Pediatric bithalamic gliomas have a distinct epigenetic signature and frequent EGFR exon 20 insertions resulting in potential sensitivity to targeted kinase inhibition

Abstract

Brain tumors are the most common solid tumors of childhood, and the genetic drivers and optimal therapeutic strategies for many of the different subtypes remain unknown. Here, we identify that bithalamic gliomas harbor frequent mutations in the EGFR oncogene, only rare histone H3 mutation (in contrast to their unilateral counterparts), and a distinct genome-wide DNA methylation profile compared to all other glioma subtypes studied to date. These EGFR mutations are either small in-frame insertions within exon 20 (intracellular tyrosine kinase domain) or missense mutations within exon 7 (extracellular ligand-binding domain) that occur in the absence of accompanying gene amplification. We find these EGFR mutations are oncogenic in primary astrocyte models and confer sensitivity to specific tyrosine kinase inhibitors dependent on location within the kinase domain or extracellular domain. We initiated treatment with targeted kinase inhibitors in four children whose tumors harbor EGFR mutations with encouraging results. This study identifies a promising genomically-tailored therapeutic strategy for bithalamic gliomas, a lethal and genetically distinct brain tumor of childhood.

Keywords: Afatinib; Bithalamic glioma; Diffuse midline glioma; EGFR; Erlotinib; Histone H3; Molecular neuropathology; Osimertinib; Pediatric cancer; Trametinib; Tyrosine kinase inhibitor.

Fig. 1

Genetic landscape of pediatric bithalamic gliomas. a, Axial T2-weighted fluid-attenuated inversion recovery (FLAIR) MR image from a 11-year-old girl (patient #3) showing an expansile mass involving the bilateral thalami (top). Histology from the stereotactic biopsy revealing a diffuse astrocytic neoplasm with pleomorphic nuclei and occasional mitotic figures (H&E stain, 400x magnification, middle). Negative immunohistochemical stain for histone H3 K27M mutant protein (400x magnification, bottom). b, Oncoprint table of the clinical features and likely pathogenic alterations identified in biopsies from thirteen children with bithalamic gliomas. Exon number where the EGFR mutations are located is annotated. c, Diagram of human EGFR protein with the location of the recurrent exon 20 small in-frame insertions within the intracellular tyrosine kinase domain and p.A289T/V missense mutations within the extracellular ligand-binding domain. UniProt P00533, RefSeq NM_005228.

Fig. 2

Multi-region sequencing of pediatric bithalamic gliomas reveals that EGFR mutations are clonal and likely an early or initiating event during tumorigenesis. a, Sequencing of a stereotactic biopsy from the right thalamus at time of initial presentation for patient #10 revealed EGFR p.A289T mutation, CDKN2C p.D95del mutation, and gains of chromosomes 1q, 5, 7, and distal Xq. At 14 months following radiation and chemotherapy, a repeat stereotactic biopsy from the right temporal lobe was performed with sequencing analysis revealing the identical EGFR p.A289T mutation and gains of chromosomes 1q, 5, and 7. b, Imaging at time of initial presentation for patient #7 revealed substantial involvement of the bilateral thalami but with more bulky disease burden in the left thalamus. A limited subtotal resection from the posterior aspect of the left thalamus was performed, with sequencing analysis performed on four independent areas of tumor from this resection specimen. The identical EGFR p.A767delinsASVD insertion within exon 20, focal homozygous deletion of BCOR, and nonsense mutation in BCORL1 was present in each of the four tumor regions.

Fig. 3

Epigenomic analysis reveals that pediatric bithalamic gliomas are a distinct glioma entity. Genome-wide DNA methylation profiles from ten bithalamic gliomas were clustered together with 616 glioma reference samples (Supplementary Table 9 [Online supplement 1]) by t-SNE analysis. Bithalamic gliomas (BITHAL) form a cluster that is distinct from all glioma entities that have been studied to date, albeit with some limited overlap with the “Glioblastoma, IDH-wildtype, RTK III subclass”, which is a poorly characterized subgroup of glioblastomas in the cerebral hemispheres of children [9]. The bithalamic gliomas are entirely distinct from their unilateral thalamic counterparts belonging to the methylation class “Diffuse midline glioma, H3 K27M-mutant” (DMG-K27), and are also distinct from the “Glioblastoma, IDH-wildtype, midline subclass” (GBM-MID) that is comprised of glioblastomas located in midline structures lacking H3 K27M mutation [9]. A-IDH, IDH-mutant astrocytoma. CONTR-HEMI, normal cerebral hemisphere. GBM-G34, diffuse hemispheric glioma with H3 G34 mutation. IHG, infantile hemispheric glioma. PXA, pleomorphic xanthoastrocytoma.

Fig. 4

EGFR mutations in pediatric bithalamic gliomas are oncogenic and confer sensitivity to specific tyrosine kinase inhibitors. a, Western blots on total cell lysate from SVG-p12 immortalized human astrocytes after lentiviral transduction with empty vector, EGFR wildtype, or EGFR mutant isoforms. b, Colony formation in soft agar of SVG-p12 immortalized human astrocytes after lentiviral transduction with empty vector, EGFR wildtype, or EGFR mutant isoforms. Images of representative wells (left) and quantitation of colony number per well (right) are shown. Error bars represent standard error from the mean of twelve replicates from two independent experiments. c, Survival plots of SVG-p12 immortalized human astrocytes transduced with empty vector, EGFR wildtype, and EGFR mutant isoforms followed by treatment with various tyrosine kinase inhibitors at the indicated doses for 72 hours. Each data point is the mean of 9 replicates from three independent experiments, and error bars show standard error of the mean. d, Comparison of kinase inhibitor efficacy in SVG-p12 immortalized human astrocytes expressing EGFR wildtype or mutant isoforms.

Fig. 5

Small molecule tyrosine kinase inhibition may extend survival for children with bithalamic gliomas. a, Imaging showing disease course for patient #1 whose therapy included both osimertinib and afatinib. Images are from axial T2/FLAIR sequences taken at different stages during treatment. b, Kaplan-Meier survival analysis of 41 children with bithalamic gliomas, including the 13 children from this study combined with all historical cases of biopsy-proven bithalamic diffuse gliomas reported in the literature (see Supplementary Table 11 [Online supplement 1]). c, Kaplan-Meier survival analysis of these 41 children with bithalamic gliomas stratified based on histologic classification as diffuse astrocytoma, anaplastic astrocytoma, or glioblastoma. d, Kaplan-Meier survival analysis of the 11 children with bithalamic gliomas harboring EGFR mutations from this study cohort stratified by those whose therapy included a small molecule EGFR or MEK kinase inhibitor (n=4) versus those children whose treatment did not include a small molecule tyrosine kinase inhibitor (n=7). p value was calculated by Log-rank (Mantel-Cox) test.