| dc.description.abstract | Introduction: The epidermal growth factor receptor (EGFR) tyrosine kinase inhibitor (TKI) osimertinib has shown to be well-tolerated and effective in T790M-positive NSCLC patients. The tissue uptake of osimertinib can be predicted with a physiologically-based pharmacokinetic model (PBPK). However, validation of the model is necessary. In this research, we aim to calculate the number of patients, with a power of 80%, required for correct prediction of simulated PK parameters and required for a whole body PK study to detect difference in brain uptake between brain metastasis (BM) and glioblastoma (GBM) patients and for model validation.
Methods: Using 200 Monte Carlo simulations, the PK profile of osimertinib in blood and tissues was predicted after dosing 80 mg every 24 hours for 7 days. For each patient, patient characteristics were taken from a normal distribution, where variability in clearance (CL) and distribution volume (V) were 20.8% and 26.8%, respectively. Brain penetration (RPT) was 25% greater in the GBM group than in the BM group due to disruption of the blood brain barrier (BBB). The power calculation for the PK parameters is based on the method of Ogungbenro et. al, was performed in NONMEN and visualized in R (package ggplot).
Results: The NONMEM model gave individual PK parameters that were used for power calculations. When 29 patients are included, PK parameter estimates (CL and V) can be estimated with an accuracy level of 20% and power >80%. If 19 patients are included in the PK study, the power is 80% to detect a minimal effect of a 1.25-fold difference in biomarker performance between primary and metastatic patients.
Conclusion: The power calculations gave the number of patients needed for correct prediction of PK parameters and for a new PK study. The power calculation for PK parameters ensures that the simulated PK parameters better match reality. In the future, the calculations could be used for other studies. | |
