The Daegu Gyeongbuk Institute of Science and Technology (DGIST) announced on the 24th that a research team led by Senior Researcher Yoonhee Lee from the Biomedical Research Division and Senior Researcher Kyu-Kwon Goo from the Intelligent Robot Research Division has developed a technology that can distinguish lung cancer gene mutations based solely on the 'hardness' of very small particles called 'exosomes' derived from cancer cells in blood, by compressing them using an atomic force microscope (AFM). This research is expected to advance into a new liquid biopsy-based lung cancer diagnostic technology, particularly as it enables fast and precise analysis of single exosomes.
Non-small cell lung cancer (NSCLC) accounts for more than 85% of all lung cancer patients and is the most common type of cancer. However, it is often difficult to detect in the early stages due to the lack of specific symptoms, and it is frequently diagnosed when it is already advanced, making treatment challenging. For this reason, non-small cell lung cancer still shows a high mortality rate, and the development of new diagnostic technologies that can facilitate early detection and treatment remains a significant challenge in the medical field. In particular, traditional tissue biopsies impose a substantial burden on patients and have limitations in repeated testing; thus, non-invasive liquid biopsy technology utilizing information from blood has been gaining attention recently.
The research team isolated exosomes from different types of cells (A549 with KRAS mutation, PC9 with EGFR mutation, and PC9/GR with EGFR resistance mutation) that possess distinct characteristics according to gene mutations in non-small cell lung cancer. The team measured the physical properties at the nano level, such as the surface strength and height-to-radius ratio of individual exosomes, with high resolution using an atomic force microscope.
As a result, the exosomes derived from A549 showed significantly higher strength, indicating that the changes in membrane lipids due to KRAS mutations were also reflected in the exosomes. In contrast, exosomes derived from PC9 and PC9/GR exhibited similar properties, confirming their association with the shared genetic background. This revealed that the physical properties of exosomes also differ according to gene mutations in cancer cells.
To classify the nano-mechanical properties of exosomes more precisely, the research team incorporated artificial intelligence (AI) technology. They visualized the height and strength information of exosomes obtained through atomic force microscopy and trained a deep learning-based convolutional neural network (CNN) model to classify the originating cells of the exosomes. As a result, exosomes derived from A549 were distinguished with a very high accuracy of 96%. This indicates the potential for a next-generation liquid biopsy platform that allows for high-precision classification based solely on the physical characteristics of exosomes without fluorescent labeling.
Yoonhee Lee and Kyu-Kwon Goo, senior researchers, noted that 'the findings of this study present a new diagnostic possibility for distinguishing lung cancers with specific gene mutations using a small amount of exosome samples' and that 'they plan to actively promote the practical application of the technology through future clinical sample validation and the integration of a high-speed atomic force microscope platform.'
These research results were published online on the 8th in 'Analytical Chemistry,' a prominent journal in the field of chemistry.
References
Analytical Chemistry (2025), DOI: https://doi.org/10.1021/acs.analchem.5c02009