Research News

Mathematical Models Yield Insights Into How Cancer Begins to Spread

Metastasis, the process by which cancer spreads from its originating location, is a critical but poorly understood phenomenon that is perhaps the most important factor in cancer being a highly lethal disease. One of the key steps in metastasis occurs when tumor cells gain the ability to migrate away from the tumor, a property known as motility. Now, Alexander Anderson, of the H. Lee Moffitt Cancer Center and Research Institute, and Philip Gerlee, a former graduate student of Dr. Anderson’s who is now at the Niels Bohr Institute in Copenhagen, have created a computer model that provides new insights into how tumor cells acquire motility.

Reporting their work in the Journal of Theoretical Biology, Drs. Anderson and Gerlee describe how they constructed their model starting with the premise that the interactions between a tumor cell and its microenvironment are critically important to metastasis. They also gave cells in the model the ability to mutate each time they divide.

One result from the model suggests that the emergence of motile cells is a rare event, but that when it does occur it dramatically alters the dynamics of the model. In response to the emergence of motile cells, tumors change shape and grow at faster rates. These changes appear to increase the invasive qualities of the tumor without requiring vast increases in cell number. The model also provided some insights into how motile tumor cells balance the nutrient requirements that would allow them to travel versus grow and divide, generating three different scenarios for how migrating cells interact with the extracellular matrix on which they travel.  The model also predicts that there is no one single phenotype or genotype that can be associated with cell motility.

In another study, this one published in the journal PLoS ONE, Sridhar Ramaswamy and his colleagues at Harvard Medical School modeled the interactions between tumor cells and fibroblasts, a component of the stroma, the normal tissues that surround a tumor. For this study, the investigators started with a large data set that they generated by growing in pair 12 different human cancer cell lines with 36 human fibroblast cell lines derived from skin and lung.

For each of these 432 pairings, the investigators compared tumor cell growth with fibroblasts to growth without fibroblasts and found that only 12% of the pairing stimulated tumor growth, while 41% inhibited tumor growth and the remaining 47% showed no effect.  Systems-level modeling of the data revealed that fibroblasts express two distinct activities and that these can either work together to modulate cancer cell proliferation or at cross-purposed to one another, which impedes cancer cell growth.

The work on the development of tumor cell motility, which was funded by the National Cancer Institute, is detailed in a paper titled, “Evolution of cell motility in an individual-based model of tumour growth.” An abstract of this paper is available at the journal's Web site.
View abstract

The work on the interaction between tumor cells and fibroblasts is described in a paper titled, “Systems-Level Modeling of Cancer-Fibroblast Interaction.”  This paper is available free-or-charge at the journal's Web site.
View paper