P-glycoprotein mediated drug-drug interaction at the blood-brain barrier on morphine brain distribution
Berfin Gülave (1), J.G. Coen van Hasselt (1), Elizabeth C.M. de Lange (1)
(1) Leiden Academic Centre for Drug Research, Leiden University, the Netherlands
Introduction/objectives: Transporter-mediated drug-drug interactions (DDIs) are studied in intestine, liver, and kidney. However, considerations of the effect of DDIs on drug exposure in the central nervous system (CNS) mediated by transporters at the blood-brain barrier (BBB) have not been extensively studied. So far, studies so far have shown contradictory impact of DDIs at the BBB [1–3].
In this study we focus on the effect of DDIs on CNS exposure to the analgesic morphine, which has as target site the brain extracellular fluid (brainECF), and thus must cross the BBB to achieve its intended effects. Transport of morphine across BBB is governed by passive diffusion, active efflux and saturable active influx [4–8]. The major efflux transporter involved in the transport of morphine at the BBB is P-glycoprotein (P-gp). Previous studies in rats have shown that inhibiting P-gp with elacridar resulted in increased brain-serum concentration ratio [9]. Because measuring the P-gp mediated DDI effect on human brainECF is more complicated than in animals, physiologically-based (PB) pharmacokinetic (PK) models can be used to evaluate the expected impact of P-gp-mediated drug interactions on morphine CNS exposure. In this context, we have previously developed the LeiCNS-PK3.0 model, which is a PBPK model that can predict unbound drug exposure in multiple central nervous system (CNS) compartments [10–12].
The aim of this study is to assess P-gp mediated DDI effect on morphine brainECF distribution, while considering the nonlinear BBB transport using the LeiCNS-PK3.0 model.
Methods: The previously published LeiCNS-PK3.0 model was extended to incorporate nonlinear P-gp mediated transport of morphine. In addition, inhibitory effects of perpetrator drugs were described by the modified Rowland-Martin equation, which uses an inhibition rate constant (Ki) or concentration at 50% inhibition (IC50). We identified 34 clinically relevant commonly used CNS and non-CNS targeted perpetrator drugs, for which we (1) calculated their expected clinical average steady-state concentrations associated with the typical clinical dosing schedule used, and (2) obtained Ki or IC50 values for P-gp from the ChEMBL database [13]. We then used the model to simulate intravenous administration of morphine for the clinical doses 10, 15 and 20 mg for 4 and 6 times a day for seven days total, with and without presence of each perpetrator drug, and compared the observed exposure measured as morphine brainECF area under the curve (AUC) and maximum concentrations (Cmax) between the exposure of morphine alone.
Results: Our results indicate only modest effects in exposure (<5% of CNS exposure) across dosing schedules for 25 out of 34 drugs. For amiodarone, amisulpride, diltiazem, haloperidol, propranolol, quetiapine, quinidine, verapamil and positive control tariquidar a clinically relevant inhibitory effect was observed (>10%). Moreover, we found that the relative AUC and Cmax at brainECF for these drugs increased with increasing dose and frequency up to 20% at 20mg, 6 times a day administration, which is likely caused by the nonlinear nature of morphine transport across the BBB.
Conclusions: Our analysis indicates that P-gp mediated DDIs may lead to clinically relevant changes in morphine CNS drug exposure. We moreover demonstrate how a model-based workflow can be used to evaluate the expected effects of DDIs in humans, through integration of drug- and systems-specific data. The results of our work can be used to further guide DDI-associated dose adjustments for morphine. The general workflow proposed, and our qualitative findings, may moreover also be relevant for other CNS targeted drugs which are substrates of the P-gp transporter.
Acknowledgements: This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No 848068. This manuscript reflects only the authors' view and the European Commission is not responsible for any use that may be made of the information it contains.
References:
[1] Kalvass JC et al. Clin Pharmacol Ther (2013) 94, 80-94
[2] Wagner CC et al. J Nucl med (2009) 50, 1954-1961
[3] Langer O et al. J Clin Pharmacol (2016) S143-S156
[4] Xie R et al. Br J Pharmacol (1999) 128, 563-568
[5] Letrent SP et al. Biochem Pharmacol (1999) 58, 951-957
[6] Chaves C et al. Curr Neuropharmacol (2017) 15, 1156-1173
[7] Tunblad K et al. Pharm Res (2003) 20, 618-623
[8] Groenendaal D et al. Br J Pharmacol (2007) 151, 701-712
[9] Letrent SP et al. Drug Metab Dispos (1999) 27, 827-834
[10] Yamamoto y et al. CPT Pharmacometrics Sys Pharmacol (2017) 6, 765-777
[11] Yamamoto Y et al. Pharm Res (2017) 34, 333-351
[12] Saleh MAA et al. J Pharmacokinet Pharmacodyn (2021) 48, 725-741
[13] Mendez D et al. Nucleic Acids Res (2019) 47, D930-D940