Approaches to handling data when a phase II trial deviates from the pre-specified Simon's two-stage design

Area of Science:

• Clinical Trials
• Biostatistics
• Oncology Drug Development

Background:

• Simon's two-stage designs are standard for early-phase cancer therapy trials.
• These designs have fixed rules and sample sizes, limiting flexibility.
• Real-world studies often face challenges with patient recruitment rates.

Purpose of the Study:

• To develop flexible approaches for phase IIA cancer trials when patient accrual deviates from Simon's designs.
• To provide statistical methods for addressing deviations in sample size and event numbers.
• To guide reporting of p-values in modified trial designs.

Main Methods:

• Examining four specific scenarios of deviation from Simon's two-stage design.
• Utilizing conditional probabilities to analyze the impact of recruitment delays.
• Developing strategies for adapting trial parameters mid-study.

Main Results:

• Conditional probabilities offer a framework to manage deviations from fixed two-stage designs.
• The proposed methods allow for adjustments to sample size and event thresholds.
• Guidance is provided on maintaining statistical integrity despite design modifications.

Conclusions:

• Flexible adaptation of Simon's two-stage designs is feasible and necessary for realistic cancer trials.
• Conditional probability methods enable robust decision-making under recruitment uncertainty.
• Clear reporting of p-values is crucial when modifying trial designs due to recruitment issues.