A two-stage Bayesian design with sample size reestimation and subgroup analysis for phase II binary response trials

Wei Zhong, Joe Koopmeiners, Brad Carlin

Research output: Contribution to journalArticle

7 Citations (Scopus)

Abstract

Frequentist sample size determination for binary outcome data in a two-arm clinical trial requires initial guesses of the event probabilities for the two treatments. Misspecification of these event rates may lead to a poor estimate of the necessary sample size. In contrast, the Bayesian approach that considers the treatment effect to be random variable having some distribution may offer a better, more flexible approach. The Bayesian sample size proposed by (Whitehead et al., 2008 [27]) for exploratory studies on efficacy justifies the acceptable minimum sample size by a "conclusiveness" condition. In this work, we introduce a new two-stage Bayesian design with sample size reestimation at the interim stage. Our design inherits the properties of good interpretation and easy implementation from Whitehead et al. (2008) [27], generalizes their method to a two-sample setting, and uses a fully Bayesian predictive approach to reduce an overly large initial sample size when necessary. Moreover, our design can be extended to allow patient level covariates via logistic regression, now adjusting sample size within each subgroup based on interim analyses. We illustrate the benefits of our approach with a design in non-Hodgkin lymphoma with a simple binary covariate (patient gender), offering an initial step toward within-trial personalized medicine.

Original languageEnglish (US)
Pages (from-to)587-596
Number of pages10
JournalContemporary Clinical Trials
Volume36
Issue number2
DOIs
StatePublished - Nov 1 2013

Fingerprint

Sample Size
Bayes Theorem
Precision Medicine
Non-Hodgkin's Lymphoma
Logistic Models
Clinical Trials
Therapeutics

Keywords

  • Bayesian design
  • Clinical trial
  • Personalized medicine
  • Predictive approach
  • Sample size reestimation
  • Subgroup analysis

Cite this

A two-stage Bayesian design with sample size reestimation and subgroup analysis for phase II binary response trials. / Zhong, Wei; Koopmeiners, Joe; Carlin, Brad.

In: Contemporary Clinical Trials, Vol. 36, No. 2, 01.11.2013, p. 587-596.

Research output: Contribution to journalArticle

@article{d768bc3725db47acb85129c40052734a,
title = "A two-stage Bayesian design with sample size reestimation and subgroup analysis for phase II binary response trials",
abstract = "Frequentist sample size determination for binary outcome data in a two-arm clinical trial requires initial guesses of the event probabilities for the two treatments. Misspecification of these event rates may lead to a poor estimate of the necessary sample size. In contrast, the Bayesian approach that considers the treatment effect to be random variable having some distribution may offer a better, more flexible approach. The Bayesian sample size proposed by (Whitehead et al., 2008 [27]) for exploratory studies on efficacy justifies the acceptable minimum sample size by a {"}conclusiveness{"} condition. In this work, we introduce a new two-stage Bayesian design with sample size reestimation at the interim stage. Our design inherits the properties of good interpretation and easy implementation from Whitehead et al. (2008) [27], generalizes their method to a two-sample setting, and uses a fully Bayesian predictive approach to reduce an overly large initial sample size when necessary. Moreover, our design can be extended to allow patient level covariates via logistic regression, now adjusting sample size within each subgroup based on interim analyses. We illustrate the benefits of our approach with a design in non-Hodgkin lymphoma with a simple binary covariate (patient gender), offering an initial step toward within-trial personalized medicine.",
keywords = "Bayesian design, Clinical trial, Personalized medicine, Predictive approach, Sample size reestimation, Subgroup analysis",
author = "Wei Zhong and Joe Koopmeiners and Brad Carlin",
year = "2013",
month = "11",
day = "1",
doi = "10.1016/j.cct.2013.03.011",
language = "English (US)",
volume = "36",
pages = "587--596",
journal = "Contemporary Clinical Trials",
issn = "1551-7144",
publisher = "Elsevier Inc.",
number = "2",

}

TY - JOUR

T1 - A two-stage Bayesian design with sample size reestimation and subgroup analysis for phase II binary response trials

AU - Zhong, Wei

AU - Koopmeiners, Joe

AU - Carlin, Brad

PY - 2013/11/1

Y1 - 2013/11/1

N2 - Frequentist sample size determination for binary outcome data in a two-arm clinical trial requires initial guesses of the event probabilities for the two treatments. Misspecification of these event rates may lead to a poor estimate of the necessary sample size. In contrast, the Bayesian approach that considers the treatment effect to be random variable having some distribution may offer a better, more flexible approach. The Bayesian sample size proposed by (Whitehead et al., 2008 [27]) for exploratory studies on efficacy justifies the acceptable minimum sample size by a "conclusiveness" condition. In this work, we introduce a new two-stage Bayesian design with sample size reestimation at the interim stage. Our design inherits the properties of good interpretation and easy implementation from Whitehead et al. (2008) [27], generalizes their method to a two-sample setting, and uses a fully Bayesian predictive approach to reduce an overly large initial sample size when necessary. Moreover, our design can be extended to allow patient level covariates via logistic regression, now adjusting sample size within each subgroup based on interim analyses. We illustrate the benefits of our approach with a design in non-Hodgkin lymphoma with a simple binary covariate (patient gender), offering an initial step toward within-trial personalized medicine.

AB - Frequentist sample size determination for binary outcome data in a two-arm clinical trial requires initial guesses of the event probabilities for the two treatments. Misspecification of these event rates may lead to a poor estimate of the necessary sample size. In contrast, the Bayesian approach that considers the treatment effect to be random variable having some distribution may offer a better, more flexible approach. The Bayesian sample size proposed by (Whitehead et al., 2008 [27]) for exploratory studies on efficacy justifies the acceptable minimum sample size by a "conclusiveness" condition. In this work, we introduce a new two-stage Bayesian design with sample size reestimation at the interim stage. Our design inherits the properties of good interpretation and easy implementation from Whitehead et al. (2008) [27], generalizes their method to a two-sample setting, and uses a fully Bayesian predictive approach to reduce an overly large initial sample size when necessary. Moreover, our design can be extended to allow patient level covariates via logistic regression, now adjusting sample size within each subgroup based on interim analyses. We illustrate the benefits of our approach with a design in non-Hodgkin lymphoma with a simple binary covariate (patient gender), offering an initial step toward within-trial personalized medicine.

KW - Bayesian design

KW - Clinical trial

KW - Personalized medicine

KW - Predictive approach

KW - Sample size reestimation

KW - Subgroup analysis

UR - http://www.scopus.com/inward/record.url?scp=84888130583&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84888130583&partnerID=8YFLogxK

U2 - 10.1016/j.cct.2013.03.011

DO - 10.1016/j.cct.2013.03.011

M3 - Article

VL - 36

SP - 587

EP - 596

JO - Contemporary Clinical Trials

JF - Contemporary Clinical Trials

SN - 1551-7144

IS - 2

ER -