Dual-Objective Bayesian Optimal Designs for a Dose-Ranging Study
- PDF / 561,589 Bytes
- 8 Pages / 504 x 719.759 pts Page_size
- 52 Downloads / 205 Views
Drug Informorion Journol. Vol. 34. pp. 421-428. 2000 Printed in che USA. All rights reserved.
DUAL-OBJECTIVE BAYESIAN OPTIMAL DESIGNS FOR A DOSE-RANGING STUDY QI ZENG Clinical Biostatistics, Merck & Co., Rahway, New Jersey
WE1 ZHU Department of Applied Mathematics and Statistics, State University of New York at Stony Brook, Stony Brook. New York
WENGKEE WONG Department of Biostatistics. University of California-Los Angeles, Los Angeles, California
The American College of Rheumatology Responder Index (ACR index) is commonly used to measure treatment eficacy in rheumatoid arthritis trials. This composite index measures the patients’ responses as improved or not improved. We use a logit model to describe the dose-response relationship and construct dual-objective optimal designs using prior information. The specific objectives of our designs are to simultaneously estimate the parameters in the dose-response curve and the minimum dose that reaches clinically meaningful significance. Key Worals: ACR Responder Index; Bayesian multiple-objectiveoptimal design; Clinically
significant dose; Efficiency plot; Logit model
INTRODUCTION ACR Responder Index VARIOUS DEFINITIONS OF therapeutic improvement have been used in rheumatoid arthritis (RA) trials. An overall assessment of therapeutic improvement usually requires multiple endpoints, such as investigatodpatients global assessment of disease activity, tenderkwollen joint counts, pain assessment, and so forth. A disadvantage of having multiple endpoints, however, is the computational complexity and interpretation involved in multiple parameter estimation and multiple hypothesis testing. The availability of a validated definition
of a binary response in RA studies is an attractive alternative to multiple endpoints. The American College of Rheumatology Preliminary Definition of Improvement in Rheumatoid Arthritis (1) is a recently published guideline to suggest such a composite score from seven major efficacy and laboratory variables. The ACR index is an attempt to maximally discriminate effective treatments from ineffective treatment protocols by summarizing several key indicators of improvement in rheumatoid arthritis. The ACR index defines a patient as a responder if the following criteria are achieved simultaneously:
A 20% improvement from baseline in a total of 68 tender joint counts, A 20% improvement from baseline in a total of 66 swollen joint counts, Reprint address: Weng Kee Wong, Department of Bio0 A 20% improvement in at least three out statistics, UCLA School of Public Health, 10833 Leof the five measurements: Investigator and Conte Ave., Los Angeles, CA 90095-1772. 0
421
Downloaded from dij.sagepub.com at UNIVERSITE DE SHERBROOKE on May 2, 2015
Qi Zeng, Wei Zhu, and Weng Kee Wong
422
patient’s global assessment of disease activity (Likert), patient’s global pain assessment (visual analog scale), Serum Creactive protein, or health assessment questionnaire disability index. This index incorporates clinically meaningful improvement in most of the relevan
Data Loading...
