Population pharmacokinetic model of high-dose methotrexate in Malaysian adults with hematological malignancies

Multiple population pharmacokinetic (PopPK) models have been developed across diverse populations and disease groups for high-dose methotrexate (HDMTX), and published external evaluations of PopPK models have demonstrated limited transferability across populations. This study aims to develop a PopPK model describing HDMTX pharmacokinetics in Malaysian adults with hematological malignancies via routine clinical data.

Methotrexate (MTX) plasma concentrations at 48, 72, and 96 h were retrospectively collected from 91 HDMTX administrations with a total of 206 concentrations. Concentrations below the lower limit of quantification (BLQ) were handled using M3. Potential covariates evaluated included demographic factors (age, sex, body surface area), renal function, liver function parameters, hematocrit and treatment-related factors (protocol and infusion duration). Renal function was explored via the estimated glomerular filtration rate (eGFR), BSA-adjusted eGFR (GFR), and creatinine clearance derived from the Cockcroft-Gault equation, to assess their relative influence on clearance. Forward inclusion followed by backwards elimination was used for covariate selection. Model adequacy was assessed using goodness-of-fit plots and visual predictive check (VPC).

A two-compartment model with literature-informed distribution parameters, first-order elimination and combined residual error best described the current data. Limited information during the distribution phase was mitigated by fixing V1, V2, and Q to the literature-derived values. Clearance (CL), normalized to the median covariate values was parameterized as: CL (L/h) = 11.13 × (Alb/39)^0.7. Albumin was a significant predictor of CL (ΔOFV = -41.81 from 119.92 in base model).

This model characterizes HDMTX pharmacokinetics in Malaysian adults with hematological malignancies, with albumin identified as a significant covariate on clearance. Simulation findings suggest that albumin may support risk stratification for delayed MTX elimination across different dosing scenarios.