Optimizing Chemotherapy Treatments Using Evolutionary Game Theory
Professor Paul Newton and Dr. Jeffrey West, two researchers at USC Viterbi, have developed a tool that uses evolutionary game theory to predict how certain cancer and non-cancer cells will respond to different chemotherapy schedules and dosages. Through using the mathematical payoffs associated with evolutionary game theory, the researchers have discovered a method which uses data based on an individual’s cell information to run mathematical models which can help identify the best course of treatment, specifically designed for each individual’s unique condition. Tumors are composed of many different kinds of cancer cells, which makes them difficult to effectively attack because of their differing growth rates. Resistant cancer cells that survive through initial rounds of treatment have less competition and more opportunity to grow because of increased space and nutrients available to them, which makes the following rounds of chemotherapy less effective upon the resistant cells unless doctors can become aware of exactly which types of cells they should be treating during each round of treatment. Newton and West have developed a tool that will hopefully allow doctors to predict which methods of treatment will be most effective throughout the chemotherapeutic process using mathematical models based on evolutionary game theory.
Patients with cancer usually receive either a maximum tolerated dose (MTD) or a low-dose metronomic (LDM) of a chemotherapy drug. Through comparing existing research on cancers treated by MTD and LDM, Newton and West were able to model the competition between cells—using evolutionary game theory. The model chooses two cells to “compete” against one another, with fitness determined by each specific type of cell’s growth rate. The players are each of the different types of cells found in an individual’s body: healthy cells, targeted cancer cells, and resistant cancer cells. Through placing different combinations of cells in competition against one another, the payoff is determined by each cell’s fitness, generating millions of combinations of competitions which represent the multitude of ways the tumor can progress over time.
The researchers then use this information from the evolutionary game theory matrixes to produce models which predict which combinations of drug concentrations and frequencies (MTD or LDM) are most effective to target each individual’s disease. Through regulating chemotherapy schedules based upon the tumor’s growth rate, Newton and West have discovered how competition among cells changes throughout one’s recovery, requiring a treatment which evolves with each individual’s changing competition among cells within their body. This new research could be pivotal in how doctors treat cancer in the future, allowing doctors to understand the outcome of competition between cells through evolutionary game theory.