Neuro-Oncology Spectral Fingerprint Biorepository: A Novel Approach Leveraging a Surgical and Clinical Detailed Repository in a Diverse Patient Population

INTRODUCTION: Despite advances in the understanding of brain cancer, surgery and extent of resection (EOR) continue to be the main predictors of patient outcome. However, currently available public biorepositories do not include extent of resection for individual patients raising important questions regarding confounding effects for some molecular markers. Further, emerging data has supported important novel biomarkers in tumor’s lipidome and optical attenuation properties which could serve as new therapeutic opportunities. METHODS: All neuro-oncology surgical cases at our institution were eligible for collection since 2016. Blood was collected from every patient pre-operatively. Cerebrospinal fluid was collected whenever surgically safe. Genomic and transcriptomic analysis was performed. Intra-operative spectral and optical attenuation data was obtained through desorption electrospray ionization mass spectrometry (DESI-MS) over 0 to 1000 m/z and optical coherence tomography, respectively, for biomarker discovery. RESULTS: A total of 912 unique patient collections have been performed including 286 gliomas and 205 other rare brain tumors. Demographic, clinical, imaging, surgical and survival data was curated for each individual patient. Extent of resection (EOR) was determined for each patient, with an average of 78.6% for chordomas, 85.9% in glioblastoma, 88.3% in other high-grade gliomas and 72% in low-grade gliomas with predictive of progression-free survival (p < 0.01) and overall survival (p < 0.01). Comprehensive sequencing has been completed for 131 gliomas. Of those, lipid and optical profiling has been completed for multiple tumor samples (330) from a total of 79 patients. CONCLUSIONS: We have curated a large biorepository of neuro-oncology tumors with detailed patient information including for the first time to our knowledge detailed surgical, imaging, spectral and optical information. Efforts are ongoing at large-scale biomarker discovery leveraging the unique spectral and optical fingerprint of each tumor to their genomic/transcriptomic profile.