Improved confidence intervals for a binomial parameter following a group sequential phase II clinical trial.

Area of Science:

• Clinical Trials Methodology
• Biostatistics
• Medical Research

Background:

• Group sequential procedures are standard in phase II clinical trials.
• Accurate interval estimation of true response rates is crucial upon trial completion.
• Existing methods, like Jennison and Turnbull (JT), can be conservative.

Purpose of the Study:

• To propose novel methods for interval estimation in group sequential phase II clinical trials.
• To improve upon the conservativeness of existing confidence intervals.
• To provide more precise estimates of the true response rate.

Main Methods:

• Utilized auxiliary statistics based on overall disease status.
• Developed new methods to address the discreteness of the underlying distribution.
• Compared proposed confidence intervals against Jennison and Turnbull (JT) intervals.

Main Results:

• Proposed confidence intervals demonstrate coefficients closer to the nominal level.
• The new methods yield confidence intervals with shorter average lengths than JT.
• Improved accuracy and efficiency in estimating true response rates.

Conclusions:

• The proposed methods offer superior interval estimation for group sequential phase II trials.
• These advancements lead to more reliable and informative clinical trial results.
• Enhanced statistical precision aids in better decision-making for treatment efficacy.