Keywords: Clinical efficacy
Room: Madison B
Clinical efficacy results from well-controlled randomized studies using pre-specified clinical endpoints are necessary but may not provide sufficient evidence to aid clinical decision-making at the level of an individual patient. A statistically significant clinical efficacy result will be boosted if a simultaneous claim can be made that a high proportion of patients on treatment are expected to benefit at pre-specified minimum clinically important criteria. Clinical decision-making as to which individual therapy should be chosen for a unique patient at a unique time (when more than one therapy is potentially available) will be aided greatly if probability of benefit to an individual patient from a prospective therapy can be easily and robustly quantified and compared among different available treatment options. Such objective quantification of treatment benefit can be used by not only the prescribers in their clinical decision-making, but also by other stakeholders including the patients themselves and payers. In this session, we will invite experts from the regulatory agencies, industry, and academia to share their viewpoints and experiences on how statistical evidence generation and communication should be optimized to enhance personalized medicine practice in the clinic. Discussions may range from optimal study designs incorporating patient stratification with biomarkers to pre-specification of statistical analysis plans that will generate clinical evidences of both treatment efficacy at the group level and effectiveness at the individual patient level.