Читать книгу Clinical Pharmacology and Therapeutics - Группа авторов - Страница 131

The above description of a clinical trial is based on a very traditional model that is generally straightforward and well‐understood by the research community but is often inflexible, time‐consuming and expensive. With this in mind, new adaptative approaches to clinical trial design have been developed that aim to be more flexible, efficient and potentially more informative. These aim for continual learning and adaptation as the data accumulate, which can result in changes to the allocation ratio, sample size, eligibility criteria and treatment arms. This has the potential to reduce the resources and time required for trial completion, reduce the number of patients exposed to inferior treatments and improve the likelihood of trial success. For example, if an interim analysis suggests a particular subgroup has a more favourable response, then the trial can be modified to solely or predominantly enrol patients from this subgroup. However, decision rules must be prespecified before the trial is commenced and are generally based around an iterative simulation process of ‘best’ and ‘worse’ case scenarios. While such adaptive trial designs can be more effective in terms of resources and time, they can be less readily accepted by regulatory agencies. They can also be limited by the potential to prematurely stop a trial for futility when the new drug is performing less well than the comparator initially but having an effect that might manifest itself later. The extent to which adaptive designs will be utilised in clinical trials remains to be fully established.