P11.30.A LONG-TERM SURVIVING GLIOBLASTOMA. CLINICAL, MOLECULAR AND SURVIVAL ANALYSES

Abstract BACKGROUND Glioblastoma (GB) has a poor prognosis with median overall survival of 12-14 months. According to literature, patients surviving more than two years has progressively increased since 2005 and represent approximately 18% of this population, being considered long-term survivors (LTS). The aim of this study is to retrospectively analyze GB LTS treated in our Neurooncology Unit. MATERIAL AND METHODS From December 2009 to March 2021, 49 GB LTS patients were identified in our oncology data base, representing 21,7% of GB patients electronically registered during the same period. Patients were diagnosed and treated for GB according to the current WHO CNS classification at the time of diagnosis. Clinical and molecular characteristics as well as treatment and survival data were analyzed. Classic prognostic factors were analyzed including contact with the subventricular zone (SVZ) which has been identified as prognostic factor in some previous studies. RESULTS Median age at diagnosis was 57 years (range: 30-75 years). Median overall survival (OS) and median time to first progression were 43.6 months and 20.7 months respectively, with 9 patients alive and 6 patients without recurrence at the time of analyses. Rate of patients surviving more than 3, 4, 5 and 7 years were 55,1%, 20,4%, 12,24% and 6,12% respectively. Patients received a median of 2 rescue treatments after first relapse (range: 0-6 lines), including systemic therapies, surgery and re-irradiation. Rates for gross total resection, subtotal resection and biopsy were 32,6%, 47% and 20,4% respectively. pMGMT methylated patients and lesions without SVZ contact rates were 73,5% and 42,8% respectively. IDH analyses was available for 61.22% of our GB LTS patients (90% were IDH wild type and 10% mutant IDH, which according to WHO CNS 2021 classification would actually be defined as Astrocytoma grade 4). No statistical differences for OS or time to first progression were found according to sex, age (70%) and contact or not with the SVZ. CONCLUSION Classical prognostic factors did not show statistical differences for OS and time to first recurrence in our GB LTS sample, which could be explained by the small number of patients in the sample and the absence of statistically comparable differentiated groups for some of the variables. Larger and prospective analyses of GB LTS are needed. Molecular analyse