Table 1.

Summary of strengths and limitations of innovative approaches to studying drugs during pregnancy and lactation

Approach	Strengths	Limitations
In vitro placenta (use of BeWo, Jeg-3, Jar, ACH-3P) In vitro lactation (use of a mouse mammary epithelial cell culture model)	Facilitate the study of placental influx and efflux transport systems, active, passive and facilitated diffusion, and drug metabolism in the placenta Facilitate the study of the directionality of passive drug transport and milk-to-plasma ratio of drugs, a critical parameter in predicting breastfed infant drug exposure through breastmilk	Limited in their ability to mimic the structure and critical physiologic functions of the human placenta; can only be used with limitations. Individual cells do not express the full portfolio of placental drug transporters (e.g. OCT3) No human in vitro systems are available yet due to the short lifespan
Ex vivo (isolated placental cotyledon) studies
1. Open and closed ex vivo circuit systems 2. Placenta on-a-chip models (based on BeWo cells)	Facilitates the understanding of mechanisms of transplacental transfer of drugs. Closed circuit systems are more physiologic, as both maternal and fetal perfusate are recirculated Replicates placental architecture and physiology, and enables precise prediction of drug transport across the placenta	Difficult to study preterm placentas, replicate placental studies, and maintain placental viability/structure over extended periods of time (days–weeks) Establishing cell-lined microfluid channels and maintaining a stable matrix between channels can be time-consuming. Individual cells do not express the full portfolio of placental drug transporters (e.g. P-glycoprotein, OCT3)
In silico
1. Population pharmacokinetic modeling 2. Maternal PBPK models 3. Fetal PBPK models 4. Lactation PBPK models	Allow the simultaneous use of sparse sampling methods and multiple covariates to explain intrasubject variability Allow prediction of maternal drug exposure, mode of action, drug–drug interactions and food–drug interactions Enable more precise estimation of fetal drug exposure by including amniotic fluid and fetal organs into PBPK models Allow prediction of infant drug exposure through breastmilk, drug–drug interactions and food–drug interactions	Involves complex data and analytical techniques; difficult to incorporate fetal drug PK as direct sampling is unethical Involves complex data and analytical techniques; not efficient at predicting fetal drug exposure Involves complex data and analytical techniques; current PBPK models do not account for placental transporters Mostly a milk-to-plasma ratio prediction component is missing
In vivo
1. Opportunistic pregnancy/lactation PK studies 2. Dedicated pregnancy and lactation PK studies 3. Microdosing pregnancy studies 4. Short-course studies in pregnancy and lactation (targeted PK studies)	Provides pregnancy- and lactation-specific PK data at the same time as in non-pregnant adults Provides expedient pregnancy- and lactation-specific PK data Subtherapeutic doses enough to allow cellular responses to be studied; have been identified as minimal risk Allows for targeted PK studies of sustained-release medications over extended periods of time (days–weeks)	Can prolong duration of the trial if pregnant and lactating women enroll at a slower pace than non-pregnant adults Direct comparison with non-pregnant adults can be difficult No direct clinical benefit to patients as doses of medications are subtherapeutic. Assumption of linearity in PK of drugs Targeted PK can miss unexpected release characteristics of medications (e.g. dose dumping)

BeWo, Jeg-3, and Jar are human choriocarcinoma cell lines, while ACH-3P is a first-trimester trophoblastic cell line; all are used for in vitro drug development

PK pharmacokinetic, PBPK physiologically based pharmacokinetic