Pharmacokinetic Model Based on Stochastic Simulation and Estimation for Therapeutic Drug Monitoring of Teicoplanin in Korean Neutropenic Hematopoietic Stem Cell Transplant Recipients

Purpose: This study aimed to develop the first population pharmacokinetic (PK) model of teicoplanin for Korean neutropenic patients after hematopoietic stem cell transplantation (HSCT), to improve the precision of therapeutic drug monitoring (TDM) and support individualized dosing strategies. Patients and Methods: TDM data comprising 568 trough concentrations were retrospectively collected from 405 post-HSCT patients at a single hospital in Korea. Because only trough samples were available, 360 simulated concentration–time points were generated from two Korean PK models at six post-dose timepoints (0.3, 1, 1.5, 2.5, 4, and 6 hr) to supplement the dataset. The combined dataset was analyzed in NONMEM to construct a base model for stochastic simulation and estimation (SSE). Covariates were identified using generalized additive modeling and stepwise covariate selection. Model performance was assessed with bootstrap and prediction-corrected visual predictive checks. Finally, the SSE procedure was implemented, with parameter re-estimation repeated 1000 times to obtain stable estimates despite the trough-only design. Results: The final model identified creatinine clearance (eGFR) and albumin as significant covariates, reflecting the effects of renal function and protein binding. Model evaluation confirmed robust predictive performance within the 720-hour TDM window, although prediction intervals widened at later times due to sparse data. Based on the SSE procedure, the final clearance estimate was 1.26 L/h, higher than values reported in non-HSCT populations (0.63– 0.69 L/h) and consistent with the elevated mean eGFR in this cohort (113 mL/min/1.73 m²), partially due to reduced muscle mass and altered serum creatinine in HSCT patients. Conclusion: This SSE-based PK model provides a practical framework for precision TDM of teicoplanin in Korean HSCT patients. By incorporating renal function and protein binding, it supports individualized dosing in this vulnerable population. Prospective validation is warranted to confirm its broader clinical application. Keywords: teicoplanin, population pharmacokinetics, therapeutic drug monitoring, hematopoietic stem cell transplantation, stochastic simulation and estimation (SSE)