A systematic review of drug development or drug licensing decisions that could inform a quantitative answer to the question is not available. The answer is based on the experience of the Medicines and Healthcare products Regulatory Agency in submissions received for scientific advice, clinical trial and marketing authorisation applications.
Group-sequential trials are the most common and longest established type of adaptive design. They are in widespread use and have been for many years. Trials with this type of design are proposed and conducted by the pharmaceutical industry and by academic groups, and are endorsed and accepted by regulatory agencies worldwide. A group-sequential trial that meets stopping criteria at an interim analysis (for 'efficacy' representing a success, or for 'futility' representing a failure) will, in many circumstances, mean that resources which would have been spent on the continuation of the trial to be used elsewhere, increasing productivity. For example, all clinical trial sponsors designing Phase III confirmatory clinical trials in oncology would consider a group-sequential design with the possibility of early stopping for efficacy. It should be noted though that running a group-sequential design does add additional costs which would not be present if interim analyses were not conducted so it is not always beneficial to use such a design.
Trials with sample size modifications are also uncontroversial. Before a trial begins there may not be enough information to make a good estimate of the required sample size and results from an interim analysis can be used to better inform the decision. This can improve productivity by avoiding resources being wasted in unnecessarily large trials, or on trials which turn out to be too small to answer the study objectives and have to then be repeated.
More recently other types of adaptive design have started to be seen, possibly encouraged by the European Medicines Agency reflection paper on the subject.
Many recent scientific advice applications have involved trials where treatment arms are discontinued after an interim analysis. For example a trial may begin with five different strengths of an experimental treatment being compared to placebo, and, based on the interim analysis, recruitment is stopped into all of the experimental strengths except for one. While many companies have sought advice on this approach the number of such trials that have been completed and submitted as part of a marketing authorisation application is limited so it is not possible to know if they would lead to productivity increases at this time. The approach could avoid the need for separate trials for dose selection and then to provide conclusive evidence of efficacy/safety for that dose, so there is potential for productivity gains. However, the need to make the dose decision quickly and without allowing wide dissemination of the interim results of the still blinded ongoing trial could mean that poorer decisions on the choice of dose are made, leading to failed trials and decreased productivity. In addition, the statistical penalties that have to be applied after dose selection to avoid increasing the number of false positive trials may also cause an increase in the number of patients needed. Therefore it is too early to say whether these newer types of adaptive design will increase productivity.