Enforced ZFP281 expression delays breast cancer initiation and can provide lifelong protection against breast cancer metastasis
Breast cancer metastatic reactivation and its links to mammary development are largely unknown. Here, using conditional knockout and overexpression in normal and HER2+ mammary glands, we show that the dormancy regulator ZFP281 promotes branching and dissemination while suppressing growth, and its overexpression can even override HER2-driven cancer initiation. Notably, while ZFP281 does not limit HER2-driven early dissemination, it constrains DCC plasticity, confining cells to a dormant mesenchymal/hybrid-like state and effectively suppressing metastasis throughout the mouse lifespan. ZFP281 is induced by estrogen, progesterone, and glucocorticoid signaling, and RNA sequencing of early lesions revealed that it regulates glutathione metabolism and ferroptosis, potentially supporting fitness during dormancy, while repressing angiogenesis, Th17-like inflammation, innate immune genes, and pro-inflammatory programs that might otherwise trigger awakening. Integrating these findings with human data, we show that DCIS and IBC (invasive breast cancer) lesions that later relapse are selectively enriched for ZFP281-regulated M-like and dormancy signatures and, after pregnancy, depleted for a ZFP281-induced stress-autophagy module, indicating that erosion or imbalance of these programs marks lesions that seed DCCs with higher reactivation potential. We propose that ZFP281 acts as a hormone-regulated dormancy gatekeeper that uncouples dissemination from growth, enforcing a metabolically fit, angiogenesis-low, immune-evasive dormant state in breast DCCs, thereby shaping the timing of metastatic relapse and potentially exploitable for durable prevention of metastasis. STATEMENT OF SIGNIFICANCEZFP281, a hormone-regulated dormancy gatekeeper, uncouples dissemination from growth and constrains DCCs into long-term arrest, defining human gene signatures that distinguish dormancy-prone from awakening-prone lesions and predict breast cancer relapse dynamics.