Abstract 5706: Epigenetic signatures in association with breast cancer mortality and prognosis that are specific to tumor subtype and menopausal status

Abstract

Abstract Background: Despite advancements in screening and treatment options for breast cancer (BC), mortality from BC has increased, accounting for 15% of new cancer deaths in the U.S. in 2020. BC is heterogeneous cancer with different molecular subtypes defined by hormone-receptor status, and the prognoses of BC patients differ by their menopausal status in conjunction with their molecular subtypes. Epigenetic dysregulation patterns can serve as principal predictors of BC progression and mortality by capturing these differences in biology that are linked to BC outcomes, leading to personalized treatment plans. In this study, we aimed to detect subtype- or menopausal-specific DNA methylation (DNAm) in BC tumor tissues across the genome that are associated with BC progression or all-cause mortality. Methods: Associations between site-specific DNAm in tumor and BC survival were evaluated using the Illumina Infinium HumanMethylation450 BeadChip array data from the Cancer Genome Atlas participants. Cox proportional hazards models were used in the strata by tumor subtypes and menopausal status, adjusting for age, race, cancer stage, tumor purity, and cell type proportion. Detected site-specific DNAm was further analyzed to identify subtype- or menopausal-specific differentially methylated regions (DMRs) and functional pathways. The validation of the results was carried out on an independent dataset. Results: We identified eight CpG probes that were associated with survival outcomes in subtype- or menopausal-specific manner. Seven probes among women with luminal A subtype were associated with all-cause mortality or progression-free interval (PFI); four probes among women with luminal B were associated with PFI; and five probes among post-menopausal women were associated with PFI. Of them, six probes presented a lower risk of all-cause mortality or BC progression with higher DNAm. The identified DMRs differed by tumor subtype and menopausal status, and a majority of DMRs were driven by several weaker associations in the same direction among proximal loci. Also, we identified subtype- or menopausal-specific functional genomic pathways. Particularly, a pathway of insulin regulation was associated with all-cause mortality and PFI among pre-menopausal women. The detected site-specific probes were validated with consistent directions and magnitudes of effects on prognosis across two independent datasets. Conclusion: We identified subtype- or menopausal-specific DNAm in BC tumor tissues, DMRs, and functional pathways in association with all-cause mortality or BC progression. Our findings warrant future studies with larger independent datasets for replication and functional implications of detected DNAm. Our detected genome-wide-associated-CpG loci could improve prognosis prediction for BC patients, thus contributing to tailored therapeutic regimens. Citation Format: Do Hyun Kim, Alexandra M. Binder, Hua Zhou, Su Yon Jung. Epigenetic signatures in association with breast cancer mortality and prognosis that are specific to tumor subtype and menopausal status [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 5706.