Abstract 7482: Mapping the temporal landscape of breast cancer using epigenetic entropy

Abstract Background: The age of a newly diagnosed breast cancer encodes valuable etiologic and prognostic information, yet it is considered a known unknown. Here we develop an entropy-based epigenetic clock that measures the age of individual breast cancers and quantifies the evolutionary-ecological forces that modulate growth. Methods: Using 450K methylation array data from 634 invasive breast cancers and 79 normal breast tissue samples, we identified 500 fluctuating CpG (fCpG) sites whose methylation states exhibit unbiased oscillations. After external validation of the fCpG set in an independent cohort (n=146), we derived an entropy-based epigenetic clock index (ECI) as a measure of tumor mitotic age. We correlated the ECI with tumor features and gene expression (GSEA) and quantified its relationship with the tumor microenvironment (CIBERSORTx). We validated the ECI in 40 patients with paired primary-metastasis samples and used mathematical models to estimate the calendar ages of individual tumors. Results: Compared to older tumors (higher ECI), younger tumors (lower ECI) had increased proliferation, higher grade, and larger size. Younger tumor age was strongly associated with the fraction of genome altered (R, -0.56; P<0.0001). Luminal A cancers were older compared to luminal B and basal cancers (P<0.001 each, Wilcoxon). Mitotically younger tumors were enriched for gene pathways related to proliferation and cell cycle control, whereas older tumors were enriched for immune pathways and immune-related signaling pathways. The extent of immune infiltration, and T-cell infiltration in particular, increased with mitotic tumor age. These findings suggest that a tumor’s effective growth rate is shaped by the competition between intrinsic growth potential and microenvironmental pressures (e.g., immune surveillance). In patients with paired primary-metastasis samples, the 500 fCpGs were strongly correlated within but not between patients (median R, 0.73 vs. 0.08; P<.0001, Wilcoxon). In 36/40 patients, the metastasis was younger than the primary, providing support for the interpretation of the ECI as a measure of tumor age. Anchoring the median age at a consensus estimate of 3 years, the inferred calendar ages among 413 cancers ranged from 0.2 to 35.2 years (IQR: 1.5-5.6 years). There were notable differences in median tumor ages by molecular subtype, ranging from 1.0 years in basal cancers to 6.5 years in luminal A cancers. Conclusions: This entropy-based epigenetic clock measures tumor growth on a scale of months to years, which contrasts with existing clocks that track longer timescales associated with biologic tissue aging and cancer risk. By delineating old and stable from young and aggressive tumors, the clock has potential applications in risk stratification and early detection. Although developed for breast cancer, the clock provides a scalable technology to characterize the broader temporal landscape of solid tumors. Citation Format: Daniel L. Monyak, Shannon T. Holloway, Graham Gumbert, Lars J. Grimm, Shelley Hwang, Jeffrey R. Marks, Darryl Shibata, Marc D. Ryser. Mapping the temporal landscape of breast cancer using epigenetic entropy [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 7482.