Multiplexed Microenvironment-Responsive Probes Enable Rapid Glioblastoma Cell Line Analysis

Accurate distinguishing the phenotype of glioblastoma (GBM) cell lines remains challenging in clinical diagnostics, particularly for rapid intraoperative assessment, due to tumor heterogeneity and the limitations of conventional methods in capturing functional cellular variations. Since the tumor microenvironment (TME) plays a pivotal role in tumor progression, it is essential for accurate cancer characterization. Herein, we developed a multiplexed optical sensing platform that exploited the distinct metabolic and physicochemical heterogeneities of the TME for cell discrimination. The platform integrated five microenvironment-responsive fluorescent probes, namely, HBTPB, CTCYS, BDPI, KLVIS, and BIDOH, designed to simultaneously monitor hydrogen peroxide, cysteine, peroxynitrite, viscosity, and pH, respectively. These parameters reflected the status of the cells. The responsive probes can sense the difference of the cells by sensing the data. By integrating this responsive multiparameter with a ResNet-based deep learning model, we achieved highly accurate identification of six cell lines comprising four phenotypically diverse GBM cell lines, normal human astrocytes, and a central nervous system tumor cell line.