Time and resources available for the development of new treatments are increasingly difficult to find, so there is a real need to demonstrate that studies offer value for money. For this reason, trials are increasingly being designed to give them the flexibility to adapt as the data emerges, and to be used to plan the extent and scope of the next phase of research. Adaptations include: varying the doses to find the optimal one for future trials; dropping ineffective treatments earlier; finding the best treatment amongst several; and deciding on the sample size of a trial using only limited information. We have a series of research areas that are tackling these issues using adaptive clinical trials. Recently we have developed novel trial designs that minimise the expected sample size and therefore the average costs of trials by using interim analyses whilst still maintaining good statistical practice of controlling the error probabilities. We are now extending our methods to tackle areas such as dual-agent dose escalation studies, dose finding trials and biomarker driven trial design. We are placing emphasis on the translational side of our work by disseminating our findings to clinical and statistical audiences, providing easy to use software and training the community to implement our methodology.
Adaptive Clinical Trials research team
- Mander, A. P., Wason, J. M. S., Sweeting, M. J. & Thompson, S. G. (2012)
Admissible two-stage designs for phase II cancer clinical trials that incorporate the expected sample size under the alternative hypothesis.
Pharmaceutical Statistics 11: 91-96
- Wason, J. M., Mander, A. P. & Thompson, S. G. (2012)
Optimal multi-stage designs for randomised clinical trials with continuous outcomes.
Statistics in Medicine 31: 301-312
- Sweeting, M.J. & Mander, A.P. (2012)
Escalation strategies for combination therapy phase I trials.
Pharmaceutical Statistics 11: 258-266
- Wason, J. M. S., Mander, A. P. & Eisen, T. G. (2011)
Using continuous tumour shrinkage endpoints in two-stage phase II cancer trial designs.
European Journal of Cancer 47: 983-989
- Mander, A. P. & Thompson, S. G. (2010)
Two-stage designs optimal under the alternative hypothesis for phase II cancer trials.
Contemporary Clinical Trials 31: 572-578