A South Korean research team from the Korea Advanced Institute of Science and Technology (KAIST), in collaboration with biotechnology firm Neogen Logic, has developed a new artificial intelligence (AI) model to design personalized cancer vaccines that enhance long-term immune memory against cancer cells.
The study, published in Science Advances on December 3, represents the first AI system capable of predicting both T-cell and B-cell responses, enabling vaccines that not only attack tumors but also remember them to prevent recurrence.
Traditional cancer vaccines have primarily focused on stimulating T-cells, which directly attack cancer cells. However, T-cells alone have limited ability to maintain long-term immunity, making it challenging to prevent relapse. The KAIST team focused on B-cells, which recognize cancer-specific neoantigens and produce antibodies, effectively creating an immune “memory” of the tumor.
The AI model was trained on the structural interactions between neoantigens derived from cancer cell DNA and B-cell receptors (BCRs). This allows the system to predict which neoantigens most effectively stimulate B-cell responses, enabling precise selection of targets for personalized vaccines. The platform has been validated through animal studies and existing clinical datasets, demonstrating enhanced antitumor activity compared with conventional approaches.
The researchers plan to submit an Investigational New Drug (IND) application to the U.S. Food and Drug Administration and aim to begin human clinical trials by 2027.
This development marks a significant advancement in precision immunotherapy, combining AI-driven neoantigen prediction with long-lasting immune protection to potentially improve outcomes for cancer patients.
