Efavirenz may be under-dosed in children

In a cohort of three to 12 year-old HIV-infected Ugandan children taking once-daily efavirenz according to 2006 World Health Organization (WHO) weight-band dosing recommendations, lower and highly variable amounts of drug entered their bloodstream compared to adults, investigators from the AntiRetroviral Research fOr Watoto (ARROW) trial team report in the advance online edition of the Journal of Acquired Immune Deficiency Syndromes.

Quirine Fillekes and collleagues suggest these findings highlight the increased risk of virological failure and urgently advocate for paediatricians in both resource-rich and resource-poor settings to move to WHO 2010 dosing guidelines for efavirenz.

WHO 2010 guidelines for HIV-infected children aged three and above recommend the use of two nucleoside reverse transcriptase inhibitors (NRTIs) and one non-nucleoside reverse transcriptase inhibitor (NNRTI). Efavirenz, an NNRTI, suitable for once-daily dosing because of its long half-life and potent antiviral activity is one of the most preferred first-line ARVs for HIV-infected children.

They recommend that efavirenz should be dosed by weight, according to the following formula:

• Liquid: 19.5mg/kg per day

• Tablet: 15mg/kg once a day

• Weight more than 40kg: 600mg once daily

The guidelines also note that the 200mg tablet version of efavirenz is scored, and can be split. Capsules can be opened and added to a small amount of food or liquid.

Little is known about how efavirenz is absorbed in children, especially in African children. Several studies have shown that a high proportion of children have sub-optimal levels blood levels of efavirenz.

The World Health Organization made weight-based dosing recommendations for children in 2006, largely based on data acquired from studies in adults.

Given the paucity of studies supporting international weight-band dosing for efavirenz, the authors undertook a pharmacokinetic (PK) sub-study to see if children in the ARROW trial were absorbing the full amount of drug when dosing according to the 2006 WHO recommendations.

ARROW, an open-label randomised trial in Uganda and Zimbabwe, is comparing routine laboratory to clinically driven monitoring strategies as well as comparing three different NRTI-based ART regimens in 1,206 HIV-infected infants and children.

41 HIV-infected Ugandan children from three to 12 years of age at two Ugandan ARROW centres taking lamivudine and abacavir twice daily with once-daily efavirenz for at least 36 weeks were enrolled in a two-period, cross-over open label pharmacokinetic study comparing twice to once daily lamivudine and abacavir. The children were not anticipated to change weight-bands within the following month.

Efavirenz was dosed according to 2006 WHO recommendations at 200, 250, 300 and 350 mg for children weighing 10<15,15<20, 20<25 and 25<30 kg, respectively; capsules of 200 and 50 mg were used, or unscored 600 mg tablets that were cut in the pharmacy.

At week 36 blood samples were taken immediately before taking ART and then at 1,2,4,6,8 and 12 hours later. After the sample evaluation (PK1) the children were switched to a once-daily (morning) regimen. Four weeks later (at week 40) sampling (PK2) was repeated and a further 24-hour sample taken.

At week 36 the median age and body weight were 7.6 (IQR: 5.6-9.1) years and 20.0 (IQR: 16.6-23.0) kg, respectively. Most were moderately stunted and wasted.

The authors reference Marzolini’s study looking at efavirenz efficacy and toxicity in adults, supporting a minimum target of 1.0 mg/L 8-12 hours after taking a dose and a maximum of 4.0 mg/L.

Efavirenz levels could be determined for all 41 children at PK1 and for 39 at PK2.

The area under the curve (AUC) is a measure used to estimate the concentrations of active drug available over time to suppress HIV. If the AUC falls too low virus levels may increase, leading to virologic rebound and the development of resistance.

The authors identified two studies suggesting that an AUC in children above 50 h.mg/L was associated with virological efficacy. In the ARROW sub-study 63% (26/41) at week 36 (PK1) and 51% (20/39) at week 40 (PK2) had an efavirenz exposure less than 50 h.mg/L. The mean AUC was 50.8 (90.8%) and 55.5 (82.7%) h.mg/L at PK1 and PK2, respectively.

17% (7/41) of the children had sub-therapeutic levels (under 1.0mg/L) of the drug eight hours and/or at 12 hours after taking the drug on one or both PK days; these findings are consistent with other studies.

59% (22/37) at PK1 and 38% at PK2 (15/39) of the children had sub-therapeutic levels of the drug 24 hours after taking it. 

Conversely toxic levels of the drug (more than 4.0 mg/L) were found in 29% (12) at eight and/or 12 hours after taking the drug: 10 children at week 36 (PK1) and 11 at week 40 (PK2).

The authors note the large number of children with efavirenz exposure outside of the therapeutic range; only 54% (22/41) were given adequate dosage (1.0-4.0 mg/L) determined by absorption eight to 12 hours after taking the drug. This is consistent with two other studies in African children with percentages of 47 and 66, they add.

No differences were seen across weight-bands suggesting use of half-tablets is acceptable.

The 2010 WHO recommendations have higher doses than those evaluated in this study or licensed for children in the 14<20, 25<30, 30<35 and 35<40 kg weight range; a result of both under-dosing found in two other studies and 50 mg capsules no longer being available. This will bring greater virological efficacy, but it also means over a third of children will be exposed to potentially toxic levels, note the authors.

The authors conclude “higher paediatric efavirenz doses, as per WHO 2010 recommendations should be used and investigated further, but may risk increasing the proportion of children with potentially toxic levels.”

Fillekes Q et al. Pediatric under-dosing of efavirenz: a pharmacokinetic study in Uganda. Advance online edition JAIDS doi: 10.1097/QAI.0b013e318235e560, 2011.