A confirmatory basket design considering non-inferiority and superiority testing

Area of Science:

• Clinical trial design
• Biostatistics
• Drug development

Background:

• Rare diseases and disease subtypes often require combined analysis for confirmatory trials.
• Novel drugs may show marginal overall efficacy but superior benefits in specific patient subgroups (baskets).
• Testing both non-inferiority and superiority hypotheses efficiently in a single trial is challenging.

Purpose of the Study:

• To propose and evaluate a two-stage adaptive trial design for rare diseases.
• To enable simultaneous testing of overall non-inferiority and subgroup superiority.
• To facilitate an efficient drug registration pathway with potential label extensions.

Main Methods:

• A two-stage adaptive design incorporating interim non-inferiority testing.
• Pruning and pooling of data between stages.
• Final stage superiority hypothesis testing on specific baskets.
• Simulation studies to examine operating characteristics.

Main Results:

• The proposed design allows for an early claim of non-inferiority.
• It enables a potential label extension based on demonstrated superiority in certain baskets.
• The design offers an improved benefit-risk profile through longer-term efficacy and safety data.
• Simulation studies confirm the design's operating characteristics.

Conclusions:

• The two-stage adaptive design provides an efficient and novel registration pathway for drugs targeting rare diseases or subtypes.
• This design is particularly suitable for emerging markets prioritizing efficient resource utilization.
• It supports claims of both overall non-inferiority and specific basket superiority.
• The approach enhances the benefit-risk profile through comprehensive efficacy and safety assessments.