Population pharmacokinetics of efavirenz in HIV and TB/HIV coinfected children: the significance of genotype-guided dosing

Objectives The current WHO weight-based dosing recommendations for efavirenz result in a wide variability of drug exposure in children. Our objectives are to characterize the effects of rifampicin- and isoniazid-containing anti-TB therapy and CYP2B6 activity on efavirenz concentrations in children, using non-linear mixed-effects modelling. Methods This is a pharmacokinetic (PK) substudy of a prospective study that examined the interactions between anti-TB therapy and efavirenz in HIV-infected children with and without TB. PK samples were obtained 4 weeks after starting efavirenz (PK1) and repeated 4 weeks after completing TB therapy (PK2) in TB/HIV coinfected patients. Drug concentrations were measured using LC-MS/MS. Composite CYP2B6 516/983/15582 genotype was determined. Population PK modelling was performed in Monolix. Simulations were performed to obtain the predicted mid-dose concentrations (C12). Results One hundred and five HIV-infected Ghanaian children (46 with TB/HIV) were included. The median age and weight were 7 years and 19 kg. The efavirenz concentrations over time were adequately described using a one-compartment model. Weight, composite CYP2B6 genotype and PK visit had a significant influence on the PK parameters, while TB therapy had no significant effect. Simulations showed adequate C12 for intermediate composite CYP2B6 metabolizers only. Conclusions Our model showed that rifampicin- and isoniazid-containing anti-TB therapy does not influence efavirenz PK parameters. On the other hand, it describes the effect of efavirenz autoinduction after completing TB treatment. In addition, dosing efavirenz in children based only on weight results in a large variability in drug exposure. We propose dose adjustments for slow and extensive composite CYP2B6 metabolizers.