Online Program Home
My Program

Abstract Details

Activity Number: 540 - SPEED: Clinical Trial Design, Longitudinal Analysis, and Other Topics in Biopharmaceutical Statistics
Type: Contributed
Date/Time: Wednesday, August 1, 2018 : 11:35 AM to 12:20 PM
Sponsor: Biopharmaceutical Section
Abstract #332849
Title: Assessing MCP-Mod Relative to Pairwise Comparisons and Trend Tests in Dose-Ranging Design and Analysis
Author(s): Anran Wang* and Fang Liu and Sammy Yuan and Man (Mandy) Jin and Meihua Wang and Akshita Chawla and Pranab Kumar Mitra and Robin Mogg
Companies: Merck & Co Inc and Merck & Co., Inc and Merck and Merck & Co., Inc. and Merck & Co. and Merck & Co Inc and Merck & Co Inc and Merck Research Laboratories
Keywords: Dose-response relationship; simulation ; MCPMod; trend test; pairwise comparison
Abstract:

Investigating dose-response relationships and identifying the appropriate dose for confirmatory studies are important and challenging goals during drug development. Inadequate design and analysis of dose response studies continue to plague many drug development programs, and are key drivers of late stage failure and post-marketing commitments related to potential dose adjustments. Analysis of dose response studies has traditionally followed two strategies: one based on multiple comparison procedure (MCP) and another based on modeling. Each strategy has shortcomings when applied separately. Multiple Comparison Procedures & Modeling (MCP-Mod), a combination of MCP principles and modeling techniques, provides the flexibility of modeling for dose estimation, while preserving the rigor associated with MCP. This presentation will examine via simulation the performance of MCP-Mod by comparing it with two common approaches used in dose response studies (pairwise comparison and linear trend tests), in terms of power to identify the dose-response signal and target dose estimates based on a minimum effective dose (MED).


Authors who are presenting talks have a * after their name.

Back to the full JSM 2018 program