Alternative views on setting clinical trial futility criteria
*Paul Gallo, Novartis Pharmaceuticals  Lu Mao, University of North Carolina  Vivian H. Shih, Novartis 

Keywords: futility, interim analysis, conditional power, predictive probability

A feature increasingly utilized in clinical trial design is to allow a trial to stop early when it seems that it will not likely achieve its primary efficacy objectives. This is commonly referred to as stopping a trial for futility, and can be motivated by ethical and financial considerations. The possibility of stopping for futility has implications for the operating characteristics of the trial design. A number of approaches have been described in the literature to set criteria for futility stopping, including rules based upon conditional power, predictive probability, beta spending functions, and others. We consider futility stopping from the point of view of quantifying and providing an objective sensible balance between risks of making incorrect decisions (e.g., stopping trials which should stop, and continuing trials which should continue), and make proposals for how specific considerations in individual trials can lead to choice of a sensible scheme. This approach is not specific to any of the particular scales in the literature such as those mentioned above, and we describe the interrelationship between criteria as they might be expressed on different scales. As futility may be evaluated multiple times in a long-term trial and the amount of information available at scheduled interim analyses may be difficult to predict in advance, we present a specific optimality criteria and discuss which of the familiar scales tend to produce schemes that are simple to describe and implement, and have better and more robust behavior across different timepoints at which futility might be evaluated within a given trial.