(596ag) Kinetic Models of Tissue Tropism in Breast Cancer Metastasis | AIChE

(596ag) Kinetic Models of Tissue Tropism in Breast Cancer Metastasis

Authors 



Metastasis is the leading cause of fatality for women diagnosed with breast cancer. The most common anatomical sites of distant tumor growth include the brain, lung, liver, and bone, and it is well known that this metastatic spread in breast cancer is not random. Rather, different clinical subtypes of breast cancer exhibit unique patterns of metastatic site preference, called tissue tropism. Given the physical and chemical diversity of these secondary tissue sites, we hypothesize that there is a relationship between the biophysical and biochemical properties of the tissue, and the ability of cells within a particular subtype of breast cancer to adhere, migrate, and grow. Here, we present Engineered Metastatic Microenvironments (EMMs), which capture the integrin-binding properties of the tissues of several common secondary sites in vivo (brain, lung, and bone). Using these EMMs, we have observed differences in cell behavior based upon the extracellular matrix (ECM) of these tissues, which can be used to determine specific mechanisms in metastatic site preference. We are building a kinetic model of the ability of cells to adhere, spread, and polarize in different subtype-EMM combinations, in order to eventually build a predictive model of tissue tropism in breast cancer patients. 

The EMMs we have designed consist of ECM proteins coupled to glass coverslips using silane chemistry. The composition of ECM varies with each EMM: brain: 1ug/cm2 of 50% fibronectin, 25% vitronectin, 20% tenascin and 5% laminin (wt%), lung: 2ug/cm2 of 3% laminin, 33% collagen IV, 15% collagen I, 15% fibronectin, and 4% tenascin (wt%), and bone: 5ug/cm2 of 99% collagen I and 1% osteopontin (wt%). We have quantified the real-time adhesion kinetics of four human breast cancer cell lines on each of these EMMs: MDA-MB-231, BT-549, MCF7 and SKBr3, which are classified into three distinct clinical subtypes: claudin low, luminal, and estrogen (ER) and progesterone receptor (PR) negative/human epidermal growth factor receptor-2 (HER2)-enriched. These clinical subtypes each have a known human, clinical metastatic profile [1]. We have imaged individual cell spreading on EMMs at high magnification, and observed qualitative differences in initial spreading. To quantify this, we developed an equation based upon Michaelis-Menten enzyme kinetics to describe cell spreading. This allows for the determination of two biologically relevant adhesion parameters: K, which is the time to half of the maximum cell spreading, and Amax, which is the maximum cell area.

Each of the cell lines shows distinct differences in their initial spreading on the three EMMs. The highest K values occur for all four cell lines on the bone, indicating that cells spread at a slower rate in the bone environment than in the lung or brain overall. We also observe faster rates of spreading and higher variability in maximum area in cell lines that are known to generally be more metastatic in vivo. This is a significant result, because an individual cell’s immediate reaction to the secondary tissue site may be critical to the eventual formation of a successful metastatic lesion. We are currently investigating whether integrin blocking will shift these rate curves, to understand mechanisms of how initial integrin-binding kinetics control the metastatic process. In addition to these well-established cell lines, we are now comparing these results with the adhesion kinetics of breast cancer samples from a local tissue bank.  Our preliminary observations suggest that even very aggressive, metastatic patient cell samples have much lower spreading kinetics than these well-established cell lines, challenging the translational use of cell lines in lab settings. We envision that this simple kinetic model of cell adhesion and spreading, combined with our work on polarization, migration, and proliferation, can be used to predict probable metastatic sites in patients, leading to patient-specific therapy and relapse site probability.

1. Kennecke, H., et al., J Clin Oncol. 28(20): p. 3271-7.