Randomized phase II selection design with order constrained strata.

Area of Science:

• Clinical Trials Methodology
• Biostatistics
• Drug Development

Background:

• Phase II trials often include heterogeneous patient subgroups, impacting statistical efficiency.
• Current randomized phase II trial designs often overlook patient stratification in sample size calculations.
• Single-arm phase II trials can improve efficiency by incorporating patient heterogeneity.

Purpose of the Study:

• To propose novel statistical methods for randomized phase II trials that leverage natural order constraints in stratified populations.
• To enhance statistical efficiency in randomized phase II selection and screening designs.
• To address the lack of consideration for patient stratification in randomized phase II trial sample size calculations.

Main Methods:

• Developed methods utilizing natural order constraints for stratified populations in randomized phase II designs.
• Focused on randomized phase II selection designs, with generalizability to screening designs.
• Considered both binary and time-to-event outcomes for statistical analysis.

Main Results:

• The proposed methods improve statistical efficiency compared to those not using order constraints.
• Demonstrated improved probabilities of correct selection in simulations and real-world examples.
• Showed a reduction in required sample sizes for randomized phase II trials.

Conclusions:

• Utilizing order constraints in stratified populations offers significant statistical advantages for randomized phase II trials.
• The developed methods provide a more efficient approach to sample size determination and correct selection.
• This work contributes to more efficient and effective drug development through improved clinical trial design.