Sample size and design considerations for phase II clinical trials with correlated observations

Daniel J. Sargent, Jeff A Sloan, Steven S. Cha

Research output: Contribution to journalArticle

6 Citations (Scopus)

Abstract

Several methods are available for the design of phase II clinical trials with binary endpoints. A primary assumption for most methods is that observations on the endpoint of interest are uncorrelated; however, this assumption is violated if an individual patient provides more than one observation on the endpoint of interest. In such cases, one solution is to use a summary measure for each patient; an alternative solution is to perform an observation-specific analysis using a technique that properly accounts for the correlation. In this paper, we investigate the effect that ignoring correlated observations can have on the design properties of the typical phase II clinical trial. In cases in which an observation-specific analysis is desirable, we propose a simple method that adjusts a standard one- or two-stage phase II design to account for loss of information due to correlated observations. Simulations demonstrate that the method ensures that type I and type II error rate design requirements are met even in the presence of strong correlation. We develop the method in the context of phase II oncology trials, but the method applies readily to other clinical areas in which multiple responses per patient are of interest. Control Clin Trials 1999;20:242-252 Copyright (C) 1999 Elsevier Science Inc.

Original languageEnglish (US)
Pages (from-to)242-252
Number of pages11
JournalControlled Clinical Trials
Volume20
Issue number3
DOIs
StatePublished - 1999

Fingerprint

Phase II Clinical Trials
Sample Size
Observation

Keywords

  • Correlated binomial
  • Multistage design

ASJC Scopus subject areas

  • Pharmacology

Cite this

Sample size and design considerations for phase II clinical trials with correlated observations. / Sargent, Daniel J.; Sloan, Jeff A; Cha, Steven S.

In: Controlled Clinical Trials, Vol. 20, No. 3, 1999, p. 242-252.

Research output: Contribution to journalArticle

@article{0b5834312d4845d09bb065176658d588,
title = "Sample size and design considerations for phase II clinical trials with correlated observations",
abstract = "Several methods are available for the design of phase II clinical trials with binary endpoints. A primary assumption for most methods is that observations on the endpoint of interest are uncorrelated; however, this assumption is violated if an individual patient provides more than one observation on the endpoint of interest. In such cases, one solution is to use a summary measure for each patient; an alternative solution is to perform an observation-specific analysis using a technique that properly accounts for the correlation. In this paper, we investigate the effect that ignoring correlated observations can have on the design properties of the typical phase II clinical trial. In cases in which an observation-specific analysis is desirable, we propose a simple method that adjusts a standard one- or two-stage phase II design to account for loss of information due to correlated observations. Simulations demonstrate that the method ensures that type I and type II error rate design requirements are met even in the presence of strong correlation. We develop the method in the context of phase II oncology trials, but the method applies readily to other clinical areas in which multiple responses per patient are of interest. Control Clin Trials 1999;20:242-252 Copyright (C) 1999 Elsevier Science Inc.",
keywords = "Correlated binomial, Multistage design",
author = "Sargent, {Daniel J.} and Sloan, {Jeff A} and Cha, {Steven S.}",
year = "1999",
doi = "10.1016/S0197-2456(98)00058-0",
language = "English (US)",
volume = "20",
pages = "242--252",
journal = "Controlled Clinical Trials",
issn = "0197-2456",
publisher = "Elsevier BV",
number = "3",

}

TY - JOUR

T1 - Sample size and design considerations for phase II clinical trials with correlated observations

AU - Sargent, Daniel J.

AU - Sloan, Jeff A

AU - Cha, Steven S.

PY - 1999

Y1 - 1999

N2 - Several methods are available for the design of phase II clinical trials with binary endpoints. A primary assumption for most methods is that observations on the endpoint of interest are uncorrelated; however, this assumption is violated if an individual patient provides more than one observation on the endpoint of interest. In such cases, one solution is to use a summary measure for each patient; an alternative solution is to perform an observation-specific analysis using a technique that properly accounts for the correlation. In this paper, we investigate the effect that ignoring correlated observations can have on the design properties of the typical phase II clinical trial. In cases in which an observation-specific analysis is desirable, we propose a simple method that adjusts a standard one- or two-stage phase II design to account for loss of information due to correlated observations. Simulations demonstrate that the method ensures that type I and type II error rate design requirements are met even in the presence of strong correlation. We develop the method in the context of phase II oncology trials, but the method applies readily to other clinical areas in which multiple responses per patient are of interest. Control Clin Trials 1999;20:242-252 Copyright (C) 1999 Elsevier Science Inc.

AB - Several methods are available for the design of phase II clinical trials with binary endpoints. A primary assumption for most methods is that observations on the endpoint of interest are uncorrelated; however, this assumption is violated if an individual patient provides more than one observation on the endpoint of interest. In such cases, one solution is to use a summary measure for each patient; an alternative solution is to perform an observation-specific analysis using a technique that properly accounts for the correlation. In this paper, we investigate the effect that ignoring correlated observations can have on the design properties of the typical phase II clinical trial. In cases in which an observation-specific analysis is desirable, we propose a simple method that adjusts a standard one- or two-stage phase II design to account for loss of information due to correlated observations. Simulations demonstrate that the method ensures that type I and type II error rate design requirements are met even in the presence of strong correlation. We develop the method in the context of phase II oncology trials, but the method applies readily to other clinical areas in which multiple responses per patient are of interest. Control Clin Trials 1999;20:242-252 Copyright (C) 1999 Elsevier Science Inc.

KW - Correlated binomial

KW - Multistage design

UR - http://www.scopus.com/inward/record.url?scp=0032919164&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0032919164&partnerID=8YFLogxK

U2 - 10.1016/S0197-2456(98)00058-0

DO - 10.1016/S0197-2456(98)00058-0

M3 - Article

C2 - 10357497

AN - SCOPUS:0032919164

VL - 20

SP - 242

EP - 252

JO - Controlled Clinical Trials

JF - Controlled Clinical Trials

SN - 0197-2456

IS - 3

ER -