(M-059) Physiologically based pharmacokinetic (PBPK) modeling of methotrexate in cerebrospinal fluid in humans

John Panetta, PhD – Biomedical Modeler, Department of Pharmacy and Pharmaceutical Sciences, St. Jude Children's Research Hospital; Markos Leggas, PhD – Director, Center for Translational Pharmacology, Department of Pharmacy and Pharmaceutical Sciences, St. Jude Children’s Research Hospital; Andrea Edginton, PhD – Professor, Hallman Director, Associate Dean (Faculty of Science), School of Pharmacy, University of Waterloo

Objectives
Methotrexate (MTX) is part of the treatment regimen for multiple cancers, including leukemia, which often metastasizes into the central nervous system (CNS). Intravenous MTX has limited distribution in the CNS (1, 2) requiring direct administration in the cerebrospinal fluid (CSF) to achieve therapeutic concentrations (2, 3). St. Jude Children’s Research Hospital (SJCRH) recently established dose-exposure relationships from retrospective data (1990 – 2019; n=22 patients with intraventricular administration, and n=15 with plasma and CSF levels) in children over 3 years and noted that there were too few CSF levels to assess in children under 3 years (2). The objective was to develop a PBPK model for MTX to predict CSF concentrations following intraventricular administration in adults and children.

Methods
Physiology-based sub-compartments of CSF were integrated into the PBPK model using PK-Sim and Mobi (version 11.2) (4). Optimization and model evaluation were based on literature data (3, 5). The ability of the model to predict CSF concentrations following intraventricular administration through the Ommaya reservoir, particularly in young children, was evaluated based on SJCRH data (2). Sensitivity analysis was performed to identify critical parameters for predicting the time CSF MTX will be above 1 μM.

Results
Following intravenous and intrathecal MTX administration, the absolute average fold error for plasma and CSF concentrations in children (median 4.8 years (5)) was 1.83, and the proportion of observed concentrations within a two-fold error range was 90%. Following Ommaya reservoir administration, 70%, 40%, and 44% of the observed plasma and CSF concentrations were within the 90% prediction interval (PI) for >10, [3, 10], and < 3-year-olds, respectively. The sensitivity analysis showed that the ratio between blood flow and CSF flow in the brain and the volume of subarachnoid space had the most influence on the time above 1 μM CSF MTX.

Conclusions
A PBPK model for MTX in CSF was developed using the literature data. The ability of the PBPK model to quantify CSF MTX levels following an Ommaya dose was lower in younger individuals. Uncertainties in the age related brain parameters likely influence accuracy and precision.

Citations: Citations:
1. Bhojwani D et al. Methotrexate-Induced Neurotoxicity and Leukoencephalopathy in Childhood Acute Lymphoblastic Leukemia. Journal of Clinical Oncology. 2014;32(9):949-59.
2. Gaietto A et al. Ommaya reservoir use in pediatric ALL and NHL: a review at St. Jude Children’s Research Hospital. Cancer Chemotherapy and Pharmacology. 2024.
3. Shapiro WR et al.. Methotrexate: Distribution in Cerebrospinal Fluid after Intravenous, Ventricular and Lumbar Injections. New England Journal of Medicine. 1975;293(4):161-6.
4. Saleh MAA et al.. Lumbar cerebrospinal fluid-to-brain extracellular fluid surrogacy is context-specific: insights from LeiCNS-PK3.0 simulations. J Pharmacokinet Pharmacodyn. 2021;48(5):725-41.
5. Evans WE et al. Methotrexate cerebrospinal fluid and serum concentrations after intermediate-dose methotrexate infusion. Clinical Pharmacology & Therapeutics. 1983;33(3):301-7.