What is fascinating about the ecology argument is that it takes into account not only cell evolution, but the environment the tumor is living in as well. By looking at both together we can potentially develop better therapies.
—Kenneth J. Pienta, MD
Kenneth J. Pienta, MD, and his colleagues at Johns Hopkins Medicine in Baltimore are using the principles of evolutionary game theory to learn how cancer cells cooperate within a tumor to gather energy and metastasize. Game theory, the mathematic study of strategic decision-making that is commonly used to predict conflict and cooperation between individuals, is increasingly being applied in science to predict cell-to-cell interactions in biology with an ecologic perspective.
In a recent study,1 Dr. Pienta used a game theory model based on biologic interactions between two types of cancer cells—hypoxic (or energy deficient) and oxygenated—to study the metabolic symbiosis in which lactate from stromal cells fuels energy generation in the oxygenated microenvironment of the tumor, while hypoxic cells use sugar glucose to produce energy and promote tumor progression.
Dr. Pienta’s research found that within certain ranges of cell mutation rates, transitions take place in which a tumor switches between different types of energy metabolic strategies. During this process might be the ideal time to disrupt cancer cell cooperation and make metastatic cells more vulnerable to cancer therapeutics, he speculated.
Dr. Pienta is the Donald S. Coffey Professor of Urology at the Johns Hopkins Brady Urological Institute, Director of the Prostate Cancer Program at the Johns Hopkins Kimmel Cancer Center; and Professor of Oncology and Professor of Pharmacology and Molecular Sciences at Johns Hopkins Medicine. In 2007, Dr. Pienta was the recipient of the American Association of Cancer Research Team Science Award and is a two-time recipient of the American Cancer Society’s Clinical Research Professor Award.
The ASCO Post talked with Dr. Pienta about his research in the ecology of cancer and how game theory can be used to target the tumor microenvironment and lead to more effective cancer therapies.
Please describe how you are using evolutionary game theory to learn how cancer cells cooperate to gather energy and multiply.
Game theory allows us to look at how cancer cells interact with each other, as well as how they interact with various other cells of the host within the tumor microenvironment. We have been using various ecologic frameworks, including cooperation theory, game theory, and population biology, to try to come up with and think about new ways to gain insights into how cancer metastasizes in people and eventually kills them.
We start with a computational model, which in our most recent paper looked at cell metabolism. We have also used population biology computational models to look at cell metastasis and cooperation theory models to look at how cancer cells share resources like cytokines, hormones, and growth factors to basically ask, how can cancer cells help or hurt each other in the microenvironment of the tumor? And as the cells grow and become hypoxic, how can they create a resource-sharing environment?
Your research has found that an intercellular signal can induce a critical transition between high and low levels of tumor glucose consumption. How does that process affect cancer growth?
In contrast to normal differentiated cells, which rely primarily on mitochondrial oxidative phosphorylation to generate the energy needed for cells to grow, many cancer cells appear to rely on aerobic glycolysis—a phenomenon called the “Warburg effect.” We have long known about the Warburg effect. The question is, why would a cancer cell that uses regular oxidative phosphorylation to create ATP and energy, use the glycolytic pathway, which uses lactate to produce ATP?
What we have suggested, and first tried to model here, is that cancer cells that are stressed can actually use both oxidative phosphorylation as well as aerobic glycolysis when there is still enough oxygen around. The modeling we have done shows that a cancer cell, in an environment that is producing lactate, can actually use both oxidative phosphorylation and aerobic glycolysis to produce energy for different cellular functions—even when there is enough oxygen around. Now we are testing this tumor metabolism coupling in our in vitro systems.
Targeting Metabolic Coupling
How can the process of metabolic coupling in tumors be broken?
We are studying that now. The question is, is there a drug that can break the shuttle somehow between oxidative phosphorylation and aerobic glycolysis? Researchers have been trying to target the Warburg effect in cancer cells as a therapy for a long time, and we don’t yet have a magic bullet to do that. But what our study results provide is some new insights into how one cancer cell can help another cancer cell survive that hypoxic environment—that insult—potentially to its own detriment.
At the risk of anthropomorphizing a disease, we view cancer as a community of cells that are interacting together—it is not every cancer cell for itself—and these cancer cells interact in a mindless manner to help each other.
Cooperation theory and game theory allows us to set up modeling of mutualism relationships, in which both parties benefit, as well as parasitism, where only one of the cells benefits and the other is hurt. For reasons we don’t yet understand, one cancer cell will parasitize another.
Are there some cancer types that are more vulnerable to metabolic uncoupling than others?
We don’t know the answer to that yet. We think that solid tumors in which hypoxic environments are created are more likely to benefit from this process, but we don’t have any proof yet.
How do you envision game theory models being used in the clinic to disrupt cancer cell cooperation?
Unfortunately, I cannot make that jump from the laboratory to the clinic yet. That said, one example of how we’ve used cooperation theory to help our own framework is we established a cooperation game theory relationship between tumor-promoting macrophages in cancer cells and showed through game theory models how they can help each other exhibiting mutualism. We found that the cancer cells release factors that help the tumor attract the tumor-associated macrophages and help them grow. Then we found how the tumor-associated macrophages secrete matrix metalloproteinases and break down the environment, so the cancer cells can escape, as well as secreting a lot of other factors like interleukin 6 that help the cancer cells grow and become resistant to therapy.
So we took those findings, proved them in vitro and in vivo, and launched a clinical trial to treat metastatic prostate cancer using a monoclonal antibody to block one of the main factors that attract macrophages, CCL2, to the tumor microenvironment. Unfortunately, the trial failed, not because the theory was wrong, but because the antibody was not strong enough.
Now we are developing therapies based on knocking out the macrophages themselves. We just have not had time to develop the interventional strategies from a metabolism standpoint.
Will it be possible to detect and interrupt cell metabolic coupling and stop cancer metastasis over the next decade?
Absolutely. Over the past couple of years, there has been a true resurgence in interest in cancer metabolism, in learning about what is going on metabolically in a tumor, and in how that process can be targeted.
I wish I could report to you that I have a new therapy based on our findings. But we are looking very hard not only at the tumor-associated macrophage interruption and targeting story, as well as the metabolism story. I am very keen on thinking about cancer as an ecosystem, and I believe that in the next 10 years there is going to be a huge step forward in this area.
A few years ago, oncologists were enthusiastic about sequencing every patient’s tumor, and that is still true today. But except in a few instances, the number of actionable targets for which a drug has been found is probably less than 20%. And while those targeted therapies are good, they are not curative. We are now recognizing that such a low percentage is just not good enough.
What is fascinating about the ecology argument is that it takes into account not only cell evolution, but the environment the tumor is living in as well. By looking at both together we can potentially develop better therapies, and we are doing this all the time. Bevacizumab (Avastin), for example, is a microenvironment-targeted drug.
We think that the answer to how cancer is going to be treated in the future is with combination therapy that targets both the malignant cells and the cancer ecosystem. ■
Disclosure: Dr. Pienta reported no potential conflicts of interest.
1. Kianercy A, Beltri R, Pienta KJ: Critical transitions in a game theoretic model of tumour metabolism. Interface Focus 4:20140014, 2014.