Integration of CSF Proteogenomics in the Diagnosis and Management of Diffuse Gliomas

Project Summary Diffuse gliomas are the most common primary malignant adult brain tumor. Lower grade gliomas (Grade 2) inevitably become malignant, and prognosis is fatal. Key clinical challenges include: 1) significant intra- and inter-tumor molecular heterogeneity; and 2) limitations in current imaging techniques for distinguishing true tumor recurrence (TTR), pseudo-progression (PP), or radiation necrosis (RN) following surgical resection and adjuvant therapy. Direct tissue sampling is invasive and may not capture the entire molecular landscape of the tumor. Non-invasive imaging (e.g. MRI) fails to detect small recurrences, is limited in resolving differential diagnoses during imaging follow-up, and does not capture tumor molecular evolution. Recently, liquid biopsy of proximal fluids has gained popularity for systemic tumors. In gliomas, both blood and cerebrospinal fluid (CSF) are viable sources of proximal fluids. While acquisition of blood is less invasive, the CSF is physiologically expected to be a superior enriched reservoir for tumor-specific biomarkers. Thus, CSF can be leveraged for more direct target identification, which can be subsequently extended to blood assays. Most molecular analyses of glioblastoma have focused on the genome, epigenome and transcriptome, but proteomic analyses are also required to understand tumor functional phenotypes like response to therapy. We have conducted shotgun proteomics on as little as 30 µL of CSF from patients with glioblastoma and primary central nervous system lymphoma, detecting a substantial concentration of distinct and identifying differentially enriched pathways for each tumor class. We hypothesize that integration of CSF-based proteogenomic analyses can improve our diagnosis of DGs, refine our management during tumor surveillance and offer insight into potential novel drug targets. We propose to conduct a multi-center study focused on CSF, with the largest prospective cohort established to-date. This project will validate our preliminary findings while establishing novel diagnostic proteomic signatures based on CSF. We will also establish the first ever CSF glycoproteome in DGs, which will be of great value for drug target development. Success in this proposal will have a significant impact on diffuse glioma management through improving preoperative diagnostic precision and subsequent surgical planning, enable early intervention for TTR while eliminating potentially unnecessary surgery for RN, and establish a template for drug target discovery in future studies.