Lung cancer is one of the most common cancers worldwide. While obesity is a strong risk factor for certain types of cancer, such as colon, breast and endometrial cancers, many epidemiological studies have consistently indicated an inverse association between body mass index (BMI) and risk of lung cancer after adjusting for other established risk factors. It is largely controversial whether the observed inverse association is due to the biological genuine effects of BMI or systematic biases. Mendelian randomization (MR) is an analytical approach that uses genetic variants as the instrumental variable (IV) to infer the causal relationship between an exposure variable and disease. However, because of the lack of efficient statistical tools, MR often requires an extremely large sample size to achiev adequate statistical power. In this proposed study, we will develop a novel statistical method that utilizes genome wide association study (GWAS) data to construct the IV for obesity traits in MR analysis. The major advantage of this approach is that it can unravel the missing heritability of obesity traits that is not accounted for by the known genetic variants, and thereby provide substantially improved statistical power. Because this study will utilize the existing individual GWAS data as well as detailed epidemiologic data from about 20,000 lung cancer cases and 20,000 controls in TRICL (Transdisciplinary Research in Cancer of the Lung), it will be conducted in an extremely cost-efficient manner. This study will provide a unique opportunity to answer the longstanding question of whether there are causal effects of obesity traits on lung cancer risk, potentially to open new avenues for further studies in understanding the etiology of lung cancer, and to provide important statistical tools for facilitating investigation of the causa relationship between risk factors and diseases in general.
Lung cancer is one of the most common cancers worldwide. Many epidemiological studies have consistently indicated an inverse association between body mass index (BMI) and risk of lung cancer. It is largely controversial whether the observed inverse association is due to the biological 'genuine' effects of BMI or systematic biases. By developing a novel and efficient statistical approach for Mendelian Randomization and taking advantage of existing individual GWAS and detailed epidemiologic data from about 20,000 lung cancer cases and 20,000 controls in TRICL (Transdisciplinary Research in Cancer of the Lung), this project will evaluate the causal role of obesity traits in lung cancer without the biases that usually plague observational epidemiological studies. This study will therefore potentially have a significant public health impact by addressing an important and controversial question about the etiology of lung cancer that, so far, has not been addressed adequately, opening new avenue for further studies in understanding the etiology of lung cancer, and providing important statistical tools for facilitating investigation of the causal relationship between risk factors an complex diseases in general.
| Wang, Tao; Xue, Xiaonan; Xie, Xianhong et al. (2018) Adjustment for covariates using summary statistics of genome-wide association studies. Genet Epidemiol 42:812-825 |
| Wang, Tao; Hosgood, H Dean; Lan, Qing et al. (2018) The Relationship Between Population Attributable Fraction and Heritability in Genetic Studies. Front Genet 9:352 |
| Doyle, Christopher R; Moon, Jee-Young; Daily, Johanna P et al. (2018) A Capsular Polysaccharide-Specific Antibody Alters Streptococcus pneumoniae Gene Expression during Nasopharyngeal Colonization of Mice. Infect Immun 86: |
| Wang, Tao; Moon, Jee-Young; Wu, Yiqun et al. (2017) Pleiotropy of genetic variants on obesity and smoking phenotypes: Results from the Oncoarray Project of The International Lung Cancer Consortium. PLoS One 12:e0185660 |
