Application of physiologically based pharmacokinetic modeling for sertraline dosing recommendations in pregnancy

Abstract

Pregnancy is a period of significant change that impacts physiological and metabolic status leading to alterations in the disposition of drugs. Uncertainty in drug dosing in pregnancy can lead to suboptimal therapy, which can contribute to disease exacerbation. A few studies show there are increased dosing requirements for antidepressants in late pregnancy; however, the quantitative data to guide dose adjustments are sparse. We aimed to develop a physiologically based pharmacokinetic (PBPK) model that allows gestational-age dependent prediction of sertraline dosing in pregnancy. A minimal physiological model with defined gut, liver, plasma, and lumped placental-fetal compartments was constructed using the ordinary differential equation solver package, 'mrgsolve', in R. We extracted data from the literature to parameterize the model, including sertraline physicochemical properties, in vitro metabolism studies, disposition in nonpregnant women, and physiological changes during pregnancy. The model predicted the pharmacokinetic parameters from a clinical study with eight subjects for the second trimester and six subjects for the third trimester. Based on the model, gestational-dependent changes in physiology and metabolism account for increased clearance of sertraline (up to 143% at 40 weeks gestational age), potentially leading to under-dosing of pregnant women when nonpregnancy doses are used. The PBPK model was converted to a prototype web-based interactive dosing tool to demonstrate how the output of a PBPK model may translate into optimal sertraline dosing in pregnancy. Quantitative prediction of drug exposure using PBPK modeling in pregnancy will support clinically appropriate dosing and increase the therapeutic benefit for pregnant women.

Fig. 1. Model prediction for steady-state sertraline plasma concentrations in nonpregnancy.

Black dots represent mean total plasma concentration in nonpregnant females (N = 11) ages 20–45 years receiving sertraline 200 mg oral tablets daily for 30 days following the last dose. Monte Carlo simulations with a daily dose of 200 mg for 30 days were run for 1000 iterations. The predicted mean is depicted by the red line. The 95% prediction interval (2.5th–97.5th percentile range of a virtual population [N = 1000]) is depicted in the red area.

Fig. 2. Pregnancy model evaluation in second and third trimester of pregnancy.

The predicted versus observed graphs for each pharmacokinetic (PK) parameter (maximum plasma concentration [C_max] and 24-h drug exposure [AUC₂₄]) is given for second (N = 8) and third trimester (N = 6). In addition, the predicted versus observed graph for percent change from second to third trimester is also shown (N = 6). The solid line represents the unity line where the predicted to observed ratio is 1. The dotted lines represent the twofold error. The black dots represent individual patient data.

Fig. 3. Predicted change in hepatic enzyme contribution to sertraline metabolism across gestation.

The nonpregnancy contribution of five cytochrome P450 (CYP) enzymes to sertraline metabolism was calculated by in vitro-in vivo extrapolation. Following incorporation of gestation-dependent increase in CYP activities, the contributions of each individual CYP enzyme out of the total hepatic clearance at various (0, 10, 20, and 30) gestational ages (GA) are demonstrated. Note that the total hepatic clearance increased over the course of pregnancy by 11%, 37%, and 63% during GA 10, 20, and 30 weeks, respectively.

Fig. 4. Predicted change in steady-state sertraline plasma concentrations with gestational age and interactive PBPK dosing tool for sertraline.

a–c The predicted sertraline plasma concentration in pregnancy (blue) compared to nonpregnancy (red) is visually depicted. Representative gestational ages (GAs) for each trimester is shown in graphs (a–c). Lines represent the mean concentration while the colored areas represent the 95% prediction interval (2.5th–97.5th percentile range of a virtual population [N = 1000]). d A screenshot for the web-based interactive PBPK dosing tool. Users can adjust various parameters including gestational age, body weight, dose and number of doses. Please note that the current version of the tool is a prototype and includes mean plasma concentration versus time profiles for illustrative purposes and does not include estimates of computed population variabilities.

Fig. 5. Workflow for pregnancy physiologically based pharmacokinetic model.

Workflow for the development of the pregnancy physiologically based pharmacokinetic (PBPK) model. Sertraline physicochemical properties were collected from DrugBank and used to calculate absorption, distribution, metabolism, and excretion (ADME) parameters. Physiological data for nonpregnancy was taken from ICRP Publication 89. A deterministic nonpregnancy PBPK model was established and calibrated with pharmacokinetic data in nonpregnancy from a calibration dataset. Following satisfactory calibration, population prediction was achieved by performing sensitivity analysis and Monte Carlo simulations. The population nonpregnancy model was extended to pregnancy by incorporating physiological changes in pregnancy^,,. The pregnancy model was simulated to predict pharmacokinetic data for the second and third trimester of pregnancy using a verification dataset. Following verification, an interactive pregnancy dosing tool was created using ‘Shiny’.

Fig. 6. Minimal pregnancy PBPK model structure for sertraline.

Physiologically based pharmacokinetic (PBPK) model structure for sertraline after oral exposure. Arrows represent blood flow between compartments. Boxes represent tissue compartments described as flow limited. Pregnancy-related tissues and remaining richly perfused and slowly perfused tissues are represented as lumped compartments. K_a represents the absorption rate constant for disappearance from lumen and appearance in the gut compartment.