P18.50.A TUMOR VOLUMES AND FEASIBILITY OF EXTENSIVE RESECTION IN NON-CONTRAST ENHANCING GLIOBLASTOMAS THAT LACK HIGH GRADE HISTOLOGICAL FEATURES: A SINGLE CENTER STUDY

Abstract BACKGROUND The clinical course and management of non-contrast enhancing glioblastomas that also lack the classic high grade histological features (non-CE GBMs) is unclear. Data on survival, clinical management including surgical therapies are sparse and conflicting. In this study we therefore investigated tumor volumes, surgical and postsurgical therapies and survival outcome of a single center cohort from the Netherlands. MATERIAL AND METHODS We identified adult cases of non-CE GBM in the Brain Tumor Center at Erasmus MC Cancer Institute in Rotterdam, the Netherlands from 2003-2024. Cases with ring-like contrast enhancement and/or central necrosis at time of diagnosis were excluded. Patients were classified as glioblastoma using a glioma-tailored NGS panel. Pre-and postoperative tumor volumes were segmented manually using open source software ITK-SNAP. Pre- and postoperative volumes were set equal when only a biopsy was performed. Clinical information and postoperative therapies were retrospectively collected from patient records. Overall survival was calculated from time of diagnostic MR scan. RESULTS We identified 77 patients with available imaging data. Median age of patients was 59 years. The presenting symptom was an epileptic seizure in 66.2% of patients. Tumor localization in 60 out of 77 patients involved mainly the temporal lobe. As initial surgical strategy 59 patients underwent a biopsy (stereotactic or open), and only 18 patients a resection (of which 5 were awake). In most cases a clinically relevant resection was considered not feasible. Of the 18 resected patients, 4 patients underwent a second resection at progression. Of the 59 biopsied patients, 5 underwent a subsequent resection, and 5 a second biopsy as the first biopsy was inconclusive. Patients that underwent a resection had higher KPS scores (≥90 in 95.7% vs 55.6%, P= 0.003), and lower rates of gliomatosis cerebri (8.7% vs 53.7%, P