The multiple comparison procedure and modeling (MCP-Mod) approach (Pinheiro et al., Analysis of dose-response studies—modeling approaches. In: Ting N (ed) Dose finding in drug development. Springer, New York, p 146–171, 2006) to confirm the proof of concept and identify the minimum effective dose (MED) in Phase II clinical trials was first introduced in the early 2000s. The MCP-Mod method has significantly transformed the way dose-finding studies are conducted. However, there are multiple issues with the dose(s) recommended by this method. First, the selected doses are often not practical for manufacturing, which requires researchers to find the feasible dose along the significant dose-response curve or return to the traditional pairwise test to find a manufacturable MED. Second, the complexity of the MCP-Mod method makes it difficult to explain to a team. Third, the algorithm encounters convergence issues. Finally, the performance of MCP-Mod is subject to whether the true dose-response relationship is close to one of the selected candidate models. In contrast, the ordinal linear contrast test (OLCT) (Zhang, A simple and efficient statistical approach for designing an early phase II clinical trial. In: Chen D-G, Jin Z, Li G, Liu A, Zhao Y (eds) New advances in statistics and data science. Springer, New York, p 176-196, 2017) has demonstrated no such issues and is considered comparable to the MCP-Mod in power and probability to select MED correctly to the MCP-Mod method. In this chapter, the authors conducted a thorough comparison of the two methods through simulations and demonstrated the superiority of the OLCT.

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Comparing MCP-Mod and Ordinal Linear Contrast Test in Dose-Finding Clinical Trials: A Thorough Examination

  • Yaohua Zhang,
  • Ning Li,
  • Naitee Ting

摘要

The multiple comparison procedure and modeling (MCP-Mod) approach (Pinheiro et al., Analysis of dose-response studies—modeling approaches. In: Ting N (ed) Dose finding in drug development. Springer, New York, p 146–171, 2006) to confirm the proof of concept and identify the minimum effective dose (MED) in Phase II clinical trials was first introduced in the early 2000s. The MCP-Mod method has significantly transformed the way dose-finding studies are conducted. However, there are multiple issues with the dose(s) recommended by this method. First, the selected doses are often not practical for manufacturing, which requires researchers to find the feasible dose along the significant dose-response curve or return to the traditional pairwise test to find a manufacturable MED. Second, the complexity of the MCP-Mod method makes it difficult to explain to a team. Third, the algorithm encounters convergence issues. Finally, the performance of MCP-Mod is subject to whether the true dose-response relationship is close to one of the selected candidate models. In contrast, the ordinal linear contrast test (OLCT) (Zhang, A simple and efficient statistical approach for designing an early phase II clinical trial. In: Chen D-G, Jin Z, Li G, Liu A, Zhao Y (eds) New advances in statistics and data science. Springer, New York, p 176-196, 2017) has demonstrated no such issues and is considered comparable to the MCP-Mod in power and probability to select MED correctly to the MCP-Mod method. In this chapter, the authors conducted a thorough comparison of the two methods through simulations and demonstrated the superiority of the OLCT.