Complex endpoints are an increasingly common challenge for clinical trial design. We evaluate three complex endpoints, the appropriate models for such endpoints and the sample size determination approaches available for these models. In this webinar we look at sample size for the proportional odds model for ordinal data, the Win Ratio statistic for composite endpoints and the MaxCombo test for complex survival analysis.
More About The Webinar
Complex endpoints have become a common issue in the design and analysis of clinical trials. Choosing the appropriate analysis and designing the trial both require careful consideration of the nature of the chosen endpoint. Three examples of complex endpoints of increasing interest in clinical trials are ordinal, composite and complex survival endpoints.
Ordinal endpoints construct the outcome assessment in terms of a ranked order of patient status. Ordinal scales are common for clinical assessment tools for outcomes such as pain and disease progression. Two commonly used approaches for analyzing ordinal data are the proportional odds and Wilcoxon (Mann-Whitney) U tests with respective strengths and weaknesses.
Composite endpoints are where multiple outcomes are used to evaluate the treatment effect. This accounts for the complexity of a disease and prevents assessment relying on measuring one outcome only. However, methods for analyzing composite endpoints in clinical trials were sparse till recently but recent papers have highlighted statistics such as the win ratio for the analysis of composite endpoints.
Survival analysis is a very important and common statistical area of interest in areas such as oncology trials. However many innovative therapies such as immunotherapies have complex survival outcomes which violate the proportional hazards assumption needed for common methods such as the log-rank test. Multiple proposals have emerged for the analysis of these complex survival datasets with the MaxCombo test being a leading tool for analyzing complex outcomes such as delayed effects.
In this webinar, we explore the complex endpoints of ordinal data, composite endpoints and non-proportional survival data and provide background on the methods available for these complex cases and the sample size determination methods available for those methods.