TY - JOUR

T1 - A novel statistical test for treatment differences in clinical trials using a response‐adaptive forward‐looking Gittins Index Rule

AU - Barnett, Helen Yvette

AU - Villar, Sofía S.

AU - Geys, Helena

AU - Jaki, Thomas

PY - 2023/3/31

Y1 - 2023/3/31

N2 - The most common objective for response-adaptive clinical trials is to seek to ensure that patients within a trial have a high chance of receiving the best treatment available by altering the chance of allocation on the basis of accumulating data. Approaches that yield good patient benefit properties suffer from low power from a frequentist perspective when testing for a treatment difference at the end of the study due to the high imbalance in treatment allocations. In this work we develop an alternative pairwise test for treatment difference on the basis of allocation probabilities of the covariate-adjusted response-adaptive randomization with forward-looking Gittins Index (CARA-FLGI) Rule for binary responses. The performance of the novel test is evaluated in simulations for two-armed studies and then its applications to multiarmed studies are illustrated. The proposed test has markedly improved power over the traditional Fisher exact test when this class of nonmyopic response adaptation is used. We also find that the test's power is close to the power of a Fisher exact test under equal randomization.

AB - The most common objective for response-adaptive clinical trials is to seek to ensure that patients within a trial have a high chance of receiving the best treatment available by altering the chance of allocation on the basis of accumulating data. Approaches that yield good patient benefit properties suffer from low power from a frequentist perspective when testing for a treatment difference at the end of the study due to the high imbalance in treatment allocations. In this work we develop an alternative pairwise test for treatment difference on the basis of allocation probabilities of the covariate-adjusted response-adaptive randomization with forward-looking Gittins Index (CARA-FLGI) Rule for binary responses. The performance of the novel test is evaluated in simulations for two-armed studies and then its applications to multiarmed studies are illustrated. The proposed test has markedly improved power over the traditional Fisher exact test when this class of nonmyopic response adaptation is used. We also find that the test's power is close to the power of a Fisher exact test under equal randomization.

KW - Applied Mathematics

KW - General Agricultural and Biological Sciences

KW - General Immunology and Microbiology

KW - General Biochemistry, Genetics and Molecular Biology

KW - General Medicine

KW - Statistics and Probability

KW - allocation probability

KW - inference

KW - nonmyopic

KW - power

KW - testing for superiority

U2 - 10.1111/biom.13581

DO - 10.1111/biom.13581

M3 - Journal article

C2 - 34669968

VL - 79

SP - 86

EP - 97

JO - Biometrics

JF - Biometrics

SN - 0006-341X

IS - 1

ER -