Innovation and excellence in science and health
The Campus houses multiple groups with expertise in all aspects of statistics relevant to the design, conduct, analysis and interpretation of clinical research.
|MRC Biostatistics Unit||The MRC Biostatics Unit is concerned with the development and application of statistical methods in medicine and biology. This includes research into the epidemiology and aetiology of disease, the development of models of the natural history of disease, the design and analysis of trials of preventive measures or of therapeutic or prophylactic agents and the evaluation of medical technology, and the development of mathematical-statistical techniques. Much of the work of the unit is performed in collaboration with other council units and other scientific workers.
|MRC Hubs for Trials Methodology Research||Focusing on large-scale pragmatic trials, the Unit’s clinical trial methodology programme has made recent advances in methods for handling missing baseline and outcome data, non-adherence to randomized treatments, design and analysis of cluster randomized trials, and within-trial cost-effectiveness analysis.
|National Cancer Registration Service (NCRS) Eastern Office ||primary role is to collect, process, store and analyse all malignant tumours and some pre-cancerous and non-malignant (benign Brain and CNS tumours only) cancers newly diagnosed in the East of England
|Novel statistical approaches
|MRC Biostatistics Unit in close collaboration with the CCTU and the BRC
- Adaptive designs e.g. dual-agent designs for phase I cancer trials, BSU investigating methodology for design and analysis of dual-agent dose escalation trials
- Analysis of complex phenotypes e.g. defining progression in phase II and III cancer trials –BSU investigating the issue of defining progression in phase II and III oncology trials. Different potential definitions of progression will be compared to find whether the current standard is sufficiently robust and informative for long-term survival
- Biomarker driven inference: biomarkers for prediction of breast cancer survival –methods for exploring biomarkers for association with survival, and for assessing their predictive benefit over existing risk factors with the aim of incorporating them into prognostic models for clinical use
- Biomarker trials in oncology: using prior information about biomarkers in lung cancer to construct suitable priors for Bayesian adaptive design trials; used to demonstrate how these designs perform (using real trial data) and will be important for evaluating the design of future trials