The Chinese Academy of Sciences has developed a bar code that can detect the clotting risk of plasma EVs, which can be used to predict complications in cancer patients.

CTOnews.com, November 14, according to the Institute of physical and Chemical Technology of the Chinese Academy of Sciences, venous thromboembolism (VTE) is a fatal complication of cancer patients. Due to the lack of accurate and effective VTE risk assessment methods, the missed diagnosis of VTE often leads to the delay of medical intervention and even the death of patients. Studies have shown that exocrine bodies (EVs) initiate exogenous coagulation through tissue factors carried on the surface, resulting in a hypercoagulable state of blood and even promoting the formation of VTE. Some studies have preliminarily discussed the procoagulant ability of plasma EVs, but how to predict the risk of clinical VTE is still a challenge.

Recently, Wang Shutao and Meng Jingxin, researchers at the Institute of physical and Chemical Technology of the Chinese Academy of Sciences, together with Zhang Shutian, a professor at Beijing Friendship Hospital affiliated to Capital Medical University, developed a bar code (PEVB) that can detect the risk of plasma EVs clotting, which provides an effective technique for predicting VTE risk in clinical cancer patients. This technology has three characteristics:

Rapid capture of EVs of plasma samples using titanium dioxide nanoflowers

In situ detection of EVs procoagulant ability by visual EVs-mediated clotting reaction (the number of positive bands)

Machine learning-assisted clinical data analysis for accurate VTE risk assessment.

The researchers used PEVB to test 167patients from eight cancers and machine learning to integrate band scores with D-dimer screening and other routine clinical information. This technology realizes the accurate prediction of VTE by screening appropriate machine learning models. The AUC is 0.993, the detection specificity is 97.1%, the sensitivity is 96.8%, and the accuracy is 97.0%. It is significantly better than the commonly used clinical VTE risk prediction tools (for example, D-dimer screening, specificity 44.8%, sensitivity 90.3%, accuracy 53.2%).

Therefore, based on bar code and detection technology, VTE risk can be evaluated intuitively, quickly and conveniently, showing potential in accurate prediction of VTE risk in cancer patients.

▲ 's schematic design of PEVB testing for accurate assessment of VTE risk related research results are published in the American Chemical Society-Nano (ACS Nano) under the title Machine-Learning-Assisted Procoagulant Extracellular Vesicle Barcode Assay toward High-Performance Evaluation of Thrombosis-Induced Death Risk in Cancer Patients. The research work is supported by the National Natural Science Foundation of China and the International Partnership Program of the Chinese Academy of Sciences.

CTOnews.com with paper link: https://pubs.acs.org/doi/epdf/10.1021/acsnano.3c04615