T-cells are central players in the immune response, and the development of immunotherapies has transformed the treatment landscape for numerous cancers. Yet, the effects of these therapies vary across cancer types and patients. Each of the millions of T-cells in each person has a unique T-cell receptor (TCR) that recognizes different antigens.
This diversity has limited our understanding of what exactly T-cells recognize in the people who respond to treatment. Understanding how TCRs recognize antigens is critical to help realize the full potential of immunotherapy. Team MATCHMAKERS aims to develop machine-learning models capable of predicting the tumor antigens recognized by T-cells in a patient's tumor. The insights gained are also expected to have significant implications beyond cancer – in infectious disease, for example, as well as autoimmunity.
“Clinically active cancer immunotherapies work by stimulating T-cell receptor recognition of cancer antigens. We can read the sequences of these TCRs, but we can’t interpret them, we can’t infer which tumor antigens are being recognized. MATCHMAKERS aims to change this and thereby promote the development of novel diagnostic and therapeutic strategies, including personalized cancer vaccines and cell therapies.”
Team MATCHMAKERS unites clinicians, patient advocates, and scientists with expertise in cancer immunology, computer science, high-throughput method development, and structural biology. This collaboration spans 10 institutions in the United States, Germany, Norway, the United Kingdom, and the Netherlands. It is one of five new teams that were announced today, representing a total investment of $125m, to tackle some of the toughest challenges in cancer research.