New quantitative approaches to precision screening: The example of lung-cancer screening

Speaker:

Dr Hormuzd A Katki, PhDSenior InvestigatorNational Cancer Institute, NIH, USA

Title: 

New quantitative approaches to precision screening: The example of lung-cancer screening

Hormuzd A. Katki received a B.S. in math from the University of Chicago and an M.S. in statistics from Carnegie-Mellon University. He received a Ph.D in biostatistics from Johns Hopkins University in 2006, where he received the Margaret Merrell Award for research by a biostatistics doctoral student. Dr. Katki joined NCI in 1999, became a principal investigator in 2009, and was appointed senior investigator upon receiving NIH scientific tenure in 2015.

Abstract: 

Lung-cancer screening may become the first cancer screening modality in the US where guidelines will recommend using an individualized risk model to decide who is eligible for screening. I will summarize the research that underlies this conclusion, then I will discuss our approaches to 3 key questions that remain. The first question is criteria for deciding on the risk-threshold at which to offer screening. The seemingly obvious answer to that question is a cost-effectiveness analysis, but in the US, costs are generally not allowed to be considered by guidelines bodies. The second question is how to update individual risk with screening test results. I will present a simple Markov model approach that uses screening test results to directly modify pre-screening individualized risk. The third key question is how to account for comorbidities and life-expectancy when selecting people for screening. A risk-based approach will generally choose older people with comorbidities, who may not live long even if their life is saved by screening. I present our individualized model for life-years gained from screening, and compare those chosen for screening by individualized life-years gained versus by individualized risk. In contrast to the current risk-based approach to precision screening, life-years gained is a benefit-based approach that is conceptually the better approach to precision-screening. Throughout the talk, I will discuss where my experience in cervical cancer screening inspired our approaches to these questions.