Impact of Ascertainment on the Risk Estimation of Second Cancers in Family Studies

• Activity Number: 292
• Type: Contributed
• Date/Time: Tuesday, August 5, 2014 : 8:30 AM to 10:20 AM
• Sponsor: Biometrics Section
• Author(s): Yun-hee Choi*+ and Laurent Briollais and Karen Kopciuk
• Companies: University of Western Ontario and Lunenfeld-Tanenbaum Research Institute and Alberta Health Services/University of Calgary
• Keywords: Progressive Multistate Model ; Competing Risks ; Family Data ; Ascertainment-corrected Likelihood ; Successive Event Times ; Lynch Syndrome

Abstract

In cancer family studies, families are often ascertained through affected individuals (i.e. the probands) where those affected individuals can have multiple cancers. Estimating the risk of first and successive cancers in these families could depend on the occurrence of multiple cancers among the probands as well as their mutation status. The goal of our study is to investigate the impact of ascertainment on risk estimation of a first and a second cancer in family studies. We propose here a general framework based on a progressive multistate model where the progression to a specific cancer stage can have multiple competing risk events to deal with complex ascertainment and provide unbiased risk estimates for first and successive cancers in family studies. We illustrate the method by an analysis of lynch syndrome families identified through the colon cancer family registries, a NIH-funded initiative. These families harbour a mutation in mismatched repair genes and members of those families are at high risk of developing multiple cancers over their lifetime including colorectal, endometrial, ovarian, stomach.