Application of promising zone based on interim conditional power in unblinded sample size re-estimation in clinical trials
CHEN Chuixiong
WANG Wenwen
HUANG Manli
HUANG Yaru
WANG Ling
XIA Jielai
LI Chen
Abstract：Objective To use the promising zone(PZ)for unblinded sample size re-estimation in adaptive design clinical trials,thereby enhancing the overall research efficiency of the trial through adaptive sample size re-estimation.Methods With an example of a superiority test for comparing two means,the PZ was employed under the two-stage adaptive design,and the fixed sample size design(Fixed)and the group sequential design(GSD)were used as methodological references.By setting different initial estimates and true values of treatment differences,the type Ⅰ error rate,power,and average sample size of the PZ were compared,and application examples were given.Results The type Ⅰ error rate of the three trial designs were approximate.The PZ could improve the overall power of the trial compared with GSD and Fixed,but it consumed a larger sample size,and was suitable for scenarios where the initial difference was overestimated.When the initial difference was correctly estimated or underestimated,the improved power was similar to that of GSD due to effective early termination of the trial,but the sample size consumed was larger than that of GSD.Conclusion The PZ can effectively improve the power of clinical trials when the initial estimated sample size is too low,thereby avoiding the inability to detect the true therapeutic difference of drugs due to insufficient power.If there are relatively reliable preliminary data that can accurately estimate the initial difference,it is not recommended to employ the PZ in adaptive design clinical trials.
Keywords：promising zoneconditional powersample size re-estimationinterim analysisadaptive design
Publication Date：2025-05-31
Online Publishing Date：2026-08-26(First online date of this platform, not the publication date of the document)
Pages：7( 668-674 )