Keywords: benefit-risk, safety, FDA
This presentation will provide a high-level statistical perspective on best practices for the pre-marketing benefit-risk assessment of a new treatment. I will discuss potential enhancements in planning prior to confirmatory clinical trials, including the utility of a program-wide safety and benefit-risk evaluation plan and an adverse event classification system that incorporates clinical importance and plausibility of effects. I will describe the importance of statistical contributions to the assessment of not only key benefits but also key risks. Examples include quantification of uncertainty around estimated risks, use of appropriate methods for designs with differing durations of follow-up across patients, and use of appropriate integrated analysis methods when data are from multiple studies. I also will discuss statistical approaches that can help weigh benefits and risks to inform regulatory decisions. Examples include visual tools that facilitate qualitative assessments and more quantitative analyses that incorporate values.