Generic Workflow to Predict Medicine Concentrations in Human Milk Using Physiologically-Based Pharmacokinetic (PBPK) Modelling—A Contribution from the ConcePTION Project

Abstract

Women commonly take medication during lactation. Currently, there is little information about the exposure-related safety of maternal medicines for breastfed infants. The aim was to explore the performance of a generic physiologically-based pharmacokinetic (PBPK) model to predict concentrations in human milk for ten physiochemically diverse medicines. First, PBPK models were developed for “non-lactating” adult individuals in PK-Sim/MoBi v9.1 (Open Systems Pharmacology). The PBPK models predicted the area-under-the-curve (AUC) and maximum concentrations (Cmax) in plasma within a two-fold error. Next, the PBPK models were extended to include lactation physiology. Plasma and human milk concentrations were simulated for a three-months postpartum population, and the corresponding AUC-based milk-to-plasma (M/P) ratios and relative infant doses were calculated. The lactation PBPK models resulted in reasonable predictions for eight medicines, while an overprediction of human milk concentrations and M/P ratios (>2-fold) was observed for two medicines. From a safety perspective, none of the models resulted in underpredictions of observed human milk concentrations. The present effort resulted in a generic workflow to predict medicine concentrations in human milk. This generic PBPK model represents an important step towards an evidence-based safety assessment of maternal medication during lactation, applicable in an early drug development stage.

Challenges Related to Acquisition of Physiological Data for Physiologically Based Pharmacokinetic (PBPK) Models in Postpartum, Lactating Women and Breastfed Infants—A Contribution from the ConcePTION Project

“…Developing PBPK models for these populations also depends on the PBPK software but are mainly developed from an existing and verified PBPK model for an adult, non-lactating population. Subsequently, this model is extended to resemble the medicine concentration profile in the targeted special population by implementing modifications in terms of characteristic physiology [64].…”

“…A currently available postpartum PBPK model is an extension of a previously developed pregnancy model with a compartment for breast tissue and for endometrium and myometrium of the uterus among other things, while pregnancy-specific compartments (e.g., foetus, placenta) are discarded [65]. Lactation PBPK models are developed from a verified adult, non-lactating population and extended with specific physiological data, such as a breast and human milk compartment [9,64,66]. The infant PBPK models are extrapolated from the adult PBPK model to represent the population, with physiological data irrespective of the type of feeding [67].…”

“…To add these extensions to the adult model, physiological data of the populations were searched for in the literature and were implemented as mathematical equations fitted to the observed data. However, many physiological data were included that were based on assumptions due to the limited data and non-systematic searches of the literature [64,65,67]. This means that data available in the literature might be missed when developing a postpartum, lactation and breastfed infant PBPK model.…”

Predicting Maternal and Infant Tetrahydrocannabinol Exposure in Lactating Cannabis Users: A Physiologically Based Pharmacokinetic Modeling Approach

“…The application of PBPK models to investigate drug pharmacokinetics in special populations, including lactating women, has garnered significant interest over the years. In a recent research endeavor, PBPK modeling was employed to characterize the breastmilk pharmacokinetics of ten drugs across diverse biopharmaceutics classifications [111]. Specifically, in this study, we developed a PBPK model to estimate THC exposure in human milk and breastfed infants.…”

Physiologically Based Pharmacokinetics Modeling in the Neonatal Population—Current Advances, Challenges, and Opportunities

“…Examples of such subpopulations include pregnant and lactating women, fetuses, neonates and infants, and other pediatric groups [2][3][4]. A major cornerstone of PBPK modeling is the incorporation of unique patient physiology, and it is thus a powerful tool for anticipating how drug PK could differ in these populations compared to populations more extensively studied [5][6][7]. Additionally, this tool can be used for in silico predictions aimed at the rational choosing of "first in population" doses for these patients, when traditional allometric scaling may be less accurate.…”