Abstract

RNA-Seq analysis of glioma tumors to reveal targetable gene fusions.

Author
Deepa Suresh Subramaniam Georgetown Lombardi Comprehensive Cancer Center, Washington, DC info_outline Deepa Suresh Subramaniam, Joanne Xiu, Shwetal Mehta, Zoran Gatalica, Jeffrey Swensen, Nader Sanai, Amy B. Heimberger
Full text
Authors Deepa Suresh Subramaniam Georgetown Lombardi Comprehensive Cancer Center, Washington, DC info_outline Deepa Suresh Subramaniam, Joanne Xiu, Shwetal Mehta, Zoran Gatalica, Jeffrey Swensen, Nader Sanai, Amy B. Heimberger Organizations Georgetown Lombardi Comprehensive Cancer Center, Washington, DC, Caris Life Sciences, Phoenix, AZ, Barrow Neurological Institute, Phoenix, AZ, The University of Texas MD Anderson Cancer Center, Houston, TX Abstract Disclosures Research Funding Other Background: Fusions involving oncogenes have been reported in gliomas and may serve as novel therapeutic targets. We aim to use RNA-sequencing to interrogate a large cohort of gliomas for targetable genetic fusions. Methods: Gliomas were profiled using the ArcherDx FusionPlex Assay at a CLIA-certified lab (Caris Life Sciences) and 52 gene targets were analyzed. Fusions with preserved kinase domains were investigated. Results: Among 404 gliomas tested, 39 (9.7%) presented potentially targetable fusions, of which 24/226 (11%) of glioblastoma (GBM), 5/42 (12%) of anaplastic astrocytoma (AA), 2/25 (8%) of grade II astrocytoma and 3 of 7 (43%) of pilocytic astrocytoma (PA) harbored targetable fusions. In GBMs, 1 of 15 (6.7%) IDH-mutated tumors had a fusion while 22 of 175 (12.6%) IDH-wild type tumors had fusions. 46 oligodendroglial tumors were profiled and no fusions were seen, which was lower than frequency of fusions in astrocytic tumors (34/300, p = 0.0236). The most frequent fusions seen involved FGFR3 (N = 12), including 10 FGFR3-TACC3 (1 AA, 6 GBM and 3 glioma NOS); 1 FGFR3-NBR1 (AA) and 1 FGFR3-BRAP (GBM). 11 fusions involving MET were seen, 10 in GBM and 1 in AA. The most common MET fusion was PTPRZ1-MET (1 in AA and 4 in GBM), followed by ST7-MET (N = 3, GBM), CAPZA2-Met (N = 2, GBM) and TPR-MET (N = 1, GBM). 8 NTRK fusions were seen; 1 involving NTRK1 (BCAN-NTRK1, PA), 6 NTRK2 (1 NOS1AP-NTRK2 in AA; GKAP1-NTRK2, KCTD8-NTRK2, TBC1D2-NTRK2 and SOSTM1-NTRK2, 1 each in GBM and 1 VCAN-NTRK2 in grade II astrocytoma) and 1 NTRK3 (EML4-NTRK3 in GBM). EGFR fusions (2 EGFR-SEPT14 and 1 EGFR-VWC2) were seen in 3 GBMs, BRAF in 3 (1 KIAA1549-BRAF, 1 LOC100093631-BRAF in PA and 1 ZSCAN23-BRAF in glioma NOS) and PDGFRA (RAB3IP-PDGFRA, in GBM) in 1. C11orf95-RELA fusions were seen in 2 of 3 grade III ependymomas but not in the 2 grade II ependymomas. Conclusions: We report targetable fusion genes involving NTRK, MET, EGFR, FGFR3, BRAF and PDGFRA including novel fusions that haven’t been previously described in gliomas (e.g., EGFR-VWC2; FGFR3-NBR1). Fusions were seen in over 10% of astrocytic tumors, while none was seen oligodendrogliomas. Identification of such kinase-associated fusion transcripts may allow us to exploit therapeutic opportunities with targeted therapies in gliomas.