Clinical aspects

Description

	• Several molecular diagnostic technologies are available for molecular diagnosis of brain tumors.
	• Most of the studies are based on brain biopsy, but blood and CSF provide material for detection of circulating tumor DNA.
	• Biomarkers also play a role in diagnosis of brain tumors.

Various molecular methods for the study of tumors are described in detail elsewhere (16). Important features applicable to brain tumors are:

Genome analysis at the molecular level. Genome analysis involves high resolution of the tumor DNA with the constitutional DNA from the patient to detect allelic imbalances by Southern blotting. This approach uses restriction fragment length polymorphism. This analysis is done on DNA extracted from fresh or frozen tumor tissue and, thus, no information is obtained at the single-cell level, and contamination of the tumor DNA with that from admixed non-neoplastic cells may create interpretational problems. However, the application of DNA polymorphism to the study of human cancers has provided important clues regarding the biological significance of imbalances in chromosomal material that are so often observed in cytogenetic studies of tumors. Allelotyping of the individual tumors present a considerable workload and would take several months if one were to apply these to the current histopathological classification systems. A diagnostic algorithm for the diagnosis of diffuse brain tumor is shown here.

Southern blot. Southern blotting can detect restriction fragment length polymorphism, but its major limitations are its low sensitivity for the detection of altered DNA (1% to 5%), the need to obtain DNA from fresh tissues, and the length of time required to obtain results.

Gel electrophoresis. This is at least 10 times as sensitive as Southern blot assay, and it can readily determine clonality in the primary diagnosis of a neoplasm.

In situ hybridization. DNA in situ hybridization methods permit investigation of genetic alterations within the context of cell morphology and tissue architecture and are important in tumor pathology. The advantages of fluorescent in situ hybridization in tumor diagnosis are as follows:

	• Chromosomal information can be obtained from nonmitotic lesions.
	• Normal cells of the same type present on the same slide serve as internal controls for the reagents as well as the hybridization procedures.
	• Data from small lesions can be obtained, whereas conventional metaphase cytogenetic analyses or mutational analysis by Southern blotting or polymerase chain reaction would be limited by dilution of the sample with normal tissue elements.
	• Although polymerase chain reaction enables recognition of alterations involving fewer cells than does fluorescent in situ hybridization, the exquisite sensitivity of polymerase chain reaction becomes a disadvantage if the clonal aberration that marks the tumor is found in a small number of cells.
	• Rapid reporting within 24 hours.
	• It can relate genetic biomarkers to tissue pathology.
	• Fluorescent in situ hybridization has been found to be a more sensitive test for detecting chromosome 10 loss than conventional karyotype analysis.

Loss of heterozygosity. This term is applied to loss of polymorphic DNA markers in tumors compared with normal cells and often indicates somatic deletion of tumor suppression genes.

Differential display of mRNA. This has been shown to be a useful technique for analyzing differential gene expression in human brain tumors. It offers a means to identify new genes of biological interest in human brain tumors such as oncogenes, tumor suppressor genes, and tumor-specific biomarkers.

Gene expression analysis. Epidermal growth factor receptor (EGFR) overexpression occurs in nearly 50% of cases of glioblastoma. High-density oligonucleotide microarrays can demonstrate that epidermal growth factor receptor-overexpressing glioblastomas have a distinct global gene transcriptional profile, which defines distinct molecular subtypes that may be important in disease stratification, as well as in the discovery and assessment of glioblastoma treatment strategies. Deletion of NFKBIA (encoding nuclear factor of κ-light polypeptide gene enhancer in B-cells inhibitor-alpha), an inhibitor of the EGFR-signaling pathway, promotes tumorigenesis in glioblastomas that do not have alterations of EGFR (05).

SAGE (Serial Analysis of Gene Expression) Genie is a logistically laid out suite of bioinformatics tools that allow automatic and reliable matches of SAGE tags to known gene transcripts. Gene expression level differences among malignant brain tumors can be archived and displayed online.

Gene expression analyses using next-generation sequencing technologies in samples of glioblastoma have led to the discovery of a variety of genes that were not previously known to be altered in these tumors. Mutations that affected amino acid 132 of IDH1 occurred in more than 70% of WHO grade II and III astrocytomas and oligodendrogliomas and in glioblastomas that developed from these lower-grade lesions (35). In this study, patients suffering from glioblastoma with IDH mutations had a better outcome than those with wild-type IDH genes and survived at least twice as long as those without these mutations.

Polymerase chain reaction techniques. Strategies using polymerase chain reaction can detect clonality with high sensitivity. This type of analysis can be done with DNA from paraffin-embedded tissues, and it produces results within a day. Combined with DNA sequencing, polymerase chain reaction can analyze tumor samples rapidly and specifically for mutations in relation to clonal expansions. Polymerase chain reaction has facilitated cancer molecular diagnosis, and some of the examples of the situations where it is useful are as follows:

	• Insufficient material obtained at biopsy.
	• Polymerase chain reaction technology can now detect and even directly sequence p53 gene structural alterations in individual or clustered cells.
	• A real-time reverse transcriptase-polymerase chain reaction can be used to assess the number of human telomerase reverse transcriptase (hTERT) transcripts in primary glial tumors. hTERT mRNA represents a diagnostic and prognostic indicator for glioblastoma patients; survival is worse in high hTERT expressors than in low hTERT expressors.

Differential display-polymerase chain reaction. ddPCR is used to isolate a cDNA fragment that is overexpressed in glioblastoma multiforme tissue as compared to normal brain tissue. Human neuron-glia-related cell adhesion molecule (hNr-CAM) is overexpressed in malignant brain tumors and can serve as a novel marker for brain tumor detection and perhaps therapy.

Comparative genomic hybridization. Comparative genomic hybridization can help in accurately identifying multiple or missing DNA sequences along the entire chromosome with the aid of "biochips" or microarrays. Array comparative genomic hybridization is a powerful technique capable of identifying both gains and losses of DNA sequences. It is labor saving and less tedious than other methods for molecular profiling of glioblastoma; it also enables the study of chromosomes at a much higher resolution than standard cytogenetic techniques.

Detection of DNA abnormalities by single-strand conformation polymorphism. This is a method to screen for DNA variations without direct sequencing. It is used for determining prognosis. It is inexpensive, but frequently misses mutations.

P53 sequencing and functional assays. These are used as prognostic indicators in specific cancers. P53 sequencing is still considered to be the "gold standard" for mutation detection but does not determine protein function. P53 functional assay confirms that the mutation is present and determines if a mutation is deleterious to the function of the gene. P53 sequencing also identifies germline mutations for Li-Fraumeni syndrome, but 1 disadvantage is that it cannot be performed on fixed tissues or blood specimens more than 24 hours old. DNA samples in Li-Fraumeni syndrome can be independently analyzed for P53 mutations using GeneChip and standard automated laser fluorescence sequencing technology. This assay may be feasible for routine molecular genetics diagnostics in determining the P53 status of childhood-tumor patients and in enabling a disease management based on the genetic background of the individual.

Application of DNA microarray technology and biochip technology to brain tumors. Microarray and biochip technology have been applied to study the differential expression of genes involved in the pathogenesis of brain tumors such as vascular endothelial growth factor, insulin growth factor binding proteins, matrix metalloproteinases, and basic fibroblast growth factor. DNA microarrays can identify gene expression differences between high-grade and low-grade glial tumors. The tissue microarray technique enables numerous tissue cores from different specimens to be reassembled into a single microarray block for parallel analytical studies of molecular alterations at the DNA, RNA, or protein level. Gene profiling using microarrays correlates with conventional histologic classification and grading with high fidelity. Application of the molecular approach to tumor classification can generate clinically meaningful patient stratification and enable the identification of previously unrecognized, clinically relevant tumor subsets.

Application of laser capture microdissection and proteomics to study of brain tumors. Laser capture microdissection provides an ideal method for extraction of cells from specimens in that the exact morphologies of both the captured cells and the surrounding tissue are preserved. Laser capture microdissection, in combination with proteomics, can be used in diagnosis of malignant neoplasms. Combination of genomic and proteomic technologies has identified several novel genes with distinct expression patterns in high-grade and low-grade gliomas and protein products of specific genes of interest have been localized in the neoplastic cells of high-grade astrocytomas.

Methods of removal of brain tumor tissue for molecular diagnostic studies. Tissue removed by ultrasonic aspiration of malignant brain tumors not only provides more accurate estimation of malignancy than routine tumor biopsies, but also is more suitable for molecular genetic studies.

Cytogenetics. Molecular cytogenetics has contributed to the study of brain tumors, encompassing the findings of classic cytogenetics, interphase- and metaphase-based fluorescent in situ hybridization studies, spectral karyotyping, and metaphase- and array-based comparative genomic hybridization. Molecular cytogenetic techniques also play a role in understanding the pathogenesis of brain tumors. Chromosomal testing can determine brain tumor therapy outcomes.

A statistical method, called Genomic Identification of Significant Targets in Cancer, has been designed for analyzing chromosomal aberrations in cancer and has been used to study chromosomal aberrations in gliomas.

Fluorescence in situ hybridization, which can detect genetic biomarkers such as 1p19q deletion, EGFR gene amplification, and BRAF rearrangement has been used as a molecular diagnostic tool for brain tumors including gliomas, embryonal neoplasms, ependymomas, and meningiomas (14).

Examination of circulating tumor cells or cell-free tumor DNA in blood. A method based on an adenoviral detection system can detect circulating tumor cells in glioblastoma based on telomerase activity, which is elevated in nearly all tumor cells but not normal cells (20). Glioblastoma patients with a higher concentration of cell-free DNA in their blood have shorter progression-free survival than those with lower concentrations, and spikes in circulating DNA seem to correlate with or even predict disease progression.

Examination of circulating tumor DNA (ctDNA) in CSF. A study of simultaneous analysis of mutations of ctDNA in different glioma subtypes, including lower-grade gliomas versus glioblastoma, found that CSF ctDNA mutations had high concordance rates with tumor DNA and that ctDNA mutations of PTEN and TP53 were commonly detected in patients with recurrent gliomas (36). Isocitrate dehydrogenase mutation was detected in most CSF ctDNA derived from isocitrate dehydrogenase mutant-diffuse astrocytomas, whereas CSF ctDNA mutations of RB1 and EGFR were found in isocitrate dehydrogenase wild-type glioblastoma. Isocitrate dehydrogenase mutation was detected in low-grade gliomas, whereas Rb1 mutation was more commonly detected in glioblastoma. Thus, CSF ctDNA detection can be an alternative method of liquid biopsy in gliomas.

Molecular imaging. PET is the most sensitive and specific technique for imaging molecular pathways in vivo in humans. PET enables quantitative assessment of brain tumor's pathophysiology, gene expression, and biochemistry. Thus, it bridges the gap between in vitro and in vivo diagnosis. Molecular imaging of brain tumors by PET helps in the diagnosis as well as monitoring of the treatment effects during follow-up.

Combined neuroimaging and DNA microarray analysis. This approach has been used to create a multidimensional map of distribution of gene expression patterns in glioblastoma that provides clinically relevant insights into tumor biology. Overexpression of epithelial growth factor receptor (EGFR) can be directly inferred by neuroimaging and validated in an independent set of tumors by immunohistochemistry. This information may be used for selecting patients who may be candidates for individualized therapies.

Biomarkers of brain tumors. Some of the molecular diagnostic technologies can be used for detection of biomarkers of brain tumors, and some diagnostics and therapeutics can be further based on the same biomarkers. Various biomarkers of brain tumors are listed in Table 1.

Table 1. Biomarkers of Brain Tumors

Cytogenetic biomarkers
	• EGFR gene amplification and BRAF rearrangement detected by fluorescence in situ hybridization
	• Loss of heterozygosity (LOH) on chromosomes 1p, 19q, 17p, and 10q
DNA biomarkers
Methylation profiling of brain tumors
	• Detection of methylation-dependent DNA sequence variation: methylSNP
	• Methylation of TMS1, an intracellular signaling molecule
IDH1 R132 or IDH2 R172 mutation
	• A gain-of-function mutation distinguishes diffuse gliomas with an IDH (isocitrate dehydrogenase) mutation from IDH wild-type glioblastomas (34). The IDH family of enzymes comprises 3 isoforms of enzymes that help break down nutrients and generate energy for cells. Mutations in IDH genes prevent cells from differentiating, or specializing, into the kind of cells they are ultimately supposed to become. When cells cannot differentiate properly, they may begin to grow out of control.
Protein biomarkers
	• Receptor protein tyrosine phosphatase
	• Serum protein fingerprinting
	• CSF protein profiling: N-myc oncoprotein, caldesmon, attractin
Metabolite biomarkers detected by magnetic resonance spectroscopy
	• 2-hydroxyglutarate (2HG)
	• N-acetylaspartate (diminished)
	• Choline
	• Lactate
MicroRNAs (miRNAs)
	• Elevated plasma levels of miR-21, miR-128, and miR-342-3p in glioblastoma patients drop to normal after surgery and chemo-radiation (32).
Biomarkers of response to therapy
	• Biomarkers to predict response to EGFR inhibitors
	• MRI biomarker for response of brain tumor to therapy
Biomarkers of susceptibility to developing a glioma
	• rs11556218 polymorphism in IL-16 gene (19).

An example of the use of biomarker studies is analysis of the promoter methylation status of key regulator genes implicated in tumor invasion, apoptosis, and inflammation as well as overall survival, therapy status, and tumor pathological features by using a methylation-specific PCR approach. The results of such approaches indicate that, compared to classic glioblastoma, tumor variations with long-term survival display distinct epigenetic characteristics, which might provide additional prognostic biomarkers for the assessment of this malignancy. Certain molecular biomarkers, particularly MGMT promoter hypermethylation, are associated with response to alkylating chemotherapy and longer survival in glioblastoma. Approximately 40% of glioblastoma patients have a MGMT (O6-methylguanine-DNA methyltransferase) promoter tumor, and this subgroup has been shown to have a better response to standard-of-care radiation and temozolomide chemotherapy. Among the biomarkers, only isocitrate dehydrogenase mutation status (prognostic), MGMT, and 1p/19q co-deletion (predictive) are routinely used for evaluation of glioma patients by clinicians in the United States as well as the United Kingdom.

The PredictMDx test is a commercially available epigenetic assay designed to test for methylation of the MGMT gene to help identify newly diagnosed glioblastoma patients who are most likely to benefit from treatment with alkylating agents and identify patients most likely to respond to targeted therapy for glioblastoma. Several studies have shown that patients suffering from glioblastoma with methylated MGMT promoter had a survival benefit when treated with temozolomide and radiotherapy compared with those who received radiotherapy only, whereas patients with MGMT promoter-unmethylated tumors had no survival benefit from chemotherapy, regardless of whether it was given at diagnosis together with radiotherapy or as a salvage treatment. Therefore, it has been suggested that elderly patients with glioblastoma who are eligible for either radiotherapy or temozolomide should undergo MGMT promoter methylation testing prior to deciding on the method of treatment (30).

More biomarkers are being tested in clinical trials, and it will be important to distinguish biomarkers that provide prognostic information from those that have predictive validity to enable future personalized therapeutic choices with minimal toxicity and better outcomes for patients with malignant gliomas (12).

Metabolic biomarkers of brain tumors. Changes in levels of certain metabolites provide a clue to the pathogenesis of brain tumors. N-acetyl-aspartyl-glutamate, a common dipeptide in brain, is significantly lower in human glioma tissues containing isocitrate dehydrogenase mutations than in gliomas without such mutations.

Complementary metabolic data that are independent from the anatomical MRI information help in the definition of glioma extension, detection of anaplastic areas, and postoperative follow-up.

MRS of the brain can detect 2-hydroxyglutarate (2HG), the oncometabolite produced in neoplasms harboring a mutation in the gene coding for isocitrate dehydrogenase (IDH). A prospective longitudinal imaging study concluded that 2HG concentration positively correlated with tumor cellularity and significantly differed between high- and lower-grade gliomas (07). These data provide a basis for incorporating 2HG MRS into clinical management of IDH-mutated gliomas.

Circulating biomarkers of brain tumors. Using patient-specific mutations as biomarkers, detectable levels of CSF tumor DNA were identified in 74% of brain tumors that abutted on CSF spaces (eg, medulloblastomas, ependymomas, and high-grade gliomas) (33).

Small extracellular vesicles produced by both glioblastoma and stromal cells are required for intercellular communication in the tumor body. As tumor small extracellular vesicles are accessible in biofluids, they are valuable biomarkers for diagnosis of brain tumors. Results of a study have shown that the content of the small extracellular vesicles mirrors the phenotypic signature of the respective glioblastoma cells, making them informative biomarkers for tumor subtyping, eg, CD44, which may be useful for personalizing treatment of glioblastoma (18).

Next generation sequencing for brain tumors. There is significant interest in new sequencing-based technologies that map genetic and epigenetic alterations in brain tumors comprehensively and at high resolution. A customized enrichment/hybrid-capture-based next generation sequencing gene panel, consisting of the entire coding and selected intronic and promoter regions of 130 genes recurrently altered in brain tumors, enables the detection of single nucleotide variations, fusions, and copy number aberrations (26). This approach will be important for treatment decision making as more therapeutic targets emerge, and classification of brain tumors is based on genetic information.

Intraoperative detection of brain tumors. There is a need for a practical method to delineate malignant brain tumors from adjacent normal tissue during surgery.

Several intraoperative diagnostic techniques have been developed for determining the resection margin in brain tumors; these include neuronavigation, magnetic resonance imaging, ultrasound, Raman spectroscopy, and optical fluorescence imaging. When combined with appropriate contrast agents, optical fluorescence imaging can provide the neurosurgeon guidance to improve resection and decrease surgical complications (15).

Fluorescence-guided surgery using preoperative oral 5-aminolevulinic acid (5-ALA) enables intraoperative visualization of malignant glioma tissue and provides the neurosurgeon with real-time guidance for differentiating tumor from normal brain to achieve a significantly higher rate of complete resection of the tumor in comparison with conventional white-light resections (11). Tozuleristide (BLZ-100), a near-infrared imaging agent composed of the peptide chlorotoxin and a near-infrared fluorophore indocyanine green, was tested in a phase 1 dose-escalation study to characterize the safety and pharmacokinetics for use in fluorescence-guided surgery in adults with suspected glioma (24). The results show safety of tozuleristide at doses up to 30 mg and suggest that tozuleristide imaging may be useful for fluorescence-guided surgery of gliomas.

Intraoperative examination of brain tumor biopsy with mass spectrometry. Current practice of histological examination of frozen tissue biopsy for characterization of brain tumor has several limitations. A rapid direct analysis and classification of molecular composition of gliomas can be done by desorption electrospray ionization-mass spectrometry (DESI-MS) imaging, multivariate statistical analysis, and machine learning (09). Results obtained by DESI-MS were correlated to preoperative MRI through neuronavigation and visualized over segmented 3D MRI tumor volume reconstruction, which demonstrated the potential of this technique to guide brain tumor surgery by providing rapid diagnosis and tumor margin assessment in near real time (10). In further development of this approach, tumor metabolite 2-hydroxyglutarate (2-HG) is detected from tissue sections of surgically resected gliomas without complex or time-consuming preparation. Imaging tissue sections with DESI-MS shows that the 2-HG signal overlaps with areas of tumor and that 2-HG levels correlate with tumor content, thereby indicating tumor margins (27).

The goal in molecular diagnostics of central nervous system tumors is to discover the genes for various tumors and to refine the methods for detection. The genetic information may be useful for determining prognosis as well as response to therapy.

Indications

Determination of genetic alterations in brain tumors. Genetic alterations that have been detected in various brain tumors and that may be part of neurogenetic syndromes are shown in Table 2. The neurofibromatosis type 2 tumor suppressor gene on chromosome 22q is inactivated in 40% of sporadic meningiomas. A homologous gene, DAL-1 (differentially expressed in adenocarcinoma of the lung), located on chromosome 18p11.3 is lost in 60% of sporadic meningiomas. This has been demonstrated by use of reverse transcription polymerase chain reaction, western blot, and immunohistochemistry analyses.

Table 2. Molecular Methods for Detection of Genetic Alterations in Brain Tumors that Occur as Part of Genetic Syndromes

Tumor(s)
	Meningioma
		• Genetic alteration or molecular mechanism
			- Allelic loss of chromosome 22q in 50% of meningiomas; loss of alleles on chromosome 1p have been associated with malignant meningiomas
		• Method of detection
			- Polymerase chain reaction: single strand conformation polymorphism, direct DNA sequencing, and microsatellite polymorphism
		• Comments
			- Some meningiomas also have loss of chromosome 22q but intact neurofibromatosis type 2. It is possible that other tumor suppressor genes exist in 22q and may be involved in the pathogenesis of CNS tumors and meningiomas as a part of neurofibromatosis 1.
	Neurofibromatosis type 1; Von Recklinghausen disease; autosomal dominant disorder
		• Genetic alteration or molecular mechanism
			- Inactivation of gene neurofibromatosis type 1, which spans 350 kb of genomic DNA in chromosomal region 17q11.2; Neurofibromatosis type 1 is tumor suppressor gene
		• Method of detection
			- Gene isolated by positional cloning
		• Comments
			- The protein encoded by neurofibromatosis type 1, neurofibromin, has a domain homologous to GTPase activating protein family and down-regulates activity.
	Neurofibromatosis type 2; autosomal dominant disorder
		• Genetic alteration or molecular mechanism
			- Allelic losses on chromosome q22 -- Inactivation of gene neurofibromatosis type 2 occurs in most of sporadic schwannomas and meningiomas.
		• Method of detection
			- Gene was identified by positional cloning
		• Comments
			- The gene encodes the protein schwannomin (merlin).
	Li Fraumeni syndrome; autosomal dominant disorder
		• Genetic alteration or molecular mechanism
			- Germline mutation of p53 gene, chromosomal location 17p13
		• Method of detection
			- P53 sequencing and functional assays
		• Comments
			- P53 gene product is a nuclear phosphoprotein that can inhibit transit of cells from G1 to S in the cell cycle. Multiple cancers including gliomas
	Von Hippel-Lindau disease; autosomal dominant disorder
		• Genetic alteration or molecular mechanism
			- P53 gene product is a nuclear phosphoprotein that can inhibit transit of cells from G1 to S in the cell cycle. Multiple cancers including gliomas; chromosomal location is 3p25, but the gene has not been identified.
		• Method of detection
			- Not listed
		• Comments
			- Gene product elongin (transcription elongation); predisposition to development of several tumors including hemangioblastomas
	Tuberose sclerosis; autosomal dominant disorder
		• Genetic alteration or molecular mechanism
			- Linked to chromosome 9q in some families and to chromosome 15 in others, but the gene has not been identified.
		• Method of detection
			- Not listed
		• Comments
			- Characterized by skin lesions, mental retardation, and seizures
	Turcot syndrome; autosomal dominant disorder
		• Genetic alteration or molecular mechanism
			- Mutation of adenomatous polyposis coli gene plus mutant DNA mismatches repair genes
		• Method of detection
			- Adenomatous polyposis coli gene determined by peripheral blood leucocyte DNA test
		• Comments
			- Primary neuroepithelial CNS tumors in patients with polyposis coli. For details see MedLink Neurology article on Turcot syndrome.
	Basal cell nevus syndrome; autosomal dominant disease
		• Genetic alteration or molecular mechanism
			- Germline mutation of PTC, a tumor suppressor gene at 9q22.3-9q31. May be associated with medulloblastoma.
		• Method of detection
			- Single strand conformation polymorphism analysis
		• Comments
			- Multiple nevoid basal-cell carcinomas of the skin, skeletal anomalies, developmental malformations, and the predisposition to other neoplastic processes
	Multiple endocrine neoplasia type 1
		• Genetic alteration or molecular mechanism
			- Putative tumor suppressor locus on 11q13
		• Method of detection
			- Linkage analysis with restriction fragment length polymorphism markers
		• Comments
			- Characterized by parathyroid adenomas, pancreatic tumors, and pituitary tumors

Genetic alterations in primary malignant brain tumors are listed in Table 3. Some of these are discussed in the following text.

Table 3. Genetic Alterations in Primary Malignant Neoplasms of the CNS

Tumor(s)
	Malignant astrocytomas
		• Genetic alteration or molecular mechanism
			- Losses of alleles at loci on chromosomes 1, 9, 10, 13, 17, 19, and 22
			- Losses of p53 tumor suppressor genes
			- Deletion of p15, p16
			- Mdm2 amplification in 15% of malignant gliomas
			- Increased sensitivity of the growth factor receptors to endogenous growth factors: fibroblast growth factor, platelet derived growth factor, tumor growth factor, and epidermal growth factor
			- MMAC1-E1, a gene involved in the progression of glioma to its most malignant form
			- MAGE-E1, a glioma-specific member of MAGE family
			- NRP/B, nuclear-restricted protein/brain is expressed in neurons but not in astrocytes.
		• Method of detection
			- Restriction fragment length polymorphism markers, polymorphic DNA probes to detect loss of heterozygosity, florescent in situ hybridization, RT-polymerase chain reaction, serial analysis of gene expression
		• Comments
			- A variety of changes is associated with transition to malignant astrocytomas (see further discussion in text).
			- MMAC1 is also found to be associated with advanced cancers of the prostate, breast, kidney, and skin. MAGE-E1 expression is also significantly increased in glioblastoma relative to human astrocytes.
	Ependymomas
		• Genetic alteration or molecular mechanism
			- Loss of chromosome 17p in children and chromosome 22q in adults are the most frequent genetic abnormalities.
		• Method of detection
			- Loss of heterozygosity analysis.
		• Comments
			- Unlike astrocytic tumors, studies involving ependymomas are rare. These tumors usually lack the characteristic cytogenic abnormalities seen in astroglial tumors.
	Oligodendrogliomas
		• Genetic alteration or molecular mechanism
			- Loss of alleles on the long arm of chromosome 19.
		• Method of detection
			- Loss of heterozygosity analysis.
		• Comments
			- A tumor suppressor gene may be present, but mutations of p53 gene are found only rarely in oligodendrogliomas.
	Primitive neuroectodermal tumors, eg, medulloblastomas
		• Genetic alteration or molecular mechanism
			- Loss of alleles on chromosome 17p is common.
			- Partial gains of chromosome 1p, 6q, 11p, and 16q
			- Amplification of N-myc and bcl-2 oncogenes in a small percentage
		• Method of detection
			- Restriction fragment length polymorphism analysis
			- Flow cytometry, measurement of N-myc amplification
		• Comments
			- A tumor suppressor gene may be present, but mutations of p53 gene are found only rarely in medulloblastomas.
			- Expression of bcl-2 is associated with suppression of programmed cell death along with their neuronal differentiation from primitive cells
	Retinoblastoma
		• Genetic alteration or molecular mechanism
			- Rb-1 gene mutation, chromosomal location 13q14, causes the disease. Rb protein is tumor suppressor.
		• Method of detection
			- Not listed
		• Comments
			- 40% of these tumors are hereditary.

Meningiomas. Genetic biomarkers have been identified by molecular characterization of meningioma that can predict tumor behavior. A small number of genetic changes are known to classify more than 85% of all meningioma and clinical trials using targeted therapy to genetic subtypes of meningioma and immunologic biomarkers are ongoing to improve diagnosis, prognosis, and therapy (25).

Benign meningiomas can recur after surgical excision, but this is difficult to accurately predict. A study has examined tumor hypoxia-regulated biomarkers, preoperative imaging, measures of proliferation, and angiogenesis in predicting patient outcome (17). Results show that hypoxia-inducible factor-1alpha (HIF-1alpha), vascular endothelial growth factor (VEGF), and MIB-1 proliferation index are significantly correlated with tumor recurrence. With further study, these molecular biomarkers may be used to predict outcome for patients with intracranial meningiomas.

Malignant meningioma is defined as a tumor invading the brain or containing abundant mitoses or other malignant features that are not present in benign meningiomas. Mutations in the NF2 gene probably account for the formation of more than half of all meningiomas. Malignant change occurs in about 12% of meningiomas and is usually seen when the tumors recur after surgery. Malignant progression of meningiomas probably involves the inactivation of tumor suppressor genes on chromosomes 1p, 9p, 10q, and 14q.

Malignant astrocytoma. The identification of genetic alterations in astrocytomas may indicate a sequential course of events associated with enhanced malignancy. The most malignant form is anaplastic astrocytoma or glioblastoma. This is a genotypically heterogenous lesion. Molecular studies enable determination of 2 types of glioblastoma: (1) a primary type that develops de novo; and (2) a secondary type that progresses from low-grade astrocytoma. Not all genetic events associated with the development of this tumor have been identified. Some of the important alterations that have been identified include the following:

	• Molecular analyses of genetic alterations in astrocytomas have been carried out to identify pathways leading to glioblastoma. Glioblastomas with p53 alterations represent tumors that progress from lower grade astrocytomas. This variant is more likely to show loss of chromosome 17p than tumors without p53 alterations. Presence of p53 immunoreactivity is significantly associated with glioblastomas arising in younger patients. Patients with grade 2 astrocytomas with histological evidence of malignant degeneration have been found to have a higher level of p53 expression than those with no evidence of progression.
	• Studies with high-resolution comparative genomic hybridization show that the loss of chromosomes 10 and 9p and the gain of chromosomes 7 and 19 are the most frequent chromosomal alterations in glioblastoma.
	• Insulin-like growth factor binding protein 5 has been shown to be overexpressed in glioblastoma, contributing to the invasiveness of the tumor and correlating with histologic grade and survival of patients.

Advanced data mining and a novel bioinformatics were used with associative analysis to accurately identify ELTD1 (epidermal growth factor, latrophilin, and 7 transmembrane domain-containing 1 on chromosome 1) as a putative glioma-associated biomarker, which may serve as an additional biomarker for gliomas in preclinical and clinical diagnosis (31).

Attempts are being made to understand the molecular relationship of anaplastic astrocytoma (grade III) to its more aggressive counterpart, glioblastoma (grade IV), which has a higher degree of angiogenesis and is more invasive. Genome-scale mRNA expression data for human malignant gliomas has revealed several pathways that are significantly different between the 2 tumors. Hypoxia-inducible factor 1A/vascular endothelial growth factor network activation is a major contributor to the increased growth and invasion displayed by glioblastoma when compared to anaplastic astrocytoma. This information is useful for diagnosis, prognosis, and grading of malignant gliomas.

Diffuse low-grade and intermediate-grade gliomas, which together make up the lower-grade gliomas (World Health Organization grades II and III), have variable clinical course that cannot be adequately predicted from histologic class. Using exome sequence, DNA copy number, DNA methylation, messenger RNA expression, microRNA expression, and targeted protein expression for genome-wide analyses of lower-grade gliomas from adults led to the conclusion that most lower-grade gliomas without an IDH mutation were molecularly and clinically like glioblastoma (06). One of the characteristic genetic alterations in lower-grade gliomas is the isocitrate dehydrogenase-1 (IDH1) mutation. Integrated genomic analyses, including whole-exome sequencing, copy number, gene expression, and DNA methylation profiling, have enabled identification of oncogenic pathways driving the transformation of grade 2 and 3 gliomas into glioblastoma (02).

Cell‐free tumor DNA CSF from glioma patients is a screening method and patients with high levels of cell‐free tumor DNA can be investigated further by larger‐scale sequencing such as by whole‐exome and whole‐genome sequencing (22). This approach was shown to detect genetic changes that were missed in some of the tumor tissue samples.

Primary glioblastoma frequently shows EGFR, PDGFRA, PTEN, TP53, NF1, and CDKN2A/B as well as TERT promoter mutations, but not IDH mutations, which are frequent in grades II and III astrocytomas, secondary glioblastoma, oligodendrogliomas, and oligoastrocytomas (01). The survival benefit associated with surgical resection differs based on IDH1 genotype in malignant astrocytic gliomas, as better prognosis is observed in the IDH1 mutant subgroup following maximal surgical resection (03). Amplification of receptor kinases such as EGFR (epithelial growth factor receptor) and PDGFRA (platelet-derived growth factor receptor alpha) are relevant to the prognosis of glioblastoma, and both may coexist in a tumor. Simultaneous inhibition of both EGFR and PDGFR is necessary for abrogation of kinase pathway activity in glioblastomas with mixed population (29). In view of the cytogenetic heterogeneity of glioblastoma, stratification for prognosis should also be based on combined assessment of cytogenetic alterations involving chromosomes 7, 9, and 10--the most frequent alteration (08).

Intratumor heterogeneity of glioblastoma is likely the key to understanding treatment failure, and an integrated genomic analysis of spatially distinct tumor fragments has been developed to uncover extensive intratumor heterogeneity (28). Phylogeny of the fragments for each patient was reconstructed, and results of the study revealed patient-specific patterns of cancer evolution to enable more effective personalized treatment design. Use of clustered regularly interspaced short palindromic repeats associated protein-9 (CRISPR-Cas9) approach has revealed mechanisms of temozolomide resistance beyond genome analysis of glioblastoma and detected genetic changes that provide a better understanding of growth and treatment resistance driven by a small subpopulation of glioblastoma stem cells (21).

Ependymoma. The mechanism of formation of ependymomas has not been elucidated. Amplification and overexpression of mdm2 may be one of the major molecular events occurring in the tumorigenesis of ependymomas.

Oligodendroglioma. Genetic alterations associated with these tumors are distinct from those found in astrocytic tumors. Mutations in CIC gene (homolog of the Drosophila gene capicua) on chromosome 19q and the FUBP1 gene encoding far-upstream element binding protein on chromosome 1p have been shown to contribute to human oligodendroglioma (04). Clinical differences in tumors with and without the -1p/-19q genotype support a genetic approach to aid diagnosis and prognosis for oligodendroglial neoplasms. Anaplastic oligodendrogliomas or oligoastrocytomas lacking 1p and 19q alleles are less aggressive and more responsive to treatment. The specific chromosomal change in oligodendroglial brain tumors is, thus, associated with good prognosis and may also identify patients who would benefit from chemotherapy treatment in addition to radiotherapy at diagnosis for long-term tumor control. Molecular analysis of tumors has shown that tumors with combined 1p and 19q loss, which are histopathologically diagnosed as anaplastic oligodendroglioma, are more frequently located in the frontal lobe and have a better outcome. Anaplastic oligodendroglial tumors with EGFR amplification are more often localized outside the frontal lobe and have a survival like that for glioblastoma. The findings could change the future of how brain cancers are diagnosed and how treatments are personalized, based on genetic makeup of the tumor.

Primitive neuroectodermal tumors. The most common chromosomal alteration is loss of loci on chromosome 17. Medulloblastomas are primitive neuroectodermal tumors that are usually present in childhood. They may occur as part of 2 inherited cancer syndromes: (1) Turcot syndrome, and (2) basal cell nevus syndrome. Microarrays have enabled the separation of medulloblastomas from morphologically identical supratentorial primitive neuroectodermal tumors. A study of genetic alterations in medulloblastoma using high-density microarrays and sequencing has revealed inactivating mutations of the histone-lysine N-methyltransferase genes MLL2 or MLL3 in 16% of the patients (23). This demonstrates an important mechanism underlying medulloblastoma and differences from adult cancers.

Response to therapy of patients with medulloblastomas is difficult to predict. Several studies have shown the clinical outcome of children with medulloblastomas is highly predictable on basis of the gene expression profiles of their tumors at diagnosis. Integrated genomic approach has enabled identification of medulloblastoma subtypes with distinct genetic profiles, pathway signatures, and clinicopathological features, which will enable a better selection and evaluation of patients in clinical trials and will support the development of molecular targeted therapies.

Use of molecular diagnostics to determine the cells of origin of a brain tumor. This is illustrated by the example of primary lymphomas of the central nervous system.

Inherent limitations of conventional cytology often result in a failure to diagnose lymphomatous meningitis in cerebrospinal fluid specimens from patients who have the disease. Polymerase chain reaction techniques enable the diagnosis of lymphoma based on the detection of clonal rearrangements of the immunoglobulin or T-cell receptor genes. Frequent p14 gene abnormalities and inactivation in primary central nervous system lymphomas are in striking contrast to the same pathological subtype of systemic lymphoma in which p14 gene abnormalities and inactivation are infrequent.

Contraindications

No contraindications are known.

Results

The potential advantages of developing molecular testing for brain tumors are as follows:

	• These methods have increased our understanding of the biology of brain tumors to enable an improved classification and new approaches to treatment. It may be possible to determine the cells of origin of some tumors.
	• Targets for alternative therapies for brain tumors will include defective tumor suppressor genes. A gene therapy strategy may be used to correct or replace the defective gene.
	• Diagnosis may be possible with a small amount of biopsy tissue or an archival tissue specimen.
	• Noninvasive diagnosis may be possible by detection of circulating tumor DNA in blood and CSF.
	• Genotyping of brain tumors may have an application in stratifying patients for clinical trials of various novel therapies.
	• Molecular diagnostic approaches enable personalized management of glioblastoma. They can be useful for determining prognosis.

The alteration of multiple networking genes by recurrent chromosomal aberrations in gliomas deregulates critical signaling pathways through multiple, cooperative mechanisms. These mutations, which are likely due to nonrandom selection of a distinct genetic landscape during gliomagenesis, are associated with patient prognosis.

Glioblastomas often have both monosomy of chromosome 10 and gains of the epidermal growth factor receptor (EGFR) gene locus on chromosome 7. Chromosome 10 losses that decrease tumor suppressor gene ANXA7 levels correspond to a rise in EGFR levels that increase tumor aggressiveness and decrease survival times. This provides a clinically relevant mechanism to augment EGFR signaling in glioblastomas beyond that resulting from amplification of the EGFR gene.

Some limitations of molecular approaches for diagnosis of brain tumors are as follows:

	• A subset of particularly aggressive high-grade gliomas exists, with no currently known molecular genetic abnormalities.
	• Clinical correlation of whole genome analysis from microdissected specimens has shown that grade II to grade III tumors, unlike glioblastoma, may contain genetically distinct subgroups with different sensitivity to the therapies (13).
	• In some studies, the overexpression of p53 does not correlate with survival. The role that this gene plays in the development and progression of gliomas remains unclear.

Molecular diagnostics provides some assessment of prognosis. Some examples are as follows:

	• Amplification of N-myc gene in neuroblastoma, particularly those forms that are highly malignant.
	• P53 may be a prognostic biomarker for glioblastomas.
	• Loss of heterozygosity for chromosomes 1p and 19q correlates with both response to therapy and improved prognosis in patients with oligodendrogliomas.

Adverse effects

There are no adverse effects of the diagnostic procedures described.

Scientific basis

• The rationale for the application of molecular techniques for diagnosis of brain tumors is based on genetic alterations in malignancy.

An understanding of the biological basis of cancer is necessary for application of molecular techniques to the detection of malignancy. Cancer is a genetic disease. Modern concepts of cancer are based on the evidence that a tumor arises from clonal progeny of a single transformed cell. This cell has multiple sites of DNA damage, both acquired and inherited, that are passed on to the progeny. These genetic changes disrupt the regulation of cell proliferation and evolution of tumor cell clones with neoplastic progression; therefore, analysis of tumors for clonal markers can provide a specific diagnosis as well as a method for following the progression and remission of the disease. The concept of tumor suppressor gene evolved from studies on mutations of the retinoblastoma susceptibility gene.

Recognition of cancer as a disorder of genes has opened the possibility of classifying tumors according to genetic alterations that underlie their pathogenesis and regulate their malignant behavior. The genetic alterations reported most often in malignant brain tumors are mutations or alterations of a tumor suppressor gene and a loss of heterozygosity for certain chromosomes. Growth factors and amplification of oncogenes are involved as well. Tumor suppressor genes regulate cell growth by counteracting the action of proto-oncogenes. Potential sites where these genes might inhibit the development of cancer include cell proliferation, differentiation, and senescence; cell to cell communication; and chromosomal stability. P53, named for its 53-kd mass, is the best known of tumor suppressor genes. The p53 gene encodes a nuclear protein that binds to and modulates the expression of genes important for DNA repair, cell division, and cell death by apoptosis. Mutations of p53 gene are the single most common genetic alteration observed in human cancers; a mutant p53 gene has been detected in nearly half of human cancers. The p53 is located on chromosome 17p13.1, which is 1 of the more frequent targets for chromosome alterations in human cancers. It is not required for normal development, but lack of p53 function raises the risk of cancer enormously. Functional single nucleotide polymorphisms in codon 72 of TP53 are a risk factor as well as a prognostic biomarker and are particularly critical for the development of glioblastoma in young patients.

P53 regulates the expression of p21 (also called WAF-1, SDI-1, and CIP-1) and p27 (also called KIP-I), and this is a link between tumor suppressor genes and cell cycle regulation. Sequential activation of C-type cyclin-dependent kinase complexes is responsible for orderly transitions through cell cycles. Another class of genes, p16 (also called MTS-1), encodes a protein that can inhibit some D-type cyclin-dependent kinases.

PTEN is another tumor suppressor gene. Ablation of PTEN tumor suppressor gene function and mutant epithelial growth receptor function gene in the adult mouse induces rapid onset of an infiltrating, high-grade malignant glioma phenotype with prominent pathological and molecular resemblance to glioblastoma in humans. A significant number of proto-oncogenes encode either a growth factor or a receptor for a growth factor (a receptor tyrosine kinase). Members of these 2 families play an important role in signal transduction processes that operate during development of the nervous system. Receptors for the following growth factors have been implicated in the development of human brain tumors: epidermal growth factor, platelet derived growth factor, fibroblast growth factor, and insulin-like growth factor 1.

Telomerase is the ribonucleoprotein that adds new repeats to ends of chromosomes and prevents shortening of telomeres with each replication cycle and, thus, allows cells to escape from the proliferative limitation of cellular senescence. Telomerase is abnormally reactivated in all major cancer types but is absent in most normal tissues. Telomerase activity can be analyzed using the telomeric repeat amplification protocol-hybridization protection assay, reverse transcription-polymerase chain reaction, and Southern blot analysis.