96 (0.83-1.12), respectively). However, lamivudine exposure was ~55% higher with the tablet formulation (AUC(0-12) dnGMR = 1.58 (1.37-1.81), C(max) dnGMR = 1.55 (1.33-1.81)). Although the clinical relevance of this finding is unclear, it highlights the impact of the formulation and the importance of conducting bioequivalence studies in target pediatric populations.