Cancer vaccines are designed to train the immune system to recognize and destroy tumors. While many vaccine strategies focus on activating “killer” CD8 T cells, another group of immune cells called CD4 helper T cells plays a critical role in sustaining long-term anti-cancer immunity and improving responses to immunotherapy.
To activate CD4 T cells, tumors must display small protein fragments on molecules called MHC class II (MHC-II). However, scientists still do not know which tumor fragments are most effectively displayed across the thousands of MHC-II variants found in different people. As a result, many cancer vaccines rely heavily on prediction rather than direct evidence, limiting their effectiveness.
Dr. Bruno’s project seeks to replace this trial-and-error process with a more precise and data-driven strategy. Using a high-throughput experimental platform, he will directly test large numbers of tumor-derived protein fragments across common MHC-II variants to identify which ones most effectively stimulate CD4 T-cell responses.
Simultaneously, Dr. Bruno will develop advanced artificial intelligence models that learn from these experimental results to improve future predictions. The system will continuously refine itself by combining laboratory testing with machine learning. This work could dramatically improve the design of cancer vaccines by enabling more accurate and personalized targeting of helper T cells, ultimately leading to stronger, longer-lasting anti-tumor immune responses.
Projects and Grants: Next-generation cancer vaccines through AI-guided MHC-II functional genomics
