Link to the University of Pittsburgh Homepage
Link to the University Library System Homepage Link to the Contact Us Form

Comparison of novel phase I clinical trial designs based on toxicity probability intervals

Yao, Yi (2017) Comparison of novel phase I clinical trial designs based on toxicity probability intervals. Master's Thesis, University of Pittsburgh. (Unpublished)

Submitted Version

Download (2MB) | Preview


Phase I clinical trial based on toxicity probability intervals is a new class of dose-finding designs characterized by integrating the concept of intervals, instead of point estimates, in detection of the maximum tolerated dose. The purpose of this article is to explore and compare the performance of three novel designs including the two-parameter logistic regression model with categorized posterior probability design (LRcat), the modified toxicity probability interval design (mTPI) and the Bayesian optimal interval design (BOIN). A thorough numeric study with eight potential scenarios was conducted to examine critical operating characteristics. Robustness of the novel designs to the change in the target interval width and mis-specified priors was investigated in a sensitivity analysis following the simulation study. In addition, we also retrospectively analyzed a recent cancer phase I clinical trial to explore the performance of these designs in real-world application
The results of our analysis showed that interval-based designs perform comparably to a traditional CRM design using posterior mean to define MTD in most scenarios. LRcat is more flexible than CRM and demonstrates robustness to the varying target toxicity interval. BOIN is safer than other designs and allocates less patients to overly-toxic levels. mTPI is more likely to allocate patients to suboptimal doses when the true MTD resides at the lowest/highest doses and performs poorly when the target interval is asymmetric.
Phase I cancer clinical trials are the indispensable step for the development of anticancer therapies. With the widespread application of phase I clinical trials, researchers and clinical investigators need up-to-date information about newly-developed phase I clinical trial methods. By providing the comparison result for a group of innovative phase I trial designs, our study facilitates choice of dose-finding method and leads to more efficient and ethical drug development to conquer cancer epidemic.


Social Networking:
Share |


Item Type: University of Pittsburgh ETD
Status: Unpublished
CreatorsEmailPitt UsernameORCID
Yao, Yiyiy43@pitt.eduYIY43
ETD Committee:
TitleMemberEmail AddressPitt UsernameORCID
Committee ChairWang, Honghow8@pitt.eduHOW8
Lin, Yanyal14@pitt.eduYAL14
Abebe, Kalebkza3@pitt.eduKZA3
Park, Seo Youngsyp8@pitt.eduSYP8
Date: 29 June 2017
Date Type: Publication
Defense Date: 12 April 2017
Approval Date: 29 June 2017
Submission Date: 28 April 2017
Access Restriction: No restriction; Release the ETD for access worldwide immediately.
Number of Pages: 52
Institution: University of Pittsburgh
Schools and Programs: School of Public Health > Biostatistics
Degree: MS - Master of Science
Thesis Type: Master's Thesis
Refereed: Yes
Uncontrolled Keywords: Phase I clinical trial; dose-finding design; interval design
Date Deposited: 29 Jun 2017 23:36
Last Modified: 29 Jun 2017 23:36


Monthly Views for the past 3 years

Plum Analytics

Actions (login required)

View Item View Item