Statistical power calculations for mixed pharmacokinetic study designs using a population approach

Area of Science:

• Pharmacokinetics
• Clinical Trial Design
• Statistical Modeling

Background:

• Population pharmacokinetic (PopPK) analysis enables simultaneous modeling of dense and sparse data.
• Simulation-based power calculations are crucial for determining optimal sample sizes to detect covariate effects.

Purpose of the Study:

• To extend the Monte Carlo Mapped Power method for sample size calculations in mixed pharmacokinetic studies (combining dense and sparse data).
• To develop a workflow for efficient and cost-effective pharmacokinetic study design.

Main Methods:

• Simulated pharmacokinetic data for a hypothetical drug and dihydroartemisinin.
• Applied the Monte Carlo Mapped Power method to assess power for varying sample sizes in dense and sparse designs.
• Evaluated study designs with >80% statistical power for cost-effectiveness.

Main Results:

• The simulation-based power calculation methodology successfully determined sample sizes for mixed pharmacokinetic study designs.
• Demonstrated applicability for both hypothetical and real-world (dihydroartemisinin) pharmacokinetic data.
• Identified cost-effective study designs meeting desired statistical power.

Conclusions:

• The extended Monte Carlo Mapped Power method is effective for sample size determination in mixed pharmacokinetic studies.
• This approach facilitates the design of robust and efficient pharmacokinetic studies.
• Contributes to optimizing clinical trial design by balancing statistical power and cost.