Population Pharmacokinetic Model Evaluation with a Small Real-World Dataset Versus a Large Virtual Dataset: Does...

Area of Science:

• Pharmacokinetics and Pharmacodynamics
• Clinical Pharmacology
• Drug Development

Background:

• Previous simulation studies indicated minimal impact of sample size on external population pharmacokinetic (PK) model evaluation.
• The applicability of these findings to real-world clinical data required validation.

Purpose of the Study:

• To validate simulation-based findings using actual clinical data.
• To assess the external evaluation of population PK models with small clinical datasets.

Main Methods:

• Collected clinical data from elderly patients receiving piperacillin/tazobactam.
• Simulated a virtual population of 1000 patients.
• Externally evaluated a population PK model using both clinical and simulated datasets.
• Assessed model performance using bias, imprecision, goodness-of-fit plots, and prediction-corrected visual predictive checks.

Main Results:

• External evaluation of a population PK model was performed on a small clinical dataset (13 patients) and a simulated dataset.
• No significant differences were observed in prediction error distributions between the clinical and simulated datasets.
• Goodness-of-fit plots and visual predictive checks indicated similar model misspecification for both datasets.

Conclusions:

• Small clinical datasets are adequate for the external evaluation of population PK models.
• Findings support the use of limited clinical data in PK model assessment.