Abstract LB308: Rewiring of methylation during embryogenesis can seed cancer decades later

Abstract As we age, cells accumulate somatic mutations, undergo epigenetic alterations, and experience telomere shortening. In older individuals, normal tissues are often overtaken by expanding clones of cells that have acquired mutations promoting proliferation. Whether epigenetic changes can also provide selective advantage and induce clonal expansion in ageing is less clear. Here, we sequenced 719 whole genomes and methylomes of single-cell colonies derived from HSCs of three healthy individuals and three breast cancer patients. Using somatic mutations, we built lineage trees of the HSCs for each individual. To establish whether loss or gain of methylation was heritable, we developed a method that accurately infers, for each CpG site, its zygotic methylation state as well as the number and timing of ancestral methylation changes that explain the observed methylation state of each somatic cell. We analysed ∼24 million CpG sites per individual and find that methylation states are, in general, stably heritable over decades of life. We find that during embryogenesis, when the embryo comprises a few thousand cells, developmental cells lay down unique methylation haplotypes, spanning hundreds of base pairs in size, that are stably inherited by descendants. This phase of rewiring the methylome has striking properties: (1) thousands of regions genome-wide show heterogeneity of methylation profiles laid down during this phase; (2) these regions are often conserved across different individuals; (3) it precedes gastrulation, so that the same changes are seen in all germ layers; (4) the process happens quickly, completing within a few generations of cell division; (5) it coincides with establishment of inactive X chromosome methylation in females; and (6) the methylation affects many key regulatory regions of the genome. To assess whether embryonic methylation rewiring can contribute to cancer development, we studied women with breast cancer who had BRCA1 promoter methylation. The tumour samples did not have germline or somatic mutation of BRCA1 (or other DNA repair genes), but had BRCA1 methylation along with the typical mutational signatures of homologous recombination deficiency. Intriguingly, we found that the same methylation change was present heterozygously in a subset of HSCs; that these HSCs all derived from one embryonic cell that pre-dated gastrulation; and that the somatic mutations defining this lineage of HSCs were also present clonally in the breast cancer. Thus, aberrant methylation of the BRCA1 promoter in embryogenesis was transmitted across germ layers, in blood and breast epithelium, and (with loss-of-heterozygosity) drove homologous recombination deficiency in a breast clone that transformed to cancer 4-6 decades later. By comparing normal healthy individuals of different ages with cancer patients, we have been able to unravel the process of normal ageing from cancer development at an unprecedented resolution. Citation Format: Lori D. Kregar, Nicholas Williams, Joe Lee, Michael Spencer-Chapman, Oleksii Nikolaienko, Emily Mitchell, Liv B. Gansmo, Per E. Lønning, Elisa Laurenti, Lucy Yates, Stian Knappskog, Jyoti Nangalia, Peter Campbell. Rewiring of methylation during embryogenesis can seed cancer decades later [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 2 (Late-Breaking, Clinical Trial, and Invited Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_2):Abstract nr LB308.