Application of nonlinear mixed effects modeling in allometric interspecies scaling of emtricitabine

Background/Aims Allometry successfully predicted human pharmacokinetic (PK) properties of nucleoside analogues due to efficient renal excretion and low plasma protein binding. We aim to characterize the allometric relationships that explain PK species differences of emtricitabine ((−)‐FTC), an anti‐HIV and anti‐HBV nucleoside analogue. Methods A 2‐compartment open model was simultaneously fit to drug concentrations after IV administration to mice, rats, woodchucks and monkeys using NONMEM ® , with body weight as a covariate according to power law. Results Including woodchuck resulted in a deficient fit and mispredicted all species (model 1). Estimating woodchuck clearance outside the allometric relationship (model 2), based on known low renal clearance of nucleoside analogues, significantly improved the fit. Accounting for circannual adipose changes with a weight multiplier parameter (model 3) further improved the fit. The allometric exponents were not significantly different from theoretical values of ¾ and 1 for flows and volumes, and human clearance was accurately predicted. Predicted human PK parameters varied by 2 fold between the models, and were 28 and 22 L/h for total and distributional clearances, and 40 and 33 L for central and peripheral volumes, respectively, for model 3. Conclusions Accounting for species‐specific properties can better characterize the allometric relationships of PK parameters, and may help predict human properties more accurately. Clinical Pharmacology & Therapeutics (2005) 77 , P80–P80; doi: 10.1016/j.clpt.2004.12.197