GENE-39. UNANTICIPATED GERMLINE CANCER SUSCEPTIBILITY MUTATIONS IDENTIFIED DURING ROUTINE NEXT GENERATION SEQUENCING OF PRIMARY CNS NEOPLASMS

Abstract

INTRODUCTION

Tumor genomic profiling is routinely performed on primary brain tumors to identify somatic mutations. Such testing may also identify clinically significant germline mutations. Several syndromes (neurofibromatosis, Li Fraumeni, Von Hippel Lindau and Turcot) are well known causes of primary CNS neoplasms but there is limited data on other germline cancer susceptibility mutations in CNS tumor patients. After a somatic mutation was unexpectedly found to be germline, we interrogated our database of somatic CNS tumor mutations.

METHODS

This IRB approved retrospective study reviewed patients at the Texas Oncology-Austin Brain Tumor Center from May 2013 through April 2017. Tumors underwent genomic profiling using a commercially available test which evaluates 315 genes using next generation sequencing. Patients' whose tumors carried a mutation which was concerning for a germline variant were tested further.

RESULTS

Tumor genomic profiling was performed on 291 CNS tumor patients. Five patients were found to carry unexpected germline variants (1.7%). Two patients had glioblastoma: 1 BRCA2 pathogenic mutation, 1 MLH1 variant of uncertain significance; 2 anaplastic astrocytomas: 1 BRCA2 pathogenic mutation, 1 with pathogenic APC and ATM mutations; 1 atypical meningioma: 1 BRCA2 pathogenic mutation. Median age was 42.2 (range 33–53). As a result of the incidental findings, medical management was altered and testing of numerous family members was recommended. Complete tumor genomic profiles, pedigrees, and additional pending germline testing will be presented.

CONCLUSIONS

In this series of primary brain tumor patients whose tumors were sequenced, unexpected germline variants were identified in the genes: BRCA2, MLH1, APC and ATM for an incidence of at least 1.7%. These results had life-saving implications for these patients and their families. This data is reasonably extrapolated to the brain tumor population at large and warrants further scrutiny of systematic approaches for identifying patients as candidates for germline testing based on tumor genomic profiling.