Analysis of multiple endpoints

Sunday afternoon, July, 8

Abstract:
This course will focus on issues arising in clinical trials with multiple endpoints, including analysis of multiple co-primary endpoints (overall analysis and analysis of individual variables) and analysis of primary endpoints in the presence of important secondary endpoints/secondary analyses (gatekeeping strategies). It provides a detailed overview of well-established methods as well as novel statistical approaches developed over the past five years. The course offers a well-balanced mix of theory and applications, including regulatory considerations and software implementation of the described statistical methods. Examples from real trials are used throughout the discussion to illustrate the statistical approaches discussed in the course.

Outline:
  1. Multiple endpoints in clinical trials (increasing popularity of multiple and composite endpoints to describe diseases with a "multidimensional" etiology)
  2. Global tests
    • Basic global tests (stepwise tests based on univariate p-values, Tukey-Ciminera-Heyse test)
    • Global tests that account for correlation structure of multiple endpoints (tests based on the assumption of multivariate normality and resampling-based tests)
  3. Inferences for individual endpoints
    • Analysis of individual endpoints (identification of treatment effect on individual components of composite endpoints)
    • Multiple tests for individual endpoints (multivariate-normal and resampling-based tests)
  4. Gatekeeping and related testing procedures in clinical trials
    • Analysis of clinical trials with multiple objectives (multiple primary and secondary endpoints, multiple dose-control comparisons, multidimensional objectives)
  5. Gatekeeping strategies in clinical trials
    • Serial and parallel gatekeeping procedures (derivation of serial and parallel gatekeeping procedures based on the closed testing principle, Bonferroni-based procedures, resampling-based procedures, stepwise procedures)
    • Gatekeeping procedures in clinical trials with multiple primary and secondary endpoints and dose-finding studies
  6. Tree gatekeeping procedures
    • Flexible multiple tests for clinical trials with hierarchical objectives (derivation of tree gatekeeping procedures based on the closed testing principle)
    • Resampling-based tree gatekeeping procedures (application of Westfall-Young methodology)
    • Tree gatekeeping procedures clinical trials with multiple endpoints
Bio sketch:
Alex Dmitrienko, Ph.D., is Principal Research Scientist, Eli Lilly and Company. He received his PhD in statistics from the University of Kentucky in 1998 and joined Eli Lilly and Company that year. He has been working on multiplicity issues arising in clinical trials since 1999 and written several papers/chapters on topics related to multiple testing. His recently published SAS Press book, "Analysis of Clinical Trials Using SAS: A Practical Guide", (written jointly with G. Molenberghs, University Hasselt; C. Chuang-Stein, Pfizer; W. Offen, Lilly) focuses on key problems arising in the context of clinical trials, including the analysis of multiple comparisons and multiple endpoints. He has also served as an editor (jointly with C. Chuang-Stein, Pfizer; Ralph D'Agostino, Boston University) of another SAS Press book, "Pharmaceutical Statistics Using SAS", that will be published in late 2006. Alex serves as an Associate Editor for Statistics in Medicine and The American Statistician.

Textbook:
Dmitrienko A, Molenberghs G, Chuang-Stein C, Offen W. Analysis of Clinical

Trials Using SAS: A Practical Guide. SAS Press: Cary, NC, 2005.