Recent Advances for Handling Composite Endpoints in Clinical Trials
Atalanta Ghosh
Janssen Research and Development
Surya Mohanty
Janssen Research and Development
Composite endpoint is commonly used in clinical trials to demonstrate overall treatment effects . A typical way to analyze composite endpoint is to count time to first event: the first occurrence of any event in the composite is considered, and any subsequent information is ignored. Implicitly, this approach only considers the timing of component events rather than clinical importance. Pocock et al. (2012, European Heart Journal) proposed a win ratio approach to analyzing composite endpoint accounting for the clinical priorities, which has drawn great interests among statistical methodologists and clinical researchers. In this session, we will invite four speakers who have been working actively on this topic. Dr. Luo (2015, Biometrics) first introduced the win ratio approach to statistical community and worked out the close form solutions to perform formal statistical inference; Dr. Behu (accepted by Biostatistics) extended the idea to a general non-parametric inference which allows different types of outcomes (e.g., binary, continuous, and time to event) to be analyzed; Dr. Lachin (accepted by Clinical Trials) proposed a new way to access the beneficial effects of a treatment on multiple event-time outcomes using Wei-Lachin statistics. Dr. Zaslavsky (2013, Pharmaceutical Statistics) brought Bayesian's solution to the issue. In summary, this session will bring speakers from various sectors together to present their latest research results and share their perspectives. The presentation topics will include most important recent advances in this field. Given the importance and broad use of composite endpoint in clinical trials,we believe the topic will be of interest to a large audience and the session will be well received. All four speakers have agreed to present if the proposal is approved. All four speakers have agreed to present if the proposal gets approved. Format of the session: four presenters and one discussant. 1) John M. Lachin, Sc.D., Research Professor at the George Washington University Biostatistics Center. Title of the talk "Application of the Wei-Lachin multivariate one-sided test to multiple event-time outcomes". 2) Xiaodong Luo, Ph.D., Assistant Professor at Icahn School of Medicine at Mount Sinai. Title of the talk "Weighted win ratio statistics". 3) Boris Zaslavsky, Ph.D., Dr.Sc., FDA, Title of the talk "Bayesian design and analysis of composite endpoints in clinical trials with multiple dependent binary outcomes". 4) Ionut Bebu, Ph.D., Assistant Professor at the George Washington University Biostatistics Center. Title of the talk "Large sample inference for a win ratio analysis of a composite outcome based on prioritized components". Discussant: Surya Mohanty, Ph.D., Head of Biostatistics for Oncology, Janssen Research and Development.