The Phenotypic Squeeze: What Can We Realistically Expect from Genetically Personalized Medicine and What Can Statisticians Do to Help Realize It?

• Activity Number: 470
• Type: Topic Contributed
• Date/Time: Wednesday, August 7, 2013 : 8:30 AM to 10:20 AM
• Sponsor: Section on Statistics in Epidemiology
• Author(s): Stephen Senn*+
• Companies: CRP-Sante
• Keywords: Personalized medicine ; Theranostics ; Genetic marker ; Variance Components ; Decision analysis ; Clinical trials

Abstract

Adding value to theranostic strategies based on genotypic information requires identification of genetic markers that outdo obvious phenotypically based predictive strategies. Body weight has some predictive value as regard concentration of active ingredient in the blood and concentration has some relevance to toxicity and efficacy but you don't need a patient's genotype to measure body weight if you have set of bathroom scales. Yet when we look at the number of pharmaceuticals that are dosed by body weight (or indeed by anything else) the list is disappointingly small. In order to understand what genetically personalized medicine can and cannot do for health care, two key branches of statistics need to be used: decision analysis and the analysis of components of variation. In this talk I will cover the contribution both can make and will argue that we need to move from a lab-push view to a clinic-pull view of personalized medicine. Instead of following promising leads that come out of in vitro studies we need to look at diseases carefully and consider both where it is plausible that theranostics can make a difference and what it would take to make it happen.