KAIST Develops AI Platform for B Cell-Reactive Cancer Vaccines, Plans 2027 Clinical Trials
核心洞察
KAIST (搜索) researchers developed an AI model that predicts both T cell (搜索) and B cell (搜索) reactivity to neoantigens (搜索), overcoming limitations of current cancer (搜索) vaccine approaches that focus primarily on T cell responses.
The AI platform analyzed over 437,000 peptides and 370 million B cell (搜索) receptor sequences, demonstrating enhanced antitumor immune responses in preclinical studies and clinical trial analyses.
KAIST (搜索) partnered with Neogenlogic (搜索) to commercialize the technology, with plans to submit an FDA IND application and enter clinical trials by 2027.
Researchers at the Korea Advanced Institute of Science and Technology (KAIST (搜索)) have developed an AI-based platform that identifies B cell (搜索)-reactive neoantigens (搜索) for personalized cancer (搜索) vaccines, addressing a critical limitation in current vaccine development approaches that focus primarily on T cell (搜索) responses. The technology, published in Science Advances, has been validated through large-scale genomic data, animal experiments, and clinical trial analyses.
Novel AI Approach Integrates T Cell and B Cell Responses
Professor Jung Kyoon Choi's research team from KAIST (搜索)'s Department of Bio and Brain Engineering, in collaboration with Neogenlogic (搜索) Co., Ltd., developed the AI model to predict neoantigens (搜索) that can activate both T cells and B cells. Unlike conventional approaches that rely primarily on T cell (搜索) reactivity predictions, this technology evaluates whether mutated peptides can form the three-dimensional structures needed for B cell (搜索) recognition and antibody production.
The AI platform was trained on an exceptionally large dataset, analyzing over 437,000 peptides for IgG binding and examining more than 370 million B cell (搜索) receptor (BCR (搜索)) sequences to identify patterns associated with antibody recognition. The model combines predicted T cell (搜索) and B cell responses into a single scoring system to prioritize neoantigens (搜索) for vaccine development.
Preclinical Validation Demonstrates Enhanced Efficacy
In mouse vaccination experiments, neoantigens (搜索) selected for B cell (搜索) reactivity produced superior effects compared to T cell (搜索)-focused vaccines alone. These vaccines triggered expansion of BCR (搜索) clones and led to faster tumor (搜索) regression in several experimental groups compared with vaccines targeting only T cell responses.
The research team validated their approach using single-cell BCR (搜索) sequencing and applied the multiomics framework to large genomic and clinical cohorts, including samples from the Cancer (搜索) Genome Atlas (TCGA) and checkpoint-blockade response datasets involving 2,074 patients. A meta-analysis of 11 personalized vaccine trials involving 1,739 neoantigens (搜索) suggested that incorporating B cell (搜索) neoepitopes may improve vaccination efficacy.
Clinical Translation Timeline and Regulatory Strategy
KAIST (搜索) has partnered with Neogenlogic (搜索), a Seoul-based biotech company, to advance the technology toward clinical development. Professor Jung Kyoon Choi, who also serves as CEO of Research and Development at Neogenlogic, stated, "We are conducting pre-clinical development of a personalized cancer vaccine platform and are preparing to submit an IND application to the FDA with the goal of entering first-in-human clinical trials in 2027."
The research addresses a gap identified by Johns Hopkins University researchers Mark Yarchoan and Elizabeth Jaffee, who noted in Nature Reviews Cancer (搜索) that "despite accumulating evidence regarding the role of B cells in tumor (搜索) immunity, most cancer vaccine clinical trials still focus only on T cell (搜索) responses."
Addressing Current Vaccine Development Limitations
Current neoantigen discovery pipelines focus on identifying peptides that bind to major histocompatibility complex (MHC (搜索)) molecules and activate T cell (搜索) receptors. However, B cell (搜索) epitopes often depend on three-dimensional protein structure rather than linear sequences, making them difficult to predict using conventional sequence-based methods.
The KAIST (搜索) platform overcomes this limitation by modeling how mutant proteins interact with B cell (搜索) receptors, evaluating structural binding characteristics to predict B cell reactivity. This approach reflects the understanding that coordinated immune responses involving both antibodies and T cells can produce stronger and more durable antitumor effects.
Industry Impact and Future Implications
The technology represents a shift from T cell (搜索)-only targeting to evaluating multiple aspects of immune response in cancer (搜索) vaccine development. Implementation will require robust structural modeling, standardized assays to measure B cell (搜索) activity, and carefully designed clinical trials that track both cellular and antibody responses.
Neoantigens (搜索) are protein fragments derived from cancer (搜索) cell mutations that serve as unique markers distinguishing cancer cells from normal tissue. Companies like Moderna and BioNTech have been advancing neoantigen-based cancer vaccine technology using mRNA platforms, conducting clinical trials alongside global pharmaceutical companies.
The KAIST (搜索) research team's findings suggest that by adding B cell (搜索) reactivity, cancer (搜索) vaccines can move beyond one-time attacks and short-term memory to become long-term immunity that "remembers" cancer, effectively preventing recurrence.