Seoul: A joint research team from the Korea Advanced Institute of Science and Technology (KAIST) and Neogenlogic, a domestic biotechnology company, has successfully developed an innovative artificial intelligence (AI) model aimed at creating a new type of personalized cancer vaccine.
According to Yonhap News Agency, the AI platform is designed to identify neoantigens unique to each cancer patient, which can help train the patient's immune system to prevent the recurrence of the illness. The team, led by Professor Choi Jung-kyoon of KAIST, explained that neoantigens are mutation-derived protein fragments specific to a patient's tumor and serve as the "fingerprints" used by vaccines to train the immune system.
Current cancer vaccines primarily focus on activating cytotoxic T cells for immediate attack, but emerging clinical evidence suggests that B cell-mediated immune memory is crucial for long-term antitumor responses and preventing recurrence. These findings were published in the December 3 edition of Science Advances, a peer-reviewed scientific journal.
The research team has developed an AI model that predicts which neoantigens will trigger a "robust B cell response" by analyzing the structural interaction patterns between mutant peptides and B cell receptors (BCRs). This study introduces the world's first AI framework capable of predicting B cell immunogenicity alongside T cell responses for personalized cancer vaccine design.
Professor Choi emphasized the study's significance in providing empirical evidence on the importance of B cells in cancer vaccine development. While the academic community recognized the importance of studying B cells, there were previously no tools available to verify this concept, Choi told Yonhap News Agency.
Neogenlogic, in a separate release, confirmed that the new technology has been validated against large-scale genomic datasets and clinical trial data from global vaccine leaders. The AI framework has been fully integrated into their proprietary discovery engine, DeepNeo.
Professor Choi also mentioned that his team is preparing an investigational new drug (IND) submission with the U.S. Food and Drug Administration (FDA) and plans to enter clinical trials in 2027. He stated that the collaboration with Neogenlogic is translating this academic breakthrough into a clinical-grade platform, enhancing the scientific rigor of neoantigen selection and moving from theoretical prediction to systematic clinical application.