AHRR hypomethylation in heavy smokers: Associations with lung cancer risk and mortality.

Abstract Hypomethylation of cg05575921 in the aryl-hydrocarbon receptor repressor (AHRR) gene is consistently and strongly associated with higher exposure to smoking (e.g., current smoking, greater number of pack years, and less time since quitting). Although cg05575921 hypomethylation has been reported to be associated with lung cancer risk and mortality, this has not been explored within heavy smokers or by histotype. We assessed cg05575921 methylation for associations with smoking behavior, prospective lung cancer risk, and mortality in 313 cases and controls from the Beta Carotene and Retinol Efficacy Trial (CARET) with ≥20 pack years of smoking and matched on age (±5 years), sex, race/ethnicity, enrollment year, smoking status, occupational asbestos exposure, and follow-up time. Methylation of cg05575921 in blood collected on average 4.3 years prior to diagnosis in cases was assayed using the 850K Illumina EPIC array. Quintiles of cg05575921 methylation were defined based on the control distribution, with the lowest quintile (Q1, referent) representing hypermethylation and the highest (Q5), hypomethylation. Among controls, increasing quintiles of cg05575921 hypomethylation were associated with current smoking status (Cochran-Armitage trend p=3.0 × 10-21), and with decreasing years since last smoked, increasing cigarettes per day, and decreasing body mass index (BMI) (ordinal ANOVA p=3.1 × 10-21, 1.3 × 10-17, and 0.002, respectively). We observed similar patterns in cases. We evaluated associations between cg05575921 methylation and lung cancer risk using logistic regression models conditioned on matching factors and adjusted for age, years since quit, and BMI. We evaluated cg05575921 methylation and lung cancer-specific mortality using Cox proportional hazards models adjusted for stage, age, sex, race, smoking status, intervention arm, asbestos exposure, and pack years smoked. We did not observe clear patterns of associations between cg05575921 methylation and lung cancer risk overall or by histotype. However, we found a linear relationship (p=0.03) between increasing cg05575921 hypomethylation and overall lung cancer mortality, with Q5 vs Q1 Hazard Ratio (HR)=1.60 (95% Confidence Interval (95% CI): 1.00-2.57) for mortality from lung cancer overall, and HR=2.09, 95% CI: 0.96-4.54 for adenocarcinoma (N=121). Our findings provide preliminary support that even among heavy smokers, cg05575921 hypomethylation is associated with more recent and extensive exposure to cigarette smoking, and that prediagnosis cg05575921 hypomethylation in lung cancer cases is associated with an increased hazard of death. However, since the CARET population represents a group who would largely qualify for lung cancer screening with annual low-dose computed tomography per the United States Preventive Task Force guidelines, our results do not provide support for the use of cg05575921 hypomethylation as an indicator for screening-based lung cancer risk stratification. Citation Format: Stefan Graw, Matt J. Barnett, Mark D. Thornquist, Gary E. Goodman, Chu Chen, Devin C. Koestler, Carmen J. Marsit, Jennifer A. Doherty, Laurie Grieshober. AHRR hypomethylation in heavy smokers: Associations with lung cancer risk and mortality [abstract]. In: Proceedings of the AACR Special Conference on Modernizing Population Sciences in the Digital Age; 2019 Feb 19-22; San Diego, CA. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2020;29(9 Suppl):Abstract nr A20.