Nanozyme-Enabled Multimodal Sensing: Visual and Rapid Profiling of Extracellular Vesicles

CD20, a transmembrane protein on the surface of lymphoma extracellular vesicles (EVs), is highly expressed and serves as an effective marker for monitoring lymphoma subtypes and evaluating the efficacy of antibody therapy. Therefore, there is an urgent need for methods to effectively enrich and accurately detect CD20 on EVs. To address these challenges, immunoaffinity magnetic bead adsorption and nanozyme-enabled multimodal sensing have been identified as effective strategies. This study demonstrates that phosphate groups in the phospholipid bilayer of EVs complex with Ti⁴⁺ on Fe₃O₄@TiO₂ magnetic beads, enabling their separation from complex samples by using an external magnetic field. The signal label CuCo-ZIF/Pt with good oxidase activity can oxidize <i>o</i>-phenylenediamine (OPD) into 2,3-diaminophenazine (DAP) with colorimetric and fluorescence signals, and the intensity of the signal is proportional to the concentration of CD20. Utilizing the Residual Network 18 with Data Augmentation (Resnet18-DA) neural network model, the artificial intelligence (AI) visual perception platform is capable of rapidly distinguishing the concentration of CD20 through color response, achieving an accuracy rate of nearly 100%. This multimodal sensing platform not only enhances the accuracy and convenience of detection but also has the potential to provide new strategies for evaluating the effectiveness of personalized tumor treatment.