All Times EDT
Keywords: Platform trials, Multiplicity, Error Rates
Recently – to some extent fueled by the global COVID-19 outbreak - master protocol trials and especially platform trials have received increased attention and are nearly unanimously considered to hold great potential for future drug development. Adaptive platform trials provide a framework to simultaneously study multiple treatments in a disease. They are multi-armed trials where interventions can enter and leave the platform based on interim analyses as well as external events, for example, if new treatments become available. Such integrated platform trials can have both exploratory and confirmatory elements and especially for the latter the probability of false positive decisions should be controlled. In this talk we will discuss the main structural sources of multiplicity in platform trials. We will discuss current controversies and the proposed statistical methodology to address these issues. Due to the complexity of platform trial designs with potentially various sub-studies allowing for multiple treatment arms and/or doses, subgroups and adaptations, it is not clear when and how to adjust for multiplicity. Which error rates are the more relevant ones considering the main objectives of the platform trial? We will discuss the pro and cons of various error rates such as the family-wise error rate (FWER), the k-FWER, the false-discovery rate (FDR) or the per-comparison error rate in the context of platform trials and how they relate to each other. We will discuss a framework on how to identify reasonables testing strategies. For example, under which circumstances sub-studies and the related decisions can be considered as independent, allowing for separate pre-defined sets of families of hypotheses. We will compare different strategies with the means of clinical trial simulations.