Response-adaptive treatment allocation for survival trials with clustered right-censored data

Area of Science:

• Biostatistics
• Clinical Trials
• Medical Research

Background:

• Comparing treatments with survival outcomes often involves clustered data, where responses within a cluster are correlated.
• Traditional balanced designs allocate equal numbers of participants to each treatment arm.
• Responsive-adaptive designs are gaining acceptance due to their efficiency and ethical advantages.

Purpose of the Study:

• To develop a novel response-adaptive treatment allocation scheme for survival trials with clustered data.
• To compare the proposed scheme against conventional balanced designs.

Main Methods:

• Developed a response-adaptive randomization procedure tailored for clustered survival data.
• Utilized simulation studies to evaluate the performance of the proposed scheme.
• Applied the method to a redesigned clinical study for validation.

Main Results:

• The proposed response-adaptive scheme allows more participants to receive the more effective treatment.
• The statistical power for comparing treatment efficacies remains highly competitive compared to balanced designs.
• The simulation study and clinical study redesign demonstrated the advantages of the new procedure.

Conclusions:

• The developed response-adaptive allocation scheme is a superior alternative to balanced designs for clustered survival trials.
• This method enhances patient outcomes by favoring the better treatment while preserving statistical rigor.
• The findings support the adoption of adaptive randomization in clinical research involving clustered data.