Parameter estimation for sigmoid E_max models in exposure-response relationship.

Area of Science:

• Pharmacokinetics and Pharmacodynamics
• Computational Biology
• Clinical Trial Design

Background:

• Population pharmacokinetic/pharmacodynamic (PK/PD) analysis is crucial for drug development.
• Understanding the limitations of PK/PD study designs is essential for accurate parameter estimation.
• The sigmoid Emax model is commonly used to describe concentration-effect relationships.

Purpose of the Study:

• To explore the limitations of population PK/PD analysis using clinical study data.
• To guide the construction of appropriate PK/PD study designs for precise parameter estimation.
• To evaluate the impact of different doses and Hill coefficients on PK/PD analysis.

Main Methods:

• Simulated data from seven escalating doses with varying Hill coefficients.
• Applied a 1-compartment PK model and a sigmoid Emax model.
• Assessed parameter estimation bias and precision using relative bias and relative root mean square error.

Main Results:

• Single-dose PK/PD parameters were accurately estimated when Cmax exceeded 85% of EC50, except for EC50 parameters at low Hill coefficients.
• Parameter estimation performance was suboptimal in multiple-dose scenarios.
• Dense sampling design highlighted the influence of concentration range relative to EC50 and sigmoidicity on estimation.

Conclusions:

• The relative range of sampled concentrations to EC50 and the sigmoidicity significantly impact PK/PD parameter estimation.
• Dense sampling designs require careful consideration of dose and drug characteristics (Hill coefficient) for optimal results.
• Further refinement of PK/PD study designs is needed, particularly for estimating EC50 and its variability.