A practical guide to pre-trial simulations for Bayesian adaptive trials using SAS and BUGS.

Christian Holm Hansen,Allan Walker,Richard A Parker,Christopher J Weir,Pamela Warner,Lucy Whitaker,Hilary O.D Critchley

doi:10.1002/pst.1897

Christian Holm Hansen, Allan Walker + Show 5 more

Open Access

https://doi.org/10.1002/pst.1897

Copy DOI

Abstract

It is often unclear what specific adaptive trial design features lead to an efficient design which is also feasible to implement. Before deciding on a particular design, it is generally advisable to carry out a simulation study to characterise the properties of candidate designs under a range of plausible assumptions. The implementation of such pre‐trial simulation studies presents many challenges and requires considerable statistical programming effort and time. Despite the scale and complexity, there is little existing literature to guide the implementation of such projects using commonly available software. This Teacher's Corner article provides a practical step‐by‐step guide to implementing such simulation studies including how to specify and fit a Bayesian model in WinBUGS or OpenBUGS using SAS, and how results from the Bayesian analysis may be pulled back into SAS and used for adaptation of allocation probabilities before simulating subsequent stages of the trial. The interface between the two software platforms is described in detail along with useful tips and tricks. A key strength of our approach is that the entire exercise can be defined and controlled from within a single SAS program.

Highlights

There has been growing realisation among statisticians that adaptive trial designs can improve the efficiency of drug development and offer advantages over conventional trial designs
There are numerous ways of designing an adaptive trial, so how do you decide how many adaptations to build in, when to adapt, what adaptation rules to implement, and so on? The task of identifying an optimum adaptive trial design is not straightforward and usually pre‐trial simulations are required to evaluate the properties of candidate designs
The freely available Bayesian inference Using Gibbs Sampling (BUGS) software was used to fit the Bayesian model with the simulated data being generated using SAS. With this motivating example in mind, we provide the framework for the dialogue between SAS and the BUGS software and illustrate how to fully automate and manage the simulation exercise using a single SAS control program

Summary

Introduction

There has been growing realisation among statisticians that adaptive trial designs can improve the efficiency of drug development and offer advantages over conventional trial designs. Fitting a Bayesian model using BUGS requires a data file, a model specification, and a set of initial values to start off the Markov Chain Monte Carlo (MCMC) iterative algorithm.[9] In addition, because the simulations are coordinated within SAS, the interactive BUGS user interface is not being used.

Results

Conclusion