Abstract PR011: Association of genetic ancestry with lung-cancer risk in White and Black ever-smokers

Abstract

Abstract Introduction: Black smokers have higher lung cancer risk, and develop cancer at earlier ages, than White smokers. It is unknown how much of this excess risk is due to race as a social construct, and how much is due to biological factors such as genetic ancestry. We examine the association of genetic ancestry with lung-cancer risk, controlling for lung-cancer risk factors and self-reported race/ethnicity, in the PLCO cancer screening trial. Methods: We used PLCO data on ever-smokers from 52,708 White smokers (2889 lung cancers) and 2585 Black smokers (206 lung cancers) with genetic ancestry determined by GRAF (https://github.com/ncbi/graf) on a set of 10,000 pre-selected fingerprinting variants. Percent ancestry for each of Europe, Africa, and East-Asia was determined for each person. Because percent ancestry is strongly associated with self-reported race/ethnicity, we also standardized the ancestry variables to have mean=0 and variance=1 for each self-reported race/ethnicity, to isolate the potential biological effect of genetic ancestry from self-reported race/ethnicity as a social construct. Because European and African ancestry are negatively correlated, we considered European minus African ancestry (EMAA) and European ancestry. We fit Cox models for time to lung cancer diagnosis, controlling for smoking history and other lung-cancer risk factors. Results: Self-reported White smokers averaged 98.1% European ancestry (standard-deviation(𝜎)=2.8%) and 0.8% (𝜎=1.7%) African ancestry, while self-reported Black smokers averaged 22.0% (𝜎=14%) European ancestry and 75.8% (𝜎=14%) African ancestry. Controlling for smoking history and all lung-cancer risk factors, standardized European ancestry was associated with increased lung-cancer risk (HR=1.13 per 1 race/ethnicity-specific standard-deviation increase in European ancestry, 95%CI: 1.02-1.25, p=0.02) that is partly counteracted by increasing EMAA (HR=0.95, 95%CI: 0.90-1.003, p=0.06). In this model, self-reported Black smokers also had increased lung-cancer risk versus self-reported White smokers (HR=1.56, 95%CI: 1.35-1.80, p<0.0001). Limiting the analysis to self-reported White smokers, European ancestry was associated with increased lung-cancer risk (HR=1.06 per 1% increase in European ancestry, 95%CI: 1.01-1.11, p=0.03) that is partly counteracted by increasing EMAA (HR=0.97, 95%CI: 0.94-1.002, p=0.07). However, limiting the analysis to self-reported Black smokers, there was no association of ancestry with lung-cancer risk (European-Ancestry: HR=0.98, p=0.7; EMAA: HR=1.01, p=0.7). Conclusions: After adjusting for lung-cancer risk factors and self-reported race/ethnicity, increasing European ancestry was associated with increasing lung-cancer risk that was partly counteracted if the difference in European and African ancestry also increased. Associations with ancestry were small and may be vulnerable to residual confounding and other epidemiologic sources of bias. Associations with ancestry could not be detected among self-reported Black smokers, suggesting that larger studies are needed. Citation Format: Courtney D. Dill, Dontray Trump, Rebecca Landy, Li Cheung, Wen-Yi Huang, Sonja Berndt, Neal Freedman, Hormuzd Katki. Association of genetic ancestry with lung-cancer risk in White and Black ever-smokers [abstract]. In: Proceedings of the 15th AACR Conference on the Science of Cancer Health Disparities in Racial/Ethnic Minorities and the Medically Underserved; 2022 Sep 16-19; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2022;31(1 Suppl):Abstract nr PR011.