LGCmain
Indian Journal of Pathology and Microbiology
Home About us Instructions Submission Subscribe Advertise Contact e-Alerts Ahead Of Print Login 
Users Online: 1361
Print this page  Email this page Bookmark this page Small font sizeDefault font sizeIncrease font size


 
  Table of Contents    
REVIEW ARTICLE  
Year : 2022  |  Volume : 65  |  Issue : 5  |  Page : 99-110
Evaluating neurosurgical biopsies for CNS tumor diagnoses: An algorithmic and pattern based approach


Department of Laboratory Medicine and Pathology, Mayo Clinic, Rochester, MN, USA

Click here for correspondence address and email

Date of Submission05-Nov-2021
Date of Decision09-Dec-2021
Date of Acceptance22-Dec-2021
Date of Web Publication11-May-2022
 

   Abstract 


The 2016 and 2021 World Health Organization (WHO) Classifications of Tumors of the Central Nervous System (CNS) reflect the importance of integrating molecular analysis into CNS tumor diagnosis and classification, adding to the complexity of any surgical neuropathology practice. On the other hand, our evolving understanding of genomic alterations across the spectrum of CNS tumors highlights the importance of utilizing traditional histological and immunohistochemical approaches to first establish as accurate a diagnosis as possible. Such an approach is also essential to recognizing the most appropriate ancillary test(s) needed for accurate classification and grading of CNS tumors. Here, we present an algorithmic approach to be considered while evaluating surgical neuropathology biopsies, which includes a recognition of main histological patterns, and incorporates clinical and radiologic features, to assist with accurate diagnosis and optimal selection of subsequent ancillary testing.

Keywords: Algorithm, CNS tumors, pattern-based

How to cite this article:
Vizcaino M A, Raghunathan A. Evaluating neurosurgical biopsies for CNS tumor diagnoses: An algorithmic and pattern based approach. Indian J Pathol Microbiol 2022;65, Suppl S1:99-110

How to cite this URL:
Vizcaino M A, Raghunathan A. Evaluating neurosurgical biopsies for CNS tumor diagnoses: An algorithmic and pattern based approach. Indian J Pathol Microbiol [serial online] 2022 [cited 2022 May 24];65, Suppl S1:99-110. Available from: https://www.ijpmonline.org/text.asp?2022/65/5/99/345038





   Introduction Top


The 2021 World Health Organization (WHO) Classification of Tumors of the Central Nervous System (CNS) is the fifth edition of the international standard for the classification of brain and spinal cord tumors. The changes made in this and the prior 2016 edition reflect the growing significance of molecular diagnostics in CNS tumor classification[1],[2] which adds to the challenge of diagnosing these neoplasms. At the same time, however, it also reflects the importance of an accurate histological evaluation as a pivotal first step in the classification and grading of these tumors. Recognizing main histological patterns, and correlation with the clinical and radiological features, helps guide selection of appropriate additional ancillary testing (immunohistochemical and/or molecular, as and when needed), and formulating an accurate and clinically-relevant integrated diagnosis. Here, we propose a practical, step-by-step approach, summarizing major features to be considered when evaluating neurosurgical specimens.


   Clinical and radiological features Top


Clinical and radiological information provides valuable clues for accurate interpretation of the findings identified in neurosurgical specimens. Since specimens in surgical neuropathology are almost invariably limited, these act as surrogates for the “macroscopic/gross” evaluation performed on other systemic surgical pathology specimens. Common features essential to narrowing the differential diagnostic considerations include the patient's age [pediatric vs adult, [Table 1]],[3] tumor location (intra- vs extra-axial, supra- vs infra-tentorial, sellar, intra-ventricular, etc.), and radiological features [Table 2]. Correlation with the imaging appearance is also essential in determining whether the most representative regions of the tumor have been appropriately represented in limited biopsies.
Table 1: Common CNS tumors by age group+

Click here to view
Table 2: Relevant CNS tumors features based on location, number, and imaging

Click here to view



   Histological features Top


A systematic approach to assess for the morphologic features in CNS tumors includes first establishing the presence of a neoplastic process, followed by evaluation of architectural and growth patterns and an assessment of the cytological features of the various cellular constituents. Identification of additional morphological findings may provide further clues to refine diagnostic considerations and indicate appropriate ancillary tests. Finally, the clinical and radiological features need to be considered. Regardless of whether review of the case starts with a morphology-first approach or a clinical-radiological review, it is important to integrate all the information available when making a final diagnosis, since all these features should generally be mutually consistent. [Table 3] provides a summary of some common CNS tumors based on location and/or radiologic features. Such an algorithmic approach, when consistently applied, would assist in formulating an accurate diagnosis while avoiding common pitfalls.[4] We propose approaching this as a series of questions to be answered through the course of histological evaluation.
Table 3: Common CNS tumors based on location and/or radiologic features

Click here to view


Is the biopsy representing abnormal tissue?

Familiarity with regional variations of the cellular constituents of different CNS regions is essential to first determine whether the targeted lesion has indeed been sampled. For instance, in a limited sample, the normal internal granular cell layer of the cerebellum or normal pineal gland may be misinterpreted as representing an embryonal neoplasm. Similarly, familiarity with the normal neurohypophysis or with normocellular cerebral, cerebellar, or spinal white matter may prevent misinterpretation as a glial neoplasm.

Is the abnormal tissue representing a non-neoplastic or a neoplastic process?

Abnormal tissue, even if representative of a mass lesion, does not necessarily indicate a neoplastic process. The misinterpretation of non-neoplastic changes may result in devastating therapeutic consequences for the patient. Accurately identifying inflammatory infiltrates, particularly macrophages, may help subsequently place nuclear atypia of glial cells into perspective. Similarly, identification of infectious micro-organisms, including parasites and viral cytopathic changes, or of abundant macrophages as seen in a demyelinating process or trauma, would lead to dramatically different therapeutic decisions.

If neoplastic, is this a primary CNS neoplasm or could this represent metastasis?

Metastases constitute the most common CNS neoplasms in adults, and the spectrum of their clinical and radiological presentations is highly variable, including as the primary presentation of an as-yet undiagnosed systemic malignancy. On the other hand, divergent differentiation may be seen in various, typically high-grade, primary CNS neoplasia (e.g. glioblastoma with epithelial differentiation, gliosarcoma, anaplastic meningioma etc.). Even if not a primary consideration upon initial histological review, the possibility of a metastasis should remain among the differentials, particularly in the presence of unusual morphological and/or immunohistochemical findings in a seemingly high-grade neoplasm and can often be readily excluded by appropriate immunohistochemical evaluation.

What is the architectural pattern?

The main architectural pattern of a CNS tumor provides valuable information regarding lineage and biological behavior. This helps not only to determine the main diagnostic category to be considered, but also to determine whether ancillary testing is needed and, if so, guide step-by-step selection of the most appropriate tests whose results would best help determine treatment and follow-up for these patients. Some of the most common architectural patterns seen in CNS tumors are summarized below and are illustrated in [Figure 1].
Figure 1: Common architectural patterns in CNS neoplasms. Circumscribed neoplasm with expansile borders and minimal to absent parenchymal infiltration (a), demonstrated by neurofilament IHC (b, *). Infiltrating glioma with parenchymal invasion (c), highlighted by neurofilament stain (d). Biphasic pattern in a case of pilocytic astrocytoma (e) with loose, myxoid areas (*) alternating with dense, compact regions (**). Nodular pattern consisting of well-defined, variably sized clusters of tumor cells (f). Pseudorosettes in a case of ependymoma, in which tumor cells show fine cytoplasmic processes oriented towards small caliber blood vessel walls (g). Palisading necrosis with peripheral aggregation of tumor cells around a central area formed by apoptotic cells and coagulative necrosis (h)

Click here to view


Solid (circumscribed)

There is minimal or absent infiltration of the neoplastic cells into the adjacent brain parenchyma, imparting a circumscribed or demarcated appearance, with a paucity of axons within the tumor demonstrated by a neurofilament immunostain [Figure 1]a, [Figure 1]b. Tumors displaying this pattern include metastases, circumscribed gliomas including those of ependymal (ependymoma, subependymoma) and astrocytic (pilocytic astrocytoma, subependymal giant cell astrocytoma (SEGA), pleomorphic xanthoastrocytoma) lineages, glioneuronal and neuronal tumors, embryonal tumors, and extra-axial tumors (such as choroid plexus papilloma, meningioma, and hemangioblastoma).

Infiltrating

The tumor cells show infiltration into the gray and/or white matter, with neurofilament demonstrating abundant preserved axons within cellular regions [Figure 1]c, [Figure 1]d. Examples of tumors predominantly showing this pattern include diffuse gliomas (glioblastoma, diffuse astrocytoma, oligodendroglioma etc.), as well as primary CNS diffuse large B-cell lymphoma (DLBCL). Non-neoplastic processes may also appear minimally to extensively infiltrative, as can be seen in inflammatory/infectious disorders, demyelinating diseases, histiocytic disorders, and vascular/metabolic/toxic insults resulting in reactive gliosis.

Mixed solid and infiltrating

It is important to recognize that several otherwise circumscribed tumors can show ill-defined borders, with single cells and small clusters apparently infiltrating adjacent brain parenchyma. Likewise, infiltrative tumors can harbor regions of solid, circumscribed-appearing growth. These findings can seem particularly exaggerated in small biopsies. Pilocytic astrocytoma, pleomorphic xanthoastrocytoma (PXA), glioblastoma/gliosarcoma, glioneuronal tumors, embryonal neoplasms, choroid plexus carcinoma, supratentorial ependymoma, germ cell tumors, primary CNS DLBCL are all examples of tumors which can show admixed regions of circumscribed and infiltrative growth.

Biphasic

Some neoplasms characteristically harbor two distinct architectural patterns or cellular elements [Figure 1]e. Pilocytic astrocytoma is probably the most recognized glial tumor to show a biphasic (compact and loose) tissue pattern. However, PXA, ganglioglioma, dysembryoplastic neuroepithelial tumor (DNT), meningioma, solitary fibrous tumor, and schwannoma are other tumors that may display biphasic architectural patterns. Even glioblastoma may harbor two or more distinct architectural patterns, as can be seen in the gliosarcoma subtype, or in the presence of regions showing divergent differentiation (such as epithelial, rhabdoid, primitive neuronal, etc.), corresponding to its erstwhile “multiforme” designation.

Lobulated, nested, or nodular

Neoplasms with lobulated or nested architecture are characterized by variably sized clusters of neoplastic cells separated by a relatively uninvolved background tissue [Figure 1]f. These may be surrounded by thin strands of fibrovascular tissue, as can be seen in metastases, pituitary adenoma, meningioma, and paraganglioma. Alternatively, the somewhat well-defined rounded aggregates of tumor cells may be present within normocellular fibrillary parenchyma. Examples of primary CNS tumors that may show variable amounts of such a pattern include oligodendroglioma, DNT, ganglioglioma, subependymoma, extraventricular neurocytoma, certain medulloblastoma subtypes (desmoplastic/nodular), pineal parenchymal tumors, germ cell tumors, and the recently codified multinodular and vacuolating neuronal tumor.[5]

Rosettes and pseudorosettes

Rosettes and pseudorosettes are formed by arrangements of neoplastic cells oriented around central non-cellular regions. These can be seen in a wide variety of CNS neoplasms, and some tumors may show a predominance of one type of rosette/pseudorosette over the other. For example, conventional ependymoma most frequently shows prominent perivascular pseudorosettes [Figure 1]g, while true ependymal rosettes are only seen in a minority of cases. Other tumors displaying prominent pseudorosette-like arrangements include astroblastoma, pituitary adenoma, paraganglioma, and metastasis of neuroendocrine origin. In contrast, true rosettes are usually more prominent in neoplasms with some extent of neuronal differentiation, such as central/extraventricular neurocytoma, rosette-forming glioneuronal tumor, pineocytoma, medulloblastoma, and embryonal tumor with multilayered rosettes. Examples of different types of rosettes include:

  • Homer Wright rosettes: tumor cells surround a delicate fibrillary central core. Tumors in which these may be seen include a spectrum of CNS embryonal tumors, neuroblastoma, medulloblastoma and pineoblastoma
  • Flexner-Wintersteiner rosettes: tumor cells showing fine cytoplasmic processes cluster around an empty or nearly empty central lumen. Tumors in which these may be seen include retinoblastoma, pineoblastoma, and medulloepithelioma
  • Ependymal rosettes/canals: tumor cells aggregate to form structures resembling small and large tubular lumina, which lack central neuropil-like material. Although present only in a minority of ependymomas, these are characteristic of ependymal differentiation.
  • Pineocytomatous and neurocytic rosettes: These are similar to the Homer-Wright rosettes in having central finely fibrillary cores; however, these are generally larger and have irregular contours. As the name suggests, these are typically seen in pineocytoma and central neurocytoma.


Papillary

As seen across the spectrum of surgical pathology specimens, tumor cells may surround thin fibrovascular cores, forming “finger-like” papillary projections. In addition to metastases, true papillary structures are prominent among certain primary CNS tumors, including choroid plexus tumors, myxopapillary ependymoma, astroblastoma, papillary glioneuronal tumor, papillary tumor of the pineal region, papillary craniopharyngioma, as well as in papillary subtypes of meningioma and ependymoma. A note of caution is due when identifying this pattern, as it may be mimicked by perivascular preservation of cells in an extensively necrotic tumor, imparting what would more accurately be described as a “pseudo-papillary” pattern. Such pseudo-papillae may be seen in metastases, particularly melanoma, as well as in any high-grade neoplasm such as glioblastoma, medulloblastoma, atypical teratoid/rhabdoid tumor (AT/RT), ependymoma, and meningioma.

Secondary structures of Scherer

Diffuse glioma cells migrating within brain parenchyma are relatively unable to penetrate glia limitans and override existing cellular constituents, resulting in the formation of “secondary” structures. Initially described by Hans Scherer in the 1930s, this finding reflects the keystone interpretation of glioma morphology in the context of a dynamic process. Secondary structures include aggregation of glioma cells in the subpial and periventricular regions, surrounding neuronal bodies (perineuronal satellitosis) and blood vessels (perivascular satellitosis), and within white matter tracts.[6] Identification of secondary structures is a key morphological clue to diagnose diffuse gliomas.

Palisading necrosis

Palisading or “pseudopalisading” necrosis, recognized for almost a century as predictive of a more aggressive clinical behavior, is a morphologic finding relatively unique to high-grade gliomas and is one of the defining features of glioblastoma [Figure 1]h. It has been shown that tumor cell palisading is an evolving phenomenon resulting from severe hypoxia secondary to vascular occlusion and intravascular thrombosis and is intimately associated with microvascular proliferation. Palisades represent migration of tumor cells away from central hypoxic regions, and are characterized by tumor cell nuclei arranged as parallel aggregates around a central pale area formed by apoptotic cells, and ultimately, coagulative necrosis.[7]

What are the cellular constituents of this neoplasm?

Fibrillary cells with irregular nuclei

The presence of cells showing fibrillary cytoplasmic processes and hyperchromatic, elongated nuclei with irregular contours typically indicates astroglial differentiation [Figure 2]a, [Figure 2]b. The quantity of fibrillary cytoplasmic processes and extent nuclear pleomorphism may vary considerably from tumor to tumor, or even within the same tumor. Apart from astrocytomas, tumors which may show prominent gliofibrillary processes but lack significant nuclear irregularity are ependymal neoplasms and pituicytoma.
Figure 2: Common cytomorphologies and additional findings in CNS neoplasms. Astrocytic morphology in an infiltrating glioma with fibrillary background and cells showing enlarged, irregular, hyperchromatic nuclei (a). In contrast, astrocytic neoplasms with piloid morphology show bland, elongated nuclei and fine bipolar cytoplasmic processes (b). Oligodendroglial and oligodendroglia-like cells show monomorphic, round, hyperchromatic nuclei surrounded by a clear artifactual halo (“fried egg” appearance) (c). Mixed glial-glioneuronal morphology shows two distinct populations of usually astrocytic and ganglion cell morphology (d). Dysmorphic features and binucleation are commonly seen in the ganglion cell component (D, arrow). Epithelioid cells show well demarcated borders and are round to polygonal with moderate to abundant cytoplasm (e). Primitive cells show scant cytoplasm and hyperchromatic, small to medium sized nuclei (f). Histiocyte-rich lesion, with numerous macrophages showing abundant vacuolated cytoplasm (g). Inflammation-rich lesion, consisting predominantly of small lymphocytes without atypia (h)

Click here to view


Oligodendroglial and oligodendroglia-like

Oligodendroglial and oligodendroglia-like cells are very monomorphic, characterized by round to oval nuclei of similar size and having a similar chromatin pattern, often with inconspicuous nucleoli [Figure 2]c. These cells may show clear perinuclear “halos” imparting the so-called “fried egg” appearance, which is an artifact secondary to cytoplasmic retraction during tissue processing, and often not as evident in intraoperative cytological preparations. Besides oligodendroglioma, other CNS neoplasms prominently showing this cytomorphology include DNT, pilocytic astrocytoma, ganglioglioma, central/extraventricular neurocytoma, rosette-forming glioneuronal tumor, pineocytoma, small cell glioblastoma, ependymoma (clear cell subtype), and paraganglioma. Two recently recognized entities that should also be considered in the differential diagnosis of oligodendroglial-like neoplasms include polymorphous low-grade neuroepithelial tumor of the young (PLNTY) and diffuse leptomeningeal glioneuronal tumor (DLGNT). PLNTY is a CNS WHO grade 1 entity that usually presents in young patients and has a strong association with seizures. Histologically, these show a diffuse growth pattern and at least a component with oligodendroglioma-like morphology, and often harbors calcifications. In contrast with oligodendroglioma, PLNTY demonstrates CD34 immunoreactivity and MAPK pathway activating alterations. DLGNT is a glioneuronal neoplasm typically showing diffuse leptomeningeal involvement without an identifiable intra-axial mass and frequently shows oligodendroglial-like cells. Molecularly, DLGNT is characterized by chromosome 1p deletion and MAPK pathway alterations, most commonly KIAA1549-BRAF fusion. Lastly, caution must also be exercised to distinguish neoplastic cells with oligodendroglia-like features from aggregates of macrophages, which may show somewhat uniform nuclei and whose cytoplasm may superficially mimic perinuclear clearing.

Mixed glial and neuronal cells

A spectrum of typically low-grade neoplasms are characterized by the presence of two distinct cell populations - one showing small to medium-sized cells with fibrillary cytoplasmic processes (i.e. glial morphology), and another formed by larger cells with amphophilic cytoplasm, often with recognizable Nissl substance, and large nuclei with open chromatin and prominent nucleoli (i.e. neuronal/ganglion cell morphology). The identification of cytological dysmorphism of neuronal cells, such as binucleated or even multinucleated forms [[Figure 2]d, arrow], is essential to distinguishing a neoplastic neuronal component from entrapped neurons within an infiltrating glioma. Alternatively, the neuronal population may show uniform round nuclei with stippled (“salt-and-pepper”) chromatin. The spectrum of neuronal and glioneuronal tumors show distinctive architectural arrangements of the glial and neuronal components, depending upon the entity (see above), such as DNT, rosette-forming glioneuronal tumor, papillary glioneuronal tumor, and ganglioglioma. Alternatively, embryonal neoplasms may show variable extents of neuronal differentiation.

Epithelioid

Epithelioid cells are characterized by cohesive groups showing moderate to abundant eosinophilic cytoplasm, demarcated cytoplasmic borders, and lacking cytoplasmic processes [Figure 2]. As the name implies, epithelioid morphology can most frequently be seen in metastatic carcinoma or melanoma. However, it is important to also be aware of primary CNS neoplasms which can show this cytomorphology, including meningioma, primary CNS melanocytic tumors, pituitary adenoma, epithelioid glioblastoma, pleomorphic xanthoastrocytoma, choroid plexus carcinoma, paraganglioma, germ cell tumors, craniopharyngioma, atypical teratoid/rhabdoid tumor (AT/RT), and chordoma.

Primitive (“Small blue cell tumor”)

There is a broad group of neoplasms composed of cells with high nucleus-to-cytoplasmic ratios, scant cytoplasm, and hyperchromatic, irregular to uniform nuclei [Figure 2]f. Metastasis from systemic tumors (such as small cell carcinoma) remains the most important diagnostic consideration among adults. However, even among pediatric patients, metastasis from neuroblastoma, rhabdomyosarcoma, non-Hodgkin lymphoma, and Ewing sarcoma remain important diagnostic considerations.[8] Primary CNS tumors composed predominantly of primitive cells include all variants of CNS embryonal neoplasms (medulloblastoma, AT/RT, pineoblastoma, and other CNS embryonal tumors). However, primitive cells may constitute a significant component of any high-grade primary CNS tumor, such as glioblastoma (with primitive neuronal component), ependymoma, and any of the pediatric-type high-grade diffuse gliomas (H3 K27-altered, H3 G34-mutant, etc.).

Histiocyte-rich

Macrophages are normally not present in brain parenchyma, and their presence should raise consideration for resolving injury, such as a sub-acute infarct, or a demyelinating disease [Figure 2]g. Other conditions in which abundant macrophages can be identified include therapy-related changes (chemotherapy, radiotherapy), infectious processes, granulomatous diseases including sarcoidosis, and vasculitides. Variable numbers of macrophages can also be associated with necrotic regions of high-grade primary CNS tumors.

Spindle cells

Spindle cells are characterized by elongated thin nuclei and long cytoplasmic processes, often arranged as fascicles. The presence of spindle cells in the CNS raises the possibility of metastatic disease from sarcomas or carcinomas showing sarcomatoid features. However, primary CNS neoplasms can also show this morphology, including meningioma, schwannoma, solitary fibrous tumor, gliosarcoma, PXA, desmoplastic infantile ganglioglioma/astrocytoma (DIG/DIA), and pituicytoma.

Additional findings


   Rosenthal fibers Top


Rosenthal fibers (RFs) are protein aggregates within astrocytes, identified by light microscopy as bright, eosinophilic, thick, undulating, seemingly extra-cellular structures. Ultrastructurally, RFs appear as non-membrane bound, amorphous, electron dense material within the astrocytic processes, surrounded by intermediate filaments, predominantly glial fibrillary acidic protein (GFAP), as well as vimentin and alpha-B-crystalline, among others. RFs are classically identified in the compact regions of pilocytic astrocytoma. However, these can also be found in non-neoplastic conditions, particularly in the setting of long-standing gliosis (“piloid gliosis”).[9] This is particularly important to recognize when evaluating tissue obtained from slowly progressive lesions, since the presence of RFs may mistakenly lead to a diagnosis of pilocytic astrocytoma, when these may in fact represent piloid gliosis adjacent to a low-grade tumor such as hemangioblastoma or craniopharyngioma, or be seen in the setting of cortical dysplasias or tuberous sclerosis.

Eosinophilic granular bodies

Eosinophilic granular bodies (EGBs) are histologically recognized as brightly eosinophilic, round bodies of variable sizes, showing coarse to fine internal granularity. By immunohistochemistry, EGBs are variably positive for alpha B-crystallin, ubiquitin and GFAP. Ultrastructurally, EGBs consist of membrane-bound round structures ranging from 50 nm to 20 microns in diameter.[10] Although EGBs are more frequently seen in lower-grade gliomas, particularly pilocytic astrocytoma, ganglioglioma, and PXA, these may also be present in high-grade astrocytomas, and so are not considered diagnostic of a specific entity.

Dense inflammation

Brain biopsies for suspected encephalitic processes are essentially no longer performed due to the availability of less invasive diagnostic options, such as cerebrospinal fluid (CSF) polymerase chain reaction (PCR). Prominent perivascular lymphocytic cuffs are a non-specific finding in several types of CNS tumors, including ganglion cell tumors, PXA, gemistocytic astrocytoma, giant cell glioblastoma, and germ cell tumors. However, the presence of large numbers of inflammatory cells in a biopsy should prompt consideration for an infectious or immune-mediated process, as well as a detailed review to exclude the possibility of demyelination or vasculitic processes [Figure 2]h. In vasculitis, the infiltration of vessel walls by inflammatory cells should be associated with evidence of significant vessel wall injury, such as fibrinoid necrosis. The concurrent presence of microglial nodules and perivascular lymphocytic cuffs may indicate the necessity to evaluate for a viral infection.[11] Another relevant finding is the presence or absence of cytological atypia within the inflammatory infiltrate, or of abnormal proportions of the T- and B-cell populations, which can indicate an associated lymphoproliferative disorder. Finally, dense lymphocytic infiltrates may obscure small numbers of diagnostic cells in the background, such as in germinoma or histiocytic neoplasms.

Is any ancillary testing needed? If so, what would be the most efficient and appropriate?

Immunohistochemistry

Immunohistochemistry (IHC) is a well-established, robust, and widely utilized ancillary diagnostic tool in surgical neuropathology, particularly in tumors. Their use can be broadly classified in two main categories: determining cell lineage, and as a surrogate for diagnostic and/or prognostic genomic alterations.[12],[13] Some of the more frequently utilized antibodies, and the expected immunophenotypes in more common CNS tumors are summarized in [Table 4].
Table 4: Frequently used IHC in CNS neoplasms

Click here to view



   Molecular testing Top


Assessment of genomic alterations provides diagnostic, prognostic, and/or predictive information in patients with brain tumors, which is increasingly used to guide patient care. Utilizing the most appropriate test based on the clinical and histological diagnostic considerations is key to providing the most beneficial information to guide appropriate therapy. It is important to recognize that each technique has its strengths and limitations, depending on the alteration being evaluated for. Therefore, it is essential to understand the role each technique can fulfil in different brain tumors, and how to integrate the findings into the practice.

Chromosomal analysis

Fluorescence in situ hybridization (FISH)

This technique involves hybridization of fluorescent-labelled specific DNA sequence probes with the tumor's DNA. As the probes can be easily created and evaluated with short turnaround times, FISH has been widely used to detect deletions, insertions, translocation breakpoints, and copy number alterations in many different diseases. Given its limited resolution, FISH has been largely replaced by chromosomal microarray analysis (CMA) when assessing for chromosomal abnormalities in CNS tumors. However, this remains a valuable tool when CMA is not available or when the amount of tissue is limited.

1p/19q co-deletion

Concurrent deletions of short arm of chromosome 1 (1p) and the long arm of chromosome 19 (19q) in a diffuse glioma harboring an IDH mutation is the defining criteria in oligodendroglioma and has long been evaluated for by FISH. Nonetheless, it is important to be aware that FISH only visualizes the region of the genome complementary to the probe used. Consequently, this technique may lead to false positive results in high-grade gliomas which may harbor partial 1p and/or 19q loss in a background of widespread complex chromosomal rearrangements. Thus, a FISH result of 1p/19q co-deletion must be correlated with histological and immunophenotypic features before definitively diagnosing oligodendroglioma.

EGFR amplification; Gain of chromosome 7/loss of chromosome 10

The Consortium to Inform Molecular and Practical Approaches to CNS Tumor Taxonomy (cIMPACT-NOW) was established in 2016 to address ongoing challenges regarding the diagnosis and classification CNS tumors. In its third update, the consortium recommended consensus minimal molecular criteria to identify an IDH-wildtype diffuse astrocytic glioma that, despite a histologically low-grade appearance, would follow a clinical course more closely resembling that of an IDH-wildtype glioblastoma.[14] These criteria, now included in the upcoming 2021 WHO Classification of CNS Tumors, are the presence of: (a) EGFR amplification, OR (b) Combined whole chromosome 7 gain and whole chromosome 10 loss (+ 7/− 10), OR (c) TERT promoter mutation.

Although FISH may be utilized to assess for EGFR amplification and +7/-10, the caveats pertaining to assessment for 1p/19q co-deletion are also applicable here, warranting caution while interpreting both positive and negative results.

Fusion probes

FISH fusion probes are used to detect common translocations involving gene regions which, when rearranged, result in different types of solid and lymphoproliferative neoplasms. Although RNA next generation sequencing (NGS) is being increasingly used, FISH remains a useful tool to assess for these alterations, particularly when the amount of available tissue is limited. One major drawback of utilizing FISH to assess for fusions is that the specificity of the probes limits the rearrangements which can be identified, and so alternate fusion partners or breakpoints with the same functional consequence may go undetected. Examples of CNS tumors that harbor characteristic gene rearrangements include pilocytic astrocytoma (particularly the KIAA1549-BRAF fusion), supratentorial ependymoma (ZFTA- or YAP1-fusion positive), intracranial mesenchymal tumor, FET-CREB fusion-positive, and CIC-rearranged sarcoma. Examples of some common FISH probes used in CNS tumors are summarized in [Table 5].
Table 5: Common FISH probes used in CNS tumors

Click here to view


Chromosomal microarray analysis (CMA)

CMA compares the entire genome of the tumor against a reference genome, identifying differences and locating regions of genomic imbalance (copy number variations (CNVs)). Given that this technique allows for whole-genome evaluation at a high resolution, it is currently the preferred method to assess for CNVs, including 1p/19q co-deletion, EGFR amplification, +7/-10, and CDKN2A/B homozygous deletion. CMA may also help detect additional alterations which further refine the diagnosis of specific entities, such as 7q34 gain/duplication, which is indicative of the KIAA1549-BRAF fusion detected in a subset of pediatric low-grade tumors, particularly pilocytic astrocytomas arising in the cerebellum.[15] Additionally, CMA can also screen for alterations predictive of therapeutic significance, such as identification of FGFR-TACC and NTRK fusions, present in a subset of low- and high-grade gliomas, which indicate potential sensitivity to FGFR tyrosine kinase inhibitor therapy.[16],[17]

Next-generation sequencing (NGS)

NGS includes a group of technologies that allow for massive parallel sequencing of various lengths of DNA or RNA sequences, or even whole-genome and whole-exome sequencing. NGS accomplishes this faster and at lower cost than its precursor, Sanger sequencing, and has dramatically changed the study of the molecular biology of a large number of CNS and non-CNS tumors.[18] In contrast to CMA, NGS-based approaches do not require prior knowledge of the genome and offer single-nucleotide resolution, making it possible to detect specific point mutations, alternatively spliced transcripts, allelic gene variants, and single nucleotide polymorphisms (SNPs). Additionally, the quantity of tissue required to perform a focused NGS assay is significantly less. For CNS tumors, a targeted NGS panel is a powerful tool to assess for specific molecular alterations which might otherwise have required multiple types of ancillary tests to confirm and/or further classify certain neoplasms, as well as provide valuable information to indicate potential treatment response.[19]

Whole-genome/Whole-exome DNA NGS

One of the advantages of NGS is the ability to interrogate multiple targeted genes simultaneously. As its name implies, whole-genome sequencing (WGS) interrogates almost the entire genome, including nuclear and mitochondrial DNA, while whole exome sequencing (WES) evaluates the 1% of the genome which is transcribed into proteins. At present, WGS and WES are used predominantly in research settings and, clinically, to assess for constitutional genetic diseases rather than for somatic mutations within tumors.

Targeted panel NGS

Solid tumors may have multiple mutations that can arise either from the outset, or due to expansion and divergent differentiation of a single clone. Additionally, metastatic tumors may harbor mutations that differ from the primary tumor and, with the surge of immunotherapy, it has become increasingly important to assess for tumor mutation burden in certain neoplasms. NGS allows testing of multiple gene mutations for diagnostic, prognostic, and predictive purposes, and is also widely used to identify high-risk populations for certain hereditary cancers, highlighting its role in personalized precision medicine.

Targeted NGS is fast becoming the most common NGS assay for evaluation of tumors. Compared to whole-genome sequencing, it limits the number of genes tested, and allows for the design of standardized panels to interrogate alterations specific for the type of neoplasm (e.g. melanoma, lung carcinoma, CNS tumors, etc.). This also allows a greater depth of sequencing, which is necessary to help identify mutations with different allelic frequencies. The ability to evaluate multiple relevant genes utilizing small amounts of tissue makes this technique suitable for cases with limited specimens, as is often the case in neurosurgical biopsies.

Single gene sequencing

As with any ancillary test, the first step in when deciding whether NGS needs to be performed for a CNS neoplasm is to determine which mutations need to be evaluated for, and how the results are going to help with further characterization. This decision is ideally based on the tumor's morphologic and immunophenotypic features, along with current guidelines for the assessment of specific tumor entities. A second key consideration is the amount of tissue available. When samples are small and/or tumor is present only focally or is of low density, evaluation of a single appropriate gene may provide the most useful information for accurate diagnosis and/or detecting alterations amenable to targeted therapy.

In contrast to NGS that sequences millions of DNA fragments simultaneously, Sanger sequencing is a “first generation” method that sequences a single DNA fragment at a time. Nonetheless, advantages of this technique include its short turnaround time and cost-effectiveness for low numbers of targets. Sanger sequencing is particularly important when the quantity of tissue available is limited and possibly insufficient for NGS and is still widely used to confirm sequence variants identified by NGS and to span regions poorly covered by NGS, such as GC-rich regions which are more resistant to sequencing due to poor capture or amplification.

On the other hand, droplet digital polymerase chain reaction (ddPCR) is a digital PCR method that uses a water-oil emulsion droplet system. ddPCR partitions nucleic acid samples into thousands of nanoliter-sized droplets that separate the template DNA molecules and essentially serve as individual test tubes or wells in which the PCR amplification takes place. This technique requires less amounts of tissue, which makes it more suitable for small specimens.

Examples include single gene sequencing to assess for IDH1, IDH2, TERT promoter, and BRAF mutations. IDH mutations are the hallmark of two types of infiltrating gliomas that have better outcomes compared to IDH-wildtype counterparts - Astrocytoma, IDH-mutant, and Oligodendroglioma, IDH-mutant and 1p/19q co-deleted. The IHC surrogate for the canonical IDH1 p.R132H mutation detects approximately 80% of all IDH mutations, but there remains a significant proportion of cases which harbor a different IDH1 or an IDH2 mutation. Current guidelines recommend assessing for non-canonical IDH mutations in patients 54 years of age or younger with negative IDH1-R132H IHC, and focused IDH sequencing can help assess this even in limited samples.[20],[21] As mentioned above, TERT promoter mutation is one of the diagnostic molecular criteria to help establish a diagnosis of IDH-wildtype glioblastoma in otherwise histologically grade 2 or 3 diffuse astrocytic gliomas (i.e. diffuse astrocytoma with molecular features of glioblastoma), and can likewise be assessed by single gene sequencing. The caveat, however, is that TERT promoter mutations can be detected in a spectrum of neoplasia within and outside the CNS (including oligodendroglioma and ependymoma). Consequently, appropriate interpretation of the result depends upon first establishing an accurate histological diagnosis, as well as demonstrating the absence of IDH1/2 mutations in the context of a diffuse glioma. BRAF V600E mutation can be present in a wide variety of CNS [Table 4] and non-CNS neoplasms. Testing for this mutation not only allows for a more refined diagnosis, but also provides predictive information regarding potential response to BRAF inhibitor therapy.

RNA NGS

NGS can also be performed at the transcriptome level, which includes assembly of RNA transcripts including messenger RNA (mRNA), ribosomal RNA (rRNA), transference RNA (tRNA), micro-RNAs, and noncoding RNAs. This type of RNA NGS can be used to detect gene fusions/rearrangements. Like DNA targeted sequencing, it is possible to create standardized panels to study specific types of tumors, evaluating for characteristic translocations/rearrangements with diagnostic and therapeutic significance. For example, among common CNS tumors, RNA NGS can be used to evaluate for the KIAA1549-BRAF fusion characteristic of pilocytic astrocytoma, MYB- or MYBL1- fusions in pediatric-type low-grade diffuse gliomas, or for ZFTA- or YAP1- fusions in supratentorial ependymoma. Common molecular markers with their associated tumor types and method of detection are shown in [Table 6].
Table 6: Common molecular markers, associated tumor types, and method of detection

Click here to view


Whole genome methylation analysis and the DKFZ/Heidelberg classifier

DNA methylation is an epigenetic mechanism to regulate gene expression by recruiting proteins involved in gene repression, or by inhibiting the binding of transcription factors to DNA. DNA methylation is essential to regulate tissue-specific gene expression, genomic imprinting, and X-chromosome inactivation, as well as influence gene activity within neoplasia. Whole genome methylation analysis involves the evaluation of patterns of DNA or histone modification by methyl groups across the genome which, in tumors, results from a combination of somatically acquired DNA methylation changes and the cell of origin.[22] Such profiling has helped to refine the diagnosis and classification of CNS tumors with ambiguous histology, or with non-conclusive or contradictory molecular profiles, and their molecular signatures are being increasingly incorporated into cancer classification systems, including in the 2021 WHO Classification of CNS Tumors.[23],[24] As our understanding of the methylation profiles across CNS tumor types evolves, whole genome methylation analysis provides a valuable tool for particularly challenging cases when all other modalities have proven inconclusive. One caveat, however, is that while methylation profiling may help establish the diagnosis of a specific entity, the detection of specific alterations necessary for therapeutic decisions still need to be established by utilizing one of the techniques described above, at least at present.

In 2018, a brain tumor methylation classifier developed by the German Cancer Research Center (DKFZ) and Heidelberg University in Heidelberg, Germany was made available through a free online tool (www.molecularneuropathology.org), to assist with the identification of distinct DNA methylation classes of CNS tumors. The raw data generated by methylation profiling performed locally can be uploaded, to be compared to methylation data from a reference cohort comprising over 2800 CNS tumors of over 80 tumor classes or subclasses. Along with methylation profiling, this tool can estimate the MGMT promoter methylation status, and provide a low-resolution copy number plot calculated from the array data of the case which can assist assessment of chromosomal copy number alterations. This classifier can be used in clinical practice to help in the diagnosis of challenging cases where clinicopathologic, histologic, and/or molecular features are ambiguous or inconclusive, as well as possibly help stratify risk groups within histologically well-defined CNS tumors.


   Conclusion Top


Here, we have described a practical approach to evaluating neurosurgical specimens, incorporating the recognition of main histological patterns, correlation with clinical and radiological features, and selection of appropriate ancillary testing (immunohistochemical and/or molecular, as and when needed). While a detailed description of each entity is beyond the scope of this review, we hope that such an algorithm will help provide a framework to assist with accurate and clinically relevant diagnoses for CNS tumors.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.



 
   References Top

1.
Louis DN, Perry A, Wesseling P, Brat DJ, Cree IA, Figarella-Branger D, et al. The 2021 WHO classification of tumors of the central nervous system: A summary. Neuro Oncol 2021;23:1231-51.  Back to cited text no. 1
    
2.
Wen PY, Packer RJ. The 2021 WHO classification of tumors of the central nervous system: Clinical implications. Neuro Oncol 2021;23:1215-7.  Back to cited text no. 2
    
3.
Ostrom QT, Cioffi G, Waite K, Kruchko C, Barnholtz-Sloan JS. CBTRUS statistical report: Primary brain and other central nervous system tumors diagnosed in the United States in 2014-2018. Neuro Oncol 2021;23 (12 Suppl 2):iii1-105.  Back to cited text no. 3
    
4.
Kleinschmidt-DeMasters BK, Prayson RA. An algorithmic approach to the brain biopsy--Part I. Arch Pathol Lab Med 2006;130:1630-8.  Back to cited text no. 4
    
5.
Pekmezci M, Stevers M, Phillips JJ, Van Ziffle J, Bastian BC, Tsankova NM, et al. Multinodular and vacuolating neuronal tumor of the cerebrum is a clonal neoplasm defined by genetic alterations that activate the MAP kinase signaling pathway. Acta Neuropathol 2018;135:485-8.  Back to cited text no. 5
    
6.
Peiffer J, Kleihues P. Hans-Joachim Scherer (1906-1945), pioneer in glioma research. Brain Pathol 1999;9:241-5.  Back to cited text no. 6
    
7.
Rong Y, Durden DL, Van Meir EG, Brat DJ. 'Pseudopalisading' necrosis in glioblastoma: A familiar morphologic feature that links vascular pathology, hypoxia, and angiogenesis. J Neuropathol Exp Neurol 2006;65:529-39.  Back to cited text no. 7
    
8.
Gregorio A, Corrias MV, Castriconi R, Dondero A, Mosconi M, Gambini C, et al. Small round blue cell tumours: Diagnostic and prognostic usefulness of the expression of B7-H3 surface molecule. Histopathology 2008;53:73-80.  Back to cited text no. 8
    
9.
Sosunov AA, McKhann GM 2nd, Goldman JE. The origin of Rosenthal fibers and their contributions to astrocyte pathology in Alexander disease. Acta Neuropathol Commun 2017;5:27.  Back to cited text no. 9
    
10.
Murayama S, Bouldin TW, Suzuki K. Immunocytochemical and ultrastructural studies of eosinophilic granular bodies in astrocytic tumors. Acta Neuropathol 1992;83:408-14.  Back to cited text no. 10
    
11.
Prayson RA, Kleinschmidt-DeMasters BK. An algorithmic approach to the brain biopsy--Part II. Arch Pathol Lab Med 2006;130:1639-48.  Back to cited text no. 11
    
12.
Edgar MA, Rosenblum MK. The differential diagnosis of central nervous system tumors: A critical examination of some recent immunohistochemical applications. Arch Pathol Lab Med 2008;132:500-9.  Back to cited text no. 12
    
13.
Louis DN, Giannini C, Capper D, Paulus W, Figarella-Branger D, Lopes MB, et al. cIMPACT-NOW update 2: Diagnostic clarifications for diffuse midline glioma, H3 K27M-mutant and diffuse astrocytoma/anaplastic astrocytoma, IDH-mutant. Acta Neuropathol 2018;135:639-42.  Back to cited text no. 13
    
14.
Brat DJ, Aldape K, Colman H, Holland EC, Louis DN, Jenkins RB, et al. cIMPACT-NOW update 3: Recommended diagnostic criteria for “Diffuse astrocytic glioma, IDH-wildtype, with molecular features of glioblastoma, WHO grade IV”. Acta Neuropathol 2018;136:805-10.  Back to cited text no. 14
    
15.
Faulkner C, Ellis HP, Shaw A, Penman C, Palmer A, Wragg C, et al. BRAF fusion analysis in pilocytic astrocytomas: KIAA1549-BRAF 15-9 fusions are more frequent in the midline than within the cerebellum. J Neuropathol Exp Neurol 2015;74:867-72.  Back to cited text no. 15
    
16.
Ellison DW, Hawkins C, Jones DT, Onar-Thomas A, Pfister SM, Reifenberger G, et al. cIMPACT-NOW update 4: Diffuse gliomas characterized by MYB, MYBL1, or FGFR1 alterations or BRAF (V600E) mutation. Acta Neuropathol 2019;137:683-7.  Back to cited text no. 16
    
17.
Lasorella A, Sanson M, Iavarone A. FGFR-TACC gene fusions in human glioma. Neuro Oncol 2017;19:475-83.  Back to cited text no. 17
    
18.
Goodwin S, McPherson JD, McCombie WR. Coming of age: Ten years of next-generation sequencing technologies. Nat Rev Genet 2016;17:333-51.  Back to cited text no. 18
    
19.
Qin D. Next-generation sequencing and its clinical application. Cancer Biol Med 2019;16:4-10.  Back to cited text no. 19
    
20.
Brat DJ, Aldape K, Colman H, Figrarella-Branger D, Fuller GN, Giannini C, et al. cIMPACT-NOW update 5: Recommended grading criteria and terminologies for IDH-mutant astrocytomas. Acta Neuropathol 2020;139:603-8.  Back to cited text no. 20
    
21.
von Deimling A, Korshunov A, Hartmann C. The next generation of glioma biomarkers: MGMT methylation, BRAF fusions and IDH1 mutations. Brain Pathol 2011;21:74-87.  Back to cited text no. 21
    
22.
Moore LD, Le T, Fan G. DNA methylation and its basic function. Neuropsychopharmacology 2013;38:23-38.  Back to cited text no. 22
    
23.
Jaunmuktane Z, Capper D, Jones DTW, Schrimpf D, Sill M, Dutt M, et al. Methylation array profiling of adult brain tumours: Diagnostic outcomes in a large, single centre. Acta Neuropathol Commun 2019;7:24.  Back to cited text no. 23
    
24.
Karimi S, Zuccato JA, Mamatjan Y, Mansouri S, Suppiah S, Nassiri F, et al. The central nervous system tumor methylation classifier changes neuro-oncology practice for challenging brain tumor diagnoses and directly impacts patient care. Clin Epigenetics 2019;11:185.  Back to cited text no. 24
    

Top
Correspondence Address:
Aditya Raghunathan
Department of Laboratory Medicine and Pathology, Mayo Clinic, 200 First St SW, Rochester, MN 55905
USA
Login to access the Email id

Source of Support: None, Conflict of Interest: None


DOI: 10.4103/ijpm.ijpm_1081_21

Rights and Permissions


    Figures

  [Figure 1], [Figure 2]
 
 
    Tables

  [Table 1], [Table 2], [Table 3], [Table 4], [Table 5], [Table 6]



 

Top
 
 
  Search
 
    Similar in PUBMED
   Search Pubmed for
   Search in Google Scholar for
 Related articles
    Email Alert *
    Add to My List *
* Registration required (free)  


    Abstract
   Introduction
    Clinical and rad...
    Histological fea...
   Rosenthal fibers
   Molecular testing
   Conclusion
    References
    Article Figures
    Article Tables

 Article Access Statistics
    Viewed110    
    Printed2    
    Emailed0    
    PDF Downloaded17    
    Comments [Add]    

Recommend this journal