Development and Evaluation of a Population Pharmacokinetic Model for High-Dose Methotrexate in Japanese Pediatric...

Methotrexate is a key chemotherapeutic agent for treating pediatric acute lymphoblastic leukemia (ALL); however, its pharmacokinetics exhibit substantial inter-individual variability. Population pharmacokinetic (PPK) models have identified body weight and renal function as major determinants of methotrexate clearance. Nonetheless, the optimal PPK model for Japanese pediatric patients, considering hydration status, has not yet been established. This study aimed to develop a PPK model of high-dose methotrexate specifically for Japanese pediatric patients with ALL. A retrospective PPK analysis was performed in pediatric patients with ALL, who were treated with high-dose methotrexate at the Tokyo Medical and Dental University Hospital. Data from 38 patients (504 methotrexate concentrations; 27 for model development and 11 for validation) were analyzed using a two-compartment model in NONMEM. Predictive performances of the newly developed and previously published models were assessed using the validation dataset. In the final model, clearance was described as a function of body surface area and hydration volume rate per body surface area (HYDR), whereas intercompartmental clearance was associated with age and HYDR. Evaluation using the validation dataset showed that the developed model had the best predictive performance among the evaluated models (median prediction error, 1.79%; median absolute prediction error, 13.3%; percentage of predictions with an absolute prediction error ≤20%, 70.3%; percentage of predictions with an absolute prediction error ≤30%, 82.9%). Simulation revealed modestly improved target-attainment rates at relatively high hydration rates. These findings highlight the importance of fluid management and support the incorporation of hydration parameters into individualized high-dose methotrexate therapy.