Dose-Response Determination in Multistage Endpoint Clinical Trials.

Area of Science:

• Clinical pharmacology
• Biostatistics
• Drug development

Background:

• Dose selection is critical in drug development, yet improper selection causes significant attrition in confirmatory studies.
• Inadequate resources and inappropriate statistical methods contribute to suboptimal dose selection in early-phase trials.

Purpose of the Study:

• To present a novel model-based dose-finding methodology for clinical trials.
• To improve the estimation of dose-response relationships and identify optimal therapeutic doses.

Main Methods:

• Utilizes all longitudinal data collected throughout a clinical trial.
• Employs a Markov chain model to handle correlated repeated measures within patients.
• Estimates the dose-response curve at any specified time point during the trial.

Main Results:

• The methodology enables estimation of target doses, such as the minimum effective dose.
• Demonstrated effectiveness through application in an actual Phase 2 clinical study.
• Validated through comprehensive simulation studies.

Conclusions:

• The proposed model-based approach enhances dose selection by leveraging all available patient data.
• This method offers a statistically robust framework for optimizing dose-ranging studies.
• Improved dose selection can potentially reduce attrition rates in later-stage clinical drug development.