Construction of an optimal destructive sampling design for noncompartmental AUC estimation

Area of Science:

• Pharmacokinetics and Toxicokinetics
• Statistical Modeling and Experimental Design

Background:

• Toxicokinetic studies often use destructive sampling, requiring careful planning.
• Accurate estimation of the area under the concentration-time curve (AUC) is crucial for toxicokinetic analysis.
• Optimizing sampling strategies can improve data precision and reduce resource use.

Purpose of the Study:

• To determine optimal sampling designs for destructive toxicokinetic studies.
• To identify the ideal number and placement of time points and replicates for AUC estimation.
• To minimize the scaled mean squared error (SMSE) of noncompartmental AUC estimates.

Main Methods:

• Utilized a prior population pharmacokinetic model.
• Employed a destructive sampling design approach.
• Developed an algorithm using Mathematica for symbolic resolution and numerical minimization.
• Evaluated designs based on varying sample sizes, time points, and animal distributions.
• Compared design efficiencies through theoretical analysis and simulations.

Main Results:

• Identified optimal sampling designs balancing accuracy and precision for AUC estimation.
• Demonstrated that design efficiency varies with sample size, number of time points, and animal distribution.
• The implemented algorithm effectively constructs, validates, and compares sampling designs.
• Provided an example application of the developed methodology.

Conclusions:

• The developed method offers efficient tools for optimizing sampling designs in destructive toxicokinetic studies.
• Accurate and precise AUC estimation can be achieved through carefully selected sampling points and replicates.
• This approach supports better experimental planning and resource allocation in toxicological research.