“A stride toward early diagnosis of lung cancer based on nanotechnology”

AI analyzes exosomes in the blood to diagnose lung cancer within 30 minutes from a single drop of blood.

This technology enables the diagnosis of even Stage 1 lung cancer, which has proved difficult to detect early, improving the survival rate.

The results from this study were published in ACS Nano, an authoritative international journal in the field of nanotechnology.

 

A novel technique allowing the diagnosis of lung cancer within 30 minutes from a single drop of blood has been developed by a Korean research team. The results of this study were published in the May issue of ACS Nano (IF: 14.5), an authoritative international journal in the field of nanotechnology.

 

The joint research team led by Professor Choi Yeonho of the School of Biomedical Engineering, College of Health Science, and Professor Kim Hyun-koo of the Department of Thoracic & Cardiovascular Surgery, Korea University Guro Hospital, analyzed the exosomes in blood, which are a biomarker for cancer diagnosis, by applying nanotechnology and AI. They were thereby able to successfully distinguish normal cells from lung cancer cells at an accuracy of 95%. This technology may be used to diagnose Stage 1 lung cancer, which is usually difficult to detect early, within 30 minutes using just one drop of blood. Therefore, the technology is expected to make great contributions to increasing the survival rate of lung cancer patients through early diagnosis.

 

▲ (From left) Professor Choi Yeonho (corresponding author), Professor Kim Hyun-koo (corresponding author), 

and Shin Hyun-ku (first author, student in the integrated Ph. D program).

 

A technology to diagnose lung cancer from blood samples has long existed. However, it has limitations in practical use, because it allows for the diagnosis of only about 50% of patients. In contrast, the new technology developed by the research team has a high accuracy. It can diagnose lung cancer at a maximum rate of 84%, and even predict the stage of lung cancer.

 

Lung cancer, often found in Stage 3 or higher where treatment is difficult, is one of the major cancers featuring a very high mortality. It is known that the diagnosis of lung cancer in its earlier stages (Stages 1 and 2) can significantly increase the survival rate. Therefore, many studies are actively conducted to develop a technology for early diagnosis. Exosomes, found in the blood, are drawing attention as biomarkers for cancer diagnosis because they contain information about tumor cells that is usually kept inaccessible deep inside the body.

 

 

The research team isolated exosomes from 20 healthy subjects and 43 patients with Stage 1 or 2 non-small cell lung cancer, and detected about 2,000 Raman spectroscopic signals by applying a nanotechnology based on surface-enhanced Raman spectroscopy. A deep learning AI model was trained using the detected signals, and then the model was able to successfully distinguish the normal cells and lung cancer cells at an accuracy of 95%. In addition, the exosomes from cancer patients were compared with the exosomes originating from lung cancer cells to classify them at a sensitivity of 84% and a specificity of 85%.

 

 

 

Professor Choi Yeonho stated, “The results demonstrated the usefulness of the diagnostic method based on the exosome analysis technique and the deep learning AI for the early diagnosis of lung cancer.” Professor Choi also added, “Our results are highly meaningful because this technology can not only be used to diagnose Stage 1 lung cancer but also determine the particular stage of lung cancer, as the analytical values and accuracy significantly increase at later stages.

 

Professor Kim Hyun-koo of the Department of Thoracic & Cardiovascular Surgery, Korea University Guro Hospital, said, “By using this technology, we can perform a blood test in advance to determine which patients may have lung cancer before CT imaging, which is associated with concerns of radiation exposure. CT imaging can therefore be reserved for only those who really need it.” Professor Kim also explained the significance of the study, “Our results are meaningful in that this technology can be used to identify even those patients with Stage 1 lung cancer at a relatively high accuracy. We hope that our technology can contribute to the early diagnosis of lung cancer, thereby increasing the survival rate of patients.”

 

The results of the study were published in the article entitled “Early-Stage Lung Cancer Diagnosis by Deep Learning-Based Spectroscopic Analysis of Circulating Exosomes.” The study was supported by the Basic Science Research Program (strategic project) funded by the National Research Foundation of Korea (NRF), and the Korea Health Technology R&D Project funded by the Korea Health Industry Development Institute (KHIDI).

 

This study was conducted by applying the technologies from Exopert Co. Ltd., which is a medical technology holding company of Korea University, managed by Professor Choi Yeonho, the CEO of the company. Exopert Co. Ltd. is a company that develops cancer diagnosis devices by using exosomes and a separation kit that can rapidly separate exosomes of high purity from the blood of cancer patients. To improve the reliability of the developed technology and commercialize it, the company is currently planning a multicenter study with five participating hospitals, including Korea University Guro Hospital, and 400 subjects including both healthy individuals and patients.