Physiologically based pharmacokinetic modelling of the impact of hepatic impairment on glucuronidation – a case study with midazolam.
Agustos C. Ozbey (1, 2), Stephan Krähenbühl (3), Pieter Annaert (1), Stephen Fowler (1), Neil Parrott (1), Kenichi Umehara (1).
(1) Pharmaceutical Sciences, Roche Pharma Research and Early Development, Roche Innovation Center Basel, Grenzacherstrasse 124, 4070, Basel, Switzerland, (2) Drug Delivery and Disposition, Department of Pharmaceutical and Pharmacological Sciences, KU Leuven, Belgium, (3) Division of Clinical Pharmacology & Toxicology, University Hospital Basel, Switzerland.
Introduction: Hepatic impairment may be associated with a reduced rate of drug metabolism. In particular, it has been demonstrated that the activities of hepatic cytochrome P450 enzymes (CYPs) are decreased in patients with hepatic impairment due to liver cirrhosis1, but few studies have examined in detail the effect of cirrhosis, and therefore hepatic impairment, on uridine glucuronosyltransferase (UGT) activities. In case of midazolam metabolism, UGTs are involved both in primary glucuronidation of midazolam and secondary glucuronidation of 1’-OH-midazolam after CYP3A oxidation.
Objective: In this project we propose to use physiologically-based pharmacokinetic (PBPK) modeling to evaluate the impact of liver cirrhosis on the primary and secondary glucuronidation of midazolam.
Method: Results from Duthaler et al (2022) clinical study including 36 patients with liver cirrhosis and 12 demographically matched patients without hepatic impairment, were used1. Study subjects received an oral dose of 2 mg of midazolam and venous blood samples were obtained 5 min, 15 min, 30 min, 45 min, 1 h, 2 h, 3 h, 4 h, 6 h, 8 h, 12 h and 24 hours after administration. Concentration of midazolam, its primary metabolites midazolam-N-glucuronide (UGT1A4) and 1’-OH-midazolam (CYP3A4/5) and its secondary metabolite 1’-OH-midazolam-O-glucuronide (UGT2B7) were determined.
A full PBPK models for midazolam, its primarily metabolites (midazolam-N-glucuronide and 1’-OH midazolam) and the secondary metabolite (1’-OH-midazolam-O-glucuronide) was established on SimCYP® software based on previously validated SimCYP® model for midazolam. Pharmacological parameters (molecular weight, pKa, logP) of midazolam-N-glucuronide, 1’-OH-midazolam and 1’-OH-midazolam-O-glucuronide were obtained in literature and if not available were predicted with ADMET predictor® based on chemical structure. Renal clearances of midazolam-N-glucuronide and 1’-OH-midazolam-O-glucuronide were estimated based on the available clinical data, and metabolic intrinsic clearance of 1’-OH-midazolam glucuronidation into 1’-OH-midazolam-O-glucuronide via UGT2B7 was measured in vitro. Finally, in vitro measurement were done to obtain midazolam and its metabolites free fraction in plasma. Total and free plasma concentration-time profiles were predicted for midazolam and all the metabolites stated as above, and the corresponding areas under the curves (AUCs) were calculated with linear trapezoidal method. The model was first verified in non-cirrhotic (control) subjects before simulating the pharmacokinetics in cirrhotic patients of the Child-Pugh (CP) classes: CP-A, CP-B and CP-C. SimCYP® integrates effects of hepatic impairment by modifying multiple physiological parameters such as: enzyme abundance in tissues, concentration of circulating plasma proteins (e.g. albumin), liver volume and renal function and hepatic portal and arterial blood flows.
The total and free AUC ratio of midazolam-N-glucuronide and 1’-OH-midazolam-O-glucuronide to midazolam and the total and free AUC ratio of midazolam-N-glucuronide to 1’-OH-midazolam-O-glucuronide metabolites were calculated to observe the impact of cirrhosis on glucuronidation (UGT1A4 and UGT2B7).
Results: The developed PBPK model was successfully able to predict the exposure to midazolam and its metabolites in non-cirrhotic and cirrhotic patients, within 2-fold. The total AUC ratio of midazolam-N-glucuronide and 1’-OH-midazolam-O-glucuronide to midazolam decreased with the cirrhosis status (Child-Pugh stage), indicating a decrease of glucuronidation due to liver cirrhosis. In comparison, the AUC ratio of midazolam-N-glucuronide to 1’-OH-midazolam-O-glucuronide metabolites was not impacted by the Child-Pugh stage, suggesting that the impact of liver cirrhosis on UGT1A4 and UGT2B7 glucuronidation may be similar to the impact on CYP3A activity. This was also predicted successfully by the model. Similar results were obtained with ratios calculated with free AUCs.
Conclusion: We were successful in showing that UGT2B7 and UGT1A4 glucuronidation is impacted by the liver cirrhosis and that PBPK modeling is able to predict this impact. We envisage that the model framework established for midazolam can be used in future to predict the effect of cirrhosis on the pharmacokinetics of other drugs cleared via UGT1A4 and UGT2B7.
Reference:
[1] Duthaler U, Bachmann F, Suenderhauf C, Grandinetti T, Pfefferkorn F, Haschke M, Hruz P, Bouitbir J, Krähenbühl S. Liver Cirrhosis Affects the Pharmacokinetics of the Six Substrates of the Basel Phenotyping Cocktail Differently. Clin Pharmacokinet. 2022 Jul;61(7):1039-1055. doi: 10.1007/s40262-022-01119-0. Epub 2022 May 16. PMID: 35570253; PMCID: PMC9287224.