INNV-05. DEVELOPMENT OF A LABEL-FREE PLASMONIC OPTICAL BIOSENSOR FOR INTRAOPERATIVE DETECTION OF TUMOR TISSUE IN GLIOBLASTOMA

Abstract Safe maximal resection of patients with glioblastoma remains a surgical challenge due to the invasive nature of the tumor and the difficulty in distinguishing tumor margins during resection. Several tools are currently used for this purpose, such as, confocal laser endomicroscopy and fluorescence-guided surgery. In this context, biosensors based on plasmonic technology using the extraordinary optical transmission (EOT) phenomenon can help to distinguish between tumor and surrounding tissue based on their optical properties. We developed a biosensor utilizing EOT through a nanostructured gold matrix with 220 nm diameter holes and conducted a prospective study in 35 GBM patients. Paired samples of tumor and peritumoral tissue were collected during surgery. Each sample’s optical imprint was analyzed, revealing significant differences in refractive indices (RI). Tumor tissue showed higher RI values (1.350, IQR 1.344–1.363) compared to peritumoral tissue (1.341, IQR 1.339–1.349) with a p-value of 0.0047. The receiver operating characteristic (ROC) curve indicated the biosensor’s ability to differentiate the tissues (AUC = 0.8779, p < 0.0001). An optimal RI cut-off point of 0.003 was determined, with the biosensor achieving 81% sensitivity and 80% specificity. Our results indicate that this plasmonic technology-based biosensor could have a significant impact on the surgical treatment of glioblastoma by providing a label-free tool for tumor margin discrimination and improving the accuracy of glioblastoma surgery. Future research will focus on optimizing the nanoarchitecture of the biosensor to further improve its sensitivity and specificity, as well as exploring its applicability to other cancer types by analyzing the biomarkers responsible for the different RIs of the footprints.