Abstract 7029: Trajectory of DNA methylation in blood up to 18 years prior to lung cancer diagnosis

Abstract

Abstract Purpose: Lung cancer (LC) has the highest mortality rate in South Korea. As prediction and early detection of lung cancer is pivotal to enhance survival rates, we attempted to investigate the trajectory of DNA methylation (DNAm) several years prior to lung cancer diagnosis by using prediagnostic blood samples. Methods: From National Cancer Center Health Examinee Cohort, established in 2003, we selected study participants who became diagnosed with LC several years after health examination, thereby collecting prediagnostic samples from 18 years to 1 year prior to LC diagnosis. Normal controls (n=150) were matched with patients (n=150) by sex, age, and the time of sample collection. Epigenome-wide DNAm data were obtained by using Infinium Human MethylationEPIC v2.0 Beadchip. We assumed non-linear relationship between DNAm changes and the time-to-diagnosis and estimated an optimal threshold point of non-linear relationship with a piecewise linear regression analysis at each CpG sites (CpGs). We used the ratio of M-values of patients and matched controls to establish the non-linear model composed of two linear models with a threshold point where the slope changes. Then, we performed functional enrichment analysis by each year prior to LC diagnosis using ReactomePA. Results: We identified threshold models with a threshold point where the slope changes from zero (from 18 years prior to LC up to the threshold point) to positive or negative (from threshold point to the point of LC diagnosis), which we termed “flat-to-change model”. Out of 790,048 CpGs, we identified a total of 7214 CpGs showing the “flat-to-change model”, of which 944, 789, 480, 413, 687, 542, 516, 1140, 315, 284, 327, 432, 83, 95, 74, 32, and 23 CpGs were found prior to 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, and 17 years before LC diagnosis, respectively. Functional enrichment analysis showed that various pathways were associated at different time point berfore LC diagnosis. CpGs of which DNAm was changed two years before LC diagnosis were associated with telomerase extension by telomerase involving ANKRD28/TERT/TERF2/PPP6R3/WRAP53. Transcriptional regulation by TP53 was associated with DNAm changes before both 5 and 14 years of LC diagnosis via POLR2A. Conclusions: We identified the trajectories of DNA methylation up to 18 years prior to LC diagnosis, and found that a greater number of CpG sites showed DNAm changes adjacent to the time point of LC diagnosis. CpG sites associated with specific time point before LC diagnosis may elucidate the mechanisms of tumorigenesis and also have potential for LC prediction markers. Citation Format: Yoon-Jung Choi, Ki-Jae Hong, Yeon-Su Lee, Woo Jin Kim, Yeol Kim, Sun Ha Jee, Jeongseon Kim. Trajectory of DNA methylation in blood up to 18 years prior to lung cancer diagnosis [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 7029.