NF-κB Epigenetic Attractor Landscape Drives Breast Cancer Heterogeneity

Abstract

Heterogeneity in breast cancer (BC) subtypes contributes to therapy resistance and recurrence. Subtype heterogeneity arises from stochastic genetic and epigenetic changes, phenotypic plasticity, and microenvironment-driven selection during tumor evolution. Here, we investigate how NF-{kappa}B epigenetic variability contributes to HER2+ BC progression. Using RNA-seq, we quantified NF-{kappa}B, TWIST1, SIP1, and SLUG expression in two BC cell lines: HCC-1954 (HER2+) and MDA-MB-231 (TNBC). Next, we built and calibrated a gene regulatory network model reproducing transcriptional interactions among these genes. The models epigenetic landscape displays two attractor basins that reproduce the HER2+ and TNBC expression profiles. Validation was performed using DHMEQ-treated cells, published patient and in vitro data. Stochastic fluctuations in NF-{kappa}B levels induce spontaneous, irreversible transitions from HER2+ to TNBC states at variable times, contributing to heterogeneity. These transitions are mediated by an unstable intermediate state that provides a noise-sensitive route. Mutations or drugs altering NF-{kappa}B availability reshape basin sizes, altering basin sizes and transition probabilities. Our work refines the attractor landscape framework, linking NF-{kappa}B dynamics to BC heterogeneity, supporting more accurate classification, prognosis, and treatment strategies.