2019 NPB Minisymposium

This minisymposium was originally published in 2019. The information provided in this minisymposium was accurate and correct at the time of initial program release. Any changes in terminology since the time of initial publication may not be reflected in this minisymposium.

Surrogate Biomarkers

The use of surrogate biomarkers for molecular alterations in tumors of the central nervous system (CNS) has increased dramatically in the last five to ten years with an even sharper rise after the 2016 update to the WHO Classification of Tumours of the Central Nervous System. These biomarkers can be diagnostic, prognostic, and/or predictive (eg, provide biological targets for treatment). Not only are they less expensive than other forms of molecular testing, but they are usually faster and easily integrated into the usual workflow of surgical pathology. Although traditional forms of molecular testing are still needed in some cases, an integrated WHO 2016 diagnosis can be reached in many cases using light microscopy alone.

The predominant method for detection of surrogate biomarkers, which vary in their mechanism of action, is through IHC antibodies. In some cases, antibodies are directed at a specific mutation. For tumors with activating or inactivating mutations, IHC can show the resulting protein overexpression or loss of expression. Herein we explore some of the more common IHC antibodies currently in use in tumor neuropathology based on their individual mechanisms of action. Please refer to Table 1 for a more comprehensive list of biomarkers and their utility.

Table 1: Surrogate biomarkers in nervous system tumors.

^{_{‡ While >10% tumor cell positivity can indicate a TP53 mutation, some gliomas harbor inactivating or truncating mutations in TP53 resulting in a total lack of positivity for p53.}}

^{_{† Loss of nuclear positivity in tumor cells must show concomitant retention of nuclear positivity in adjacent or infiltrated nonneoplastic cells such as endothelial or inflammatory cells.}}

^{_{* While cytoplasmic positivity can be seen widely for these antibodies, nuclear positivity is needed to suggest pathologic mutations in tumor cells.}}

Mutation-Specific Antibodies

With the development of specific antibodies, mutant protein products can be identified through simple IHC detection. Amazingly, many of these antibodies detect seemingly small antigenic changes resulting from point mutations. Examples of some of the more commonly utilized mutation-specific antibodies include mutant IDH1 R132H protein in IDH-mutant gliomas and mutant histone 3 (H3.1 or H3.3) protein products (both K27M and G34R/V) in diffuse midline gliomas or hemispheric IDH-wildtype gliomas, respectively.

Missense mutations in either the IDH1 or IDH2 genes define infiltrating lower grade gliomas including astrocytomas and oligodendrogliomas, as well as a small percentage of glioblastomas, most of which arise from progression of these lower grade forms (ie, secondary glioblastoma). Because these gliomas are associated with dramatically better survival times than their IDH-wildtype counterparts and are otherwise biologically and genetically distinct, the WHO 2016 uses this feature as a major stratification point in the classification of diffuse gliomas. As such, integrated WHO diagnoses are now entirely dependent on knowing the status of IDH1/2 (eg, anaplastic astrocytoma, IDH-mutant; oligodendroglioma, IDH-mutant, and 1p/19q-codeleted; glioblastoma, IDH-wildtype; etc). As an extremely useful screening tool, a commercial IHC antibody is now available that can recognize the most common (>90%) IDH mutation: mutant IDH1 R132H protein. IHC for mutant IDH1 R132H is highly sensitive and specific and should be utilized in the workup of any infiltrating glioma, particularly in adult and adolescent patients; IDH mutations are uncommon in young children. Because ~10% of infiltrating lower grade gliomas will show alternative IDH1 or IDH2 mutations (eg, IDH1 R132C, IDH1 R132G, etc), in cases that are IDH-nonimmunoreactive, follow-up molecular genetic testing for other IDH mutations should be performed on all grade II or grade III astrocytomas and oligodendrogliomas, all gliomas with ATRX loss, and glioblastomas in patients under the age of 55 (IDH-mutation is seen in <1% of glioblastomas in patients 55 and over).

The genes HIST1H3B and H3F3A encode histone 3 protein variants H3.1 and H3.3, respectively. The K27M mutation (sometimes listed as K28M depending on amino acid numbering) is characteristic of diffuse midline gliomas seen in both children and adults, irrespective of location (thalamus, pons, and spinal cord are most common). A mutant-specific antibody that can detect either H3.1 or H3.3 K27M mutations is highly sensitive and specific. Strong and diffuse nuclear positivity is key to interpretation of this antibody, such that background cytoplasmic (most often in histiocytes) and axonal staining should be ignored (Image 1). Importantly, the WHO defines all diffuse midline gliomas with H3 K27M mutation as WHO grade IV even if the biopsy otherwise resembles a WHO grade II or III astrocytoma using traditional histologic grading. As the entity name implies, however, there must be a diffusely infiltrative growth pattern, it must be centered in a midline location, it must have an astrocytoma appearance, and it must be positive for the mutation by IHC or sequencing. A recently published clarification by the Consortium to Inform Molecular and Practical Approaches to CNS Tumor Taxonomy – Not Officially WHO (cIMPACT-NOW) reflects reports since the WHO 2016 was published, showing that H3 K27M mutations are also rarely found in nonmidline gliomas, gangliogliomas, pilocytic astrocytomas, and ependymomas, tumors that don’t necessarily share the automatic WHO grade IV biology of the diffuse midline gliomas. Liberal use of this antibody is suggested, particularly in gliomas with any midline component, and can be performed in conjunction with a trimethylation-specific antibody (H3K27me3). Confirmatory molecular testing can be performed in equivocal cases but needs to interrogate both genes.

Other H3F3A mutations have been described in biologically distinct nonmidline/hemispheric high-grade infiltrating gliomas: H3 G34R and G34V; reports suggest that the prognosis for such tumors is intermediate between that of IDH-mutant glioblastoma and diffuse midline glioma, H3 K27M-mutant. New commercially available antibodies against H3 G34R/V mutant proteins are now being used at some academic medical centers with broader use on the horizon. Currently, this is not a separate diagnostic entity in the WHO scheme, but this is likely to change in future versions.

Increased Protein Expression Secondary to Signal Pathway Activating Alterations or Protein Stabilization

When molecular genetic alterations lead to activation of specific cell signaling pathways, they are often reflected by increased protein expression in downstream markers. Some examples of this include IHC for p53 in gliomas, L1CAM or p65RELA in ependymomas, and LIN28 in embryonal tumors with multilayered rosettes.

Mutations in the TP53 gene are commonly found in IDH-mutated astrocytomas and can also be seen in some IDH-wildtype glioblastomas and H3 K27M-mutant diffuse midline gliomas, while they are usually not found in oligodendrogliomas. The majority of TP53 glioma mutations result in abnormally prolonged half-life for the protein product or increased protein stabilization and can be identified as a significant increase in nuclear p53 positivity by IHC. For higher sensitivity and specificity, >10% of tumor cells must show strong nuclear positivity, although most TP53 mutant gliomas show widespread p53 expression. However, rarely, nonsense mutations or deletions in TP53 can result in a loss of protein expression and lack of p53 immunopositivity entirely (Image 2). Other uses for p53 IHC include subclassification of SHH pathway medulloblastomas into TP53-mutant high-risk cases versus TP53-wildtype low-risk cases and identification of TP53-mutant choroid plexus carcinoma in Li-Fraumeni syndrome (Table 1).

IHC using antibodies against L1 cell adhesion molecule (L1CAM) or p65RELA can be used as surrogate markers for RELA fusion-positive supratentorial ependymomas most commonly seen in young children; this integrated diagnosis is now a distinct WHO 2016 entity. Increased L1CAM protein expression is seen downstream of NF-kB activation and is driven by the characteristic C11orf95-RELA fusion seen in these tumors. While sensitive, increased expression of L1CAM is not entirely specific and has been found in other neoplasms. To increase specificity, there must be strong and diffuse cytoplasmic positivity and the tumor should clearly be a supratentorial ependymoma using classic histopathologic and IHC criteria. Nevertheless, follow-up molecular testing is required to confirm the RELA fusion.

Embryonal tumor with multilayered rosettes (ETMR), C19MC-altered, an aggressive CNS embryonal neoplasm in infants, represents another new WHO 2016 entity and is molecularly defined by amplification of the C19MC oncogenic miRNA cluster. This amplification results in upregulation of the LIN28 gene and increased LIN28 protein expression. Immunostaining for LIN28 in ETMR tumors typically shows strong positivity within the multilayered rosettes, small-cell tumor areas, and/or papillary/tubular structures while showing little to no staining within neuropil-rich areas. LIN28 expression is not entirely specific as focal positivity can be seen in tumors such as atypical teratoid/rhabdoid tumor and some high-grade gliomas, but strong and diffuse cytoplasmic positivity lends support to a diagnosis of ETMR, particularly when the histopathology is otherwise classic (eg, primitive cells, ependymoblastic rosettes, neuronal/neurocytic maturation with abundant neuropil). Ultimately, follow-up molecular testing interrogating C19MC is required to make a definitive diagnosis.

Loss of Protein Expression Secondary to Inactivating Alterations

Loss of protein expression by IHC in tumor cells is a surrogate for inactivating or truncating molecular alterations in CNS tumors, and retention of staining in nonneoplastic cells (eg, endothelial cells, entrapped neurons, inflammatory cells, etc) provides a critical internal positive control. Some examples include IHC for ATRX in gliomas and INI1 or BRG1 in atypical teratoid/rhabdoid tumors (Image 3).

Inactivating mutations of the b-thalassemia/mental retardation syndrome X-linked gene (ATRX) are found predominately in IDH-mutant astrocytomas (most examples harboring both ATRX and TP53 mutations). When staining these astrocytomas with an antibody against ATRX, loss of nuclear positivity in tumor cells with retention in nonneoplastic cells is strongly associated with the presence of an ATRX mutation or other inactivating mechanism. It is now becoming standard practice to perform immunostaining for IDH1 R132H in conjunction with p53 and ATRX in the workup of infiltrating gliomas. Also, because ATRX mutations are virtually mutually exclusive with codeletion of chromosomes 1p/19q, if clear loss of ATRX immunopositivity is identified, interrogation of 1p/19q deletion status is not required to rule out an oligodendroglioma; ATRX staining is retained in nearly all oligodendrogliomas. ATRX mutations are also seen concurrently in about 15% of H3 K27M-mutant diffuse midline gliomas and nearly all H3 G34R/V-mutant gliomas, often in conjunction with p53 overexpression.

The vast majority of atypical teratoid/rhabdoid tumors (AT/RT) harbor biallelic somatic inactivation of the SMARCB1 gene (INI1). Therefore, IHC for INI1 will show nuclear loss in tumor cells and retained positivity in nonneoplastic cells. Extremely rare AT/RTs show inactivation of the SMARCA4 gene (BRG1) instead. IHC in BRG1-altered AT/RTs will show loss of nuclear positivity for BRG1 in tumor cells with retained positivity in nonneoplastic cells. In these tumors, INI1 is retained in both tumor cells and nonneoplastic cells.

Abnormal Nuclear Localization of Proteins

Interpretation of some surrogate biomarkers relies on abnormal protein localization, particularly pathologic nuclear translocation of otherwise cytoplasmic proteins. The molecular mechanisms behind the abnormal localization vary widely, but the IHC interpretation is fairly straightforward. For beta-catenin IHC in medulloblastomas and adamantinomatous craniopharyngiomas, as well as STAT6 IHC in solitary fibrous tumor/hemangiopericytomas (SFT/HPCs), nuclear positivity is evidence of abnormal protein translocation/localization, reflecting underlying genetic alterations that define various diagnostic entities and tumor variants.

Beta-catenin protein is encoded by the CTNNB1 gene, and nuclear translocation of beta-catenin can be seen in multiple tumor types by varying mechanisms. In WNT-activated medulloblastomas, nuclear translocation of beta-catenin can be detected by immunostaining. However, this can be focal and thus challenging to interpret. Confirmation of WNT activation in medulloblastomas can be obtained with additional surrogate biomarkers such as ALK or LEF1 (Table 1) or more definitively via molecular confirmation. In adamantinomatous craniopharyngiomas, mutations in the CTNNB1 gene result in aberrant beta-catenin nuclear translocation that can be very focal in small clusters or even single cells.

Meningeal SFT/HPC is characterized by in-frame NAB2-STAT6 gene fusions. NAB2 protein is typically found in nuclei, and STAT6 is typically cytoplasmic, but the fusion product results in abnormal translocation of STAT6 into the nucleus (Image 4). Nuclear STAT6 immunopositivity is almost 100% sensitive and specific for a diagnosis of SFT/HPC, and staining should be performed whenever this entity is in the differential.

Methylation-Specific Antibodies

The most commonly used methylation-specific antibody in evaluation of CNS tumors interrogates trimethylation of H3 K27 (H3K27me3). This antibody is most helpful when used in association with mutant H3 K27M IHC because the mutant protein product is associated with loss of the repressive trimethyl on lysine 27. Therefore, tumors with H3 K27M mutations will show nuclear positivity for mutant H3 K27M and loss of nuclear expression of H3K27me3 with retention in nonneoplastic cells. Loss of H3K27me3 expression is not entirely specific for diffuse midline gliomas and can also be seen in posterior fossa ependymomas (the PFA group of ependymomas, specifically), as well as malignant peripheral nerve sheath tumors (MPNSTs). In contrast, a mosaic pattern of H3K27me3 expression, wherein some tumor nuclei appear positive and others negative, is now considered a nonspecific pattern.

• The use of surrogate IHC biomarkers for molecular alterations in CNS neoplasms can assist with neuropathologic diagnosis, enhance prognostic accuracy, and even suggest potential therapeutic targets.
• Antibodies directed at point mutations in the H3F3A gene include mutant H3 K27M and mutant H3 G34R/V that can identify biologically distinct diffuse midline gliomas and hemispheric gliomas, respectively.
• Increased protein expression of L1CAM occurs downstream of NF-kB activation, which is driven by C11orf95-RELA fusions in supratentorial ependymomas in children (RELA fusion-positive).
• While >10% tumor cell positivity for p53 is highly suggestive of a mutation in the TP53 gene in gliomas, rare nonsense mutations or deletions in TP53 result in a total lack of immunopositivity, and occasional false positives may also be encountered.
• Loss of nuclear positivity for INI1, or less commonly BRG1, is seen in atypical teratoid/rhabdoid tumors and is evidence of biallelic somatic inactivation of SMARCB1 or SMARCA4, respectively.
• Nuclear STAT6 immunopositivity is almost 100% sensitive and specific for a diagnosis of solitary fibrous tumor/hemangiopericytoma.
• Loss of nuclear expression of H3K27me3 is seen in conjunction with H3 K27M mutations in diffuse midline gliomas.

1. Bechet D, Gielen GGH, Korshunov A, et al. Specific detection of methionine 27 mutation in histone 3 variants (H3K27M) in fixed tissue from high-grade astrocytomas. Acta Neuropathol. 2014;128:733-41.
2. Gessi M, Capper D, Sahm F, et al. Evidence of H3 K27M mutations in posterior fossa ependymomas. Acta Neuropathol. 2016;135:635-37.
3. Haque F, Varlet P, Puntonet J, et al. Evaluation of a novel antibody to define histone 3.3 G34R mutant brain tumours. Acta Neuropathologica Communications. 2017;5:45.
4. Hasselblatt M, Gesk S, Oyen F, et al. Nonsense mutation and inactivation of SMARCA4 (BRG1) in an Atypical Teratoid/Rhabdoid Tumor showing retained SMARCB1 (INI1) expression. Am J Surg Pathol. 2011;35(6):933-5.
5. Korshunov A, Capper D, Reuss D, et al. Histologically distinct neuroepithelial tumors with histone 3 G34 mutation are molecularly similar and comprise a single nosologic entity. Acta Neuropathol. 2016;131(1):137-46.
6. Korshunov A, Ryzhova M, Jones DTW, et al. LIN28A immunoreactivity is a potent diagnostic marker of embryonal tumor with multilayered rosettes (ETMR). Acta Neuropathol. 2012;124:875-81.
7. Louis DN, Giannini C, Capper D, 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(4):639-42.
8. 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):803-20.
9. Parker M, Mohankumar KM, Punchihewa C, et al. C11orf95-RELA fusions drive oncogenic NF-kB signaling in ependymoma. Nature. 2014;506:451-5.
10. 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;129:133-46.
11. Schweizer L, Koelsche C, Sahm F, et al. Meningeal hemangiopericytoma and solitary fibrous tumors carry the NAB2-STAT6 fusion and can be diagnosed by nuclear expression of STAT6 protein. Acta Neuropathol. 2013;125:651-8.
12. Sekine S, Shibata T, Kokubu A, et al. Craniopharyngiomas of adamantinomatous type harbor beta-catenin gene mutations. Am J Pathol. 2002;161(6):1997-2001.
13. Spence T, Sin-Chan P, Picard D, et al. CNS-PNETs with C19MC amplification and/or LIN28 expression comprise a distinct histogenetic diagnostic and therapeutic entity. Acta Neuropathol. 2014;128:291-303.
14. Tanboon J, Williams EA, Louis DN. The diagnostic use of immunohistochemical surrogates for signature molecular genetic alterations in glioma. J Neuropath Exp Neurol. 2016;75(1):4-18.
15. Torres-Mora J, Dry S, Li X, Binder S, Amin M, Folpe AL. Malignant melanotic schwannian tumor. A clinicopathologic, immunohistochemical, and gene expression profiling study of 40 cases, with a proposal for the reclassification of “melanotic schwannoma.” Am J Surg Pathol. 2014;38(1):94-105.