Four Placental-Clearance Methods Give Limited Fetal Exposure Predictions
A nine-drug comparison found that reasonably accurate maternal PBPK predictions did not ensure agreement with umbilical-cord measurements. Placental-clearance estimates remained a major source of uncertainty.
Testing the Placental Input
A Central South University-led study compared four ways of estimating transplacental clearance for maternal–fetal physiologically based pharmacokinetic (PBPK) models. The researchers selected nine drugs with suitable supporting data, developed non-pregnant adult models in Simcyp and extended them to pregnancy.
The four approaches used ex vivo human placental perfusion, quantitative structure–property relationships, a Caco-2 permeability-based in vitro–in vivo correlation, or midazolam-based calibration combined with Caco-2 scaling. Model performance was checked against maternal plasma and umbilical-cord concentrations.
Maternal Agreement Did Not Resolve the Fetal Prediction
Adult AUC and peak-concentration predictions were within 0.5- to 2-fold of observed values, and most maternal concentration measurements fell within the simulated 90% prediction intervals. Fetal prediction remained less consistent.
The study defined predictive success as more than half of a drug’s observed umbilical-cord concentrations lying within the simulated 90% prediction interval. Success rates were 33.3% for placental perfusion, 28.6% for midazolam-calibrated Caco-2 scaling, 22.2% for the structure–property approach and 14.3% for the Caco-2 correlation.
The comparison identifies placental-clearance estimation as a major uncertainty in fetal exposure modeling. Stronger agreement for maternal concentrations coexisted with substantially lower success rates for fetal predictions.