'World's Largest' Cancer Map Constructed... KAIST "Contributing to Immunotherapy and Prognosis Prediction"

KAIST has established the world's largest cancer database. This database is expected to contribute to analyzing the impact of cellular states on the treatment and prognosis of cancer patients.

On the 22nd, KAIST announced that a joint research team involving Professor Jong Eun Park from the Graduate School of Medical Science and Engineering and Professor Jung Kyun Choi from the Department of Bio and Brain Engineering (with Dr. Junho Kang and Dr. Junhyeong Lee as first authors) constructed the world's largest single-cell and spatial transcriptome database of cancer tissues. Based on this, they reported important cellular ecosystem types for predicting the prognosis of immunotherapy in collaboration with Professor Sehoon Lee’s research team at Samsung Seoul Hospital.

Single-cell and spatial transcriptomes are data that analyze the expression patterns of all genes at the individual cell level or within the three-dimensional tissue structure.

Schematic diagram of the data processing workflow for detailed analysis of single-cell cancer transcriptome data. Provided by KAIST

Accumulating digital cancer information goes beyond data production; establishing methods for data collection and management and operating big data can become significant competitive advantages. Strategically, it is considered a crucial task to collect and integrate both domestically produced data linked to precise clinical information and large-scale international data that can foster understanding of diversity.

In particular, cancer has the characteristic of evolving autonomously within the human body, making it essential to understand the cellular heterogeneity and intercellular interactions that constitute the cellular ecosystem within cancer tissues.

Single-cell and spatial transcriptomes quantitatively measure and represent the three-dimensional arrangement and interactions of cells composing the microenvironment. This is meaningful because it expands the concept of microenvironmental heterogeneity to the ecosystem level, allowing it to be stored and analyzed in the form of digital information.

The KAIST research team constructed a “pan-cancer single-cell atlas” that comprehensively maps cellular data for all cancers by associating single-cell transcriptome data from 1,000 tissue samples obtained from cancer patients and normal tissue samples from over 500 individuals with more than 30 types of cancers to elucidate cancer cell ecosystem types at the pan-cancer level.

First, the research team, including internal medicine specialists, directly collected the data and defined about 100 cellular states constituting cancer tissues by reprocessing metadata and classifying cancer types.

Next, based on their occurrence frequencies, they classified the states of tissues by cancer type and analyzed the impact of each cellular state on treatment and prognosis in cancer patients using large cohort data such as the United States’ public cancer patient database (TCGA).

In particular, they confirmed that the interferon-associated ecosystem, including components of tertiary lymphoid structures, was effective in predicting the response to immune checkpoint inhibitor therapy in various cancers treated with immune checkpoint inhibitors, such as the lung cancer cohort from Samsung Seoul Hospital’s research team.

Tertiary lymphoid structures are formations similar to lymph nodes but do not form in healthy tissues; they are organized immune cell structures formed in areas with chronic inflammation, infection, or cancer.

Professor Jong Eun Park said, “The joint research team succeeded in building the world’s largest cancer tissue database through this study,” adding, “This is expected to have a significant impact on predicting the prognosis of immunotherapy.”

He further stated, “Above all, these research results will be effective in selecting treatment candidates for immune checkpoint inhibitors, which show very good therapeutic responses in a small number of patients but cause immune-related side effects in some cases.”

Meanwhile, the joint research team conducted the study with support from the Korea Research Foundation’s Next-Generation Bio Promising Universal Technology Research Support Project and Excellent New Researcher Project, the Korea Health Industry Development Institute’s Research-Centered Hospital Development Project, the Convergent Physician-Scientist Training Project, and the POSCO Science Fellowship.

The research results were also published in the international academic journal Nature Communications on May 14.

This content was produced with the assistance of AI translation services.