Tamara Jordens1,2, Maqsood Yaqub3, Mathilde Kouwenhoven4,5, Idris Bahce6, Harry Hendrikse2,3,7,8, Imke Bartelink1,2
1Department of Clinical Pharmacology and Pharmacy, Amsterdam UMC Location VUmc, 2Cancer Center Amsterdam, Imaging and Biomarkers, 3Department of Radiology and Nuclear Medicine, Amsterdam UMC Location VUmc, 4Department of Neurology, Amsterdam UMC Location VUmc, 5Cancer Center Amsterdam, Brain Tumor Center Amsterdam, 6Department of Pulmonary Medicine, Amsterdam UMC Location VUmc, 7Department of Hospital Pharmacy, Erasmus MC, University Medical Center Rotterdam, 8Department of Radiology and Nuclear Medicine, Erasmus MC, University Medical Center Rotterdam
Introduction: Up to 70% of non-small cell lung cancer (NSCLC) patients develop brain metastases (BM) during the course of their disease[1]. Osimertinib, a tyrosine kinase inhibitor (TKI), is a treatment option in EGFR mutation positive NSCLC patients with brain metastases, however at present no clear exposure-response relationship is known yet based on plasma data from clinical trials[2]. Target-site concentrations in the lung and brain tumor could be used in the prediction of response and therapeutic success in patients. In addition, physiologically-based pharmacokinetic (PBPK) modeling can be used to gain insights into drug distribution over various tissues in the body, including concentrations at target-sites. TKIs are known to be substrates for active efflux transporters present at the blood-brain-barrier (BBB), including P-glycoprotein and breast cancer resistance protein as found in vitro[3]. However, to what extend this affects reaching of effective in vivo brain tumor concentrations with therapeutic dosing is unknown. Objectives: Therefore, in this study we developed a semi-mechanistic PBPK model that describes the drug permeability of TKIs over the BBB, in order to predict brain concentrations of TKIs. Methods: An established whole-body PBPK model was extended to describe drug permeability over the BBB into healthy brain, cerebrospinal fluid (CSF) and brain metastasis tissue[4, 5]. The model extension included both passive and active transport over the BBB, as well as EGFR target-binding in the primary tumor, brain metastasis and relevant healthy organs. Moreover, changes in the BBB microvascular environment by the brain tumor was described, including increased microvascular blood flow, increased vessel density and increased microvascular volume in the brain, hence describing a separate blood-tumor barrier (BTB). Model parameter values were based on in vitro or in vivo findings in literature. However, since in vitro parameters for transport over the BBB inaccurately describe observed data, correction factors describing the system-differences in transport between in vitro cell-based cultures and NSCLC-BM patients were estimated. Estimation of the correction factors was done in NONMEM using quantitative PET-scan data of microdose radiolabeled [11C]C-osimertinib and therapeutic steady state concentrations measured in cerebrospinal fluid available in literature [6-9]. Results: Correction factors estimated based on available literature data showed that separate corrections were needed for passive and active transport of [11C]C-osimertinib over the healthy BBB, the BTB and the blood-CSF-barrier (BCSFB). For healthy BBB final estimates (RSE%) were 22.3 (7.9%) and 104 (3.2%) for influx and efflux respectively, while for the tumor these were 3.07 (0.2%) and 304 (1.9%), indicating that in vitro derived measures on passive and active transport underpredicted brain penetration and require large corrections for healthy brain and brain tumors. In contrast, however, the final estimates for the correction factors for influx and efflux over the BCSFB were found to be 0.843 (8.2%) and 0.492 (19.2%) respectively, indicating that the in vitro derived values overpredict CSF penetration by 1.2-2 fold. Model predictions of therapeutic PK steady state concentrations with the final estimates mentioned were within 2-fold of observed plasma concentrations in literature (Cssmean,pred = 558 nM vs Cssmean,obs = 485 nM, Prediction error = 14%)[10]. Drug regimen simulations indicated that with standard 80mg QD dosing osimertinib concentrations in the brain metastasis above mutated-EGFR IC50 values were achieved. However, these concentrations did not surpass IC50 values required to target a 50/50 expression ratio of wild-type/mutated-EGFR in the brain tumor, which is a mutation ratio often observed in the clinic. Conclusions: The extended PBPK-BBB model presented in this work has been developed to accurately predict osimertinib concentrations across various tissues at both microdose and therapeutic levels. This model could therefore serve a first step in the direction of personalized dosing of TKIs in brain tumor patients and may serve as a platform to assess the applicability of TKIs for brain tumor treatment. Moreover, this work shows that quantitative PET-data can be used to guide model building and validate model performance and accuracy, while in vitro BBB predictions require careful evaluation and correction before clinical application.
1. Schoenmaekers, J., et al., Dynamics of eligibility criteria for central nervous system metastases in non-small cell lung cancer randomized clinical trials over time: A systematic review. Crit Rev Oncol Hematol, 2021. 166: p. 103460. 2. Boosman, R.J., et al., Exposure-Response Analysis of Osimertinib in EGFR Mutation Positive Non-Small Cell Lung Cancer Patients in a Real-Life Setting. Pharm Res, 2022. 39(10): p. 2507-2514. 3. Colclough, N., et al., Preclinical Comparison of the Blood-brain barrier Permeability of Osimertinib with Other EGFR TKIs. Clin Cancer Res, 2021. 27(1): p. 189-201. 4. Bartelink, I.H., et al., Physiologically Based Pharmacokinetic (PBPK) Modeling to Predict PET Image Quality of Three Generations EGFR TKI in Advanced-Stage NSCLC Patients. Pharmaceuticals (Basel), 2022. 15(7). 5. van der Gaag, S., et al., Physiologically Based Pharmacokinetic Model of Tyrosine Kinase Inhibitors to Predict Target Site Penetration, with PET-Guided Verification. CPT Pharmacometrics Syst Pharmacol, 2025. 6. Ekman, S., et al., Brain exposure of osimertinib in patients with epidermal growth factor receptor mutation non-small cell lung cancer and brain metastases: A positron emission tomography and magnetic resonance imaging study. Clin Transl Sci, 2023. 16(6): p. 955-965. 7. Varrone, A., et al., A PET study in healthy subjects of brain exposure of (11)C-labelled osimertinib – A drug intended for treatment of brain metastases in non-small cell lung cancer. J Cereb Blood Flow Metab, 2020. 40(4): p. 799-807. 8. Yamaguchi, H., et al., A phase II study of Osimertinib for patients with radiotherapy-naive CNS metastasis of non-small cell lung cancer: treatment rationale and protocol design of the OCEAN study (LOGIK 1603/WJOG 9116L). BMC Cancer, 2020. 20(1): p. 370. 9. Xing, L., et al., Biomarkers of Osimertinib Response in Patients with Refractory, EGFR-T790M-positive Non-Small Cell Lung Cancer and Central Nervous System Metastases: The APOLLO Study. Clin Cancer Res, 2020. 26(23): p. 6168-6175. 10. Pilla Reddy, V., et al., Development, Verification, and Prediction of Osimertinib Drug-Drug Interactions Using PBPK Modeling Approach to Inform Drug Label. CPT Pharmacometrics Syst Pharmacol, 2018. 7(5): p. 321-330.
Reference: PAGE 33 (2025) Abstr 11484 [www.page-meeting.org/?abstract=11484]
Poster: Drug/Disease Modelling - Oncology