Improved two-stage group sequential procedures for testing a secondary endpoint after the primary endpoint achieves significance

Huiling Li, Jianming Wang, Xiaolong Luo, Janis Grechko, Christopher Jennison

Research output: Contribution to journalArticle

Abstract

In two-stage group sequential trials with a primary and a secondary endpoint, the overall type I error rate for the primary endpoint is often controlled by an α-level boundary, such as an O'Brien-Fleming or Pocock boundary. Following a hierarchical testing sequence, the secondary endpoint is tested only if the primary endpoint achieves statistical significance either at an interim analysis or at the final analysis. To control the type I error rate for the secondary endpoint, this is tested using a Bonferroni procedure or any α-level group sequential method. In comparison with marginal testing, there is an overall power loss for the test of the secondary endpoint since a claim of a positive result depends on the significance of the primary endpoint in the hierarchical testing sequence. We propose two group sequential testing procedures with improved secondary power: the improved Bonferroni procedure and the improved Pocock procedure. The proposed procedures use the correlation between the interim and final statistics for the secondary endpoint while applying graphical approaches to transfer the significance level from the primary endpoint to the secondary endpoint. The procedures control the familywise error rate (FWER) strongly by construction and this is confirmed via simulation. We also compare the proposed procedures with other commonly used group sequential procedures in terms of control of the FWER and the power of rejecting the secondary hypothesis. An example is provided to illustrate the procedures.

Original languageEnglish
Pages (from-to)893-902
Number of pages10
JournalBiometrical Journal
Volume60
Issue number5
Early online date7 Jun 2018
DOIs
Publication statusPublished - 1 Sep 2018

Fingerprint

Group Sequential
Sequential Procedure
Testing
Familywise Error Rate
Bonferroni
Type I Error Rate
Interim Analysis
Sequential Testing
Group Testing
Sequential Methods
Significance level
Statistical Significance
Statistics

Keywords

  • closed testing
  • graphical approach
  • group sequential
  • hierarchical testing
  • multiple endpoints

ASJC Scopus subject areas

  • Statistics and Probability
  • Statistics, Probability and Uncertainty

Cite this

Improved two-stage group sequential procedures for testing a secondary endpoint after the primary endpoint achieves significance. / Li, Huiling; Wang, Jianming; Luo, Xiaolong; Grechko, Janis; Jennison, Christopher.

In: Biometrical Journal, Vol. 60, No. 5, 01.09.2018, p. 893-902.

Research output: Contribution to journalArticle

@article{e94d43fa99f54203b4c8d2a9872a69b9,
title = "Improved two-stage group sequential procedures for testing a secondary endpoint after the primary endpoint achieves significance",
abstract = "In two-stage group sequential trials with a primary and a secondary endpoint, the overall type I error rate for the primary endpoint is often controlled by an α-level boundary, such as an O'Brien-Fleming or Pocock boundary. Following a hierarchical testing sequence, the secondary endpoint is tested only if the primary endpoint achieves statistical significance either at an interim analysis or at the final analysis. To control the type I error rate for the secondary endpoint, this is tested using a Bonferroni procedure or any α-level group sequential method. In comparison with marginal testing, there is an overall power loss for the test of the secondary endpoint since a claim of a positive result depends on the significance of the primary endpoint in the hierarchical testing sequence. We propose two group sequential testing procedures with improved secondary power: the improved Bonferroni procedure and the improved Pocock procedure. The proposed procedures use the correlation between the interim and final statistics for the secondary endpoint while applying graphical approaches to transfer the significance level from the primary endpoint to the secondary endpoint. The procedures control the familywise error rate (FWER) strongly by construction and this is confirmed via simulation. We also compare the proposed procedures with other commonly used group sequential procedures in terms of control of the FWER and the power of rejecting the secondary hypothesis. An example is provided to illustrate the procedures.",
keywords = "closed testing, graphical approach, group sequential, hierarchical testing, multiple endpoints",
author = "Huiling Li and Jianming Wang and Xiaolong Luo and Janis Grechko and Christopher Jennison",
year = "2018",
month = "9",
day = "1",
doi = "10.1002/bimj.201700231",
language = "English",
volume = "60",
pages = "893--902",
journal = "Biometrical Journal",
issn = "0323-3847",
publisher = "Wiley-VCH Verlag",
number = "5",

}

TY - JOUR

T1 - Improved two-stage group sequential procedures for testing a secondary endpoint after the primary endpoint achieves significance

AU - Li, Huiling

AU - Wang, Jianming

AU - Luo, Xiaolong

AU - Grechko, Janis

AU - Jennison, Christopher

PY - 2018/9/1

Y1 - 2018/9/1

N2 - In two-stage group sequential trials with a primary and a secondary endpoint, the overall type I error rate for the primary endpoint is often controlled by an α-level boundary, such as an O'Brien-Fleming or Pocock boundary. Following a hierarchical testing sequence, the secondary endpoint is tested only if the primary endpoint achieves statistical significance either at an interim analysis or at the final analysis. To control the type I error rate for the secondary endpoint, this is tested using a Bonferroni procedure or any α-level group sequential method. In comparison with marginal testing, there is an overall power loss for the test of the secondary endpoint since a claim of a positive result depends on the significance of the primary endpoint in the hierarchical testing sequence. We propose two group sequential testing procedures with improved secondary power: the improved Bonferroni procedure and the improved Pocock procedure. The proposed procedures use the correlation between the interim and final statistics for the secondary endpoint while applying graphical approaches to transfer the significance level from the primary endpoint to the secondary endpoint. The procedures control the familywise error rate (FWER) strongly by construction and this is confirmed via simulation. We also compare the proposed procedures with other commonly used group sequential procedures in terms of control of the FWER and the power of rejecting the secondary hypothesis. An example is provided to illustrate the procedures.

AB - In two-stage group sequential trials with a primary and a secondary endpoint, the overall type I error rate for the primary endpoint is often controlled by an α-level boundary, such as an O'Brien-Fleming or Pocock boundary. Following a hierarchical testing sequence, the secondary endpoint is tested only if the primary endpoint achieves statistical significance either at an interim analysis or at the final analysis. To control the type I error rate for the secondary endpoint, this is tested using a Bonferroni procedure or any α-level group sequential method. In comparison with marginal testing, there is an overall power loss for the test of the secondary endpoint since a claim of a positive result depends on the significance of the primary endpoint in the hierarchical testing sequence. We propose two group sequential testing procedures with improved secondary power: the improved Bonferroni procedure and the improved Pocock procedure. The proposed procedures use the correlation between the interim and final statistics for the secondary endpoint while applying graphical approaches to transfer the significance level from the primary endpoint to the secondary endpoint. The procedures control the familywise error rate (FWER) strongly by construction and this is confirmed via simulation. We also compare the proposed procedures with other commonly used group sequential procedures in terms of control of the FWER and the power of rejecting the secondary hypothesis. An example is provided to illustrate the procedures.

KW - closed testing

KW - graphical approach

KW - group sequential

KW - hierarchical testing

KW - multiple endpoints

UR - http://www.scopus.com/inward/record.url?scp=85052735377&partnerID=8YFLogxK

U2 - 10.1002/bimj.201700231

DO - 10.1002/bimj.201700231

M3 - Article

VL - 60

SP - 893

EP - 902

JO - Biometrical Journal

JF - Biometrical Journal

SN - 0323-3847

IS - 5

ER -