Population Pharmacokinetics and Optimal Sampling Strategy for Model-Based Precision Dosing of Melphalan in Patients...

High-dose melphalan is an important component of conditioning regimens for patients undergoing hematopoietic stem cell transplantation. The current dosing strategy based on body surface area results in a high incidence of oral mucositis and gastrointestinal and liver toxicity. Pharmacokinetically guided dosing will individualize exposure and help minimize overexposure-related toxicity.

The purpose of this study was to develop a population pharmacokinetic model and optimal sampling strategy.

A population pharmacokinetic model was developed with NONMEM using 98 observations collected from 15 adult patients given the standard dose of 140 or 200 mg/m² by intravenous infusion. The determinant-optimal sampling strategy was explored with PopED software. Individual area under the curve estimates were generated by Bayesian estimation using full and the proposed sparse sampling data. The predictive performance of the optimal sampling strategy was evaluated based on bias and precision estimates. The feasibility of the optimal sampling strategy was tested using pharmacokinetic data from five pediatric patients.

A two-compartment model best described the data. The final model included body weight and creatinine clearance as predictors of clearance. The determinant-optimal sampling strategies (and windows) were identified at 0.08 (0.08-0.19), 0.61 (0.33-0.90), 2.0 (1.3-2.7), and 4.0 (3.6-4.0) h post-infusion. An excellent correlation was observed between area under the curve estimates obtained with the full and the proposed four-sample strategy (R ² = 0.98; p < 0.01) with a mean bias of -2.2% and precision of 9.4%. A similar relationship was observed in children (R ² = 0.99; p < 0.01).

The developed pharmacokinetic model-based sparse sampling strategy promises to achieve the target area under the curve as part of precision dosing.