Background:  Correct dosing of HIV protease inhibitors (PI) in HIV-infected children is a big challenge due to age-dependent changes in pharmacokinetics and the relative paucity of the data. More particular, there are hardly any data for Asian children using PI.

Method:  At baseline, the 50 subjects participating in this open-label single-arm study received lopinavir/ritonavir (LPV/r) (230/57.5 mg/m²) and saquinavir (SQV) (50 mg/kg). Pre-dose plasma concentrations (C_min) were prospectively collected at weeks 12, 24, 36, 48, 60, 72, 84, and 96. Subjects with C_min <0.1 mg/L for LPV or 0.02 mg/L for SQV were excluded from analysis because of suspected nonadherence. Physicians adjusted dosages according to clinical, growth, and C_min. C_min above the 50% inhibitory concentration (IC₅₀) of >1.0 mg/L for LPV and >0.28 mg/L for SQV were desired. The dosage per body surface area (LPV) or per kilogram body weight (SQV) was calculated for each time-point. The mean C_min were calculated per time-point; intervariability was calculated as a coefficient of variance. The intrasubject variability was calculated as the standard deviation (SD) divided by the mean over the different time points. A paired t-test was used to detect differences between time points in C_min.

Results: The median age at baseline was 9.3 years (IQR 7.1 to 11.2). For each time-point, 42 to 48 samples were available for analysis (total 348 for LPV and 353 for SQV). The overall mean (SD) of the C_min were 5.52 (3.85) and 1.37 (1.24) mg/L for LPV and SQV, respectively. There was no significant change over time for the C_min levels. At week 96, 5.5% and 3.7% had C_min >IC₅₀ for LPV and SQV. The average dosage for SQV decreased from 43 mg/kg at baseline to 34 mg/kg at week 96. The LPV dose was almost equal—228 mg/m² and 229 mg/m²—at the same time-points. The median (IQR) intravariability (coefficient of variance) over 96 weeks was 55% (41 to 76) and 67% (54 to 85) for LPV and SQV, respectively. The median intervariability was 69% (60 to 74) for LPV and 81% (76 to 94) for SQV.

Conclusions:  The mean C_min for LPV, and SQV, were consistent over time with a large inter- and intrasubject variability. Despite the dose reduction of 32% from standard dosing for SQV at week 96, the plasma levels indicate that further dose reduction is possible. Similarly, the high LPV C_min suggests that dose reduction is reasonable. Further research is warranted to establish a more appropriate dose for this Asian population.