A Bayesian Framework for Optimizing Amikacin Therapy in Critically Ill Patients With Cancer

Area of Science:

• Pharmacology
• Oncology
• Critical Care Medicine

Background:

• Amikacin (AMK) treats gram-negative infections in intensive care units (ICUs).
• High interindividual variability in AMK levels can cause toxicity or ineffectiveness.
• Need for optimized AMK dosing in critically ill cancer patients.

Purpose of the Study:

• Establish a roadmap for AMK therapeutic drug monitoring in critically ill cancer patients.
• Develop a Bayesian estimator for bedside application.
• Individualize AMK doses to improve efficacy and safety.

Main Methods:

• Observational retrospective study of oncological ICU patients receiving AMK.
• Nonlinear mixed-effects modeling of plasma concentrations.
• Covariate analysis including patient data and comorbidities.
• Comparison with existing AMK dosing methods using Bland-Altman analysis.

Main Results:

• A one-compartment model with linear elimination best described AMK pharmacokinetics.
• Estimated glomerular filtration rate (eGFR) was a significant covariate of clearance (CL), explaining 16% of variability.
• Body weight influenced the volume of distribution (4% variability).
• The developed model reduced bias in individual CL estimates and was implemented in DoseMeRx.

Conclusions:

• A Bayesian estimation method effectively individualizes AMK doses for critically ill cancer patients.
• Further refinement is possible with additional renal function biomarkers.
• Improved predictive performance can enhance AMK therapy in this population.