An example of optimal phase II design for exposure response modelling

Area of Science:

• Clinical Trials Methodology
• Pharmacometrics
• Biostatistics

Background:

• Phase II clinical studies require robust designs for effective exposure-response analysis.
• Traditional dose-response analysis may not fully capture individual variability.

Purpose of the Study:

• To apply optimal design methodology for Phase II clinical study design.
• To optimize the relationship between clinical endpoint and drug exposure (Area Under the Curve - AUC).

Main Methods:

• Utilized optimal design methodology to compare candidate Phase II study designs.
• Employed an algorithm for finding optimal designs based on D-optimality and standard error (SE).
• Modeled exposure-response using a sigmoidal Emax with baseline model.

Main Results:

• Recommended increasing the sample size from 540 to 700 participants.
• An optimal design with 700 participants showed a 16% gain over the reference design (equivalent to 812 individuals).
• Design performance was acceptable, but suboptimal for very flat exposure-response relationships.

Conclusions:

• Optimal design methodology enables prospective assessment of Phase II study performance.
• Revised study designs ensure greater value from subsequent exposure-response modeling.
• Methodology facilitates informed decisions on sample size and design for clinical trials.