SURG-82. Stereotactic biopsy in IDH-wild type glioblastoma (GBM) - a survival analysis and interrogation of prognostic variables

Abstract Maximal safe cytoreduction remains the optimal surgical strategy in GBM. Despite this, a significant proportion of patients are not amenable to resection. Stereotactic biopsy remains essential to provide an integrated diagnosis and guide further treatment. There is a paucity of recent literature examining survival variables of this cohort post the WHO 2021 classification. Single centre retrospective study of stereotactic biopsies in IDHwt-GBM patients between January 2014-December 2024. Patient demographic, performance status (PS), radiological, histological and outcome data were collected. MGMT-hypermethylated tumours were stratified into 5-10% vs >10% cohorts. Univariate and Kaplan-Meier survival analyses were performed. Basal ganglia, thalamus, brainstem and corpus callosum defined midline structures. 394 cases were identified. Median age was 66 with a male:female ratio=1.77:1, 285 (72.5%) cases were PS0-1. Multifocal tumours accounted for 65.9% (259) of cases, of which 99 were multicentric. 184(46.7%) patients had midline involvement. Median survival(mOS) with further treatment and best supportive care was 7.5 months (95% CI: 6.47-8.23) and 2.1 months (95% CI: 1.60-2.53). Kaplan-Meier and univariate survival analysis demonstrated significant improved survival distributions for age less than 65 (Logrank-Mantel-Cox, p=0.0004) and pre-biopsy PS 0-1 (Logrank-mantel-cox, p<0.0001). These were also predictive of undergoing further treatment. Involvement of midline structures had reduced mOS compared to non-midline lesions (Logrank Mantel-Cox, 0.0005). MGMT hypermethylation was associated with improved mOS in patient undergoing adjuvant treatment (8.57 vs 7.63months, Logrank Mantel-Cox p=0.0145). Whilst degree of MGMT-hypermethylation did not demonstrate significance in mOS for the adjuvant treatment cohort (8.2 vs 8.6 months, p=0.1828), >10% MGMT-hypermethylation cohort demonstrated improved 2-year OS (17.5% vs 6.3%). Despite recent advances in therapeutics, inoperable GBM remains a challenging entity to treat. Here we quantify pertinent peri-operative factors which may be utilised to council patients and manage prognostic expectations. Poorer prognostic cohorts may be better served with enrolment in clinical trials or best supportive care.