Keywords: MCP-Mod, Dose finding, Minimum effect dose, Simulation
Investigating dose-response relationships and identifying the appropriate dose for confirmatory studies are important and challenging goals during the 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. 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. We examine via extensive simulation the performance of MCP-Mod for both continuous and binary endpoints by comparing it with several other common approaches used in dose response studies. The evaluation includes the power to identify the dose-response signal and target dose estimates based on a minimum effective dose (MED), and the response estimates at the dose level selected for confirmatory studies.