Applying the Estimand Framework: Case Studies in Neuroscience
摘要
Neuroscience is a heterogeneous therapeutic area with many different facets and challenges. Different types of endpoints are utilized in neuroscience, and many different problems arise, especially with intercurrent events and missing data. The estimand framework is, therefore, helpful in many neuroscience indications, as it allows a structured approach to address such challenges. This chapter focuses on several typical neuroscience indications, including multiple sclerosis (MS), pain and migraine, Parkinson’s disease (PD), and Alzheimer’s disease (AD). It discusses and explains typical challenges arising in these heterogeneous diseases and, for each one, lays out a path forward based on the estimand framework. As expected, the strategies for addressing intercurrent events are the central focus and will be specifically discussed. A treatment policy strategy is frequently proposed and justifiable but not always. There are also examples where a composite variable strategy, a hypothetical strategy, or even a principal stratum strategy seems reasonable.