Breast Tumor Stiffness and Bone Metastasis: How “Soil Selects Soil”

Abstract

Breast cancer is the most common cancer amongst women globally and continues to produce considerable harm. The microenvironment of primary breast tumors plays a well-described role in promoting the growth and progression of the disease; however, despite our wealth of knowledge concerning how the biochemical and biophysical properties of primary tumors influence cancer cells in situ, very little is known about whether this influence persists after dissemination to metastatic sites, such as the skeleton. Bone metastases inflict the greatest morbidity associated with breast cancer, and they affect a majority of women with advanced disease. Although predictive gene signatures of osteolytic metastasis have been identified, their genesis remains obscure. Current models propose that bone metastases originate from rare subclones that arise stochastically during clonal evolution of primary tumors. In contrast, we reveal a deterministic origin of the osteolytic phenotype that lies in the response of breast cancer cells to mechanical stimuli in their primary microenvironment. Here, we show that primary tumor stiffness encodes a “mechanical memory” which instructs cancer cells to adopt and maintain distinct biophysical properties, in addition to promoting osteolytic bone metastasis. We present a “mechanical conditioning score” comprised of mechanically-regulated genes in order to proxy tumor stiffness response clinically, and we show that it is associated with bone-specific metastasis. Using a discovery approach, we trace mechanical memory to the mechanotransductive activation of RUNX2, an osteogenic gene bookmarker and bone metastasis driver. This combination of traits allows for the stable transactivation of osteolytic target genes which persists after cancer cells disseminate from their activating environment. Using genetic, epigenetic, and functional approaches, RUNX2-mediated mechanical memory can be simulated, repressed, selected, or extended. Taken together, these results show that the primary tumor microenvironment can determine the metastatic microenvironment, i.e. "soil selects soil."