gBOIN: a unified model-assisted phase I trial design accounting for toxicity grades, and binary or continuous end points

Rongji Mu, Ying Yuan, Jin Xu, Sumithra J Mandrekar, Jun Yin

Research output: Contribution to journalArticle

3 Citations (Scopus)

Abstract

The landscape of oncology drug development has recently changed with the emergence of molecularly targeted agents and immunotherapies. These new therapeutic agents appear more likely to induce multiple low or moderate grade toxicities rather than dose limiting toxicities. Various model-based dose finding designs and toxicity severity scoring systems have been proposed to account for toxicity grades, but they are difficult to implement because of the use of complicated dose–toxicity models and the requirement to refit the model at each decision of dose escalation and de-escalation. We propose a generalized Bayesian optimal interval design, gBOIN, that accommodates various existing toxicity grade scoring systems under a unified framework. As a model-assisted design, gBOIN derives its optimal decision rule on the basis of the exponential family of distributions but is carried out in a simple way as the algorithm-based design: its decision of dose escalation and de-escalation involves only a simple comparison of the sample mean of the end point with two prespecified dose escalation and de-escalation boundaries. No model fitting is needed. We show that gBOIN has the desirable finite property of coherence and a large sample property of consistency. Numerical studies show that gBOIN yields good performance that is comparable with or superior to that of some existing, more complicated model-based designs. A Web application for implementing gBOIN is freely available from http://www.trialdesign.org.

Original languageEnglish (US)
Pages (from-to)289-308
Number of pages20
JournalJournal of the Royal Statistical Society. Series C: Applied Statistics
Volume68
Issue number2
DOIs
StatePublished - Feb 1 2019

Fingerprint

Phase I Trial
End point
Toxicity
Binary
Dose
Scoring
Immunotherapy
Dose Finding
Model-based Design
Oncology
Model
Model Fitting
Sample mean
Exponential Family
Decision Rules
Web Application
Numerical Study
Drugs
Limiting
Likely

Keywords

  • Bayesian adaptive design
  • Dose finding
  • Exponential family distribution
  • Maximum tolerated dose

ASJC Scopus subject areas

  • Statistics and Probability
  • Statistics, Probability and Uncertainty

Cite this

@article{62ef3584b93d415a9939f3a59e4f3312,
title = "gBOIN: a unified model-assisted phase I trial design accounting for toxicity grades, and binary or continuous end points",
abstract = "The landscape of oncology drug development has recently changed with the emergence of molecularly targeted agents and immunotherapies. These new therapeutic agents appear more likely to induce multiple low or moderate grade toxicities rather than dose limiting toxicities. Various model-based dose finding designs and toxicity severity scoring systems have been proposed to account for toxicity grades, but they are difficult to implement because of the use of complicated dose–toxicity models and the requirement to refit the model at each decision of dose escalation and de-escalation. We propose a generalized Bayesian optimal interval design, gBOIN, that accommodates various existing toxicity grade scoring systems under a unified framework. As a model-assisted design, gBOIN derives its optimal decision rule on the basis of the exponential family of distributions but is carried out in a simple way as the algorithm-based design: its decision of dose escalation and de-escalation involves only a simple comparison of the sample mean of the end point with two prespecified dose escalation and de-escalation boundaries. No model fitting is needed. We show that gBOIN has the desirable finite property of coherence and a large sample property of consistency. Numerical studies show that gBOIN yields good performance that is comparable with or superior to that of some existing, more complicated model-based designs. A Web application for implementing gBOIN is freely available from http://www.trialdesign.org.",
keywords = "Bayesian adaptive design, Dose finding, Exponential family distribution, Maximum tolerated dose",
author = "Rongji Mu and Ying Yuan and Jin Xu and Mandrekar, {Sumithra J} and Jun Yin",
year = "2019",
month = "2",
day = "1",
doi = "10.1111/rssc.12263",
language = "English (US)",
volume = "68",
pages = "289--308",
journal = "Journal of the Royal Statistical Society. Series C: Applied Statistics",
issn = "0035-9254",
publisher = "Wiley-Blackwell",
number = "2",

}

TY - JOUR

T1 - gBOIN

T2 - a unified model-assisted phase I trial design accounting for toxicity grades, and binary or continuous end points

AU - Mu, Rongji

AU - Yuan, Ying

AU - Xu, Jin

AU - Mandrekar, Sumithra J

AU - Yin, Jun

PY - 2019/2/1

Y1 - 2019/2/1

N2 - The landscape of oncology drug development has recently changed with the emergence of molecularly targeted agents and immunotherapies. These new therapeutic agents appear more likely to induce multiple low or moderate grade toxicities rather than dose limiting toxicities. Various model-based dose finding designs and toxicity severity scoring systems have been proposed to account for toxicity grades, but they are difficult to implement because of the use of complicated dose–toxicity models and the requirement to refit the model at each decision of dose escalation and de-escalation. We propose a generalized Bayesian optimal interval design, gBOIN, that accommodates various existing toxicity grade scoring systems under a unified framework. As a model-assisted design, gBOIN derives its optimal decision rule on the basis of the exponential family of distributions but is carried out in a simple way as the algorithm-based design: its decision of dose escalation and de-escalation involves only a simple comparison of the sample mean of the end point with two prespecified dose escalation and de-escalation boundaries. No model fitting is needed. We show that gBOIN has the desirable finite property of coherence and a large sample property of consistency. Numerical studies show that gBOIN yields good performance that is comparable with or superior to that of some existing, more complicated model-based designs. A Web application for implementing gBOIN is freely available from http://www.trialdesign.org.

AB - The landscape of oncology drug development has recently changed with the emergence of molecularly targeted agents and immunotherapies. These new therapeutic agents appear more likely to induce multiple low or moderate grade toxicities rather than dose limiting toxicities. Various model-based dose finding designs and toxicity severity scoring systems have been proposed to account for toxicity grades, but they are difficult to implement because of the use of complicated dose–toxicity models and the requirement to refit the model at each decision of dose escalation and de-escalation. We propose a generalized Bayesian optimal interval design, gBOIN, that accommodates various existing toxicity grade scoring systems under a unified framework. As a model-assisted design, gBOIN derives its optimal decision rule on the basis of the exponential family of distributions but is carried out in a simple way as the algorithm-based design: its decision of dose escalation and de-escalation involves only a simple comparison of the sample mean of the end point with two prespecified dose escalation and de-escalation boundaries. No model fitting is needed. We show that gBOIN has the desirable finite property of coherence and a large sample property of consistency. Numerical studies show that gBOIN yields good performance that is comparable with or superior to that of some existing, more complicated model-based designs. A Web application for implementing gBOIN is freely available from http://www.trialdesign.org.

KW - Bayesian adaptive design

KW - Dose finding

KW - Exponential family distribution

KW - Maximum tolerated dose

UR - http://www.scopus.com/inward/record.url?scp=85048328881&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85048328881&partnerID=8YFLogxK

U2 - 10.1111/rssc.12263

DO - 10.1111/rssc.12263

M3 - Article

AN - SCOPUS:85048328881

VL - 68

SP - 289

EP - 308

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 - 2

ER -