Diffuse Gliomas for Nonneuropathologists: The New Integrated Molecular Diagnostics.
This review is intended to provide readers with a brief synopsis on (1) the new diffuse glioma diagnostic algorithm, (2) common genetic alterations found in diffuse gliomas, and (3) a comparison of various approved genetic tests, with an emphasis on targeted next-generation sequencing (NGS).
A NEW DIAGNOSTIC APPROACH TO DIFFUSE GLIOMAS: THE 2016 WHO UPDATE
Before the introduction of the "integrated" molecular and histology diagnostic guidelines, diffuse gliomas were classified into one of the following categories based on histology (2): astrocytoma, oligodendroglioma, or mixed glioma. Along with the histologic type, the tumors were (and still are) graded to WHO II, III, and IV, which were regarded as forming the cornerstone of predicting tumor aggressiveness and patient survival. The criteria for grade III are increasing "anaplastic" features, such as increasing cell density, mitosis, and nuclear anaplasia. The criteria for grade IV (glioblastoma [GBM]) are the presence of microvascular proliferation and/or tumor necrosis, in addition to anaplastic features. As such, much room for subjectivity and variable interpretation existed in the pre-2016 WHO criteria, including no cutoff criteria for distinguishing grade II from grade III. Furthermore, one of the hallmarks of diffuse glioma is the high degree of heterogeneity in intratumoral histology. The latter is particularly true for high-grade gliomas (III or IV), introducing tumor sampling as another factor contributing to variable readings. The challenges and the limitations of this scheme are reflected in such subcategories as "GBM with oligodendroglial component," as well as "mixed oligoastrocytomas," which constituted a large proportion of diffuse glioma diagnoses, pre-2016.
The 2016 WHO update combines histology with tumor genetic information to create a specific diagnostic algorithm (Figure 1). (1) First, the histologic information is used to identify a diffuse glioma/tumor grade. Then, the IDH mutation status is tested, first by immunohistochemistry (IHC) for IDH1 R132H (~90% of all IDH mutations), for which mutant-specific antibody is available (Figure 2). If the tumor is negative for IDH1 R132H, sequencing is requested to identify other minor IDH1/IDH2 mutations (~10%). For GBM patients older than 55 years, sequencing is not recommended because of the very low probability for detecting other IDH mutations in this group (see below).
For IDH-mutant lower-grade gliomas (grades II and III), a set of genetic parameters can distinguish astrocytomas from oligodendrogliomas. For oligodendrogliomas (all are IDH mutant), demonstration of loss of heterozygosity due to chromosomal arm 1p and 19q deletion (1p19q codel) is required. The 1p19q codel is the result of an unbalanced whole-arm chromosomal translocation (with the loss of the derivative chromosome) t(1p;19q) and is a molecular hallmark of oligodendroglioma. Fluorescence in situ hybridization (FISH) is the most commonly used technique for determining 1p19q codel, but other techniques of loss of heterozygosity determination (a method that detects whole-arm deletions) can be used (see below).
Conversely, IDH-mutant astrocytomas (~80% of all astrocytomas) are defined by mutations of ATRX and TP53 (and lack of 1p19q codel). ATRX and TP53 mutations are not present in oligodendrogliomas. For lower-grade gliomas lacking IDH mutations either by IHC or sequencing (~20% of astrocytomas), the diagnosis of IDH-wild-type astrocytoma is established.
For GBM, IDH mutation status is established by IHC (and sequencing if IHC negative in patients <55 years). Although most GBMs are IDH-wild-type, 5% to 15% of GBMs are IDH mutant. Most IDH-mutant GBMs are "secondary" GBMs, meaning most have evidence of progression from previous lower-grade gliomas. Most IDH-wild-type GBMs are "primary" GBMs, meaning they arose de novo. As explained below, the 2 groups ("IDH-mutant" and "IDH- wild-type") of astrocytomas and GBMs have distinct genetic and clinical profiles.
An important contribution of the new integrated diagnosis is the elimination of the mixed oligoastrocytoma category (Figure 3). (3) This is an example of how incorporation of molecular parameters has resolved frequently encountered diagnostic dilemmas in neuropathology. There are no lineage-specific immunohistochemical markers useful for distinguishing astrocytomas from oligodendrogliomas; thus, tumor typing had relied on morphologic identification of astrocytic or oligodendroglial components in a given glioma. Importantly, the new integrated diagnosis based on ATRX and TP53 mutations and 1p19q codel as signature lineage markers for astrocytoma and oligodendroglioma, respectively, shows that the 2 parameters are mutually exclusive, thus completely eliminating the mixed oligoastrocytoma category. In cases of mismatching morphology and genetic parameters (ie, morphology suggests one type but the molecular parameters point to another), the final integrated diagnosis will rely on the molecular determinants.
IDH MUTATIONS, EARLY DRIVING MUTATIONS PREVALENT IN LOWER-GRADE DIFFUSE GLIOMAS: HISTORY AND CLINICAL UTILITY
Among the most important cancer mutations that changed the practice of diffuse glioma diagnosis significantly are a group of missense mutations (a point mutation leading to amino acid change) involving the tricarboxylic acid cycle enzymes IDH1 and IDH2. Studies have shown that once the diagnosis of diffuse glioma is established based on clinical, radiographic, and histologic appearances, determining the IDH mutation status is the single most important step that informs the final diagnosis as well as prognosis.
IDH1 mutations were first discovered in 2008 during unbiased genomic analysis of 22 human GBM samples using NGS technologies. (4) Parsons et al (4) have reported recurrent mutations in the active site of IDH1 in 12% of GBMs, mostly in secondary GBMs in young patients, which were associated with an increase in overall survival. Subsequently, in 2009 Yan et al (5) published a large-scale study sequencing IDH1 and related IDH2 genes in 455 CNS tumors and 494 non-CNS tumors. They identified mutations affecting amino acid 132 (arginine) of IDH1 in more than 70% of WHO grades II and III astrocytomas and oligodendrogliomas, and in secondary glioblastomas. Tumors without mutations in IDH1 (R132) often had mutations affecting the analogous amino acid (R172) of the IDH2 gene. In contrast, IDH mutations were not found in pediatric gliomas, in metastatic tumors, or in nontumor conditions. Thus, IDH mutations have been established as a unique biomarker for adult lower-grade gliomas and a subset of GBMs.
Of note, IDH mutations are also found in certain non-CNS tumors, including acute myelogenous leukemia, myelodysplastic syndrome, intrahepatic cholangiocarcinoma, chondrosarcoma, angioimmunoblastic T-cell lymphoma, and prostate carcinoma. (6,7) In particular, IDH-mutant acute myelogenous leukemias (~20% of all acute myelogenous leukemias) are being actively investigated as a model system in which to develop IDH-targeted therapies, with recent approval of the oral IDH2 inhibitor enasidenib in the United States. (8) These hematopoietic and solid tumors with IDH mutations appear to share similar oncogenic mechanisms--that is, mutant IDH-mediated generation of the oncometabolite 2-hydroxyglutarate (2-HG), leading to genome-wide DNA and histone hypermethylation (see below)--and several different IDH inhibitors are being currently tested for gliomas.
IDH mutations appear to be an early oncogenic event in gliomas, with subsequent mutations critically contributing to lineage-dependent tumor development. Although delineating the underlying mechanisms and therapeutic targeting will be in the works, the practical importance of the discovery of IDH mutations in glioma diagnosis, classification, and patient risk stratification cannot be overemphasized. In both lower-grade astrocytomas and in GBM, patients with IDH-mutant tumors have highly significant survival benefits. For example, in a study of 558 cases of grade II/III diffuse gliomas, Olar et al (9) have found that although survival based on WHO grade alone (pre-2016 criteria) hardly distinguished WHO grade II gliomas from WHO grade III gliomas, survival based on the IDH mutation status alone (regardless of WHO grade) separated the 2 groups much more significantly (Figure 4). Furthermore, among IDH-mutant tumors, WHO grade II patients had no survival advantage over grade III patients, a surprising finding with practical significance (Figure 5). These results have been reproduced by other groups, including the Heidelberg group. (10)
BIOLOGY OF THE IDH ENZYMES: GENERATION OF 2-HG AND CREATION OF A GLIOMA-CpG ISLAND METHYLATOR PHENOTYPE BY MUTANT IDH
IDHs are homodimeric enzymes in the Krebs cycle that normally catalyze the oxidative carboxylation of isocitrate to [alpha]-ketoglutarate ([alpha]-KG; Figure 6). (6,11,12) They play a role in the metabolism of glucose, fatty acids, and glutamine, and contribute to the maintenance of the normal cellular redox state. Two isoforms (the cytosolic IDH1 and the mitochondrial IDH2) are found mutated in gliomas and other tumors. The mutant proteins in diffuse gliomas originating from various mutations (IDH1 R132H in ~90% and other IDH1 R132 or IDH2 R172 mutations in ~10%) display the same biochemical and metabolic activity. Though initially thought to be a loss-of-function mutation with reduced levels of normal metabolite, subsequent studies have shown that the mutant IDH proteins acquired a gain-of-function neoenzymatic activity ("neomorphic" mutation). (13) The mutant IDH directly works on [alpha]-KG and enables its conversion to an "oncometabolite" 2-HG in a NADPH-dependent manner. Of the 2 stereoisomers (L and D), the product of the mutant IDH is exclusively the D isomer (D-2-HG). Very low levels of D and L isomers are produced by normal cells, with overlapping and opposing functions described previously. (6,14) Many of the biologic sequelae of the IDH mutations can be ascribed to the concomitant loss of normal metabolite [alpha]-KG and the gain of the abnormal metabolite 2-HG at very high (millimolar) levels. Furthermore, 2-HG binds to and functions as a competitive inhibitor of the enzymes that are dependent on [alpha]-KG.
By far the most significant biochemical process affected by 2-HG appears to be the interference of [alpha]-KG-dependent dioxygenases and hydroxylases that are critical in maintaining DNA and histone proteins in a demethylated state. The prime targets include the TET family proteins (TET2) and the Jumonji C histone demethylases. The net effect is the creation of hypermethylated DNA across the genome, referred to as the "glioma-CpG island methylator phenotype (G-CIMP)". (15) There is a direct causal relationship between IDH mutation and G-CIMP, because the characteristic methylator phenotype can be created by the mutant IDH protein experimentally in vitro and in vivo. Identification of IDH mutations thus has provided a highly interesting model system in which somatic mutations lead to abnormal cellular metabolism, and ultimately to an altered global epigenetic state, setting in motion a number of tumorigenesis events.
Another important facet of IDH mutations is 2-HG-mediated modulation of prolyl hydroxylases, in particular the [alpha]-KG-dependent prolyl hydroxylase, EGLN. (16) EGLN is a major regulator of hypoxia-induced factor (HIF-1[alpha]), a transcription factor that activates an array of genes important in the cellular response to hypoxia. HIF-1 is a heterodimer comprising HIF-1[alpha] and HIF-1[beta] subunits. HIF-1[alpha] is active under hypoxic conditions but becomes unstable and degraded by von Hippel-Lindau protein (VHL, an ubiquitin E3 ligase) in the presence of oxygen. EGLN enables HIF-1[alpha] proteosomal degradation by creating a binding site for the VHL ubiquitin-ligase complex. Although it was originally reported to increase HIF-1[alpha] activity, subsequent work in the Kaelin Laboratory has shown that (D) 2-HG activates EGLN, causing a reduction in HIF-1[alpha] signaling. (16) Because various isoforms of HIF proteins have been implicated in the proinvasive and proangiogenic properties of malignant (high-grade) gliomas, the pathogenetic implications of the actions of mutant IDH on HIF protein degradation are unclear and may even seem counterintuitive. It is also entirely feasible that not all biologic aspects of the IDH mutations contribute to tumor formation and progression, and that the uniformly favorable outcome of IDH-mutant gliomas may in part be attributable to the balancing acts of the IDH mutant proteins/2-HG that contribute to the less aggressive behavior of the tumor. Caution must be exercised in therapeutically targeting IDH-mutant tumors.
[alpha]-THALASSEMIA/MENTAL RETARDATION SYNDROME X-LINKED GENE AND TP53 MUTATIONS ARE MARKERS OF IDH-MUTANT ASTROCYTOMA
Mutations of the chromatin-remodeling protein [alpha]-thalassemia/mental retardation syndrome X-linked gene (ATRX) has been associated with "alternative lengthening of telomeres" (ALT), a telomerase-independent mechanism used by cancer cells to overcome replicative senescence, in a manner dependent on chromosomal homologous recombination. (17) Missense and truncating mutations of ATRX are common (~90%) in IDH-mutant astrocytomas, and can be detected by IHC (loss of expression), and this information is built into the 2016 WHO diagnostic algorithm (Figure 1). ATRX mutations in diffuse gliomas often coexist with TP53 mutations in adults, and with H3.3 histone mutations in children and young adults (see below). More importantly, ATRX mutations are not seen in oligodendrogliomas or in GBM; thus, it is useful as an astrocytic lineage marker. Of note, ATRX mutations are mutually exclusive with the related telomerase reverse transcriptase (TERT) promoter mutations, which are prevalent in oligodendrogliomas and GBM. (18) TERT mutations are associated with telomerase activation, and thus telomerase-dependent telomere lengthening. TERT mutations are detected by DNA sequencing (no IHC surrogate).
Perhaps one of the most prevalent mutations found in cancer is that of the tumor suppressor gene TP53. The transcription factor p53 functions as a tumor suppressor by acting on DNA repair mechanisms. Although they are not unique to gliomas, TP53 mutations are useful markers for astrocytic lineage in the context of IDH mutations (Figure 1). Most TP53 mutations in glioma are missense mutations that affect the half-life of the protein, and thus are detectable by IHC using antibodies against normal p53. (19)
IDH-WILD-TYPE ASTROCYTOMAS: HOW DO THEY FARE?
Having delineated "IDH-mutant glioma" as a new entity, there has been much effort to understand the IDH-wild-type diffuse glioma group with regard to its genetic, epigenetic, biologic, and clinical parameters. Evidence supports the conclusion that most IDH-wild-type lower-grade astrocytomas may essentially behave as a "molecular GBM." (20,21) This is perhaps one of the most striking examples of how far our understanding of glioma biology has evolved. Previously, patient survival in gliomas was primarily predicated based on the tumor grade (II, III, or IV). Now, with the new integrated diagnosis approach, the boundaries between WHO grades are blurring. (22) Specifically, molecular profiling of large numbers of brain tumors, including the Cancer Genome Atlas network studies, has demonstrated that the large majority of lower-grade astrocytomas without an IDH mutation have genomic alterations and clinical behavior strikingly similar to those of the primary (ie, IDH-wild-type) GBMs (Figure 7). These studies have shown that most IDH-wild-type lower-grade astrocytomas essentially behave like IDH-wild-type GBM. (20) In other words, the single most important predictor of overall survival in adults with diffuse glioma is IDH mutation status, and WHO grading under most circumstances may have limited significance as a predictor of tumor behavior.
THE GENOMIC PROFILES OF IDH-WILD-TYPE GLIOMAS
The genomic profile of the IDH-wild-type GBM is characterized by certain signature mutations of the genes involved in cell growth, cell cycle/proliferation, and survival. These include the growth factor receptor gene EGFR, which is amplified (copy number variation) in approximately 40% of the GBMs, with approximately half of these also carrying a structural variation resulting from an in-frame deletion creating a unique extracellular EGFR domain called variant III (EGFRvIII).
Other frequent types of mutations in IDH-wild-type GBM involve deletions in the cell cycle regulatory genes CDKN2A/B (p16). Mutations of CDKN2A/B have also been associated with other types of gliomas (including lower-grade gliomas and, rarely, pilocytic astrocytoma-like gliomas (21)) that appear to be a harbinger of malignant transformation and aggressive behavior.
The phosphatase PTEN gene normally functions as a negative regulator of the cell survival kinase pathway PI3K, downstream of AKT. PTEN-inactivating mutations and deletions occur in approximately 20% of GBMs. Amplifications and single-nucleotide variations in the various regulatory and catalytic subunits of the PI3K kinase (PIK3R, PIK3C, and PIK3C2) are also found in approximately 20% of GBMs. Resulting overactivation of the PI3K pathway likely cooperates with other mutations, such as EGFR and CDKN2A/B, and drives oncogenesis and tumor progression in GBM.
Additional less frequent mutations found in IDH-wild-type GBM include genes of the growth factor receptor PDGFR, the Ras proto-oncogene regulatory protein NF1, and the tumor suppressor gene TP53. These mutations have been used to profile and classify malignant (high-grade: III and IV) gliomas in earlier Cancer Genome Atlas work (PDGFR and TP53 for "proneural" subtype and NF1 for "mesenchymal" subtype, with favorable and unfavorable prognosis, respectively). (23) However, these mutations have not been incorporated into the 2016 WHO diagnostic algorithm for GBM subtyping.
Thus, the characteristic gene alterations found in IDH-wild-type gliomas are those involved in canonical receptor tyrosine kinase pathways and downstream signaling pathways involved in cell growth, survival, and proliferation. Most of these genomic abnormalities are not unique to CNS tumors, and they have been targeted therapeutically by way of monoclonal antibodies and small-molecule inhibitors as for non-CNS tumors. However, therapeutic benefits have not yet been achieved with these agents, demonstrating unique challenges associated with CNS tumors. Likewise, clinical trials involving immunotherapy against the tumor-specific antigen EGFRvIII have not yielded a significant response.
H3K27M, A NOVEL MUTATION PREVALENT IN DIFFUSE MIDLINE GLIOMAS OF CHILDREN AND YOUNG ADULTS
Another group of new genetic alterations found in diffuse gliomas is that involving histone H3 genes encoding H3.3 and H3.1 proteins. (24,25) The 4 histone proteins H2A, H2B, H3, and H4 are important scaffolding proteins for DNA packaging into structural units called nucleosomes. Histone tails are the sites at which numerous modifications, such as methylation, occur, which in turn are associated with either activation or inhibition of transcription of associated genes. Histone methylation is tightly controlled by a balance between histone methyltransferases and demethylases, with the latter removing methyl groups from lysine residues on histone tails that are monomethylated, bimethylated, or trimethylated. Alterations of this fine balance have significant effects on gene expression.
"Diffuse midline glioma with H3K27M mutation" is a new entity in the 2016 WHO update. Missense mutations leading to substitution of the lysine residue at position 27 for methionine (K27M) on H3.3 (majority) or H3.1 (minority) inhibit the normal trimethylation (me3) of the protein (H3K27me3). H3K27M midline gliomas are characterized by histologically high-grade diffuse gliomas (with protean morphologic presentations) (26) associated with aggressive clinical behavior (another "molecular GBM"). They arise within midline structures, such as thalamus, pons (brainstem), cerebellum, and spinal cord, whereas another missense mutation involving H3.3, H3G34R/V (glycine at 34 to arginine or valine) is associated with rarer hemispheric diffuse gliomas in this age group. (27)
Detection of H3K27M mutation is greatly facilitated by IHC with a mutant-specific antibody (reacts with the predominant H3.3 K27M, as well as H3.1 K27M), in conjunction with loss of immunoreactivity for H3K27me3 in tumor cells. Although H3K27M typically heralds aggressive clinical behavior, rare instances of nonaggressive (and even nonmidline) gliomas have been reported, usually associated with low-grade or nondiffuse histology. Furthermore, H3K27M-positive ependymoma is a rare but distinct entity. (28,29) Thus, the significance of H3K27M mutation in nondiffuse astrocytomas remains unclear.
With regard to its utility in the diagnostic workup of diffuse gliomas of the adult age group, H3K27M mutations are associated with p53 and ATRX mutations, (30) but they are mutually exclusive with IDH (or most EGFR) mutations. These features are reflected in the analysis of age distribution of the mutations (Figure 8), indicating an age dependence of the mutations. However, as more cases are found, the spectrum of H3K27M gliomas is being expanded. H3K27M mutation is being increasingly recognized in adults, with uniform midline presentation and poor prognosis. (31) Many of these cases have atypical and confusing (many presenting as low-grade) histologic presentations, and as such, a high index of suspicion should be exercised. These cases also strengthen the argument for adopting NGS as the first line of investigation (see below).
MGMT PROMOTER METHYLATION STATUS, A REQUIRED TEST FOR ALL GBMs THAT PREDICTS CHEMOTHERAPY RESPONSE
The standard of care for GBM is surgery, radiotherapy, and chemotherapy. The introduction of the new alkylating agent temozolomide has resulted in an approximately 3- to 6-month increase in overall survival, and temozolomide has become integral to various glioma treatment protocols. Importantly, a landmark study published in 2005 revealed that epigenetic modification (methylation) of the promoter of the DNA repair enzyme O-6-methylguanine-DNA methyltransferase (MGMT) correlated significantly with improved survival following temozolomide treatment. (32) Hypermethylation of the promoter silences expression of the MGMT gene, thus disabling the DNA repair mechanism in tumor cells damaged by alkylating agents. Hypermethylated MGMT is found in approximately 45% of all GBMs and predicts a favorable prognosis. Although currently there is no alternative treatment that can be offered to the methylation-negative group, clinicians routinely inquire about MGMT methylation status. MGMT methylation can be detected by methylation-specific polymerase chain reaction, and among several techniques employed currently, pyrosequencing appears to be the method of choice.
THE CURRENT APPROACH FOR DIFFUSE GLIOMA WORKUP: TAILORED APPROACH VERSUS TARGETED NGS
Currently, the workflow of diffuse glioma diagnosis in most institutions is as follows:
1. Histologic confirmation of the nature of the tumor as a diffuse glioma or closely related entity on a hematoxylineosin stain preparation.
2. IHC for IDH1 R132H
3. IHC for p53 and/or ATRX, if available.
4. FISH for 1p19q codel, if histology suggests the possibility of an oligodendroglioma or if mutant IDH is detected.
5. If IDH1 (R132H) is negative by IHC, sequencing for IDH1/IDH2 mutations is requested.
6. If histology suggests grade IV (GBM), MGMT methylation test is ordered.
Diagnostic neuropathology, like other subspecialties in surgical pathology, is confronted with an ever-expanding list of molecular (or surrogate) tests for diagnostic workup. Although the new WHO guideline (Figure 1) specifies which genetic parameters should be determined, it does not indicate the specific methods. This adds to the challenge, because most practicing pathologists are in a resource-poor environment with regulatory, cost, and other considerations. There are several factors to be considered in choosing the tests, including test sensitivity/specificity, turnaround time (with the goal of generating a final diagnosis within a reasonable time frame), and insurance reimbursement, as well as the workflow/algorithm that is efficient for the specific laboratory and personnel, including the pathologist.
IMMUNOHISTOCHEMISTRY FOR IDH1 R132H, ATRX, P53, AND H3K27M
The advantages of IHC are obvious: inexpensive, short turnaround time, sensitive at the individual cell level, and accessible to most pathologists. (19) The availability of mutant-specific antibodies for IDH1 R132H and H3K27M has greatly facilitated the identification of these new tumor entities. Immunohistochemistry provides cell-specific information that equips neuropathologists with a great advantage. For example, staining for IDH1 R132H offers a rare opportunity to visualize individual neoplastic glial cells and distinguish them from preexisting brain cells, because this mutant protein is one of the rare tumor-specific antigens (Figure 9). Identification of the IDH1 R132H protein in otherwise uncharacteristic tissue can help establish a diagnosis of diffuse glioma, and exclude the possibility of other (morphologically similar, in a limited tissue sample) entities, such as ganglioglioma (33) or nonneoplastic conditions (inflammation or infection).
Immunohistochemical detection of ATRX mutations or TP53 mutations also greatly facilitates identification of IDH-mutant astrocytomas. ATRX mutations are inactivating mutations leading to a loss of protein expression in tumor cell nuclei that can be detected by IHC. ATRX is normally expressed in nonneoplastic cells (endothelial cells, inflammatory cells, and preexisting brain cells) in these tissues. Given that IHC is readily available and practical in most institutions (as opposed to FISH or sequencing), a diffuse glioma diagnostic algorithm using ATRX (and IDH1) as the first line of investigation has been proposed. (3)
Most TP53 mutations are missense mutations that lead to increased protein half-life, which translates to immunohistochemically detectable p53 expression. However, the pattern of p53 protein expression in TP53-mutated glioma is not uniform with regard to staining intensity and the number of positive cells. Furthermore, although strong nuclear p53 immunoreactivity is unique to neoplastic cells, nontumor cells under stress conditions can also become p53 positive (false positive). In addition, rare truncating mutations do not lead to protein overexpression (false negative).
Thus, there are downsides to IHC as well, such as false-positive/false-negative results in tissues that are artefactually damaged (ATRX and IDH1), (3) a lack of clear cutoff criteria for immunoreactivity (p53), (19) and cross-reactivity for nonneoplastic tissue (for example, microglial staining in H3K27M IHC). Therefore, although IHC will not be completely replaced by sequencing, there are enough caveats associated with IHC (methodologic, interpretational, and algorithm-wise) that make targeted NGS an attractive alternative.
1p19q CODEL: A COMPARISON OF DIFFERENT TESTS FOR OLIGODENDROGLIOMA
Documenting chromosomal arms 1p and 19q codeletion is currently required to establish the diagnosis of oligodendroglioma. The most frequently used technique is FISH, in which the codeletion status is inferred based on the probes used to cover a short segment of each of the chromosomal arms, 1p and 19q. (34) Fluorescence in situ hybridization is a more demanding technique than IHC (or sequencing), with an average failure rate of approximately 5%. Also, FISH is expensive, with a long turnaround time. On the other hand, it is entirely feasible to diagnose oligodendroglioma based on sequencing data, such as the presence of IDH and TERT mutations, as well as other less frequent but oligodendroglioma-specific mutations, such as far upstream elementbinding protein (FUSBP) and the homolog of the Drosophila gene capicua (CIC), that contribute to oncogenesis in the setting of loss of heterozygosity. (35) The absence of ATRX and TP53 mutations in an IDH-mutated tumor also strongly point toward oligodendroglioma.
TARGETED NGS APPROACH: THE INEVITABLE FUTURE?
Given the complexity outlined, we can take an unbiased NGS approach for all gliomas rather than relying on individual hand-picked tests chosen based on availability, cost, and other considerations. There are several different types of NGS panels available commercially or institutionally. These include exome sequencing (no target genes), targeted cancer NGS gene panels (not CNS specific), and a CNS (brain tumor)-specific targeted NGS panel. Obvious advantages of NGS include:
1. NGS simplifies the workflow.
2. Testing is not incumbent on another (no diagnostic algorithm).
3. Reliance on a pattern of gene alteration rather than a single gene alteration or several gene alterations, increasing the confidence level.
4. Diagnosis of an unsuspected entity with atypical presentations.
5. Discovery of "drug-able" targets.
6. Could be cost-saving by obviating the need for individual dedicated assays.
The disadvantages include long turnaround time (~2 weeks), and the fact that the fees are not universally reimbursed by insurance. There are a number of institutions that accept outside cases at a nominal cost (~$1600). Glioseq, developed at the University of Pittsburgh, (36) is one example (http://mgp.upmc.com/Applications/mgp/Home/ Test/GlioSeq_Details, accessed November 16, 2017). At present, Glioseq is the only targeted NGS panel that is tailored to brain tumors.
There are clear advantages for using a targeted NGS panel, such as Glioseq, that is capable of detecting a wide range of gene alterations, including single-nucleotide variations, (small) indel, copy number variations, and structural variations (fusions/rearrangements), all in a single workflow. The panel includes IDH, ATRX, TP53, H3K27M, and TERT, as well as the genes frequently altered in IDH-wild-type gliomas (EGFR, CDKN2A/B, PTEN, PIK3, and NF1). Alterations in FUSBP and CIC, located in 1p and 19q, respectively, are likely to be found in oligodendrogliomas.
Glioseq covers BRAF fusions and BRAF V600E mutation, 2 important gene alterations found in pediatric gliomas, with resulting constitutive activation of the RAF-MEK-MAPK signaling pathway. BRAF fusions are now becoming virtually synonymous with the diagnosis of pilocytic astrocytoma (37) (WHO grade I), the most frequent CNS tumor in children. BRAF fusions/rearrangements are detectable by FISH, but the tests are not available commercially.
BRAF V600E is present in a wide range of brain tumors, including ganglioglioma (~30%), pleomorphic xanthoastrocytoma (~40%), and epithelioid GBM (~50%), and is mutually exclusive with BRAF fusions. A BRAF V600E mutant-specific antibody is available for IHC which is widely used in diagnostic neuropathology (in combination with sequencing) and in dermatopathology (BRAF V600E is found in ~50% of melanomas).
Demonstration of the presence or absence of signature genetic alterations is an essential aspect of today's diagnostic pathology and is becoming increasingly important in arriving at the correct tumor diagnosis. In neuropathology, these cases with atypical clinical, radiographic, and pathologic presentations; tumors arising from deep locations; and tumors with limited tissue sampling are not infrequent. The process of brain tumor diagnosis is far more complex, with many overlaps in clinical, demographic, radiographic, neurosurgical, and histologic features. Currently, the field is scrambling to configure algorithms for various (suspected) brain tumor entities, with each institution dealing with unique challenges, mostly in a limited-resource environment. Thus, there is an enormous advantage in examining a wide range of predetermined sets of gene abnormalities (casting a wider net), such as NGS targeted cancer panels, to arrive at a meaningful personalized diagnosis. This approach will also simplify and streamline the workflow for the laboratory and the pathologist, will eventually prove to be cost-saving, and will enable the establishment of a large database generated using a standard test tool. The latter will undoubtedly help further our understanding of the biology of brain tumors and likely increase the opportunity to improve their outcomes.
The author is grateful to Cynthia Hawkins, MD, PhD, and Scott Ryall, BSc, for providing Figure 8; Rosemarie Didonato, MD, for photographing the case for Figure 9; Cynthia Hawkins, MD, PhD, and Celia Brosnan, PhD, for reading the manuscript; and Joseph Locker, MD, PhD, for helpful discussions. Many excellent papers on the subject of diffuse glioma integrated diagnostics could not be cited because of the lack of space.
(1.) Louis DN, Perry A, Reifenberger G, et al. The 2016 World Health Organization Classification ofTumors of the Central Nervous System: a summary. Acta Neuropathol. 2016;131(6):803-820.
(2.) Louis DN, Ohgaki H, Wiestler OD, et al. The 2007 WHO classification of tumours of the central nervous system. Acta Neuropathol. 2007;114(2):97-109.
(3.) 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. 2015; 129(1):133-146.
(4.) Parsons DW, Jones S, Zhang X, et al. An integrated genomic analysis of human glioblastoma multiforme. Science. 2008;321(5897):1807-1812.
(5.) Yan H, Parsons DW, Jin G, et al. IDH1 and IDH2 mutations in gliomas. N Engl J Med. 2009;360(8):765-773.
(6.) Cairns RA, Mak TW. Oncogenic isocitrate dehydrogenase mutations: mechanisms, models, and clinical opportunities. Cancer Discov. 2013;3(7):730-741.
(7.) Lemonnier F, Cairns RA, Inoue S, et al. The IDH2 R172K mutation associated with angioimmunoblastic T-cell lymphoma produces 2HG in T cells and impacts lymphoid development. Proc Natl Acad Sci USA. 2016;113(52): 15084-15089.
(8.) Kim ES. Enasidenib: first global approval. Drugs. 2017;77(15):1705-1711.
(9.) 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. 2015;129(4):585-596.
(10.) Reuss DE, Mamatjan Y, Schrimpf D, et al. IDH mutant diffuse and anaplastic astrocytomas have similar age at presentation and little difference in survival: a grading problem for WHO. Acta Neuropathol. 2015;129(6):867-873.
(11.) Dimitrov L, Hong CS, Yang C, Zhuang Z, HeissJD. New developments in the pathogenesis and therapeutic targeting of the IDH1 mutation in glioma. Int J MedSci. 2015;12(3):201-213.
(12.) Masui K, Cavenee WK, Mischel PS. Cancer metabolism as a central driving force of glioma pathogenesis. Brain Tumor Pathol. 2016;33(3):161-168.
(13.) Dang L, White DW, Gross S, et al. Cancer-associated IDH1 mutations produce 2-hydroxyglutarate. Nature. 2009;462(7274):739-744.
(14.) Wahl DR, Venneti S. 2-Hydoxyglutarate: D/Riving Pathology in gLiomaS. Brain Pathol. 2015;25(6):760-768.
(15.) Noushmehr H, Weisenberger DJ, Diefes K, et al. Identification of a CpG island methylator phenotype that defines a distinct subgroup of glioma. Cancer Cell. 2010;17(5):510-522.
(16.) Koivunen P, Lee S, Duncan CG, et al. Transformation by the (R)-enantiomer of 2-hydroxyglutarate linked to EGLN activation. Nature. 2012;483(7390):484-488.
(17.) Nguyen DN, Heaphy CM, de Wilde RF, et al. Molecular and morphologic correlates of the alternative lengthening of telomeres phenotype in high-grade astrocytomas. Brain Pathol. 2013;23(3):237-243.
(18.) Eckel-Passow JE, Lachance DH, Molinaro AM, et al. Glioma groups based on 1p/19q, IDH, and TERT promoter mutations in tumors. N Engl J Med. 2015; 372(26):2499-2508.
(19.) Tanboon J, Williams EA, Louis DN. The diagnostic use of immunohistochemical surrogates for signature molecular genetic alterations in gliomas. J Neuropathol Exp Neurol. 2016;75(1):4-18.
(20.) Cancer Genome Atlas Research Network, Brat DJ, Verhaak RG, et al. Comprehensive, integrative genomic analysis of diffuse lower-grade gliomas. N Engl J Med. 2015;372(26):2481-2498.
(21.) Ceccarelli M, Barthel FP, Malta TM, et al. Molecular profiling reveals biologically discrete subsets and pathways of progression in diffuse glioma. Cell. 2016;164(3):550-563.
(22.) Ramaswamy V, Taylor MD. Fall of the optical wall: freedom from the tyranny of the microscope improves glioma risk stratification. Cancer Cell. 2016; 29(2):137-138.
(23.) Comprehensive genomic characterization defines human glioblastoma genes and core pathways. Nature. 2008;455(7216):1061-1068.
(24.) Buczkowicz P, Hoeman C, Rakopoulos P, et al. Genomic analysis of diffuse intrinsic pontine gliomas identifies three molecular subgroups and recurrent activating ACVR1 mutations. Nat Genet. 2014;46(5):451-456.
(25.) Khuong-Quang DA, Buczkowicz P, Rakopoulos P, et al. K27M mutation in histone H3.3 defines clinically and biologically distinct subgroups of pediatric diffuse intrinsic pontine gliomas. Acta Neuropathol. 2012;124(3):439-447.
(26.) Solomon DA, Wood MD, Tihan T, et al. Diffuse midline gliomas with histone H3-K27M mutation: a series of 47 cases assessing the spectrum of morphologic variation and associated genetic alterations. Brain Pathol. 2016; 26(5):569-580.
(27.) Ryall S, Tabori U, Hawkins C. A comprehensive review of paediatric low-grade diffuse glioma: pathology, molecular genetics and treatment. Brain Tumor Pathol. 2017;34(2):51-61.
(28.) Panwalkar P, Clark J, Ramaswamy V, et al. Immunohistochemical analysis of H3K27me3 demonstrates global reduction in group-A childhood posterior fossa ependymoma and is a powerful predictor of outcome. Acta Neuropathol. 2017;134(5):705-714.
(29.) Ryall S, Guzman M, Elbabaa SK, et al. H3 K27M mutations are extremely rare in posterior fossa group A ependymoma. Childs NervSyst. 2017;33(7):1047-1051.
(30.) Schwartzentruber J, Korshunov A, Liu XY, et al. Driver mutations in histone H3.3 and chromatin remodelling genes in paediatric glioblastoma. Nature. 2012; 482(7384):226-231.
(31.) Meyronet D, Esteban-Mader M, Bonnet C, et al. Characteristics of H3 K27M-mutant gliomas in adults. Neuro Oncol. 2017;19(8):1127-1134.
(32.) Hegi ME, Diserens AC, Gorlia T, et al. MGMT gene silencing and benefit from temozolomide in glioblastoma. N Engl J Med. 2005;352(10):997-1003.
(33.) Horbinski C, Kofler J, Yeaney G, et al. Isocitrate dehydrogenase 1 analysis differentiates gangliogliomas from infiltrative gliomas. Brain Pathol. 2011;21(5): 564-574.
(34.) Horbinski C, Miller CR, Perry A. Gone FISHing: clinical lessons learned in brain tumor molecular diagnostics over the last decade. Brain Pathol. 2011;21(1): 57-73.
(35.) Bettegowda C, Agrawal N, Jiao Y, et al. Mutations in CIC and FUBP1 contribute to human oligodendroglioma. Science. 2011;333(6048):1453-1455.
(36.) Nikiforova MN, Wald AI, Melan MA, et al. Targeted next-generation sequencing panel (GlioSeq) provides comprehensive genetic profiling of central nervous system tumors. Neuro Oncol. 2016;18(3):379-387.
(37.) Horbinski C. To BRAF or not to BRAF: is that even a question anymore? J Neuropathol Exp Neurol. 2013;72(1):2-7.
(38.) Sturm D, Witt H, Hovestadt V, et al. Hotspot mutations in H3F3A and IDH1 define distinct epigenetic and biological subgroups of glioblastoma. Cancer Cell. 2012;22(4):425-437.
(39.) Ryall S, Krishnatry R, Arnoldo A, et al. Targeted detection of genetic alterations reveal the prognostic impact of H3K27M and MAPK pathway aberrations in paediatric thalamic glioma. Acta Neuropathol Commun. 2016; 4(1):93.
Sunhee C. Lee, MD
Accepted for publication December 5, 2017.
Published as an Early Online Release May 18, 2018.
From the Department of Pathology, Albert Einstein College of Medicine, and the Department of Neuropathology, Montefiore Medical Center, Bronx, New York.
The author has no relevant financial interest in the products or companies described in this article.
Presented at the 16th Spring Seminar of the Korean Pathologists Association of North America (KOPANA); March 3, 2017; San Antonio, Texas.
Corresponding author: Sunhee C. Lee, MD, Albert Einstein College of Medicine, 1300 Morris Park Ave, Bronx, NY 10461 (email: email@example.com).
Caption: Figure 1. A simplified algorithm for classification of the diffuse gliomas based on histologic and genetic features (see text and 2016 World Health Organization [WHO] criteria on tumors of the central nervous system [CNS] for details). Abbreviations: IDH, isocitrate dehydrogenase; NOS, not otherwise specified. Reproduced from Louis DN, Perry A, Reifenberger G, et al. The 2016 World Health Organization Classification of Tumors of the Central Nervous System: a summary. Acta Neuropathol. 2016;131(6):803-820, with permission from Springer. (1)
Caption: Figure 2. Isocitrate dehydrogenase 1 (IDH1) R132H immunoreactivity in a diffuse astrocytoma. Note the immunoreactivity in most cells representing neoplastic cells. The staining is in the cytoplasm, and also in the nucleus to some degree. The unstained cells in the brain parenchyma represent nonneoplastic cells (arrows: endothelial cells, preexisting brain cells, and inflammatory cells; original magnification approximately X700).
Caption: Figure 3. The integrated diagnostic approach no longer recognizes a mixed glioma category: Changes from initial (pre-2016) to integrated (World Health Organization 2016) diagnoses in 100 patients. Oligodendrogliomas (O), oligoastrocytomas (OA), astrocytoma (A), and glioblastoma (GBM) are shown. Width of bars indicates relative proportions of the initial tumor groups. Abbreviations: A-IDHmut, IDH-mutant astrocytoma; A-IDHwt, IDH-wild-type astrocytoma; IDH, isocitrate dehydrogenase. Reproduced from 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. 2015;129(1):133-146, with permission from Springer. (3)
Caption: Figure 4. Stratification of lower-grade gliomas (LGGs) into subsets defined by presence or absence of isocitrate dehydrogenase (IDH) mutation leads to subgroups with distinct prognostic characteristics. A, Overall survival among 558 grades II to III diffuse gliomas stratified by World Health Organization (WHO) grade. B, Overall survival among 558 grades II to III diffuse gliomas stratified by IDH mutation status. C, Overall survival among 558 grades II to III diffuse gliomas stratified by IDH mutation and 1p19q codeletion status. Abbreviations: m, mutant; non-co-del, without 1p19q codeletion; wt, wild-type. Reproduced from 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. 2015;129(4):585-596, with permission from Springer. (9)
Caption: Figure 5. No survival difference between World Health Organization (WHO) grades II and III among isocitrate dehydrogenase (IDH)-mutant (IDHm) gliomas. A, Overall survival among IDH-mutated diffuse gliomas (n =475) stratified by WHO grade. B, Overall survival among IDH-wild-type diffuse gliomas (n =83) stratified by WHO grade. Abbreviation: IDHwt, wild-type IDH. Reproduced from 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. 2015;129(4):585-596, with permission from Springer. (9)
Caption: Figure 6. Metabolic reprogramming in isocitrate dehydrogenase (IDH)-mutated (mIDH) gliomas. Mutations in IDH play an important role in gliomas through their neomorphic activity that converts a-ketoglutarate ([alpha]-KG) to an oncometabolite, 2-hydroxyglutarate (2-HG). 2-Hydroxyglutarate stimulates the activity of EGLN prolyl-hydroxylases, enhancing cellular proliferation through degradation of hypoxia-inducible factor (HIF), and it also inhibits [alpha]-KG-dependent dioxygenases, including Jumonji C histone lysine demethylases (KDM) and the ten-eleven translocation (TET) family of 5'-methylcytosine hydroxylases, leading to a methylator phenotype. Conversion of [alpha]-KG to 2-HG decreases intracellular NADPH levels, which contributes to oncogenesis by creating a pro-oxidant state. Abbreviations: GSH, reduced glutathione; mut, mutant; ROS, reactive oxygen species; RTK: receptor tyrosine kinase; TERT, telomerase reverse transcriptase. Reproduced from Masui et al (12) with permission from the Japan Society of Brain Tumor Pathology.
Caption: Figure 7. Most lower-grade gliomas (LGGs) without IDH mutation are "molecular GBM": frequencies in the LGG molecular subtypes of mutation events that are commonly found in primary glioblastoma (GBM). These groups include LGG with isocitrate dehydrogenase (IDH) mutation and 1p19q codeletion (codel; n =85), LGG with IDH mutation but without codel (n = 141), and LGG without IDH mutation (n =56). Abbreviations: SNV, single-nucleotide variant; SV, structural variant. Reproduced from Cancer Genome Atlas Research Network, Brat DJ, Verhaak RG, et al. Comprehensive, Integrative genomic analysis of diffuse lower-grade gliomas. N Engl J Med. 2015;372(26):2481-2498. Reprinted with permission from the Massachusetts Medical Society. (20)
Caption: Figure 8. Frequency of mutations in pediatric high-grade gliomas according to the age of diagnosis. The y-axis indicates the percentage of each age category that contains a certain mutation. Samples categorized as wild type are negative for both histone mutations (K27M and G34R/V) and isocitrate dehydrogenase 1 (IDH1; R132H). Data are extrapolated from previously published studies, (24,30,38,39) and from Ryall S. and Hawkins C, unpublished data (2017).
Caption: Figure 9. Isocitrate dehydrogenase 1 (IDH1) R132H antibody can identify individual infiltrating tumor cells at the edge of the tumor. A, Hematoxylin-eosin stain reveals cerebral cortex with minimally increased cellularity. B, Ki-67 stain fails to detect positive cells. C, IDH1 stain reveals individually scattered positive cells, confirming the diagnosis of IDH-mutant glioma (a case of IDH1-mutant, 1p19q codeleted oligodendroglioma; original magnification X100).
|Printer friendly Cite/link Email Feedback|
|Author:||Lee, Sunhee C.|
|Publication:||Archives of Pathology & Laboratory Medicine|
|Date:||Jul 1, 2018|
|Previous Article:||The 16th Spring Seminar of the Korean Pathologists Association of North America: Critical Review and Appraisal of Recently Described Entities and...|
|Next Article:||Immunohistochemical Expression of p16 in Melanocytic Lesions: An Updated Review and Meta-analysis.|