2019 PharmSci 360
Although FDA recommends clinical lactation studies to assess the drug transfer into breast milk if the drug is used by women of reproductive age, there are challenges in conducting these studies such as difficulty in enrolling lactating women. To fill the knowledge gap, we utilized in vitro-in vivo extrapolation to predict human milk/plasma (M/P) concentration ratios of 71 drugs. By incorporating ionization fractions, protein/lipids binding in human plasma and milk, and Caco-2 cell bidirectional permeability data into the IVIVE model, the M/P ratios can be predicted with an absolute prediction fold error (APFE) of 1.62 fold and 2.48 fold for passive-diffusion drugs (N = 42) and transporter substrates (N = 29), which are superior to previously reported methods (pH partition method 2.75 fold, Meskin’s method 3.13 fold, phase distribution method 2.37 fold, and Koshimichi method 2.76 fold). The IVIVE method can be used to support lactation risk assessment for passive-diffusion drugs.