Identification of SNVs via targeted comprehensive genome profiling in patients with brain cancer.

Abstract

e14047 Background: Advanced brain tumors comprise an aggressive class of malignancies and are characterized by poor patient survival. In general, the 5-yr survival rate for people ≥40 years is about 21%. These tumors also appear to be recalcitrant to available SOC, hence the need for deeper insights to recognize actionable genetic alterations in order to make reliable assessments of diagnosis, prognosis, therapy response, and resistance. Comprehensive targeted Next Generation Sequencing (NGS) has allowed significant advances in precision oncology by pinpointing actionable alterations in oncogenes, tumor suppressor genes, and cellular signaling pathways. Methods: Retrospective NGS-based genome profiling was performed on ccfDNA (circulating cell-free DNA) and genomic DNA (FFPE/fresh tissue) samples from 8 brain tumor patients OncoIndx proprietary gene panel that targets exonic regions of 600 cancer related genes. Illumina compatible libraries were prepared by target hybridization method and sequenced on Nextseq 2000 platform in pair-end fashion. Variant calling was performed using an in-house developed bioinformatics pipeline. The 8 patients comprised 1 each for glioma, anaplastic astrocytoma, pilocytic astrocytoma and parietal SOL (Space Occupying Lesion) and 2 each for GBM and ependymoma. Results: Novel mutations such as GNASA366S, BRAFM590K and RAD54LD592V were observed in glioblastoma. Additionally, novel SNVs such as STK11G215V, CREBBPG1465E, GNA11G208C and SF3B1R549S were observed in ependymoma along with as yet unreported amplification of KIT, KDR and PDGFRA. NOTCH3 double mutation C428S and C455S was identified for the first time in this study in anaplastic astrocytoma. EGFRH835L, as yet unreported in glioma or ependymoma, was also identified here. Strikingly, some of these alterations are actionable: preclinical data links GNAS to MAPK and Wnt signaling creating a potential workspace for these targeted agents. Similarly, loss of STK11 sensitizes 3D gliomaspheres to metformin and mutations in SF3B1 are known to cause cellular sensitivity to PRMT5i. SF3B1 R549S mutation has not been reported yet and it will be crucial to assess the effect this alteration has on protein function. Furthermore, amplification of KIT, KDR and PDGFRA is known to create a sensitive niche for TKI in adenoid cystic carcinoma and may also have potential in ependymoma where these amplifications have been reported for the first time in this study. The EGFR H835L finding has clinical implication since L833V/H835L co-mutation has shown remarkable efficacy to Afatinib in LUAD. Conclusions: We report, for the first-time actionable mutations in key genes related to critical signaling pathways in a mixed population of brain tumor patients. The data highlights that comprehensive NGS-based deep assessment can lead to clinically viable insights in this population of patients with brain tumors.