Journal of biopharmaceutical statistics
-
In randomized controlled trials, patients are recruited and randomly allocated to treatments. Patients are never randomly sampled from large population of patients on treatments under study. ⋯ Our theoretical results and Monte Carlo simulation study suggest that the permutation test following Pocock-Simon's covariate-adaptive randomization can be a useful alternative to traditional population-based tests in a confirmatory randomized controlled trial with important prognostic factors. The proposed procedure is illustrated with modified data from the randomized placebo-controlled trial of pirfenidone.
-
This manuscript discusses optimal cost-effective designs for Phase II proof of concept (PoC) trials. Unlike a confirmatory registration trial, a PoC trial is exploratory in nature, and sponsors of such trials have the liberty to choose the type I error rate and the power. The decision is largely driven by the perceived probability of having a truly active treatment per patient exposure (a surrogate measure to development cost), which is naturally captured in an efficiency score to be defined in this manuscript. ⋯ The idea is applied to derive optimal trial-level, program-level, and franchise-level design strategies. The study is not meant to provide any general conclusion because the settings used are largely simplified for illustrative purposes. However, through the examples provided herein, a reader should be able to gain useful insight into these design problems and apply them to the design of their own PoC trials.
-
The concordance correlation coefficient (CCC) is an index that is commonly used to assess the degree of agreement between observers on measuring a continuous characteristic. Here, a CCC for longitudinal repeated measurements is developed through the appropriate specification of the intraclass correlation coefficient from a variance components linear mixed model. A case example and the results of a simulation study are provided.
-
For experimental anticancer agents that may have both cytostatic and cytotoxic effects, assessment of response rates alone may not capture the full impact of the treatment. Oncologists are therefore interested in assessing both response and stable disease rates in early phase clinical trials of such therapies. We describe the design of a single-arm, Phase II clinical trial for the simultaneous evaluation of objective response and stable disease (lack of early tumor progression) rates using standard RECIST criteria. ⋯ A design is chosen that satisfies the desired type I error constraint and has sufficient statistical power at several selected points within the alternative hypothesis space using a restricted search algorithm. An early stopping rule for lack of efficacy is incorporated. The method is illustrated by the design of a Phase II clinical trial in head and neck cancer.