Abstract. In recent years, novel therapies for treating cancer by means of a patient’s own immune system have emerged. Checkpoint-blockade immunotherapies are designed to enable a patient’s immune cells to recognize and destroy tumor cells. The process of recognition is based on specific protein binding interactions between the immune cells and cancer cells. Because these interactions depend on mutations in the cancer genome, immune recognition becomes an evolutionary problem. In this talk, I will present a new mathematical model of tumor evolution based on the fitness cost of tumor cells due to immune recognition. The model successfully predicts tumor response to checkpoint blockade immunotherapy, as shown in patient cohorts with melanoma and lung cancer. Our results highlight evolutionary similarities between cancer and viral pathogens and suggest general concepts of predictive analysis in fast-evolving systems.
Short Bio. Marta Łuksza is an assistant professor at the Icahn School of Medicine, Mount Sinai in New York. She completed her Ph.D in Computer Science at the Freie Universität and the Max Planck Institute for Molecular Genetics in Berlin. Her research focuses on the evolutionary dynamics of fast-evolving systems, such us viruses, cancer and the immune system. She has developed predictive computational models for the evolution of cancer and the human influenza virus. She consults the WHO Influenza Unit in their bi-annual vaccine composition meetings.