Identifying optimal approaches to early termination in two-stage biomarker validation studies

Alexander M. Kaizer, Joe Koopmeiners

Research output: Contribution to journalArticle

Abstract

Group sequential study designs have been proposed as an approach to conserve resources in biomarker validation studies. Typically, group sequential study designs allow both ‘early termination to reject the null hypothesis’ and ‘early termination for futility’ if there is evidence against the alternative hypothesis. In contrast, several researchers have advocated for using group sequential study designs that allow only early termination for futility in biomarker validation studies because of the desire to obtain a precise estimate of marker performance at study completion. This suggests a loss function that heavily weights the precision of the estimate that is obtained at study completion at the expense of an increased sample size when there is strong evidence against the null hypothesis. We propose a formal approach to comparing designs that allow early termination for futility, superiority or both by developing a loss function that incorporates the expected sample size under the null and alternative hypotheses, as well as the mean-squared error of the estimate that is obtained at study completion. We then use our loss function to compare several candidate designs and derive optimal two-stage designs for a recently reported validation study of a novel prostate cancer biomarker.

Original languageEnglish (US)
Pages (from-to)187-199
Number of pages13
JournalJournal of the Royal Statistical Society. Series C: Applied Statistics
Volume66
Issue number1
DOIs
StatePublished - Jan 1 2017

Fingerprint

Early Termination
Biomarkers
Group Sequential
Sequential Design
Loss Function
Completion
Null hypothesis
Sample Size
Estimate
Two-stage Design
Prostate Cancer
Conserve
Alternatives
Mean Squared Error
Null
Resources
Termination
Loss function
Evidence
Design

Keywords

  • Diagnostic biomarkers
  • Group sequential testing
  • Optimal designs
  • Prostate cancer

Cite this

Identifying optimal approaches to early termination in two-stage biomarker validation studies. / Kaizer, Alexander M.; Koopmeiners, Joe.

In: Journal of the Royal Statistical Society. Series C: Applied Statistics, Vol. 66, No. 1, 01.01.2017, p. 187-199.

Research output: Contribution to journalArticle

@article{fe5a885970494e6c865aad36832f9d56,
title = "Identifying optimal approaches to early termination in two-stage biomarker validation studies",
abstract = "Group sequential study designs have been proposed as an approach to conserve resources in biomarker validation studies. Typically, group sequential study designs allow both ‘early termination to reject the null hypothesis’ and ‘early termination for futility’ if there is evidence against the alternative hypothesis. In contrast, several researchers have advocated for using group sequential study designs that allow only early termination for futility in biomarker validation studies because of the desire to obtain a precise estimate of marker performance at study completion. This suggests a loss function that heavily weights the precision of the estimate that is obtained at study completion at the expense of an increased sample size when there is strong evidence against the null hypothesis. We propose a formal approach to comparing designs that allow early termination for futility, superiority or both by developing a loss function that incorporates the expected sample size under the null and alternative hypotheses, as well as the mean-squared error of the estimate that is obtained at study completion. We then use our loss function to compare several candidate designs and derive optimal two-stage designs for a recently reported validation study of a novel prostate cancer biomarker.",
keywords = "Diagnostic biomarkers, Group sequential testing, Optimal designs, Prostate cancer",
author = "Kaizer, {Alexander M.} and Joe Koopmeiners",
year = "2017",
month = "1",
day = "1",
doi = "10.1111/rssc.12163",
language = "English (US)",
volume = "66",
pages = "187--199",
journal = "Journal of the Royal Statistical Society. Series C: Applied Statistics",
issn = "0035-9254",
publisher = "Wiley-Blackwell",
number = "1",

}

TY - JOUR

T1 - Identifying optimal approaches to early termination in two-stage biomarker validation studies

AU - Kaizer, Alexander M.

AU - Koopmeiners, Joe

PY - 2017/1/1

Y1 - 2017/1/1

N2 - Group sequential study designs have been proposed as an approach to conserve resources in biomarker validation studies. Typically, group sequential study designs allow both ‘early termination to reject the null hypothesis’ and ‘early termination for futility’ if there is evidence against the alternative hypothesis. In contrast, several researchers have advocated for using group sequential study designs that allow only early termination for futility in biomarker validation studies because of the desire to obtain a precise estimate of marker performance at study completion. This suggests a loss function that heavily weights the precision of the estimate that is obtained at study completion at the expense of an increased sample size when there is strong evidence against the null hypothesis. We propose a formal approach to comparing designs that allow early termination for futility, superiority or both by developing a loss function that incorporates the expected sample size under the null and alternative hypotheses, as well as the mean-squared error of the estimate that is obtained at study completion. We then use our loss function to compare several candidate designs and derive optimal two-stage designs for a recently reported validation study of a novel prostate cancer biomarker.

AB - Group sequential study designs have been proposed as an approach to conserve resources in biomarker validation studies. Typically, group sequential study designs allow both ‘early termination to reject the null hypothesis’ and ‘early termination for futility’ if there is evidence against the alternative hypothesis. In contrast, several researchers have advocated for using group sequential study designs that allow only early termination for futility in biomarker validation studies because of the desire to obtain a precise estimate of marker performance at study completion. This suggests a loss function that heavily weights the precision of the estimate that is obtained at study completion at the expense of an increased sample size when there is strong evidence against the null hypothesis. We propose a formal approach to comparing designs that allow early termination for futility, superiority or both by developing a loss function that incorporates the expected sample size under the null and alternative hypotheses, as well as the mean-squared error of the estimate that is obtained at study completion. We then use our loss function to compare several candidate designs and derive optimal two-stage designs for a recently reported validation study of a novel prostate cancer biomarker.

KW - Diagnostic biomarkers

KW - Group sequential testing

KW - Optimal designs

KW - Prostate cancer

UR - http://www.scopus.com/inward/record.url?scp=84973657635&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84973657635&partnerID=8YFLogxK

U2 - 10.1111/rssc.12163

DO - 10.1111/rssc.12163

M3 - Article

VL - 66

SP - 187

EP - 199

JO - Journal of the Royal Statistical Society. Series C: Applied Statistics

JF - Journal of the Royal Statistical Society. Series C: Applied Statistics

SN - 0035-9254

IS - 1

ER -