The Role of Game Theory in Cancer Treatment?
https://jamanetwork.com/journals/jamaoncology/article-abstract/2696342
The article, titled “Optimizing Cancer Treatment Using Game Theory,” constructs a unique bond between a physician’s therapy and cancer cells’ biological resistance strategies. It assigns physicians’ approach to systematic cancer treatment as “player 1” and the cancer’s invariable evolution of resistance as “player 2.” Some interesting factors that play a significant role in this study include:
- The only player that actually “plays” is the physician and his or her treatment plan. Consequently, only the physician player can anticipate future events. Because the cancer population merely reacts to the hand that is dealt to them, and unpredictably at that, it would be inaccurate to assert that the cancer cells have a so-called “game plan”; they simply adapt.
- Next, the article describes the relationship between the two players as “Stackelberg” dynamics: a situation in which player one plays his move and player two (the follower) responds, in this case, to player one’s therapy of choice.
The physicians are essentially playing a game: they enter the arena as a predator, looking for ways to maximize their efficiency in preying for cancer cells. The end goal – to eliminate the prey. This can be seen somewhat as an impossible goal, as although cancer cells are not players capable of making conscious choices, they are still “active” players in that they respond to treatment plans laid out by player one. An important note here is that while the cancer cells do develop resistance to treatments over time, the physicians also evolve in a sense by varying and experimenting with new treatment plans. As a Stackelberg game, this relationship allows physicians to loosen their focus on immediate responses that is dependent on precision medicine. It also allows them to incorporate this relationship into treatment plans in order to gear the treatments more specifically toward patients and new resistances in cancer populations, a concept termed “treatment-specific evolutionary dynamics. By establishing a unique game theory approach to cancer treatment, physicians are better able to use a dynamic approach to tackling cancer, and consequently to keep an open mind in exploring different methods of doing so.
The ideas of the article are based on the game theory topics from last week’s lectures. In particular, the article studied the interconnectedness found between the behavior (treatment) of the physician and the response of the cancer cells. While this is not the case where both players have the freedom to choose their route, it still highlights a specific relationship between the two parties, ultimately allowing us to form basic models in terms of game theory. Specifically, in framing the contest so that the cancer cells’ response becomes player two, the article “assume[s] that most malignant tumors contain multiple subpopulations with varying sensitivities to available therapies,” which gives a cause-and-effect relationship that we can model, as we did with the different examples shown in class.
this is cool.