Biostatistics and Bioinformatics in Clinical Trials
Summary of Key Points
• The process of conducting cancer research must change in the face of prohibitive costs and limited patient resources.
• Biostatistics has a tremendous impact on the level of science in cancer research, especially in the design and conducting of clinical trials.
• The Bayesian statistical approach to clinical trial design and conduct can be used to develop more efficient and effective cancer studies.
• Modern technology and advanced analytic methods are directing the focus of medical research to subsets of disease types and to future trials across different types of cancer.
• A consequence of the rapidly changing technology for generating “omics” data is that biological assays are often not stable long enough to discover and validate a model in a clinical trial.
• Bioinformaticians must use technology-specific data normalization procedures and rigorous statistical methods to account for sample collecting, batch effects, multiple testing, confounding covariates, and any other potential biases.
• Best practices in developing prediction models include public access to the information, rigorous validation of the model, and model lockdown prior to its use in patient care management.
