An optimal stratified Simon two-stage design

Area of Science:

• Clinical Trials
• Biostatistics
• Oncology

Background:

• Phase II oncology trials increasingly evaluate targeted therapies in specific populations.
• Stratification based on molecular characteristics is crucial for these targeted trials.
• Existing adaptive designs need refinement for personalized medicine.

Purpose of the Study:

• To analyze and extend the Jones and Holmgren (JH) adaptive two-stage design for Phase II oncology trials.
• To introduce novel methods for controlling the familywise error rate (FWER) in stratified trials.
• To develop an optimal design minimizing expected sample size in targeted therapy evaluations.

Main Methods:

• Detailed study of the JH adaptive design, which uses interim analysis for population enrichment.
• Development of alternative FWER control methods (strong and weak sense).
• Introduction of a novel optimal design focused on minimizing expected sample size.

Main Results:

• The JH design allows adaptive enrichment of Phase II trial populations based on molecular characteristics.
• Extended methods provide robust control over familywise error rates in stratified designs.
• A new optimal design is proposed, offering efficiency gains through minimized sample size.

Conclusions:

• The enhanced JH design provides a valuable framework for Phase II trials in stratified medicine.
• The proposed methods improve the statistical rigor and efficiency of evaluating targeted oncology therapies.
• This work supports the development of personalized medicine through adaptive trial designs.