Adaptive Blinded Bayesian Sample Size Re-Determination for Clinical Trials: Extensions and Risk Minimizations

• Activity Number: 486
• Type: Contributed
• Date/Time: Wednesday, August 7, 2013 : 8:30 AM to 10:20 AM
• Sponsor: Biopharmaceutical Section
• Author(s): Andrew Hartley*+
• Companies: PPD, Inc
• Keywords: Sample Size Adjustment ; Bayesian ; Blinded ; Decision Theory

Abstract

Blinded sample size re-determination (SSR) provides a mechanism for sponsors of pharmaceuticals to increase sample sizes of ongoing clinical trials. In its 2010 draft guidance on adaptive designs, the FDA remarked that this SSR "should be considered for most trials," since the SSR displays a low potential for bias and requires little logistical support. Following this guidance and reports of increased industry interest, Hartley (2012) developed a method of blinded SSR for trials comparing two independent normal means. The method updates the population probability distributions of both the treatment effect and the within-treatment variance. He showed the method to reduce an expected loss function better than does another method. This presentation extends the previous work. An enhancement to the method is given, to weight prior evidence more heavily. The enhanced method is compared to additional, established blinded SSR methods with respect to a wider assortment of expected loss functions, including decision theoretic ones which account for the trial's enrollment speed and treatment period duration, and the investigational product's remaining patent life.