Escalation with Overdose Control using a Toxicity Score for Personalized Maximum Tolerated Dose in Phase I Clinical Trials Open Access

Zhuang, Run (2016)

Permanent URL:


The goal of many cancer Phase I clinical trials is to establish the level of dose toxicity tolerance in order to estimate the maximum tolerated dose (MTD) used for future trials. Common methods are divided into rule-based and model-based methods for dose escalation. The most popular method currently used is the 3+3 rule-based design. This method is simple and convenient, but has been shown to predict the true MTD in about 35% of trials. Model-based designs establish a dose response relationship that is equal to a pre-specified probability of dose limiting toxicity (DLT). Escalation with overdose control (EWOC) is a Bayesian adaptive design for selecting dose levels in cancer Phase I clinical trials while controlling the posterior probability of exceeding the MTD. We extend EWOC to incorporate a novel toxicity score with the addition of covariates for personalized medicine (EWOC-NETS with covariates). Under our extensions, we found significant differences in the estimated MTD between EWOC, EWOC-NETS, and EWOC-NETS with covariates for 4 simulated scenarios. Our developed method is also safer with respect to toxicity and overdosing.

Table of Contents

List of Tables i

List of Figures ii

1 Introduction 1

1.1 Rule-Based Methods in Phase I Trials 2

1.2 Model-Based Methods in Phase I Trials 3

1.3 Escalation with Overdose Control (EWOC) 4

1.4 Extensions and Goals 6

2 Methods 7

2.1 EWOC 8

2.2 NETS 10

2.3 EWOC-NETS 12

2.4 EWOC-NETS with a Discrete Covariate 14

2.5 EWOC-NETS with a Continuous Covariate 16

2.6 Simulation Settings and Methods 17

3 Results 21

3.1 Discrete Covariate 21

3.2 Continuous Covariate 24

4 Discussion 28

5 Bibliography 30

6 Appendix 32

About this Master's Thesis

Rights statement
  • Permission granted by the author to include this thesis or dissertation in this repository. All rights reserved by the author. Please contact the author for information regarding the reproduction and use of this thesis or dissertation.
Subfield / Discipline
  • English
Research Field
Committee Chair / Thesis Advisor
Committee Members
Partnering Agencies
Last modified

Primary PDF

Supplemental Files