Better decision making in drug development through adoption of formal prior elicitation.

With the continued increase in the use of Bayesian methods in drug development, there is a need for statisticians to have tools to develop robust and defensible informative prior distributions. Whilst relevant empirical data should, where possible, provide the basis for such priors, it is often the case that limitations in data and/or our understanding may preclude direct construction of a data-based prior. Formal expert elicitation methods are a key technique that can be used to determine priors in these situations.

Epithelial desquamation observed in a phase I study of an oral cathepsin C inhibitor (GSK2793660).

Aims: Cathepsin C (CTSC) is necessary for the activation of several serine proteases including neutrophil elastase (NE), cathepsin G and proteinase 3. GSK2793660 is an oral, irreversible inhibitor of CTSC that is hypothesized to provide an alternative route to achieve NE inhibition and was tested in a Phase I study.

Methods: Single escalating oral doses of GSK2793660 from 0.

The perils with the misuse of predictive power.

In early drug development, especially when studying new mechanisms of action or in new disease areas, little is known about the targeted or anticipated treatment effect or variability estimates. Adaptive designs that allow for early stopping but also use interim data to adapt the sample size have been proposed as a practical way of dealing with these uncertainties. Predictive power and conditional power are two commonly mentioned techniques that allow predictions of what will happen at the end of the trial based on the interim data.

Practical usage of O'Brien's OLS and GLS statistics in clinical trials.

Multivariate techniques of O'Brien's OLS and GLS statistics are discussed in the context of their application in clinical trials. We introduce the concept of an operational effect size and illustrate its use to evaluate power. An extension describing how to handle covariates and missing data is developed in the context of Mixed models.