A decision-theoretic approach to Bayesian clinical trial design and evaluation of robustness to prior-data conflict

Bayesian clinical trials allow taking advantage of relevant external information through the elicitation of prior distributions, which influence Bayesian posterior parameter estimates and test decisions. However, incorporation of historical information can have harmful consequences on the trial's frequentist (conditional) operating characteristics in case of inconsistency between prior information and the newly collected data. A compromise between meaningful incorporation of historical information and strict control of frequentist error rates is therefore often sought. Our aim is thus to review and investigate the rationale and consequences of different approaches to relaxing strict frequentist control of error rates from a Bayesian decision-theoretic viewpoint. In particular, we define an integrated risk which incorporates losses arising from testing, estimation, and sampling. A weighted combination of the integrated risk addends arising from testing and estimation allows moving smoothly between these two targets. Furthermore, we explore different possible elicitations of the test error costs, leading to test decisions based either on posterior probabilities, or solely on Bayes factors. Sensitivity analyses are performed following the convention which makes a distinction between the prior of the data-generating process, and the analysis prior adopted to fit the data. Simulation in the case of normal and binomial outcomes and an application to a one-arm proof-of-concept trial, exemplify how such analysis can be conducted to explore sensitivity of the integrated risk, the operating characteristics, and the optimal sample size, to prior-data conflict. Robust analysis prior specifications, which gradually discount potentially conflicting prior information, are also included for comparison. Guidance with respect to cost elicitation, particularly in the context of a Phase II proof-of-concept trial, is provided.

[1]  D. Lindley.  The choice of sample size , 1997 .

[2]  Bradley P Carlin,et al.  Commensurate Priors for Incorporating Historical Information in Clinical Trials Using General and Generalized Linear Models. , 2012, Bayesian analysis.

[3]  Fulvio De Santis.  Sample Size Determination for Robust Bayesian Analysis , 2006 .

[4]  Annette Kopp-Schneider,et al.  Power gains by using external information in clinical trials are typically not possible when requiring strict type I error control , 2019, Biometrical journal. Biometrische Zeitschrift.

[5]  Christian P Robert,et al.  Revised evidence for statistical standards , 2014, Proceedings of the National Academy of Sciences.

[6]  Vahid Montazerhodjat,et al.  Is the Fda Too Conservative or Too Aggressive?: A Bayesian Decision Analysis of Clinical Trial Design , 2015, Journal of Econometrics.

[7]  Carlos Alberto de Bragança Pereira,et al.  Adaptative significance levels using optimal decision rules: Balancing by weighting the error probabilities , 2016 .

[8]  Andrew P Grieve,et al.  How to test hypotheses if you must , 2015, Pharmaceutical statistics.

[9]  Nicky Best,et al.  Practical experiences of adopting assurance as a quantitative framework to support decision making in drug development , 2018, Pharmaceutical statistics.

[10]  Silvia Calderazzo,et al.  Quantification of prior impact in terms of effective current sample size , 2019, Biometrics.

[11]  J. Berger,et al.  Testing a Point Null Hypothesis: The Irreconcilability of P Values and Evidence , 1987 .

[12]  Anthony O'Hagan,et al.  Robust meta‐analytic‐predictive priors in clinical trials with historical control information , 2014, Biometrics.

[13]  N. Lazar,et al.  The ASA Statement on p-Values: Context, Process, and Purpose , 2016 .

[14]  Fang Chen,et al.  Use of historical control data for assessing treatment effects in clinical trials , 2014, Pharmaceutical statistics.

[15]  A O'Hagan,et al.  Bayesian Assessment of Sample Size for Clinical Trials of Cost-Effectiveness , 2001, Medical decision making : an international journal of the Society for Medical Decision Making.

[16]  James O. Berger,et al.  Rejection odds and rejection ratios: A proposal for statistical practice in testing hypotheses , 2015, Journal of mathematical psychology.

[17]  Roger J Lewis,et al.  Bayesian decision-theoretic group sequential clinical trial design based on a quadratic loss function: a frequentist evaluation , 2007, Clinical trials.

[18]  Roland Fisch,et al.  Bayesian Design of Proof-of-Concept Trials , 2015, Therapeutic innovation & regulatory science.

[19]  H M James Hung,et al.  Adapting the sample size planning of a phase III trial based on phase II data , 2006, Pharmaceutical statistics.

[20]  Satrajit Roychoudhury,et al.  Beyond p-values: A phase II dual-criterion design with statistical significance and clinical relevance , 2018, Clinical trials.

[21]  Leonhard Held,et al.  Adaptive power priors with empirical Bayes for clinical trials , 2017, Pharmaceutical statistics.

[22]  Frank E. Harrell,et al.  Inference and Decision Making for 21st-Century Drug Development and Approval , 2019, The American Statistician.

[23]  J. Berger.  Robust Bayesian analysis : sensitivity to the prior , 1990 .

[24]  Joseph G Ibrahim,et al.  Bayesian clinical trial design using historical data that inform the treatment effect. , 2019, Biostatistics.

[25]  Sujit K. Sahu,et al.  A Bayesian method of sample size determination with practical applications , 2006 .

[26]  L. M. Berliner,et al.  Robust Bayes and Empirical Bayes Analysis with $_\epsilon$-Contaminated Priors , 1986 .