Population pharmacokinetic assessment of the effect of food on piperaquine bioavailability in patients with uncomplicated malaria

Abstract

Previously published literature reports various impacts of food on the oral bioavailability of piperaquine. The aim of this study was to use a population modeling approach to investigate the impact of concomitant intake of a small amount of food on piperaquine pharmacokinetics. This was an open, randomized comparison of piperaquine pharmacokinetics when administered as a fixed oral formulation once daily for 3 days with (n=15) and without (n=15) concomitant food to patients with uncomplicated Plasmodium falciparum malaria in Thailand. Nonlinear mixed-effects modeling was used to characterize the pharmacokinetics of piperaquine and the influence of concomitant food intake. A modified Monte Carlo mapped power approach was applied to evaluate the relationship between statistical power and various degrees of covariate effect sizes of the given study design. Piperaquine population pharmacokinetics were described well in fasting and fed patients by a three-compartment distribution model with flexible absorption. The final model showed a 25% increase in relative bioavailability per dose occasion during recovery from malaria but demonstrated no clinical impact of concomitant intake of a low-fat meal. Body weight and age were both significant covariates in the final model. The novel power approach concluded that the study was adequately powered to detect a food effect of at least 35%. This modified Monte Carlo mapped power approach may be a useful tool for evaluating the power to detect true covariate effects in mixed-effects modeling and a given study design. A small amount of food does not affect piperaquine absorption significantly in acute malaria.

FIG 1

Final structural model for piperaquine population pharmacokinetics in fasting (n = 15) and fed (n = 15) patients with uncomplicated P. falciparum malaria in Thailand. CL, elimination clearance; F, relative oral bioavailability; ktr, transit absorption rate constant; Q, intercompartment clearances; V_C, apparent volume of distribution of the central compartment; V_P, apparent volume of distribution of the peripheral compartments; MTT, mean absorption transit time; n, number of transit compartments [MTT = (n + 1)/ktr].

FIG 2

Box plots (interquartile ranges with 2.5 to 97.5 percentiles) showing the effect of estimated food effects on mean absorption transit time (MTT) and relative bioavailability (F). Vertical dashed lines indicate no effect and ±25% effects.

FIG 3

Goodness-of-fit diagnostics of the final population pharmacokinetic model of piperaquine in fasting (n = 15) and fed (n = 15) patients with uncomplicated P. falciparum malaria. Broken lines, locally weighted least-squares regression; solid lines, line of identity; broken horizontal lines, lower limit of quantification. The observed concentrations, population predictions, and individual predictions were transformed into their logarithms (base 10).

FIG 4

Visual predictive check of the final model describing the population pharmacokinetics of piperaquine in fasting (n = 15) and fed (n = 15) patients with uncomplicated P. falciparum malaria. The inset shows piperaquine simulations at 0 to 72 h. Open circles, observed data points; solid lines, 5th, 50th, and 95th percentiles of the observed data; shaded areas, 95% confidence intervals of simulated (n = 2,000) 5th, 50th, and 95th percentiles. Broken horizontal lines are the lower limit of quantification. Venous plasma piperaquine concentrations were transformed into their logarithms (base 10).

FIG 5

Modified Monte Carlo mapped power approach, showing the statistical power to detect various degrees of covariate effects with the final pharmacokinetic model and the given study design (15 fed and 15 fasting patients).