Abstract:
|
Dose response modeling is an important step in drug development to confirm the existence of an efficacious dose and to investigate the extent of a dose response effect. One objective in dose response modeling is to identify the minimum effective dose (MED), which is defined as the lowest dose (among the pre-specified doses) that are deemed efficacious per a pre-specified criteria. We investigate three approaches to identify the MED in a dose ranging study, including step-down trend test, Dunnett Step-Up (SU) multiple testing procedure and the MCP-Mod approach. The trend test and the Dunnett SU procedure are frequently used in studying dose-response relationships, and we have adapted these two methods to suit the purpose of identifying the MED. The MCP-Mod approach combines multiple comparison procedures with modeling techniques, and can be directly used for identifying the MED. Via simulation, we compare the power of finding the MED and other operating characteristics using these three approaches under a variety of dose response curves and simulation settings. Furthermore, the advantages and disadvantages of the different approaches will be discussed.
|
ASA Meetings Department
732 North Washington Street, Alexandria, VA 22314
(703) 684-1221 • meetings@amstat.org
Copyright © American Statistical Association.