Glioma Susceptibility 1

A number sign (#) is used with this entry because glioma can develop as part of Li-Fraumeni syndrome-1 (LFS1; 151623), a cancer predisposition syndrome caused by mutation in the TP53 gene (191170) on chromosome 17p13.

Gliomas are central nervous system neoplasms derived from glial cells and comprise astrocytomas, glioblastoma multiforme, oligodendrogliomas, ependymomas, and subependymomas. Glial cells can show various degrees of differentiation even within the same tumor (summary by Kyritsis et al., 2010).

Ependymomas are rare glial tumors of the brain and spinal cord (Yokota et al., 2003).

Subependymomas are unusual tumors believed to arise from the bipotential subependymal cell, which normally differentiates into either ependymal cells or astrocytes. They were characterized as a distinct entity by Scheinker (1945). They tend to be slow-growing, noninvasive, and located in the ventricular system, septum pellucidum, cerebral aqueduct, or proximal spinal cord (summary by Ryken et al., 1994).

Gliomas are known to occur in association with several other well-defined hereditary tumor syndromes such as mismatch repair cancer syndrome (276300), melanoma-astrocytoma syndrome (155755), neurofibromatosis-1 (NF1; 162200) and NF2 (101000), and tuberous sclerosis (TSC1; 191100). Familial clustering of gliomas may occur in the absence of these tumor syndromes, however.

Genetic Heterogeneity of Susceptibility to Glioma

Other glioma susceptibilities include GLM2 (613028), caused by variation in the PTEN gene (601728) on chromosome 10q23; GLM3 (613029), caused by variation in the BRCA2 gene (600185) on chromosome 13q12; GLM4 (607248), mapped to chromosome 15q23-q26.3; GLM5 (613030), mapped to chromosome 9p21; GLM6 (613031), mapped to chromosome 20q13; GLM7 (613032), mapped to chromosome 8q24; GLM8 (613033), mapped to chromosome 5p15; and GLM9, caused by variation in the POT1 gene (606478) on chromosome 7q31.

Somatic mutation, disruption, or copy number variation of the following genes or loci may also contribute to the formation of glioma: ERBB (EGFR; 131550), ERBB2 (164870), LGI1 (604619), GAS41 (602116), GLI (165220), DMBT1 (601969), IDH1 (147700), IDH2 (147650), BRAF (164757), PARK2 (602544), TP53 (191170), RB1 (614041), PIK3CA (171834), 10p15, 19q, and 17p13.3.

King and Eisinger (1966) described glioma multiforme of the frontal lobes in father and daughter with development of symptoms at age 50 and 34 years, respectively. Armstrong and Hanson (1969) described 3 sibs who died of brain glioma in adulthood. In a study of cancer mortality during childhood in sibs, Miller (1971) found 8 pairs of nontwin sibs with brain tumor versus 0.9 expected. There were 8 other families versus 0.9 expected in which 1 child died of brain tumor and another died of cancer of bone or muscle. Thuwe et al. (1979) observed 6 cases of brain glioma and a possible seventh on an isolated Swedish coastal island. The affected persons were related as cousins, all in different sibships. One instance of parental consanguinity, the lack of parent-child transmission, and the longtime isolation of the population suggest recessive inheritance. In further studies in this island community, Thuwe (1984) reported 4 closely related cases of brain tumor. It was found that 30 probands with brain tumor were more often the product of a consanguineous marriage than were controls and a higher proportion could be traced to a common ancestor living in the 1600s. It was concluded that genetic factors play a role, although a single major gene seemed unlikely.

Schianchi and Kraus-Ruppert (1980) described affected father and son, suggesting autosomal dominant inheritance.

In a highly inbred Arab family in Israel, Chemke et al. (1985) observed 5 cases of glioblastoma multiforme in 2 sibships. Curiously, all were male and in all the tumor was located on the right side of the brain. The ages of presentation ranged from 4 to 11 years. 'Astrocytoma type 3' was the histologic diagnosis.

Leblanc et al. (1986) described father and son operated on at ages 26 and 37 years, respectively, for mixed oligodendrocytic-astrocytic glioma. Heuch and Blom (1986) reported glioblastoma multiforme in 2 brothers, aged 65 and 68 years, and in their paternal aunt, aged 81. The father had died of tuberculosis before age 40. True multicentric origin, consistent with a hereditary basis, was observed in 1 of the 3 cases.

Clarenbach et al. (1979) described the simultaneous occurrence of fourth ventricular subependymomas in monozygotic twins, both of whom became symptomatic at 22 years of age. Honan et al. (1987) described subependymomas in 3 out of 11 sibs. Ryken et al. (1994) reported the occurrence of fourth ventricular subependymomas in a father and son.

Tijssen (1987) referred to an international register of familial brain tumors maintained in Leyden. Vieregge et al. (1987) gave an extensive review of reported cases of familial glioma with or without other malformations. They reported a family in which members of several generations had one or another abnormality: father and son had glioma; another man and his daughter had colonic polyps; and skeletal abnormalities in the form of short stature and exostoses were present in some members. Munoz et al. (1988) described a brother and sister without a history of phacomatosis or cerebral tumors who developed malignant tumors with ependymal and choroidal differentiation. The girl presented at 28 months with a tumor of the posterior fossa, and the boy presented at 15 months with a tumor of the left cerebral hemisphere. Duhaime et al. (1989) reported histologically identical glioblastoma multiforme in 2 sibs, aged 2 and 5 years, whose symptoms developed simultaneously.

On the basis of segregation analyses in families with multiple glioma patients, autosomal recessive and multifactorial mendelian models have been suggested (Malmer et al., 2001; de Andrade et al., 2001).

In a review of the Utah population database for individuals with primary brain tumors, including 744 astrocytomas and 658 glioblastomas, Blumenthal and Cannon-Albright (2008) found significant excess of affected first-degree relatives among patients with astrocytomas and glioblastomas as a group (relative risk (RR) of 3.29) and for astrocytomas (RR of 3.82) and glioblastomas (RR of 2.29) considered separately. Among second-degree relatives, only astrocytoma showed a significant RR of 1.91 (p = 0.03). Analysis of the data using a genealogic index of familiarity (GIF) showed significant excess relatedness for astrocytomas and glioblastomas as a group and for the astrocytoma subgroup, but not for the glioblastoma subgroup. The results suggested that there is a strongly heritable contribution to astrocytoma risk and nominal contribution to glioblastoma risk.

Marie et al. (2001) found that OLIG2 (606386) expression was upregulated in neoplastic oligodendrocytes, but not in neoplastic astrocytes or in other brain tumor cells, and suggested its use as a specific marker in the diagnosis of oligodendroglial tumors.

Prenatal Diagnosis

The case of prenatal diagnosis ascertained via ultrasound reported by Geraghty et al. (1989) illustrated the occurrence of glioblastoma multiforme in the fetus.

Von Deimling et al. (1995) proposed a simplified model for the pathogenesis of human gliomas. Reviewing their work and that of others, they suggested 3 distinct pathways. The first pathway, which leads to pilocytic astrocytomas (WHO grade I), is caused by loss of heterozygosity for chromosome 17q, presumably unmasking mutations in the NF1 (613113) gene. The second pathway begins with LOH at 17p, unmasking mutations in p53 (191170) and leading to astrocytoma (WHO grade II). Further LOH at 13q, 19q and 9p, unmasking mutations in RB1 (614041), p16 and p15, provides a further step on the second pathway, giving rise to grade III astrocytomas. The final step on the second pathway, LOH on chromosome 10 and perhaps other chromosomes, culminates in glioblastoma multiforme type 1 (WHO grade IV). The third independent pathway begins with LOH at chromosomes 10 and 9p, followed by gene amplification of EGFR (131550), CDK4 (123829), MDM2 (164785), and SAS (181035), and culminating in glioblastoma multiforme type 2 (WHO grade IV).

Bogler et al. (1995) reviewed the role of the p53 gene in the initiation and progression of human gliomas. Pollack et al. (2002) found that overexpression of p53 in malignant gliomas during childhood was strongly associated with an adverse outcome, independent of clinical prognostic factors and histologic findings.

By gene expression profiling, Liang et al. (2005) found significant upregulation of genes involved in macrophage activation, hypoxia, extracellular matrix remodeling, and cell proliferation from 25 different glioblastoma multiforme specimens compared to normal brain tissue. These findings paralleled the histologic observations of macrophage infiltration, tissue necrosis, tumor vasculature, and proliferation of tumor cells. Gene expression patterns were always more closely related to different specimens from the same tumor than to that of any other tumor, suggesting molecular heterogeneity between tumors. Immunohistochemical studies confirmed increased expression of the FABP7 gene (602965), known to be involved in the establishment of the radial glial system in the developing brain, and showed that expression was associated with decreased survival, particularly in younger patients, in 2 unrelated cohorts totaling 105 patients. Transfection of FABP7 into glioma cells in vitro resulted in a 5-fold increase in cell migration compared to control cells, suggesting a functional correlation.

Bao et al. (2006) showed that cancer stem cells contribute to glioma radioresistance through preferential activation of the DNA damage checkpoint response and an increase in DNA repair capacity. The fraction of tumor cells expressing CD133 (604365), a marker for both neural stem cells and brain cancer stem cells, was enriched after radiation in gliomas. In both cell culture and the brains of immunocompromised mice, CD133-expressing glioma cells survived ionizing radiation in increased proportions relative to most tumor cells, which lack CD133. CD133-expressing tumor cells isolated from both human glioma xenografts and primary patient glioblastoma specimens preferentially activated the DNA damage checkpoint in response to radiation, and repaired radiation-induced DNA damage more effectively than CD133-negative tumor cells. In addition, Bao et al. (2006) found that the radioresistance of CD133-positive glioma stem cells could be reversed with a specific inhibitor of the CHK1 (603078) and CHK2 (604373) checkpoint kinases. Bao et al. (2006) concluded that CD133-positive tumor cells represent the cellular population that confers glioma radioresistance and could be the source of tumor recurrence after radiation. Targeting DNA damage checkpoint response in cancer stem cells may overcome this radioresistance and provide a therapeutic model for malignant brain cancers.

Piccirillo et al. (2006) reported that bone morphogenic proteins (BMPs), among which BMP4 (112262) elicits the strongest effect, trigger a significant reduction in the stem-like, tumor-initiating precursors of human glioblastomas. Transient in vitro exposure to BMP4 abolished the capacity of transplanted glioblastoma cells to establish intracerebral glioblastomas. Most importantly, in vivo delivery of BMP4 effectively blocked the tumor growth and associated mortality that occurred in 100% of mice after intracerebral grafting of human glioblastoma cells. Piccirillo et al. (2006) demonstrated that BMPs activate their cognate receptor BMPRs and trigger the SMAD (see 601595) signaling cascade in cells isolated from human glioblastomas. This is followed by a reduction in proliferation, and increased expression of markers of neural differentiation, with no effect on cell viability. The concomitant reduction in clonogenic ability, in the size of the CD133-positive population, and in the growth kinetics of glioblastoma cells indicated that BMP4 reduces the tumor-initiating cell pool of glioblastomas. These findings showed that the BMP-BMPR signaling system, which controls the activity of normal brain stem cells, may also act as a key inhibitory regulator of tumor-initiating, stem-like cells from glioblastomas. The results also identified BMP4 as a novel, noncytotoxic therapeutic effector, which may be used to prevent growth and recurrence of glioblastomas in humans.

Savaskan et al. (2008) found that human primary gliomas showed increased expression of XCT, encoded by the SLC7A11 gene (607933), that was associated with increased glutamate secretion compared to normal brain tissue. Further studies suggested that gliomas secrete glutamate via XCT channels, thereby causing neuronal cell death. Genetic or pharmacologic inhibition of Xct in rats with gliomas abrogated neurodegeneration, attenuated perifocal edema, and prolonged survival. These findings indicated a crucial role for XCT in glioma-induced neurodegeneration and brain edema, corroborating the concept that edema formation may be in part a consequence of peritumoral cell death.

Carro et al. (2010) used reverse engineering and an unbiased interrogation of a glioma-specific regulatory network to reveal the transcriptional module that activates expression of mesenchymal genes in malignant glioma. Two transcription factors, C/EBP-beta (189965) and STAT3 (102582), emerged as synergistic initiators and master regulators of mesenchymal transformation. Ectopic coexpression of C/EBP-beta and STAT3 reprogrammed neural stem cells along the aberrant mesenchymal lineage, whereas elimination of the 2 factors in glioma cells led to collapse of the mesenchymal signature and reduced tumor aggressiveness. In human glioma, expression of C/EBP-beta and STAT3 correlated with mesenchymal differentiation and predicted poor clinical outcome. Carro et al. (2010) concluded that the activation of a small regulatory module is necessary and sufficient to initiate and maintain an aberrant phenotypic state in cancer cells.

Sheng et al. (2010) used a genomewide RNAi screen in mouse glioma cells to identify activators of the transcription factor ATF5 (606398), which is highly expressed in glioma cells. The results indicated that FRS2 (607743), PAK1 (602590), and CREB3L2 (608834) are components of RAS-MAPK- or PI3K-activated pathways that regulate ATF5 expression, and that this pathway is required for viability of malignant glioma cells. Further studies indicated that ATF5 promoted survival through upregulation of MCL1 (159552), an antiapoptotic factor. The ATF5 pathway was also found to promote survival in other human cancer cell lines. Analysis of human malignant glioma samples indicated that ATF5 expression inversely correlated with disease prognosis. The RAF kinase inhibitor sorafenib suppressed ATF5 expression in glioma stem cells and inhibited malignant glioma growth in human cell culture and mouse models. The findings demonstrated that ATF5 is essential in genesis of malignant glioma.

Ricci-Vitiani et al. (2010) showed that a variable number (20 to 90%, mean 60.7%) of endothelial cells in glioblastoma carry the same genomic alteration as tumor cells, indicating that a significant portion of the vascular endothelium has a neoplastic origin. The vascular endothelium contained a subset of tumorigenic cells that produced highly vascularized anaplastic tumors with areas of vasculogenic mimicry in immunocompromised mice. In vitro culture of glioblastoma stem cells in endothelial conditions generated progeny with phenotypic and functional features of endothelial cells. Likewise, orthotopic or subcutaneous injection of glioblastoma stem cells in immunocompromised mice produced tumor xenografts, the vessels of which were primarily composed of human endothelial cells. Selective targeting of endothelial cells generated by glioblastoma stem cells in mouse xenografts resulted in tumor reduction and degeneration, indicating the functional relevance of the glioblastoma stem cell-derived endothelial vessels. Ricci-Vitiani et al. (2010) concluded that their findings described a novel mechanism for tumor vasculogenesis and may explain the presence of cancer-derived endothelial-like cells in several malignancies.

Wang et al. (2010) independently demonstrated that a subpopulation of endothelial cells within glioblastomas harbor the same somatic mutations identified within tumor cells. In addition, the authors demonstrated that the stem cell-like CD133 (604365)-positive fraction includes a subset of vascular endothelial cadherin (CD144)-expressing cells that show characteristics of endothelial progenitors capable of maturation into endothelial cells. Extensive in vitro and in vivo lineage analyses, including single cell clonal studies, further showed that a subpopulation of the CD133-positive stem-like cell fraction is multipotent and capable of differentiation along tumor and endothelial lineages, possibly via an intermediate CD133-positive/CD144-positive progenitor cell. Wang et al. (2010) asserted that their findings were supported by genetic studies of specific exons selected from the Cancer Genome Atlas, quantitative FISH, and comparative genomic hybridization data that demonstrated identical genomic profiles in the CD133-positive tumor cells, their endothelial progenitor derivatives, and mature endothelium. Exposure to the clinical antiangiogenesis agent bevacizumab or to a gamma-secretase inhibitor as well as knockdown small hairpin RNA (shRNA) studies demonstrated that blocking VEGF (192240) or silencing VEGFR2 (191306) inhibits the maturation of tumor endothelial progenitors into endothelium but not the differentiation of CD133-positive cells into endothelial progenitors, whereas gamma-secretase inhibition or NOTCH1 (190198) silencing blocks the transition into endothelial progenitors. Wang et al. (2010) concluded that their data provided novel perspectives on the mechanisms of failure of antiangiogenesis inhibitors. The lineage plasticity and capacity to generate tumor vascularization of the putative cancer stem cells within glioblastoma were novel findings that provided insight into the biology of gliomas and the definition of cancer stemness, as well as the mechanisms of tumor neoangiogenesis.

Johnson et al. (2010) identified subgroups of human ependymoma and then performed genomic analyses and found subgroup-specific alterations that included amplifications and homozygous deletions of genes not previously implicated in ependymoma. They then used cross-species genomics to select cellular compartments most likely to give rise to subgroups of ependymoma and compared human tumors and mouse neural stem cells, isolated from different regions, specifically with an intact or deleted Cdkn2a (600160)/Cdkn2b (600431) locus. The transcriptome of human supratentorial ependymomas with amplified EPHB2 (600997) and deleted CDKN2A/CDKN2B matched only that of embryonic cerebral Cdkn2a/Cdkn2b -/- mouse neuronal stem cells. Activation of Ephb2 signaling in Cdkn2a/Cdkn2b -/- mouse neuronal stem cells, but not other neural stem cells, generated the first mouse model of ependymoma, which was highly penetrant and accurately modeled the histology and transcriptome of 1 subgroup of human supratentorial tumor (subgroup D). Comparative analysis of matched mouse and human tumors revealed selective deregulation in the expression and copy number of genes that control synaptogenesis, pinpointing disruption of this pathway as a critical event in the production of this ependymoma subgroup.

Singh et al. (2012) reported that a small subset of GBMs (3.1%; 3 of 97 tumors examined) harbors oncogenic chromosomal translocations that fuse in-frame the tyrosine kinase coding domains of fibroblast growth factor receptor (FGFR) genes (FGFR1, 136350 or FGFR3, 134934) to the transforming acidic coiled-coil (TACC) coding domains of TACC1 (605301) or TACC3 (605303), respectively. The FGFR-TACC fusion protein displayed oncogenic activity when introduced into astrocytes or stereotactically transduced in the mouse brain. The fusion protein, which localizes to mitotic spindle poles, has constitutive kinase activity and induces mitotic and chromosomal segregation defects and triggers aneuploidy. Inhibition of FGFR kinase corrected the aneuploidy, and oral administration of an FGFR inhibitor prolonged survival of mice harboring intracranial FGFR3-TACC3-initiated glioma. Singh et al. (2012) concluded that FGFR-TACC fusions could potentially identify a subset of GBM patients who would benefit from targeted FGFR kinase inhibition.

Schwartzentruber et al. (2012) sequenced the exomes of 48 pediatric glioblastoma samples. Somatic mutations in the H3.3-ATRX (300032)-DAXX (603186) chromatin remodeling pathway were identified in 44% of tumors (21 of 48). Recurrent mutations in H3F3A (601128), which encodes the replication-independent histone-3 variant H3.3, were observed in 31% of tumors, and led to amino acid substitutions at 2 critical positions within the histone tail (K27M, G34R/G34V) involved in key regulatory posttranslational modifications. Mutations in ATRX and DAXX, encoding 2 subunits of a chromatin remodeling complex required for H3.3 incorporation at pericentric heterochromatin and telomeres, were identified in 31% of samples overall, and in 100% of tumors harboring a G34R or G34V H3.3 mutation. Somatic TP53 (191170) mutations were identified in 54% of all cases, and in 86% of samples with H3F3A and/or ATRX mutations. Screening of a large cohort of gliomas of various grades and histologies (n = 784) showed H3F3A mutations to be specific to glioblastoma multiforme and highly prevalent in children and young adults. Furthermore, the presence of H3F3A/ATRX-DAXX/TP53 mutations was strongly associated with alternative lengthening of telomeres and specific gene expression profiles. Schwartzentruber et al. (2012) stated that this was the first report to highlight recurrent mutations in a regulatory histone in humans, and concluded that their data suggested that defects of the chromatin architecture underlie pediatric and young adult glioblastoma multiforme pathogenesis.

Wu et al. (2012) reported that a K27M mutation occurring in either H3F3A or HIST1H3B (602819) was observed in 78% of diffuse intrinsic pontine gliomas (DIPGs) and 22% of non-brain-stem gliomas.

Lewis et al. (2013) reported that human DIPGs containing the K27M mutation in either histone H3.3 (H3F3A) or H3.1 (HIST1H3B) display significantly lower overall amounts of H3 with trimethylated lysine-27 (H3K27me3) and that histone H3K27M transgenes are sufficient to reduce the amounts of H3K27me3 in vitro and in vivo. Lewis et al. (2013) found that H3K27M inhibits the enzymatic activity of the Polycomb repressive complex-2 (PRC2) through interaction with the EZH2 (601573) subunit. In addition, transgenes containing lysine-to-methionine substitutions at other known methylated lysines (H3K9 and H3K36) are sufficient to cause specific reduction in methylation through inhibition of SET domain enzymes. Lewis et al. (2013) proposed that K-to-M substitutions may represent a mechanism to alter epigenetic states in a variety of pathologies.

Diffuse intrinsic pediatric gliomas (DIPGs) are rare, highly aggressive brainstem tumors. Over 70% of DIPGs harbor a somatic K27M mutation in the H3F3A gene, a substitution associated with a poor prognosis and diminished survival. Funato et al. (2014) used a human embryonic stem cell system to model this tumor, and showed that H3.3K27M expression synergizes with loss of p53 and activation of PDGFRA (173490) in neural progenitor cells derived from human embryonic stem cells, resulting in neoplastic transformation. Genomewide analyses indicated a resetting of the transformed precursors to a developmentally more primitive stem cell state, with evidence of major modifications of histone marks at several master regulator genes. Drug screening assays identified a compound targeting the protein menin (613733) as an inhibitor of tumor cell growth in vitro and in mice.

Kim et al. (2015) identified a key role for serine and glycine metabolism in the survival of brain cancer cells within the ischemic zones of gliomas. In human glioblastoma multiforme, SHMT2 (138450) and GLDC (238300) are highly expressed in the pseudopalisading cells that surround necrotic foci. Kim et al. (2015) found that SHMT2 activity limits that of PKM2 (179050) and reduces oxygen consumption, eliciting a metabolic state that confers a profound survival advantage to cells in poorly vascularized tumor regions. GLDC inhibition impairs cells with high SHMT2 levels, as the excess glycine not metabolized by GLDC can be converted to the toxic molecules aminoacetone and methylglyoxal. Kim et al. (2015) concluded that SHMT2, which is required for cancer cells to adapt to the tumor environment, also renders these cells sensitive to glycine cleavage system inhibition.

Flavahan et al. (2016) showed that human IDH1 (147700) and IDH2 (147650) mutant gliomas exhibit hypermethylation at cohesin (see 606462) and CTCF (604167)-binding sites, compromising binding of this methylation-sensitive insulator protein. Reduced CTCF binding is associated with loss of insulation between topologic domains and aberrant gene activation. Flavahan et al. (2016) specifically demonstrated that loss of CTCF at a domain boundary permits a constitutive enhancer to interact aberrantly with the receptor tyrosine kinase gene PDGFRA, a prominent glioma oncogene. Treatment of IDH mutant gliomaspheres with a demethylating agent partially restored insulator function and downregulated PDGFRA. Conversely, CRISPR-mediated disruption of the CTCF motifs in IDH wildtype gliomaspheres upregulated PDGFRA and increased proliferation.

The most common activating mutation of EGFR in glioblastoma is deletion of exons 2 through 7, which generates a constitutively active EGFR, termed EGFRvIII, that induces phosphorylation of STAT3 to drive tumorigenesis. Using RNA sequencing analysis, Western blot analysis, and deletion and knockdown experiments, Jahani-Asl et al. (2016) found that OSMR (601743) was highly expressed in a STAT3-dependent manner in EGFRvIII-expressing human brain tumor stem cells (BTSCs) and mouse astrocytes compared with controls. Chromatin immunoprecipitation and sequencing showed that STAT3 occupied the promoter of the OSMR gene. There was significant overlap among OSMR-, STAT3-, and EGFRvIII-dependent target genes. Immunohistochemical analysis demonstrated that OSMR and EGFRvIII formed a coreceptor complex at the cell membrane, and gp130 (IL6ST; 600694) and wildtype EGFR were not required for the interaction. OSM (165095) signaling induced phosphorylation and activation of EGFR, leading to EGFR-OSMR interaction. Knockdown of OSMR inhibited proliferation of BTSCs and astrocytes. Furthermore, knockdown of Osmr suppressed tumor growth in SCID mice injected with EgfrvIII-expressing astrocytes or BTSCs. Jahani-Asl et al. (2016) concluded that OSMR is a cell surface receptor that defines a feed-forward mechanism with EGFRvIII and STAT3 in glioblastoma pathogenesis.

Pilocytic Astrocytoma

Jones et al. (2013) described whole-genome sequencing of 96 pilocytic astrocytomas, with matched RNA sequencing for 73 samples, conducted by the International Cancer Genome Consortium PedBrain Tumor Project. Jones et al. (2013) identified recurrent activating mutations in FGFR1 and PTPN11 (176876) and novel NTRK2 (600456) fusion genes in noncerebellar tumors. Novel BRAF (164757)-activating changes were also observed. MAPK pathway alterations affected all tumors analyzed, with no other significant mutations identified, indicating that pilocytic astrocytoma is predominantly a single-pathway disease. Notably, Jones et al. (2013) identified the same FGFR1 mutations in a subset of H3F3A (601128)-mutated pediatric glioblastoma with additional alterations in the NF1 gene (613113).

Gilchrist and Savard (1989) described the familial occurrence of ependymoma in 2 sisters and a maternal male cousin. Karyotypic analysis of the tumor from 1 sister showed mosaicism for the loss of one chromosome 22.

Approximately 30% of ependymomas are said to have monosomy 22 as revealed by cytogenetic studies (Griffin et al., 1992; Ransom et al., 1992; Sainati et al., 1992).

Yokota et al. (2003) reported a non-neurofibromatosis type II (101000) Japanese family in which 2 of 4 sibs had cervical spinal cord ependymoma and 1 of the 4 had schwannoma. Loss of heterozygosity (LOH) studies in 2 of the patients showed a common allelic loss at 22q11.2-qter. The findings suggested the existence of a tumor suppressor gene on chromosome 22 related to the tumorigenesis of familial ependymal tumors. The NF2 gene (607379) maps to chromosome 22q12.

Kyritsis et al. (1994) identified germline mutations in the TP53 gene (see, e.g., 191170.0042) in 6 of 19 patients with multifocal glioma, all of whom had a family history of cancer. In addition, germline TP53 mutations were found in 3 of 19 patients with unifocal glioma and a family history of cancer. No mutations were detected in a patient with unifocal glioma and another malignancy or in 12 control patients with unifocal glioma and no second malignancies or family history of cancer. Patients with mutations were younger than other patients in the same group. Kyritsis et al. (1994) concluded that germline TP53 mutations are frequent in patients with multifocal glioma, glioma and another primary malignancy, and glioma associated with a family history of cancer, particularly if these factors are combined.

Chen et al. (1995) found somatic mutations in the TP53 gene in 8 of 22 adult glioma tissue specimens and germline mutations in 2 of these 8 patients. Both patients with germline mutations developed glioblastoma multiforme before the age of 31, compared to the median age of greater than 50 for glioma patients. Family history was not available for these patients. TP53 mutations were not found in 16 glial tumors occurring in children or in benign meningiomas. The findings suggested that TP53 germline mutations may identify a subset of young adults predisposed to the development of high-grade astrocytic tumors.

In a family in which several individuals had glioblastome multiforme and additional family members had multiple cancer types including some consistent with Li-Fraumeni syndrome (151623), Tachibana et al. (2000) identified a germline mutation in the p53 gene (R248Q; 191170.0010). The authors concluded that point mutations of p53 may be responsible for some apparent familial glioma cases.

Modifier Genes

Among 254 patients with glioblastoma multiforme, El Hallani et al. (2009) found an association between a pro72 allele in the TP53 gene (191170.0005) and earlier age at onset. The pro/pro genotype was present in 20.6% of patients with onset before age 45 years compared to in 6.5% of those with onset after age 45 years (p = 0.002) and 5.9% among 238 controls (p = 0.001). The findings were confirmed in an additional cohort of 29 patients. The variant did not have any impact on overall patient survival. Analysis of tumor DNA from 73 cases showed an association between the pro allele and a higher rate of somatic TP53 mutations.

Somatic Mutations

To identify the genetic alterations in glioblastoma multiforme (GBM), Parsons et al. (2008) sequenced 20,661 protein-coding genes, determined the presence of amplifications and deletions using high-density oligonucleotide arrays, and performed gene expression analyses using next-generation sequencing technologies in 22 human tumor samples. This comprehensive analysis led to the discovery of a variety of genes that were not known to be altered in GBMs. Most notably, Parsons et al. (2008) found recurrent mutations in the active site of isocitrate dehydrogenase-1 (IDH1; 147700) in 12% of GBM patients. Mutations in IDH1 occurred in a large fraction of young patients and in most patients with secondary GBMs and were associated with an increase in overall survival. Parsons et al. (2008) concluded that their studies demonstrated the value of unbiased genomic analyses in the characterization of human brain cancer and identified a potentially useful genetic alteration for the classification and targeted therapy of GBMs. Parsons et al. (2008) found that the hazard ratio for death among 79 patients with wildtype IDH1, as compared to 11 with mutant IDH1, was 3.7 (95% confidence interval, 2.1 to 6.5; p less than 0.001). The median survival was 3.8 years for patients with mutated IDH1, as compared to 1.1 years for patients with wildtype IDH1. Parsons et al. (2008) found that a majority of tumors analyzed had alterations in genes encoding components of each of the TP53 (191170), RB1 (614041), and PI3K (see 171834) pathways.

The Cancer Genome Atlas Research Network (2008) reported the interim integrative analysis of DNA copy number, gene expression, and DNA methylation aberrations in 206 glioblastomas and nucleotide sequence alterations in 91 of the 206 glioblastomas. The authors found that p53 itself showed mutation or homozygous deletion in 35% of tumors and that there was altered p53 signaling in 87% of tumors, as demonstrated by homozygous deletion or mutations in CDKN2A (600160) in 49% of tumors, amplification of MDM2 (164785) in 14%, and amplification of MDM4 (602704) in 7%. The authors also observed that the RTK/RAS/PI3K signaling pathway was altered in 88% of glioblastomas. EGFR (131550) mutation or amplification was present in 45%, PDGFRA (173490) amplification was present in 13%, and MET (164860) amplification was present in 4%. (ERBB2 (164870) mutation was reported in 8%; in an erratum, the group stated that the somatic mutations reported in ERBB2 were actually an artifact of DNA amplification and were not validated in unamplified DNA.) Furthermore, NF1 (613113) was found to be an important gene in glioblastoma as mutation or homozygous deletion of the NF1 gene was present in 18% of tumors. Somatic mutation in the PI3K complex was frequently identified. In particular, novel somatic mutations were identified in the PIK3R1 gene (171833) that result in disruption of the important C2-iSH2 interaction between PIK3R1 and PIK3CA (171834). The RB signaling pathway was found to be altered in 78% of glioblastomas, with RB itself mutated in 11% of tumors. Of note, the Cancer Genome Atlas Research Network (2008) found a link between MGMT (156569) promoter methylation and hypermutator phenotype consequent to mismatch repair deficiency in treated glioblastomas. The methylation status of MGMT predicts sensitivity to temozolomide, an alkylating agent used to treat glioblastoma patients. In those patients who also have mutation in the mismatch repair pathway, treatment with an alkylating agent was associated with characteristic C-G and A-T transversions in non-CpG sites, raising the possibility that patients who initially respond to treatment with alkylating agents may evolve not only treatment resistance but also a mismatch repair-defective hypermutator phenotype.

Bredel et al. (2011) analyzed 790 human glioblastomas for deletions, mutations, or expression of NFKBIA (164008) and EGFR. They further studied the tumor suppressor activity of NFKBIA in tumor cell culture and compared the molecular results with the outcome of glioblastoma in 570 affected individuals. Bredel et al. (2011) found that NFKBIA is often deleted but not mutated in glioblastomas; most deletions occur in nonclassical subtypes of the disease. Deletion of NFKBIA and amplification of EGFR show a pattern of mutual exclusivity. Restoration of the expression of NFKBIA attenuated the malignant phenotype and increased the vulnerability to chemotherapy of cells cultured from tumors with NFKBIA deletion; it also reduced the viability of cells with EGFR amplification but not of cells with normal gene dosages of both NFKBIA and EGFR. Deletion and low expression of NFKBIA were associated with unfavorable outcomes. Patients who had tumors with NFKBIA deletion had outcomes that were similar to those in patients with tumors harboring EGFR amplification. These outcomes were poor as compared with the outcomes in patients with tumors that had normal gene dosages of NFKBIA and EGFR. Bredel et al. (2011) suggested a 2-gene model that was based on expression of NFKBIA and O(6)-methylguanine DNA methyltransferase (156569) being strongly associated with the clinical course of the disease, and concluded that deletion of NFKBIA has an effect that is similar to the effect of EGFR amplification in the pathogenesis of glioblastoma and is associated with comparatively short survival.

Frattini et al. (2013) described a computational platform that integrates the analysis of copy number variations and somatic mutations, and unravels the landscape of in-frame gene fusions in glioblastoma. The authors found mutations with loss of heterozygosity in LZTR1 (600574), encoding an adaptor of CUL3 (603136)-containing E3 ligase complexes. Mutations and deletions disrupt LZTR1 function, which restrains the self-renewal and growth of glioma spheres that retain stem cell features. Loss-of-function mutations in CTNND2 (604275) target a neural-specific gene and are associated with the transformation of glioma cells along the very aggressive mesenchymal phenotype. Frattini et al. (2013) also reported recurrent translocations that fuse the coding sequence of EGFR to several partners, with EGFR/SEPT14 (612140) being the most frequent functional gene fusion in human glioblastoma. EGFR/SEPT14 fusions activate STAT3 (102582) signaling and confer mitogen independence and sensitivity to EGFR inhibition.

Mutations in IDH1 and IDH2

Yan et al. (2009) determined the sequence of the IDH1 (147700) gene and related IDH2 (147650) gene in 445 CNS tumors and 494 non-CNS tumors. The enzymatic activity of the proteins that were produced from normal and mutant IDH1 and IDH2 genes was determined in cultured glioma cells that were transfected with these genes. Yan et al. (2009) identified mutations that affected amino acid 132 of IDH1 (see 147700.0001) in more than 70% of World Health Organization (WHO) grade II and III astrocytomas and oligodendrogliomas and in glioblastomas that developed from these lower-grade lesions. Tumors without mutations in IDH1 often had mutations affecting the analogous amino acid (R172) of the IDH2 gene. Tumors with IDH1 or IDH2 mutations had distinctive genetic and clinical characteristics, and patients with such tumors had a better outcome than those with wildtype IDH genes. Each of the 4 tested IDH1 and IDH2 mutations reduced the enzymatic activity of the encoded protein. Yan et al. (2009) concluded that mutations of NADP(+)-dependent isocitrate dehydrogenases encoded by IDH1 and IDH2 occur in a majority of several types of malignant gliomas.

De Carli et al. (2009) found that IDH1 mutations were more commonly found in adult patients with gliomas (38%; 155 of 404) compared to children with gliomas (5%; 4 of 73). No IDH2 mutations were found in 73 children with gliomas. IDH1 mutations in adults were significantly associated with lower tumor grade, increased survival, and younger age. Children with tumors bearing IDH1 mutations were older than children with mutation-negative tumors. The findings suggested that pediatric and adult gliomas differ biologically.

In a retrospective study of 49 progressive astrocytomas, 42 (86%) of which had somatic mutations in the IDH1 gene, Dubbink et al. (2009) found that the presence of IDH1 mutations was significantly associated with increased patient survival (median survival, 48 vs 98 months), but did not affect outcome of treatment with temozolomide.

Bralten et al. (2011) found that overexpression of IDH1-R132H (147700.0001) in established glioma cell lines resulted in decreased proliferation and more contact-dependent cell migration compared to wildtype. Intracerebral injection of IDH1-R132H in mice, as compared to injection of wildtype, resulted in increased survival and even absence of tumor in 1 mouse. Reduced cellular proliferation was associated with accumulation of D-2-hydroxyglutarate that is produced by the R132H variant protein. The decreased proliferation was not associated with increased apoptosis, but was associated with decreased AKT1 (164730) activity. The findings indicated that R132H dominantly reduces aggressiveness of established glioma cell lines in vitro and in vivo. Bralten et al. (2011) noted that the findings were apparently contradictory because the presence of an IDH1 mutation was thought to contribute to tumorigenesis; the authors suggested that IDH1 mutations may be involved in tumor initiation and not in tumor progression. IDH1-mutant tumors are typically low-grade and often slow-growing.

Chromosome 7

In a series of human glioblastoma cell lines, Henn et al. (1986) found that the most striking and consistent chromosomal finding was an increase in copy number of chromosome 7. In all of the cell lines, ERBB-specific mRNA (EGFR; 131550) was increased to levels even higher than expected from the number of chromosomes 7 present. These changes were not found in benign astrocytomas. Previously, Downward et al. (1984) presented evidence that oncogene ERBB may be derived from the gene coding for EGFR.

Bigner et al. (1988) determined that double minute chromosomes, indicating the presence of gene amplification, are found in about 50% of malignant gliomas. Most tumors with double minute chromosomes contain 1 of 5 amplified genes, most often the EGFR gene on chromosome 7. Following up on the observation that the EGFR gene is amplified in 40% of malignant gliomas, Wong et al. (1992) characterized the rearrangements in 5 malignant gliomas. In one they found deletion of most of the extracytoplasmic domain of the receptor. The 4 other tumors had internal deletions of the gene.

Using array CGH, Pfister et al. (2008) found that 30 (45%) of 66 low-grade pediatric astrocytomas contained a somatic copy number gain at chromosome 7q34 spanning the BRAF (164757) locus, among others. These changes were associated with increased BRAF mRNA, and further studies showed evidence for activation of the MAPK1 (176948) pathway and downstream targets, such as ERK1/2 (see, e.g., 176872) and CCND1 (168461). Four (6%) of the tumors had an activating BRAF somatic mutation (V600E; 164757.0001). Among 26 adult tumors, 16 (62%) had copy number gains of the BRAF locus. Other changes in the 66 pediatric tumors including large somatic trisomies of chromosomes 5 (6 of 66) and 7 (4 of 66). Initial in vitro pharmacologic studies suggested that inhibition of the MAPK pathway may be possible.

Yu et al. (2009) found that 42 (60%) of 70 sporadic pilocytic astrocytomas had rearrangements of the BRAF gene. Two additional tumors with no rearrangement carried a BRAF mutation. Twenty-two of 36 tumors with BRAF rearrangements had corresponding amplification of the neighboring HIPK2 gene (606868). However, 14 of 36 tumors with BRAF rearrangement had no detectable HIPK2 gene amplification. Six of 20 tumors demonstrated HIPK2 amplification without apparent BRAF rearrangement or mutation. Only 12 (17%) of the 70 tumors lacked detectable BRAF or HIPK2 alterations. Yu et al. (2009) concluded that BRAF rearrangement represents the most common genetic alteration in sporadic pilocytic astrocytomas.

Chromosome 10

Bigner et al. (1988) concluded that the most frequent chromosomal changes in malignant gliomas are gains of chromosome 7 and losses of chromosome 10. Loss of 1 copy of chromosome 10 is a common event in high-grade gliomas. Rearrangement and loss of at least some parts of the second copy, especially in the 10q23-q26 region, has been demonstrated in approximately 80% of glioblastoma multiforme tumors (Bigner and Vogelstein, 1990).

Chromosome 10 was implicated in glioblastoma multiforme by Fujimoto et al. (1989), who found loss of constitutional heterozygosity in tumor samples from 10 of 13 patients in whom paired tumor and lymphocyte DNA samples were screened. In a search for submicroscopic deletions in chromosome 10, Fults and Pedone (1993) performed a RFLP analysis in 30 patients, using markers that had been mapped accurately on chromosome 10 by genetic linkage studies. Loss of heterozygosity (LOH) at one or more loci was found in 15 of the 30 patients. In 7 cases, LOH was found at every informative locus. LOH was confined to a portion of the long arm in 6 patients; the smallest region of overlap among these 6 deletions was flanked by markers D10S12 proximally and D10S6 distally, a 33.4-cM region mapped physically near the telomere, 10q25.1-qter.

Karlbom et al. (1993) analyzed a panel of glial tumors consisting of 11 low-grade gliomas, 9 anaplastic gliomas, and 29 glioblastomas for loss of heterozygosity by examining at least one locus for each chromosome. The frequency of allele loss was highest among the glioblastomas, suggesting that genetic alterations accumulate during glial tumor development. The most common genetic alteration was found to involve allele losses of chromosome 10 loci, these being found in all glioblastomas and in 3 anaplastic tumors. Deletion mapping analysis revealed partial loss of chromosome 10 in 8 glioblastomas and 2 anaplastic tumors in 3 distinct regions: one telomeric region on 1p and both telomeric and centromeric locations on 10q. These data suggested to Karlbom et al. (1993) the existence of multiple chromosome 10 tumor suppressor gene loci whose inactivation is involved in the malignant progression of glioma. In studies of 20 gliomas with microsatellite markers from chromosome 10, the locus that exhibited the most loss (69%) was the region bordered by D10S249 and D10S558 and inclusive of D10S594, with a linkage distance of 3 cM (Kimmelman et al., 1996). This region was known to be deleted in various grades of tumor, including low- and high-grade tumors. Kimmelman et al. (1996) suggested that chromosome region 10p15 is involved in human gliomas of diverse grades and that this region may harbor genes important in the development of and progression to the malignant phenotype.

See the DMBT1 gene (601969), so designated for 'deleted in malignant brain tumors,' for a discussion of a gene on 10q25.3-q26.1 that showed intragenic homozygous deletions in medulloblastoma and glioblastoma multiforme tumor tissue and in brain tumor cell lines (Mollenhauer et al., 1997). Chernova