Andrada, Maria de

Abstract In order to select samples that are optimally informative for sequence analysis, studies for the Genetic Epidemiology Consortium, we implemented three strategies for analysis, taking into account the known family histories of lung cancer. Susceptibility for lung cancer arises from a complex interplay between largely unknown genetic factors and smoking behavior. In order to optimize our ability to identify genetic factors for lung cancer we have recruited individuals with extensive family histories of lung cancer and also collected information about their smoking behaviors. For this study, we have modeled family history to identify individuals to select for sequencing using three approaches. First, we used an expert opinion approach in which multiple investigators evaluated pedigree structures with family history to reach a consensus priority in selecting individuals. Second we developed an approximation that assumes a low penetrance among noncarriers and uses known cancer status in first degree relatives. Third, we used the cancer, smoking behaviors and family information along with a likelihood that models penetrance according to genetic susceptibility and an interaction with smoking to estimate the probabilities individuals carry susceptibility factors. In this analysis we compare results among these methods. We evaluate effectiveness of these methods using simulation-based approaches. Citation Format: Christopher Amos, Joan Bailey-Wilson, Susan Pinney, Maria de Andrada, Colette Gaba, Diptasri Mandal, Ming You, Margaret Spitz, Marshall Anderson. Optimal strategies for selecting samples for sequencing families with a common cancer: The Genetic Epidemiology of Lung Cancer Consortium. [abstract]. In: Proceedings of the Eleventh Annual AACR International Conference on Frontiers in Cancer Prevention Research; 2012 Oct 16-19; Anaheim, CA. Philadelphia (PA): AACR; Cancer Prev Res 2012;5(11 Suppl):Abstract nr A106.