Continuing the trend from the 2016 revised 4th edition of the WHO Classification of Tumors of the Central Nervous System (CNS), the recent 2021 5th edition incorporates more molecular data into the classification and grading of many entities and introduces some new entities based on molecular alterations. Insight gathered from DNA methylation array tumor profiling has been a powerful tool in refining and occasionally defining certain entities. Updates to many tumor types, including glioneuronal, pineal region, embryonal, and mesenchymal tumors, are incorporated into the new classification. This brief review focuses mainly on glioma and meningioma classification, highlighting changes from the prior classification and their impact on diagnostic pathology practices.1,2
A major change in the adult infiltrating glioma paradigm is defining glioblastoma as an IDH1/IDH2-wildtype infiltrating astrocytic glioma in the adult setting. Thus, the term secondary glioblastoma for an IDH-mutant astrocytoma that presents with or has progressed with aggressive histology (ie, tumor necrosis or microvascular proliferation) is now termed astrocytoma, IDH-mutant, CNS WHO grade 4. In the 2016 WHO classification, the diagnosis of glioblastoma required the histologic features of tumor necrosis and/or microvascular proliferation. In the 2021 WHO classification, these features are not required if at least one of the following molecular features are identified: EGFR amplification or combined chromosome 7 gain and chromosome 10 loss or TERT promoter mutation (hotspots c.-124 (referred to as C228T) and c.-146 (referred to as C250T)). Even in the absence of increased mitotic activity, necrosis, or microvascular proliferation, IDH-wildtype infiltrating astrocytic gliomas with these molecular features have been shown to behave aggressively with overall survival times comparable to that of histologically classic glioblastoma, IDH-wildtype, CNS WHO grade 4.3,4,5
The nomenclature and grading scheme of IDH-mutant astrocytoma also has been revised. The terms diffuse astrocytoma and anaplastic astrocytoma to designate grade 2 and 3 tumors have been discontinued, and a grade 4 designation has been added with the following resultant classification:
- Astrocytoma, IDH-mutant, CNS WHO grade 2
- Astrocytoma, IDH-mutant, CNS WHO grade 3
- Astrocytoma, IDH-mutant, CNS WHO grade 4
Designation of grade 2 and 3 tumors continues to be made by increased anaplasia and proliferative activity in grade 3 tumors, though a precise cutoff for mitotic activity is not provided, making the distinction somewhat subjective. Grade 4 designation is achieved by any of the following features: tumor necrosis, microvascular proliferation, or homozygous loss of CDKN2A and/or CDKN2B. Thus, even in the setting of a low-grade appearing IDH-mutant astrocytoma without significant mitotic activity, homozygous deletion of CDKN2A and/or CDKN2B would result in a grade 4 designation. Testing for CDKN2A/B status is therefore necessary on all IDH-mutant astrocytomas.
The grading and nomenclature for oligodendroglioma, IDH-mutant, 1p/19q-codeleted is largely unchanged, though anaplastic is no longer recommended to denote a CNS WHO grade 3 tumor. While the 2016 WHO classification specified at least 6 mitotic figures per 10 high-power fields (³ 2.5 mitotic figures per mm2) for grade 3 designation, the 2021 WHO classification has removed this hard cutoff, given that literature does not support a clear distinction by either mitotic count or Ki67 index, though soft parameters are offered. While CDKN2A/B status is not formally incorporated into the grading criteria, CDKN2A/B homozygous deletion has been reported in a small subset of grade 3 oligodendrogliomas, and not in grade 2, where it was associated with poor outcomes.6
Figure 1 and Figure 2 demonstrate the 2016 to 2021 changes in nomenclature and grading of adult infiltrating gliomas.
Given the changes discussed above, initial diagnostic workup of an adult hemispheric infiltrating glioma might start with immunohistochemistry for IDH1 p.R132H (accounting for ~ 90% of IDH1/IDH2 mutations in this setting), ATRX, p53, and Ki67 (particularly in the setting of lower-grade histology).7 If IDH1 p.R132H is negative for mutant protein expression, reflex sequencing of both IDH1 and IDH2 should be performed in patients < 55 years of age, as well as considered in older patients with lower grade histology. In IDH-wildtype tumors with lower grade histology, additional testing to assess EGFR amplification (eg, FISH, NGS, chromosomal microarray), Ch7 gain + Ch10 loss (eg, FISH, chromosomal microarray), and TERT promoter mutations (sequencing techniques) should be employed. In the setting of an IDH-mutant tumor, loss of ATRX expression is diagnostic of an IDH-mutant astrocytoma, and CDKN2A/B status should be assessed for grading (eg, FISH, chromosomal microarray). In the setting of retained or equivocal ATRX expression, testing for 1p/19q status is indicated (eg, FISH, chromosomal microarray). As the performance of ATRX immunohistochemistry is known to be technically challenging, some groups go straight to 1p/19q status for stratification of IDH-mutant diffuse gliomas (see Figure 2). Though rare, cases of true dual molecular oligodendroglioma and astrocytoma differentiation are reported, with separate components of morphologic and molecular IDH-mutant astrocytoma (ATRX loss, 1p/19q-intact) and IDH-mutant oligodendroglioma (ATRX intact, 1p/19q-codeleted).8 The diagnosis would require a high degree of suspicion and separate testing of each component, involving good communication with the molecular/FISH diagnostician. TERT promoter mutation would also be expected in the oligodendroglial component but not the astrocytic component.
In the setting of midline infiltrating gliomas, additional testing for H3F3A (H3.3 isoform) (most common), HIST1H3B/C (H3.1 isoform), and HIST2H3A/C (H3.2 isoform) (very rare) H3 histone mutations should be performed to exclude/diagnose diffuse midline glioma, H3 K27-altered, CNS WHO grade 4. Initial assessment by immunohistochemistry to evaluate for concurrent loss of H3 p.K27me3 expression and H3 p.K27M mutant protein expression (covering canonical H3.1-3 K27M mutations) can identify the vast majority of cases.9,10 In the setting of an H3-wildtype diffuse midline glioma, aberrant EZHIP overexpression also results in the K27-altered phenotype. Rarely, H3 p.K27I mutations are also reported, which would require sequencing for identification. Additionally, a new subtype characterized by EGFR alterations is described. Importantly, loss of H3 p.K27me3 expression would be expected in all subtypes. Of note, in NGS reporting, the H3 p.K27M mutation is reported as p.K28M due to differences in amino acid numbering (the initiating methionine is counted in HGVS nomenclature, whereas histone amino acids are often described with the protein sequences lacking this initial amino acid). The p.K27M legacy nomenclature is maintained in the 2021 WHO classification to uphold consistency with prior nomenclature.
Figure 1: 2016 CNS WHO adult infiltrating gliomas.
Figure 2: 2021 CNS WHO adult infiltrating gliomas.
Pediatric infiltrating (diffuse) gliomas are stratified into low-grade and high-grade categories, which are largely diagnosed by their molecular drivers, often requiring ancillary testing (eg, NGS panel including fusion detection, directed FISH, and/or tumor methylation profiling).
Pediatric-type diffuse low-grade gliomas include the following:
- Diffuse astrocytoma, MYB- or MYBL1-altered, CNS WHO grade 1
- Angiocentric glioma, CNS WHO grade 1
- Polymorphous low-grade neuroepithelial tumor of the young (PLNTY), CNS WHO grade 1
- Diffuse low-grade glioma, MAPK pathway-altered
Diffuse astrocytoma, MYB- or MYBL1-altered, CNS WHO grade 1 tumors are rare seizure-associated gliomas that harbor structural rearrangements resulting in MYB or MYBL1 fusion products, often with PCDHGA1, MMP16, and MAML2. More rarely, MYB::QKI fusion is seen, though this is more classically associated with and should raise the possibility of angiocentric glioma, CNS WHO grade 1. PLNTY, another seizure-associated entity, typically occurs in younger individuals, and often shows infiltrative growth pattern with oligodendroglioma-like areas, prominent CD34 reactivity, and MAPK pathway alterations, most commonly involving BRAF, FGFR2, and FGFR3.2 Finally, diffuse low-grade glioma, MAPK pathway-altered is a heterogeneous group of low-grade infiltrating gliomas that are IDH-wildtype, H3-wildtype, and CDKN2A-intact and most often harbor BRAF V600E or alterations in FGFR1 (point mutations, internal tandem duplications of the tyrosine kinase domain, or fusions).2
Pediatric-type diffuse high-grade gliomas include the following:
- Diffuse midline glioma, H3 K27-altered, CNS WHO grade 4
- Diffuse hemispheric glioma, H3 G34-mutant, CNS WHO grade 4
- Diffuse pediatric-type high-grade glioma, H3-wildtype and IDH-wildtype, CNS WHO grade 4
- Infant-type hemispheric glioma
In the case of diffuse pediatric-type high-grade glioma, H3- and IDH-wildtype, CNS WHO grade 4, three main subtypes are stratified by methylation profiling: RTK1 (enriched for PDGFRA alterations or in association with Lynch syndrome or constitutional mismatch repair deficiency syndrome), RTK2 (enriched for EGFR amplification and TERT promoter mutation), and MYCN (often harboring MYCN amplification). Infant-type hemispheric glioma usually occurs in the first year of life and harbors typical receptor tyrosine kinase fusions (commonly involving NTRK1, NTRK2, NTRK3, ALK, ROS1, or MET) making them potentially sensitive to targeted therapies. While they typically demonstrate high-grade features, formal grading has not been established as outcome data are lacking.2
Within the category of circumscribed astrocytic gliomas, a new entity of interest is high-grade astrocytoma with piloid features (HGAP), for which a definitive diagnosis is achieved by methylation profiling. This entity occurs over a broad age range and anatomic CNS distribution. It often occurs de novo, but has also been reported to progress from a prior lower-grade glioma, often pilocytic astrocytoma. The combination of a MAPK pathway alteration, ATRX mutation and/or loss of ATRX expression, and homozygous deletion of CDKN2A/B in the correct morphologic setting is particularly suggestive of this diagnosis.11 Formal grading has not been established, but clinical behavior roughly corresponding to CNS WHO grade 3 is suggested.2
Additional proposed gliomas supported by tumor methylation profiling and presented in the literature since the publication of WHO CNS 2021 classification include high-grade glioma with pleomorphic and pseudopapillary features (HPAP), neuroepithelial tumor with PATZ1 fusion (considered most compatible with a glioma by some), and glial tumor with BCOR fusion (not to be confused with CNS tumor with BCOR internal tandem duplication (ITD), an embryonal tumor of childhood). HPAP, as its name implies, is described as having generally high-grade but widely varied histologic features, often with pseudopapillary features and a predominately non-infiltrative growth pattern. It typically occurs in young adults, has a distinct methylation signature, and demonstrates near-universal monosomy 13, as well as recurrent alterations in TP53, NF1, NF2, RB1, and BRAF and lack of CDKN2A/B deletion.12 In the initial description of neuroepithelial tumor with PATZ1 fusion, the entity occurred in children, demonstrated varied glial or glioneuronal morphology, and harbored recurrent PATZ1 fusion with either MN1 or EWSR1 gene partners.13 Gliomas with BCOR fusions (most commonly EP300::BCOR; rarely CREBBP::BCOR, MEAF6::CXXC5, and BCOR stop mutations) have been described over a broad age range, as opposed to BCOR-ITD tumors occurring in young children, and with diverse histologic appearances.14
Briefly, changes in ependymoma classification include stratification by anatomic location (supratentorial, posterior fossa, and spinal) and further incorporation of molecular findings. The molecularly defined ependymomas include supratentorial ependymoma with either ZFTA (previously referred to as C11orf95) or YAP1 fusion, posterior fossa group A ependymoma with EZHIP or H3 K27M mutations and frequent chromosome 1q gain, posterior fossa group B ependymoma with chromosomal instability, and spinal ependymoma with MYCN amplification. Immunohistochemistry for H3 p.K27me3 expression is a useful adjunct for identifying posterior fossa group A ependymoma, as H3 K27M histone and EZHIP mutations or overexpression both result in reduced trimethylation expression at this site.15 Of note, myxopapillary ependymoma (previously designated grade 1) is now considered a CNS WHO grade 2 entity given is propensity for recurrence and morbidity.2
Minor changes to the meningioma classification are implemented in the 2021 classification. While the mitotic activity corresponding to grade 2 and 3 designation is technically unchanged, greater emphasis on precise quantification is specified. Given the variance in high-power field surface area between microscopes, reporting of mitotic figures per mm2 aids in standardized reporting and grading, with ≥ 2.5/mm2 and ≥ 12.5/mm2 corresponding to grade 2 and 3, respectively. Additional independent criteria for grade 3 designation include the identification of a TERT promoter mutation or CDKN2A and/or CDKN2B homozygous deletion. Regarding surrogate immunohistochemistry, a recent study demonstrated good correlation with p16 loss of expression in histologically higher grade meningiomas, though poor correlation in histologically grade 1 meningiomas. As such, p16 immunohistochemistry was not recommended for CDKN2A/B status screening in histologic grade 1 tumors.16
Lastly, the WHO CNS 2021 classification abandons the use of Roman numerals (ie, I, II, III, or IV) to designate tumor grade and advocates the exclusive use of Arabic numerals (ie, 1, 2, 3 or 4). This change was made to reduce the risk of typographical or reading errors in the reporting, made more important given the diffuse and anaplastic designations are no longer included in the nomenclature for IDH-mutant gliomas, and anaplastic terminology has likewise been removed from other glioma entities (eg, IDH-wildtype gliomas, pleomorphic xanthoastrocytoma (for CNS WHO grade 3), and ependymoma (for CNS WHO grade 3).
The described changes and additions to the new WHO CNS 2021 classification highlight the necessity of advanced diagnostic molecular testing in most tumor types, often requiring a range of techniques to identify the multitude of possible genetic aberrations, including point mutations, structural variants, copy number changes, deletions and duplications, protein expression abnormalities, chromosomal aberrations, and epigenetic changes, among others. Particularly in the case of pediatric CNS tumor diagnosis, some institutions employ multiple techniques in tandem at initial evaluation, including immunohistochemistry, large DNA/RNA next-generation sequencing panels with fusion analysis, chromosomal microarray, and tumor methylation profiling, to arrive at an accurate and timely diagnosis. As described above, many tumor types have overlapping histologic phenotypes and overlapping mutation profiles. This makes diagnosis of many entities challenging even with robust data at hand and necessitates the use of tumor methylation profiling, which currently has limited but expanding clinical availability.
- WHO Classification of Tumours Editorial Board. Central nervous system tumours. WHO classification of tumours series, 4th ed. International Agency for Research on Cancer; 2016.
- WHO Classification of Tumours Editorial Board. Central nervous system tumours. WHO classification of tumours series, 5th ed., vol. 6. International Agency for Research on Cancer; 2021. https://publications.iarc.fr/601
- Aoki K, Nakamura H, Suzuki H, et al. Prognostic relevance of genetic alterations in diffuse lower-grade gliomas. Neuro Oncol. 2018;20(1):66-77.
- Stichel D, Ebrahimi A, Reuss D, et al. Distribution of EGFR amplification, combined chromosome 7 gain and chromosome 10 loss, and TERT promoter mutation in brain tumors and their potential for the reclassification of IDHwt astrocytoma to glioblastoma. Acta Neuropathol. 2018;136(5):793-803.
- Wijnenga MMJ, Dubbink HJ, French PJ, et al. Molecular and clinical heterogeneity of adult diffuse low-grade IDH wild-type gliomas: assessment of TERT promoter mutation and chromosome 7 and 10 copy number status allows superior prognostic stratification. Acta Neuropathol. 2017;134(6):957-959.
- Appay R, Dehais C, Maurage CA, et al. CDKN2A homozygous deletion is a strong adverse prognosis factor in diffuse malignant IDH-mutant gliomas. Neuro Oncol. 2019;21(12):1519-1528.
- Brat DJ, Aldape K, Bridge JA, et al. Molecular biomarker testing for the diagnosis of diffuse gliomas. Arch Pathol Lab Med. 2022;146(5):547-574.
- Mizuno R, Homma T, Adachi JI, et al. True anaplastic oligoastrocytoma with dual genotype: illustrative case. J Neurosurg Case Lessons. 2022;4(3):CASE22146.
- Castel D, Philippe C, Kergrohen T, et al. Transcriptomic and epigenetic profiling of ‘diffuse midline gliomas, H3 K27M-mutant’ discriminate two subgroups based on the type of histone H3 mutated and not supratentorial or infratentorial location. Acta Neuropathol Commun. 2018;6(1):117.
- Huang T, Garcia R, Qi J, et al. Correction: Detection of histone H3 K27M mutation and post-translational modifications in pediatric diffuse midline glioma via tissue immunohistochemistry informs diagnosis and clinical outcomes. Oncotarget. 2019;10(28):2788.
- Reinhardt A, Stichel D, Schrimpf D, et al. Anaplastic astrocytoma with piloid features, a novel molecular class of IDH wildtype glioma with recurrent MAPK pathway, CDKN2A/B and ATRX alterations. Acta Neuropathol. 2018;136(2):273-291.
- Pratt D, Abdullaev Z, Papanicolau-Sengos A, et al. High-grade glioma with pleomorphic and pseudopapillary features (Hpap): a proposed type of circumscribed glioma in adults harboring frequent TP53 mutations and recurrent monosomy 13. Acta Neuropathol. 2022;143(3):403-414.
- Alhalabi KT, Stichel D, Sievers P, et al. PATZ1 fusions define a novel molecularly distinct neuroepithelial tumor entity with a broad histological spectrum. Acta Neuropathol. 2021;142(5):841-857.
- Wu Z, Rajan S, Chung HJ, et al. Molecular and clinicopathologic characteristics of gliomas with EP300:BCOR fusions. Acta Neuropathol. 2022;144(6):1175-1178.
- Jain SU, Do TJ, Lund PJ, et al. PFA ependymoma-associated protein EZHIP inhibits PRC2 activity through a H3 K27M-like mechanism. Nat Commun. 2019;10(1):2146.
- Tang V, Lu R, Mirchia K, et al. Loss of p16 expression is a sensitive marker of CDKN2A homozygous deletion in malignant meningiomas. Acta Neuropathol. 2023;145(4):497-500.
Sahara Cathcart, MD is an assistant professor at the University of Nebraska Medical Center in Omaha, Nebraska. She is board certified in Anatomic-Clinical Pathology, Neuropathology, and Molecular-Genetic Pathology. She serves as associate director of the pathology residency program and is heavily involved in resident, fellow, and med student education. Her research interests include translational study brain tumors and neurodegenerative disease.
Allison Cushman-Vokoun, MD, PhD is the medical director of both the Molecular Diagnostics Laboratory and the Warren G. Sanger Human Genetics Laboratory at the University of Nebraska Medical Center/Nebraska Medicine in Omaha, Nebraska. She is one of the leaders for the Precision Medicine Initiative at Nebraska Medicine, and she started the Molecular Genetic Pathology Fellowship Program at UNMC. She is a full-time molecular pathologist with research interests in the translational evaluation of molecular biomarkers in various malignancies. She has happily served on two terms of the CAP Personalized Health Care Committee.