From Therapeutic Drug Monitoring to Model-Informed Precision Dosing: A Review of Busulfan Dosing Optimization in Pediatric Hematopoietic Stem Cell Transplantation.

Abstract

Background/Objectives: Busulfan is a cornerstone conditioning agent for pediatric hematopoietic stem cell transplantation (HSCT), yet its narrow therapeutic index and substantial pharmacokinetic (PK) variability complicate dosing. This review synthesizes current evidence on pediatric busulfan PK, exposure-guided dosing, and the transition from conventional therapeutic drug monitoring (TDM) to model-informed precision dosing (MIPD). Methods: We conducted a focused narrative search of PubMed, Web of Science, and ScienceDirect from database inception to August 2026 and synthesized relevant literature on pediatric busulfan PK, exposure-response relationships, TDM, population PK modeling, Bayesian methods, and emerging quantitative approaches. Results: Body size is the primary determinant of busulfan clearance in children, while maturation, underlying disease, concomitant medications, and pharmacogenetic variation may contribute additional variability, although their effects are inconsistent across populations. Busulfan exposure is associated with HSCT outcomes and toxicity; however, reported targets differ by dosing regimen, conditioning intensity, disease category, and exposure metric. Reliable exposure estimation requires accurate sampling records and validated bioanalysis, while limited sampling strategies may improve clinical feasibility. Integrating population PK models with Bayesian estimation enables individualized dose adjustment, although routine implementation requires external validation and compatibility with local workflows. Emerging approaches, including physiologically based PK, PK/PD, and machine learning models, show promise but lack sufficient clinical validation for routine use. Conclusions: TDM remains the foundation of pediatric busulfan precision dosing, while MIPD provides a framework for integrating patient characteristics, measured concentrations, and validated models. Standardized workflows, population-appropriate exposure targets, prospective validation, and evidence of improved clinical outcomes are essential for broader implementation.