Seoul, Jan 2 (NationPress) A collaborative research initiative involving the Korea Advanced Institute of Science and Technology (KAIST) and the biotech company Neogenlogic has unveiled an innovative artificial intelligence (AI) model designed to create a novel type of personalized cancer vaccine, announced the team on Friday.

This advanced AI platform specializes in identifying neoantigens that are unique to each individual cancer patient, effectively training the patient’s immune system to prevent the recurrence of the disease, according to reports from Yonhap news agency.

Neoantigens—protein fragments derived from mutations specific to a patient’s tumor—serve as the “fingerprints” utilized by vaccines to educate the immune system, the research team, led by Professor Choi Jung-kyoon of KAIST, stated in a press release.

The statement further elaborated, “Current vaccines primarily focus on stimulating cytotoxic T cells for immediate attack, but emerging clinical studies suggest that B cell-mediated immune memory is crucial for long-lasting, effective antitumor responses and minimizing recurrence.”

These findings were published in the December 3 issue of Science Advances, a peer-reviewed journal.

The research team indicated that their AI model can predict which neoantigens will elicit a strong B cell response by analyzing the structural interactions between mutant peptides and B cell receptors (BCRs).

“This study presents the first AI framework globally capable of predicting both B cell immunogenicity and T cell responses for the creation of customized cancer vaccines,” they claimed.

Professor Choi emphasized the study's importance, noting it provides concrete evidence regarding the role of B cells in cancer vaccine development.

“While the academic field recognized the need to investigate B cells for developing cancer vaccines, there were no existing tools to substantiate this concept,” Choi told Yonhap News Agency.

Neogenlogic, in a separate announcement, stated that the new technology has been validated against extensive genomic datasets and clinical trial data from leading global vaccine developers, asserting that the AI framework is fully integrated into its proprietary discovery engine, DeepNeo.

Choi revealed that his team is preparing a submission for an investigational new drug (IND) to the U.S. Food and Drug Administration (FDA), with intentions to commence clinical trials in 2027.

“In partnership with Neogenlogic, we are advancing this academic discovery into a clinically applicable platform,” he added. “Our proprietary AI enhances the scientific rigor in selecting neoantigens, transitioning from theoretical predictions to systematic clinical applications.”