A Cornell-led team used Levy distribtion analysis to characterize the heterogeneity of tumor cell motility. Using this statistical modeling technique more commonly used in physics and economics, the team was able to demonstrate how the heterogeneity of cancer cells can be influenced by their chemical environment—namely by interactions with a specific protein, which leads to tumor growth.
As part of the study, the team fabricated a microfluidic chip containing four identical three-channel devices. They then put breast tumor cells and the chemokine protein CCL19 into each device and then used open-source software to analyze the cancer cell behavior. Details on the study were published recently in Integrative Biology.
"It's pretty tough to treat cancer. A lot of people in the field believe that is because of the diversity in the cancer population," said senior author Mingming Wu. "While immune cells are rounded and kind of similar and move in the same way, cancer cells are different in shape and move at different speed. We know that fast movers are very lethal. How would you quantify that heterogeneity?"
In order to model the cancer cell trajectories, the team used Root, an open-source software for performing statistical analysis in high-energy physics and in certain economics applications. The researchers found that the presence of chemokine caused the targeted cancer cells to move faster, and heterogeneity increased.
"It is similar to how we as a society are trying to make the population more diverse, because we know that if the population is diverse, it's more robust, more healthy," Wu said. "I think that cancer is the same way. They are making their population more diverse, more indestructible."
A treatment that inhibits the receptor to CCL19 could potentially decrease the invasiveness of tumor cells, although that might also cause the cancer cells to adopt new, even stealthier strategies to survive, Wu added.