EPID-03. PATHOGENIC GERMLINE MUTATIONS IN GLIOMA PATIENTS

Abstract

Abstract BACKGROUND No consensus germline testing guidelines currently exist for glioma patients. Hence such testing is not routinely performed. The prevalence and type of germline pathogenic variants in these brain tumors remains unknown. DESIGN/ METHODS A retrospective review of patients treated at Baylor College of Medicine with paired tumor/normal sequencing using the Tempus xT tumor/normal matched approach from August 2018- April 2022 was performed. Corresponding clinical data was collected for these patients. RESULTS We identified 152 glioma patients of which 15 (9.8%) had pathogenic germline variants. Pathogenic germline variants were seen in 11/84 (13.1%) of Glioblastoma, IDH wildtype, 3/42 (7.1%) of Astrocytoma, IDH mutant and 1/26 (3.8%) of Oligodendroglioma, IDH mutant and 1p/19q co-deleted patients. Pathogenic variants in BRCA2, MUTYH and CHEK2 were seen more commonly (3/15 or 20% each). BRCA1 variants were seen in 2/15 (13%) patients, with variants in NF1, ATM, MSH2, and MSH3 occurring in one patient (7%) each. Second hit somatic variants were seen in 3/15 patients (20%). A second somatic hit was seen in NF1, MUTYH and MSH2 in one patient (7%) each. Referral to genetics was performed in 6/15 (40%) patients with pathogenic germline variants. Median overall survival was 1.6 years for glioblastoma, IDH wildtype patients with a pathogenic germline variant compared to 1.79 years for glioblastoma, IDH wildtype patients without it (p = 0.67). CONCLUSIONS Although not routinely performed in glioma patients, pathogenic germline variants occurred in ~10% of our patients. Only 40% of these patients were referred to genetics. These findings suggest a possible overlooked opportunity for determination of hereditary cancer syndromes with impact on surveillance as well as potential broader treatment options. Further research to confirm the occurrence and types of pathogenic germline variants in patients with IDH wildtype compared to IDH mutant tumors is necessary.