Laura Dickinson (1), Quirine Fillekes (2), Tim R. Cressey (3,4), Kulkanya Chokephaibulkit (5), Gerry Davies (1), Victor Musiime (6), Philip Kasirye (7), Saye Khoo (1), David Burger (2), Sarah Walker (8)
(1) University of Liverpool, Liverpool, UK; (2) Radboud University Nijmegen Medical Centre, Nijmegen, the Netherlands; (3) Program for HIV Prevention and Treatment (IRD URI 174), Faculty of Associated Medical Sciences, Chiang Mai University, Chiang Mai, Thailand; (4) Harvard School of Public Health, Boston, MA, USA; (5) Faculty of Medicine Siriraj Hospital, Mahidol University, Bangkok, Thailand; (6) Joint Clinical Research Centre, Kampala, Uganda; (7) PIDC, Mulago Hospital, Kampala, Uganda; (8) MRC Clinical Trials Unit, London, UK
Objectives: Zidovudine (ZDV) is a component of HIV therapy for children but limited PK data are available on the widely applied twice daily dose. WHO guidelines aim to provide a simplified dosing approach based on weight bands. In 2010, the WHO increased ZDV doses in several weight bands compared to 2006 guidelines. Our aim was to develop a model to describe ZDV plasma concentrations in HIV-infected children and estimate exposures following WHO dosing.
Methods: Rich and sparse data were combined from 6 paediatric studies. A total of 773 plasma concentrations were available from 100 children (n=63 girls) stable on ZDV tablets (n=54 PK profiles) or syrup (n=73 PK profiles) dosed according to WHO or National guidelines. Median (range) ZDV dose was 150mg (60-300) [9mg/kg (5.1-24.0)]. Median age and weight were 5yr (1-18) and 18kg (6-59). Nonlinear mixed effects modelling (NONMEM v.7.2) was applied to estimate ZDV population PK. Covariates that could potentially influence ZDV PK were evaluated. Validity of the model was assessed using visual predictive check. Comparison of ZDV exposures based on WHO 2006 and 2010 guidelines was performed using simulations.
Results: ZDV PK was described by a 2-compartment model with first-order (tablet; ka 3.0h-1) or zero-order (syrup; D2 0.7h) absorption. Weight on clearance and volume of distribution parameters using allometric scaling improved the fit. A positive linear relationship was observed between ZDV CL/F and age. ZDV CL/F, Q/F, Vc/F and Vp/F were 62, 7L/h/18.3 kg and 66, 53L/18.3 kg, respectively (IIV CL/F 32%). Of the observed ZDV concentrations 91% were within the 90% prediction interval. Mean ZDV AUC0-12 was 25% higher compared to published adult data [1] (2.81 vs. 2.24mg.h/L). Simulated mean AUC0-12 for WHO 2010 guidelines (2.52-3.55mg.h/L) were higher than those for 2006 guidelines (2.10-3.13mg.h/L) in the majority of weight bands due to increased doses. Stratifying simulated ZDV AUC0-12 according to arbitrary low and high thresholds of half and double the average adult exposure resulted in 0-7% and 0-3% of AUC0-12 below 1.12mg.h/L and 3-16% and 6-25% above 4.48mg.h/L for 2006 and 2010 guidelines, respectively.
Conclusions: Mean ZDV exposure was higher in children than adults. Additional safety and tolerability data at higher WHO 2010 doses are needed to establish a toxicity threshold for ZDV. Impact on active intracellular ZDV triphosphate requires investigation; however, predicted plasma exposures are reassuring.
References:
[1] Aurobindo Pharma Limited. Zidovudine 100 mg Capsules. Summary of Product Characteristics 2011 (http://www.medicines.org.uk/emc/medicine/24327/SPC/zidovudine%20100mg%20capsules/)
Reference: PAGE 22 (2013) Abstr 2726 [www.page-meeting.org/?abstract=2726]
Poster: Paediatrics