Zinnia P. Parra-Guillen (1), Markus Joerger (2), Benjamin Berger (3), Manuel Haschke (3), Charlotte Kloft (1)
Zinnia P. Parra-Guillen (1), Markus Joerger (2), Benjamin Berger (3), Manuel Haschke (3), Charlotte Kloft (1)
Objectives: Erlotinib is a tyrosine kinase inhibitor widely used in the treatment of non small-cell lung cancer. Similar to other molecules of this class, large inter-patient variability has been shown in clinical practice [1]. Erlotinib is mainly metabolised by CYP3A4 and to a lesser extent by CYP1A2, both enzymes varying largely across individuals. Therefore, enzymatic information could be used to individualise therapy and reduce variability. A prospective, nonrandomized, pharmacological cohort study was performed to assess the association between erlotinib exposure and cytochrome activity using midazolam and caffeine as probe drugs for characterising CYP3A4 and CYP1A2 phenotype, respectively (NCT01402089). The aim of this work was to describe the pharmacokinetics (PK) of the two probe drugs used in the study and their major metabolites (OH-midazolam and paraxanthine), as well as to develop the base model for erlotinib and its main metabolite, OSI-420.
Methods: Patients received erlotinib 150 mg orally once daily (n=36, dose reductions were allowed). Information regarding dosing times were extracted from the patients’ dosing diaries. On day 1 and after an overnight fast, 2 mg oral midazolam and 100 mg oral caffeine after 36 h of abstinence were administered. Plasma samples for the probe drugs, erlotinib and the metabolites were collected after 1,2,3,4 and 6 h. Additional measurements of erlotinib and OSI-420 were available at steady-state. Data analysis was performed using NONMEM 7.3.
Results: Midazolam PK was best described with a 2 compartment disposition model, while caffeine and erlotinib were adequately described using a 1 compartment model. Due to the fast drug absorption of midazolam and erlotinib, this process could not be identified based only on the available data, and the absorption rate constants were fixed to the literature values of 3.18 [2] and 1.09 h-1 [3] respectively. Midazolam and erlotinib clearance were estimated to be solely dependent of metabolite formation (additional elimination was not identifiable). Finally, paraxanthine pharmacokinetics could not be characterised due to the short collection sampling times (only absorption phase).
Conclusions: Midazolam and caffeine probe drug PK have been adequately described. Different relevant metrics such as clearance, area under the curve or parent/metabolite ratio can be now computed and explored as potential covariates to explain inter-individual variability in erlotinib clearance.
References:
[1] H.J. Klümpen, C.F. Samer, R.H.J. Mathijssen et al. Moving towards dose individualization of tyrosine kinase inhibitors. Cancer Treatment Reviews, 37: 251-260 (2011).
[2] A. van Rongen, L. Kervezee, M.J.E. Brill et al. Population Pharmacokinetic Model characterizing 24-Hour Variation in the Pharmacokinetics of Oral and Intravenous Midazolam in Healthy Volunteers. CPT: Pharmacometrics Syst. Pharmacol., 4: 454-464 (2015).
[3] E. Petit Jean, T. Buclin, M. Guidi et al. Erlotinib: Another Candidate for the Therapeutic Drug Monitoring of Targeted Therapy of Cancer? A Pharmacokinetic and Pharmacodynamic Systematic Review of Literature. Ther. Drug Monit. 37: 2-21 (2015).
Reference: PAGE 25 () Abstr 6005 [www.page-meeting.org/?abstract=6005]
Poster: Drug/Disease modeling - Oncology