Advertisement
Research ArticleArticle

Development of a Novel Maternal-Fetal Physiologically Based Pharmacokinetic Model I: Insights into Factors that Determine Fetal Drug Exposure through Simulations and Sensitivity Analyses

Zufei Zhang, Marjorie Z. Imperial, Gabriela I. Patilea-Vrana, Janak Wedagedera, Lu Gaohua and Jashvant D. Unadkat
Drug Metabolism and Disposition August 2017, 45 (8) 920-938; DOI: https://doi.org/10.1124/dmd.117.075192

Abstract

Determining fetal drug exposure (except at the time of birth) is not possible for both logistical and ethical reasons. Therefore, we developed a novel maternal-fetal physiologically based pharmacokinetic (m-f-PBPK) model to predict fetal exposure to drugs and populated this model with gestational age–dependent changes in maternal-fetal physiology. Then, we used this m-f-PBPK to: 1) perform a series of sensitivity analyses to quantitatively demonstrate the impact of fetoplacental metabolism and placental transport on fetal drug exposure for various drug-dosing regimens administered to the mother; 2) predict the impact of gestational age on fetal drug exposure; and 3) demonstrate that a single umbilical venous (UV)/maternal plasma (MP) ratio (even after multiple-dose oral administration to steady state) does not necessarily reflect fetal drug exposure. In addition, we verified the implementation of this m-f-PBPK model by comparing the predicted UV/MP and fetal/MP AUC ratios with those predicted at steady state after an intravenous infusion. Our simulations yielded novel insights into the quantitative contribution of fetoplacental metabolism and/or placental transport on gestational age–dependent fetal drug exposure. Through sensitivity analyses, we demonstrated that the UV/MP ratio does not measure the extent of fetal drug exposure unless obtained at steady state after an intravenous infusion or when there is little or no fluctuation in MP drug concentrations after multiple-dose oral administration. The proposed m-f-PBPK model can be used to predict fetal exposure to drugs across gestational ages and therefore provide the necessary information to assess the risk of drug toxicity to the fetus.

Introduction

Fetal exposure to drugs has become increasingly common. This can be attributed to the rising use of therapeutic drugs among pregnant women (Mitchell et al., 2011) as pre-existing maternal conditions (e.g., epilepsy, asthma) or conditions that developed during pregnancy (e.g., gestational diabetes and hypertension) must be treated to ensure the health and welfare of the mother and therefore her fetus. Sometimes, it is the unborn child that is the target of the treatment [e.g., to prevent maternal-fetal human immunodeficiency virus (HIV) transmission] (McGowan and Shah, 2000). Consequently, the ability to quantitatively evaluate fetal exposure to drugs and risk of toxicity, not only at term but also earlier during pregnancy when the fetus is most vulnerable to teratogens, is needed.

Unlike the general population, fetal exposure to drugs ingested by the pregnant mother cannot be readily studied prior to birth for ethical and logistical reasons. Even at the time of birth, assessment of fetal exposure to drugs is limited to a single cord plasma concentration measurement and reported as the umbilical vein (UV)/maternal plasma (MP) drug concentration ratio. As shown here, in most clinical scenarios, this UV/MP ratio does not reflect the extent of fetal drug exposure relative to that in the mother. Nor does this ratio provide information on fetal drug exposure over time [i.e., fetal plasma area under the curve (AUCf)] or maximum fetal plasma drug concentration (Cmax,f) that often drives drug efficacy and/or toxicity in the fetus. Further, since cord blood sampling is limited to term, fetal drug exposure during early gestation remains unknown. To overcome these gaps in knowledge, mechanistic understanding of the determinants of fetal drug exposure and noninvasive approaches to predict fetal exposure across gestational ages (GAs), such as physiologically based pharmacokinetic (PBPK) modeling and simulation, are needed.

Although a number of attempts to develop fetal PBPK models have been made (Clewell et al., 2007; Yoon et al., 2011; Loccisano et al., 2013), upon closer examination, except for a recently published model (De Sousa Mendes et al., 2016, 2017), the majority of these models are not intended to predict fetal exposure to therapeutic agents prescribed to pregnant women. Their limitations include the following: 1) incomplete inclusion of pregnancy-caused changes in maternal and fetal physiology; 2) not accounting for the alterations in maternal drug disposition; and 3) exclusion of fetal body compartments important for the disposition of therapeutic drugs. Recently, our laboratory has successfully refined and verified a mechanistic maternal pregnancy PBPK model (m-PBPK) that can predict the maternal disposition of drugs cleared by one or more cytochrome P450 enzymes during pregnancy (Ke et al., 2012, 2013, 2014). However, this model contains only a lumped tissue compartment representing the placenta and fetus. Therefore, the goals of the current investigation were as follows: 1) to develop a maternal-fetal PBPK (m-f-PBPK) model by incorporating a physiologically relevant fetal-PBPK into our previously verified m-PBPK; 2) to quantitatively demonstrate the impact of fetoplacental metabolism and placental transport on fetal drug exposure; 3) to quantitatively predict the impact of GA on fetal drug exposure; and 4) to show that the UV/MP ratio after a single dose or after multiple dosing (even at steady state) does not necessarily represent the extent of fetal drug exposure.

Materials and Methods

Model Structure and General Assumptions.

Briefly, an m-f-PBPK model was built using MATLAB (R2014b; MathWorks, Natick, MA) by adding the placenta, the amniotic fluid compartment, and fetal organs important in drug disposition (e.g., liver and kidney) and distribution (e.g., brain) (Fig. 1) to our verified m-PBPK model (Ke et al., 2012, 2013, 2014). The remaining fetal organs were lumped into a single compartment referred to as “rest of body.” Our model also accounted for the marked differences in fetal circulation compared with that in the mother (Polin et al., 2004). For instance, it is the venous blood that carries oxygenated blood (via the UV) from the placenta to the fetus. The majority of this flow bypasses the fetal liver through the ductus venosus before perfusing fetal tissues.

Fig. 1.
Fig. 1.

Schematic diagram of the full m-f-PBPK model. Solid arrows indicate tissue blood flows, whereas dashed arrows indicate clearances. f/F, fetal; PD, passive diffusion; M, maternal; P, placental; A, amniotic fluid; met, metabolism; renal, renal excretion; reabsorp, amniotic fluid swallowing. CLPF/FP and CLPM/MP represent unidirectional transporter-mediated clearances.

General model assumptions include the following:

  • The bidirectional maternal-placental and fetal-placental unbound transplacental passive diffusion clearances across the placenta are equal and always present.

  • For a given drug, the magnitude of CLPD,u is directly proportional to the placenta villous surface area, which increases with GA.

  • The UV plasma drug concentration represents the systemic fetal plasma venous drug concentration.

  • Fetal renal clearance is negligible during the first 20 weeks of gestation (Polin et al., 2004). After week 20, it consists of only glomerular filtration clearance, which can be estimated from fetal plasma protein binding and inulin clearance estimated in preterm (weeks 23–40) and term neonates (within first 14 days of life).

  • Compared with fetal swallowing, the movement of amniotic fluid between the amniotic sac and maternal circulation is negligible (Gilbert and Brace, 1993). Therefore, fluid transfer between these two compartments was considered to be zero.

Collection and Analyses of Fetal Physiologic Parameters.

To populate the m-f-PBPK model with fetal physiologic parameters, a systematic literature search was carried out using PUBMED to obtain these parameters (Table 1). The search strategy was aimed to identify cohort studies whereby the parameters of interest was longitudinally examined during gestation. Data from the control arm of case-control studies and healthy subjects of cross-sectional studies were considered for inclusion. Other inclusion criteria included the following: 1) human; 2) uncomplicated singleton pregnancies; and 3) otherwise uneventful pregnancies with conditions thought not to affect the parameter of interest (e.g., preterm birth data were used to estimate fetal renal function). If the data were not tabulated and only graphs were present, individual data points were digitized using Digitizer (a free MATLAB tool available on http://www.mathworks.com/matlabcentral/). When multiple qualified studies were available for a physiologic parameter of interest, data were pooled, stratified by GA (measured in weeks from the first day of the last menstrual cycle), and summarized using the approach previously published (Abduljalil et al., 2012). For a given GA, the overall sample size weighted mean parameter value Embedded Imagefrom different studies was calculated as follows: Embedded Image, where Embedded Imageis the number of subjects in the ith study and Embedded Imageis the mean value from that study. In cases whereby both mean values and variability (S.D. or S.E.) of various GA groups were provided, overall S.D. was calculated as Embedded ImageEmbedded Image, where Embedded Imageis the S.D. in the ith study and, Embedded Imageis the number of subjects in the ith study, and Embedded Image is the mean value from that study. When individual measurements associated with a given GA were not available, the average values computed from formulae across publications were used instead.

View this table:
TABLE 1

GA-dependent key fetal and selected maternal physiologic parameters

Data analysis was performed using Excel (2010; Microsoft, Redmond, WA). In general, polynomial, exponential, or power function equations were chosen to describe the longitudinal changes in parameters during development. The choice of the polynomial degree was determined by fitting various polynomials to data using nonlinear regression. If a higher order of polynomial equation did not reduce R2 value and/or if it departed from the original data in comparison with a lower degree by visual check, then the lower one was chosen. Exponential and power equations were chosen to describe parameters when polynomials did not adequately fit rapidly time-varying parameters.

Sensitivity Analyses to Identify Key Determinants of Fetal Drug Exposure.

To identify the quantitative impact of key factors (e.g., fetal metabolic clearance) that influence maternal-placental/fetal plasma concentration-time (C-T) profiles of a drug at term (week 40), we conducted a series of simulations of two hypothetical drugs, X and Y (Table 2), using our newly developed m-f-PBPK model. Drug X was designed to be a neutral compound (e.g., the HIV nucleoside drugs zidovudine and didanosine, which are predominately un-ionized at physiologic pH) with intermediate permeability across the placenta and minimal plasma protein binding. Therefore, all the variables and parameters discussed here for drug X should be read as unbound values. A drug with these characteristics was chosen to quantitatively illustrate the impact of fetoplacental metabolism and placental transport on fetal exposure to drugs. Drug Y was designed to represent highly lipophilic, neutral drugs with high permeability across the biologic membrane. It was significantly bound to plasma albumin. Consequently, its maternal-fetal distributional equilibrium was affected by differences in maternal versus fetal plasma albumin concentrations. Relevant examples include protease inhibitors and many biopharmaceutics classification system class I and II drugs. These drugs are cleared by cytochrome P450 enzymes that have altered activity during pregnancy (Isoherranen and Thummel, 2013).

View this table:
TABLE 2

Summary of drug-specific parameters used in the simulations at term (week 40)

Additional assumptions made for the hypothetical drugs X and Y are as follows:

  • Drug X is neutral, follows linear kinetics, and has negligible binding in the MP and fetal plasma and in the placenta. Therefore, all concentrations and clearances of drug X represent their corresponding unbound values.

  • Drug Y is neutral, exhibits linear kinetics, and binds to plasma albumin. Its binding in the placenta tissue is the same as that in the MP.

  • Maternal absorption of drug X or Y is first order and does not change during pregnancy [i.e., first-order absorption rate constant (ka), fraction absorbed (Fa), and fraction escaping gut metabolism or intestinal bioavailability (Fg)].

  • Maternal and fetal tissue-to-plasma partition coefficients are identical for both drugs and remain constant throughout pregnancy.

  • Except where indicated, the fetal renal clearance of drug X or Y is negligible.

  • Drug X or Y swallowed by the fetus (i.e., the amniotic fluid) is instantly and completely absorbed from the fetal intestine and is not metabolized there.

Although some of the above assumptions were made to simplify the simulations (e.g., neutrality thus zero ionization; unaltered maternal absorption during pregnancy), others were made (e.g., fetal renal secretion of drugs) because these values cannot be determined.

Except where indicated, the simulations were conducted using our m-f-PBPK model, where only one parameter was changed at a time (Tables 3 and 4) at week 40.

View this table:
TABLE 3

Drug X clearance values used in various scenarios

View this table:
TABLE 4

Drug Y clearance values used in various scenarios

Effect of GA on the Fetal-MP Pharmacokinetics.

We simulated the impact of GA on exposure of the maternal-fetal unit to drug X or Y under various scenarios (Table 5). Week 20 and week 40 were chosen, respectively, to represent the GA when fetal skin keratinization begins and when cord blood sampling is possible. Fetal metabolic clearance of drug X, when present, was assumed to be directly proportional to fetal body volume as the metabolism of HIV nucleoside drugs (i.e., phosphorylation) occurs throughout the body. Where invoked, placental apical efflux clearance (CLPM) was assumed to be mediated by P-glycoprotein (P-gp), and the magnitude of this clearance was assumed to be proportional to the expression of P-gp in the placenta. At term, P-gp–mediated CLPM was arbitrarily set at 20% of transplacental passive diffusion clearance (CLPD). In the absence of placental P-gp expression data at week 20, we assumed that placental P-gp expression decreases by 5-fold based on our first-trimester data on placental P-gp expression (Mathias et al., 2005). This change in expression was then scaled up to the whole placenta based on change in the GA-dependent placental volume (Abduljalil et al., 2012).

View this table:
TABLE 5

GA-dependent changes in the fetal metabolic (CLf0) and placental efflux (CLPM) clearances of drugs X and Y

Results

Fetal Physiologic Parameters.

The time-variant fetal physiologic parameters used to populate the m-f-PBPK model show that the GA-dependent changes in fetal physiologic parameters are pronounced (Table 1). For example, umbilical venous blood flow (i.e., fetal placental blood flow) increases by approximately 6.2-fold (from 3.3 to 20.2 l/h) from week 20 to week 40. Some fetal physiologic values change with GA in an opposite direction to the corresponding values in the mother. For example, the fetal plasma albumin concentrations increase with GA, whereas the reverse is true for the mother.

Impact of Maternal Metabolism and Placental Passive Drug Permeability on Fetal Exposure to Drug X or Y in the Absence of Placental/Fetal Metabolism or Placental Transport.

As expected, after continuous intravenous infusion of drug X or Y, both steady state (steady stateinf) maternal venous plasma (MP) and fetal venous plasma (fetal plasma) concentrations and the time to reach steady state after an intravenous infusion (steady-stateinf) were inversely dependent on the maternal systemic clearance (CLm0) when Vss was held constant (Fig. 2, a and b; Fig. 3, a and b). In contrast, for both drugs, the MP C-T profile, but not the fetal plasma C-T profile, was independent of the CLPD of the drug as were the maternal and fetal steady stateinf plasma concentrations (Fig. 4a; Fig. 5a). Although total fetal plasma steady-stateinf concentrations of drug Y were consistently higher than those in the mother (Fig. 3, a and b; Fig. 5, a and b), for both drugs the corresponding unbound steady-stateinf UV/MP ratio remained at unity irrespective of the magnitude of CLm0 or CLPD values (Fig. 2b; Fig. 4b for drug X; data not shown for drug Y). However, for both drugs, the time to reach fetal steady-stateinf plasma concentration was prolonged with lower CLm0 or CLPD values (Fig. 2b; Fig. 3b; Fig. 4b; Fig.. 5b). Of note, the simulated UV/MP ratio of drug X (which is not bound to plasma proteins) or Y (after correcting for plasma protein binding) matched those predicted by our steady-stateinf model (see Supplemental Material).

Fig. 2.
Fig. 2.

Impact of changes in CLm0 on fetal and maternal drug X plasma concentration and UV/MP ratio. Changes in CLm0 of drug X significantly influenced maternal-fetal drug X plasma C-T profiles at week 40. (a) After infusion (16.7 mg/h, i.v.) at week 40, decreasing CLm0 from 45 l/h (red) to 4.5 l/h (blue) increased the steady-state maternal (solid lines) and fetal (dashed lines) plasma concentration of drug X as well as the time to reach steady state. Inset shows the curves on a semilogarithmic scale. (b) The corresponding UV/MP ratios indicate that at steady stateinf these ratios do not change with changes in CLm0 (45 l/h, red; 4.5 l/h, blue). (c) After a single oral dose (400 mg), increasing CLm0 from 4.5 l/h (blue) to 45 l/h (red) resulted in lower MP drug concentrations (solid lines) and subsequently lower fetal plasma drug X concentrations (dashed lines). (d) Corresponding changes in UV/MP ratio indicate that higher CLm0 (red) led to greater time-dependent fluctuations in the UV/MP ratio as well as a lager UV/MP ratio at distributional equilibrium. (e) Under a multiple oral dosing regimen (133.3 mg; τ = 8 hours) lower CLm0 (blue) not only prolonged the time to reach steady state but also resulted in greater extent of drug accumulation. (f) In addition, lower CLm0 (blue) led to fewer fluctuations in UV/MP ratio within a dosing interval compared with higher CLm0 (red). When the dosing interval was increased (400 mg; τ = 24 hours), the effect of reduction in CLm0 on drug accumulation (g) or variation in UV/MP ratio (h) was dampened. In panels (e) and (g), the predicted F/M AUC ratio remained at unity despite the changes in CLm0. See Table 3 for the clearance values used in these simulations.

Fig. 3.
Fig. 3.

Impact of changes in CLm0 on fetal and maternal drug Y plasma concentration and UV/MP ratio. Changes in CLm0 of drug Y significantly influenced maternal-fetal drug Y plasma C-T profiles at week 40. After infusion (0.625 mg/h, i.v.) at week 40, a 10-fold decrease in maternal hepatic intrinsic clearance (from 3327 to 332.7 l/h) resulted in a decrease in CLm0 from 43 l/h (red) to 12 l/h (blue). (a) The resultant steady-stateinf maternal (solid lines) and fetal (dashed lines) plasma concentration of drug Y as well as the time to reach steady state increased. (b) The corresponding UV/MP ratios indicate that at steady stateinf these ratios stay at 1.2 and do not change with varying CLm0 values (43 l/h, red; 12 l/h, blue). (c) After a single oral dose (15 mg), increasing CLm0 from 12 l/h (blue) to 43 l/h (red) resulted in lower MP drug concentrations (solid lines) and subsequently lower fetal plasma drug X concentrations (dashed lines). (d) Corresponding changes in UV/MP ratio indicate that higher CLm0 (red) led to greater fluctuations in the UV/MP ratio as well as a larger UV/MP ratio at distributional equilibrium. (e) After multiple oral doses (3.75 mg; τ = 4 hours), lower CLm0 (blue) not only prolonged the time to reach steady state but also resulted in much greater drug accumulation. (f) In addition, lower CLm0 values (blue) led to fewer fluctuations in UV/MP ratio (within a dosing interval) when compared with higher CLm0 values (red). When the dosing interval was increased (15 mg; τ = 24 hours), the effect of reduction in CLm0 on dose accumulation (g) or variations in UV/MP ratio (h) was dampened. The inset shows the curves on a semilogarithmic scale. In panels (e) and (g), the predicted unbound F/M AUC ratio remained at unity despite the changes in CLm0. See Table 4 for the clearance values used in these simulations.

Fig. 4.
Fig. 4.

Impact of changes in CLPD on fetal and maternal drug X plasma concentration and UV/MP ratio. Changes in CLPD of drug X significantly influenced the fetal (dashed lines) but not the maternal (solid lines) drug X plasma C-T profile at week 40. (a) After infusion (16.7 mg/h, i.v.) at week 40, decreasing CLPD from 18 l/h (red) to 1.8 l/h (blue) did not affect the maternal (the red and blue solid lines overlap) or fetal steady-stateinf plasma concentrations of the drug. (a and b) The corresponding UV/MP ratios indicate that at steady stateinf these ratios do not change with alterations in CLPD (18 l/h, red; 1.8 l/h, blue), but the time to reach the steady-state ratio was prolonged. (c) After a single oral dose (400 mg), increasing CLPD from 1.8 l/h (blue) to 18 l/h (red) significantly modified the shape of fetal plasma C-T curves (dashed lines) but not MP drug concentrations (solid lines; blue and red lines overlap). (d) Corresponding changes in UV/MP ratio indicate that higher CLPD values (red) not only shortened the time to reach distributional equilibrium but also reduced distributional equilibrium UV/MP ratio compared with lower CLPD (blue). After multiple oral doses (400 mg; τ = 24 hours), alterations in CLPD did not affect the MP drug X C-T curve but significantly changed the shape of fetal plasmas in the drug X C-T profile. (e) Higher CLPD values (red) resulted in greater fluctuations in fetal plasma drug X concentration within a dosing interval compared with lower CLPD values (blue). (f) In contrast, higher CLPD (red) produced fewer fluctuations in UV/MP ratio within a dosing interval compared with lower CLPD (blue). Insets in (d) and (f) show the F/M AUC ratios at lower CLPD (blue) or higher CLPD (red). The predicted F/M AUC ratio remained at unity despite changes in CLPD after single (c) or multiple (e) oral doses of drug X. See Table 3 for the clearance values used in these simulations.

Fig. 5.
Fig. 5.

Impact of changes in CLPD on fetal and maternal drug Y plasma concentration and UV/MP ratio. Changes in CLPD of drug Y significantly influenced drug Y fetal (dashed lines), but not maternal (solid lines), plasma C-T profile at week 40. (a) After infusion (0.625 mg/h, i.v.) at week 40, decreasing CLPD values from 22.5 l/h (red) to 2.25 l/h (blue) did not affect the maternal (the red and blue solid lines overlap) or fetal steady stateinf plasma concentration of the drug. (b) The corresponding UV/MP ratios indicate that at steady-stateinf these ratios do not change with alterations in CLPD (22.5 l/h, red; 2.25 l/h, blue), but the time to reach the steady-state ratio was prolonged. (c) After a single oral dose (15 mg), decreasing CLPD from 22.5 l/h (red) to 2.25 l/h (blue) significantly modified the shape of fetal plasma C-T curve (dashed lines) but not MP drug concentrations (solid lines; blue and red lines overlap). (d) Corresponding changes in UV/MP ratio indicate that higher CLPD values (red) not only shortened the time to reach distributional equilibrium but also reduced distributional equilibrium UV/MP ratio compared with lower CLPD (blue). Under a multiple oral dosing regimen (15 mg; τ = 24 hours), alterations in CLPD values did not affect the MP drug Y C-T curve but significantly changed the shape of the fetal plasma drug Y C-T profile. (e) Higher CLPD values (red) resulted in greater fluctuations in fetal plasma drug Y concentration within a dosing interval compared with lower CLPD values (blue). (f) In contrast, higher CLPD values (red) produced fewer fluctuations in UV/MP ratio within a dosing interval compared with lower CLPD values (blue) as transplacental distributional equilibirum of drug Y was quickly attained after an oral dose. Insets in (d) and (f) show the unbound F/M AUC ratios at lower CLPD (blue) or higher CLPD (red). The predicted unbound F/M AUC ratio remained at unity despite changes in CLPD after single (c) or multiple (e) oral doses of drug Y. See Table 4 for the clearance values used in these simulations.

After a single oral dose of drug X or Y, fetal drug exposure (AUCf) was inversely proportional to CLm0 (Fig. 2c; Fig. 3c) but independent of CLPD (Fig. 4c; Fig. 5c), although the fetal plasma time to reach maximum drug concentration (tmax) and Cmax,f were affected by both clearances. As expected, their UV/MP ratios varied over time until maternal-fetal distributional equilibrium was achieved (Fig. 2d; Fig. 3d; Fig. 4d; Fig. 5d). Interestingly, for both drugs (but drug X more than drug Y), its distributional equilibrium UV/MP ratio was greater than the expected value of unity in the absence of fetoplacental metabolism or drug transport (Supplemental eq. 1). For drug Y, this deviation from unity persisted after correcting for plasma protein binding (data not shown). Furthermore, for both drugs, the deviation from unity became significantly dampened with decrease in CLm0 (Fig. 2d; Fig. 3d) or increase in CLPD (Fig. 4d; Fig. 5d).

After multiple oral doses, as expected, the time to reach steady state, and the accumulation and fluctuations in steady-state MP drug concentrations were inversely proportional to CLm0 (Fig. 2, e and g; Fig. 3, e and g) but were independent of changes in CLPD (Fig. 4e; Fig. 5e). However, higher CLPD resulted in greater fluctuations in steady-state fetal plasma drug concentrations within a dosing interval (Fig. 4e; Fig. 5e). Overall, the effect of CLm0 or CLPD on the UV/MP ratio remained the same after single or multiple oral dosing.

Impact of Dosing Interval (τ) on UV/MP Ratio of Drug X or Y in the Absence of Placental/Fetal Metabolism or Placental Transport.

As expected, shortening τ resulted in greater drug accumulation (Fig. 2, e versus g; Fig. 3, e versus g) but fewer fluctuations in the UV/MP ratio within a dosing interval (Fig. 2, f versus h; Fig. 3, f versus h). As in the above scenarios, the observed distributional equilibrium UV/MP ratio of drug X and Y remained higher than the expected unbound steady-stateinf UV/MP ratio of 1.0, and the extent of this deviation was inversely related to CLm0 (Fig. 2, f and h; Fig. 3, f and h).

Impact of Placental Drug Transport, Fetal Metabolism, and Placental Metabolism on Fetal Exposure to Drug X or Y.

Overall, variations in these pathways (Table 3) produced significant changes in fetal plasma C-T profiles of drug X without affecting maternal pharmacokinetics (PK) (data not shown). As expected, after a single oral dose of 400 mg of drug X, fetal plasma concentrations, Cmax,f, and placental concentrations (data not shown) of drug X were inversely correlated with its metabolism in the placenta [placental metabolic clearance (CLp0)] (Fig. 6a1), in the fetus [fetal metabolic clearance (CLf0)] (Fig. 6b1), or both (Fig. 6c1). Irrespective of the magnitude of CLp0 relative to CLPD, the effect of change in CLp0 on the magnitude of reduction in AUCf and placental AUC (AUCp) were identical (Fig. 6a2). In contrast, an increase of the same magnitude in fetal metabolism (CLf0) resulted in a greater reduction in AUCf compared with AUCp (Fig. 6b2). For example, when CLp0 equaled CLPD, AUCp and AUCf were both reduced by 50%, whereas when CLf0 equaled CLPD, a greater reduction in AUCf was seen compared with AUCp (67% versus 33%) (Fig. 6, a2 versus b2). When both metabolic processes were present (CLp0 and CLf0) and equal to CLPD, the reduction in the Cmax,f (Fig. 6c1), AUCp (60%), and AUCf (80%) was even greater (Fig. 6c2). The addition of placental apical uptake clearance (CLMP) or CLPM on the apical side of the placenta altered fetoplacental exposure to drug X in opposite directions (Fig. 6, d versus e). Although increasing CLMP resulted in higher fetal plasma drug concentrations and Cmax,f (Fig. 6d1) as well as increased fetal/MP (F/M) and placental/MP (P/M) AUC ratios (Fig. 6d2), increasing CLPM lowered Cmax,f (Fig. 6e1) and reduced F/M and placental/MP (P/M) AUC ratios (Fig. 6e2). The predicted F/M and P/M AUC ratios quantitatively matched those predicted by our steady-stateinf model (Supplemental eqs. 1 and 2, respectively). However, even under fetal-maternal distributional equilibrium the UV/MP ratio (Fig. 6, a3–e3) deviated from its respective steady-stateinf F/M drug concentration ratio and therefore did not represent the F/M AUC ratio.

Fig. 6.
Fig. 6.

Impact of changes in feto-placental metabolism and placental transport on fetal plasma and placental drug X concentration, P/M AUC, F/M AUC, and UV/MP ratios after a single 400-mg oral dose of drug X. Changes in fetoplacental clearance pathways differentially impacted fetal exposure to drug X, P/M plasma AUC ratio (hatched bar), F/M plasma AUC ratio (solid bar), and the UV/MP ratio after a single 400-mg oral dose of drug X at week 40. The CLPD of drug X was held at 1.8 l/h. The clearance pathway variance is indicated at the extreme left of the first panel of each row. The indicated clearance was set at 0, 0.18, 0.9, or 1.8 l/h, respectively (0%, 10%, 50%, or 100% of CLPD; black, red, green, and blue lines, respectively). Other than CLMP (d1), increasing any of the indicated clearance pathways resulted in lower fetal plasma drug X concentrations (a1–c1 and e1). When CLf0 was present, the resultant F/M plasma AUC ratio was lower (b2 and c2) than when only CLP0 was present (a2). In all cases, the predicted UV/MP ratio at distributional equilibrium was significantly greater than its steady-stateinf value (Supplemental eq. 1). The UV/MP ratio at distributional equilibrium decreased with an increase in CLp0, CLf0,CLf0 plus CLp0, or CLPM (a3, b3, c3, and e3, respectively) and increased with an increase in CLMP (d3). See Table 3 for the clearance values used in these simulations.

Overall, after a single oral dose of drug Y, within the test range, variations in the same set of drug Y clearance pathways (Table 4) produced similar changes in fetal plasma drug concentrations (Fig. 7, a1–e1) without altering the MP C-T curve of drug Y (data not shown). As was the case with drug X, both P/M and F/M AUC ratios decreased as these clearance pathways became larger with the exception of CLMP, which was positively correlated with P/M and F/M ratios. After accounting for binding, the impact of fetal clearance was larger than that of placental clearance on F/M and P/M AUC ratios (Fig. 7, a2–e2). In addition, the predicted drug Y UV/MP ratio at distributional equilibrium (Fig. 7, a3–e3) was consistently higher than the expected steady-stateinf UV/MP ratios (Supplemental eq. 1).

Fig. 7.
Fig. 7.

Impact of changes in fetoplacental metabolism and placental transport on fetal plasma and placental drug Y concentrations, and P/M AUC, F/M AUC, and UV/MP ratios after a single 15-mg oral dose of drug Y. Changes in fetoplacental clearance pathways differentially impacted fetal exposure to drug Y, P/M plasma unbound AUC ratio (hatched bar), F/M plasma unbound AUC ratio (solid bar), and the UV/MP ratio after a single 15-mg oral dose of drug Y at week 40. The CLPD of drug Y was held at 22.5 l/h. The clearance pathway varied is indicated at the extreme left of the first panel of each row. The indicated clearance was set at 0, 2.25, 11.3, or 22.5 l/h, respectively (0%, 10%, 50%, or 100% of CLPD; black, red, green, and blue lines, respectively). Other than CLMP (d1), increasing any of the indicated clearance pathways resulted in lower fetal plasma drug X concentrations (a1–c1 and e1). When CLf0 was present, the F/M plasma unbound AUC ratio was lower (b2 and c2) than when only CLP0 was present (a2). In all cases, the predicted UV/MP ratio at distributional equilibrium was greater than its expected steady-stateinf value (Supplemental eq. 1). The UV/MP ratio at distributional equilibrium decreased with an increase in CLp0, CLf0, CLf0 plus CLp0, or CLPM (a3, b3, c3, and e3, respectively) and increased with an increase in CLMP (d3). See Table 4 for the clearance values used in these simulations.

Impact of GA on Fetal Disposition of Drug X.

Maternal and fetal plasma C-T profiles after a single 400-mg oral dose of drug X at weeks 20 and 40 were simulated using the m-f-PBPK model under various scenarios outlined in Table 3. GA significantly altered fetal plasma C-T profile while minimally affecting maternal PK (Fig. 8). Furthermore, the impact of GA on fetal exposure to drug X depended on the clearance mechanisms within the fetoplacental unit. In scenario 1 (Fig. 8, a and b), where drug X was assumed to passively diffuse across the placenta without placental drug transport or irreversible clearance in the fetoplacental unit (e.g., metabolism), despite a modest 16% decrease in Cmax.f, the F/M plasma AUC ratio remained at unity and did not change with advancing GA. In scenario 2 (Fig. 8, c and d), the addition of fetal metabolism produced significant reductions in both the Cmax,f and AUCf, resulting in F/M AUC ratio of 0.48 and 0.50 at weeks 20 and 40, respectively. In scenario 3 (Fig. 8, e and f), where P-gp–mediated CLPM was assumed to be 20% of CLPD with no fetal metabolism, advancing GA from week 20 to week 40, and the associated decrease in CLPM resulted in substantial changes in the shape of the fetal C-T curve as well as in increase in Cmax,f (by 118.5%) and fetal AUC (by 3.6-fold). Finally, in scenario 4 (Fig. 8, g and h), the combination of both fetal metabolism and placental efflux resulted in the lowest Cmax,f and AUCf at both GAs.

Fig. 8.
Fig. 8.

Impact of GA on maternal and fetal drug X plasma concentration and F/M AUC ratio after a single 400-mg oral dose of drug X at week 20 (red) or week 40 (blue). The effect of GA on fetal exposure to drug X varies with fetoplacental clearance mechanisms involved. Maternal (solid) and fetal (dashed) C-T profiles as well as F/M plasma AUC ratios were simulated after a single 400 mg oral dose of drug X at week 20 (red) and week 40 (blue) under different scenarios. In scenario 1, where no irreversible loss of drug X occurred in the fetoplacental unit, the advancement of GA did not significantly affect fetal-maternal drug disposition (a) or the F/M plasma AUC ratio of unity (b). In scenario 2, the addition of fetal metabolism (increased proportionally with the fetal body volume as GA increased) significantly reduced fetal plasma drug concentrations (c versus a) and resulted in an ∼50% decrease in F/M AUC plasma ratio at both GAs (d). In scenario 3, placental P-gp efflux clearance decreased with GA (based on reported P-gp expression and changes in placental volume with GA). Consequently, fetal exposure to drug X increased with GA, reflected by higher fetal plasma drug concentrations (e) and increased F/M plasma AUC ratios (f). Finally, in scenario 4, where both fetal metabolism and placental efflux were present, a further reduction in fetal exposure was observed at both GAs (g and h). See Table 3 and Table 5 for the clearance values used in these simulations.

Impact of GA on Fetal Disposition of Drug Y.

Overall, GA had a marked effect on maternal-fetal plasma C-T curves of drug Y. In scenario 1, where neither fetoplacental metabolism nor placental drug transport was present and maternal hepatic unbound intrinsic clearance remained independent of GA, increasing GA resulted in slightly lower MP concentrations of drug Y (Fig. 9a), as evidenced by a 19% reduction in MP AUC from week 20 to week 40. Over the same period, fetal plasma AUC increased significantly by 67% (Fig. 9b). This decrease in AUCm, in conjunction with the increase in AUCf, resulted in a pronounced ∼100% increase in F/M AUC ratio from week 20 to 40. However, after correcting for plasma protein binding, unbound F/M AUC ratios remained unity at both GAs (Fig. 9c). In scenario 2 when a 100% induction in maternal CLint,u,hep at week 40 relative to that at week 20 was assumed, the resultant higher CLm0 gave rise to lower the maternal plasma AUC (AUCm) and thus a 34% decrease in term AUCf (Fig. 9, d and e) compared with scenario 1. Of note, F/M AUC ratios, both total and unbound, were identical to those in scenario 1 (Fig. 9f) with unbound F/M AUC ratios being unity. Like drug X, the introduction of CLPM (20% of CLPD at term) in scenario 3 produced markedly lower fetal plasma drug concentrations and hence much smaller AUCf at week 20 compared with week 40 (Fig. 9h). Unbound F/M AUC ratios reduced to lower-than-unity values (∼0.2 and ∼0.8 at week 20 and 40, respectively) (Fig. 9i). Last, in scenario 4, the combination of placental efflux and increasing CLm0 with GA resulted in a pronounced difference in maternal-fetal plasma drug C-T curves between week 20 and week 40. The increase in CLm0 from week 20 to week 40 in GA led to significantly lower maternal drug Y concentrations and a greater than 50% reduction in AUCm at week 40. When compared with scenario 3, this scenario had much lower Cmax,f and AUCf values at week 40 (Fig. 9, j and k), whereas the F/M AUC ratios were of the same magnitude (Fig. 9l).

Fig. 9.
Fig. 9.

Impact of GA on maternal and fetal drug Y plasma concentration and F/M AUC ratio after a single 15-mg oral dose of drug Y at week 20 (red) and week 40 (blue). The effect of GA on fetal exposure to drug Y varies with the fetoplacental clearance mechanisms involved. Maternal (solid) and fetal (dashed) C-T profiles, maternal (solid) and fetal (hatched) plasma AUCs, as well as F/M [total (solid) and unbound (hatched)] AUC ratios were simulated after a single 15-mg oral dose of drug Y at week 20 (red) and week 40 (blue) under different scenarios. In scenario 1, where no irreversible loss of drug Y occurred in the fetoplacental unit under a constant maternal hepatic unbound intrinsic clearance (CLint,u,hep), the advancement of GA resulted in an increased fetal plasma AUC despite the decrease in MP AUC (a and b) but did not affect the F/M plasma AUC unbound ratio of unity (c). In scenario 2, the assumed 100% higher CLint,u,hep at week 40 compared with week 20 significantly decreased maternal-fetal plasma drug Y concentrations (hence, AUCs) at term (d versus a; e versus b), whereas F/M AUC plasma ratios at both GAs were identical to those in scenario 1 (f versus c). In scenario 3, placental P-gp–mediated efflux clearance decreased with GA. Consequently, fetal exposure to drug Y increased with GA, which is reflected by higher fetal plasma drug concentrations (g) and increased F/M plasma AUC ratios (i). Finally, in scenario 4, the increase in fetal exposure from week 20 to week 40 persisted in the presence of both a decrease in placental efflux and an increase in CLm0 with GA, but to a smaller extent compared with scenario 3 (k versus h). Note that F/M AUC plasma ratios at both GAs were identical to those in scenario 3 (l versus i). See Table 4 and Table 5 for the clearance values used in these simulations.

Discussion

Here we present a novel maternal-fetal PBPK model that, to our knowledge, allows for the first time the capability to predict the fetal disposition of pharmaceutical drugs at various GAs. The model incorporated the unique fetal vascular physiology and allows future incorporation of placental transport and metabolism within the fetoplacental unit (in progress in our laboratory). This model has been verified by us, at term, where the predicted and observed maternal and fetal plasma concentrations of theophylline and zidovudine were in a good agreement (Zhang and Unadkat, 2017). However, since such verification data can only be collected at term and does not speak to which factors might affect this ratio (after a single dose or at steady state), we have used this novel m-f-PBPK model to quantitatively demonstrate the factors that can affect this ratio and when this ratio can and cannot be used to estimate fetal exposure to drugs (Figs. 2–9). Although some of these results are obvious from fundamental pharmacokinetic principles, others are not. These unexpected results are highlighted below and summarized in Supplemental Table 2. Also, note that all fetoplacental clearance pathways below refer to clearance values after accounting for binding.

When Does UV/MP Ratio Reflect Fetal Drug Exposure?

Contrary to a widely held belief, the UV/MP ratio, even at distributional equilibrium, does not indicate fetal drug exposure relative to that in the mother (AUCf/AUCm) (Figs. 2–7). The only exception is when the drug is administered to steady state via an infusion or when MP concentrations, at steady state, do not fluctuate much after multiple oral administration situations that occur infrequently in the clinic. The same UV/MP ratio is often deemed to infer the mechanisms by which the drug crosses the placenta. A greater than unity ratio is often interpreted as accumulation of the drug in the fetal compartment (Else et al., 2011). In contrast, a ratio of less than unity is interpreted as low fetal exposure and is attributed to low maternal drug concentrations (Chappuy et al., 2004a,b), fetal metabolism (Ngan Kee et al., 2009), and/or placental efflux (Marzolini and Kim, 2005; Else et al., 2011). However, as shown by our simulations, such interpretations can be false. Although fetal/placental metabolism or placental transport processes cannot be discounted based on a single UV/MP ratio, such deviation from unity is more likely due to the time-dependent distributional kinetics of the drugs across the placenta.

In the absence of fetal/placental metabolism or placental transport, the unbound steady-stateinf UV/MP ratio and therefore the unbound F/M AUC ratio after single or multiple doses (assuming linearity) should be unity (Supplemental eq. 1). Indeed, it is (Figs. 2-7). In all other cases, the degree of deviation of this ratio from unity (sometimes by an order of magnitude) not only varied with time but also depended on the magnitude of CLPD (Figs. 4 and 5) and CLm0 (Figs. 2 and 3). The extent of this deviation decreased when the fetal compartment was allowed to rapidly equilibrate with the maternal compartment. Even in these cases where no active placental uptake was invoked, the UV/MP ratio at distributional equilibrium still exceeded unity (Fig. 2d, ∼1.7; Fig. 3d, 1.5). Although puzzling at first glance, these observations can be explained by the multicompartmental PK of drugs X and Y. During the postdistributive phase, the peripheral/central compartment drug concentration ratio increases as the central compartment clearance is increased or as the intercompartmental clearance is decreased (for details, see Supplemental Material). The above conclusions also hold true for multiple dosing regimens (Figs. 2–5). Our simulations are consistent with literature reports. For example, after the last maternal dose, the UV/MP ratio of zidovudine (half-life 1.1 hours) (Collins and Unadkat, 1989) range from 0.18 to 17.2 (Chappuy et al., 2004a), whereas that of theophylline (half-life ∼8 hours) is much less variable (from 0.98 to 1.59) (Ron et al., 1984). Clearly, in the case of zidovudine, interpreting UV/MP ratios of much greater than 1.0 as indicative of active maternal-fetal transport would be incorrect. Similarly, another frequently reported ratio, the amniotic fluid/UV drug concentration ratio, varied with time and did not reflect fetal drug exposure (see Supplemental Material).

Unbound F/M AUC Ratio Is Determined by the Magnitude of Placental Clearance Relative to Fetoplacental and/or Placental Transport Clearance.

Another common misconception about fetal drug exposure (i.e., AUCf) is that AUCf is mainly driven by MP drug concentrations and not by passive placental transfer and/or fetoplacental metabolism. The reasoning behind it is that fetal/placental clearances are thought to be minor compared with maternal clearance of the drug, due to the small size of the fetal liver [only about 9.5% of maternal liver weight at term (Abduljalil et al., 2012)] or the limited metabolic capacities of the placenta and the fetal liver. Therefore, it is often assumed that when active transport across the placenta is absent, the unbound F/M AUC ratio will approximate unity. Although MP drug concentrations do drive fetal plasma drug concentrations, the above reasoning about F/M AUC ratio is false because it fails to recognize an additional critical factor, CLPD. It is the ratio of these two clearances (i.e., the magnitude of fetoplacental clearances relative to CLPD and not relative to CLm0) that determines the unbound F/M AUC ratio. These determinants of fetal drug exposure, however, have been largely overlooked by others (Hill and Abramson, 1988; Marzolini and Kim, 2005; Myllynen et al., 2007; Bernick and Kane, 2012) and are discussed below with examples (Figs. 6 and 7).

For relative polar drugs with intermediate or low CLPD (e.g., drug X), the introduction of fetal/placental metabolism and/or placental transport can significantly alter fetal drug exposure if their magnitude is significant relative to CLPD (Fig. 6). It is important to stress here that the reference point is CLPD and not CLmo. Consistent with our simulations, due to fetal metabolism, the unbound steady-stateinf F/M drug concentration ratios of two predominantly polar dideoxynucleoside HIV drugs, didanosine and zalcitabine, were 0.48 and 0.63 in the chronically catheterized maternal-fetal macaques (Pereira et al., 1994; Tuntland et al., 1996). Furthermore, our simulations yielded another novel insight. The site of drug clearance (placental, fetal, or both) had a differential impact on AUCf and AUCp (Fig. 6; Supplemental eqs. 1 and 2). Because any drug taken by mother has to first traverse the placenta to reach fetal circulation, the placenta essentially behaves like the “fetal gut.” If metabolism or drug efflux/influx occurs only in the placenta, then the change in placental and fetal drug exposure will be identical and will depend on the degree of presystemic “first pass” (Fig. 6, a, d, and e). Instead, if metabolism (of equal magnitude) occurs in the fetus, this will result in a greater reduction in AUCf compared with AUCp (Fig. 6, b and c). In essence, metabolism in the fetus magnifies the reduction in fetal drug exposure caused by placental clearance. This “site effect” may have clinical implications as fetal toxicity can occur via toxicity to the placenta (e.g., formaldehyde) (Pidoux et al., 2015). In these cases, fetal toxicity cannot be readily ruled out even when fetal exposure to the toxin/drug is low. In contrast, when the CLPD of drugs is much greater than fetal/placental metabolism or placental transport (e.g., lipophilic drugs such as drug Y), the placenta becomes “transparent.” In other words, fetal and maternal compartments rapidly equilibrate. Therefore, AUCf will approach or equal AUCm after single-dose or multiple-dose administration of the drug (assuming linearity) (Fig. 7b). In reality, for lipophilic drugs with limited fetoplacental metabolic capacities, a negligible effect on AUCf is expected when CLf0 or CLp0 values are varied. However, even for such drugs, in the presence of significant fetal/placental metabolism and/or placental transport relative to CLPD, fetal drug exposure will be significantly altered (e.g., remifentanil) (Egan, 1995; Coonen et al., 2010). Several lipophilic HIV protease inhibitors that are P-gp substrates (e.g., ndinavir, ritonavir, and saquinavir) (Unadkat et al., 2004) have UV/MP ratios that are considerably lower than unity (Marzolini and Kim, 2005). However, as pointed out earlier, these non–steady-stateinf ratios should be interpreted with caution. Nevertheless, a role of placental P-gp efflux likely contributes to their low fetal exposure as these drugs are excellent substrates of P-gp (van der Sandt et al., 2001; McCormack and Best, 2014).

The Impact of GA on Fetal Drug Exposure.

The expression of fetoplacental drug-metabolizing enzymes and placental transporters are known to change with GA (Myllynen et al., 2009). For this reason, fetal drug exposure (with no change in maternal dosage regimen) is likely to change with GA (Figs. 8 and 9). But these changes are not intuitive. When drug X passively diffused across the placenta without being transported or metabolized in the fetoplacental unit (Fig. 8, a and b), the progression of gestation did not affect AUCf because the latter was solely driven by AUCm, which remained virtually the same from week 20 to week 40. In scenario 2, where only fetal metabolic clearance was present, AUCf was significantly reduced and appeared nearly equal between week 20 and week 40 (Fig. 8d). This is because the CLf0/CLPD ratio remained similar from week 20 to week 40 (0.53 and 0.50, respectively) due to the fact that fetal body volume (and therefore fetal metabolic clearance) and the placental surface area (and therefore CLPD) increased with GA in a nearly parallel fashion (Table 1). Placental P-gp expression is significantly higher during early gestation versus term (Mathias et al., 2005). This gestational effect in P-gp expression resulted in lower fetal exposure to drug X at week 20, whereas at week 40, due to lower placental P-gp efflux clearance, fetal exposure to drug X increased by 1.9-fold (Fig. 8, e and f). When both clearance pathways were present, the change in placental P-gp expression was a major determinant of the GA effect on drug X AUCf (Fig. 8, g and h).

Challenges for the Next Generation of Fetal PBPK Models.

Like other fetal PBPK models, our m-f-PBPK model also has some limitations. The use of our model to predict fetal exposure to drugs prior to week 20 may be limited for the following reasons. First, data on many fetal physiologic parameters prior to week 20 (Table 1) are not available. Second, fetal skin is not completely keratinized and is highly permeable during the first half of gestation (Polin et al., 2004). Therefore, drugs that extensively partition into the skin/subcutaneous layer may readily cross into the amniotic fluid. Such movement of drugs could be incorporated into a future iteration of this model. In order for the model to be useful for drugs that are extensively metabolized by the fetus or transported/metabolized in the placenta, the model will need to be populated with GA-dependent changes in the expression of these enzymes and transporters. Such studies, using quantitative targeted proteomics, are in progress in our laboratory.

Currently, the lack of critical data hinders further development of fetal PBPK models. First, fetoplacental physiologic data across developmental stages are much needed, including the substantial changes in placental and fetal circulation and fetal organ sizes and composition. Obtaining such information will rely on the advancements in noninvasive physiologic measurement techniques and devices. Before such information becomes available, cross-species extrapolation through PBPK modeling and simulation can be conducted (Poet et al., 2010; Yoon et al., 2011) . However, this approach will not work when there are significant interspecies differences in fetoplacental metabolism and placental transport. Another challenge lies in obtaining data for PBPK model validation. As detailed above, the UV/MP ratio can change dramatically with time. Therefore, we propose that the maternal dosing regimen, the time after the last maternal dose when maternal and UV drug plasma concentrations are obtained, be recorded and the unbound drug concentration in these samples be measured.

In summary, through simulations we have shown that even when fetoplacental metabolic or placental transport clearance is small, it can significantly determine fetal drug exposure provided that the magnitude of these clearances is comparable to the CLPD of the drug (likely for hydrophilic drugs). In addition, we have shown that the single time point UV/MP ratio (except at steady state after an intravenous infusion or when maternal concentrations do not fluctuate much after multiple oral administration), routinely reported in the literature, cannot be used as an indicator of F/M plasma drug exposure ratio even at distributional equilibrium (after single or multiple doses). Therefore, one promising alternative is to dynamically estimate fetal drug exposure in humans at term and earlier in gestation through PBPK models such as the one presented here. However, prior to using the proposed model, it needs to be verified with fetal exposure data. In our companion article (Zhang and Unadkat, 2017), we describe such a verification using midazolam, zidovudine, and theophylline.

Acknowledgments

We thank Dr. Masoud Jamei for help and support on the model. We also thank Dr. Nina Isoherranen for valuable discussions.

Authorship Contributions

Participated in research design: Zhang and Unadkat.

Conducted experiments: Zhang and Imperial.

Contributed new reagents or analytic tool: Patilea-Vrana, Wedagedera, and Gaohua.

Performed data analysis: Zhang.

Wrote or contributed to the writing of the manuscript: Zhang and Unadkat

Footnotes

    • Received January 25, 2017.
    • Accepted May 25, 2017.

  • ↵1 Current affiliation: Merck Sharp & Dohme Corp., Kenilworth, New Jersey.

  • ↵2 Current affiliation: Department of Bioengineering and Therapeutic Sciences, University of California San Francisco, San Francisco, California.

  • The current work was supported by National Institutes of Health National Institute on Drug Abuse [Grant P01DA032507]. Z.Z. was supported by the Office of Women's Health, US Food and Drug Administration [ORISE Fellowship] for part of the submitted work. No other potential conflicts of interest relevant to this article are reported.

  • https://doi.org/10.1124/dmd.117.075192.

  • ↵Embedded ImageThis article has supplemental material available at dmd.aspetjournals.org.

Abbreviations

AUCf
fetal plasma area under the curve
AUCm
maternal plasma area under the curve
AUCp
placental area under the curve
Cmax,f
maximum fetal plasma drug concentration
CLf0
fetal metabolic clearance
CLMP
placental apical uptake clearance
CLm0
maternal systemic clearance
CLp0
placental metabolic clearance
CLPM
placental apical efflux clearance
CLPD
transplacental passive diffusion clearance
CLPD,u
unbound transplacental passive diffusion clearance
C-T
concentration-time
Fa
fraction absorbed
Fg
intestinal bioavailability
F/M
fetal/maternal plasma ratio
GA
gestational age
HIV
human immunodeficiency virus
ka
first-order absorption rate constant
m-f-PBPK
maternal-fetal physiologically based pharmacokinetic
MP
maternal plasma
m-PBPK
maternal pregnancy physiologically based pharmacokinetic
PBPK
physiologically based pharmacokinetic
P-gp
P-glycoprotein
PK
pharmacokinetics
P/M
placental/maternal plasma ratio
steady stateinf
steady state after an intravenous infusion
UV
umbilical vein
Vss
volume of distribution at steady state

References

  1. ↵
    1. Abduljalil K,
    2. Furness P,
    3. Johnson TN,
    4. Rostami-Hodjegan A, and
    5. Soltani H
    (2012) Anatomical, physiological and metabolic changes with gestational age during normal pregnancy: a database for parameters required in physiologically based pharmacokinetic modelling. Clin Pharmacokinet 51:365–396.
    OpenUrlCrossRefPubMed
    1. Acharya G,
    2. Erkinaro T,
    3. Mäkikallio K,
    4. Lappalainen T, and
    5. Rasanen J
    (2004) Relationships among Doppler-derived umbilical artery absolute velocities, cardiac function, and placental volume blood flow and resistance in fetal sheep. Am J Physiol Heart Circ Physiol 286:H1266–H1272.
    OpenUrlCrossRefPubMed
    1. Arant Jr BS
    (1978) Developmental patterns of renal functional maturation compared in the human neonate. J Pediatr 92:705–712.
    OpenUrlCrossRefPubMed
    1. Archie JG,
    2. Collins JS, and
    3. Lebel RR
    (2006) Quantitative standards for fetal and neonatal autopsy. Am J Clin Pathol 126:256–265.
    OpenUrlCrossRefPubMed
    1. Barry M,
    2. Back D,
    3. Ormesher S,
    4. Beeching N, and
    5. Nye F
    (1993) Metabolism of didanosine (ddI) by erythrocytes: pharmacokinetic implications. Br J Clin Pharmacol 36:87–88.
    OpenUrlCrossRefPubMed
    1. Bellotti M,
    2. Pennati G,
    3. De Gasperi C,
    4. Bozzo M,
    5. Battaglia FC, and
    6. Ferrazzi E
    (2004) Simultaneous measurements of umbilical venous, fetal hepatic, and ductus venosus blood flow in growth-restricted human fetuses. Am J Obstet Gynecol 190:1347–1358.
    OpenUrlCrossRefPubMed
  7. ↵
    1. Bernick SJ and
    2. Kane S
    (2012) Drug transfer to the fetus and to the breastfeeding infant: what do we know? Curr Drug Deliv 9:350–355.
    OpenUrlCrossRefPubMed
    1. Boito S,
    2. Struijk PC,
    3. Ursem NT,
    4. Stijnen T, and
    5. Wladimiroff JW
    (2002) Umbilical venous volume flow in the normally developing and growth-restricted human fetus. Ultrasound Obstet Gynecol 19:344–349.
    OpenUrlAbstract/FREE Full Text
  9. ↵
    1. Chappuy H,
    2. Tréluyer J-M,
    3. Jullien V,
    4. Dimet J,
    5. Rey E,
    6. Fouché M,
    7. Firtion G,
    8. Pons G, and
    9. Mandelbrot L
    (2004a) Maternal-fetal transfer and amniotic fluid accumulation of nucleoside analogue reverse transcriptase inhibitors in human immunodeficiency virus-infected pregnant women. Antimicrob Agents Chemother 48:4332–4336.
    OpenUrlPubMed
  10. ↵
    1. Chappuy H,
    2. Tréluyer J-M,
    3. Rey E,
    4. Dimet J,
    5. Fouché M,
    6. Firtion G,
    7. Pons G, and
    8. Mandelbrot L
    (2004b) Maternal-fetal transfer and amniotic fluid accumulation of protease inhibitors in pregnant women who are infected with human immunodeficiency virus. Am J Obstet Gynecol 191:558–562.
    OpenUrlCrossRefPubMed
  11. ↵
    1. Clewell RA,
    2. Merrill EA,
    3. Gearhart JM,
    4. Robinson PJ,
    5. Sterner TR,
    6. Mattie DR, and
    7. Clewell 3rd HJ
    (2007) Perchlorate and radioiodide kinetics across life stages in the human: using PBPK models to predict dosimetry and thyroid inhibition and sensitive subpopulations based on developmental stage. J Toxicol Environ Health A 70:408–428.
    OpenUrlPubMed
  12. ↵
    1. Collins JM and
    2. Unadkat JD
    (1989) Clinical pharmacokinetics of zidovudine. An overview of current data. Clin Pharmacokinet 17:1–9.
    OpenUrlAbstract/FREE Full Text
  13. ↵
    1. Coonen JB,
    2. Marcus MA,
    3. Joosten EA,
    4. van Kleef M,
    5. Neef C,
    6. van Aken H, and
    7. Gogarten W
    (2010) Transplacental transfer of remifentanil in the pregnant ewe. Br J Pharmacol 161:1472–1476.
    OpenUrlCrossRefPubMed
    1. Coulthard MG
    (1985) Maturation of glomerular filtration in preterm and mature babies. Early Hum Dev 11:281–292.
    OpenUrlPubMed
    1. Cussen L,
    2. Scurry J,
    3. Mitropoulos G,
    4. McTigue C, and
    5. Gross J
    (1990) Mean organ weights of an Australian population of fetuses and infants. J Paediatr Child Health 26:101–103.
    OpenUrlAbstract
    1. De Sousa Mendes M,
    2. Hirt D,
    3. Vinot C,
    4. Valade E,
    5. Lui G,
    6. Pressiat C,
    7. Bouazza N,
    8. Foissac F,
    9. Blanche S,
    10. Le MP, et al.
    (2015) Prediction of human fetal pharmacokinetics using ex vivo human placenta perfusion studies and physiologically based models. Br J Clin Pharmacol 81:646–657
    OpenUrlPubMed
  17. ↵
    1. De Sousa Mendes M,
    2. Lui G,
    3. Zheng Y,
    4. Pressiat C,
    5. Hirt D,
    6. Valade E,
    7. Bouazza N,
    8. Foissac F,
    9. Blanche S,
    10. Treluyer J-M, et al.
    (2017) A physiologically-based pharmacokinetic model to predict human fetal exposure for a drug metabolized by several CYP450 pathways. Clin Pharmacokinet 56:537–550.
    OpenUrlCrossRefPubMed
  18. ↵
    1. Egan TD
    (1995) Remifentanil pharmacokinetics and pharmacodynamics. A preliminary appraisal. Clin Pharmacokinet 29:80–94.
    OpenUrl
  19. ↵
    1. Else LJ,
    2. Taylor S,
    3. Back DJ, and
    4. Khoo SH
    (2011) Pharmacokinetics of antiretroviral drugs in anatomical sanctuary sites: the fetal compartment (placenta and amniotic fluid). Antivir Ther 16:1139–1147.
    OpenUrlCrossRefPubMed
    1. FitzSimmons J,
    2. Chinn A, and
    3. Shepard TH
    (1988) Normal length of the human fetal gastrointestinal tract. Pediatr Pathol 8:633–641.
    OpenUrlAbstract/FREE Full Text
    1. Flo K,
    2. Wilsgaard T, and
    3. Acharya G
    (2010) Longitudinal reference ranges for umbilical vein blood flow at a free loop of the umbilical cord. Ultrasound Obstet Gynecol 36:567–572.
    OpenUrlPubMed
  22. ↵
    1. Gilbert WM and
    2. Brace RA
    (1993) Amniotic fluid volume and normal flows to and from the amniotic cavity. Semin Perinatol 17:150–157.
    OpenUrlCrossRefPubMed
    1. Gitlin D and
    2. Perricelli A
    (1970) Synthesis of serum albumin, prealbumin, alpha-foetoprotein, alpha-1-antitrypsin and transferrin by the human yolk sac. Nature 228:995–997.
    OpenUrlFREE Full Text
    1. Gruenwald P and
    2. Minh HN
    (1961) Evaluation of body and organ weights in perinatal pathology. II. Weight of body and placenta of surviving and of autopsied infants. Am J Obstet Gynecol 82:312–319.
    OpenUrlPubMed
    1. Guignard JP,
    2. Torrado A,
    3. Da Cunha O, and
    4. Gautier E
    (1975) Glomerular filtration rate in the first three weeks of life. J Pediatr 87:268–272.
    OpenUrlCrossRefPubMed
    1. Hansen K,
    2. Sung CJ,
    3. Huang C,
    4. Pinar H,
    5. Singer DB, and
    6. Oyer CE
    (2003) Reference values for second trimester fetal and neonatal organ weights and measurements. Pediatr Dev Pathol 6:160–167.
    OpenUrlAbstract/FREE Full Text
    1. Hansen NB,
    2. Oh W,
    3. LaRochelle F, and
    4. Stonestreet BS
    (1983) Effects of maternal ritodrine administration on neonatal renal function. J Pediatr 103:774–780.
    OpenUrlPubMed
    1. Haugen G,
    2. Kiserud T,
    3. Godfrey K,
    4. Crozier S, and
    5. Hanson M
    (2004) Portal and umbilical venous blood supply to the liver in the human fetus near term. Ultrasound Obstet Gynecol 24:599–605.
    OpenUrlAbstract/FREE Full Text
  29. ↵
    1. Hill MD and
    2. Abramson FP
    (1988) The significance of plasma protein binding on the fetal/maternal distribution of drugs at steady-state. Clin Pharmacokinet 14:156–170.
    OpenUrlCrossRefPubMed
  30. ↵
    1. Isoherranen N and
    2. Thummel KE
    (2013) Drug metabolism and transport during pregnancy: how does drug disposition change during pregnancy and what are the mechanisms that cause such changes? Drug Metab Dispos 41:256–262.
    OpenUrlCrossRefPubMed
    1. Kanto J,
    2. Sjövall S,
    3. Erkkola R,
    4. Himberg JJ, and
    5. Kangas L
    (1983) Placental transfer and maternal midazolam kinetics. Clin Pharmacol Ther 33:786–791.
    OpenUrlPubMed
  32. ↵
    1. Ke AB,
    2. Nallani SC,
    3. Zhao P,
    4. Rostami-Hodjegan A, and
    5. Unadkat JD
    (2012) A PBPK model to predict disposition of CYP3A-metabolized drugs in pregnant women: verification and discerning the site of CYP3A induction. CPT Pharmacometrics Syst Pharmacol 1:e3.
    OpenUrlCrossRefPubMed
  33. ↵
    1. Ke AB,
    2. Nallani SC,
    3. Zhao P,
    4. Rostami-Hodjegan A,
    5. Isoherranen N, and
    6. Unadkat JD
    (2013) A physiologically based pharmacokinetic model to predict disposition of CYP2D6 and CYP1A2 metabolized drugs in pregnant women. Drug Metab Dispos 41:801–813.
    OpenUrlPubMed
  34. ↵
    1. Ke AB,
    2. Nallani SC,
    3. Zhao P,
    4. Rostami-Hodjegan A, and
    5. Unadkat JD
    (2014) Expansion of a PBPK model to predict disposition in pregnant women of drugs cleared via multiple CYP enzymes, including CYP2B6, CYP2C9 and CYP2C19. Br J Clin Pharmacol 77:554–570
    OpenUrlAbstract/FREE Full Text
    1. Kenny JF,
    2. Plappert T,
    3. Doubilet P,
    4. Saltzman DH,
    5. Cartier M,
    6. Zollars L,
    7. Leatherman GF, and
    8. St John Sutton MG
    (1986) Changes in intracardiac blood flow velocities and right and left ventricular stroke volumes with gestational age in the normal human fetus: a prospective Doppler echocardiographic study. Circulation 74:1208–1216.
    OpenUrlCrossRef
    1. Kessler J,
    2. Rasmussen S,
    3. Godfrey K,
    4. Hanson M, and
    5. Kiserud T
    (2008) Longitudinal study of umbilical and portal venous blood flow to the fetal liver: low pregnancy weight gain is associated with preferential supply to the fetal left liver lobe. Pediatr Res 63:315–320.
    OpenUrlPubMed
    1. Kiserud T,
    2. Rasmussen S, and
    3. Skulstad S
    (2000) Blood flow and the degree of shunting through the ductus venosus in the human fetus. Am J Obstet Gynecol 182:147–153.
    OpenUrlAbstract/FREE Full Text
    1. Knupp CA,
    2. Shyu WC,
    3. Dolin R,
    4. Valentine FT,
    5. McLaren C,
    6. Martin RR,
    7. Pittman KA, and
    8. Barbhaiya RH
    (1991) Pharmacokinetics of didanosine in patients with acquired immunodeficiency syndrome or acquired immunodeficiency syndrome-related complex. Clin Pharmacol Ther 49:523–535.
    OpenUrlCrossRefPubMed
    1. Krauer B,
    2. Dayer P, and
    3. Anner R
    (1984) Changes in serum albumin and alpha 1-acid glycoprotein concentrations during pregnancy: an analysis of fetal-maternal pairs. Br J Obstet Gynaecol 91:875–881.
    OpenUrlCrossRefPubMed
    1. Lees C,
    2. Albaiges G,
    3. Deane C,
    4. Parra M, and
    5. Nicolaides KH
    (1999) Assessment of umbilical arterial and venous flow using color Doppler. Ultrasound Obstet Gynecol 14:250–255.
    OpenUrlCrossRefPubMed
  41. ↵
    1. Loccisano AE,
    2. Longnecker MP,
    3. Campbell JL Jr.,
    4. Andersen ME, and
    5. Clewell HJ 3rd.
    (2013) Development of PBPK models for PFOA and PFOS for human pregnancy and lactation life stages. J Toxicol Environ Health A 76:25–57.
    OpenUrlAbstract/FREE Full Text
  42. ↵
    1. Marzolini C and
    2. Kim RB
    (2005) Placental transfer of antiretroviral drugs. Clin Pharmacol Ther 78:118–122.
    OpenUrlAbstract/FREE Full Text
  43. ↵
    1. Mathias AA,
    2. Hitti J, and
    3. Unadkat JD
    (2005) P-glycoprotein and breast cancer resistance protein expression in human placentae of various gestational ages. Am J Physiol Regul Integr Comp Physiol 289:R963–R969.
    OpenUrlCrossRefPubMed
    1. Mayhew TM,
    2. Ohadike C,
    3. Baker PN,
    4. Crocker IP,
    5. Mitchell C, and
    6. Ong SS
    (2003) Stereological investigation of placental morphology in pregnancies complicated by pre-eclampsia with and without intrauterine growth restriction. Placenta 24:219–226.
    OpenUrlAbstract/FREE Full Text
  45. ↵
    1. McCormack SA and
    2. Best BM
    (2014) Protecting the fetus against HIV infection: a systematic review of placental transfer of antiretrovirals. Clin Pharmacokinet 53:989–1004.
    OpenUrlCrossRefPubMed
  46. ↵
    1. McGowan JP and
    2. Shah SS
    (2000) Prevention of perinatal HIV transmission during pregnancy. J Antimicrob Chemother 46:657–668.
    OpenUrlPubMed
  47. ↵
    1. Mitchell AA,
    2. Gilboa SM,
    3. Werler MM,
    4. Kelley KE,
    5. Louik C, and
    6. Hernandez-Diaz S
    (2011) Medication use during pregnancy, with particular focus on prescription drugs: 1976-2008. Am J Obstet Gynecol 205:51.e1–e8.
    OpenUrlCrossRefPubMed
    1. Moniz CF,
    2. Nicolaides KH,
    3. Bamforth FJ, and
    4. Rodeck CH
    (1985) Normal reference ranges for biochemical substances relating to renal, hepatic, and bone function in fetal and maternal plasma throughout pregnancy. J Clin Pathol 38:468–472.
    OpenUrlAbstract/FREE Full Text
  49. monograph D.
  50. ↵
    1. Myllynen P,
    2. Immonen E,
    3. Kummu M, and
    4. Vähäkangas K
    (2009) Developmental expression of drug metabolizing enzymes and transporter proteins in human placenta and fetal tissues. Expert Opin Drug Metab Toxicol 5:1483–1499.
    OpenUrlAbstract/FREE Full Text
  51. ↵
    1. Myllynen P,
    2. Pasanen M, and
    3. Vähäkangas K
    (2007) The fate and effects of xenobiotics in human placenta. Expert Opin Drug Metab Toxicol 3:331–346.
    OpenUrlPubMed
    1. Nagata S,
    2. Koyanagi T,
    3. Horimoto N,
    4. Satoh S, and
    5. Nakano H
    (1990) Chronological development of the fetal stomach assessed using real-time ultrasound. Early Hum Dev 22:15–22.
    OpenUrlPubMed
  53. ↵
    1. Ngan Kee WD,
    2. Khaw KS,
    3. Tan PE,
    4. Ng FF, and
    5. Karmakar MK
    (2009) Placental transfer and fetal metabolic effects of phenylephrine and ephedrine during spinal anesthesia for cesarean delivery. Anesthesiology 111:506–512.
    OpenUrlPubMed
    1. Parulekar SG
    (1991) Sonography of normal fetal bowel. J Ultrasound Med 10:211–220.
    OpenUrlPubMed
    1. Pasman SA,
    2. van den Brink CP,
    3. Kamping MA,
    4. Adama van Scheltema PN,
    5. Oepkes D, and
    6. Vandenbussche FP
    (2009) Total blood volume is maintained in nonhydropic fetuses with severe hemolytic anemia. Fetal Diagn Ther 26:10–15.
    OpenUrlCrossRefPubMed
  56. ↵
    1. Pereira CM,
    2. Nosbisch C,
    3. Winter HR,
    4. Baughman WL, and
    5. Unadkat JD
    (1994) Transplacental pharmacokinetics of dideoxyinosine in pigtailed macaques. Antimicrob Agents Chemother 38:781–786.
    OpenUrlCrossRefPubMed
  57. ↵
    1. Pidoux G,
    2. Gerbaud P,
    3. Guibourdenche J,
    4. Thérond P,
    5. Ferreira F,
    6. Simasotchi C,
    7. Evain-Brion D, and
    8. Gil S
    (2015) Formaldehyde crosses the human placenta and affects human trophoblast differentiation and hormonal functions. PLoS One 10:e0133506.
    OpenUrlCrossRefPubMed
  58. ↵
    1. Poet TS,
    2. Kirman CR,
    3. Bader M,
    4. van Thriel C,
    5. Gargas ML, and
    6. Hinderliter PM
    (2010) Quantitative risk analysis for N-methyl pyrrolidone using physiologically based pharmacokinetic and benchmark dose modeling. Toxicol Sci 113:468–482.
    OpenUrlAbstract/FREE Full Text
  59. ↵
    1. Polin RA,
    2. Fox WW, and
    3. Abman SH
    (2004) Fetal and Neonatal Physiology, W.B. Saunders Co., Philadelphia, PA.
  60. ↵
    1. Ron M,
    2. Hochner-Celnikier D,
    3. Menczel J,
    4. Palti Z, and
    5. Kidroni G
    (1984) Maternal-fetal transfer of aminophylline. Acta Obstet Gynecol Scand 63:217–218.
    OpenUrlCrossRefPubMed
    1. Rubesova E,
    2. Vance CJ,
    3. Ringertz HG, and
    4. Barth RA
    (2009) Three-dimensional MRI volumetric measurements of the normal fetal colon. AJR Am J Roentgenol 192:761–765.
    OpenUrlPubMed
    1. Rudolph AM,
    2. Heymann MA,
    3. Teramo KAW,
    4. Barrett CT, and
    5. Raiha NCR
    (1971) Studies on the circulation of the previable human fetus. Pediatr Res 5:452–465.
    OpenUrlCrossRefPubMed
    1. Sutton MS,
    2. Theard MA,
    3. Bhatia SJ,
    4. Plappert T,
    5. Saltzman DH, and
    6. Doubilet P
    (1990) Changes in placental blood flow in the normal human fetus with gestational age. Pediatr Res 28:383–387.
    OpenUrlCrossRefPubMed
    1. Tchirikov M,
    2. Rybakowski C,
    3. Hüneke B,
    4. Schoder V, and
    5. Schröder HJ
    (2002) Umbilical vein blood volume flow rate and umbilical artery pulsatility as ‘venous-arterial index’ in the prediction of neonatal compromise. Ultrasound Obstet Gynecol 20:580–585.
    OpenUrlCrossRefPubMed
    1. Tchirikov M,
    2. Rybakowski C,
    3. Hüneke B, and
    4. Schröder HJ
    (1998) Blood flow through the ductus venosus in singleton and multifetal pregnancies and in fetuses with intrauterine growth retardation. Am J Obstet Gynecol 178:943–949.
    OpenUrlCrossRefPubMed
  66. ↵
    1. Tuntland T,
    2. Nosbisch C,
    3. Baughman WL,
    4. Massarella J, and
    5. Unadkat JD
    (1996) Mechanism and rate of placental transfer of zalcitabine (2′,3′-dideoxycytidine) in Macaca nemestrina. Am J Obstet Gynecol 174:856–863.
    OpenUrlAbstract/FREE Full Text
    1. Tuntland T,
    2. Odinecs A,
    3. Pereira CM,
    4. Nosbisch C, and
    5. Unadkat JD
    (1999) In vitro models to predict the in vivo mechanism, rate, and extent of placental transfer of dideoxynucleoside drugs against human immunodeficiency virus. Am J Obstet Gynecol 180:198–206.
    OpenUrlCrossRefPubMed
  68. ↵
    1. Unadkat JD,
    2. Dahlin A, and
    3. Vijay S
    (2004) Placental drug transporters. Curr Drug Metab 5:125–131.
    OpenUrlCrossRefPubMed
    1. van den Anker JN,
    2. de Groot R,
    3. Broerse HM,
    4. Sauer PJ,
    5. van der Heijden BJ,
    6. Hop WC, and
    7. Lindemans J
    (1995) Assessment of glomerular filtration rate in preterm infants by serum creatinine: comparison with inulin clearance. Pediatrics 96:1156–1158.
    OpenUrlAbstract/FREE Full Text
  70. ↵
    1. van der Sandt IC,
    2. Vos CM,
    3. Nabulsi L,
    4. Blom-Roosemalen MC,
    5. Voorwinden HH,
    6. de Boer AG, and
    7. Breimer DD
    (2001) Assessment of active transport of HIV protease inhibitors in various cell lines and the in vitro blood--brain barrier. AIDS 15:483–491.
    OpenUrlAbstract/FREE Full Text
    1. Veille JC,
    2. Hanson RA,
    3. Tatum K, and
    4. Kelley K
    (1993) Quantitative assessment of human fetal renal blood flow. Am J Obstet Gynecol 169:1399–1402.
    OpenUrlCrossRefPubMed
    1. Velasque LS,
    2. Estrela RCE,
    3. Suarez-Kurtz G, and
    4. Struchiner CJ
    (2007) A new model for the population pharmacokinetics of didanosine in healthy subjects. Braz J Med Biol Res 40:97–104.
    OpenUrlPubMed
    1. Wang Y and
    2. Zhao S
    (2010) Vascular Biology of the Placenta, Morgan & Claypool Life Sciences, San Rafael, CA.
    1. Weiner CP,
    2. Sipes SL, and
    3. Wenstrom K
    (1992) The effect of fetal age upon normal fetal laboratory values and venous pressure. Obstet Gynecol 79:713–718.
    OpenUrlPubMed
  75. ↵
    1. Yoon M,
    2. Schroeter JD,
    3. Nong A,
    4. Taylor MD,
    5. Dorman DC,
    6. Andersen ME, and
    7. Clewell 3rd HJ
    (2011) Physiologically based pharmacokinetic modeling of fetal and neonatal manganese exposure in humans: describing manganese homeostasis during development. Toxicol Sci 122:297–316.
    OpenUrlAbstract/FREE Full Text
  76. ↵
    1. Zhang Z and
    2. Unadkat JD
    (2017) Verification of a maternal-fetal physiologically based pharmacokinetic model for passive placental permeability drugs. Drug Metab Dispos DOI:10.1124/dmd.116.073957 [published ahead of print].
Back to top

In this issue

Download PDF
Article Alerts
Email Article
Citation Tools

Research ArticleArticle

A PBPK Model of Factors Determining Fetal Drug Exposure

Zufei Zhang, Marjorie Z. Imperial, Gabriela I. Patilea-Vrana, Janak Wedagedera, Lu Gaohua and Jashvant D. Unadkat
Drug Metabolism and Disposition August 1, 2017, 45 (8) 920-938; DOI: https://doi.org/10.1124/dmd.117.075192
Reddit logo Twitter logo Facebook logo Mendeley logo

Jump to section

Advertisement