2

Abstract: Glioblastoma (GBM, WHO grade IV astrocytoma) is among the most common adult brain tumors and one that is invariably fatal. GBM is classified as either primary (de novo) or secondary in origin. Secondary GBMs are derived from previously lower grade (WHO grades II or III) gliomas. While secondary GBMs present with similar clinical characteristics as their primary counterparts, the molecular pathways involved in their pathogenesis distinguish the two and have functional consequences for their behavior. Although a large number of histologic markers are routinely utilized to distinguish primary from secondary GBM, advances in genomic and bioinformatics techniques have drastically improved classification of high-grade gliomas and our understanding of the molecular pathways that influence tumor behavior and response to treatment. The significant influence of molecular identity on tumor behavior has been recognized by the most recent WHO classification of CNS tumors, wherein specific molecular markers have been integrated as part of tumor subtype identification process, as a supplement to traditional histological analysis. Indeed, the change heralds a new era for neuro-oncology, one that is moving toward targeted therapeutics as part of the standard of care. Thus, a comprehensive grasp of this diverse landscape is necessary. In this chapter, we provide an overview of our latest understanding of the molecular diversity of GBM, specifically as it pertains to primary and secondary GBMs, and how it influences prognostication and therapeutic decision-making.

Keywords: Alpha thalassemia/mental retardation syndrome X-linked (ATRX); Isocitrate dehydrogenase (IDH); Low-grade glioma; Secondary glioblastoma

Introduction

Glioblastoma (GBM, WHO grade IV astrocytoma) is the most common malignant primary brain tumor among adults. Despite aggressive therapy, the current median survival is approximately 15 months (1). In addition to the diffusely infiltrative nature of these tumors, which prevents complete surgical resection, tumor recurrence and ultimate patient demise is also largely attributed to the significant molecular and cellular heterogeneity of these lesions, which inevitably results in treatment resistance and tumor recurrence. GBMs are further classified into primary (de novo) and secondary tumors that, while they present with similar clinical characteristics, are derived from previously lower grade (WHO grades II or III) gliomas. While both categories are diffuse in nature, the molecular pathways involved, along with functional tumor behavior, treatment strategy, and clinical outcomes are different (2, 3). Although clinical and imaging biomarkers can be used to distinguish primary from secondary GBM, advances in genomic and bioinformatics techniques have drastically improved classification of high-grade gliomas and our understanding of the molecular pathways that influence tumor behavior and response to treatment. The significant influence of molecular identity on tumor behavior has been recognized by the most recent WHO classification of CNS tumors, wherein specific molecular markers have been integrated as part of tumor subtype identification process, as a supplement to traditional histological analysis (4). Indeed, the change heralds a new era for neuro-oncology, one that is moving toward targeted therapeutics as part of the standard of care. Thus, a comprehensive grasp of this diverse landscape is necessary. In this chapter, we provide an overview of our latest understanding of the molecular diversity of GBM, specifically as it pertains to primary and secondary GBMs, and how it influences prognostication and therapeutic decision-making.

Distinguishing Primary and Secondary GBMs

Primary and secondary GBMs are histologically indistinguishable. Historically, the distinction between the two has been based on clinical history. With a more in-depth understanding of the genetic, epigenetic, and molecular profile of these tumors, however, the distinction has become clearer (Table 1) (5).

TABLE 1 Key Characteristics of IDH-Wildtype and IDH-Mutant Glioblastomas (adapted from Ref. (5).)

	IDH-WT GBM	IDH-mutant GBM
Synonym	Primary glioblastoma	Secondary glioblastoma
Precursor lesion	Identified de novo	Diffuse astrocytoma Anaplastic astrocytoma
Proportion of glioblastomas	~90%	~10%
Median age at diagnosis	~62 years	~44 years
M:F ratio	1.42:1	1.05:1
Median length of clinical history at diagnosis	4 months	15 months
Median overall survival
Surgery + radiotherapy	9.9 months	24 months
Surgery + RT + CTX	15 months	31 months
Location	Supratentorial	Preferentially frontal
Necrosis	Extensive	Limited
TERT promoter mutations	72%	26%
TP53 mutations	27%	81%
ATRX mutations	Exceptional	71%
EGFR amplification	35%	Exceptional
PTEN mutations	24%	Exceptional

ATRX, adult thalassemia mental retardation x-linked; CTX, chemotherapy; EGFR, epidermal growth factor receptor; GBM, Glioblastoma multiforme; IDH, Isocitrate dehydrogenase; PTEN, phosphatase and tensin homolog; TERT, telomerase reverse transcriptase; TP53, tumor protein 53; RT, radiotherapy.

EPIDEMIOLOGY OF SECONDARY GBM

The incidence of secondary GBMs based on clinical and imaging criteria is somewhat lower than that estimated by isocitrate dehydrogenase (IDH) status (5% vs. 6–13%, respectively) (2, 6, 7). Furthermore, patients with a clinical diagnosis of secondary GBM are on average 17 years younger than those with primary GBM (2, 7); this bias toward a younger patient cohort correlates very closely with IDH1 status, as patients with IDH mutations are substantially younger (8, 9). The clinical course is substantially longer in patients with IDH-mutant GBM, indicative of a less aggressive behavior (2, 6, 8).

ANATOMIC PREVALENCE OF SECONDARY GBM

Interestingly, IDH-mutant GBM has a predilection for the frontal lobe and typically present with seizure rather than neurological deficit. The same has been demonstrated for IDH-mutant Grade II astrocytomas and oligodendrogliomas, including tumor with 1p/19q co-deletion (10). These findings support a hypothesis that the precursor cell of origin among IDH-mutant tumor subtypes is shared, and suggest that these tumors may arise from mutations within a cell population that is independent of the cell populations at risk during development of de novo GBM (11).

MOLECULAR LANDSCAPE OF SECONDARY GBM

Amplification of the EGFR gene and activating mutations of its protein product are hallmarks of primary GBM and appear to be exclusive of TP53 mutations (12). PTEN amplification and loss of chromosome 10 are additional features of primary GBMs (3, 13). Both primary and secondary GBMs have in common loss of heterozygosity (LOH) at chromosome 10q (14–16); although PTEN is also located on chromosome 10, mutations in this gene are only observed in primary GBM. Therefore, additional genetic events must be responsible for oncogenesis of high-grade gliomas that is shared among both primary and secondary tumors.

One of the earliest events, if not the initial event, in gliomagenesis is mutation of the IDH1 or IDH2 gene. Mutations in the promoter of the telomerase reverse transcriptase (TERT) gene lead to enhanced telomerase activity, which results in maintenance of telomere length and promotion of cell survival. Interestingly, TERT mutation is shared among both primary and secondary GBMs, potentially rendering this mutation as an early event in the process of tumorigenesis (17). In addition to these mutations, secondary GBM originating from a lower grade astrocytoma will frequently display mutations in the TP53 and ATRX (adult thalassemia mental retardation x-linked) genes, while anaplastic tumors arising from a lower grade oligodendroglioma lineage will have co-deletions of 1p and19q (2, 3, 18). There are several key signaling pathways involved in this transformation as well, and knowledge of mutations in genes involved in these processes and pathways is critical for an in-depth understanding of the biology of secondary GBM and in working toward targeted therapeutics. We will review these pathways in detail below.

Molecular Classification of GBMs Based on Gene Expression

In 2010, Verhaak and colleagues analyzed somatic mutations, DNA copy-number alterations, and gene expression profiling to group GBMs into discrete categories (19). Through this work, they were able to establish four subtypes of GBMs (Classic, Proneural, Neural, and Mesenchymal) based on the specific clustering of molecular and gene expression profiles. The Classic category demonstrated a greater preponderance of EGFR amplification, decreased rates of TP53 mutation, along with p16INK4A and p14ARF deletion. Histologically, the Classic subtype demonstrated features more consistent with astrocytes. The Proneural category was found to have a greater rate of PDGFR amplification, TP53 mutation, LOH, and IDH1 mutation. These tumors had histological features most consistent with oligodendrocytes. Moreover, patients harboring the Proneural subtypes were younger and responded better to therapy. The Neural subtype was found to have a greater degree of neuronal marker expression and the histology was consistent with a combination of oligodendroglial, astrocytoic, and neuronal features. The Mesenchymal subtype was found to have a greater degree of NF1 mutations, along with alterations of PTEN and Akt. Histologically, these tumors demonstrated a greater degree of necrosis and inflammatory features. Furthermore, astroglial and microglial cell signatures were commonly noted. This landmark study established the concept of differential behavior of GBMs that may be similar histologically but differ substantially from a molecular and gene expression perspective.

Mechanisms of Gliomagenesis

Gliomagenesis is a multicomponent process involving several genetic mutations affecting numerous molecular pathways (Figure 1). When considering tumor phylogeny, IDH mutation is critical to deciphering whether the identified tumor is a primary GBM or a GBM arising from secondary progression of a lower grade glioma. It is now established that while IDH mutations are early events in the process of gliomagenesis in secondary GBM, additional genes and their end products are altered during this process and these include ATRX mutation, loss of tumor suppressor genes such as TP53 and RB1, and mutations in the promoter of TERT (5). Alterations of chromosomes 1, 7, 10, and 19, each harboring a distinct subset of tumor suppressor/promoter genes, are pivotal as well. Distinct pathways that have been identified as part of the core drivers of gliomagenesis include the EGFR/PTEN/Akt/mTOR, TP53/MDM2/p14ARF, and the p16INK4a/RB1 pathways, which will be elaborated upon in the subsequent sections.

Figure 1 Molecular pathways to gliomagenesis. While the cell of origin in glioma is yet to be identified, large-scale expression and copy-number analyses have determined multiple molecular processes that result in glioma formation. Primary glioblastomas (and most Grade I gliomas) arise via an IDH-independent pathway. Conversely, IDH mutation is an early if not initiating event in the development of of low-grade astrocytomas and oligodendrogliomas. By definition, secondary glioblastomas arise from malignant degeneration of an IDH-mutant lower grade tumor.

IDH AND GLIOMA INITIATION

First reported by Parsons and colleagues in 2008, a number of recent studies have since confirmed recurrent somatic mutations in the IDH1 and IDH2 genes (R132H and R172K as the canonical mutations in these genes, respectively) in a significant proportion of patients with gliomas. Further, patients who harbored tumors with an IDH mutation exhibit distinct disease characteristics relative to patients with a glioma with wild-type (WT) IDH. In 615 WHO grade II/III gliomas, IDH mutations were identified in 79% of the patient tumors (17). In another series of 457 WHO grade II/III gliomas, 80.7% of the patients were found to harbor an IDH mutation (20). The Cancer Genome Atlas Research Network found an IDH mutation in 226 (80.1%) of 282 WHO grade II/III gliomas (21). Based on these results, the WHO now recognizes IDH mutation as a critical biomarker in the classification of gliomas (4).

The IDH enzymes catalyze the oxidative conversion of isocitrate to α-ketoglutarate (α-KG). IDH mutations confer a gain-of-function neomorphic activity, converting α-KG to R-2-hydroxyglutarate (R-2-HG), instead of its racemic enantiomer S-2-HG. Although 2-HG is a trace metabolic product in normal cells, it is markedly elevated in IDH-mutant gliomas and in other malignancies, such as acute myeloid leukemia (22–24). The oncogenic effect of IDH mutation is thought to be twofold. First, 2-HG is considered an oncometabolite that may play a role in the process of glioma development, and progression or resistance to treatment. Although the exact role of IDH1 mutation in gliomagenesis had initially been hampered by difficulties in establishing in vitro cultures with IDH1 mutations (25), recent reports have demonstrated that increased levels of 2-HG result in increased activity of HIF-1-α and increased levels of its downstream targets such as VEGF. In addition, 2HG also affects collagen maturation, resulting in defective basement membranes that are potentially pivotal to glioma progression (25). Second, IDH mutation results in decreased production of α-KG, which impairs the function of many α-KG-dependent dioxygenases, including but not limited to histone demethylases (e.g., collagen prolyl-4-hydroxylase, prolyl hydroxylases, and the ten-eleven translocation (TET) family of DNA hydroxylases) (26). Change in histone methylation is thought to also interfere with the terminal differentiation of cells and may predispose cells harboring mutant IDH to malignant transformation (27). Based on the above evidence, IDH1/2 mutations have been termed as lineage markers by some authors (11), and it is now accepted as a more definitive marker of secondary GBM than any other clinical or pathological criterion (28).

ATRX, TP53, AND 1p/19q

The great majority of low-grade astrocytomas carry a p53 mutation while most oligodendrogliomas demonstrate loss of chromosomes 1p and 19q (26, 29–33). Biopsy-based studies suggest that the IDH1 mutation occurs prior to either p53 mutation or 1p and 19q loss (26, 33). Following IDH-mediated oncogenesis, acquisition of p53 and ATRX mutations occurs in the setting of development of an astrocytoma (34, 35), while loss of chromosomes 1p and 19q occurs in the setting of development of an oligodendroglioma. While both subgroups are capable with time of undergoing further malignant degeneration, the current WHO classification system only considers progression to secondary GBM as an endpoint of astrocytoma progression. It is conceivable that all GBMs that harbor an IDH mutation are secondary tumors. In one study, the small subgroup of patients with primary GBM carrying an IDH mutation (3.4%) was younger than noncensored primary GBM patients and harbored frequent p53 mutations and an absence of EGFR amplification, features consistent with secondary GBMs (8). These findings suggest that these tumors could represent cases of a rapidly progressive secondary GBM, rather than a true primary GBM. Conversely, it can be argued that all GBMs harboring a WT IDH are biologically primary GBMs: cases of secondary GBM without an IDH mutation likely represent a progression from an undergraded, lower grade, or anaplastic glioma (8). These assumptions are borne out by recent data that show that gliomas lacking mutation in IDH or having chromosomal loss at 1p and 19q cluster by expression analysis and DNA copy-number profiling (21) and portend a severe prognosis (17). With an increased understanding of molecular markers and their incorporation into clinical trials, the disparity between molecular markers and histopathology-based diagnostics methods becomes more evident. For now, the current WHO classification system posits that, despite histopathological features such as neo-vascularity and necrosis, a high-grade glioma with IDH1 mutation and 1p/19q co-deletion should be considered an anaplastic oligodendroglioma. Conversely, from a biological perspective, a histological anaplastic astrocytoma with WT IDH is now considered a GBM (36). These modifications in the classification system have been corroborated by outcomes data emerging from clinical trials. Together, these findings confirm the integral role of IDH and 1p/19q status in determining patient survival.

TERT PROMOTER MUTATION

Mutations in the TERT gene are thought to prevent cell senescence through increased telomere length, thus promoting tumorigenesis in several cancers, including GBM (37). The contribution of TERT mutation to tumor aggressiveness however is not clear. Focusing on a sample of GBM cases, Mosrati et al. found that TERT promoter mutation was associated with a shorter overall survival (37). Interestingly, this mutation was found in both primary and secondary tumors. More recently, Eckel-Passow et al. found that, while GBMs had a higher proportion of TERT mutations in isolation (74% of cases) or had neither TERT or IDH mutations or loss of chromosome 1p and 19q (what they termed “triple negative” tumors, making up17% of cases), lower grade gliomas were much less likely to be “triple negative” (7% of cases) or harbor a TERT mutation in isolation (10% of cases) (16). These findings suggest that while TERT promoter mutation is integral to tumorigenesis and may contribute to the overall aggressiveness of the tumor, its role is modified by other key mutations.

THE G-CIMP PHENOTYPE

Methylation of the promoter region of the MGMT gene is more frequently found in secondary GBMs compared to primary GBMs (75% vs. 36%) (38), and it is frequently associated with mutations in IDH1/2 and TP53 and utilized as a strong predictive marker for response to chemotherapy in GBM patients. In fact, IDH mutation has been shown to mediate widespread changes in chromosome structure and remodeling of the DNA methylome, resulting in the establishment of the glioma CpG island methylator phenotype (G-CIMP). Introduction of mutant IDH1 into primary human astrocytes was found to be sufficient to alter specific histone methylation marks and induce extensive DNA hypermethylation in a manner that resembles the changes observed in G-CIMP+ lower grade gliomas. Furthermore, the epigenomic alterations resulting from mutant IDH1 activate specific gene expression programs that are associated with G-CIMP+ proneural glioblastoma, but not other glioblastoma subtypes, and are associated with longer survival. Based on these data, IDH mutation is likely the molecular basis of G-CIMP in gliomas, highlighting the interplay between genomic and epigenomic changes in cancers including GBM.

In GBM, the proneural subtype is predominantly associated with IDH1/2 mutations and these are further subclassified as either CIMP+ or CIMP- (of which the G-CIMP+ shows better prognosis). The proneural subtype by itself, however, appears to bear little prognostic significance unless considered in association with the IDH1/2 mutation status (39). In fact, Turcan et al. have demonstrated that the IDH1 mutation alone is capable of remodeling the genomic methylation profile of the tumor, thus promoting the CIMP+ profile (40). Interestingly, WT IDH1 status promoted hypomethylation at numerous foci and CIMP- low-grade gliomas lacked IDH1 mutation. In addition, decreased expression of ATRX is associated with downregulation of MGMT expression via promoter hypermethylation (41). Therefore, ATRX mutation status not only predicts cell of origin but also has a significant prognostic role as well (34, 42, 43).

Genetics of Glioma Progression

EGFR/PTEN/AKT/MTOR PATHWAY

Activation of the PI3K/Akt pathway results in increased cell proliferation via downregulation of p27, thereby influencing cell-cycle progression (44), inactivation of pro-apoptotic genes (45), and increased transcription of pro-survival genes under the influence of NFkB (46). PI3K is recruited to the cell surface and activated through EGFR. Once phosphorylated, PI3K activates PIP3 via phosphorylation, which induces activation of downstream molecules such as Akt—a serine/threonine kinase (47)—promoting cell survival and proliferation (48).

EGFR is a tyrosine kinase growth factor receptor situated in the cell membrane. Amplification of the EGFR gene and mutation of the protein product are key contributors to the activation of this receptor tyrosine kinase (RTK) pathway in primary GBM. The most common of the EGFR-activating mutants is the EGFRvIII variant, in which gene mutation results in a truncated protein product that is constitutively active. Mutations in Akt itself, however, are not common in gliomas (49).

PTEN is the second most commonly mutated tumor suppressor gene in all cancers after p53 (50), and PTEN mutation is found in approximately 40% of GBMs, predominantly in the primary form (51). PTEN is a tumor suppressor and one of its functions is dephosphorylation of PIP3, thus preventing activation of Akt and mTOR (47). Through this role, PTEN is central in inhibiting cell proliferation and regulating the ability of cells in migration and invasion (52). Loss of PTEN function, either through genetic or epigenetic modifications, is a common component of the Akt/PI3K/mTOR activation pathway in cancer.

TP53/MDM2/P14ARF PATHWAY

Although mutations of the TP53 gene have been identified in both primary and secondary GBMs, its role appears to be predominantly related to the latter, where the mutation is an early event in gliomagenesis (2). While p53 mutations in primary GBM appear to involve all exons indiscriminately, they are predominantly focused at codons 248 and 273, particularly involving CpG sites, in secondary GBM (2). This discrepancy suggests that p53 mutation in secondary GBM is a specific and stereotyped event in secondary GBM ontology, while p53 mutation in primary GBM is potentially a consequence of widespread genomic instability (3).

MDM2 amplification appears to be specific to primary GBMs that lack the p53 mutation (53, 54). In normal cells, WT p53 induces the expression of MDM2, which in turn inhibits the function of WT p53. Furthermore, WT p53 inhibits the function of p14ARF, which would normally inhibit the downregulation of p53 by MDM2. This autoregulatory loop is disrupted when any of the above is dysfunctional, adversely affecting cell-cycle control, DNA damage repair, cell proliferation/differentiation, and neovascularization (55).

P16INK4A/RB1 PATHWAY

Either through homozygous deletion or promoter methylation, the alteration of p16INK4a is an important step in both primary and secondary GBMs (56). Conversely, methylation of the RB1 promoter, correlating with decreased RB1 expression, is more specific to secondary GBM (57). The p16INK4a/RB1 pathway is critical to cell-cycle control (58), as RB1 regulates the progression of the cell cycle from G1 to the S phase by preventing the release of the E2F transcription factor. The latter enables the transcription of genes required for cell-cycle progression, in addition to p14ARF. The phosphorylation of RB1, via the CDK4/cyclin D complex, inhibits this function enabling the progression of the cell cycle along with increased p53 expression via the activated p14ARF. WT p16INK4a serves as an additional checkpoint by binding to CDK4 and inhibiting the function of the CDK4/cyclin D complex. Therefore, altered expression of any of these genes results in an inability to control cell-cycle progression. The central role of cell-cycle regulation in the genesis of secondary GBM has also been confirmed with cDNA expression profile analysis (59).

Effect of Treatment on Glioma Transformation

By virtue of the inherent heterogeneity of these tumors, it is expected that not all of the cells within a glioma will respond to chemotherapy and radiation, inevitably resulting in tumor progression/recurrence. Further, recent evidence suggests that chemotherapy and radiation may actually result in mutations that promote tumor cell survival. This pro-mutational ability has been most extensively studied in the setting of temozolamide (TMZ) and ionizing radiation.

TEMOZOLAMIDE AND LGG PROGRESSION

An alkylating agent, TMZ is an integral component of the standard treatment regimen for patients with GBM. Accumulating evidence from numerous studies suggests that acquired treatment resistance following TMZ administration is multifactorial and rooted in transcriptional, metabolomic, genomic, and epigenomic changes that lead to this phenotype (60–67).

Costello and colleagues undertook genome sequence analysis of 23 initial and matched recurrent human gliomas to address two questions: (i) What is the extent to which mutations in initial tumors differ from mutations in their subsequent recurrent tumors? (ii) How does chemotherapy with TMZ affect the mutational profile of recurrent tumors? The authors found an average of 33 somatic coding mutations in each initial tumor, of which an average of 54% were also detected at recurrence (shared mutations), including mutations in IDH1, TP53, and ATRX. All other somatic mutations were identified only in the initial tumor or only in the recurrent tumor from a given patient (private mutations), though overall, the initial and recurrent gliomas displayed a broad spectrum of genetic relatedness. Interestingly, in multiple patients, the recurrent tumors shared ≤25% of mutations detected in the initial tumors, suggesting that these tumors were seeded by cells derived from the initial tumor at an early stage of its evolution, and that tumor recurrence can occur as the result of either linear or branched evolution.

Their findings regarding the effect of TMZ on tumor evolution and recurrence were as striking. Although the initial tumors and most of the recurrent tumors in their cohort had 0.2 to 4.5 mutations per megabase (Mb), 6 of the 10 patients treated with TMZ had recurrent tumors that were hypermutated; that is, they harbored 31.9 to 90.9 mutations per Mb. Overall, 97% of these were C>T/G>A transitions predominantly occurring at CpC and CpT dinucleotides, which is a signature of TMZ-induced mutagenesis distinct from nonhypermutated tumors. Further, acquisition of DNA mismatch repair (MMR) pathway dysfunction, which results in resistance to TMZ, appeared to exacerbate hypermutation in the face of continued TMZ therapy. The authors postulated that introduction of thousands of de novo mutations could drive the evolution of TMZ-resistant glioma cells to higher states of malignant potential. Indeed, all six recurrent tumors that showed evidence of TMZ-induced hypermutation underwent malignant progression to GBM. Many of these tumors developed mutations in pathways described as critical to gliomagenesis, including Akt-mTOR and the p16/RB. Treatment-induced somatic mutations were recently longitudinally studied in a patient with a 5-year survival period following initial diagnosis (68). Using whole exome sequencing, the investigators demonstrated that each successive therapy selected for resistant clones of tumor cells and that these had arisen via the process of chromothripsis. In addition, this approach enabled the provision of personalized therapy for this patient, based on the identification of target clonal populations sensitive to available treatment, which was critical for this long-term survival. Given the evidence derived from such analyses, it is clear that the genome of GBMs is dynamic and in order to offer true personalized treatment, the genome of each successive tumor population must be investigated thoroughly.

Stepaneko et al. extended these findings with in vitro studies that demonstrated that long-term exposure of glioma cells to TMZ induces chromosomal instability, leading to alteration of cell growth, invasiveness, migration, and response to retreatment (69). Among the TMZ-resistant cell lines, some responded to temsirolimus, an mTOR inhibitor. Interestingly, although TMZ has been shown to induce the transformation of glioma nonstem-like cells into glioma stem-like cells, the sensitivity of both differentiated and stem-like cells to TMZ was similar (70, 71). These findings further highlight the importance of the evolution of the genetic network that infers TMZ resistance in GBM.

EFFECT OF RADIATION ON GLIOMA BEHAVIOR

The introduction of radiation therapy to the armamentarium of therapy in patients with GBM has been a significant contribution. However, similar to TMZ, radiation is thought to promote malignant progression of gliomas as well. Based on transcription profiling of patient-derived radiation-resistant GBM cells, the mesenchymal subtype was the most commonly identified (72). In vitro studies have also demonstrated a proneural to mesenchymal transition among oligodendroglioma cell cultures that were irradiated (73). The authors proposed that the activation of the STAT3 transcription factor following radiation was contributory, given that its inhibition prevented this transition. Furthermore, Jak2 inhibition in mice undergoing radiation prolonged their survival. Alternative mechanisms such as activation of the TNF-α/NFkB pathway may also be involved (72). Other post-translational effects of radiation exposure, such as the stabilization of HIF-1α, promoting angiogenesis, have been proposed (74). Therefore, a combination of intrinsic cell changes and modifications to the tumor microenvironment may be responsible for the radiation-induced malignant progression noted in gliomas.

Conclusion

The recent publication of the modified WHO classification for CNS tumors, integrating molecular signatures into histological-based classifications, is timely and reflects the field’s evolution. Based on our understanding of the vast intratumoral heterogeneity among GBMs, the logical next step is to establish biomarkers that would be predictive of treatment response, identify clonal populations that are potentially resistant to therapy, and develop combination therapies tailored to the specific pathways involved within the entirety of the tumor. Analysis of initial and recurrent tumor samples may be helpful for better clonal evolution analysis.

Acknowledgment: Sunit Das is supported by grants and awards from the American College of Surgeons, Canadian Institutes of Health Research, and Megan’s Walk.

Conflict of interest: The authors declare no potential conflicts of interest with respect to research, authorship, or publication of this manuscript.

Copyright and permission statement: To the best of our knowledge, the materials included in this chapter do not violate copyright laws. All original sources have been appropriately acknowledged and/or referenced. Where relevant, appropriate permissions have been obtained from the original copyright holder(s).

REFERENCES

1. Stupp R, Mason WP, van den Bent MJ, Weller M, Fisher B, Taphoorn MJ, et al. Radiotherapy plus concomitant and adjuvant temozolomide for glioblastoma. N Engl J Med. 2005;352(10):987–96. http://dx.doi.org/10.1056/NEJMoa043330
2. Ohgaki H, Dessen P, Jourde B, Horstmann S, Nishikawa T, Di Patre PL, et al. Genetic pathways to glioblastoma: A population-based study. Cancer Res. 2004;64(19):6892–9. http://dx.doi.org/10.1158/0008-5472.CAN-04-1337
3. Ohgaki H, Kleihues P. Genetic pathways to primary and secondary glioblastoma. Am J Pathol. 2007;170(5):1445–53. http://dx.doi.org/10.2353/ajpath.2007.070011
4. Louis DN, Perry A, Reifenberger G, von Deimling A, Figarella-Branger D, Cavenee WK, et al. The 2016 World Health Organization classification of tumors of the central nervous system: A summary. Acta Neuropathol. 2016;131(6):803–20. http://dx.doi.org/10.1007/s00401-016-1545-1
5. Costello JF, Plass C, Arap W, Chapman VM, Held WA, Berger MS, et al. Cyclin-dependent kinase 6 (CDK6) amplification in human gliomas identified using two-dimensional separation of genomic DNA. Cancer Res. 1997;57(7):1250–4.
6. Ohgaki H, Kleihues P. Population-based studies on incidence, survival rates, and genetic alterations in astrocytic and oligodendroglial gliomas. J Neuropathol Exp Neurol. 2005;64(6):479–89. http://dx.doi.org/10.1093/jnen/64.6.479
7. Dropcho EJ, Soong SJ. The prognostic impact of prior low grade histology in patients with anaplastic gliomas: A case-control study. Neurology. 1996;47(3):684–90. http://dx.doi.org/10.1212/WNL.47.3.684
8. Nobusawa S, Watanabe T, Kleihues P, Ohgaki H. IDH1 mutations as molecular signature and predic-tive factor of secondary glioblastomas. Clin Cancer Res. 2009;15(19):6002–7. http://dx.doi.org/10.1158/1078-0432.CCR-09-0715
9. Ichimura K, Pearson DM, Kocialkowski S, Backlund LM, Chan R, Jones DT, et al. IDH1 mutations are present in the majority of common adult gliomas but rare in primary glioblastomas. Neuro Oncol. 2009;11(4):341–7. http://dx.doi.org/10.1215/15228517-2009-025
10. Lai A, Kharbanda S, Pope WB, Tran A, Solis OE, Peale F, et al. Evidence for sequenced molecular evolu-tion of IDH1 mutant glioblastoma from a distinct cell of origin. J Clin Oncol. 2011;29(34):4482–90. http://dx.doi.org/10.1200/JCO.2010.33.8715
11. Ohgaki H, Kleihues P. The definition of primary and secondary glioblastoma. Clin Cancer Res. 2013;19(4):764–72. http://dx.doi.org/10.1158/1078-0432.CCR-12-3002
12. Watanabe K, Tachibana O, Sata K, Yonekawa Y, Kleihues P, Ohgaki H. Overexpression of the EGF receptor and p53 mutations are mutually exclusive in the evolution of primary and second-ary glioblastomas. Brain Pathol. 1996;6(3):217–23; discussion 23–4. http://dx.doi.org/10.1111/j.1750-3639.1996.tb00848.x
13. Fujisawa H, Reis RM, Nakamura M, Colella S, Yonekawa Y, Kleihues P, et al. Loss of heterozygosity on chromosome 10 is more extensive in primary (de novo) than in secondary glioblastomas. Lab Invest. 2000;80(1):65–72. http://dx.doi.org/10.1038/labinvest.3780009
14. Ichimura K, Schmidt EE, Miyakawa A, Goike HM, Collins VP. Distinct patterns of deletion on 10p and 10q suggest involvement of multiple tumor suppressor genes in the development of astrocytic gliomas of different malignancy grades. Genes Chromosomes Cancer. 1998;22(1):9–15. http://dx.doi.org/10.1002/(SICI)1098-2264(199805)22:1%3C9::AID-GCC2%3E3.0.CO;2-1
15. Karlbom AE, James CD, Boethius J, Cavenee WK, Collins VP, Nordenskjold M, et al. Loss of hetero-zygosity in malignant gliomas involves at least three distinct regions on chromosome 10. Human Genet. 1993;92(2):169–74. http://dx.doi.org/10.1007/BF00219686
16. Rasheed BK, McLendon RE, Friedman HS, Friedman AH, Fuchs HE, Bigner DD, et al. Chromosome 10 deletion mapping in human gliomas: A common deletion region in 10q25. Oncogene. 1995;10(11):2243–6.
17. Eckel-Passow JE, Lachance DH, Molinaro AM, Walsh KM, Decker PA, Sicotte H, et al. Glioma groups based on 1p/19q, IDH, and TERT promoter mutations in tumors. N Engl J Med. 2015;372(26):2499–508. http://dx.doi.org/10.1056/NEJMoa1407279
18. Nakamura M, Yang F, Fujisawa H, Yonekawa Y, Kleihues P, Ohgaki H. Loss of heterozygosity on chro-mosome 19 in secondary glioblastomas. J Neuropathol Exp Neurol. 2000;59(6):539–43. http://dx.doi.org/10.1093/jnen/59.6.539
19. Verhaak RG, Hoadley KA, Purdom E, Wang V, Qi Y, Wilkerson MD, et al. Integrated genomic analysis identifies clinically relevant subtypes of glioblastoma characterized by abnormalities in PDGFRA, IDH1, EGFR, and NF1. Cancer Cell. 2010;17(1):98–110. http://dx.doi.org/10.1016/j.ccr.2009.12.020
20. Ceccarelli M, Barthel FP, Malta TM, Sabedot TS, Salama SR, Murray BA, et al. Molecular profiling reveals biologically discrete subsets and pathways of progression in diffuse glioma. Cell. 2016;164(3):550–63. http://dx.doi.org/10.1016/j.cell.2015.12.028
21. Cancer Genome Atlas Research Network, Brat DJ, Verhaak RG, Aldape KD, Yung WK, Salama SR, et al. Comprehensive, integrative genomic analysis of diffuse lower-grade gliomas. N Engl J Med. 2015;372(26):2481–98. http://dx.doi.org/10.1056/NEJMoa1402121
22. Dang L, White DW, Gross S, Bennett BD, Bittinger MA, Driggers EM, et al. Cancer-associated IDH1 mutations produce 2-hydroxyglutarate. Nature. 2009;462(7274):739–44. http://dx.doi.org/10.1038/nature08617
23. Fathi AT, Nahed BV, Wander SA, Iafrate AJ, Borger DR, Hu R, et al. Elevation of urinary 2-hydroxy-glutarate in IDH-mutant glioma. Oncologist. 2016;21(2):214–19. http://dx.doi.org/10.1634/theoncologist.2015-0342
24. Fathi AT, Sadrzadeh H, Borger DR, Ballen KK, Amrein PC, Attar EC, et al. Prospective serial evaluation of 2-hydroxyglutarate, during treatment of newly diagnosed acute myeloid leukemia, to assess disease activity and therapeutic response. Blood. 2012;120(23):4649–52. http://dx.doi.org/10.1182/blood-2012-06-438267
25. Sasaki M, Knobbe CB, Itsumi M, Elia AJ, Harris IS, Chio, II, et al. D-2-hydroxyglutarate produced by mutant IDH1 perturbs collagen maturation and basement membrane function. Genes Dev. 2012;26(18):2038–49. http://dx.doi.org/10.1101/gad.198200.112
26. Watanabe T, Nobusawa S, Kleihues P, Ohgaki H. IDH1 mutations are early events in the development of astrocytomas and oligodendrogliomas. Am J Pathol. 2009;174(4):1149–53. http://dx.doi.org/10.2353/ajpath.2009.080958
27. Lu C, Ward PS, Kapoor GS, Rohle D, Turcan S, Abdel-Wahab O, et al. IDH mutation impairs histone demethylation and results in a block to cell differentiation. Nature. 2012;483(7390):474–8. http://dx.doi.org/10.1038/nature10860
28. Parsons DW, Jones S, Zhang X, Lin JC, Leary RJ, Angenendt P, et al. An integrated genomic analysis of human glioblastoma multiforme. Science. 2008;321(5897):1807–12. http://dx.doi.org/10.1126/science.1164382
29. Louis DN, Ohgaki H, Wiestler OD, Cavenee WK, Burger PC, Jouvet A, et al. The 2007 WHO classifi-cation of tumours of the central nervous system. Acta Neuropathol. 2007;114(2):97–109. http://dx.doi.org/10.1007/s00401-007-0243-4
30. Okamoto Y, Di Patre PL, Burkhard C, Horstmann S, Jourde B, Fahey M, et al. Population-based study on incidence, survival rates, and genetic alterations of low-grade diffuse astrocytomas and oligoden-drogliomas. Acta Neuropathol. 2004;108(1):49–56. http://dx.doi.org/10.1007/s00401-004-0861-z
31. Watanabe T, Nakamura M, Kros JM, Burkhard C, Yonekawa Y, Kleihues P, et al. Phenotype versus genotype correlation in oligodendrogliomas and low-grade diffuse astrocytomas. Acta Neuropathol. 2002;103(3):267–75. http://dx.doi.org/10.1007/s004010100464
32. Reifenberger G, Louis DN. Oligodendroglioma: Toward molecular definitions in diagnostic neuro-oncology. J Neuropathol Exp Neurol. 2003;62(2):111–26. http://dx.doi.org/10.1093/jnen/62.2.111
33. Kim YH, Nobusawa S, Mittelbronn M, Paulus W, Brokinkel B, Keyvani K, et al. Molecular classifica-tion of low-grade diffuse gliomas. Am J Pathol. 2010;177(6):2708–14. http://dx.doi.org/10.2353/ajpath.2010.100680
34. Jiao Y, Killela PJ, Reitman ZJ, Rasheed AB, Heaphy CM, de Wilde RF, et al. Frequent ATRX, CIC, FUBP1 and IDH1 mutations refine the classification of malignant gliomas. Oncotarget. 2012;3(7): 709–22. http://dx.doi.org/10.18632/oncotarget.588
35. Liu XY, Gerges N, Korshunov A, Sabha N, Khuong-Quang DA, Fontebasso AM, et al. Frequent ATRX mutations and loss of expression in adult diffuse astrocytic tumors carrying IDH1/IDH2 and TP53 mutations. Acta Neuropathol. 2012;124(5):615–25. http://dx.doi.org/10.1007/s00401-012-1031-3
36. Aldape K, Zadeh G, Mansouri S, Reifenberger G, von Deimling A. Glioblastoma: Pathology, molecular mechanisms and markers. Acta Neuropathol. 2015;129(6):829–48. http://dx.doi.org/10.1007/s00401-015-1432-1
37. Mosrati MA, Malmstrom A, Lysiak M, Krysztofiak A, Hallbeck M, Milos P, et al. TERT promoter muta-tions and polymorphisms as prognostic factors in primary glioblastoma. Oncotarget. 2015;6(18): 16663–73. http://dx.doi.org/10.18632/oncotarget.4389
38. Nakamura M, Watanabe T, Yonekawa Y, Kleihues P, Ohgaki H. Promoter methylation of the DNA repair gene MGMT in astrocytomas is frequently associated with G:C --> A:T mutations of the TP53 tumor suppressor gene. Carcinogenesis. 2001;22(10):1715–19. http://dx.doi.org/10.1093/carcin/22.10.1715
39. Brennan CW, Verhaak RG, McKenna A, Campos B, Noushmehr H, Salama SR, et al. The somatic genomic landscape of glioblastoma. Cell. 2013;155(2):462–77. http://dx.doi.org/10.1016/j.cell.2013.09.034
40. Turcan S, Rohle D, Goenka A, Walsh LA, Fang F, Yilmaz E, et al. IDH1 mutation is sufficient to estab-lish the glioma hypermethylator phenotype. Nature. 2012;483(7390):479–83. http://dx.doi.org/10.1038/nature10866
41. Cai J, Chen J, Zhang W, Yang P, Zhang C, Li M, et al. Loss of ATRX, associated with DNA methylation pattern of chromosome end, impacted biological behaviors of astrocytic tumors. Oncotarget. 2015;6(20):18105–15. http://dx.doi.org/10.18632/oncotarget.3906
42. Wiestler B, Capper D, Holland-Letz T, Korshunov A, von Deimling A, Pfister SM, et al. ATRX loss refines the classification of anaplastic gliomas and identifies a subgroup of IDH mutant astrocytic tumors with better prognosis. Acta Neuropathol. 2013;126(3):443–51. http://dx.doi.org/10.1007/s00401-013-1156-z
43. Cai J, Yang P, Zhang C, Zhang W, Liu Y, Bao Z, et al. ATRX mRNA expression combined with IDH1/2 mutational status and Ki-67 expression refines the molecular classification of astrocytic tumors: Evidence from the whole transcriptome sequencing of 169 samples samples. Oncotarget. 2014;5(9):2551–61. http://dx.doi.org/10.18632/oncotarget.1838
44. Narita Y, Nagane M, Mishima K, Huang HJ, Furnari FB, Cavenee WK. Mutant epidermal growth fac-tor receptor signaling down-regulates p27 through activation of the phosphatidylinositol 3-kinase/ Akt pathway in glioblastomas. Cancer Res. 2002;62(22):6764–9.
45. Datta SR, Dudek H, Tao X, Masters S, Fu H, Gotoh Y, et al. Akt phosphorylation of BAD couples survival signals to the cell-intrinsic death machinery. Cell. 1997;91(2):231–41. http://dx.doi.org/10.1016/S0092-8674(00)80405-5
46. Ozes ON, Mayo LD, Gustin JA, Pfeffer SR, Pfeffer LM, Donner DB. NF-kappaB activation by tumour necrosis factor requires the Akt serine-threonine kinase. Nature. 1999;401(6748):82–5. http://dx.doi.org/10.1038/43466
47. LoPiccolo J, Blumenthal GM, Bernstein WB, Dennis PA. Targeting the PI3K/Akt/mTOR pathway: Effective combinations and clinical considerations. Drug Resist Updat. 2008;11(1–2):32–50. http://dx.doi.org/10.1016/j.drup.2007.11.003
48. Bao ZS, Chen HM, Yang MY, Zhang CB, Yu K, Ye WL, et al. RNA-seq of 272 gliomas revealed a novel, recurrent PTPRZ1-MET fusion transcript in secondary glioblastomas. Genome Res. 2014;24(11): 1765–73. http://dx.doi.org/10.1101/gr.165126.113
49. Carpten JD, Faber AL, Horn C, Donoho GP, Briggs SL, Robbins CM, et al. A transforming mutation in the pleckstrin homology domain of AKT1 in cancer. Nature. 2007;448(7152):439–44. http://dx.doi.org/10.1038/nature05933
50. Stokoe D. Pten. Curr Biol. 2001;11(13):R502. http://dx.doi.org/10.1016/S0960-9822(01)00303-7
51. Tohma Y, Gratas C, Biernat W, Peraud A, Fukuda M, Yonekawa Y, et al. PTEN (MMAC1) mutations are frequent in primary glioblastomas (de novo) but not in secondary glioblastomas. J Neuropathol Exp Neurol. 1998;57(7):684–9. http://dx.doi.org/10.1097/00005072-199807000-00005
52. Tamura M, Gu J, Matsumoto K, Aota S, Parsons R, Yamada KM. Inhibition of cell migration, spread-ing, and focal adhesions by tumor suppressor PTEN. Science. 1998;280(5369):1614–17. http://dx.doi.org/10.1126/science.280.5369.1614
53. Reifenberger G, Liu L, Ichimura K, Schmidt EE, Collins VP. Amplification and overexpression of the MDM2 gene in a subset of human malignant gliomas without p53 mutations. Cancer Res. 1993;53(12):2736–9.
54. Biernat W, Kleihues P, Yonekawa Y, Ohgaki H. Amplification and overexpression of MDM2 in pri-mary (de novo) glioblastomas. J Neuropathol Exp Neurol. 1997;56(2):180–5. http://dx.doi.org/10.1097/00005072-199702000-00009
55. Stott FJ, Bates S, James MC, McConnell BB, Starborg M, Brookes S, et al. The alternative product from the human CDKN2A locus, p14(ARF), participates in a regulatory feedback loop with p53 and MDM2. EMBO J. 1998;17(17):5001–14. http://dx.doi.org/10.1093/emboj/17.17.5001
56. Nakamura M, Watanabe T, Klangby U, Asker C, Wiman K, Yonekawa Y, et al. p14ARF deletion and methylation in genetic pathways to glioblastomas. Brain Pathol. 2001;11(2):159–68. http://dx.doi.org/10.1111/j.1750-3639.2001.tb00388.x
57. Nakamura M, Yonekawa Y, Kleihues P, Ohgaki H. Promoter hypermethylation of the RB1 gene in glioblastomas. Lab Invest. 2001;81(1):77–82. http://dx.doi.org/10.1038/labinvest.3780213
58. Sherr CJ, Roberts JM. CDK inhibitors: Positive and negative regulators of G1-phase progression. Genes Dev. 1999;13(12):1501–12. http://dx.doi.org/10.1101/gad.13.12.1501
59. Tso CL, Freije WA, Day A, Chen Z, Merriman B, Perlina A, et al. Distinct transcription profiles of primary and secondary glioblastoma subgroups. Cancer Res. 2006;66(1):159–67. http://dx.doi.org/10.1158/0008-5472.CAN-05-0077
60. Lamoral-Theys D, Le Mercier M, Le Calve B, Rynkowski MA, Bruyere C, Decaestecker C, et al. Long-term temozolomide treatment induces marked amino metabolism modifications and an increase in TMZ sensitivity in Hs683 oligodendroglioma cells. Neoplasia. 2010;12(1):69–79. http://dx.doi.org/10.1593/neo.91360
61. Svilar D, Dyavaiah M, Brown AR, Tang JB, Li J, McDonald PR, et al. Alkylation sensitivity screens reveal a conserved cross-species functionome. Mol Cancer Res. 2012;10(12):1580–96. http://dx.doi.org/10.1158/1541-7786.MCR-12-0168
62. Ye F, Zhang Y, Liu Y, Yamada K, Tso JL, Menjivar JC, et al. Protective properties of radio-chemoresistant glioblastoma stem cell clones are associated with metabolic adaptation to reduced glucose dependence. PLoS One. 2013;8(11):e80397. http://dx.doi.org/10.1371/journal.pone.0080397
63. Auger N, Thillet J, Wanherdrick K, Idbaih A, Legrier ME, Dutrillaux B, et al. Genetic alterations asso-ciated with acquired temozolomide resistance in SNB-19, a human glioma cell line. Mol Cancer Ther. 2006;5(9):2182–92. http://dx.doi.org/10.1158/1535-7163.MCT-05-0428
64. Hiddingh L, Raktoe RS, Jeuken J, Hulleman E, Noske DP, Kaspers GJ, et al. Identification of temozolo-mide resistance factors in glioblastoma via integrative miRNA/mRNA regulatory network analysis. Sci Rep. 2014;4:5260. http://dx.doi.org/10.1038/srep05260
65. Kumar DM, Patil V, Ramachandran B, Nila MV, Dharmalingam K, Somasundaram K. Temozolomide-modulated glioma proteome: Role of interleukin-1 receptor-associated kinase-4 (IRAK4) in chemo-sensitivity. Proteomics. 2013;13(14):2113–24. http://dx.doi.org/10.1002/pmic.201200261
66. Sun S, Wong TS, Zhang XQ, Pu JK, Lee NP, Day PJ, et al. Protein alterations associated with temozolo-mide resistance in subclones of human glioblastoma cell lines. J Neuro Oncol. 2012;107(1):89–100. http://dx.doi.org/10.1007/s11060-011-0729-8
67. Anderson JC, Duarte CW, Welaya K, Rohrbach TD, Bredel M, Yang ES, et al. Kinomic exploration of temozolomide and radiation resistance in Glioblastoma multiforme xenolines. Radiother Oncol. 2014;111(3):468–74. http://dx.doi.org/10.1016/j.radonc.2014.04.010
68. Erson-Omay EZ, Henegariu O, Omay SB, Harmanci AS, Youngblood MW, Mishra-Gorur K, et al. Longitudinal analysis of treatment-induced genomic alterations in gliomas. Genome Med. 2017;9(1):12. http://dx.doi.org/10.1186/s13073-017-0401-9
69. Stepanenko AA, Andreieva SV, Korets KV, Mykytenko DO, Baklaushev VP, Huleyuk NL, et al. Temozolomide promotes genomic and phenotypic changes in glioblastoma cells. Cancer Cell Int. 2016;16:36. http://dx.doi.org/10.1186/s12935-016-0311-8
70. Auffinger B, Tobias AL, Han Y, Lee G, Guo D, Dey M, et al. Conversion of differentiated cancer cells into cancer stem-like cells in a glioblastoma model after primary chemotherapy. Cell Death Differ. 2014;21(7):1119–31. http://dx.doi.org/10.1038/cdd.2014.31
71. Fouse SD, Nakamura JL, James CD, Chang S, Costello JF. Response of primary glioblastoma cells to therapy is patient specific and independent of cancer stem cell phenotype. Neuro Oncol. 2014;16(3):361–71. http://dx.doi.org/10.1093/neuonc/not223
72. Bhat KP, Balasubramaniyan V, Vaillant B, Ezhilarasan R, Hummelink K, Hollingsworth F, et al. Mesenchymal differentiation mediated by NF-kappaB promotes radiation resistance in glioblastoma. Cancer Cell. 2013;24(3):331–46. http://dx.doi.org/10.1016/j.ccr.2013.08.001
73. Lau J, Ilkhanizadeh S, Wang S, Miroshnikova YA, Salvatierra NA, Wong RA, et al. STAT3 blockade inhibits radiation-induced malignant progression in glioma. Cancer Res. 2015;75(20):4302–11. http://dx.doi.org/10.1158/0008-5472.CAN-14-3331
74. Kim YH, Yoo KC, Cui YH, Uddin N, Lim EJ, Kim MJ, et al. Radiation promotes malignant progression of glioma cells through HIF-1alpha stabilization. Cancer Lett. 2014;354(1):132–41. http://dx.doi.org/10.1016/j.canlet.2014.07.048