Yingying Tian(1), Christian Mattonet(2), Lucia Nogova(2), Uwe Fuhr(1)
(1)Department I of Pharmacology, University Hospital of Cologne, Cologne, Germany; (2)Lung Cancer Group Cologne, Center for Integrated Oncology, University Hospital Cologne, Cologne, Germany
Background: The combination of the mTOR inhibitor everolimus and the multi-kinase inhibitor sorafenib may increase anti-tumor efficacy, therefore the drugs are intended to be given in combination. Some factors including variable oral bioavailability and drug-drug interactions lead to high inter-individual pharmacokinetic variability of everolimus. Understanding the sources of variability may support personalized cancer management.
Objective: To develop a pharmacokinetic model for everolimus able to describe its exposure profile in patients with solid tumors.
Methods: Pharmacokinetic profiles of everolimus were obtained from a phase-I clinical trial. In its initial dose finding phase, 17 patients with relapsed solid tumors were treated with escalating everolimus doses (2.5, 5, 7.5 or 10 mg daily). A fixed dose of sorafenib 400 mg bid was added from day 15. Additionally, 13 KRAS mutated non-small cell lung cancer patients were treated in an extension phase with the maximum tolerated dose 7.5 mg/day everolimus with 400 mg sorafenib bid thereafter until progression. Everolimus concentrations were measured on day 5, 14, and 29 (pre-dose, 0.5, 1, 2, 3, 4, 8, 12, 24 h relative to morning dose) and less dense sampling was taken for the follow-up. Data were analyzed using nonlinear mixed-effects modeling implemented in NONMEM V7.3.0. The impact of different covariates such as demographics and laboratory tests on the pharmacokinetic of everolimus were evaluated and quantified.
Results: A two-compartment model with linear absorption and elimination was developed. It predicted that the total clearance would increase 55% in the presence of sorafenib, which was comparable to non-compartment analysis that the AUC and Cmax of everolimus showed a 20 – 40% reduction. Among the covariate relationships tested, everolimus pharmacokinetic characteristics were not influenced by age, body weight or sex, but clearance decreased with higher blood bilirubin. By including a power covariate relationship model the unexplained inter-individual variability could be reduced by 13.4% for clearance.
Conclusion: The current model was suitable to predict the pharmacokinetics of everolimus. It allows a better description of everolimus exposure when co-administered with sorafenib in solid tumor patients. The mechanism for the lower exposure to everolimus when co-administered with sorafenib remains to be elucidated.
Reference: PAGE 25 (2016) Abstr 5867 [www.page-meeting.org/?abstract=5867]
Poster: Drug/Disease modeling - Oncology