Application of PBPK modeling in predicting maternal and fetal pharmacokinetics of levetiracetam during pregnancy

Physiologically-based pharmacokinetic modeling to predict the exposure and to assess pharmacodynamics of daptomycin in infants within 1 year old

Daptomycin is widely used in pediatric patients for serious infections caused by Gram-positive bacteria, however, studies regarding its safety and efficacy in infants within 1 year old are very limited. A physiologically-based pharmacokinetic (PBPK) model of daptomycin was built for children aged 1-17 years old and extrapolated to infants within 1 year old to evaluate pharmacodynamics (PD) based on efficacy and safety considerations. Monte Carlo Simulations (MCSs) were conducted to determine the probabilities of target attainment (PTA) and cumulative fractions of response (CFR) of daptomycin. The pharmacokinetic (PK) of daptomycin did not differ much in the population of infants within 1 year of age, with peak plasma concentration (Cmax) and area under the curve (AUC) maintained at an approximate level at all months of age, while the average trough concentration of daptomycin was 3.49 μg/mL when 10 mg/kg daptomycin was given, and 4.98 ug /mL at 15 mg/kg. According to the results of the MCSs, 10mg/kg daptomycin provides good antimicrobial effect for S.pneumoniae and MSSA. With the increase of dosage, the CFR value of daptomycin against MRSA, E.faecalis and E.faecium also gradually reached >90 %, except for E.faecalis with an average CFR of only 82.94 % at 12mg/kg. This is a daptomycin PBPK model in infants within 1 year of age, dose regimen higher than 10 mg/kg should be recommended for this population in the treatment of MRSA, E. faecalis, and E. faecalis.

PBPK Modeling: Empowering Drug Development and Precision Dosing in China

Physiologically based pharmacokinetic (PBPK) modeling, a cornerstone of model-informed drug development and model-informed precision dosing, simulates drug disposition in the human body by integrating physiological, biochemical, and physicochemical parameters. While PBPK modeling has advanced globally since the 1970s, China's adoption of this technology has followed a distinctive path, characterized by accelerated growth over the past 2 decades. This review provides a comprehensive analysis of China's contributions to PBPK modeling, addressing knowledge gaps in publication trends, application domains, and platform preferences. A systematic literature search yielded 266 original PBPK research articles from PubMed up to August 08, 2024. The analysis revealed that drug disposition and drug-drug interaction studies constitute the largest proportion of PBPK analyses in China. Chinese universities and hospitals emerge as the leading contributors to PBPK research among institutions in China. Although established commercial PBPK platform such as GastroPlus and Simcyp remain popular within the Chinese pharmaceutical industry, open-source platforms like PK-Sim are gaining significant traction in PBPK applications across China. This review underscores the transformative potential of PBPK modeling in drug development within China, offering valuable insights into future directions and challenges in the field.

Quantification of sertraline maternal/fetal ratio and amniotic fluid concentration using a pregnancy physiologically based pharmacokinetic model

AIMS

Selective serotonin reuptake inhibitors (SSRIs) are indicated for a variety of psychiatric conditions which may require treatment during pregnancy. Knowledge of appropriate SSRI dosages that maintain maternal therapeutic benefit and minimize fetal risk are needed. Assessment of fetal exposure to drugs is challenging since sampling is often limited to a single concentration from the umbilical cord at delivery. Physiologically based pharmacokinetic (PBPK) modelling provides a non-invasive approach to quantify exposure in pregnancy.

METHODS

We incorporated sertraline clearances mediated by passive diffusion, placental efflux transporters P-glycoprotein (P-gp) and breast cancer resistance protein (BCRP) into our previously published pregnancy PBPK model for sertraline. Simulations were performed for various sertraline doses (25-200 mg) at 40 weeks gestational age to predict the minimum (Cmin), maximum (Cmax) and average (Cavg) sertraline maternal and fetal plasma concentrations and evaluated them against observed maternal and cord concentrations obtained at delivery from five clinical studies.

RESULTS

The accuracy of the PBPK predictions as indicated by the average fold error (AFE) value for Cmax, Cmin and Cavg for maternal plasma sertraline concentrations at delivery was 1.7, 1.2 and 1.4, respectively. The AFE for the Cmax, Cmin and Cavg for cord blood sertraline concentration at delivery was 1.2, 1 and 1.1, respectively. The AFE for cord-maternal sertraline concentration ratio at delivery for Cmax, Cmin and Cavg was 0.7, 0.9 and 0.8, respectively.

CONCLUSIONS

The PBPK model we developed may serve as a guide for maternal sertraline dose adjustment during pregnancy considering changes in exposures for both mother and fetus.

Coupling Pre- and Postnatal Infant Exposures with Physiologically Based Pharmacokinetic Modeling to Predict Cumulative Maternal Levetiracetam Exposure During Breastfeeding

BACKGROUND AND OBJECTIVE

Although breastfeeding ensures optimal infant development and maternal health, mothers taking medications may abandon breastfeeding because of uncertainties regarding toxicity to infants. Current methods in predicting infant risk to maternal medication exposure do not account for breastfeeding-related variability or in utero exposure via the umbilical cord (UC). Previously, our workflow integrated variability in infant anatomy and physiology, breast milk intake volume, and drug concentrations in breast milk using physiologically based pharmacokinetic (PBPK) modeling. The upper area under the curve ratio (UAR) was then calculated to assess infant risk from maternal drug. Herein, we enhanced this workflow by coupling pre- and postnatal exposures to predict the overall levetiracetam exposure in breastfeeding infants.

METHODS

A published pediatric PBPK model of levetiracetam was used to simulate an infant population (n = 100). Daily infant doses were simulated using a weight-normalized milk intake model to calculate volumes ingested across age groups, alongside literature-derived or simulated milk concentrations across maternal doses to predict infant concentrations. Published UC concentrations were used to develop a cord-coupled neonatal model (CCM), which was integrated with the PBPK and milk intake models and evaluated by comparing observed and simulated infant blood concentrations using a 90% prediction interval (PI).

RESULTS

UC concentration data from 14 mothers were used to develop the CCM. A total of 16 paired (known milk concentrations) and two unpaired (unknown milk concentrations) individual infant concentrations were identified for evaluating the model along with population values of 64 infants from two age groups (2-4 and 7-31 days). The CCM improved the predictions overall compared with the original workflow, largely due to improvements for the youngest age group evaluated. Overall, 83% (10 of 12) of the individual infant plasma concentrations were successfully captured within the 90% PI for the paired, quantifiable (i.e. above the limit of quantification) evaluation datasets. After administration of a maternal dose of levetiracetam 2000 mg, the calculated UAR ranged from 0.13 to 0.27 for the 95th percentile infants.

CONCLUSIONS

To our knowledge, this is the first report to combine prenatal levetiracetam exposures from the UC and postnatal exposures from breastfeeding to predict overall infant drug exposure. The results indicate that infant exposure in infants aged 0-7 days may approach therapeutic levels of levetiracetam in the highest-risk infants (i.e. 95th percentile), with a low likelihood of adverse effects based on published clinical studies. This integrated modeling approach provides a more holistic analysis of neonatal exposures. It can be applied in future studies to derive the UAR of drugs administered during breastfeeding to identify infants at risk of potential toxicity.

Preclinical pharmacokinetic studies and prediction of human PK profiles for Deg-AZM, a clinical-stage new transgelin agonist

Introduction

Deglycosylated azithromycin (Deg-AZM), a newly developed Class I drug with good therapeutic effects on slow transit constipation, is a small-molecule transgelin agonist that has been approved for clinical trials in 2024. The preclinical pharmacokinetic profile of Deg-AZM was investigated to support further development.

Methods

A LC-MS/MS method was established and validated to detected the concentration of Deg-AZM in various biological samples. In vivo tests such as pharmacokinetic studies in rats and dogs, tissue distribution studies in rats, and extraction studies in rats were conducted to investigated the preclinical pharmacokinetic behaviors of Deg-AZM comprehensively. The plasma protein rate of Deg-AZM was determined by rapid equilibrium dialysis method in vitro. The metabolic stability and metabolite profile of Deg-AZM was assessed using pooled mice, rats, dogs, monkeys and humans microsomes in vitro. The PK profiles of Deg-AZM in human was predicted based on physiologically based pharmacokinetic (PBPK) models.

Results

The plasma protein binding rates of Deg-AZM were lower in mice and rats, higher in dogs, and moderate in humans. The metabolic process of Deg-AZM was similar in rat and human liver microsomes. From Pharmacokinetic studies in rats and dogs, Deg-AZM was rapidly absorbed into the blood and then quickly eliminated. Plasma exposure of Deg-AZM was dose dependent with no accumulation after continuous gavage administration. In addition, there is no significant gender difference in the pharmacokinetic behavior of Deg-AZM. Deg-AZM was widely distributed in the tissues without obvious accumulation, and mainly excreted from the urinary excretion pathway. Furthermore, the pharmacokinetic profiles of Deg-AZM in humans showed dose dependency.

Conclusion

The pharmacokinetic profiles of Deg-AZM was fully explored, these results could provide valuable information to support the first-in-human dosage prediction and phase I clinical design.

Physiologically-based pharmacokinetic models to predict drug exposure during pregnancy

As pregnant women are constantly exposed to drugs during pregnancy, either to treat long-term conditions or acute illnesses, drug safety is a major concern for the fetus and the mother. Clinical trials are rarely made in this population due to strict regulation and ethical reasons. However, drug pharmacokinetic (PK) parameters vary during pregnancy with an increase in distribution volume, renal clearance and more. In addition, the fetal distribution should be evaluated with the importance of placental diffusion, both active and passive. Therefore, there is a recent interest in the use of physiologically-based pharmacokinetic (PBPK) modeling to characterize these changes and complete the sparse data available on drug PK during pregnancy. Indeed, PBPK models integrate drug physicochemical and physiological parameters corresponding to each compartment of the body to estimate drug concentrations. This review establishes an overview on the current use of PBPK models in drug dosage determination for the pregnant woman, fetal exposure and drug interactions in the fetal compartment.

Quantitation of levetiracetam concentrations in plasma and saliva samples by ultra-performance liquid chromatography-tandem mass spectrometry: Application to therapeutic drug monitoring for pregnant women with epilepsy

Although levetiracetam (LEV) has favorable linear pharmacokinetic properties, therapeutic drug monitoring (TDM) is necessary for pregnant women with epilepsy. This study aims to build a simple, reliable, and sensitive ultra-performance liquid chromatography-tandem mass spectrometry (UPLC-MS/MS) method for determining LEV concentrations in plasma and saliva samples, to support the routine TDM of LEV in Chinese pregnant women with epilepsy. The stable isotope-labeled LEV-d6 was used as the internal standard. The extracted samples were analyzed using a UPLC-MS/MS system with positive electrospray ionization. Mobile phase A was water containing 5 mM ammonium acetate and 0.1% formic acid, and phase B was 1:1 methanol-acetonitrile with 0.1% formic acid. The method was validated and utilized to determine LEV concentrations in non-pregnant and pregnant patients with epilepsy. The developed method was validated in both plasma and saliva samples over a concentration range of 0.1-50 μg/mL. The intra- and inter-batch accuracy for LEV ranged from -7.0% to 2.9%, with precisions between 2.7% and 9.3%. In pregnant patients, the mean dose-standardized LEV trough plasma concentrations were significantly lower than those in non-pregnant patients (4.73 ± 2.99 vs. 7.74 ± 3.59 ng/mL per mg/day; P < 0.0001). It is recommended that the TDM of LEV should be routinely performed during the different stages of pregnancy.

PBPK Modeling Approach to Predict the Behavior of Drugs Cleared by Metabolism in Pregnant Subjects and Fetuses

This study aimed to develop a physiologically based pharmacokinetic (PBPK) model that simulates metabolically cleared compounds' pharmacokinetics (PK) in pregnant subjects and fetuses. This model accounts for the differences in tissue sizes, blood flow rates, enzyme expression levels, plasma protein binding, and other physiological factors affecting the drugs' PK in both the pregnant woman and the fetus. The PBPKPlus™ module in GastroPlus® was used to model the PK of metoprolol, midazolam, and metronidazole for both non-pregnant and pregnant groups. For each of the three compounds, the model was first developed and validated against PK data in healthy non-pregnant volunteers and then applied to predict the PK in the pregnant groups. The model accurately described the PK in both the non-pregnant and pregnant groups and explained well the differences in the plasma concentration due to pregnancy. When available, the fetal plasma concentration, placenta, and fetal tissue concentrations were also predicted reasonably well at different stages of pregnancy. The work described the use of a PBPK approach for drug development and demonstrates the ability to predict differences in PK in pregnant subjects and fetal exposure for metabolically cleared compounds.

In Silico and In Vivo Pharmacokinetic Evaluation of 84-B10, a Novel Drug Candidate against Acute Kidney Injury and Chronic Kidney Disease

Acute kidney injury (AKI) and chronic kidney disease (CKD) have become public health problems due to high morbidity and mortality. Currently, drugs recommended for patients with AKI or CKD are extremely limited, and candidates based on a new mechanism need to be explored. 84-B10 is a novel 3-phenylglutaric acid derivative that can activate the mitochondrial protease, Lon protease 1 (LONP1), and may protect against cisplatin-induced AKI and unilateral ureteral obstruction- or 5/6 nephrectomy [5/6Nx]-induced CKD model. Preclinical studies have shown that 84-B10 has a good therapeutic effect, low toxicity, and is a good prospect for further development. In the present study, the UHPLC-MS/MS method was first validated then applied to the pharmacokinetic study and tissue distribution of 84-B10 in rats. Physicochemical properties of 84-B10 were then acquired in silico. Based on these physicochemical and integral physiological parameters, a physiological based pharmacokinetic (PBPK) model was developed using the PK-Sim platform. The fitting accuracy was estimated with the obtained experimental data. Subsequently, the validated model was employed to predict the pharmacokinetic profiles in healthy and chronic kidney injury patients to evaluate potential clinical outcomes. Cmax in CKD patients was about 3250 ng/mL after a single dose of 84-B10 (0.41 mg/kg), and Cmax,ss was 1360 ng/mL after multiple doses. This study may serve in clinical dosage setting in the future.