The Role of Molecular Genetic Pathology in the Diagnosis And

1

Contemporary Diagnosis of Gliomas using Biomarkers

Daniel J. Brat, MD, PhD1

1Department of Pathology and Laboratory Medicine, Winship Cancer Institute,

Emory University School of Medicine, Atlanta, GA

Address for Correspondence/Reprints: Daniel J. Brat, MD, PhD Department of Pathology and Laboratory Medicine Emory University Hospital, H-195 1364 Clifton Rd. NE Atlanta, GA 30322 Phone: 404-712-1266; Fax: 404-727-3133; e-mail: [email protected] 2

Introduction:

Gliomas are the largest and most diverse group of primary brain tumors 1,2 and are subdivided into distinct classes according to clinical, neuroimaging, histopathologic and molecular genetic characteristics3,4. The diffuse gliomas are a subset defined by widely

infiltrative properties and their tendency toward biologically progression. Although they

can occur throughout the neuraxis and at any age, they most frequently arise within the

cerebral hemispheres of adults. The primary morphologic classes of diffuse gliomas are

astrocytomas, oligodendrogliomas and oligoastrocytomas, and these are graded

according to World Health Organization (WHO) criteria as grades II-IV1.

The diagnosis of glial neoplasms has been established primarily by histopathological

examination since the early 1900s, when Bailey and Cushing first classified them based

on their presumed histogenesis 5,6. In these histologic schemes, diffuse astrocytomas

are recognized by their irregular, elongated hyperchromatic nuclei and high degree of

fibrillarity 1. Mitotic activity is associated with a shorter survival and is used as a grading

criterion to distinguish infiltrating astrocytoma, WHO grade II from anaplastic

astrocytoma, WHO grade III. Likewise, necrosis and microvascular proliferation signal even more aggressive behavior and serve as criteria for GBM, WHO grade IV7. By

contrast, oligodendrogliomas have round, regular nuclei, perinuclear halos and a

delicate branching (“chicken-wire”) vasculature. Increased mitoses (≥ 6 per 10 high

power fields), necrosis and microvascular proliferation are used as criteria to distinguish

oligodendroglioma, WHO grade II from anaplastic oligodendroglioma, WHO grade III 8.

The WHO has also recognized oligoastrocytomas, which show both astrocytic and 3

oligodendroglial morphology, and grading schemes for this class have largely followed those of oligodendroglioma.

Many studies have demonstrated low reproducibility and interobserver concordance in the diagnosis of diffuse gliomas, leading to confusion in clinical management. Similarly, correlations of histologic class with molecular markers, clinical behavior and response to therapies have been highly variable 7,9. Pilocytic astrocytomas, gangliogliomas and

PXAs can also occasionally pose diagnostic challenges, yet their proper recognition is critical, since the treatment and prognosis differ from those of diffuse gliomas. Over the past 20 years, investigations of genomic alterations in glial neoplasms have greatly informed our understanding of molecular classes of disease and have led to our current use of biomarker-driven glioma classification10-13.

IDH Mutations Subdivide Infiltrating Gliomas in Adults into Distinct Subsets

The mutational status of isocitrate dehydrogenase 1 and 2 (IDH1 and IDH2) is now recognized as a major discriminator of biological classes among the diffuse gliomas14-16.

When mutations occur in IDH1 or IDH2, the mutant enzyme develops a preferential affinity for alpha-ketoglutarate instead of isocitrate, which leads to the production and accumulation of the oncometabolite 2-hydroxyglutarate17. Both IDH1 and IDH2 mutations occur in infiltrating gliomas, though IDH2 mutations are much less common than IDH1 15,18. IDH1 mutations are found in approximately 70-80% of histologic grade II and III infiltrating gliomas and secondary GBMs, yet are much less frequent in primary

GBMs (~ 5%) 14,15,19-21. The most frequent IDH1 mutation is R132H, which occurs at codon 132 and leads to exchange of the amino acid arginine for histidine 15,21-24. 4

Infiltrating gliomas occurring in young adults are more likely to harbor an IDH1 mutation

than those in the elderly 15,21,23,24, but this mutation is rarely found in gliomas arising in

young children 18,25. IDH mutant gliomas progress more slowly over time than those

lacking IDH mutations, suggesting that these are biologically distinct forms of disease.

Therefore, diffuse gliomas with IDH mutations are associated with a better prognosis,

grade for grade, than those without 15,24. While IDH1 mutations are frequent in diffuse

gliomas, they are rarely found (or absent) in other CNS neoplasms, including

ependymomas, pilocytic astrocytomas, gangliogliomas and PXAs 15,21-23. Thus, the

finding of an IDH mutation in a primary glial neoplasm strongly suggests that it is a

diffusely infiltrative glioma. Since IDH1 mutations are present in both diffuse gliomas

that show astrocytic differentiation and in those that show oligodendroglial

differentiation, they are thought to occur early in gliomagenesis, before those molecular

alterations that promote specific differentiation 15,21,22, such as TP53 and ATRX

mutations in astrocytomas, or 1p/19q co-deletion, CIC, FUBP1 and TERT promoter

mutations in oligodendrogliomas 21,24.

IDH Mutant Astrocytomas have TP53 mutations and ATRX alterations

Among diffuse gliomas that are IDH mutant, there is a relatively strict molecular dichotomy: one subset, accounting for about 30-40% of IDH mutant gliomas, will demonstrate 1p/19q co-deletion; the other larger subset will be defined largely by TP53 mutations and loss of ATRX (Figure 1)16. Studies based on morphologic class have

consistently shown that the large majority of grade II and III diffuse astrocytomas and

secondary glioblastomas have TP53 mutations 15,21,26. When restricted to IDH-mutated 5

infiltrating astrocytomas, an even higher percentage have TP53 mutations, since IDH

wild-type diffuse gliomas have a lower frequencies of TP53 mutation. The Cancer

Genome Atlas project demonstrated that 94% of IDH mutant grade II and III diffuse

gliomas that lacked 1p/19q co-deletion had TP53 mutations16.

The combination of IDH and TP53 mutations is strongly coupled to inactivating

alterations in Alpha Thalassemia/Mental Retardation Syndrome X-linked (ATRX), a

gene that encodes a chromatin remodeling regulator. Nearly all diffuse gliomas with an

ATRX mutation had an IDH1 mutation as well. In tumors with IDH and ATRX mutations

present, 94% had a TP53 mutation. ATRX mutations are associated with the Alternative

Lengthening of Telomeres (ALT) phenotype27, an alternative mechanism for maintaining

telomere length in tumors that do not have constitutive telomerase activity 28. In establishing the diagnosis of an IDH mutant diffuse glioma, loss of immunohistochemical staining for ATRX in neoplastic cells is an excellent surrogate for

ATRX gene inactivation and supports the diagnosis of an astrocytoma29. The evidence

to date indicates that the molecular signature of IDH mutant astrocytoma includes TP53

mutation and ATRX alteration and is associated with ALT 16,19,29.

The grading of astrocytomas has traditionally been performed based on morphologic

features, and the presence of mitotic activity has been use to stratify tumors as grade II

and III. These histopathologic grading schemes have included both IDH mutant and IDH

wild type astrocytomas that were classified based on morphology. Grading schemes

may need to be reconsidered for IDH mutant astrocytomas that are classified based on

TP53 mutations or ATRX loss. It is not entirely clear whether mitotic activity or other 6

molecular, morphologic or clinical feature will optimally stratifies IDH mutant

astrocytomas 30.

Oligodendrogliomas are Diffuse Gliomas with IDH Mutations and 1p/19q Co- deletion

A second form of diffuse glioma is characterized by IDH mutations and 1p/19q co-

deletion, and shows a strong association with oligodendroglioma histology (Figure

1)31,32. While past studies based on morphologic classification of disease have emphasized the tight correlation of 1p/19q co-deletion and the oligodendroglioma

morphology, more recent interpretations have emphasized that the combination of IDH

mutation and 1p/19q co-deletion is the molecular signature of this disease rather than

an association 16,29,31. IDH mutant diffuse gliomas with 1p/19q co-deletion show a

slightly higher percentage of IDH2 mutations than those without co-deletion, yet the

percentage is still small compared to IDH1 mutations 16. Deletion of the 1p and 19q

chromosomal arms occurs through an unbalanced translocation that results in the

formation of der(1;19) (q10;p10) 33. Inactivating mutations of the tumor suppressor

genes FUBP1 and CIC, on chromosomes 1p and 19q, respectively, occur secondary to

this unbalanced translocation 34. Among diffuse gliomas, these mutations are specific

for oligodendrogliomas with 1p/19q co-deletion, with a frequency ranging from 46 to

83% for CIC and 20-30% for FUBP1 16,31. They are rare or absent in infiltrating astrocytomas of all grades 27,31,34,35 and are mutually exclusive with mutations in TP53

and ATRX 27. Other genes that are mutated in IDH mutant, 1p/19q co-deleted gliomas

include Notch1, PIK3CA and PIK3R1 16,34,36. 7

Oligodendrogliomas also have a mechanism for maintaining telomere length.

While the ALT serves this purpose in astrocytomas, it is rarely seen in

oligodendrogliomas. Instead, activating mutations in the telomerase reverse

transcriptase (TERT) promoter are present in nearly all oligodendrogliomas 28,37. In the

TCGA study of diffuse grade II and III diffuse glioma, 96% of cases with IDH mutations

and 1p/19q co-deletion showed TERT mutations, while only 4% of IDH mutant gliomas

without 1p/19q co-deletion showed this mutation 16. Thus, the molecular signature of oligodendroglioma includes mutations of IDH and the TERT promoter, together with co- deletion of 1p/19q16,37. With this emerging definition of disease, the optimal morphologic or molecular prognostic and predictive markers will need to be defined 30.

Oligoastrocytoma

Previous WHO classifications have recognized grade II, III and IV diffuse gliomas with mixed histology (oligoastrocytoma, grades II and III, and GBM with oligodendroglioma component, grade IV)1. However, numerous investigations have indicated that IDH-

mutant gliomas are either 1p/19q co-deleted or TP53 mutant, with few gaps or overlaps,

and reflect two distinct molecular lineages that are best represented by

oligodendroglioma (TERT mutation and 1p/19q co-deletion) or astrocytoma (TP53

mutation and ATRX loss)29,36,38. In the TCGA analysis of grades II and III diffuse

gliomas, the majority of tumors characterized morphologically as oligoastrocytomas

were IDH mutant and had TP53 mutations, while smaller numbers were 1p/19q co-

deleted or were IDH wild-type16. Similarly, studies of GBM-O have concluded that these tumors are either IDH mutant, combined with either 1p/19q co-deletion or TP53 8

mutation, or alternatively, are IDH wild-type and contain genetic alterations typical of classic GBM, such as EGFR amplification and PTEN loss.39-41 Combined, these studies

have not provided evidence for a genetic signature specific for gliomas of mixed

histology (oligoastroctyomas), but rather indicate that these mixed histology gliomas

consist of discrete molecular entities. 36,38,42 Since the diagnoses of oligoastrocytoma

and GBM-O have been associated with a high degree of interobserver variation and

molecular alterations and clinical outcomes have not been consistent, the use molecular

signatures to define astrocytoma and oligodendroglioma lineage should improve

interobserver concordance. Biomarker-driven diagnosis also seems likely to reduce the

diagnosis of “oligoastrocytoma” and the confusion related to its clinical management.

IDH Wild-type Diffuse Gliomas in Adults are Biologically Aggressive

The majority of infiltrating gliomas that are IDH wild-type are primary GBMs12,15.

Approximately 95% of primary GBMs lack an IDH mutation, compared with only 20-30%

of grade II and III infiltrating gliomas15. However, the subset of diffuse gliomas that are

IDH wild-type and grade II-III by histological criteria (lacking necrosis and microvascular hyperplasia) represent a clinically important subset, because their genetic alterations and clinical behavior are similar to those of IDH wild type (primary) GBM 27. For example, IDH wild-type grade II and III diffuse gliomas have a much lower frequency of

TP53 mutations and do not demonstrate 1p/19q co-deletion 24. TP53 mutations are

present in only 15-25% of IDH wild type grade II-III astrocytomas as compared to 94%

of IDH mutant astrocytomas 15,21. Furthermore, diffuse grade II-III gliomas that are IDH

wild-type more frequently show molecular genetic alterations similar to GBM, such as 9

EGFR, PTEN, NF1, RB1, CDKN2A alterations, than those tumors that have an IDH

mutation 15,16. TERT promoter mutations are frequent in IDH wild type grade II-III gliomas and GBMs (83%), but are rare in IDH1-mutant infiltrating astrocytomas

28,37.Since IDH-mutated and IDH wild-type astrocytomas have disimilar sets of mutations and clinical behaviors, they appear to arise by distinct oncogenic pathways and represent separate diseases, despite their histopathologic similarity.

Primary vs Secondary Glioblastoma

The clinical presentation of GBM may be primary, with a de novo grade IV neoplasm, or

secondary, in which the GBM arises over time from a lower grade infiltrating glioma 43.

The large majority of GBMs are primary (90-95%) while secondary GBMs are less common (5-10%) 6. These two clinical presentations of GBM have molecular correlates

as well, with most primary GBMs being IDH wild-type and most secondary GBMs being

IDH-mutant. Primary, IDH wild-type GBMs typically demonstrate an accumulation of molecular alterations within three dominant genetic families: 1) receptor tyrosine kinase

(RTK); 2) Retinoblastoma (RB1); and 3) p53. EGFR amplification, PTEN deletion and mutations of the PI3K subunits are common alterations in the RTK family that are commonly associated with primary GBMs 12,44. Approximately 40-50% of GBMs show

amplification of EGFR and about half of these have the variant III (vIII) mutation. The vIII

mutation causes truncation of the receptor, which leads to constitutive tyrosine kinase activity 6. PTEN deletion or mutation occurs in about 80% of GBMs 6. Other members of

this pathway that are commonly altered in primary GBMs include: PDGFRA, which is

amplified in 18%; phosphatidylinositol-3-OH kinase (PI(3)K), which is mutated in 15%; 10

and NF1, with 18% showing mutations and/or deletions. Among the p53 family of

genetic alterations, the most frequent are mutations and deletions of CDKN2A/ARF

(49%), amplification of MDM2 (14%) and mutations and deletions of TP53 (35%).

Alterations in the RB1 pathway include mutations and deletions of CDKN2A/p16 (52%),

amplification of CDK4 (18%) and mutations and deletions of RB1 (11%)12. In the evolution of IDH wild-type, primary GBM, the precise sequence of oncogenic events involving RTK, RB1 and p53 has not been determined.

In contrast, secondary tumors show a high frequency of IDH1 and TP53

mutations, together with ATRX inactivation43. Approximately 85% of secondary GBMs

harbor an IDH1 mutation while 62% have a TP53 mutation 15. Since TP53 mutation is

thought to occur early in IDH mutant astrocytoma development, additional alterations

are likely associated with biological progression45.

MGMT Promoter Methylation

Standard therapy for GBM includes radiation and chemotherapy with

temozolomide, which acts by crosslinking DNA by alkylating multiple sites including the

O6 position of guanine 46. DNA crosslinking is reversed by the DNA repair enzyme

MGMT (O6-methylguanine-DNA methyltransferase). Therefore, low levels of MGMT

would be expected to be enhanced response to alkylating agents. The expression level

of MGMT is determined in large part by the methylation status of the gene’s promoter.

Epigenetic silencing of MGMT occurs in 40% to 50% of GBMs and can be assessed by

its promoter methylation status on PCR-based tests of genomic DNA. MGMT promoter methylation is a strong predictor of prolonged survival, independent of other clinical 11

factors or treatment 47 and MGMT promoter methylation is associated with prolonged

progression-free and overall survival in patients with GBM treated with chemotherapy

and radiation therapy 47.

Like IDH mutations, MGMT promoter methylation occurs with similar frequencies

in grades II, III and IV diffuse astrocytomas and is thought to be an early event in

gliomagenesis 48-50 49,51 47,49-51. Secondary GBMs show a particularly high frequency of

MGMT promoter methylation 49,50. In low grade diffuse astrocytomas and secondary

GBMs, MGMT promoter methylation is tightly correlated with TP53 mutation and overexpression of the p53 protein 48,50. In one study, 92% of grade II diffuse

astrocytomas with MGMT promoter methylation harbored a TP53 mutation, compared

to only 39% of those that were unmethylated. Similarly, secondary GBMs with TP53

mutations had a greater percentage with MGMT promoter methylation that those

without (92% vs 25%, respectively) 50.

Pediatric Gliomas

High grade infiltrating astrocytomas in pediatric patients are vastly different than those

found in adult patients.52,53 Though histologically similar, these two groups can be

distinguished by their locations, clinical behavior, and mutation and gene expression

profiles. Pediatric GBMs, for example, nearly always form de novo and rarely progress

from a lower grade glioma 54. IDH1 mutations are uncommon in pediatric GBMs, occurring in less than 10% 54, and are typically absent in high grade gliomas of young

children 55. Conversely, mutations in H3F3A, ATRX and DAXX (death-domain

associated protein) are more typical of pediatric GBMs 52,54,56. These genes encode 12

proteins with roles in chromatin remodeling. H3F3A encodes for H3.3, a replication-

independent histone variant that is normally recruited to DNA via the ATRX-DAXX heterodimer 56. Mutations in H3.3 lead to decreased methylation of H3 histones 57,58 and

usually involve amino acid substitutions at K27 and G34. Approximately 40% of

pediatric GBMs harbor H3F3A mutations and nearly all (95%) also carry an ATRX

mutation. A smaller percent of these (4.2%) had a DAXX mutation. In this same study, approximately half (54%) of the GBMs had a TP53 mutation and there was significant overlap between TP53 mutations and ATRX, H3F3A or DAXX mutations. Alternative lengthening of telomeres (ALT) was associated with loss of ATRX expression and was most frequent in tumors that had combined mutations in ATRX, H3F3A and TP53 56.

The site of H3F3A mutations in pediatric high grade glioma is associated with patient

age and tumor location: those with mutations at K27 tend to occur in young children

(median age 11 yrs) and in midline locations (pons and thalamus); those with mutations

at G34 occur in teenagers and young adults and arose from the cerebral hemispheres

56. A mutual exclusivity of H3F3A and IDH1 mutations has been described in pediatric

high grade gliomas 54,56.

Low grade gliomas of childhood are also distinct from those of adults. Among

those pediatric gliomas that appear diffusely infiltrative, a lower percentage contain IDH

mutations11. However, the IDH wild-type status does not necessarily have the same implications for biologically aggressive behavior in low grade diffuse gliomas of childhood. A recent study of oligodendrogliomas in children found that only 18% of those cases tested had IDH R132H mutations and only 25% of cases had 1p/19q co- deletion 59. With the definition of oligodendrogliomas of adults emphasizing IDH 13

mutation and 1p/19q co-deletion as a definitive signature, the neuropathologic diagnosis

of pediatric oligodendroglioma will not likely follow the same paradigm.

A large and important subset of pediatric low grade gliomas are better- circumscribed and more indolent in clinical behavior, including pilocytic astrocytomas,

WHO grade I, gangliogliomas, WHO grade I, and pleomorphic xanthoastrocytomas

(PXAs), WHO grade II. In addition to their general lack of infiltration, pilocytic astrocytomas, gangliogliomas and PXAs also differ from diffuse gliomas in their mutational profiles. They typically lack IDH mutations, TP53 mutations and 1p/19q co- deletions 21,23,60. In contrast, these tumors are most often defined genetically by the

presence of activating alterations of BRAF, an oncogene that regulates cell growth,

proliferation and survival and is a member of the ERK/MAPK pathway 61. A subset of

pilocytic astrocytomas (60-70%) demonstrate a specific fusion event between BRAF

and KIAA1549 (BRAF:KIAA1549) that results from a tandem duplication at 7q34 and

leads to BRAF activation 60,62. This fusion event occurs more commonly in tumors that

are midline and infratentorial in location. BRAF V600E missense mutations also occur in

pilocytic astrocytomas, but are not as common (9%) 61,63. While BRAF alterations are

most common in sporadic pilocytic astrocytomas, they are not characteristic of pilocytic

astrocytomas that arise in the setting of neurofibromatosis 1, since ERK/MAPK pathway

activation in these tumors is directly due to the loss of proper NF1 function11.

BRAF (V600E) mutations occur frequently in PXAs (60%) and gangliogliomas

(50%), and like BRAF fusions, result in activation of the ERK/MAPK pathway 61,63,64.

Although the histologic diagnosis of these tumors is straightforward in classic cases, the

finding of BRAF mutations may be helpful in distinguishing these low grade tumors from 14

diffusely infiltrative gliomas in diagnostically challenging cases, and could be clinically

important given the contrast in prognosis and treatment between these groups 60,61.

Ancillary Studies Used in the Diagnosis of Gliomas

Ancillary studies such as immunohistochemistry, and molecular and cytogenetic testing

are now used routinely in the diagnosis of gliomas10. Immunostains commonly used to

determine lineage and prognostic subsets of the diffuse gliomas include those that

recognize the IDH1 mutant protein, ATRX and the p53 protein 29. Immunohistochemistry for IDH1 mutant protein is positive in those gliomas that have the R132H mutation, which accounts for over 90% of all IDH mutations in diffuse gliomas 65. Because

immuohistochemisty can detect rare positive cells among abundant non-neoplastic

ones, this method has comparable sensitivity to PCR or sequencing 23,66-68. Since

immunohistochemistry is less expensive and widely available, it is often used in favor of

other methods. However, this antibody only recognizes the R132H IDH1 mutation, so

that other less common IDH1 mutations and all IDH2 mutations, albeit rare, will be

missed with this test. Since the designation of a diffuse glioma in adults as IDH wild type

has taken on heightened clinical and therapeutic significance, a negative

immunohistochemical test for the IDH mutant protein should be followed by mutational

analysis of IDH1 and IDH2.

The p53 immunostain is targeted against the normal p53 protein. TP53 mutation leads

to decreased degradation of p53 protein oligomers, which includes mutant and wild-type

gene products, allowing their immunohistochemical detection within the nucleus. TP53

mutations are frequent in grade II and grade III astrocytomas, but not in 15

oligodendrogliomas and are nearly mutually exclusive with 1p/19 co-deletion. The p53

immunostain is therefore useful as a marker for astrocytic differentiation, with strong,

diffuse staining favoring the diagnosis of astrocytoma over oligodendroglioma. Although

this immunostain can be a useful aid in determining lineage, it is not entirely specific for

TP53 mutations or for neoplastic disease, since the p53 protein can be upregulated by

other mechanisms 69,70.

Mutations and inactivating deletions of ATRX are a marker of astrocytic lineage among the IDH mutant gliomas and are mutually exclusive with 1p/19q co-deletion.

Immunohistochemistry for ATRX demonstrates a loss of protein expression in neoplastic nuclei that harbor mutations, while expression is retained in non-neoplastic cells within the sample (e.g., endothelial cells).29 Among the pediatric gliomas,

antibodies to specific histone marks (H3K27me3) for high grade gliomas and to mutant

BRAF (V600E) are useful for accurately subtyping disease.

Promoter methylation analysis of MGMT is performed routinely for high grade

astrocytomas, typically by methylation-specific PCR 47. While immunohistochemistry for

MGMT protein expression is also available, correlations with PCR results, response to

therapy and patient survival have not been strong 71-75.

Fluorescence in situ hybridization (FISH) is commonly used to assess copy number

alterations of single loci, including EGFR and PDGFRA amplification, PTEN and

CDKN2A deletions in high grade astrocytomas and 1p/19q co-deletion in

oligodendrogliomas. As molecular techniques become more advanced and the number

of clinically relevant molecular alterations increases, molecular diagnostic labs will

continue to transition towards a panel-based assessment of mutations and high 16

resolution, whole genome analysis of copy number alterations. Genes that will need to

be considered for inclusion in a diagnostic panel approach for gliomas include IDH1,

IDH2, TP53, ATRX, CIC, FUBP1, TERT promoter, NOTCH1, PIK3CA, PIK3R1, EGFR,

PDGFRA, PTEN, NF1, RB1, H3F3A, DAXX and BRAF. Whole genome assessment of

copy number alterations is capable of detecting 1p/19q status; chromosome 7 gains and

chromosome 10 losses; amplifications of EGFR, PDGFRA and c-MET; deletions of

PTEN and CDKN2A/B; alterations of BRAF and other less frequent gains, losses and

unbalanced translocations.

Future Direction of Diagnosis and Management of Gliomas

Because gliomas are increasingly being defined by their molecular profiles, biomarkers

will be relied on to a greater extent to establish neuropathologic diagnoses. For example

in the guidelines established by an international group of neuropathologists for the 4th

Edition of the WHO Classification of Central Nervous System Tumors, it was concluded

that histology alone may be sufficient for the diagnosis of some tumors, while others will

require a more “integrated” diagnosis, which would include the histological classification,

the WHO grade and the results of molecular studies 76. In this layered format, the

“Integrated Diagnosis” would form the top line, and while underneath would be

categories for histological classification, WHO grade and molecular data. For each tumor entity, it will be determined whether molecular information would be necessary, suggested or not needed for the diagnosis. Furthermore, if molecular testing is needed or suggested, entity-specific guidelines will provide a list of appropriate molecular tests and a format for reporting results. Another important recommendation was that 17

selected pediatric tumors be separated from their histologically identical adult counterparts, on the basis of their distinct molecular profiles. It is hoped that these recommendations, once in practice, will lead to more objective, reproducible diagnoses that reflect the biological nature of each tumor.

18

References:

1. Louis DN, Ohgaki, H., Wiestler, O.D., Cavenee, W.K. WHO Classification of Tumours of the Central Nervous System. 4th ed. Lyon: Intl. Agency for Research; 2007. 2. Ostrom QT, Gittleman H, Farah P, et al. CBTRUS statistical report: Primary brain and central nervous system tumors diagnosed in the United States in 2006-2010. Neuro Oncol. Nov 2013;15 Suppl 2:ii1-56. 3. Bourne TD, Schiff D. Update on molecular findings, management and outcome in low-grade gliomas. Nat Rev Neurol. Dec 2010;6(12):695-701. 4. Appin CL, Brat DJ. Molecular genetics of gliomas. Cancer J. Jan-Feb 2014;20(1):66-72. 5. Bailey P, Cushing H. A classification of the tumors of the glioma group on a histogenic basis with a correlated study of prognosis. Philadelphia, Pennsylvania: JB Lippincott Co; 1926. 6. Perry A, Brat DJ. Practical Surgical Neuropathology: A Diagnostic Approach. Philadelphia, Pennsylvania: Churchill Livingstone; 2010. 7. Brat DJ, Prayson RA, Ryken TC, Olson JJ. Diagnosis of malignant glioma: role of neuropathology. J Neurooncol. Sep 2008;89(3):287-311. 8. Giannini C, Scheithauer BW, Weaver AL, et al. Oligodendrogliomas: reproducibility and prognostic value of histologic diagnosis and grading. J Neuropathol Exp Neurol. Mar 2001;60(3):248-262. 9. Coons SW, Johnson PC, Scheithauer BW, Yates AJ, Pearl DK. Improving diagnostic accuracy and interobserver concordance in the classification and grading of primary gliomas. Cancer. Apr 1 1997;79(7):1381-1393. 10. Brat DJ, Cagle PT, Dillon DA, et al. Template for Reporting Results of Biomarker Testing of Specimens From Patients With Tumors of the Central Nervous System. Arch Pathol Lab Med. Feb 2 2015. 11. Rodriguez FJ, Lim KS, Bowers D, Eberhart CG. Pathological and molecular advances in pediatric low-grade astrocytoma. Annu Rev Pathol. Jan 24 2013;8:361-379. 12. Brennan CW, Verhaak RG, McKenna A, et al. The somatic genomic landscape of glioblastoma. Cell. Oct 10 2013;155(2):462-477. 13. Theeler BJ, Yung WK, Fuller GN, De Groot JF. Moving toward molecular classification of diffuse gliomas in adults. Neurology. Oct 30 2012;79(18):1917- 1926. 14. Parsons DW, Jones S, Zhang X, et al. An integrated genomic analysis of human glioblastoma multiforme. Science. Sep 26 2008;321(5897):1807-1812. 15. Yan H, Parsons DW, Jin G, et al. IDH1 and IDH2 mutations in gliomas. N Engl J Med. Feb 19 2009;360(8):765-773. 16. The_Cancer_Genome_Atlas_Research_Network. Comprehensive, Integrative Genomic Analysis of Diffuse Lower Grade Gliomas. N Engl J Med. 2015;(In Press). 17. Dang L, White DW, Gross S, et al. Cancer-associated IDH1 mutations produce 2-hydroxyglutarate. Nature. Dec 10 2009;462(7274):739-744. 19

18. Hartmann C, Meyer J, Balss J, et al. Type and frequency of IDH1 and IDH2 mutations are related to astrocytic and oligodendroglial differentiation and age: a study of 1,010 diffuse gliomas. Acta Neuropathol. Oct 2009;118(4):469-474. 19. Liu XY, Gerges N, Korshunov A, et al. Frequent ATRX mutations and loss of expression in adult diffuse astrocytic tumors carrying IDH1/IDH2 and TP53 mutations. Acta Neuropathol. Nov 2012;124(5):615-625. 20. Ohgaki H, Kleihues P. Genetic profile of astrocytic and oligodendroglial gliomas. Brain tumor pathology. Jul 2011;28(3):177-183. 21. Watanabe T, Nobusawa S, Kleihues P, Ohgaki H. IDH1 mutations are early events in the development of astrocytomas and oligodendrogliomas. Am J Pathol. Apr 2009;174(4):1149-1153. 22. Balss J, Meyer J, Mueller W, Korshunov A, Hartmann C, von Deimling A. Analysis of the IDH1 codon 132 mutation in brain tumors. Acta Neuropathol. Dec 2008;116(6):597-602. 23. Mellai M, Piazzi A, Caldera V, et al. IDH1 and IDH2 mutations, immunohistochemistry and associations in a series of brain tumors. J Neurooncol. Nov 2011;105(2):345-357. 24. Metellus P, Coulibaly B, Colin C, et al. Absence of IDH mutation identifies a novel radiologic and molecular subtype of WHO grade II gliomas with dismal prognosis. Acta Neuropathol. Dec 2010;120(6):719-729. 25. Zhang J, Wu G, Miller CP, et al. Whole-genome sequencing identifies genetic alterations in pediatric low-grade gliomas. Nat Genet. Jun 2013;45(6):602-612. 26. Okamoto Y, Di Patre PL, Burkhard C, et al. Population-based study on incidence, survival rates, and genetic alterations of low-grade diffuse astrocytomas and oligodendrogliomas. Acta Neuropathol. Jul 2004;108(1):49-56. 27. Jiao Y, Killela PJ, Reitman ZJ, et al. Frequent ATRX, CIC, FUBP1 and IDH1 mutations refine the classification of malignant gliomas. Oncotarget. Jul 2012;3(7):709-722. 28. Killela PJ, Reitman ZJ, Jiao Y, et al. TERT promoter mutations occur frequently in gliomas and a subset of tumors derived from cells with low rates of self- renewal. Proceedings of the National Academy of Sciences. 2013;110(15):6021- 6026. 29. Reuss DE, Sahm F, Schrimpf D, et al. ATRX and IDH1-R132H immunohistochemistry with subsequent copy number analysis and IDH sequencing as a basis for an "integrated" diagnostic approach for adult astrocytoma, oligodendroglioma and glioblastoma. Acta Neuropathol. Jan 2015;129(1):133-146. 30. Olar A, Wani KM, Alfaro-Munoz KD, et al. IDH mutation status and role of WHO grade and mitotic index in overall survival in grade II-III diffuse gliomas. Acta Neuropathol. Feb 21 2015. 31. Yip S, Butterfield YS, Morozova O, et al. Concurrent CIC mutations, IDH mutations, and 1p/19q loss distinguish oligodendrogliomas from other cancers. J Pathol. Jan 2012;226(1):7-16. 32. Aldape K, Burger PC, Perry A. Clinicopathologic aspects of 1p/19q loss and the diagnosis of oligodendroglioma. Arch Pathol Lab Med. Feb 2007;131(2):242-251. 20

33. Jenkins RB, Blair H, Ballman KV, et al. A t(1;19)(q10;p10) mediates the combined deletions of 1p and 19q and predicts a better prognosis of patients with oligodendroglioma. Cancer research. Oct 15 2006;66(20):9852-9861. 34. Bettegowda C, Agrawal N, Jiao Y, et al. Mutations in CIC and FUBP1 contribute to human oligodendroglioma. Science. Sep 9 2011;333(6048):1453-1455. 35. Sahm F, Koelsche C, Meyer J, et al. CIC and FUBP1 mutations in oligodendrogliomas, oligoastrocytomas and astrocytomas. Acta Neuropathol. Jun 2012;123(6):853-860. 36. Cryan JB, Haidar S, Ramkissoon LA, et al. Clinical multiplexed exome sequencing distinguishes adult oligodendroglial neoplasms from astrocytic and mixed lineage gliomas. Oncotarget. Sep 30 2014;5(18):8083-8092. 37. Eckel-Passow JE, Lachance DH, Molinaro AM, et al. Glioma groups based on TERT promoter mutation, IDH mutation, and 1p/19q codeletion. N Engl J Med. 2015;(in press). 38. Sahm F, Reuss D, Koelsche C, et al. Farewell to oligoastrocytoma: in situ molecular genetics favor classification as either oligodendroglioma or astrocytoma. Acta Neuropathol. Oct 2014;128(4):551-559. 39. Appin CL, Gao J, Chisolm C, et al. Glioblastoma with oligodendroglioma component (GBM-O): molecular genetic and clinical characteristics. Brain Pathol. Jul 2013;23(4):454-461. 40. Hegi ME, Janzer RC, Lambiv WL, et al. Presence of an oligodendroglioma-like component in newly diagnosed glioblastoma identifies a pathogenetically heterogeneous subgroup and lacks prognostic value: central pathology review of the EORTC_26981/NCIC_CE.3 trial. Acta Neuropathol. Jun 2012;123(6):841- 852. 41. Laxton RC, Popov S, Doey L, et al. Primary glioblastoma with oligodendroglial differentiation has better clinical outcome but no difference in common biological markers compared with other types of glioblastoma. Neuro Oncol. Dec 2013;15(12):1635-1643. 42. Wiestler B, Capper D, Sill M, et al. Integrated DNA methylation and copy-number profiling identify three clinically and biologically relevant groups of anaplastic glioma. Acta Neuropathol. Jul 10 2014. 43. Ohgaki H, Kleihues P. The definition of primary and secondary glioblastoma. Clinical cancer research : an official journal of the American Association for Cancer Research. Feb 15 2013;19(4):764-772. 44. The_Cancer_Genome_Atlas_Network. Comprehensive genomic characterization defines human glioblastoma genes and core pathways. Nature. Oct 23 2008;455(7216):1061-1068. 45. Wakimoto H, Tanaka S, Curry WT, et al. Targetable signaling pathway mutations are associated with malignant phenotype in IDH-mutant gliomas. Clinical cancer research : an official journal of the American Association for Cancer Research. Jun 1 2014;20(11):2898-2909. 46. Stupp R, Mason WP, van den Bent MJ, et al. Radiotherapy plus concomitant and adjuvant temozolomide for glioblastoma. N Engl J Med. Mar 10 2005;352(10):987-996. 21

47. Hegi ME, Diserens AC, Gorlia T, et al. MGMT gene silencing and benefit from temozolomide in glioblastoma. N Engl J Med. Mar 10 2005;352(10):997-1003. 48. Komine C, Watanabe T, Katayama Y, Yoshino A, Yokoyama T, Fukushima T. Promoter hypermethylation of the DNA repair gene O6-methylguanine-DNA methyltransferase is an independent predictor of shortened progression free survival in patients with low-grade diffuse astrocytomas. Brain Pathol. Apr 2003;13(2):176-184. 49. Mellai M, Monzeglio O, Piazzi A, et al. MGMT promoter hypermethylation and its associations with genetic alterations in a series of 350 brain tumors. J Neurooncol. May 2012;107(3):617-631. 50. 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. Oct 2001;22(10):1715-1719. 51. Esteller M, Garcia-Foncillas J, Andion E, et al. Inactivation of the DNA-repair gene MGMT and the clinical response of gliomas to alkylating agents. N Engl J Med. Nov 9 2000;343(19):1350-1354. 52. Sturm D, Bender S, Jones DT, et al. Paediatric and adult glioblastoma: multiform (epi)genomic culprits emerge. Nat Rev Cancer. Feb 2014;14(2):92-107. 53. Gajjar A, Pfister SM, Taylor MD, Gilbertson RJ. Molecular insights into pediatric brain tumors have the potential to transform therapy. Clinical cancer research : an official journal of the American Association for Cancer Research. Nov 15 2014;20(22):5630-5640. 54. Sturm D, Witt H, Hovestadt V, et al. Hotspot mutations in H3F3A and IDH1 define distinct epigenetic and biological subgroups of glioblastoma. Cancer cell. Oct 16 2012;22(4):425-437. 55. Pollack IF, Hamilton RL, Sobol RW, et al. IDH1 mutations are common in malignant gliomas arising in adolescents: a report from the Children's Oncology Group. Childs Nerv Syst. Jan 2011;27(1):87-94. 56. Schwartzentruber J, Korshunov A, Liu XY, et al. Driver mutations in histone H3.3 and chromatin remodelling genes in paediatric glioblastoma. Nature. Feb 9 2012;482(7384):226-231. 57. Chan KM, Fang D, Gan H, et al. The histone H3.3K27M mutation in pediatric glioma reprograms H3K27 methylation and gene expression. Genes & development. May 1 2013;27(9):985-990. 58. Lewis PW, Muller MM, Koletsky MS, et al. Inhibition of PRC2 activity by a gain- of-function H3 mutation found in pediatric glioblastoma. Science. May 17 2013;340(6134):857-861. 59. Rodriguez FJ, Tihan T, Lin D, et al. Clinicopathologic features of pediatric oligodendrogliomas: a series of 50 patients. Am J Surg Pathol. Aug 2014;38(8):1058-1070. 60. Korshunov A, Meyer J, Capper D, et al. Combined molecular analysis of BRAF and IDH1 distinguishes pilocytic astrocytoma from diffuse astrocytoma. Acta Neuropathol. Sep 2009;118(3):401-405. 22

61. Schindler G, Capper D, Meyer J, et al. Analysis of BRAF V600E mutation in 1,320 nervous system tumors reveals high mutation frequencies in pleomorphic xanthoastrocytoma, ganglioglioma and extra-cerebellar pilocytic astrocytoma. Acta Neuropathol. Mar 2011;121(3):397-405. 62. Jones DT, Kocialkowski S, Liu L, et al. Tandem duplication producing a novel oncogenic BRAF fusion gene defines the majority of pilocytic astrocytomas. Cancer research. Nov 1 2008;68(21):8673-8677. 63. Dias-Santagata D, Lam Q, Vernovsky K, et al. BRAF V600E mutations are common in pleomorphic xanthoastrocytoma: diagnostic and therapeutic implications. PloS one. 2011;6(3):e17948. 64. Dougherty MJ, Santi M, Brose MS, et al. Activating mutations in BRAF characterize a spectrum of pediatric low-grade gliomas. Neuro Oncol. Jul 2010;12(7):621-630. 65. Capper D, Weissert S, Balss J, et al. Characterization of R132H mutation- specific IDH1 antibody binding in brain tumors. Brain Pathol. Jan 2010;20(1):245- 254. 66. Lee D, Suh YL, Kang SY, Park TI, Jeong JY, Kim SH. IDH1 mutations in oligodendroglial tumors: comparative analysis of direct sequencing, pyrosequencing, immunohistochemistry, nested PCR and PNA-mediated clamping PCR. Brain Pathol. May 2013;23(3):285-293. 67. Loussouarn D, Le Loupp AG, Frenel JS, et al. Comparison of immunohistochemistry, DNA sequencing and allele-specific PCR for the detection of IDH1 mutations in gliomas. International journal of oncology. Jun 2012;40(6):2058-2062. 68. Takano S, Tian W, Matsuda M, et al. Detection of IDH1 mutation in human gliomas: comparison of immunohistochemistry and sequencing. Brain tumor pathology. Apr 2011;28(2):115-123. 69. van Meyel DJ, Ramsay DA, Casson AG, Keeney M, Chambers AF, Cairncross JG. p53 mutation, expression, and DNA ploidy in evolving gliomas: evidence for two pathways of progression. J Natl Cancer Inst. Jul 6 1994;86(13):1011-1017. 70. Watanabe K, Sato K, Biernat W, et al. Incidence and timing of p53 mutations during astrocytoma progression in patients with multiple biopsies. Clinical cancer research : an official journal of the American Association for Cancer Research. Apr 1997;3(4):523-530. 71. Brell M, Tortosa A, Verger E, et al. Prognostic significance of O6-methylguanine- DNA methyltransferase determined by promoter hypermethylation and immunohistochemical expression in anaplastic gliomas. Clinical cancer research : an official journal of the American Association for Cancer Research. Jul 15 2005;11(14):5167-5174. 72. Esteller M, Hamilton SR, Burger PC, Baylin SB, Herman JG. Inactivation of the DNA repair gene O6-methylguanine-DNA methyltransferase by promoter hypermethylation is a common event in primary human neoplasia. Cancer research. Feb 15 1999;59(4):793-797. 23

73. Maxwell JA, Johnson SP, Quinn JA, et al. Quantitative analysis of O6- alkylguanine-DNA alkyltransferase in malignant glioma. Molecular cancer therapeutics. Oct 2006;5(10):2531-2539. 74. Preusser M, Charles Janzer R, Felsberg J, et al. Anti-O6-methylguanine- methyltransferase (MGMT) immunohistochemistry in glioblastoma multiforme: observer variability and lack of association with patient survival impede its use as clinical biomarker. Brain Pathol. Oct 2008;18(4):520-532. 75. Rodriguez FJ, Thibodeau SN, Jenkins RB, et al. MGMT immunohistochemical expression and promoter methylation in human glioblastoma. Applied immunohistochemistry & molecular morphology : AIMM / official publication of the Society for Applied Immunohistochemistry. Jan 2008;16(1):59-65. 76. Louis DN, Perry A, Burger P, et al. International Society Of Neuropathology-- Haarlem consensus guidelines for nervous system tumor classification and grading. Brain Pathol. Sep 2014;24(5):429-435.

Figure Legend

Figure 1. Overview of altered genes in specific types of gliomas. AA = anaplastic

astrocytoma, GBM = glioblastoma, PXA = pleomorphic xanthoastrocytoma.