Isis Van herteryck 1,2, Evelyn Dhont 1,3, Mathias Devreese 2, Bram De Wilde 4, Eline Hermans 1,2,5, Peter De Paepe 1,6, Zhiyuan Tan 1, Pieter De Cock 1,7
1 Department of Basic and Applied Medical Sciences, Faculty of Medicine and Health Sciences, Ghent University (Ghent, Belgium), 2 Department of Pathobiology, Pharmacology and Zoological Medicine, Faculty of Veterinary Medicine, Ghent University (Merelbeke, Belgium), 3 Department of Paediatric Intensive Care, Ghent University Hospital (Ghent , Belgium), 4 Department of Paediatric Hematology-Oncology, Ghent University Hospital (Ghent, Belgium ), 5 Department of Paediatric Nephrology, Ghent University Hospital (Ghent , Belgium ), 6 Department of Emergency Medicine, Ghent University Hospital (Ghent , Belgium ), 7 Department of Pharmacy, Ghent University Hospital (Ghent , Belgium )
Introduction:
Amikacin is a critical antibiotic for treating severe infections in immunocompromised children. Pediatric patients are particularly vulnerable due to developmental variability affecting dose optimization1. In severely ill children, pathophysiological changes such as augmented renal clearance (ARC) can further alter drug exposure2,3. Amikacin’s narrow therapeutic window necessitates precise dosing to ensure antibacterial efficacy while minimizing nephrotoxicity and ototoxicity. Traditional dosing recommendations target the maximum concentration (Cmax) to minimum inhibitory concentration (MIC) ratio, but there is a shift towards using area under the curve (AUC)/MIC as the efficacy metric4,5.
Objectives:
This study aimed to (i) characterize amikacin pharmacokinetics (PK) and (ii) evaluate the impact of different PK/PD targets on PK/PD target attainment, and (iii) develop an evidence-based dosing nomogram.
Methods:
An observational PK study was conducted at Ghent University Hospital in children ≤16 years, admitted to the pediatric intensive care unit or oncology ward unit, receiving standard of care amikacin therapy (25 mg/kg intravenously over 30 minutes every 24 hours). Population PK analysis was performed using MONOLIX (Lixoft, version 2024R1). Model validation employed bootstrapping (n=1000) and the prediction-corrected visual predictive check. Probability of target attainment (PTA) was evaluated with the MrgSolve package in RStudio (Posit, version 4.5.2), simulating 3,000 patients per subgroup. Optimal dosing regimens were derived based on efficacy targets (AUC/MIC ≥ 62.5 and/or Cmax/MIC ≥ 8) and the safety threshold (Cmin ≤ 3 mg/L). An MIC of 2 mg/L for Enterobacter was evaluated in this analysis2,3.
Results:
A total of 210 samples from 38 patients (median [range] age 4.2 [0.05–14] years and eGFR of 151 [48-284] mL/min/1.73 m²) were analyzed. A two-compartment PK model with weight allometric scaling on clearance and volume of distribution parameters best described the data. Furthermore, clearance was further significantly influenced by the estimated glomerular filtration rate (eGFR), the combined creatinine–cystatin C equation by Zappitelli outperformed other pediatric formulas for eGFR. Oncology disease reduced the central volume of distribution (coefficient of -0.44). Population PK parameters were: clearance 8.25 L/h, central volume 24 L, intercompartmental flow 1.82 L/h, and peripheral volume 14.76 L. Simulations demonstrated that the current dosing regimen achieved a PTA of 66.8% at an MIC of 2 mg/L. Dosing requirements were strongly dependent on the PK/PD target used. Higher doses were required to meet the AUC/MIC > 62.5 target, resulting in PTA values of 82–90% across subgroups, approaching the recommended 90% threshold.
Conclusions:
A population pharmacokinetic model of amikacin in critically ill and oncology children was successfully developed. A two-compartment model best described amikacin disposition in this population. A creatinine- and cystatin C-based eGFR estimate (Zappitelli formula) provided the most accurate description of amikacin elimination. Dosing strategies incorporating eGFR and oncology status improved target attainment compared with current regimens. Regimens based on AUC/MIC targets are more stringent than Cmax/MIC-based approaches in this population and result in lower PTA values at higher MICs.
References:
[1] H. K. Batchelor and J. F. Marriott, “Paediatric pharmacokinetics: Key considerations,” British ournal of Clinical Pharmacology, vol.79, pp. 395-404,(2015).
[2] A. Alwhaibi, M. M. Almutairi, S. Alsanea, M. Al-Jeraisy, A. Alsultan, and M. Abouelkheir, “Optimizing Gentamicin Dosing in Pediatric Oncology Patients,” Pediatric Infectious Disease Journal, (2025).
[3] A. F. Zuppa and J. S. Barrett, “Pharmacokinetics and Pharmacodynamics in the Critically Ill Child,” vol. 55, pp.735-755, (2008).
[4] N. Dia et al., “Dose Optimization of Amikacin in the Emergency Department: A Population Pharmacokinetics Simulation Study,” Therapeutic Drug Monitoring, vol. 47, pp. 353–362, (2025).
[5] M. A. A. van der Veer et al., “Amikacin dosing in neonates: evaluation of target attainment using a simplified and complex pharmacokinetic model-derived dosing regimen in clinical practice,” Antimicrobial Agents and Chemotherapy, vol. 69, (2025).
Reference: PAGE 34 (2026) Abstr 12013 [www.page-meeting.org/?abstract=12013]
Poster: Drug/Disease Modelling - Oncology