Abstract 827: Estimating breast tissue-specific epigenetic age using next-generation methylation sequencing data

Abstract

Abstract Epigenetic age captures both genetic and environmental influences across time on cellular functions and is an indicator of biological aging. Epigenetic age may deviate from chronological age substantially in individuals. Epigenetic age is also tissue-specific. Emerging evidence suggests that female breast tissue ages faster than other parts of the body according to epigenetic age estimation using the "Horvath Clock" model. The Horvath method is based on the DNA methylation of 353 CpG loci on the outdated Illumina microarray platforms. The increasing availability of next-generation sequencing data calls for method development that uses DNA methylation sequencing data to estimate tissue-specific epigenetic age. We developed a new method to estimate breast tissue-specific epigenetic aging using next-generation methylation sequencing data and assessed the difference between epigenetic and chronological ages, known as epigenetic age acceleration (EAA), in different breast tissue types. The Illumina TruSeq Methyl Capture EPIC Sequencing technology was used to obtain DNA methylation profiles of approximately 3.3 million CpG sites in 111 tumor, 48 matched adjacent normal, and 462 normal breast tissue samples. A total of approximately 1.4 million CpG sites remained after quality control. Following the Horvath approach, we used an elastic net penalized regression model to regress chronological age on CpG sites in normal breast tissue and defined a new set of 247 clock CpGs specific to breast tissue with randomly divided training (n = 370) and testing (n = 92) data sets. We estimated breast tissue-specific epigenetic age and EAA in tumor, adjacent, and normal breast tissue. We found that breast tissue-specific epigenetic age was positively correlated with chronological age (r=0.87; P<2.2X10-16). Neither normal nor adjacent normal breast tissue showed a significant EAA. However, tumor breast tissue had a significant and increased EAA (median = 8.0 years; P=6.9X10-9). While triple-negative breast tumors showed no significant EAA, hormone receptive-positive and Her2-positive breast tumors had a significant and increased EAA (median=9.5 and 13.1 years; P=1X10-5 and 0.02, respectively). Results of this new model were compared to similar results using the Horvath Clock model and were found to be qualitatively consistent. Further research is needed to determine whether epigenetic age acceleration in normal breast tissue is predictive of breast cancer risk and how breast cancer risk factors influence the rate of acceleration. Citation Format: James R. Castle, Nan Lin, Jingpeng Liu, Chi Wang, Yunlong Liu, Chunyan He. Estimating breast tissue-specific epigenetic age using next-generation methylation sequencing data [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 827.