A small n sequential multiple assignment randomized trial design for use in rare disease research

Area of Science:

• Clinical trial design
• Rare disease research
• Biostatistics

Background:

• Conducting clinical trials for rare diseases is challenging due to small patient populations.
• No standard therapy exists for the rare disease studied.
• A novel trial design was needed to efficiently evaluate multiple treatments.

Purpose of the Study:

• To illustrate the design of a small n sequential multiple assignment randomized trial (snSMART).
• To detail the sample size estimation and operating characteristics of the snSMART design.
• To present a viable clinical trial methodology for rare diseases.

Main Methods:

• The study employed computer simulations to investigate the performance of weighted Z statistics.
• The snSMART design allows for sequential treatment assignments and adaptive sample size allocation.
• Operating characteristics were evaluated under various weighting schemes for different trial stages.

Main Results:

• The snSMART design demonstrated increased statistical power compared to traditional single-stage trial designs.
• Simulation results showed how power is influenced by the weighting assigned to each stage of the trial.
• The effectiveness of the snSMART approach was quantified through performance metrics.

Conclusions:

• The snSMART design is a promising approach for rare disease clinical trials.
• This design is particularly suitable when multiple treatments are being considered and small sample sizes are required.
• The snSMART methodology offers an efficient strategy for rare disease drug development.