This half-day course focuses on a broad class of statistical problems arising in subgroup analysis.  The first part of the course deals with exploratory subgroup analysis, i.e., subgroup search/identification methods that can be applied in late-phase clinical trials.  The discussion of exploratory subgroup analysis begins with an overview of recent approaches to subgroup identification in the context of personalized medicine. These approaches fall in the four subtypes: global modeling of the outcome function, direct modeling of treatment contrast, identifying optimal treatment regimes, and direct subgroup search. The overview will be followed by a detailed review of the SIDES method (Subgroup Identification based on Differential Effect Search) introduced in Lipkovich et al. (2011), Lipkovich and Dmitrienko (2014, 2015).  SIDES is based on recursive partitioning and can be used in prospective and retrospective subgroup analysis.  Key elements of SIDES will be discussed, including generation of multiple promising subgroups based on different splitting criteria, choice of optimal values of complexity parameters via cross-validation, evaluation of variable importance and using variable importance indices for pre-screening covariates, and addressing Type I error rate inflation using a resampling-based method.
 
The second part focuses on the issues related to confirmatory subgroup analysis, i.e., analysis of pre-specified subgroups in the context of confirmatory Phase III clinical trials.  This will include a summary of general principles of confirmatory subgroup analysis (interpretation of findings in several patient populations based on the influence and interaction conditions) introduced in Millen at el. (2012, 2014) and Dmitrienko, Millen and Lipkovich (2015).  In addition, a review and comparison of multiplicity adjustment methods used in confirmatory subgroup analysis will be provided, including non-parametric and parametric procedures (Dmitrienko et al., 2009; Dmitrienko, D’Agostino and Huque, 2013).

Multiple case studies will be used to illustrate the principles and statistical methods introduced in this course, including design and analysis of Phase III trials with target subgroups and biomarker discovery in Phase III development programs.  Software tools for implementing the subgroup analysis methods in clinical trials will be presented, including the SIDES package developed by the authors.

Instructors:
Alex Dmitrienko, PhD
Executive Director, Center for Statistics in Drug Development
Quintiles Inc., United States

Ilya Lipkovich, PhD
Senior Director, Center for Statistics in Drug Development, Innovation
Quintiles Inc., United States

For further information on this tutorial click here