(W-028) Development and application of physiologically based pharmacokinetic model of darunavir on pregnant population

Thi Tham BUI, Master Degree of Pharmacy – PhD student, Chungnam National Univesity; Thi Lien NGO, Doctor of Philosophy – Postdoc, Chungnam National Univesity; Hwi-Yeol YUN, PhD, Adjunct Professor – Professor, Chungnam National Univesity; Jung-woo CHAE, PhD, Associate Professor – Professor, Chungnam National Univesity

Objectives: In pregnant women, the safety or efficacy of the chronic medications, such as anti-human immunodeficiency virus (HIV) drug may be altered compared to the non-pregnant population due to maternal physiological and anatomical changes. Our study thus aimed to create a physiologically based pharmacokinetic (PBPK) model that predicts the exposure of darunavir (DRV), an HIV-1 protease inhibitor, combined with a booster (ritonavir - RTV) on healthy subjects and to apply it on pregnant population.

Methods: All the physiochemical and PK properties of DRV were collected from literature; the clinical studies from relevant databases (PubMed, ClinicalTrials.gov, EU Clinical Trial Register and the reference lists) were extracted for PK data. The physiological parameters of each population were provided within the database of PK-Sim® and Mobi® softwares. The model of DRV was first developed in a healthy population when the drug was administered alone or in combination with RTV. A previously developed and fully validated PBPK model of RTV was used in this step without any modification. The kinetics and reactions of the involved enzyme (cytochrome P450 - CYP) and transporter (P-glycoprotein – P-gp), and glomerular filtration process of DRV were considered in the model. Structural parameters of the DRV model were estimated by fitting the model with observed clinical PK data. Subsequently, this base model was applied on a pregnancy population simulated by the software [1,2], with the parameters for pregnancy scaled based on literature.

Results: There were 25 eligible studies on HIV-negative healthy subjects, among which 17 were used for training and 8 for testing the model. Nineteen studies were selected on HIV positive patients, among which 10, 6 and 3 studies provide PK data of DRV/RTV use on the non-pregnant, pregnant, and both populations, respectively. These studies investigated the PK of DRV administered intravenously and orally (including tablet, solution, and suspension), of single or multiple doses, with or without oral RTV use concomitantly. Some PK parameters were refined to get the best fit, such as Michaelis–Menten kinetics (via K_m and V_max values) described the metabolism or transport of DRV by CYP3A4 or P-gp, respectively. More than 97% of the predicted AUC_last and C_max values were within the 2-fold-dimensions of the observed values. The values of mean relative deviation of predicted plasma level and geometric mean fold error of AUC_last and C_max calculated to assess the model’s performance were acceptable. This model predicted well the PK profiles of DRV when being administered to HIV-positive, non-pregnant and pregnant subjects.

Conclusions: Our developed model can help better understand the exposure of DRV. The final validated PBPK model in pregnancy will be applied for dose selection and as a basis for further PBPK model development of other medication use in the pregnant population.

Citations:

[1] Dallmann A, Ince I, Meyer M, Willmann S, Eissing T, Hempel G. Gestation-specific changes in the anatomy and physiology of healthy pregnant women: an extended repository of model parameters for physiologically based pharmacokinetic modeling in pregnancy. Clin Pharmacokinet 2017;56:1303–30. https://doi.org/10.1007/s40262-017-0539-z.

[2] Dallmann A, Solodenko J, Ince I, Eissing T. Applied concepts in PBPK modeling: how to extend an open systems pharmacology model to the special population of pregnant women. CPT Pharmacometrics Syst Pharmacol 2018;7:419–31. https://doi.org/10.1002/psp4.12300.