Population Pharmacokinetics of Praziquantel in Pregnant and Lactating Filipino Women Infected with Schistosoma japonicum

Abstract

An estimated 40 million women of reproductive age are infected with one of three species of the waterborne parasite Schistosoma spp. Treatment with praziquantel (PZQ) via mass drug administration (MDA) campaigns is the mainstay of schistosomiasis control for populations living in areas of endemicity. The World Health Organization recommends that pregnant and lactating women be included in schistosomiasis MDA programs, and several recent studies have evaluated the safety and efficacy of PZQ use during pregnancy. To date, there are no data describing PZQ pharmacokinetics (PK) during pregnancy or among lactating postpartum women. As part of a randomized controlled trial investigating the safety and efficacy of PZQ during human pregnancy, we examined the PK of this therapeutic drug among three distinct cohorts of women infected with S. japonicum in Leyte, Philippines. Specifically, we studied the PK properties of PZQ among early- and late-gestation pregnant women (n = 15 each) and lactating postpartum women (n = 15) with schistosomiasis. We found that women in early pregnancy had increased apparent clearance and lower area-under-the-curve (AUC_0-24) values that may be related to physiological changes in drug clearance and/or changes in oral bioavailability. There was no relationship between body weight and apparent clearance. The mean ± standard deviation partition ratio of plasma to breast milk was 0.36. ± 0.13. The estimated median infant PZQ daily dose would be 0.037 mg/kg of body weight ingested from breast milk, which is significantly lower than the dosage required for antischistosomal activity and not known to be harmful to the infant. Our PK data do not support the suggestion to delay breastfeeding 72 h after taking PZQ. Results can help inform future drug efficacy studies in pregnant and lactating women with schistosomiasis.

Keywords: PK; Schistosoma japonicum; breast milk; lactation; pharmacokinetics; praziquantel; pregnancy; schistosomiasis.

FIG 1

Study flow design.

FIG 2

Median and individual PZQ concentration time profiles.

FIG 3

(A and B) The observed-predicted plots for the PZQ concentrations in plasma (A) and breast milk (B) after the Bayesian step. The median parameter values for each patient have been used. The observed-predicted data are plotted on a log-log plot for both outputs and are shown in the inserts. The regression line for plasma in panel A is given by observed = 0.016 + 1.04 · predicted; r² = 0.604. The regression line for breast milk in panel B is given by observed = 0.015 + 0.953 · predicted; r² = 0.468.

FIG 4

Residual plots for plasma concentrations. The average residuals did not vary from zero; P = 0.88 for weighted residual error versus predicted concentrations (A) and for weighted residual error versus time (B). The solid line in panels A and B is the loess regression. The residuals were normally distributed as assessed using D’Agostino, Shapiro-Wilk, and Kolmogorov-Smirnof tests (C). P > 0.05.

FIG 5

(A and B) The relationship between weight and clearance/F (A) and weight and volume/F (B). The volume is the volume of the central compartment. Neither relationship is statistically significant, with r = 0.03 (P = 0.84) and 0.16 (P = 0.29) for clearance/F and volume/F, respectively. The broken line is the loess line.

FIG 6

(A to D) Box plots showing the relationship between various stages of pregnancy and dose (A), volume of the central compartment/F (B), clearance/F (C) and the area under the concentration-time curve (AUC_0–24) in panel D. There was no relationship between the stage and pregnancy and the absolute dose (mg) and volume/F (P = 0.2072 and 0.626, respectively). Women in the early pregnancy group have a higher clearance/F than other women (P = 0.016 for all groups) and a lower AUC_0–24 (P = 0.01 for all groups).

FIG 7

(A and B) Monte Carlo simulations showing the drug exposure in plasma (A) and breast milk (B) from 1,000 lactating women. The lines represent the 5th, 25th, 50th, 75th, and 95th centiles, and the gray shading represents the confidence interval around each centile. In panel C, the AUC_0–24 in plasma versus breast milk in each simulated woman is shown with black open circles. The AUC_0–24 from each of the 15 patients in the study are shown with solid red circles.