Benjamin R Saville1, Jason T Connor2, Gregory D Ayers3, JoAnn Alvarez3. 1. Department of Biostatistics, School of Medicine, Vanderbilt University, Nashville, TN, USA b.saville@vanderbilt.edu. 2. Berry Consultants, Austin, TX, USA College of Medicine, University of Central Florida, Orlando, FL, USA. 3. Department of Biostatistics, School of Medicine, Vanderbilt University, Nashville, TN, USA.
Abstract
BACKGROUND: Bayesian predictive probabilities can be used for interim monitoring of clinical trials to estimate the probability of observing a statistically significant treatment effect if the trial were to continue to its predefined maximum sample size. PURPOSE: We explore settings in which Bayesian predictive probabilities are advantageous for interim monitoring compared to Bayesian posterior probabilities, p-values, conditional power, or group sequential methods. RESULTS: For interim analyses that address prediction hypotheses, such as futility monitoring and efficacy monitoring with lagged outcomes, only predictive probabilities properly account for the amount of data remaining to be observed in a clinical trial and have the flexibility to incorporate additional information via auxiliary variables. LIMITATIONS: Computational burdens limit the feasibility of predictive probabilities in many clinical trial settings. The specification of prior distributions brings additional challenges for regulatory approval. CONCLUSIONS: The use of Bayesian predictive probabilities enables the choice of logical interim stopping rules that closely align with the clinical decision-making process.
BACKGROUND: Bayesian predictive probabilities can be used for interim monitoring of clinical trials to estimate the probability of observing a statistically significant treatment effect if the trial were to continue to its predefined maximum sample size. PURPOSE: We explore settings in which Bayesian predictive probabilities are advantageous for interim monitoring compared to Bayesian posterior probabilities, p-values, conditional power, or group sequential methods. RESULTS: For interim analyses that address prediction hypotheses, such as futility monitoring and efficacy monitoring with lagged outcomes, only predictive probabilities properly account for the amount of data remaining to be observed in a clinical trial and have the flexibility to incorporate additional information via auxiliary variables. LIMITATIONS: Computational burdens limit the feasibility of predictive probabilities in many clinical trial settings. The specification of prior distributions brings additional challenges for regulatory approval. CONCLUSIONS: The use of Bayesian predictive probabilities enables the choice of logical interim stopping rules that closely align with the clinical decision-making process.
Authors: David J Wilber; Carlo Pappone; Petr Neuzil; Angelo De Paola; Frank Marchlinski; Andrea Natale; Laurent Macle; Emile G Daoud; Hugh Calkins; Burr Hall; Vivek Reddy; Giuseppe Augello; Matthew R Reynolds; Chandan Vinekar; Christine Y Liu; Scott M Berry; Donald A Berry Journal: JAMA Date: 2010-01-27 Impact factor: 56.272
Authors: Wolfgang Wick; Susan Dettmer; Anne Berberich; Tobias Kessler; Irini Karapanagiotou-Schenkel; Antje Wick; Frank Winkler; Elke Pfaff; Benedikt Brors; Jürgen Debus; Andreas Unterberg; Martin Bendszus; Christel Herold-Mende; Andreas Eisenmenger; Andreas von Deimling; David T W Jones; Stefan M Pfister; Felix Sahm; Michael Platten Journal: Neuro Oncol Date: 2019-01-01 Impact factor: 12.300
Authors: Wills C Dunham; Matthew B Weinger; Jason Slagle; Mias Pretorius; Ashish S Shah; Tarek S Absi; Matthew S Shotwell; Marc Beller; Erica Thomas; Cindy L Vnencak-Jones; Robert E Freundlich; Jonathan P Wanderer; Warren S Sandberg; Miklos D Kertai Journal: J Cardiothorac Vasc Anesth Date: 2019-09-10 Impact factor: 2.628
Authors: E Vincent S Faustino; Veronika Shabanova; Leslie J Raffini; Sarah B Kandil; Simon Li; Matthew G Pinto; Jill M Cholette; Sheila J Hanson; Marianne E Nellis; Cicero T Silva; Ranjit Chima; Anjali Sharathkumar; Kimberly A Thomas; Tara McPartland; Joana A Tala; Philip C Spinella Journal: Crit Care Med Date: 2021-03-01 Impact factor: 9.296
Authors: Jason T Connor; Kristine R Broglio; Valerie Durkalski; William J Meurer; Karen C Johnston Journal: Trials Date: 2015-03-04 Impact factor: 2.279
Authors: Steffen Ventz; Leah Comment; Bill Louv; Rifaquat Rahman; Patrick Y Wen; Brian M Alexander; Lorenzo Trippa Journal: Neuro Oncol Date: 2022-02-01 Impact factor: 13.029