BIOM-59. NOVEL BIOMARKER DISCOVERY AND DIAGNOSTICS FOR INTRA-AXIAL BRAIN TUMORS, USING CSF PROTEOMICS

Abstract BACKGROUND Invasive brain sampling is typically necessary for reliable diagnosis and prognostication of intra-axial tumors but carries risk morbidity. Liquid biopsy proximal fluids may mitigate this risk. Through direct contact with the tumor microenvironment and, as an ultra-filtrate plasma, cerebrospinal fluid be ideal matrix. Reflecting phenotype, proteomic analyses are critical. Here we identified diagnostic CSF signatures putatively novel biomarkers glioblastoma (GBM), metastases (BM), central nervous system lymphoma (CNSL). METHODS samples were retrospectively retrieved from Penn State Neuroscience Biorepository profiled using shotgun proteomics MStern approach. Proteomic machine learning classifiers survival analyses. RESULTS With little 30 µL CSF, 755 unique proteins recovered across 73 (22 GBM, 17 BM, 14 CNSL, 20 NPH). Proteomic-based malignancy area under receiver operating characteristic (AUROC) 0.94 distinguished between entities AUROC ≥0.95. More clinically relevant triplex classifiers, comprised just 3 proteins, 0.75-0.89. Novel identified, including GAP43, TFF3 CACNA2D2, characterized single-cell RNA sequencing. DISCUSSION Reliable classification malignancies low volumes feasible, allowing longitudinal surveillance. Based on emerging evidence, upfront implantation reservoirs in patients warrants consideration.