Abstract 4150: Nicotine dependence impacts the relationship between genetic variants and risk of lung adenocarcinoma

Lung cancer, the leading cause of cancer death, accounted for 27% of cancer deaths for Americans in 2013. Despite its negative impact on public health, effective early detection tools for lung cancer are still under development. Several genetic variations associated with lung cancer risk and smoking behavior have been identified in some genome-wide association studies. It is unclear, however, if the gene variants affect lung cancer risk directly or indirectly through nicotine dependence. In an attempt to elucidate the relationships between lung cancer, genetic variants and smoking, we applied mediation analysis to quantify the impact of nicotine dependence. We evaluated 37 single nucleotide polymorphisms (SNPs) in nine candidate genes (TERT, CLPTM1L, CHRNB3, CHRNA6, CHRNA5/A3/B4, CYP1A1, and TP53), previously reported to be associated with lung cancer risk or with smoking behavior. A total of 661 lung adenocarcinoma (ADC) cases and 1,347 controls with a smoking history obtained from the Environment and Genetics in Lung Cancer Etiology (EAGLE) case-control study were included in the study. Results show that nicotine dependence is a mediator of the association between lung ADC and gene variations of CHRNA3 and CHRNB4 and accounts for approximately 20% of this relationship. The association between two CHRNA3 SNPs (rs1051730 and rs12914385) and risk for lung ADC reached a genome-wide significance (p-value=1.9x10-10 and 1.1x10-10, respectively). Also, these two SNPs had the most significant indirect effect on lung ADC risk through nicotine dependence (p=7x10-5 and 3x10-4). Gene variations rs2736100 and rs2853676 in TERT and rs401681 and rs31489 in CLPTM1L had a significant direct association on lung ADC without an indirect effect through nicotine dependence. In addition, rs6474414 in CHRNB3 and rs4646421 in CYP1A1, were associated with nicotine dependence but not directly with lung ADC risk. Our findings provide valuable information for understanding the pathogenesis and for developing potentially effective screening tools for lung ADC. These identified SNPs may improve prediction accuracy of lung ADC at earlier stages, which is essential to enhancing survival rates. Translating these findings into public health practice may lead to tailored and customized smoking cessation interventions for Individuals with genetic variants, which have both direct and indirect effect through nicotine dependence on lung ADC. Citation Format: Tung-Sung Tseng, Jong Y. Park, Jovan