Acta Neuropathol

DOI 10.1007/s00401-015-1432-1

REVIEW

Glioblastoma: pathology, molecular mechanisms and markers

Kenneth Aldape1 · Gelareh Zadeh1,2 · Sheila Mansouri1 ·
Guido Reifenberger3 · Andreas von Deimling4

Received: 25 February 2015 / Revised: 14 April 2015 / Accepted: 22 April 2015

© Springer-Verlag Berlin Heidelberg 2015

Abstract Recent advances in genomic technology have (IDH1/2 wt), and secondary (IDH1/2 mut) GBMs. In this
led to a better understanding of key molecular alterations review, we describe major clinically relevant genetic and
that underlie glioblastoma (GBM). The current WHO- epigenetic abnormalities in GBM—such as mutations in
based classification of GBM is mainly based on histologic IDH1/2, EGFR, PDGFRA, and NF1 genes—altered meth-
features of the tumor, which frequently do not reflect the ylation of MGMT gene promoter, and mutations in hTERT
molecular differences that describe the diversity in the biol- promoter. These markers may be incorporated into a more
ogy of these lesions. The current WHO definition of GBM refined classification system and applied in more accurate
relies on the presence of high-grade astrocytic neoplasm clinical decision-making process. In addition, we focus on
with the presence of either microvascular proliferation and/ current understanding of the biologic heterogeneity and
or tumor necrosis. High-throughput analyses have identi- classification of GBM and highlight some of the molecular
fied molecular subtypes and have led to progress in more signatures and alterations that characterize GBMs as histo-
accurate classification of GBM. These findings, in turn, logically defined. We raise the question whether IDH-wild
would result in development of more effective patient strat- type high grade astrocytomas without microvascular prolif-
ification, targeted therapeutics, and prediction of patient eration or necrosis might best be classified as GBM, even
outcome. While consensus has not been reached on the if they lack the histologic hallmarks as required in the cur-
precise nature and means to sub-classify GBM, it is clear rent WHO classification. Alternatively, an astrocytic tumor
that IDH-mutant GBMs are clearly distinct from GBMs that fits the current histologic definition of GBM, but which
without IDH1/2 mutation with respect to molecular and shows an IDH mutation may in fact be better classified as
clinical features, including prognosis. In addition, recent a distinct entity, given that IDH-mutant GBM are quite dis-
findings in pediatric GBMs regarding mutations in the his- tinct from a biological and clinical perspective.
tone H3F3A gene suggest that these tumors may represent
a 3rd major category of GBM, separate from adult primary Keywords Glioblastoma · TCGA · G-CIMP · IDH1/2 ·
MGMT

* Kenneth Aldape
ken.aldape@uhn.ca Introduction
1
Princess Margaret Cancer Centre and MacFeeters-Hamilton
Centre for Neuro-Oncology Research, 101 College St., Glial tumors can be divided into two major categories based
Toronto, ON M5G 1L7, Canada on the degree of invasiveness into the surrounding brain
2
Division of Neurosurgery, Toronto Western Hospital, tissue; gliomas with diffuse infiltration of the brain paren-
University of Toronto, Toronto, Canada chyma are referred to as “diffuse gliomas”, to be contrasted
3
Department of Neuropathology, Heinrich, Heine University with gliomas with more “circumscribed” growth behavior.
Düsseldorf, Düsseldorf, Germany Diffuse gliomas share the ability to infiltrate surrounding
4
Department of Neuropathology, Institute of Pathology, normal brain parenchyma, and unfortunately, inevitably
German Cancer Research Center, Heidelberg, Germany recur even after gross total resection [136]. Given their

13
 Acta Neuropathol

extensive infiltrative nature, major goals for neurosurgery of malignancies are approached. In turn, molecular signa-
include cytoreduction, to the extent that is possible, as well tures that are found to either describe fundamental biologic
as obtaining tissue for accurate diagnosis. Another feature behavior or correlate clinically with patient outcome—fol-
of diffuse gliomas is the notion that low-grade tumors of lowing administration of either cytotoxic or molecularly
WHO grade II over time not only recur but also tend to pro- targeted agents—become candidates to enter classification
gress to high-grade (anaplastic) gliomas of WHO grade III criteria as circumstances warrant. Due to this shift, tumors
and eventually secondary GBM of WHO grade IV, leading are classified not only based on a static concept of how
to rapid clinical deterioration. GBM is considered incura- they “look” under the microscope, but rather by incorpo-
ble, with a median survival of 15 months following aggres- rating molecular markers relevant to current therapeutic
sive combination of therapies including maximal-safe modalities. Clear proof of principle for such approaches
surgical resection, adjuvant radiation therapy (RT) with has been demonstrated in therapies targeting EGFR mutant
concurrent and adjuvant temozolomide (TMZ) treatment non-small-cell lung cancer [103], HER2-amplified breast
[154]. Many tumors respond poorly to conventional chemo- cancer [149], lung cancer harboring the EML4–ALK trans-
therapy and radiation, and those for which tumor control is location [89], chronic myelogenous leukemia (CML) har-
accomplished often lack a durable therapy response [107]. boring the BCR–ABL translocation [36], and BRAF mutant
Therefore, development of new diagnostic approaches and melanoma [26]. Annotations of molecular alterations are
especially more effective treatment strategies is urgently more routinely being incorporated into histopathologic
needed. Targeting driver molecular aberrations is the most diagnosis where appropriate [34, 104] and have facilitated
promising therapeutic advancement, as seen with successes therapeutic decision making, progressively decreasing the
of “personalized” and targeted therapies in other cancer time frame from target discovery to therapy [25]. Molecu-
types. In this context, we provide in this review an update lar initiatives, including the Cancer Genome Atlas (TCGA),
on the state of our knowledge in this field, focusing on how have described fundamental aspects of the biologic under-
understanding of the molecular heterogeneity of GBM has pinnings of GBM [18] and lower grade gliomas (TCGA
been and could be utilized for classification of these tumors network, unpublished data). The TCGA project was initi-
into molecular subtypes that could potentially improve out- ated by the NIH and is a comprehensive, coordinated, mul-
comes for specific tumor subsets. ticenter effort that applies multiple innovative genomic
To date, gliomas are classified largely based on their analysis tools to understand the genetics and epigenetics
histopathological characteristics and while clinical and of cancer. More than 20 cancer types, including more than
radiological features of the tumors are at times taken into 10,000 samples, will undergo detailed genomic characteri-
account, the present WHO classification is mainly based zation and further incorporated with bioinformatic and data
on histological features. Histologic criteria for high-grade analysis components that will enable researchers to apply
infiltrating astrocytic tumor (at least grade III) include this information for prevention, diagnosis, and treatment of
hypercellularity, nuclear atypia, and mitotic activity. Fur- cancer. Unfortunately, although such molecular alterations
thermore, a GBM diagnosis requires, in addition, either have led to extensive clinical progress for many cancer
microvascular proliferation and/or tumor necrosis. How- types, to date these alterations have not been incorporated
ever, many aspects of these histologic features remain into clinical decision making where ultimately the subtype
poorly correlated with key molecular drivers and path- classification can be matched with efficacious therapeutic
ways. For example, the presence or absence of IDH muta- options. In addition, more detailed characterization of the
tions cannot be distinguished on pure morphologic grounds genomic alterations that are clinically relevant still need to
in GBM. In addition, among IDH wild-type high-grade be established in GBM to fully implement and maximize
gliomas (which account for over 90 % of GBM), the key information from these high-throughput genomic studies.
molecular chromosomal changes are shared between histo-
logic grade III (anaplastic astrocytoma) and GBM (histo-
logic grade IV) tumors. In addition, clinical and biological Clinical diagnosis of GBM
variability is thought to exist within each grade and each
tumor entity, suggesting that identification of molecular Based on guidelines of the World Health Organization
factors which contribute to this variation would be invalu- (WHO) for classification of central nervous system tumors
able for the development of targeted therapies. In other [101], diffuse gliomas are divided into three grades:
words, histopathologically defined GBM in fact may rep- WHO grades II, III, and IV, with WHO grade IV diffuse
resent multiple subtypes based on molecular features or glioma being synonymous with GBM. Diffuse gliomas
signatures. occur more commonly in adults than in children and are
The emergence of molecular signatures in cancer can the most common intrinsic primary brain tumors that dis-
iteratively present a shift in the way diagnosis and treatment play a wide range of clinical behaviors, ranging from slow

13
Acta Neuropathol

clinical progression in patients with WHO grade II tumors, definitions. For example, TERT promoter mutation, PTEN
to very short median survival times of 12–18 months in tumor suppressor gene mutation, and high-level gene
patients with WHO grade IV tumors (GBM); however, amplification of certain proto-oncogenes—most commonly
long-term survival extending the span of three years has the epidermal growth factor receptor (EGFR) gene—are
been described in a fraction of GBM patients [88]. Diffuse hallmark alterations in primary GBMs, while mutations of
gliomas of WHO grade II or III are further divided into IDH1/2, TP53, and ATRX are frequent in secondary GBMs
several histologic entities, including astrocytoma/anaplas- [97, 118]. Going further, several recent studies have utilized
tic astrocytoma, oligodendroglioma/anaplastic oligoden- high-throughput genomic, epigenomic, and transcriptomic
droglioma, and oligoastrocytoma/anaplastic oligoastrocy- approaches for detailed molecular characterization of glio-
toma. The most common diffuse glioma, however, is GBM mas [18, 66, 166, 176]. The identification of distinctive and
(WHO grade IV), accounting for 45–50 % of all primary highly recurrent molecular alterations has begun to clarify
intrinsic brain tumors [38, 101, 161], with the vast major- some of this diversity and introduce new concepts in tumor
ity of GBMs arising de novo as “primary GBMs”. GBMs classification. Further, these studies provide insights for
that develop by progression from a pre-existing glioma of improvement of current therapeutic strategies and develop-
WHO grade II or III are less common and are referred to as ment of a new paradigm for the management of this deadly
“secondary GBMs” [118]. Most primary GBMs manifest in malignancy.
elderly patients, while secondary GBMs most commonly Large-scale molecular profiling of diffuse gliomas has
affect younger patients prior to the age of 45 years. taken place in individual laboratories [13, 123, 129], at the
Histopathologically, several patterns exist, including national level in the US by TCGA network [23], and at the
giant cell GBM, small cell GBM, and gliosarcoma. Glio- international level within the International Cancer Genome
sarcoma can be observed at initial diagnosis or at recur- Consortium (ICGC) [74]. GBM was one of the early tumor
rence, and appears to have similar genetic aberrations types that was investigated by TCGA and characteriza-
as GBM, although MGMT methylation may be less fre- tion of the genome and transcriptome of these tumors has
quently present [92], and EGFR mutations may be less provided a detailed insight of their genomic landscape and
common as well [56]. Another pattern that may be seen revealed the major molecular alterations that may contrib-
is termed “GBM with oligodendroglioma component” ute to disease pathobiology and progression [18, 23, 166].
(GBM-O), where the tumor, at least regionally, appears While many of the findings from TCGA were confirmatory
similar to anaplastic oligodendroglioma. These tumors are and relied on the foundation set by prior studies, insights
easily distinguished from GBM by the presence or absence gained from TCGA data are partially based on the ability
of 1p/19q co-deletion, which while controversial, in our to integrate data from diverse molecular platforms (mRNA,
view effectively defines this differential diagnosis as glio- miRNA, DNA copy number, mutational data, protein
blastoma versus oligodendroglioma. Specifically, GBM-O expression, DNA methylation) on a focused set of tumor
is distinguished from anaplastic oligodendroglioma (AO) samples. TCGA and other large-scale analyses have dem-
by the absence of 1p/19q deletion, and the presence of IDH onstrated that GBM, as histologically defined, is a hetero-
mutation and 1p/19q deletion effectively defines AO and is geneous tumor type at the molecular level and is potentially
therefore incompatible with the diagnosis of GBM-O. To sub-classifiable into distinct biologic entities based on
put it a different way, high-grade gliomas with IDH muta- molecular pathogenesis and “driver” lesions (i.e., molecu-
tion and whole-arm 1p/19q co-deletion should in our view lar changes that are required for tumorigenesis and pro-
be classified as AO grade III. For further discussion on this gression). While such comprehensive genome-wide stud-
point, the reader is referred to the companion article on ies have provided useful insights for the characterization
oligodendroglial tumors in the cluster in this issue of Acta and classification of tumors, their experimental limitations
Neuropathologica. need to be taken into consideration when drawing conclu-
sions. Such limitations include the retrospective nature of
the experimental design and the fact that patients involved
Integrated genomic analysis of GBM in these studies were not uniformly treated. In addition, the
impact of patient selection with respect to tumors with suf-
Traditionally, GBM is separated into 2 major classes as ficient material for multidimensional profiling is unknown,
“primary” and “secondary” GBM. Primary GBM was sug- as is the potential for bias from the fact that samples were
gested as generally presenting without a known clinical derived primarily from academic oncology centers. Fur-
precursor, while secondary GBM was a result of molecu- thermore, although some prognostic markers are emerg-
lar progression and increased malignancy grade of a lower ing from these studies, there is a great demand for the
grade glioma over time. Ongoing and recent advances identification of bona fide predictive markers that would
have demonstrated molecular correlates of these clinical improve the treatment process for personalized care and

13
 Acta Neuropathol

these markers await identification. At this point, translation

of key novel biomarkers discovered by initiatives such as
TCGA into clinically useful tests is yet to be fully accom-
plished. That said, these efforts have led to improved under-
standing of the molecular signature of GBM and other
diffuse gliomas and have revealed a number of consistent
alterations in genes and pathways, including mutations in
specific genes, modified pathway component expression
signatures, and altered DNA methylation patterns [23, 117,
129, 166], but point to the still unmet need to incorporate
these findings into the clinic to identify predictive markers
to improve outcome for patients with GBM.

Common pathways disrupted in GBM

In the past two decades, a large number of recurring molec-

ular alterations have been identified in gliomas and particu-
larly in GBMs, which enable characterization of diffuse
gliomas and better understanding of glioma landscape and
pathways that are disrupted in this malignancy. In the ini- Fig. 1  Alterations in the RTK/RAS/PI3 K signaling pathway in
tial TCGA report, Sanger sequencing was combined with GBM. Several genes that encode proteins involved in the RTK/RAS/
PI3 K signaling pathway are considerably altered in GBM. Genes that
array-based platforms to analyze alterations in 601 genes are most frequently amplified in this pathway are epidermal growth
from 91 samples. This study investigated the gene expres- factor receptor (EGFR) and platelet-derived growth factor receptor α
sion, DNA methylation, DNA copy number, in addition to (PDGFRA), two transmembrane receptors with tyrosine kinase activ-
coding and non-coding RNA expression profiles. Results ity. The most commonly deleted gene in the RTK pathway is phos-
phatase and tensin homolog (PTEN), a tumor suppressor that inhibits
from TCGA studies and contributions made by individual phosphatidylinositol-3 kinase (PI3 K) signaling such as retinoblas-
labs have revealed a number of genetic abnormalities and toma (RB1), a cell cycle inhibitor of PARK2, a regulator of dopamin-
as a result, specific patterns have emerged that suggest the ergic cell death, and neurofibromin 1 (NF1), a negative regulator of
the RAS signal transduction pathway. The most commonly mutated
involvement of specific molecular and signaling pathways
genes in this pathway are PTEN, NF1, EGFR, and PIK3R1, and
in the development and progression of glial tumors. These PIK3CA. This figure was adapted from The Cancer Genome Atlas
include loss of CDKN2A, RB1, and TP53 tumor suppressor Research Network [23]
genes, in addition to alterations in genes involved in these
pathways or regulated by these tumor suppressor proteins
[30, 63, 100, 131, 171]. Mutations in the IDH1, ATRX, and (phosphorylates and inactivates Rb) and Mdm2 (p53 inhib-
p53 genes are considered molecular hallmarks of diffuse itor), are often up-regulated by gene amplification, suggest-
and anaplastic astrocytomas (WHO grades II and III) as ing the involvement of alternative mechanisms for disrup-
well as secondary GBMs [27, 77, 100, 106], and interest- tion of p53 and Rb signaling pathways, as observed in the
ingly, TP53 mutations also occur in nearly all instances of majority of GBMs [23].
the rare giant cell GBM variant [110]. Integrated genomic In addition to alterations in tumor suppressive path-
studies have revealed that in the majority of GBMs, the ways, activation of oncogenic pathways such as those
functions of p53 (87 % of GBM patients) and retinoblas- involving receptor tyrosine kinases (RTKs) are well
toma (Rb) (78 % of GBM patients) pathways are disrupted known to be one of the most common genetic alterations
either by mutations or gene copy number alterations [23]. In in malignant gliomas (Fig. 1). The role of these drivers of
addition, mutations in genes encoding upstream regulators glioma has been demonstrated and their importance was
of Rb, but not necessarily the RB1 gene itself, have been revealed in a number of studies using mouse models. In
known for some time to be characteristic of gliomas [23, adult GBMs, high-level genomic amplification (~40 %)
63, 123]. For example, in a fraction of anaplastic gliomas occurs in the EGFR gene, often along with constitutively
and particularly in GBMs, the CDKN2A gene is homozy- activating mutations in this protein’s ectodomain mainly
gously deleted; CDKN2A locus encodes both Ink4A and through the variant III (vIII) deletion event [46, 93, 114,
Arf proteins, which are crucial activators of Rb and p53, 123, 178]. Although the changes leading to the EGFRvIII
respectively [22, 123, 126, 163]. In addition, upstream mutation are complex and heterogeneous, they are consid-
repressors of p53 and Rb signaling pathways, such as Cdk4 ered late events following amplification of EGFR. Overall,

13
Acta Neuropathol

EGFRvIII is found in approximately 30–50 % of glioma between NF1 gene alteration and the mesenchymal GBM
cases in which EGFR amplification is present. Histopatho- subclass (see below).
logically, the pattern recognized as “small cell” GBM is The identification of point mutations in codon 132 of
enriched for EGFRvIII-positive tumors [128]. In addition, isocitrate dehydrogenase I (IDH1) (and less commonly
in a smaller proportion of adult GBMs (~13 %), high-level codon 172 of IDH2) in gliomas has provided a fundamen-
amplification of the platelet-derived growth factor receptor tal new insight into our understanding of the biology, as
alpha gene (PDGFRA) has also been detected [114]. Simi- well as the molecular classification of these tumors [123].
larly, constitutively activating deletion mutants in PDG- Such mutations are frequent in WHO grade II and III dif-
FRA have been demonstrated in receptor-amplified GBMs fuse gliomas (70–90 %) and secondary GBMs (85 %),
[121]. PDGFRA gene amplification appears to be a com- but are rarely found in patients with traditionally referred
mon genomic alteration in the RTK pathway, exerting a to as “primary” GBMs (5 %) [58, 181]. While the distinc-
significant impact on pediatric GBMs and diffuse intrinsic tion between primary and secondary pathways to GBM
pontine gliomas (DIPG) [125, 126, 182]. Although much was originally based on the different clinical history, it has
less frequent in GBMs, high-level amplification of the become evident that both are molecularly distinct GBM
MET proto-oncogene has also been shown [114, 123, 126]. entities with absence or presence of IDH1/2 mutations
More importantly, activating genetic alterations can occur being the most important molecular discriminator. Further-
simultaneously in multiple RTKs within individual GBMs, more, IDH1/2 mutations are generally found to positively
with distinct cellular subpopulations containing amplified correlate with other genetic abnormalities common to dif-
receptor genes [151, 158]. This finding suggests that the fuse gliomas such as TP53 and ATRX mutations in astro-
targeting of single RTKs in an effort to neutralize onco- cytoma and 1p/19q co-deletion in oligodendroglial tumors,
genic signaling may in some cases prove futile, and drugs while they display an inverse correlation with EGFR gene
targeting multiple RTKs activated in GBM may confer amplification and monosomy of chromosome 10, altera-
greater treatment efficacy in some settings as opposed to tions that more commonly occur in primary GBMs [181].
drugs targeting single RTKs. Finally, although high-level Therefore, the molecular pathways that lead to the devel-
amplification of RTK genes is not frequent in WHO grade opment of low-grade gliomas and secondary GBMs that
II and III gliomas, their pathogenesis is often associated they evolve into are clearly distinct from those giving rise
with elevated PDGF signaling and PDGFRA phosphoryla- to primary GBMs. On this point, the prior designation of
tion [33, 52]. These findings suggest that the lack of suc- “primary GBM” is likely misleading, since IDH wild-type
cess in anti-EGFR treatment trials of GBM may be in part lower grade gliomas—especially anaplastic astrocyto-
due to the high degree of heterogeneity and complexity of mas—are often genomically identical and likely represent
RTK biology in gliomas. precursors to IDH wild-type GBM. Therefore, “primary”
The majority of GBMs exhibit activation of the extended GBM likely undergo molecular evolution from lower grade
PI3 K–AKT–mTOR and RAS–MAPK signaling pathways lesions (Brat et al. Comprehensive, integrative genomic
[114] and these are therefore, considered to be common analysis of diffuse lower grade gliomas, in press). As a first
oncogenic alterations in these tumors. Deregulating muta- pass, while GBMs with IDH1/2 mutations are relatively
tions in these pathways include mutations in genes encod- uncommon, IDH1/2 mutant secondary GBMs represent a
ing either the catalytic (PIK3CA) or regulatory (PIK3R1) completely different biologic entity compared to the major-
domains of PI3 K, which in turn induce the activity of ity of GBMs which do not harbor mutations in IDH1/2, i.e.,
these enzymes (~15 % of adult GBMs), as well as deletions most primary GBMs. In addition, anaplastic astrocytomas
and/or silencing mutations in PTEN, the primary negative that are IDH wild type are for practical purposes best con-
regulator of the PI3 K-AKT signaling pathway (~30 % of sidered as GBM, since these tumors show genomic hall-
cases). Beyond genetic alterations of PTEN, additional epi- marks of GBM (loss of chromosome 10, gain of chromo-
genetic and miRNA-based regulation of PTEN repression some 7, and EGFR amplification) and clinically behave as
have also been described in diffuse gliomas, although they GBMs.
are more common in WHO grade II and III gliomas (50–
60 %) [52, 71, 82, 108, 175]. Mutations in the Ras antag-
onist protein neurofibromin 1 (NF1) are thought to be the Transcriptional subtypes of GBM
cause of neurofibromatosis type 1, a cancer predisposition
syndrome mainly characterized by frequent neurofibro- The availability of high-throughput genomic platforms
mas and astrocytomas [55]. Recent studies, however, have for mRNA expression profiling since the late 1990s has
demonstrated NF1 somatic gene mutation or deletion in resulted in some experience and published data attempting
15–18 % of “primary” GBMs [23, 123], and a major con- to identify patterns of gene expression and to codify these
clusion from the TCGA effort was demonstration of a link into subtypes, with subsequent correlative studies layering

13
 Acta Neuropathol

additional genetic and genomic aberrations. Initially, work

on gliomas focused mainly on high-grade tumors such as
GBM and comprehensive transcriptional analysis was used
to identify molecular correlates for known clinical and/or
pathological distinctions, such as the corresponding WHO
grade, primary versus secondary GBM, and astrocytic ver-
sus oligodendroglial morphology [49, 80, 83, 96, 133, 137,
146, 160, 162]. Subsequent profiling studies have success-
fully identified distinct molecular signatures for diffuse gli-
omas and revealed specific subclasses within GBMs.
In 2006, Phillips et al. [129] examined differential
Fig. 2  Transcriptional subtypes of glioblastomas based on Phillips
expression of markers associated with clinical outcome and and Verhaak classification. Gene expression-based molecular clas-
used K-means clustering to delineate gene signatures in sification of GBM into proneural, neural, classical, and mesenchy-
WHO grade III and IV diffuse gliomas. Three major sub- mal subtypes by Verhaak et al. and into proneural, proliferative, and
mesenchymal subtypes by Phillips et al. Integrated genomic analysis
classes of GBM emerged based on this analysis: proneural,
demonstrate patterns of somatic mutations and DNA copy number
mesenchymal, and proliferative. This classification bares alterations. Aberrations in EGFR, NF1, and PDGFRA/IDH1 genes
similarity to earlier sub-classification of prognostically rel- each define the classical, mesenchymal, and proneural subtypes,
evant high-grade gliomas [47]. In addition, markers associ- respectively
ated with one of the major subtypes (the mesenchymal sig-
nature), including YLK40 and VEGF, had previously been
applied to distinguish GBMs from lower grade gliomas [48,
51, 137]. In its initial description, the proneural signature
was shown to be associated with a better outcome (although
later this was discovered to be confounded by the fact that
IDH1/2 mutant gliomas invariably appear to be proneural)
and expresses marker genes associated with neurogen-
esis. In a study by Aiguo et al. unsupervised analysis of
transcriptome profiles from 159 glioma samples predicted
two major groups of gliomas (oligodendroglioma-rich and
GBM-rich) that were further separable into six hierarchi-
cally nested subtypes [95]. The initial TCGA expression
profiling report described four GBM subtypes termed
proneural, neural, classical, and mesenchymal [166]. This
study also documented genomic associations, with classi-
cal, proneural, and mesenchymal tumors strongly enriched
for aberrations in EGFR, PDGFRA and IDH1 or IDH2, and
NF1 genes, respectively (Fig. 2).
The proneural subtype is mainly described by mutations
in PDGFRA or in IDH1/2, whereas the classical subtype Fig. 3  Progression of IDH-wild type and IDH-mutant gliomas. IDH-
is characterized by amplification/mutation of the EGFR mutant gliomas (right) go through an ordered sequence of genetic
modifications. Upon acquisition of IDH1/2 mutations and hypermeth-
gene, and mutations in neurofibromin 1 (NF1) are mainly ylation of CpG islands (CIMP) in the glial progenitor cells, a subset
found in the mesenchymal subtype. The proneural GBM of these cells acquires secondary mutations in TP53 and ATRX, which
is further subdivided into glioma CpG (G–CIMP)-positive result in the development of astrocytomas and eventual progression
and -negative subgroups based on the characteristic DNA to ‘secondary’ glioblastoma. Co-deletion of 1p and 19q occurs in the
other subset of glial cells, along with TERT promoter mutation, lead-
methylation patterns that are directly linked to the IDH1/2 ing to formation of oligodendrogliomas. IDH-wild type gliomas (left)
mutational status [117]. The mesenchymal signature is progress via acquisition of different molecular alterations and most
mainly regulated by the expression of the transcription commonly present as glioblastoma. However, the designation as ‘pri-
factor signal transducer and activator of transcription 3 mary’ glioblastoma may not be entirely accurate, as IDH-wild type
lower grade astrocytoma, although not common, is well-described
(STAT3), CCAAT/enhancer-binding protein-β (C/EBPβ),
and transcriptional co-activator with PDZ-binding motif
(TAZ), which have also been associated with poor clinical as negative and positive genetic regulators of mesenchymal
outcome [15, 45]. Furthermore, recent studies have shown transformation, respectively [32, 45]. Whether the proneu-
that CTNND2 (encoding catenin-δ2) and RHPN2 function ral and mesenchymal signatures, as well as the other

13
Acta Neuropathol

transcriptional subtypes of GBMs, could serve as predic- number of physiological processes such as hypoxia sens-
tors of patient outcome is being investigated by a number ing, histone demethylation, and changes in DNA methyla-
of groups. While overall the proneural subclass is associ- tion, among others [98]. A distinctive and nearly invari-
ated with better outcome, a finer examination suggests that able feature of IDH1/2 mutant gliomas is the glioma CpG
proneural GBMs can be subdivided into IDH1/2 mutant island methylator phenotype (G-CIMP) [117]. Baysan et al.
G-CIMP-positive and -negative subsets and once IDH1/2 [12] applied unsupervised clustering of TCGA methyla-
G-CIMP status is controlled for, the proneural class has no tion data from 368 GBM samples, showing that G-CIMP-
prognostic advantage compared to other IDH1/2 wild-type positive expression signatures were linked with mutant
GBMs [18]. IDH1 expression and correlated with better prognosis. For
While the two major subclasses, proneural and mesen- the detection of IDH1/2 mutant gliomas—where approxi-
chymal, appear to be reproducibly defined and character- mately 15 % of cases are not detected using the IDH-
ized and may describe important biology, the implemen- R132H antibody—DNA sequencing of antibody-negative
tation of these gene expression signatures into clinical cases has provided more accurate diagnosis and prediction
diagnosis has not been accomplished. Indeed, subclass of patient outcome and prognosis. This is especially use-
assignment has been shown to be unstable and change ful in younger GBM patients, as IDH1/2 mutation is more
following surgical resection and radiochemotherapy [14, common in this patient group.
129]. Moreover, a recent study analyzing expression sig- In addition to providing insights about the origin of
natures of single cells within GBM samples showed sub- gliomas, the mutational status of IDH1/2 serves as a prog-
stantial intratumoral heterogeneity of expression subclasses nostic marker in patients with WHO grade II and III glio-
within each tumor [124]. Based on these considerations, mas [58, 123, 141, 181] and GBMs [172]. While IDH1/2
the mRNA expression profile of glial tumors may represent wild-type GBMs (as well as most anaplastic gliomas that
an average of a heterogeneous mix of transcriptional sig- do not have an IDH mutation) exhibit a pattern of genetic
natures and therefore, alternative aberrations, including the changes that are associated with primary GBMs—such
genomic and epigenomic profiles of tumors may represent as gain of chromosome 7, loss of chromosome 10, and
more stable metrics for tumor classification. EGFR amplification—this pattern is not characteristic of
IDH1/2 mutant GBMs. Unresolved issues remain, related
to understanding of the specific driver molecular changes
Clinically relevant genetic and epigenetic in IDH1/2 wild-type GBM, made complex by the fact that
abnormalities in GBM some of the prototypical changes include gains and losses
of whole chromosomes or chromosomal arms (e.g., losses
Isocitrate dehydrogenase 1 and 2 (IDH1/2) genes of chromosome 10 and the short arm of chromosome 9,
and gain of chromosome 7). Intertwined within this issue
One of the most important discoveries resulting from high- is the fact that the expression subtypes within IDH1/2 wild-
throughput genomic studies, which has led to remodeling type GBM (e.g., proneural versus mesenchymal) often have
of our understanding of gliomas including GBMs, was largely similar genomic changes, with the exception of the
the identification of mutations in the metabolic enzymes classical subtype which harbors large-scale amplification
isocitrate dehydrogenase 1 and 2 (IDH1/2) [123]. In this of EGFR at the genomic level. One recent study attempted
landmark paper, the majority of tumor samples bearing to address this issue in an interesting manner by using
this mutation (5/6) were classified as secondary GBMs, mathematical modeling of genomic changes in IDH1/2
suggesting that IDH1/2 mutation could serve as a genetic wild-type/G-CIMP-negative GBM using data available
marker for this GBM type. Mutant IDH1/2 alleles identi- from TCGA coupled with experimental mouse modeling.
fied in gliomas result in enzymes with a neomorphic func- Their results suggested that gain of PDGFA (chromosome
tion [31], whereby the mutant enzymes have acquired the 7) and loss of PTEN (chromosome 10) are likely initial
ability to catalyze the NADPH-dependent reduction of driver events, and that a hierarchy of expression subtypes
α-KG to the (R)-enantiomer of 2-hydroxyglutarate (2-HG), exists. It is also likely that PDGFA drives a proneural phe-
that is the same stereoisomer of 2-HG seen in D-2-HG. notype, which can be followed by loss of NF1 function
In fact, Dang et al. [31] showed that IDH1/2 mutant cells to promote a subsequent mesenchymal phenotype [122].
had high levels of 2-HG, as is also found in primary IDH1 Taken together, the collective data clearly show, although
mutant gliomas and in the serum of IDH1/2 mutant acute histologically similar, IDH1/2-mutant and wild-type GBMs
myeloid leukemia (AML) patients [54, 169]. are clearly distinct diseases on a genomic basis and under-
It is thought that expression of mutant IDH1/2 proteins standing the biological contribution of these mutations
results in inhibition of α-KG-dependent dioxygenases may help in the diagnosis and design of treatment strate-
by 2-HG. Enzymes that are α-KG dependent regulate a gies (Fig. 3). On this point, the development of therapies

13
 Acta Neuropathol

specific for mutant IDH1/2 appears to be a practical lead the cytokine co-receptor gp130, and thereby, induce the
for molecularly driven therapies and show promising pre- expression of wild-type EGFR in the neighboring cells
clinical results, either as a small molecular inhibitor or as a [73]. Therefore, intratumoral heterogeneity and cooperativ-
vaccine approach targeting the R132H protein as a tumor- ity may be the key for EGFRvIII function in GBM. Altera-
specific neoantigen [135, 144]. Additional clinical devel- tions in the EGFR gene have been found in other cancer
opment of these approaches, including plans to address types such as non-small-cell lung cancer (NSCLC), but
blood–brain barrier penetration of these targeting agents, is the type of genetic alterations found in EGFR in GBM
likely to yield important information and hopefully thera- are distinct from those associated with other cancers. For
peutic advances in the coming years. example, focal EGFR amplification occurs at an extremely
high rate in gliomas (>20 copies) and the majority of other
Epidermal growth factor receptor (EGFR) mutations—such as the EGFRvIII mutation and missense
and EGFRvIII mutations—are located in the extracellular domain [23,
93], while in most non-glioma cancers they are found in
Approximately 40 % of primary GBMs carry amplification the intracellular domain [75]. It should be noted that EGFR
of the EGFR gene [70, 76, 143]. In addition, about 50 % of amplification and EGFRvIII expression may not persist in
GBMs with EGFR amplification also harbor a mutation in cultured cells as in primary tumors, but recent studies have
this gene that codes for EGFRvIII—a constitutively active successfully passaged EGFRvIII-expressing GBM xeno-
variant of EGFR that is supposed to promote tumor growth grafts both in vivo as well as in vitro by growing them in
and is potentially associated with a worse clinical outcome stem cell culture conditions [153]. Therefore, long-term
[86, 87]. Interestingly, established prognostic factors in EGFRvIII expression may in fact be possible and is associ-
GBM, e.g., the Radiation Therapy Oncology Group’s recur- ated with differentiation and/or the developmental stage of
sive partitioning analysis (RTOG-RPA) class, were not pre- the tumor.
dictive of outcome in EGFRvIII-positive GBMs [127]. The The EGFRvIII mutation has become clinical rel-
EGFRvIII mutation involves an intragenic gene rearrange- evance as this deletion mutation generates a novel pep-
ment that is generated by an in-frame deletion of exons tide sequence that may serve as an immunogenic tumor-
2–7, which encode part of the extracellular domain of this specific target, which can be exploited in a peptide-based
protein [40, 62, 147]. A number of studies have shown that vaccination strategy. Initial results from single-arm trials
ectopic overexpression of EGFRvIII in glioma cell lines employing EGFRvIII-specific vaccination provided prom-
results in constitutive autophosphorylation and activa- ising results in comparison to historical controls [138].
tion of the Shc–Grb2–Ras and class I PI3 K pathways [69, The efficacy of EGFRvIII-targeted vaccination in newly
113], induces tumorigenicity [69], cell proliferation [113], diagnosed GBM patients is currently being investigated in
and resistance to apoptosis through modulation of Bcl- the prospective randomized ACT IV trial (EUDRA-CT#:
XL gene expression [111]. Interestingly, the tumorigenic 2011-006068-32).
effects of EGFRvIII overexpression are not recapitulated
by overexpression of the wild-type EGFR. Furthermore, Platelet‑derived growth factor alpha (PDGFRA)
both EGFRvIII and wild-type EGFR proteins have been
detected in the nucleus and are thought to drive transcrip- In approximately 30 % of human gliomas, expression of
tional and signaling pathways that contribute to cell prolif- genes associated with platelet-derived growth factor recep-
eration and DNA damage repair [168]. Notably, although tor (PDGFR) signaling and genes involved in oligoden-
EGFRvIII is well known to induce cell proliferation, it is drocyte development (OLIG2, NKX2-2, and PDGF), are
only expressed in a fraction of GBM cells [116]. A num- observed and are thought to be hallmarks of the proneural
ber of recent studies have suggested a model for functional signature in GBM [17]. Amplification of the alpha-type
heterogeneity, where a small number of EGFRvIII-positive PDGFR (PDGFRA) gene is found in 15 % of all tumors,
cells not only drive their own proliferation but also enhance mainly in the proneural subtype of GBM [129, 166] and
the proliferation of their neighboring cells that express approximately 40 % of tumors harboring gene amplifica-
wild-type EGFR. In a study conducted by Inda et al. it was tion contain an intragenic deletion in this gene, termed
found that wild-type EGFR-expressing cells exhibit accel- PDGFRAΔ8,9 [28], where in-frame deletion of 243 base
erated proliferation due to a paracrine mechanism driven by pairs of exons 8 and 9 results in a truncated extracellular
EGFRvIII-expressing cells. They demonstrated that human domain [121]. In addition to this deletion, in-frame gene
glioma tissues, glioma cell lines, glioma stem cells, and fusion of the extracellular domain of KDR/VEGFR-2 and
immortalized mouse Ink4a/Arf (−/−) astrocytes express- the intracellular domain of PDGFRA has also been found,
ing EGFRvIII also expressed cytokines such as IL-6 and/ and both of these mutant proteins were shown to be con-
or leukemia inhibitory factor (LIF), which in turn activate stitutively active, display transforming ability and could be

13
Acta Neuropathol

inhibited using inhibitors of PDGFRA. Point mutations in models of NF1 [170]. Other studies using genetically
PDGFRA have also been detected but are generally rare engineered mouse models have shown that loss of NF1 in
[23]. In addition, PDGFR signaling can be activated upon glial cells, in combination with a germ line TP53 muta-
up-regulation of PDGF ligands (A–D) in approximately tion, results in astrocytomas [183] and further progress to
30 % of glioma tumor samples and cell lines. The expres- GBM upon deletion of PTEN [90]. More recent work has
sion of PDGFRB, however, seems to be limited to prolifer- revealed that the same combination of genetic alterations
ating endothelial cells in GBM [33, 43, 64, 99, 150]. in these tumor suppressor genes in neural stem/progenitor
Similar to EGFR and EGFRvIII, amplification of PDGF cells is necessary and sufficient to induce astrocytoma for-
and PDGFR seems to promote aggressive glioma growth. mation [1]. Loss of NF1 gene function has been implicated
Assanah et al. [3, 4] demonstrated that transduction of cells in the development of the mesenchymal phenotype for
of the subventricular zone (SVZ) of the lateral ventricle of GBM [166]. These findings emphasize the heterogeneity
neonatal rats with a retrovirus expressing PDGF yielded and the contribution of cell type-specific effects of various
large, diffuse tumors that resembled GBM. They found that genetic alterations to the development of GBM.
in these tumors, both infected and uninfected PDGFRα+-
expressing progenitors massively proliferated, suggesting MGMT promoter methylation
that PDGF expression leads to tumor formation through
both autocrine and paracrine signaling mechanisms, driv- Promoters of several genes at specific loci are hypermeth-
ing the evolution of heterogeneous malignant gliomas. ylated in GBM and frequently result in altered expression
These results raise the possibility that cells distinct from of tumor suppressor genes, such as cyclin-dependent kinase
the initially transformed cells of origin within the tumor inhibitor 2A (CDKN2A), RB1, PTEN, and TP53, among
environment can eventually become tumor cells and sug- others [2, 6, 29, 112]. One of the clinically most impor-
gest a model of glioma evolution that is different from the tant DNA methylation markers in GBMs is the promoter
generally accepted view of linear gliomagenesis [44]. of MGMT (encoding O6-methylguanine-DNA methyltrans-
ferase), which is found in approximately 40 % of primary
Neurofibromatosis type 1 gene (NF1) GBM patients and is associated with transcriptional silenc-
ing of the MGMT gene. Hypermethylation of the MGMT
Large-scale sequencing analysis by the TCGA has shown promoter was demonstrated to serve as a predictive marker
that in approximately 15 % of glioma samples the NF1 for alkylating chemotherapy in GBMs [61, 174]. MGMT
gene is inactivated by genetic loss or mutation [123], and is a DNA repair enzyme and modulation of sensitivity to
NF1 mutations are most common in the mesenchymal sub- alkylating agents can be explained by the ability of this
type of GBM [166]. Inactivation of NF1 protein can also enzyme to restore guanine from O-6-methylguanine, which
arise from excessive proteasomal degradation mediated is the type of genomic lesion induced by alkylating agents
by hyperactivation of PKC [23, 109]. Neurofibromin 1 is used for chemotherapy drugs such as temozolomide (TMZ)
the product of NF1 gene and is a tumor suppressor that (Fig. 4). A number of clinical trials and cohort studies have
negatively regulates Ras and mTOR signaling pathways shown that promoter methylation of the MGMT gene is
in astrocytomas. In fact, experiments using NF1-deficient associated with prolonged progression-free and overall sur-
primary murine astrocytes have revealed that loss of NF1 vival in patients who were treated with alkylating agents
causes increased cell proliferation and migration that is [41, 60, 61, 65, 172].
dependent on hyperactivation of mTOR mediated by Ras A seminal trial conducted by the EORTC examined con-
signaling. In this setting, mTOR induces rapamycin-sen- current/adjuvant TMZ treatment during and after radiother-
sitive activation of Rac1 GTPase, independent of elonga- apy compared to radiotherapy alone for newly diagnosed
tion factor 4E-binding protein 1(4EBP-1)/S6 kinase (S6 K) GBM patients [155]. While the trial was overall positive,
[140]. Stat3 is another downstream target of NF1 that is analysis of a subset of samples in this trial showed that
regulated in an mTORC1- and Rac1-dependent manner and patients with glioma tumors harboring MGMT promoter
increases cyclinD1 expression [9]. methylation benefited from chemotherapy almost exclu-
Using genetically engineered mouse models, it was sively [61]. Similar results have been found in elderly
found that targeted homozygous loss of NF1 in astrocytes patients, showing improved outcome with chemotherapy
is not sufficient to induce tumor formation, although it is treatment in MGMT promoter-methylated tumors, while
sufficient to increase cell growth both in vitro and in vivo worse survival was associated with unmethylated tumors
[8]. Furthermore, NF1−/− astrocytes were shown to develop [105, 174], suggesting that MGMT promoter methylation is
optic gliomas in NF1+/− brains of mice [7, 184] and low not a prognostic, but instead a predictive marker. Additional
levels of cAMP expression in the stroma cause induction work has reconfirmed the predictive value of MGMT pro-
of optic glioma formation in genetically engineered mouse moter methylation for response to chemotherapy in IDH1/2

13
 Acta Neuropathol

patients are often treated with TMZ regardless of methyla-

tion status. However, MGMT methylation status may be
useful in these patients to distinguish pseudoprogression
(PsPD) from true progression [16]. PsPD is a pathological
phenomenon in malignant glioma patients that are treated
with combination radiotherapy and TMZ. PsPD generally
occurs within a few months from radiochemotherapy and
appears as an increase in tumor size in radiological imag-
ing; however, it is not accompanied with worsening of the
neurological signs and symptoms. PsPD was recorded in
21 (91 %) of 23 patients with methylated MGMT promoter
and 11 (41 %) of 27 patients with unmethylated MGMT
promoter (P = 0.0002). In pediatric gliomas, both the fre-
quency (16–50 %) [20, 35, 94, 152] and the prognostic or
predictive significance of MGMT silencing remain to be
determined [35, 94].
Epigenome-wide analysis of DNA methylation pat-
terns in glioma tumors has improved our understanding
of glioma biology and has contributed to the advance-
ment of tumor classification [117, 156]. Recently, algo-
rithms have been developed that enable assessment of the
Fig. 4  MGMT promoter methylation as a predictive marker for TMZ three biomarkers, 1p/19q co-deletion, G-CIMP status,
treatment. TMZ is an oral alkylating agent used as a chemotherapeu- and MGMT promoter methylation, using Illumina Infin-
tic treatment for GBMs. TMZ causes DNA lesions such as O6-meth-
ylguanine (O6-meG), and N3-methyladenine and N7-methylguanine ium HumanMethylation450 (450 K) data [5, 117, 156].
(N3-meA, N7-meG). O6-meG DNA methyltransferase (MGMT) Hybridization of tumor DNA to these arrays allows one to
restores the guanine to normal by removing the O6-alkylguanine, profile methylation of up to 450,000 CpG sites distributed
and thereby, promoting tumor cell survival. MGMT function may be across the human genome as well as analyzing genome-
impaired by gene deletion or suppression of its expression by hyper-
methylation of its promoter. Specifically in glioblastomas, IDH1/2 wide copy number changes [5, 68, 156]. In addition, this
mutations cause the CpG island methylator phenotype (CIMP) which method is suitable for analysis of formalin-fixed and par-
may involve MGMT methylation as part of this phenomenon. Loss of affin-embedded (FFPE) tissue samples [68]. Wiestler et al.
MGMT-mediated DNA repair may lead to DNA strand breaks, apop- [177] assessed the reliability and value of this technology
tosis, autophagy, and tumor cell death
and demonstrated its diagnostic and prognostic accuracy
in determining G-CIMP, 1p/19q co-deletion, and MGMT
wild-type GBMs, while this marker is prognostic, albeit promoter methylation status in the biomarker cohort of the
more commonly, in IDH1/2 mutant anaplastic gliomas prospective NOA-04 trial. Further optimization and eluci-
[165, 173]. It is important to note that there is extensive dation of MGMT methylation testing may yield additional
overlap between MGMT methylation status and G-CIMP, clinical relevance of this important biomarker.
and while MGMT methylation is present in a subset of
G-CIMP-negative GBMs, it is found in almost all cases of hTERT promoter mutation
G-CIMP-positive tumors [5]. Importantly, MGMT promoter
methylation is a clinically important predictive marker for Human telomerase is a ribonucleoprotein that regulates
guiding adjuvant therapy in elderly GBM patients [61, 105, the length of telomeric DNA at the ends of chromosomes
119, 174]. In this patient group, MGMT methylation sta- and therefore, plays an important role in cellular immor-
tus has emerged as a predictive marker to determine best talization and oncogenesis. One of the hallmarks of cancer
therapy and inclusion of TMZ. In this setting, the MGMT is deregulation of telomere maintenance and this process
promoter methylation status helps to stratify patients into is regulated by the enzyme telomerase, which is active in
those who should be treated with radiotherapy only, i.e., 90 % of all advanced cancers. Telomerase reverse tran-
patients with MGMT promoter unmethylated tumors, and scriptase (TERT) is the catalytic subunit of the telomer-
those who should be treated with TMZ chemotherapy or ase complex and its expression is associated with poor
combined TMZ/radiotherapy, i.e., patients with MGMT outcome in most tumors such as breast cancer, sarcomas,
promoter-methylated tumors [61, 105, 119, 174]. How- and brain tumors [37, 50, 102, 139, 159]. Recent findings
ever, the importance of MGMT methylation testing in non- have established frequent mutations in the promoter of
elderly GBM patients remains a matter of debate, as these TERT in a number of cancer types, including melanomas,

13
Acta Neuropathol

liposarcomas, bladder cancer, and gliomas [53, 67, 81, 86, xanthoastrocytoma and one-third of gangliogliomas, and
167]. Interestingly, genomic analysis of gliomas has shown occasional pilocytic astrocytoma [142]. In GBMs, BRAF
that TERT promoter mutations occur in 70–80 % of pri- V600E mutations have been detected in approximately
mary GBMs and in more than 70 % of oligodendrogliomas, 5 % of the cases [85, 142]. A higher frequency of BRAF
but are less frequent in IDH1/2 mutant diffuse and ana- mutation has been reported in GBMs with histological fea-
plastic astrocytomas as well as IDH1/2 mutant (secondary) tures of epitheloid differentiation, i.e., “epitheloid GBMs”,
GBMs that instead carry frequent ATRX mutations [11, 72, which preferentially manifest in children and young adults
115]. TERT promoter mutations are also rare in pediatric and carry BRAF V600E mutations in more than 50 % of the
GBMs characterized by histone H3.3 (H3F3A) mutations, cases (7 of 13) [84]. While treatment of pediatric low-grade
which often are associated with TP53 and ATRX/DAXX astrocytoma patients with sorafenib, a multikinase inhibitor
mutations [160]. targeting BRAF, VEGFR, PDGFR, and c-KIT, resulted in
Recently, Killela et al. [81] assessed the association unexpected acceleration of tumor growth, even in patients
between IDH1/2 mutation and TERT promoter mutations with BRAF mutant tumors [79], a recent case report showed
across several glioma subtypes. The joint influence of complete regression of a BRAF V600E mutant pediatric
IDH1/2 mutation and TERT promoter mutation on over- GBM following treatment with the BRAF inhibitor vemu-
all survival (OS) was examined and three common glioma rafenib [134]. Thus, molecular testing for BRAF mutation,
subtypes were delineated; astrocytomas of WHO grade II either by DNA sequencing or by immunohistochemistry
and III, oligodendrogliomas of WHO grade II and III, and using a BRAF V600E-specific antibody [24], may uncover
GBMs. In general, TERT promoter mutations predicted a potentially active novel targeted therapy option in a small
poorer OS in GBMs without IDH1/2 mutations. Additional fraction of GBM patients.
studies by Simon et al. and Labussiere et al. demonstrated
that TERT promoter mutation signature could serve as a
novel independent prognostic factor for poor outcome in Comparison of molecular features of GBMS
primary GBMs. Their findings, however, suggest that the in pediatric versus adult patients
prognostic effect of TERT promoter mutation is independ-
ent of the mutation status of IDH1/2 in GBMs [91, 148]. Childhood GBM is much less common in absolute num-
On the other hand, TERT promoter mutations maybe asso- bers than the adult form; however, it is relatively a much
ciated with longer survival in patients with IDH1/2 mutant more frequent primary CNS tumor as a proportion of all
gliomas, as they are closely linked with the prognostically brain tumors (children 0–19 years: GBMs, 2.9 %, malig-
favorable 1p/19q co-deletion in oligodendroglial tumors [3, nant gliomas NOS, 11.7 %; all age groups: GBMs 15.4 %)
152]. Analysis of TERT promoter mutation also serves as [120]. The 2-year survival rate for GBM in children is
a novel prognostic marker for primary GBM patients and approximately 12 %, making this disease a leading cause
more recently, combined analysis of TERT promoter muta- of cancer-related deaths in children [19]. A number of stud-
tion, EGFR amplification, and IDH1/2 mutation has ena- ies have indicated that distinct genetic mechanisms play
bled identification of distinct classes of adult GBM [81]. a role in the pathogenesis of pediatric and adult GBMs
TERT promoter mutation testing may have a dual role [42, 57, 126, 157, 179] and although in-depth analysis
in molecular classification of gliomas based on IDH1/2 have demonstrated alterations in three key signaling path-
mutation status: within the IDH1/2 mutant tumors, TERT ways—including TP53, PI3 K/Akt, and Rb—and identified
mutation could possibly serve as a surrogate/confirmatory discrete transcriptional subtypes in adults, little is known
marker for 1p/19q co-deletion, as the two are highly corre- about alterations in these pathways in pediatric GBMs.
lated. For the purposes of IDH1/2 wild-type GBMs, TERT Two recent studies have attempted to identify somatic
mutation appears to be found in the majority of cases, but mutations specific to GBM patients who are younger than
those which do not have promoter mutations (referred to in 19 years of age at the time of diagnosis. These studies were
a recent report as “triple negative” (negative for all 3 mark- the first ones to discover somatic mutations in the histone
ers: IDH mutation, 1p/19q co-deletion, and TERT muta- H3.3-alpha-thalassemia X-linked mental retardation pro-
tion) may be clinically distinct from those GBMs which are tein (ATRX)–death domain-associated protein (DAXX)
“single-positive” (TERT-mutant only) [39]. chromatin remodeling pathway that lead to changes in the
chromatin architecture and play a major role in pediatric
BRAF mutation GBM pathogenesis in approximately 44 % of tumors [145,
180]. Recurrent somatic mutations in H3F3A, the gene
Activating missense mutations at the BRAF hotspot which encodes the replication-independent histone 3 (H3)
codon 600, most commonly the V600E, are common in variant H3.3, result in amino acid changes mainly in two
several neuroepithelial tumors, including pleomorphic residues within the histone tail; K27 M or G24R/G34 V.

13
 Acta Neuropathol

This mutation was found predominantly in GBM and was

more prevalent in children than adults; however, recent
findings point to approximately 5 % of adult GBM patients
also carrying this lesion [132] and likewise, H3F3A muta-
tion in adult GBM patients is associated with ATRX muta-
tions. In addition, somatic mutations in TP53 were found in
54 % of all cases and in 86 % of cases harboring mutations
in H3F3A and/or ATRX. The mutations in H3.3/ARTX/
DAXX/TP53 were also found to associate with changes
in the telomere lengthening and specific gene expression
profiles, suggesting that changes in the chromatin archi-
tecture contribute to the pathogenesis of childhood GBM.
Other studies that analyzed molecular profiles of pediatric
high-grade gliomas (HGG) have also suggested the exist-
ence of molecularly diverse subsets of pediatric GBMs [42,
Fig. 5  Major classes of glioma based on differences in pediatric or
57, 125, 126]. Another alteration found in pediatric GBMs adult genomic alterations. Somatic mutations in the histone H3.3/
is a higher amplification frequency of the PDGFRA gene ATRX/DAXX chromatin remodeling pathway are mainly found in
that is associated with activation of a PDGFRA-driven gene pediatric glioma patients. Higher amplification frequency of PDG-
expression signature [125, 126, 130]. FRA gene and frequent gain of chromosome 1q are also common in
pediatric gliomas. In adults, gain of chromosome 7 and loss of chro-
Another major difference between adult and pediatric mosome 10 are highly prevalent, in addition to TERT promoter muta-
GBM is the concomitant gain of chromosome 7 and loss tion, EGFR amplification, and IDH1/2 mutation
of chromosome 10 in most adult tumors (on average 84 %)
[18, 23]. Additional genomic abnormalities that occur at
higher frequency in adult than in childhood GBMs include differences between pediatric and adult tumors suggest
gains of chromosomes 19 and 20, and losses that affect that pediatric GBMs are in fact distinct entities on a bio-
chromosomes 9p, 22q, 13q, 14q, and 6q [18, 23]. Such logic level. Even though the histology overlaps between
chromosomal imbalances are generally found at lower pediatric and adult GBM, the genetic signatures indicate
frequency in childhood GBMs, and a proportion of these that these should not be “lumped” together into a single
tumors (~15 %) does not contain any detectable copy num- entity. In addition, recent data show that tumors morpho-
ber alterations [10, 78, 126, 156]. Pediatric tumors also logically classified as GBM in children actually represent
display more frequent gain of chromosome arm 1q com- very distinct subsets, based on molecular criteria (Illumina
pared to the adult counterparts, while they rarely harbor 450 k methylation profiling, DNA copy number analysis,
gain of chromosome 7 and loss of chromosome 10 (Fig. 5) and mutational analysis). Specifically, pediatric GBM that
[10, 126]. In addition, TERT promoter mutations occur at showed evidence of amplification a known oncogene and/
a much lower rate (3–11 %) in pediatric GBMs [81, 86], or K27 M mutation in histone H3.3 showed a particularly
which instead frequently display mutations in the H3.3/ poor prognosis, while tumors without evidence of these
ATRX/DAXX and consequent alternative lengthening of genetic lesions were prognostically more favorable, with a
telomeres (ALT) [59, 145]. With respect to DNA methyla- 3-year overall survival rate of approximately 70 % [87].
tion signatures in pediatric GBMs, Sturm et al. [156] per-
formed genomic DNA methylation profiling of 59 pediat-
ric and 77 adult tumors and identified distinct epigenetic Conclusion
GBM subgroups that were closely linked to specific genetic
alterations. One of the identified subtypes was the IDH1/2 Recently, aberrations in genes and molecular pathways in
mutant group, which is directly associated with global GBMs have provided a biological basis to establish appro-
hypermethylation (G-CIMP positive), while the H3F3A– priate clinically relevant biomarkers and point to the need
G34 group is linked to a hypomethylated signature of the for development of new therapeutic opportunities. We are
genome (G-CIMP negative). In light of the identification of at a point where progress in molecular classification of
distinct genetic and epigenetic differences between pedi- GBMs has provided useful insights for the development of
atric and adult GBMs, and the recently identified correla- more effective targeted therapeutics. Several clinically rel-
tion between these changes, it would be essential to fully evant molecular markers are well established and serve in
understand the molecular differences between the adult and the clinic as standard of care for patients diagnosed with
pediatric tumors to establish treatments specifically target- glioma tumors. For example, the status of MGMT promoter
ing GBMs in the two age groups. The substantial molecular methylation in GBMs (especially those detected in elderly

13
Acta Neuropathol

patients), 1p and 19q co-deletions in anaplastic oligoden- Infinium methylation BeadChip identifies two distinct CpG
drogliomas, and IDH1/2 mutations, now play major roles regions associated with gene silencing and outcome, yielding a
prediction model for comparisons across datasets, tumor grades,
in tumor diagnostics and/or clinical decision making [21, and CIMP-status. Acta Neuropathol 124:547–560. doi:10.1007/
105, 164, 174]. Meanwhile, multiplatform analyses of the s00401-012-1016-2
genetic, epigenetic, and transcriptional profiles have proven 6. Baeza N, Weller M, Yonekawa Y, Kleihues P, Ohgaki H (2003)
useful in refining the classification of brain tumors and pre- PTEN methylation and expression in glioblastomas. Acta Neu-
ropathol 106:479–485. doi:10.1007/s00401-003-0748-4
dicting patient outcome. Recent studies on pediatric GBM 7. Bajenaru ML, Hernandez MR, Perry A, Zhu Y, Parada LF,
have demonstrated that these tumors, which are frequently Garbow JR, Gutmann DH (2003) Optic nerve glioma in mice
driven by epigenetic changes in histone H3.3, may repre- requires astrocyte Nf1 gene inactivation and Nf1 brain het-
sent a 3rd major category of GBM, in addition to IDH1/2 erozygosity. Cancer Res 63:8573–8577
8. Bajenaru ML, Zhu Y, Hedrick NM, Donahoe J, Parada LF, Gut-
mutant (secondary) and IDH1/2 wild-type (primary) GBMs mann DH (2002) Astrocyte-specific inactivation of the neurofi-
in adults. With these molecular insights, it is hoped that fur- bromatosis 1 gene (NF1) is insufficient for astrocytoma forma-
ther improvements in molecular assays would bring them tion. Mol Cell Biol 22:5100–5113
to the clinic and be sought after as clinically indicated. 9. Banerjee S, Byrd JN, Gianino SM, Harpstrite SE, Rodriguez
FJ, Tuskan RG, Reilly KM, Piwnica-Worms DR, Gutmann DH
These techniques might soon become more widely avail- (2010) The neurofibromatosis type 1 tumor suppressor controls
able, easier to standardize, and become more cost effective. cell growth by regulating signal transducer and activator of tran-
Furthermore, the current histology-based diagnosis of brain scription-3 activity in vitro and in vivo. Cancer Res 70:1356–
tumors will increasingly be supplemented with molecular 1366. doi:10.1158/0008-5472.CAN-09-2178
10. Bax DA, Mackay A, Little SE, Carvalho D, Viana-Pereira M,
diagnostic tests to enable a biology-based classification and Tamber N, Grigoriadis AE, Ashworth A, Reis RM, Ellison
improve patient stratification that will hopefully be incor- DW et al (2010) A distinct spectrum of copy number aberra-
porated in carefully designed clinical trials. It is hoped that tions in pediatric high-grade gliomas. Clin Cancer Res Off J
this approach of precision diagnostics–therapeutics can Am Assoc Cancer Res 16:3368–3377. doi:10.1158/1078-0432.
CCR-10-0438
lead to step-by-step improvements of outcome where effec- 11. Baylin SB, Esteller M, Rountree MR, Bachman KE, Schuebel
tive therapeutics are appropriately “matched” with molecu- K, Herman JG (2001) Aberrant patterns of DNA methylation,
larly defined patient subsets. Even with the current excite- chromatin formation and gene expression in cancer. Hum Mol
ment in molecular classification, we remain a significant Genet 10:687–692
12. Baysan M, Bozdag S, Cam MC, Kotliarova S, Ahn S, Wall-
distance from substantive improvements, and ultimately a ing J, Killian JK, Stevenson H, Meltzer P, Fine HA (2012)
cure, for patients with GBM. Proper classification and bio- G-cimp status prediction of glioblastoma samples using mRNA
logic understanding, while a key feature of personalized expression data. PLoS One 7:e47839. doi:10.1371/journal.
therapy, is only one component and in itself is of limited pone.0047839
13. Beroukhim R, Getz G, Nghiemphu L, Barretina J, Hsueh T, Lin-
value unless matched by parallel successes in the develop- hart D, Vivanco I, Lee JC, Huang JH, Alexander S et al (2007)
ment of companion drugs and modalities for the overall Assessing the significance of chromosomal aberrations in can-
goal of improved patient outcomes. cer: methodology and application to glioma. Proc Natl Acad Sci
USA 104:20007–20012. doi:10.1073/pnas.0710052104
14. Bhat KP, Balasubramaniyan V, Vaillant B, Ezhilarasan R, Hum-
melink K, Hollingsworth F, Wani K, Heathcock L, James JD,
References Goodman LD et al (2013) Mesenchymal differentiation medi-
ated by NF-kappaB promotes radiation resistance in glioblas-
1. Alcantara Llaguno S, Chen J, Kwon CH, Jackson EL, Li Y, toma. Cancer Cell 24:331–346. doi:10.1016/j.ccr.2013.08.001
Burns DK, Alvarez-Buylla A, Parada LF (2009) Malignant 15. Bhat KP, Salazar KL, Balasubramaniyan V, Wani K, Heath-
astrocytomas originate from neural stem/progenitor cells in a cock L, Hollingsworth F, James JD, Gumin J, Diefes KL, Kim
somatic tumor suppressor mouse model. Cancer Cell 15:45–56. SH et al (2011) The transcriptional coactivator TAZ regulates
doi:10.1016/j.ccr.2008.12.006 mesenchymal differentiation in malignant glioma. Genes Dev
2. Amatya VJ, Naumann U, Weller M, Ohgaki H (2005) TP53 25:2594–2609. doi:10.1101/gad.176800.111
promoter methylation in human gliomas. Acta Neuropathol 16. Brandes AA, Franceschi E, Tosoni A, Blatt V, Pession A, Tallini
110:178–184. doi:10.1007/s00401-005-1041-5 G, Bertorelle R, Bartolini S, Calbucci F, Andreoli A et al (2008)
3. Assanah M, Lochhead R, Ogden A, Bruce J, Goldman J, MGMT promoter methylation status can predict the incidence
Canoll P (2006) Glial progenitors in adult white matter are and outcome of pseudoprogression after concomitant radio-
driven to form malignant gliomas by platelet-derived growth chemotherapy in newly diagnosed glioblastoma patients. J Clin
factor-expressing retroviruses. J Neurosci Off J Soc Neurosci Oncol Off J Am Soc Clin Oncol 26:2192–2197. doi:10.1200/
26:6781–6790. doi:10.1523/JNEUROSCI.0514-06.2006 JCO.2007.14.8163
4. Assanah MC, Bruce JN, Suzuki SO, Chen A, Goldman JE, 17. Brennan C, Momota H, Hambardzumyan D, Ozawa T, Tandon
Canoll P (2009) PDGF stimulates the massive expansion of A, Pedraza A, Holland E (2009) Glioblastoma subclasses
glial progenitors in the neonatal forebrain. Glia 57:1835–1847. can be defined by activity among signal transduction path-
doi:10.1002/glia.20895 ways and associated genomic alterations. PLoS One 4:e7752.
5. Bady P, Sciuscio D, Diserens AC, Bloch J, van den Bent MJ, doi:10.1371/journal.pone.0007752
Marosi C, Dietrich PY, Weller M, Mariani L, Heppner FL et al 18. Brennan CW, Verhaak RG, McKenna A, Campos B, Noushmehr
(2012) MGMT methylation analysis of glioblastoma on the H, Salama SR, Zheng S, Chakravarty D, Sanborn JZ, Berman

13
 Acta Neuropathol

SH et al (2013) The somatic genomic landscape of glioblas- J, Kwak EL et al (2010) Rapid targeted mutational analysis
toma. Cell 155:462–477. doi:10.1016/j.cell.2013.09.034 of human tumours: a clinical platform to guide personalized
19. Broniscer A, Gajjar A (2004) Supratentorial high-grade astrocy- cancer medicine. EMBO Mol Med 2:146–158. doi:10.1002/
toma and diffuse brainstem glioma: two challenges for the pedi- emmm.201000070
atric oncologist. Oncologist 9:197–206 35. Donson AM, Addo-Yobo SO, Handler MH, Gore L, Foreman
20. Buttarelli FR, Massimino M, Antonelli M, Lauriola L, Nozza NK (2007) MGMT promoter methylation correlates with sur-
P, Donofrio V, Arcella A, Oliva MA, Di Rocco C, Giangasp- vival benefit and sensitivity to temozolomide in pediatric glio-
ero F (2010) Evaluation status and prognostic significance of blastoma. Pediatr Blood Cancer 48:403–407. doi:10.1002/
O6-methylguanine-DNA methyltransferase (MGMT) promoter pbc.20803
methylation in pediatric high grade gliomas. Child’s Nerv- 36. Druker B (2001) Signal transduction inhibition: results from
ous Sys ChNS Off J Int Soc Pediatr Neurosurg 26:1051–1056. phase I clinical trials in chronic myeloid leukemia. Semin
doi:10.1007/s00381-010-1191-1 Hematol 38:9–14
21. Cairncross G, Wang M, Shaw E, Jenkins R, Brachman D, Buck- 37. Ducrest AL, Szutorisz H, Lingner J, Nabholz M (2002) Regula-
ner J, Fink K, Souhami L, Laperriere N, Curran W et al (2013) tion of the human telomerase reverse transcriptase gene. Onco-
Phase III trial of chemoradiotherapy for anaplastic oligodendro- gene 21:541–552. doi:10.1038/sj.onc.1205081
glioma: long-term results of RTOG 9402. J Clin Oncol Off J Am 38. Dunn GP, Rinne ML, Wykosky J, Genovese G, Quayle SN,
Soc Clin Oncol 31:337–343. doi:10.1200/JCO.2012.43.2674 Dunn IF, Agarwalla PK, Chheda MG, Campos B, Wang A
22. Cairncross JG, Ueki K, Zlatescu MC, Lisle DK, Finkelstein et al (2012) Emerging insights into the molecular and cellular
DM, Hammond RR, Silver JS, Stark PC, Macdonald DR, Ino basis of glioblastoma. Genes Dev 26:756–784. doi:10.1101/
Y et al (1998) Specific genetic predictors of chemotherapeutic gad.187922.112
response and survival in patients with anaplastic oligodendro- 39. Eckel-Passow J, Lachance D, Walsh K, Decker P, Sicotte H,
gliomas. J Natl Cancer Inst 90:1473–1479 Pekmezci M, Molinaro A, Rice T, Kosel M, Smirnov I et al
23. Cancer Genome Atlas Research N (2008) Comprehensive (2014) TERT promoter mutation, IDH mutation and 1p/19q
genomic characterization defines human glioblastoma genes codeletion define five glioma molecular groups with spe-
and core pathways. Nature 455:1061–1068. doi:10.1038/ cific clinical characteristics and germline variant associations.
nature07385 Neuro-Oncology; Abstracts from the 19th Annual Scientific
24. Capper D, Preusser M, Habel A, Sahm F, Ackermann U, Schin- Meeting of the Society for Neuro-Oncology, City
dler G, Pusch S, Mechtersheimer G, Zentgraf H, von Deimling A 40. Ekstrand AJ, Sugawa N, James CD, Collins VP (1992) Ampli-
(2011) Assessment of BRAF V600E mutation status by immu- fied and rearranged epidermal growth factor receptor genes in
nohistochemistry with a mutation-specific monoclonal antibody. human glioblastomas reveal deletions of sequences encoding
Acta Neuropathol 122:11–19. doi:10.1007/s00401-011-0841-z portions of the N- and/or C-terminal tails. Proc Natl Acad Sci
25. Chabner BA (2011) Early accelerated approval for highly USA 89:4309–4313
targeted cancer drugs. New Engl J Med 364:1087–1089. 41. Esteller M, Garcia-Foncillas J, Andion E, Goodman SN,
doi:10.1056/NEJMp1100548 Hidalgo OF, Vanaclocha V, Baylin SB, Herman JG (2000)
26. Chapman PB, Hauschild A, Robert C, Haanen JB, Ascierto P, Inactivation of the DNA-repair gene MGMT and the clini-
Larkin J, Dummer R, Garbe C, Testori A, Maio M et al (2011) cal response of gliomas to alkylating agents. New Eng J Med
Improved survival with vemurafenib in melanoma with BRAF 343:1350–1354. doi:10.1056/NEJM200011093431901
V600E mutation. New Eng J Med 364:2507–2516. doi:10.1056/ 42. Faury D, Nantel A, Dunn SE, Guiot MC, Haque T, Hauser
NEJMoa1103782 P, Garami M, Bognar L, Hanzely Z, Liberski PP et al (2007)
27. Chung R, Whaley J, Kley N, Anderson K, Louis D, Menon Molecular profiling identifies prognostic subgroups of pedi-
A, Hettlich C, Freiman R, Hedley-Whyte ET, Martuza R et al atric glioblastoma and shows increased YB-1 expression in
(1991) TP53 gene mutations and 17p deletions in human astro- tumors. J Clin Oncol Off J Am Soc Clin Oncol 25:1196–1208.
cytomas. Genes Chromosom Cancer 3:323–331 doi:10.1200/JCO.2006.07.8626
28. Clarke ID, Dirks PB (2003) A human brain tumor-derived 43. Fleming TP, Saxena A, Clark WC, Robertson JT, Oldfield EH,
PDGFR-alpha deletion mutant is transforming. Oncogene Aaronson SA, Ali IU (1992) Amplification and/or overexpres-
22:722–733. doi:10.1038/sj.onc.1206160 sion of platelet-derived growth factor receptors and epider-
29. Costello JF, Berger MS, Huang HS, Cavenee WK (1996) Silenc- mal growth factor receptor in human glial tumors. Cancer Res
ing of p16/CDKN2 expression in human gliomas by methyla- 52:4550–4553
tion and chromatin condensation. Cancer Res 56:2405–2410 44. Fomchenko EI, Dougherty JD, Helmy KY, Katz AM, Pietras A,
30. Costello JF, Plass C, Arap W, Chapman VM, Held WA, Berger Brennan C, Huse JT, Milosevic A, Holland EC (2011) Recruited
MS, Su Huang HJ, Cavenee WK (1997) Cyclin-dependent cells can become transformed and overtake PDGF-induced
kinase 6 (CDK6) amplification in human gliomas identified murine gliomas in vivo during tumor progression. PLoS One
using two-dimensional separation of genomic DNA. Cancer 6:e20605. doi:10.1371/journal.pone.0020605
Res 57:1250–1254 45. Frattini V, Trifonov V, Chan JM, Castano A, Lia M, Abate F,
31. Dang L, Jin S, Su SM (2010) IDH mutations in glioma and Keir ST, Ji AX, Zoppoli P, Niola F et al (2013) The integrated
acute myeloid leukemia. Trends Mol Med 16:387–397. landscape of driver genomic alterations in glioblastoma. Nat
doi:10.1016/j.molmed.2010.07.002 Genet 45:1141–1149. doi:10.1038/ng.2734
32. Danussi C, Akavia UD, Niola F, Jovic A, Lasorella A, Pe’er D, 46. Frederick L, Wang XY, Eley G, James CD (2000) Diversity
Iavarone A (2013) RHPN2 drives mesenchymal transformation and frequency of epidermal growth factor receptor mutations in
in malignant glioma by triggering RhoA activation. Cancer Res human glioblastomas. Cancer Res 60:1383–1387
73:5140–5150. doi:10.1158/0008-5472.CAN-13-1168-T 47. Freije WA, Castro-Vargas FE, Fang Z, Horvath S, Cloughesy
33. Di Rocco F, Carroll RS, Zhang J, Black PM (1998) Platelet- T, Liau LM, Mischel PS, Nelson SF (2004) Gene expression
derived growth factor and its receptor expression in human oli- profiling of gliomas strongly predicts survival. Cancer Res
godendrogliomas. Neurosurgery 42:341–346 64:6503–6510. doi:10.1158/0008-5472.CAN-04-0452
34. Dias-Santagata D, Akhavanfard S, David SS, Vernovsky K, 48. Fuller GN, Hess KR, Rhee CH, Yung WK, Sawaya RA, Bruner
Kuhlmann G, Boisvert SL, Stubbs H, McDermott U, Settleman JM, Zhang W (2002) Molecular classification of human diffuse

13
Acta Neuropathol

gliomas by multidimensional scaling analysis of gene expres- glioblastoma patients treated with temozolomide. Clin Cancer
sion profiles parallels morphology-based classification, corre- Res Off J Am Assoc Cancer Res 10:1871–1874
lates with survival, and reveals clinically-relevant novel glioma 61. Hegi ME, Diserens AC, Gorlia T, Hamou MF, de Tribolet N,
subsets. Brain Pathol 12:108–116 Weller M, Kros JM, Hainfellner JA, Mason W, Mariani L et al
49. Fuller GN, Rhee CH, Hess KR, Caskey LS, Wang R, Bruner (2005) MGMT gene silencing and benefit from temozolomide
JM, Yung WK, Zhang W (1999) Reactivation of insulin-like in glioblastoma. New Eng J Med 352:997–1003. doi:10.1056/
growth factor binding protein 2 expression in glioblastoma mul- NEJMoa043331
tiforme: a revelation by parallel gene expression profiling. Can- 62. Heimberger AB, Hlatky R, Suki D, Yang D, Weinberg J, Gilbert
cer Res 59:4228–4232 M, Sawaya R, Aldape K (2005) Prognostic effect of epidermal
50. Gertler R, Rosenberg R, Stricker D, Friederichs J, Hoos A, Wer- growth factor receptor and EGFRvIII in glioblastoma multi-
ner M, Ulm K, Holzmann B, Nekarda H, Siewert JR (2004) forme patients. Clin Cancer Res Off J Am Assoc Cancer Res
Telomere length and human telomerase reverse transcriptase 11:1462–1466. doi:10.1158/1078-0432.CCR-04-1737
expression as markers for progression and prognosis of colorec- 63. Henson JW, Schnitker BL, Correa KM, von Deimling A, Fass-
tal carcinoma. J Clin Oncol Off J Am Soc Clin Oncol 22:1807– bender F, Xu HJ, Benedict WF, Yandell DW, Louis DN (1994)
1814. doi:10.1200/JCO.2004.09.160 The retinoblastoma gene is involved in malignant progression
51. Godard S, Getz G, Delorenzi M, Farmer P, Kobayashi H, Des- of astrocytomas. Annals Neurol 36:714–721. doi:10.1002/
baillets I, Nozaki M, Diserens AC, Hamou MF, Dietrich PY ana.410360505
et al (2003) Classification of human astrocytic gliomas on the 64. Hermanson M, Funa K, Hartman M, Claesson-Welsh L, Heldin
basis of gene expression: a correlated group of genes with angi- CH, Westermark B, Nister M (1992) Platelet-derived growth
ogenic activity emerges as a strong predictor of subtypes. Can- factor and its receptors in human glioma tissue: expression of
cer Res 63:6613–6625 messenger RNA and protein suggests the presence of autocrine
52. Gorovets D, Kannan K, Shen R, Kastenhuber ER, Islamdoust and paracrine loops. Cancer Res 52:3213–3219
N, Campos C, Pentsova E, Heguy A, Jhanwar SC, Mellinghoff 65. Herrlinger U, Rieger J, Koch D, Loeser S, Blaschke B, Kort-
IK et al (2012) IDH mutation and neuroglial developmental fea- mann RD, Steinbach JP, Hundsberger T, Wick W, Meyermann
tures define clinically distinct subclasses of lower grade diffuse R et al (2006) Phase II trial of lomustine plus temozolomide
astrocytic glioma. Clin Cancer Res Off J Am Assoc Cancer Res chemotherapy in addition to radiotherapy in newly diagnosed
18:2490–2501. doi:10.1158/1078-0432.CCR-11-2977 glioblastoma: UKT-03. J Clin Oncol Off J Am Soc Clin Oncol
53. Griewank KG, Murali R, Schilling B, Scholz S, Sucker A, Song 24:4412–4417. doi:10.1200/JCO.2006.06.9104
M, Susskind D, Grabellus F, Zimmer L, Hillen U et al (2013) 66. Hoadley KA, Yau C, Wolf DM, Cherniack AD, Tamborero D,
TERT promoter mutations in ocular melanoma distinguish Ng S, Leiserson MD, Niu B, McLellan MD, Uzunangelov V
between conjunctival and uveal tumours. Br J Cancer 109:497– et al (2014) Multiplatform analysis of 12 cancer types reveals
501. doi:10.1038/bjc.2013.312 molecular classification within and across tissues of origin. Cell
54. Gross S, Cairns RA, Minden MD, Driggers EM, Bittinger MA, 158:929–944. doi:10.1016/j.cell.2014.06.049
Jang HG, Sasaki M, Jin S, Schenkein DP, Su SM et al (2010) 67. Horn S, Figl A, Rachakonda PS, Fischer C, Sucker A, Gast A,
Cancer-associated metabolite 2-hydroxyglutarate accumulates Kadel S, Moll I, Nagore E, Hemminki K et al (2013) TERT pro-
in acute myelogenous leukemia with isocitrate dehydroge- moter mutations in familial and sporadic melanoma. Science
nase 1 and 2 mutations. J Exp Med 207:339–344. doi:10.1084/ 339:959–961. doi:10.1126/science.1230062
jem.20092506 68. Hovestadt V, Remke M, Kool M, Pietsch T, Northcott PA, Fischer
55. Gutmann DH, Rasmussen SA, Wolkenstein P, MacCollin MM, R, Cavalli FM, Ramaswamy V, Zapatka M, Reifenberger G et al
Guha A, Inskip PD, North KN, Poyhonen M, Birch PH, Fried- (2013) Robust molecular subgrouping and copy-number profil-
man JM (2002) Gliomas presenting after age 10 in individuals ing of medulloblastoma from small amounts of archival tumour
with neurofibromatosis type 1 (NF1). Neurology 59:759–761 material using high-density DNA methylation arrays. Acta Neu-
56. Han SJ, Yang I, Tihan T, Prados MD, Parsa AT (2010) Primary ropathol 125:913–916. doi:10.1007/s00401-013-1126-5
gliosarcoma: key clinical and pathologic distinctions from glio- 69. Huang HS, Nagane M, Klingbeil CK, Lin H, Nishikawa R, Ji
blastoma with implications as a unique oncologic entity. J Neu- XD, Huang CM, Gill GN, Wiley HS, Cavenee WK (1997) The
rooncol 96:313–320. doi:10.1007/s11060-009-9973-6 enhanced tumorigenic activity of a mutant epidermal growth
57. Haque T, Faury D, Albrecht S, Lopez-Aguilar E, Hauser P, Gar- factor receptor common in human cancers is mediated by
ami M, Hanzely Z, Bognar L, Del Maestro RF, Atkinson J et al threshold levels of constitutive tyrosine phosphorylation and
(2007) Gene expression profiling from formalin-fixed paraffin- unattenuated signaling. J Biol Chem 272:2927–2935
embedded tumors of pediatric glioblastoma. Clin Cancer Res 70. Hurtt MR, Moossy J, Donovan-Peluso M, Locker J (1992)
Off J Am Assoc Cancer Res 13:6284–6292. doi:10.1158/1078- Amplification of epidermal growth factor receptor gene in glio-
0432.CCR-07-0525 mas: histopathology and prognosis. J Neuropathol Exp Neurol
58. Hartmann C, Meyer J, Balss J, Capper D, Mueller W, Chris- 51:84–90
tians A, Felsberg J, Wolter M, Mawrin C, Wick W et al (2009) 71. Huse JT, Brennan C, Hambardzumyan D, Wee B, Pena J, Rou-
Type and frequency of IDH1 and IDH2 mutations are related hanifard SH, Sohn-Lee C, le Sage C, Agami R, Tuschl T et al
to astrocytic and oligodendroglial differentiation and age: a (2009) The PTEN-regulating microRNA miR-26a is amplified
study of 1,010 diffuse gliomas. Acta Neuropathol 118:469–474. in high-grade glioma and facilitates gliomagenesis in vivo.
doi:10.1007/s00401-009-0561-9 Genes Dev 23:1327–1337. doi:10.1101/gad.1777409
59. Heaphy CM, de Wilde RF, Jiao Y, Klein AP, Edil BH, Shi C, 72. Ideraabdullah FY, Vigneau S, Bartolomei MS (2008) Genomic
Bettegowda C, Rodriguez FJ, Eberhart CG, Hebbar S et al imprinting mechanisms in mammals. Mutat Res 647:77–85.
(2011) Altered telomeres in tumors with ATRX and DAXX doi:10.1016/j.mrfmmm.2008.08.008
mutations. Science 333:425. doi:10.1126/science.1207313 73. Inda MM, Bonavia R, Mukasa A, Narita Y, Sah DW, Vanden-
60. Hegi ME, Diserens AC, Godard S, Dietrich PY, Regli L, Oster- berg S, Brennan C, Johns TG, Bachoo R, Hadwiger P et al
mann S, Otten P, Van Melle G, de Tribolet N, Stupp R (2004) (2010) Tumor heterogeneity is an active process maintained
Clinical trial substantiates the predictive value of O-6-meth- by a mutant EGFR-induced cytokine circuit in glioblastoma.
ylguanine-DNA methyltransferase promoter methylation in Genes Dev 24:1731–1745. doi:10.1101/gad.1890510

13
 Acta Neuropathol

74. International Cancer Genome C, Hudson TJ, Anderson W, Artez (2007) Long-term survival with glioblastoma multiforme. Brain
A, Barker AD, Bell C, Bernabe RR, Bhan MK, Calvo F, Eerola J Neurol 130:2596–2606. doi:10.1093/brain/awm204
I et al (2010) International network of cancer genome projects. 89. Kwak EL, Bang YJ, Camidge DR, Shaw AT, Solomon B, Maki
Nature 464:993–998. doi:10.1038/nature08987 RG, Ou SH, Dezube BJ, Janne PA, Costa DB et al (2010)
75. Janne PA, Engelman JA, Johnson BE (2005) Epidermal growth Anaplastic lymphoma kinase inhibition in non-small-cell
factor receptor mutations in non-small-cell lung cancer: impli- lung cancer. New Eng J Med 363:1693–1703. doi:10.1056/
cations for treatment and tumor biology. J Clin Oncol Off J Am NEJMoa1006448
Soc Clin Oncol 23:3227–3234. doi:10.1200/JCO.2005.09.985 90. Kwon CH, Zhao D, Chen J, Alcantara S, Li Y, Burns DK,
76. Jaros E, Perry RH, Adam L, Kelly PJ, Crawford PJ, Kalbag Mason RP, Lee EY, Wu H, Parada LF (2008) Pten haploinsuf-
RM, Mendelow AD, Sengupta RP, Pearson AD (1992) Prog- ficiency accelerates formation of high-grade astrocytomas. Can-
nostic implications of p53 protein, epidermal growth factor cer Res 68:3286–3294. doi:10.1158/0008-5472.CAN-07-6867
receptor, and Ki-67 labelling in brain tumours. Br J Cancer 91. Labussiere M, Boisselier B, Mokhtari K, Di Stefano AL, Rahim-
66:373–385 ian A, Rossetto M, Ciccarino P, Saulnier O, Paterra R, Marie Y
77. Jiao Y, Killela PJ, Reitman ZJ, Rasheed AB, Heaphy CM, de et al (2014) Combined analysis of TERT, EGFR, and IDH sta-
Wilde RF, Rodriguez FJ, Rosemberg S, Oba-Shinjo SM, Naga- tus defines distinct prognostic glioblastoma classes. Neurology
hashi Marie SK et al (2012) Frequent ATRX, CIC, FUBP1 and 83:1200–1206. doi:10.1212/WNL.0000000000000814
IDH1 mutations refine the classification of malignant gliomas. 92. Lee D, Kang SY, Suh YL, Jeong JY, Lee JI, Nam DH (2012)
Oncotarget 3:709–722 Clinicopathologic and genomic features of gliosarcomas. J
78. Jones C, Perryman L, Hargrave D (2012) Paediatric and adult Neurooncol 107:643–650. doi:10.1007/s11060-011-0790-3
malignant glioma: close relatives or distant cousins? Nature 93. Lee JC, Vivanco I, Beroukhim R, Huang JH, Feng WL, DeBiasi
reviews. Clin Oncol 9:400–413. doi:10.1038/nrclinonc.2012.87 RM, Yoshimoto K, King JC, Nghiemphu P, Yuza Y et al (2006)
79. Karajannis MA, Legault G, Fisher MJ, Milla SS, Cohen KJ, Epidermal growth factor receptor activation in glioblastoma
Wisoff JH, Harter DH, Goldberg JD, Hochman T, Merkelson A through novel missense mutations in the extracellular domain.
et al (2014) Phase II study of sorafenib in children with recur- PLoS Med 3:e485. doi:10.1371/journal.pmed.0030485
rent or progressive low-grade astrocytomas. Neuro-oncology 94. Lee JY, Park CK, Park SH, Wang KC, Cho BK, Kim SK (2011)
16:1408–1416. doi:10.1093/neuonc/nou059 MGMT promoter gene methylation in pediatric glioblastoma:
80. Karcher S, Steiner HH, Ahmadi R, Zoubaa S, Vasvari G, Bauer analysis using MS-MLPA. Child’s nervous system: ChNS.
H, Unterberg A, Herold-Mende C (2006) Different angiogenic Off J Int Soc Pediatr Neurosurg 27:1877–1883. doi:10.1007/
phenotypes in primary and secondary glioblastomas. Int J Can- s00381-011-1525-7
cer J Int Cancer 118:2182–2189. doi:10.1002/ijc.21648 95. Li A, Walling J, Ahn S, Kotliarov Y, Su Q, Quezado M, Ober-
81. Killela PJ, Reitman ZJ, Jiao Y, Bettegowda C, Agrawal N, Diaz holtzer JC, Park J, Zenklusen JC, Fine HA (2009) Unsuper-
LA Jr, Friedman AH, Friedman H, Gallia GL, Giovanella BC vised analysis of transcriptomic profiles reveals six glioma
et al (2013) TERT promoter mutations occur frequently in glio- subtypes. Cancer Res 69:2091–2099. doi:10.1158/0008-5472.
mas and a subset of tumors derived from cells with low rates CAN-08-2100
of self-renewal. Proc Natl Acad Sci USA 110:6021–6026. 96. Liang Y, Diehn M, Watson N, Bollen AW, Aldape KD, Nicho-
doi:10.1073/pnas.1303607110 las MK, Lamborn KR, Berger MS, Botstein D, Brown PO et al
82. Kim H, Huang W, Jiang X, Pennicooke B, Park PJ, Johnson MD (2005) Gene expression profiling reveals molecularly and clini-
(2010) Integrative genome analysis reveals an oncomir/onco- cally distinct subtypes of glioblastoma multiforme. Proc Natl
gene cluster regulating glioblastoma survivorship. Proc Natl Acad Sci USA 102:5814–5819. doi:10.1073/pnas.0402870102
Acad Sci USA 107:2183–2188. doi:10.1073/pnas.0909896107 97. Liu XY, Gerges N, Korshunov A, Sabha N, Khuong-Quang
83. Kim S, Dougherty ER, Shmulevich I, Hess KR, Hamilton SR, DA, Fontebasso AM, Fleming A, Hadjadj D, Schwartzentruber
Trent JM, Fuller GN, Zhang W (2002) Identification of com- J, Majewski J et al (2012) Frequent ATRX mutations and loss
bination gene sets for glioma classification. Mol Cancer Ther of expression in adult diffuse astrocytic tumors carrying IDH1/
1:1229–1236 IDH2 and TP53 mutations. Acta Neuropathol 124:615–625.
84. Kleinschmidt-DeMasters BK, Aisner DL, Birks DK, Foreman doi:10.1007/s00401-012-1031-3
NK (2013) Epithelioid GBMs show a high percentage of BRAF 98. Loenarz C, Schofield CJ (2008) Expanding chemical biol-
V600E mutation. Am J Surg Pathol 37:685–698. doi:10.1097/ ogy of 2-oxoglutarate oxygenases. Nat Chem Biol 4:152–156.
PAS.0b013e31827f9c5e doi:10.1038/nchembio0308-152
85. Knobbe CB, Reifenberger J, Reifenberger G (2004) Muta- 99. Lokker NA, Sullivan CM, Hollenbach SJ, Israel MA, Giese NA
tion analysis of the Ras pathway genes NRAS, HRAS, KRAS (2002) Platelet-derived growth factor (PDGF) autocrine signal-
and BRAF in glioblastomas. Acta Neuropathol 108:467–470. ing regulates survival and mitogenic pathways in glioblastoma
doi:10.1007/s00401-004-0929-9 cells: evidence that the novel PDGF-C and PDGF-D ligands
86. Koelsche C, Sahm F, Capper D, Reuss D, Sturm D, Jones DT, may play a role in the development of brain tumors. Cancer Res
Kool M, Northcott PA, Wiestler B, Bohmer K et al (2013) Dis- 62:3729–3735
tribution of TERT promoter mutations in pediatric and adult 100. Louis DN (1994) The p53 gene and protein in human brain
tumors of the nervous system. Acta Neuropathol 126:907–915. tumors. J Neuropathol Exp Neurol 53:11–21
doi:10.1007/s00401-013-1195-5 101. Louis DN, Ohgaki H, Wiestler OD, Cavenee WK, Burger PC,
87. Korshunov A, Ryzhova M, Hovestadt V, Bender S, Sturm D, Jouvet A, Scheithauer BW, Kleihues P (2007) The 2007 WHO
Capper D, Meyer J, Schrimpf D, Kool M, Northcott PA et al classification of tumours of the central nervous system. Acta
(2015) Integrated analysis of pediatric glioblastoma reveals Neuropathol 114:97–109. doi:10.1007/s00401-007-0243-4
a subset of biologically favorable tumors with associated 102. Lu L, Zhang C, Zhu G, Irwin M, Risch H, Menato G, Mitidieri
molecular prognostic markers. Acta Neuropathol 129:669–678. M, Katsaros D, Yu H (2011) Telomerase expression and tel-
doi:10.1007/s00401-015-1405-4 omere length in breast cancer and their associations with adju-
88. Krex D, Klink B, Hartmann C, von Deimling A, Pietsch T, vant treatment and disease outcome. Breast Cancer Res BCR
Simon M, Sabel M, Steinbach JP, Heese O, Reifenberger G et al 13:R56. doi:10.1186/bcr2893

13
Acta Neuropathol

103. Lynch TJ, Bell DW, Sordella R, Gurubhagavatula S, Oki- 118. Ohgaki H, Kleihues P (2013) The definition of primary and sec-
moto RA, Brannigan BW, Harris PL, Haserlat SM, Supko JG, ondary glioblastoma. Clin Cancer Res Off J Am Assoc Cancer
Haluska FG et al (2004) Activating mutations in the epidermal Res 19:764–772. doi:10.1158/1078-0432.CCR-12-3002
growth factor receptor underlying responsiveness of non-small- 119. Olson RA, Brastianos PK, Palma DA (2011) Prognostic and
cell lung cancer to gefitinib. New Eng J Med 350:2129–2139. predictive value of epigenetic silencing of MGMT in patients
doi:10.1056/NEJMoa040938 with high grade gliomas: a systematic review and meta-analysis.
104. MacConaill LE, Campbell CD, Kehoe SM, Bass AJ, Hatton C, J Neurooncol 105:325–335. doi:10.1007/s11060-011-0594-5
Niu L, Davis M, Yao K, Hanna M, Mondal C et al (2009) Pro- 120. Ostrom QT, Gittleman H, Liao P, Rouse C, Chen Y, Dowling
filing critical cancer gene mutations in clinical tumor samples. J, Wolinsky Y, Kruchko C, Barnholtz-Sloan J (2014) CBTRUS
PLoS One 4:e7887. doi:10.1371/journal.pone.0007887 statistical report: primary brain and central nervous system
105. Malmstrom A, Gronberg BH, Marosi C, Stupp R, Frappaz D, tumors diagnosed in the United States in 2007–2011. Neuro-
Schultz H, Abacioglu U, Tavelin B, Lhermitte B, Hegi ME oncology 16(Suppl 4):1–63. doi:10.1093/neuonc/nou223
et al (2012) Temozolomide versus standard 6-week radio- 121. Ozawa T, Brennan CW, Wang L, Squatrito M, Sasayama T,
therapy versus hypofractionated radiotherapy in patients older Nakada M, Huse JT, Pedraza A, Utsuki S, Yasui Y et al (2010)
than 60 years with glioblastoma: the Nordic randomised, PDGFRA gene rearrangements are frequent genetic events in
phase 3 trial. Lancet Oncol 13:916–926. doi:10.1016/ PDGFRA-amplified glioblastomas. Genes Dev 24:2205–2218.
S1470-2045(12)70265-6 doi:10.1101/gad.1972310
106. Mashiyama S, Murakami Y, Yoshimoto T, Sekiya T, Hayashi K 122. Ozawa T, Riester M, Cheng YK, Huse JT, Squatrito M, Helmy
(1991) Detection of p53 gene mutations in human brain tumors K, Charles N, Michor F, Holland EC (2014) Most human
by single-strand conformation polymorphism analysis of poly- non-GCIMP glioblastoma subtypes evolve from a common
merase chain reaction products. Oncogene 6:1313–1318 proneural-like precursor glioma. Cancer Cell 26:288–300.
107. Masui K, Cloughesy TF, Mischel PS (2012) Review: molecu- doi:10.1016/j.ccr.2014.06.005
lar pathology in adult high-grade gliomas: from molecular 123. Parsons DW, Jones S, Zhang X, Lin JC, Leary RJ, Angenendt
diagnostics to target therapies. Neuropathol Appl Neurobiol P, Mankoo P, Carter H, Siu IM, Gallia GL et al (2008) An inte-
38:271–291. doi:10.1111/j.1365-2990.2011.01238.x grated genomic analysis of human glioblastoma multiforme.
108. McBride SM, Perez DA, Polley MY, Vandenberg SR, Smith JS, Science 321:1807–1812. doi:10.1126/science.1164382
Zheng S, Lamborn KR, Wiencke JK, Chang SM, Prados MD 124. Patel AP, Tirosh I, Trombetta JJ, Shalek AK, Gillespie SM,
et al (2010) Activation of PI3 K/mTOR pathway occurs in most Wakimoto H, Cahill DP, Nahed BV, Curry WT, Martuza RL
adult low-grade gliomas and predicts patient survival. J Neu- et al (2014) Single-cell RNA-seq highlights intratumoral het-
rooncol 97:33–40. doi:10.1007/s11060-009-0004-4 erogeneity in primary glioblastoma. Science 344:1396–1401.
109. McGillicuddy LT, Fromm JA, Hollstein PE, Kubek S, Berouk- doi:10.1126/science.1254257
him R, De Raedt T, Johnson BW, Williams SM, Nghiemphu P, 125. Paugh BS, Broniscer A, Qu C, Miller CP, Zhang J, Tatevossian
Liau LM et al (2009) Proteasomal and genetic inactivation of RG, Olson JM, Geyer JR, Chi SN, da Silva NS et al (2011)
the NF1 tumor suppressor in gliomagenesis. Cancer Cell 16:44– Genome-wide analyses identify recurrent amplifications of
54. doi:10.1016/j.ccr.2009.05.009 receptor tyrosine kinases and cell-cycle regulatory genes in dif-
110. Meyer-Puttlitz B, Hayashi Y, Waha A, Rollbrocker B, Bostrom fuse intrinsic pontine glioma. J Clin Oncol Off J Am Soc Clin
J, Wiestler OD, Louis DN, Reifenberger G, von Deimling A Oncol 29:3999–4006. doi:10.1200/JCO.2011.35.5677
(1997) Molecular genetic analysis of giant cell glioblastomas. 126. Paugh BS, Qu C, Jones C, Liu Z, Adamowicz-Brice M, Zhang
Am J Pathol 151:853–857 J, Bax DA, Coyle B, Barrow J, Hargrave D et al (2010) Inte-
111. Nagane M, Levitzki A, Gazit A, Cavenee WK, Huang HJ (1998) grated molecular genetic profiling of pediatric high-grade
Drug resistance of human glioblastoma cells conferred by a gliomas reveals key differences with the adult disease. J Clin
tumor-specific mutant epidermal growth factor receptor through Oncol Off J Am Soc Clin Oncol 28:3061–3068. doi:10.1200/
modulation of Bcl-XL and caspase-3-like proteases. Proc Natl JCO.2009.26.7252
Acad Sci USA 95:5724–5729 127. Pelloski CE, Ballman KV, Furth AF, Zhang L, Lin E, Sulman
112. Nakamura M, Yonekawa Y, Kleihues P, Ohgaki H (2001) Pro- EP, Bhat K, McDonald JM, Yung WK, Colman H et al (2007)
moter hypermethylation of the RB1 gene in glioblastomas. Lab Epidermal growth factor receptor variant III status defines clini-
Investig J Tech Method Pathol 81:77–82 cally distinct subtypes of glioblastoma. J Clin Oncol Off J Am
113. Narita Y, Nagane M, Mishima K, Huang HJ, Furnari FB, Cave- Soc Clin Oncol 25:2288–2294. doi:10.1200/JCO.2006.08.0705
nee WK (2002) Mutant epidermal growth factor receptor sign- 128. Perry A, Aldape KD, George DH, Burger PC (2004) Small cell
aling down-regulates p27 through activation of the phosphati- astrocytoma: an aggressive variant that is clinicopathologically
dylinositol 3-kinase/Akt pathway in glioblastomas. Cancer Res and genetically distinct from anaplastic oligodendroglioma.
62:6764–6769 Cancer 101:2318–2326. doi:10.1002/cncr.20625
114. Network TC (2013) Corrigendum: comprehensive genomic 129. Phillips HS, Kharbanda S, Chen R, Forrest WF, Soriano RH,
characterization defines human glioblastoma genes and core Wu TD, Misra A, Nigro JM, Colman H, Soroceanu L et al
pathways. Nature 494:506. doi:10.1038/nature11903 (2006) Molecular subclasses of high-grade glioma predict prog-
115. Ng HH, Bird A (1999) DNA methylation and chromatin modifi- nosis, delineate a pattern of disease progression, and resemble
cation. Curr Opin Genet Dev 9:158–163 stages in neurogenesis. Cancer Cell 9:157–173. doi:10.1016/j.
116. Nishikawa R, Sugiyama T, Narita Y, Furnari F, Cavenee WK, ccr.2006.02.019
Matsutani M (2004) Immunohistochemical analysis of the 130. Puget S, Philippe C, Bax DA, Job B, Varlet P, Junier MP,
mutant epidermal growth factor, deltaEGFR, in glioblastoma. Andreiuolo F, Carvalho D, Reis R, Guerrini-Rousseau L et al
Brain Tumor Pathol 21:53–56 (2012) Mesenchymal transition and PDGFRA amplification/
117. Noushmehr H, Weisenberger DJ, Diefes K, Phillips HS, Pujara mutation are key distinct oncogenic events in pediatric diffuse
K, Berman BP, Pan F, Pelloski CE, Sulman EP, Bhat KP et al intrinsic pontine gliomas. PLoS One 7:e30313. doi:10.1371/
(2010) Identification of a CpG island methylator phenotype that journal.pone.0030313
defines a distinct subgroup of glioma. Cancer Cell 17:510–522. 131. Reifenberger J, Reifenberger G, Liu L, James CD, Wechsler
doi:10.1016/j.ccr.2010.03.017 W, Collins VP (1994) Molecular genetic analysis of

13
 Acta Neuropathol

oligodendroglial tumors shows preferential allelic deletions on 145. Schwartzentruber J, Korshunov A, Liu XY, Jones DT, Pfaff E,
19q and 1p. Am J Pathol 145:1175–1190 Jacob K, Sturm D, Fontebasso AM, Quang DA, Tonjes M et al
132. Reuss DE, Sahm F, Schrimpf D, Wiestler B, Capper D, Koels- (2012) Driver mutations in histone H3.3 and chromatin remod-
che C, Schweizer L, Korshunov A, Jones DT, Hovestadt V et al elling genes in paediatric glioblastoma. Nature 482:226–231.
(2014) ATRX and IDH1-R132H immunohistochemistry with doi:10.1038/nature10833
subsequent copy number analysis and IDH sequencing as a 146. Shai R, Shi T, Kremen TJ, Horvath S, Liau LM, Cloughesy TF,
basis for an “integrated” diagnostic approach for adult astrocy- Mischel PS, Nelson SF (2003) Gene expression profiling identi-
toma, oligodendroglioma and glioblastoma. Acta Neuropathol. fies molecular subtypes of gliomas. Oncogene 22:4918–4923.
doi:10.1007/s00401-014-1370-3 doi:10.1038/sj.onc.1206753
133. Rickman DS, Bobek MP, Misek DE, Kuick R, Blaivas M, Kur- 147. Shinojima N, Tada K, Shiraishi S, Kamiryo T, Kochi M, Naka-
nit DM, Taylor J, Hanash SM (2001) Distinctive molecular mura H, Makino K, Saya H, Hirano H, Kuratsu J et al (2003)
profiles of high-grade and low-grade gliomas based on oligonu- Prognostic value of epidermal growth factor receptor in patients
cleotide microarray analysis. Cancer Res 61:6885–6891 with glioblastoma multiforme. Cancer Res 63:6962–6970
134. Robinson GW, Orr BA, Gajjar A (2014) Complete clinical 148. Simon M, Hosen I, Gousias K, Rachakonda S, Heidenreich B,
regression of a BRAF V600E-mutant pediatric glioblastoma Gessi M, Schramm J, Hemminki K, Waha A, Kumar R (2014)
multiforme after BRAF inhibitor therapy. BMC Cancer 14:258. TERT promoter mutations: a novel independent prognostic fac-
doi:10.1186/1471-2407-14-258 tor in primary glioblastomas. Neuro-oncology. doi:10.1093/
135. Rohle D, Popovici-Muller J, Palaskas N, Turcan S, Grommes neuonc/nou158
C, Campos C, Tsoi J, Clark O, Oldrini B, Komisopoulou E 149. Slamon D, Pegram M (2001) Rationale for trastuzumab (Her-
et al (2013) An inhibitor of mutant IDH1 delays growth and ceptin) in adjuvant breast cancer trials. Semin Oncol 28:13–19
promotes differentiation of glioma cells. Science 340:626–630. 150. Smith JS, Wang XY, Qian J, Hosek SM, Scheithauer BW, Jen-
doi:10.1126/science.1236062 kins RB, James CD (2000) Amplification of the platelet-derived
136. Sahm F, Capper D, Jeibmann A, Habel A, Paulus W, Troost D, growth factor receptor-A (PDGFRA) gene occurs in oligoden-
von Deimling A (2012) Addressing diffuse glioma as a systemic drogliomas with grade IV anaplastic features. J Neuropathol
brain disease with single-cell analysis. Arch Neurol 69:523– Exp Neurol 59:495–503
526. doi:10.1001/archneurol.2011.2910 151. Snuderl M, Fazlollahi L, Le LP, Nitta M, Zhelyazkova BH,
137. Sallinen SL, Sallinen PK, Haapasalo HK, Helin HJ, Helen PT, Davidson CJ, Akhavanfard S, Cahill DP, Aldape KD, Beten-
Schraml P, Kallioniemi OP, Kononen J (2000) Identification of sky RA et al (2011) Mosaic amplification of multiple receptor
differentially expressed genes in human gliomas by DNA micro- tyrosine kinase genes in glioblastoma. Cancer Cell 20:810–817.
array and tissue chip techniques. Cancer Res 60:6617–6622 doi:10.1016/j.ccr.2011.11.005
138. Sampson JH, Heimberger AB, Archer GE, Aldape KD, Fried- 152. Srivastava A, Jain A, Jha P, Suri V, Sharma MC, Mallick S,
man AH, Friedman HS, Gilbert MR, Herndon JE 2nd, McLen- Puri T, Gupta DK, Gupta A, Sarkar C (2010) MGMT gene pro-
don RE, Mitchell DA et al (2010) Immunologic escape after moter methylation in pediatric glioblastomas. Child’s nervous
prolonged progression-free survival with epidermal growth system: ChNS. Off J Int Soc Pediatr Neurosurg 26:1613–1618.
factor receptor variant III peptide vaccination in patients with doi:10.1007/s00381-010-1214-y
newly diagnosed glioblastoma. J Clin Oncol Off J Am Soc Clin 153. Stockhausen MT, Broholm H, Villingshoj M, Kirchhoff M,
Oncol 28:4722–4729. doi:10.1200/JCO.2010.28.6963 Gerdes T, Kristoffersen K, Kosteljanetz M, Spang-Thomsen
139. Sanders RP, Drissi R, Billups CA, Daw NC, Valentine MB, M, Poulsen HS (2011) Maintenance of EGFR and EGFRvIII
Dome JS (2004) Telomerase expression predicts unfavorable expressions in an in vivo and in vitro model of human glioblas-
outcome in osteosarcoma. J Clin Oncol Off J Am Soc Clin toma multiforme. Exp Cell Res 317:1513–1526. doi:10.1016/j.
Oncol 22:3790–3797. doi:10.1200/JCO.2004.03.043 yexcr.2011.04.001
140. Sandsmark DK, Zhang H, Hegedus B, Pelletier CL, Weber JD, 154. Stupp R, Hegi ME, Mason WP, van den Bent MJ, Taphoorn MJ,
Gutmann DH (2007) Nucleophosmin mediates mammalian tar- Janzer RC, Ludwin SK, Allgeier A, Fisher B, Belanger K et al
get of rapamycin-dependent actin cytoskeleton dynamics and (2009) Effects of radiotherapy with concomitant and adjuvant
proliferation in neurofibromin-deficient astrocytes. Cancer Res temozolomide versus radiotherapy alone on survival in glio-
67:4790–4799. doi:10.1158/0008-5472.CAN-06-4470 blastoma in a randomised phase III study: 5-year analysis of the
141. Sanson M, Marie Y, Paris S, Idbaih A, Laffaire J, Ducray F, El EORTC-NCIC trial. Lancet Oncol 10:459–466. doi:10.1016/
Hallani S, Boisselier B, Mokhtari K, Hoang-Xuan K et al (2009) S1470-2045(09)70025-7
Isocitrate dehydrogenase 1 codon 132 mutation is an important 155. Stupp R, Mason WP, van den Bent MJ, Weller M, Fisher B,
prognostic biomarker in gliomas. J Clin Oncol Off J Am Soc Taphoorn MJ, Belanger K, Brandes AA, Marosi C, Bogdahn
Clin Oncol 27:4150–4154. doi:10.1200/JCO.2009.21.9832 U et al (2005) Radiotherapy plus concomitant and adjuvant
142. Schindler G, Capper D, Meyer J, Janzarik W, Omran H, Herold- temozolomide for glioblastoma. New Eng J Med 352:987–996.
Mende C, Schmieder K, Wesseling P, Mawrin C, Hasselblatt doi:10.1056/NEJMoa043330
M et al (2011) Analysis of BRAF V600E mutation in 1,320 156. Sturm D, Witt H, Hovestadt V, Khuong-Quang DA, Jones DT,
nervous system tumors reveals high mutation frequencies in Konermann C, Pfaff E, Tonjes M, Sill M, Bender S et al (2012)
pleomorphic xanthoastrocytoma, ganglioglioma and extra-cer- Hotspot mutations in H3F3A and IDH1 define distinct epige-
ebellar pilocytic astrocytoma. Acta Neuropathol 121:397–405. netic and biological subgroups of glioblastoma. Cancer Cell
doi:10.1007/s00401-011-0802-6 22:425–437. doi:10.1016/j.ccr.2012.08.024
143. Schlegel J, Merdes A, Stumm G, Albert FK, Forsting M, Hynes 157. Suri V, Das P, Pathak P, Jain A, Sharma MC, Borkar SA, Suri A,
N, Kiessling M (1994) Amplification of the epidermal-growth- Gupta D, Sarkar C (2009) Pediatric glioblastomas: a histopatho-
factor-receptor gene correlates with different growth behaviour logical and molecular genetic study. Neuro-oncology 11:274–
in human glioblastoma. Int J Cancer J Int Cancer 56:72–77 280. doi:10.1215/15228517-2008-092
144. Schumacher T, Bunse L, Pusch S, Sahm F, Wiestler B, Quandt 158. Szerlip NJ, Pedraza A, Chakravarty D, Azim M, McGuire
J, Menn O, Osswald M, Oezen I, Ott M et al (2014) A vaccine J, Fang Y, Ozawa T, Holland EC, Huse JT, Jhanwar S et al
targeting mutant IDH1 induces antitumour immunity. Nature (2012) Intratumoral heterogeneity of receptor tyrosine kinases
512:324–327. doi:10.1038/nature13387 EGFR and PDGFRA amplification in glioblastoma defines

13
Acta Neuropathol

subpopulations with distinct growth factor response. Proc Natl 171. Watanabe T, Yokoo H, Yokoo M, Yonekawa Y, Kleihues P,
Acad Sci USA 109:3041–3046. doi:10.1073/pnas.1114033109 Ohgaki H (2001) Concurrent inactivation of RB1 and TP53
159. Tabori U, Ma J, Carter M, Zielenska M, Rutka J, Bouffet E, Bar- pathways in anaplastic oligodendrogliomas. J Neuropathol Exp
tels U, Malkin D, Hawkins C (2006) Human telomere reverse Neurol 60:1181–1189
transcriptase expression predicts progression and survival in 172. Weller M, Felsberg J, Hartmann C, Berger H, Steinbach JP,
pediatric intracranial ependymoma. J Clin Oncol Off J Am Soc Schramm J, Westphal M, Schackert G, Simon M, Tonn JC et al
Clin Oncol 24:1522–1528. doi:10.1200/JCO.2005.04.2127 (2009) Molecular predictors of progression-free and overall sur-
160. Tanwar MK, Gilbert MR, Holland EC (2002) Gene expression vival in patients with newly diagnosed glioblastoma: a prospec-
microarray analysis reveals YKL-40 to be a potential serum tive translational study of the German Glioma Network. J Clin
marker for malignant character in human glioma. Cancer Res Oncol Off J Am Soc Clin Oncol 27:5743–5750. doi:10.1200/
62:4364–4368 JCO.2009.23.0805
161. Thakkar JP, Dolecek TA, Horbinski C, Ostrom QT, Lightner 173. Wick W, Meisner C, Hentschel B, Platten M, Schilling A, Wies-
DD, Barnholtz-Sloan JS, Villano JL (2014) Epidemiologic and tler B, Sabel MC, Koeppen S, Ketter R, Weiler M et al (2013)
molecular prognostic review of glioblastoma. Cancer Epide- Prognostic or predictive value of MGMT promoter methylation
miol Biomark Prevent Pub Am Assoc Cancer Res Cosponsored in gliomas depends on IDH1 mutation. Neurology 81:1515–
Am Soc Prevent Oncol 23:1985–1996. doi:10.1158/1055-9965. 1522. doi:10.1212/WNL.0b013e3182a95680
EPI-14-0275 174. Wick W, Platten M, Meisner C, Felsberg J, Tabatabai G, Simon
162. Tso CL, Freije WA, Day A, Chen Z, Merriman B, Perlina A, M, Nikkhah G, Papsdorf K, Steinbach JP, Sabel M et al (2012)
Lee Y, Dia EQ, Yoshimoto K, Mischel PS et al (2006) Distinct Temozolomide chemotherapy alone versus radiotherapy alone
transcription profiles of primary and secondary glioblastoma for malignant astrocytoma in the elderly: the NOA-08 ran-
subgroups. Cancer Res 66:159–167. doi:10.1158/0008-5472. domised, phase 3 trial. Lancet Oncol 13:707–715. doi:10.1016/
CAN-05-0077 S1470-2045(12)70164-X
163. Ueki K, Ono Y, Henson JW, Efird JT, von Deimling A, Louis 175. Wiencke JK, Zheng S, Jelluma N, Tihan T, Vandenberg S, Tam-
DN (1996) CDKN2/p16 or RB alterations occur in the major- guney T, Baumber R, Parsons R, Lamborn KR, Berger MS et al
ity of glioblastomas and are inversely correlated. Cancer Res (2007) Methylation of the PTEN promoter defines low-grade
56:150–153 gliomas and secondary glioblastoma. Neuro-oncology 9:271–
164. van den Bent MJ, Brandes AA, Taphoorn MJ, Kros JM, Kou- 279. doi:10.1215/15228517-2007-003
wenhoven MC, Delattre JY, Bernsen HJ, Frenay M, Tijssen CC, 176. Wiestler B, Capper D, Hovestadt V, Sill M, Jones DT, Hartmann
Grisold W et al (2013) Adjuvant procarbazine, lomustine, and C, Felsberg J, Platten M, Feiden W, Keyvani K et al (2014)
vincristine chemotherapy in newly diagnosed anaplastic oli- Assessing CpG island methylator phenotype, 1p/19q codele-
godendroglioma: long-term follow-up of EORTC brain tumor tion, and MGMT promoter methylation from epigenome-wide
group study 26951. J Clin Oncol Off J Am Soc Clin Oncol data in the biomarker cohort of the NOA-04 trial. Neuro-oncol-
31:344–350. doi:10.1200/JCO.2012.43.2229 ogy. doi:10.1093/neuonc/nou138
165. van den Bent MJ, Erdem-Eraslan L, Idbaih A, de Rooi J, Eil- 177. Wiestler B, Capper D, Sill M, Jones DT, Hovestadt V, Sturm
ers PH, Spliet WG, den Dunnen WF, Tijssen C, Wesseling P, D, Koelsche C, Bertoni A, Schweizer L, Korshunov A et al
Sillevis Smitt PA et al (2013) MGMT-STP27 methylation status (2014) Integrated DNA methylation and copy-number profiling
as predictive marker for response to PCV in anaplastic Oligo- identify three clinically and biologically relevant groups of ana-
dendrogliomas and Oligoastrocytomas. A report from EORTC plastic glioma. Acta Neuropathol 128:561–571. doi:10.1007/
study 26951. Clin Cancer Res Off J Am Assoc Cancer Res s00401-014-1315-x
19:5513–5522. doi:10.1158/1078-0432.CCR-13-1157 178. Wong AJ, Ruppert JM, Bigner SH, Grzeschik CH, Humphrey
166. Verhaak RG, Hoadley KA, Purdom E, Wang V, Qi Y, Wilkerson PA, Bigner DS, Vogelstein B (1992) Structural alterations of the
MD, Miller CR, Ding L, Golub T, Mesirov JP et al (2010) Inte- epidermal growth factor receptor gene in human gliomas. Proc
grated genomic analysis identifies clinically relevant subtypes Natl Acad Sci USA 89:2965–2969
of glioblastoma characterized by abnormalities in PDGFRA, 179. Wong KK, Tsang YT, Chang YM, Su J, Di Francesco AM,
IDH1, EGFR, and NF1. Cancer Cell 17:98–110. doi:10.1016/j. Meco D, Riccardi R, Perlaky L, Dauser RC, Adesina A et al
ccr.2009.12.020 (2006) Genome-wide allelic imbalance analysis of pediat-
167. Vinagre J, Almeida A, Populo H, Batista R, Lyra J, Pinto V, ric gliomas by single nucleotide polymorphic allele array.
Coelho R, Celestino R, Prazeres H, Lima L et al (2013) Fre- Cancer Res 66:11172–11178. doi:10.1158/0008-5472.
quency of TERT promoter mutations in human cancers. Nature CAN-06-2438
Commun 4:2185. doi:10.1038/ncomms3185 180. Wu G, Broniscer A, McEachron TA, Lu C, Paugh BS, Becks-
168. Wang SC, Hung MC (2009) Nuclear translocation of the epi- fort J, Qu C, Ding L, Huether R, Parker M et al (2012) Somatic
dermal growth factor receptor family membrane tyrosine histone H3 alterations in pediatric diffuse intrinsic pontine glio-
kinase receptors. Clin Cancer Res Off J Am Assoc Cancer Res mas and non-brainstem glioblastomas. Nat Genet 44:251–253.
15:6484–6489. doi:10.1158/1078-0432.CCR-08-2813 doi:10.1038/ng.1102
169. Ward PS, Patel J, Wise DR, Abdel-Wahab O, Bennett BD, 181. Yan H, Parsons DW, Jin G, McLendon R, Rasheed BA, Yuan W,
Coller HA, Cross JR, Fantin VR, Hedvat CV, Perl AE et al Kos I, Batinic-Haberle I, Jones S, Riggins GJ et al (2009) IDH1
(2010) The common feature of leukemia-associated IDH1 and and IDH2 mutations in gliomas. New Eng J Med 360:765–773.
IDH2 mutations is a neomorphic enzyme activity converting doi:10.1056/NEJMoa0808710
alpha-ketoglutarate to 2-hydroxyglutarate. Cancer Cell 17:225– 182. Zarghooni M, Bartels U, Lee E, Buczkowicz P, Morrison A,
234. doi:10.1016/j.ccr.2010.01.020 Huang A, Bouffet E, Hawkins C (2010) Whole-genome pro-
170. Warrington NM, Gianino SM, Jackson E, Goldhoff P, Garbow filing of pediatric diffuse intrinsic pontine gliomas highlights
JR, Piwnica-Worms D, Gutmann DH, Rubin JB (2010) Cyclic platelet-derived growth factor receptor alpha and poly (ADP-
AMP suppression is sufficient to induce gliomagenesis in a ribose) polymerase as potential therapeutic targets. J Clin
mouse model of neurofibromatosis-1. Cancer Res 70:5717– Oncol Off J Am Soc Clin Oncol 28:1337–1344. doi:10.1200/
5727. doi:10.1158/0008-5472.CAN-09-3769 JCO.2009.25.5463

13
 Acta Neuropathol

183. Zhu Y, Guignard F, Zhao D, Liu L, Burns DK, Mason RP, 184. Zhu Y, Harada T, Liu L, Lush ME, Guignard F, Harada C,
Messing A, Parada LF (2005) Early inactivation of p53 tumor Burns DK, Bajenaru ML, Gutmann DH, Parada LF (2005)
suppressor gene cooperating with NF1 loss induces malig- Inactivation of NF1 in CNS causes increased glial progeni-
nant astrocytoma. Cancer Cell 8:119–130. doi:10.1016/j. tor proliferation and optic glioma formation. Development
ccr.2005.07.004 132:5577–5588. doi:10.1242/dev.02162

13