Optimizing randomized trial designs to distinguish which subpopulations benefit from treatment

Michael Aaron Rosenblum, M. J. Van Der Laan

Research output: Contribution to journalArticle

Abstract

It is a challenge to evaluate experimental treatments where it is suspected that the treatment effect may only be strong for certain subpopulations, such as those having a high initial severity of disease, or those having a particular gene variant. Standard randomized controlled trials can have low power in such situations. They also are not optimized to distinguish which subpopulations benefit from a treatment. With the goal of overcoming these limitations, we consider randomized trial designs in which the criteria for patient enrollment may be changed, in a preplanned manner, based on interim analyses. Since such designs allow data-dependent changes to the population enrolled, care must be taken to ensure strong control of the familywise Type I error rate. Our main contribution is a general method for constructing randomized trial designs that allow changes to the population enrolled based on interim data using a prespecified decision rule, for which the asymptotic, familywise Type I error rate is strongly controlled at a specified level α. As a demonstration of our method, we prove new, sharp results for a simple, two-stage enrichment design. We then compare this design to fixed designs, focusing on each design's ability to determine the overall and subpopulation-specific treatment effects.

Original languageEnglish (US)
Pages (from-to)845-860
Number of pages16
JournalBiometrika
Volume98
Issue number4
DOIs
StatePublished - Dec 2011

Fingerprint

Randomized Trial
Type I Error Rate
Treatment Effects
Therapeutics
Population
Fixed Design
Randomized Controlled Trial
Randomized Controlled Trials
Dependent Data
Decision Rules
Design
Randomized trial
disease severity
Genes
Demonstrations
Gene
Evaluate
methodology

Keywords

  • Adaptive design
  • Enrichment design
  • Group sequential design
  • Optimization
  • Patient-oriented research
  • Randomized trial
  • Subpopulation

ASJC Scopus subject areas

  • Agricultural and Biological Sciences(all)
  • Agricultural and Biological Sciences (miscellaneous)
  • Statistics and Probability
  • Mathematics(all)
  • Applied Mathematics
  • Statistics, Probability and Uncertainty

Cite this

Optimizing randomized trial designs to distinguish which subpopulations benefit from treatment. / Rosenblum, Michael Aaron; Van Der Laan, M. J.

In: Biometrika, Vol. 98, No. 4, 12.2011, p. 845-860.

Research output: Contribution to journalArticle

@article{5f1d9453c5d2466ca5d1b4a6c751bf1e,
title = "Optimizing randomized trial designs to distinguish which subpopulations benefit from treatment",
abstract = "It is a challenge to evaluate experimental treatments where it is suspected that the treatment effect may only be strong for certain subpopulations, such as those having a high initial severity of disease, or those having a particular gene variant. Standard randomized controlled trials can have low power in such situations. They also are not optimized to distinguish which subpopulations benefit from a treatment. With the goal of overcoming these limitations, we consider randomized trial designs in which the criteria for patient enrollment may be changed, in a preplanned manner, based on interim analyses. Since such designs allow data-dependent changes to the population enrolled, care must be taken to ensure strong control of the familywise Type I error rate. Our main contribution is a general method for constructing randomized trial designs that allow changes to the population enrolled based on interim data using a prespecified decision rule, for which the asymptotic, familywise Type I error rate is strongly controlled at a specified level α. As a demonstration of our method, we prove new, sharp results for a simple, two-stage enrichment design. We then compare this design to fixed designs, focusing on each design's ability to determine the overall and subpopulation-specific treatment effects.",
keywords = "Adaptive design, Enrichment design, Group sequential design, Optimization, Patient-oriented research, Randomized trial, Subpopulation",
author = "Rosenblum, {Michael Aaron} and {Van Der Laan}, {M. J.}",
year = "2011",
month = "12",
doi = "10.1093/biomet/asr055",
language = "English (US)",
volume = "98",
pages = "845--860",
journal = "Biometrika",
issn = "0006-3444",
publisher = "Oxford University Press",
number = "4",

}

TY - JOUR

T1 - Optimizing randomized trial designs to distinguish which subpopulations benefit from treatment

AU - Rosenblum, Michael Aaron

AU - Van Der Laan, M. J.

PY - 2011/12

Y1 - 2011/12

N2 - It is a challenge to evaluate experimental treatments where it is suspected that the treatment effect may only be strong for certain subpopulations, such as those having a high initial severity of disease, or those having a particular gene variant. Standard randomized controlled trials can have low power in such situations. They also are not optimized to distinguish which subpopulations benefit from a treatment. With the goal of overcoming these limitations, we consider randomized trial designs in which the criteria for patient enrollment may be changed, in a preplanned manner, based on interim analyses. Since such designs allow data-dependent changes to the population enrolled, care must be taken to ensure strong control of the familywise Type I error rate. Our main contribution is a general method for constructing randomized trial designs that allow changes to the population enrolled based on interim data using a prespecified decision rule, for which the asymptotic, familywise Type I error rate is strongly controlled at a specified level α. As a demonstration of our method, we prove new, sharp results for a simple, two-stage enrichment design. We then compare this design to fixed designs, focusing on each design's ability to determine the overall and subpopulation-specific treatment effects.

AB - It is a challenge to evaluate experimental treatments where it is suspected that the treatment effect may only be strong for certain subpopulations, such as those having a high initial severity of disease, or those having a particular gene variant. Standard randomized controlled trials can have low power in such situations. They also are not optimized to distinguish which subpopulations benefit from a treatment. With the goal of overcoming these limitations, we consider randomized trial designs in which the criteria for patient enrollment may be changed, in a preplanned manner, based on interim analyses. Since such designs allow data-dependent changes to the population enrolled, care must be taken to ensure strong control of the familywise Type I error rate. Our main contribution is a general method for constructing randomized trial designs that allow changes to the population enrolled based on interim data using a prespecified decision rule, for which the asymptotic, familywise Type I error rate is strongly controlled at a specified level α. As a demonstration of our method, we prove new, sharp results for a simple, two-stage enrichment design. We then compare this design to fixed designs, focusing on each design's ability to determine the overall and subpopulation-specific treatment effects.

KW - Adaptive design

KW - Enrichment design

KW - Group sequential design

KW - Optimization

KW - Patient-oriented research

KW - Randomized trial

KW - Subpopulation

UR - http://www.scopus.com/inward/record.url?scp=82255183905&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=82255183905&partnerID=8YFLogxK

U2 - 10.1093/biomet/asr055

DO - 10.1093/biomet/asr055

M3 - Article

C2 - 23049131

AN - SCOPUS:82255183905

VL - 98

SP - 845

EP - 860

JO - Biometrika

JF - Biometrika

SN - 0006-3444

IS - 4

ER -