Molecular alterations of low-grade gliomas in young patients: Strategies and platforms for routine evaluation

Abstract

In recent years, it has been established that molecular biology of pediatric low-grade gliomas (PLGGs) is entirely distinct from adults. The majority of the circumscribed pediatric gliomas are driven by mitogen-activated protein kinase (MAPK) pathway, which has yielded important diagnostic, prognostic, and therapeutic biomarkers. Further, the Consortium to Inform Molecular and Practical Approaches to CNS Tumor Taxonomy (cIMPACT) Steering Committee in their fourth meeting, suggested including a panel of molecular markers for integrated diagnosis in “pediatric-type” diffuse gliomas. However, a designated set of platforms for the evaluation of these alterations has yet not been mentioned for easier implementation in routine molecular diagnostics. Herein, we have reviewed the relevance of analyzing these markers and discussed the strategies and platforms best apposite for clinical laboratories.

Pediatric low-grade gliomas (PLGGs) are a heterogeneous set of entities and the most common form of primary central nervous system (CNS) tumor, accounting for over 30% in childhood.^1,2 They have long been recognized as distinct from those arising in older adolescents and adults, both in their pathological characteristics and biological behavior.³ PLGGs include glial, glioneuronal, and neuronal tumors: pilocytic astrocytoma (PA), dysembryoplastic neuroepithelial tumor (DNET), ganglioglioma (GG), pleomorphic xanthoastrocytoma (PXA), subependymal giant cell astrocytoma (SEGA), diffuse astrocytoma (DA), diffuse oligodendroglial tumors (d-OT), and other rare tumors.⁴

Although PLGGs have an excellent prognosis, they present a significant challenge to the treating oncologist due to various comorbidities associated with tumors or to their treatment. Also, a substantial proportion of cases, especially with subtotal resection (STR) show multiple recurrences.⁵ Clinical outcomes can vary by histology as well as patient’s age, sex, and tumor location.^6,7 Owing to their heterogeneous nature as well as the erraticism of clinical outcome, there is a formidable need for classification of PLGGs on an integrated histomolecular basis for accurate diagnosis, prognostication, and future clinical trials.^5,8 Recent large-scale genomic analysis has highlighted that most PLGGs harbor alterations in major oncogenic pathways like the mitogen-activated protein kinase (MAPK) pathway and PI3K/AKT/mTOR pathway.⁹ MAPK pathway upregulation via the myriad of alterations (fusions and mutations) is the most important underlying event in tumorigenesis and hence PLGGs are now being considered as a “one pathway disease.”

The KIAA1549-BRAF fusion and BRAF V600E mutation are the most frequent (>70%) genetic alterations in circumscribed, slow-growing tumors.^8,10–12 Additionally, rarer alterations affecting RAS/MAPK signaling, including those involving FGFR1/2/3, NTRK2, RAF1, ALK, and ROS1 as well as non-RAS/MAPK alterations, such as MYB, MYBL1, IDH1, H3F3A mutations, and CDKN2A/B homozygous deletion have been identified in small numbers of cases.⁸ These findings have major diagnostic implications and would help expedite the use of better-targeted therapies.⁹ In addition, the prognostic role of these alterations has also been observed. Ryall et al in a landmark study on 1000 cases suggested a risk stratification of PLGGs by the type of molecular alteration: fusion or single-nucleotide variation (SNV)-driven. Cases with fusions were younger, had lower grade histology, few recurrences, while SNV-driven tumors were older, displayed higher grade histology, and had more recurrences.¹³ The Consortium to Inform Molecular and Practical Approaches for CNS Taxonomy (cIMPACT) Update 4 proposed that molecular analysis should be incorporated into histologically based tiered classification schema for all LGGs and glial-neuronal tumors.¹⁴ It has provided anecdotal evidence that WHO grade II pediatric-type diffuse gliomas (MYB-altered or MYBL1-altered) behave more like WHO grade I.¹⁴

However, the WHO classification of tumors does not specify a designated platform for the detection of such molecular alterations. In routine clinical/diagnostic settings, especially on a case-to-case basis, high-throughput gold standard platforms might be economically challenging or not available either due to lack of expertise or limited resources. Hence, in this review, we aim to encapsulate the diagnostic, prognostic, and predictive molecular biomarkers in PLGGs and consolidate the key advantages and challenges of various techniques which can be used for routine assessment, ranging from gold standard high-throughput techniques to robust economic methods.

Current Understanding of Molecular Alterations in PLGGs

BRAF V600E

BRAF is a constituent of the RAS/RAF/MEK/ERK pathway and has a crucial role in transmitting extracellular signals to the nucleus. In human cancers, over 90% of BRAF mutations harbor a single amino acid substitution of valine by glutamic acid at codon 600 (p.V600E), due to c.1799T>A base variant.¹⁵ The BRAF V600E acts as a phosphomimic in the MAPK signaling cascade, activating downstream transcription factors committed to cell differentiation, proliferation, apoptosis, and senescence.^15,16 This mutation has been reported in approximately 40%-80% of PXAs, 18%-33% of GGs, approximately 20%-25% of DNETs and 9% of PAs, and in rare cases of low-grade diffuse gliomas in children.^17,18 It has been reported in ~10% of glioblastomas (GBs), especially the epithelioid variant.^19,20

PLGGs with BRAF V600E mutation have been shown to have a worse overall survival (OS) and progression-free survival (PFS) relative to other PLGGs.^21,22 Furthermore, tumors that harbor BRAF V600E mutation in conjunction with CDKN2A homozygous deletion are more likely to transform to high-grade astrocytomas. The most common among LGGs which show such events include the transformation of PXAs into epithelioid GBs.²³

FDA-approved kinase inhibitor, PLX4032 (vemurafenib), has been launched and used effectively for the targeted treatment in metastatic melanomas.²⁴ A favorable response to BRAF inhibitors has also been documented in BRAF V600E-mutant PXAs, brainstem GGs, and pediatric GBs.^25–27

KIAA1549-BRAF

KIAA1549-BRAF fusion is the most common event identified in PLGGs.²⁸ It results from focal tandem duplication at locus 7q34, causing exclusion of BRAF’s N regulatory domain and thereby leads to constitutive upregulation of the RAS/MAPK pathway.¹² This alteration is dominantly seen in >70% PAs majorly located in the posterior fossa/cerebellum but seldom in the supratentorial region.^12,28 It is also a common feature of diffuse leptomeningeal glioneuronal tumor (DLGNT) and rarely recognized in GGs and 1p/19q co-deleted d-OTs. Studies have documented that the dominant fusion is 1-16/9-18 (49%), followed by 1-15/9-18 (35%), and 1-16/11-18 (8%), while two of these fusions have less frequent breakpoints: 1-15/11-18 (6%) and 1-17/10-18 (1%).^11,29,30 Among these fusions, 15-9 is found in midline location outside of the cerebellum (in hypothalamus, optic pathway, and brainstem) while 16-9, in few cases, is reported to be found in the cerebellum.³¹ Lin et al found that the presence of this fusion is linked with a relatively good prognosis among PLGGs, yet uncertain in the case of PAs.³⁰ Overall, the prognostic significance of these fusions needs multivariate analysis, since some studies claim to show a positive outcome, while others have reported no effect.^11,32 Other less common fusions of the BRAF gene include FAM131-BRAF, RNF130-BRAF, CLCN6-BRAF, MKRN1-BRAF, GNAI1-BRAF, and GTF2I-BRAF.^11,12,33

Tumors harboring this fusion are mostly well-circumscribed, amenable to gross total resection, and have excellent OS for instance in cases of PAs arising in the cerebellum.¹² Hawkins et al demonstrated that KIAA1549-BRAF fusion is an essential prognostic marker even in partially resected pediatric low grade astrocytomas (PLGAs) and not dependent on location, pathology, and age.³⁴ Presently, MEK inhibitors like selumetinib, trametinib (NCT03363217), cobimetinib (NCT02639546), and binimetinib (NCT02285439) are undergoing clinical trials for PLGGs (NF1-pLGG, KIAA1549-BRAF-fused) with evidence that fusions as these do not respond to first-generation RAF inhibitors which instead lead to paradoxical activation of MAPK pathway.¹² KIAA1549-BRAF fusion is a significant diagnostic, prognostic, and therapeutic marker in PLGGs.

MYB

MYB proto-oncogene protein (c-MYB) has a significant role in cell maturation, proliferation, can determine lineage fate, and exhibits proto-oncogenic functions in both human leukemia and solid tumors.^35,36 The MYB oncogene codes a nuclear DNA-binding leucine zipper transcription factor and experiences posttranslational modifications like ubiquitylation, phosphorylation, acetylation, sumoylation, thereby altering levels of DNA-binding or transcriptional activation.^12,37 Angiocentric gliomas (AGs) and pediatric MYB/MYBL1-altered DAs are associated with favorable prognosis.³⁸ Rearrangements of MYB or MYBL1 genes have been substantially detected in specific clinicopathological subgroups of IDH-wt/H3-wt diffuse gliomas from a large pediatric-based cohort thereby strictly indicating the need of instilling their diagnostic testing.¹⁴ In the last decade, noteworthy findings have revealed that MYB alterations are identified in higher frequencies in AGs (87%) as well as pediatric DAs (41%).³⁹ The most common fusion partner QKI also has oncogenic implications in various cancers including gallbladder, prostate cancer, gastric cancer, and balanced reciprocal translocations reported in few patients with T-cell acute lymphoblastic leukaemia (ALL).³⁹ Other fusion partners detected in the rest of the PLGGs include MMP16, PCDHGA1, and MAML2.¹² Targeting MYB with small molecule inhibitors is tricky as of now but transcriptional targets of MYB-QKI-like KIT and/or CDK6 can be approached for targeted therapy.^39,40 MYB-QKI and its connection with H3K27ac enhancer elements also generates an ambit of therapeutic inhibition by incorporating BET bromodomain or CDK7 inhibition in an indirect fashion.^41,42

MYBL1

MYBL1 (MYB proto-oncogene-like 1) is an allied family member of MYB and regulates transcription which results in proliferation and differentiation MYB and MYBL1 alterations mostly occur in pediatric tumors with a median age of 5 years, they are majorly enriched in the cerebral hemispheres and infrequent in diencephalon or brainstem.^8,43,44 MYBL1-driven tumors depicting a partial duplication with truncation of its C-terminal regulatory were first described by Ramkissoon et al in 28% (5/18) of DAs.^45,46 In a study from St. Jude Children’s Research Hospital on 46 gliomas, it was established by t-distributed stochastic neighbor embedding (t-SNE) analysis that MYB/MYBL1-altered gliomas formed a single cluster, regardless of the histopathologic diagnosis of anatomic location. The 10-year PFS and OS rates were 89.6% and 95.2%, respectively.⁴³

FGFR gene family alterations

Fibroblast growth factor receptors (FGFRs) are transmembrane receptor tyrosine kinases with a poignant role in glioma tumorigenesis.³³ The FGFR1 gene on chromosome 8p11.23 has emerged as a recurrently altered oncogene in a diverse spectrum of primary glial and glioneuronal tumors.^5,12,33,47 FGFR1 alterations are known to occur either as FGFR1 mutations, FGFR1-TACC fusions, or FGFR1-TKD.^5,33

FGFR1 hotspot (N546 and K656) mutations have been identified in approximately 20%-30% of DNETs and 5% of PAs and rarely are germline events.^11,33,47,48 An association between midline location and a higher frequency of these mutations has been noted.¹² FGFR1-TACC fusions are most commonly seen in extra ventricular neurocytoma.⁴⁹ FGFR1-TKD duplication has been observed in low-grade astrocytomas (including PA), mostly in an extracerebellar region, and in DNET.⁴⁷ All the above alterations cause RAS/MAPK pathway upregulation as a result of FGFR1 autophosphorylation.^5,33 FGFR inhibitors (AZ4547, dovitinib, PD173074, ponatinib) are considered proficient in decreasing the growth of pediatric diffuse midline glioma, H3K27M-mutant cells in vitro relative to temozolomide.⁵⁰ Efficient programming of clinical trials will rely on the improved classification of tumors with these FGFR alterations.^51,52

None of the FGFR alterations are restricted to PLGGs and have rarely been identified in DAs, d-OTs, and other adult tumors.⁴⁶ Thus, broader genomic testing (loss of CDKN2A/2B, TERT promoter mutation, H3K27M mutation, etc.) should be strongly considered in cases with the presence of any atypical features.⁵¹

Other Rare Alterations

KRAS mutation

—A small percentage (1%-5%) of LGGs harbor mutations in KRAS. Among PLGGs, most of these mutations have been identified in PAs.⁵³ Although not diagnostic, identification of these mutations may pave the path for targeted therapies, given the success of targeted therapy against KRAS in other cancer.⁸

PIK3CA mutations.

—Among low-grade gliomas, PIK3CA mutations in exon 9 (E545K, c.1633G>A), have been exclusively found in Rosette-forming glioneuronal tumors (RGNT) and DNETs.⁵⁴ It is known to be a hallmark mutation in the histologic classification of such mixed histology and may probably contribute to therapeutic course of action.⁵⁴

PTPN11 mutation.

—A tyrosine phosphatase adaptor protein of the RAS/MAPK pathway, mutation of which is associated with Noonan Syndrome.⁵⁵ These mutations have been found in conjunction with FGFR1 alterations in PLGGs and also activate PI3K/AKT/mTOR pathway.¹¹ Hence, it has been postulated that both mTOR inhibitors and MAPK pathway inhibitors might work in such tumors.

NTRK fusions.

—NTRK (neurotrophic tyrosine receptor kinase) fusions are rare and result in constitutional activation of both PI3K/AKT/mTOR as well as RAS/MAPK pathways due to aberrant dimerization of its kinase domain. NTRK fusions with fusion partners SLMAP, QKI, NACC2 have been identified in PAs as well as in other PLGGs.^11,46

ALK fusions.

—Fusion events of anaplastic lymphoma kinase (ALK) gene result in constitutive expression of ALK resulting in upregulation of PI3K/AKT/mTOR as well as RAS/MAPK pathways. ALK alterations have rarely been reported in gliomas, most common of which include CCDC88A-ALK and PPP1CB-ALK.^8,56,57

ROS1 fusions.

—ROS1 is an orphan receptor tyrosine kinase akin to ALK1, and its fusions result in activation of the RAS/MAPK pathway as well as JAK/STAT and PI3K pathways. GOPC-ROS1 is the most common among ROS1 fusions and has been detected in GB, while KLC1-ROS1 in PLGGs.⁵⁸

Although alterations in ALK, ROS1, and NTRK are rare, the agents for targeted therapy are already developed, approved, and are available, due to their higher prevalence in other adult malignancies.^8,59 ALK, ROS, and NTRK have also been identified in a subset of infant gliomas which the authors termed as hemispheric RTK-driven.⁵⁷

CRAF fusions.

—CRAF or RAF1, a serine/threonine kinase, and is a principal component of the MAPK pathway. Reported fusion partners include QKI and SRGAP3, which result in oncogenic fusions, which activate both MAPK and PI3K signaling pathways. It has been rarely detected in PLGGs.²⁹

Secondary Alteration in PLGGs (CDKN2A Deletion)

Homozygous deletion of CDKN2A has been observed in a small fraction of PLGGs.⁶⁰ Tumors especially PXA harboring both BRAF V600E and CDKN2A deletion have worse OS and PFS and are prone to transformation into secondary high-grade gliomas.^22,60,61 Co-occurrence of CDKN2A deletion with BRAF fusions has been described in anaplastic astrocytoma with piloid features.⁶² Thus, the presence of CDKN2A warrants close clinical follow-up.

Strategies for Assessment of Molecular Alterations in PLGGs

Currently, a wide array of clinically certified laboratory methods are used to molecularly profile PLGGs. To provide the best patient care possible, it is important to carefully select the assays used as well as to monitor their validity and accuracy. However, no “gold standard” exists for testing the potential molecular events and various strategies may be used depending on tissue quality/quantity and budget. Majority of PLGGs harbor a single driver alteration, which may be a point mutation, gene fusion, or insertion/deletion. We suggest four strategies to identify these alterations: (i) sequential or parallel single gene testing using immunohistochemistry (IHC), Fluorescent in-situ hybridization (FISH) assay, Sanger sequencing, droplet digital PCR (ddPCR), or RT-PCR-based assay in resource restrained laboratories for detection of most of the alterations; (ii) upfront tumor next-generation sequencing (NGS) (targeted gene panel); (iii) supplement single gene testing with NanoString assay and or targeted sequencing; (iv) DNA methylation array. The selection of approach may depend on accessibility to a platform, resource constraints, quantity of tissue, and turnaround time. Most reliable, easy, and economical techniques best suited for diagnostic and prognostic assessment may be implemented.

Molecular characterization of almost two-thirds of PLGGs can be performed by simple and robust tests to detect BRAF fusions (FISH, RT-PCR) and BRAF V600E mutation (IHC, Sanger sequencing, RT-PCR)¹³ BRAF V600E-mutant gliomas like PXA or PAs with atypical features should be assessed for CDKN2A deletion.⁶⁰ IHC for IDHR132H, ATRX, H3K27M, and H3G34R/V mutations should be performed in cases with DA- or d-OTs-like histology.^12,13 KIAA1549-BRAF fusion and BRAF V600E-negative PAs and glioneuronal tumors can be assessed by NanoString assay as a majority of the cases harbor gene fusions.¹³ The remaining cases can be processed by targeted sequencing for the detection of other known SNVs.

Common Testing Platforms Used for Molecular Profiling of PLGGs

IHC is available to detect BRAF V600E mutation, MYB alterations, and FGFR1 mutations.^36,39,47,63 The IHC for BRAF V600E is done using clone VE1 (catalog no. 790-4855; 1/100 titer; clone VE1) which can be performed on Ventana BenchMark Ultra platform (Ventana Medical Systems, Inc., Tucson, AZ, USA).⁶³ The sensitivity and specificity of this technique are reported to be more than 94% in determining BRAF V600E mutation in GB.⁶³ The antibody designed for FGFR1 (cat. no. BS5569; 1:500, polyclonal rabbit anti-FGFR1; Bioworld, St Louis Park, MN, USA) can also be used by the Ventana platform.⁶⁴ Lehtinen et al have suggested that immunohistochemical data of FGFR1 are a potential prognostic marker in PAs and might have a role as a therapeutic target.⁶⁵ MYB’s monoclonal antibody is designed for its N-terminal region (Abcam EP769Y, Cambridge, MA, USA).³⁷ There is prevailing evidence that the sensitivity and specificity of MYB IHC are 82% and 86% in adenoid cystic carcinoma, however, its sensitivity and specificity have not been discussed in low-grade gliomas as per the latest resources.³⁷

Advantages of IHC include quick turnaround time, use of little material, low-price, relative ease of deployment and is useful for specimens with scattered tumor cells intermingled with abundant nonneoplastic cells.²⁷ However, standardization of IHC relies on certain subjective parameters like tissue preservation, fixation, endogenous peroxidase activity, temperature, and primary and secondary antibodies.⁶³ Finally, no IHC marker is 100% sensitive or specific, and their reliability can be compromised by necrosis and thermal cautery artifacts.⁶⁶

FISH assay is widely available and allows for visualization of both gene fusions and copy number events and can be used on formalin-fixed paraffin-embedded (FFPE) material but is relatively labor-intensive, and can only test for a single alteration at a time (or if multiplexed, a handful of loci). It cannot fully cover whole-arm losses or gains (occasionally resulting in false positives).⁶⁶ In PLGGs, FISH has been used in the identification of BRAF, FGFR1, ALK/ROS1/NTRK1/2/3 fusions, and MYB/MYBL1 alterations.^{12,32,37,45,47,57,67} It can also be used for identifying CDKN2A deletions. FISH may be complicated when targeted genes have multiple fusion partners and/or multiple breakpoints (eg, BRAF).⁶⁸ The interpretation needs an experienced pathologist or cytogeneticist.⁶⁹ Nevertheless, FISH can be used as the first line of screening and the doubtful cases may further be reconfirmed by NanoString assay.⁸

Sanger sequencing is considered as the gold standard because of its cost-efficiency, easy data analysis, and feasibility with FFPE samples.⁷⁰ The platform has been used for the assessment of BRAF V600E and FGFR mutations.^12,46,69 However, it is unable to detect mutations with <10%-20% mutant alleles because of its lower analytic sensitivity/higher threshold.^27,69 Additionally, in GGs and other glioneuronal tumors, it has been reported that Sanger sequencing fails if there is a low percentage of mutated ganglion cells, but such cases can be detected by allele-specific quantitative PCR and IHC.²⁷

Droplet digital PCR (ddPCR) is a powerful single-molecule counting strategy to identify trace (850 pg) amounts of genetic material. The platform can measure low amplitude copy number variations (CNVs) as well as perform accurate counting of alleles from DNA isolated from a mixture of heterogeneous cell populations.^28,71 Appay et al demonstrated 100% sensitivity and specificity of ddPCR to assess the KIAA1549-BRAF fusion from very low amounts of DNA isolated from FFPE specimens using RNA sequencing as a gold standard.²⁸ The authors concluded that evidence for BRAF duplication is both necessary and sufficient to predict KIAA1549-BRAF fusion. ddPCR has also been used for the detection of several molecular alterations in different diseases including the detection of the internal tandem duplication of the tyrosine kinase domain of the FGFR1 gene in DNETs.⁷² It should be incorporated in routine practice to save time, money, and precious tissue, barring the demerit that it is difficult to multiplex and can only evaluate a fixed number of genetic alterations.²⁸

Real-time PCR is one of the most widely practiced techniques, conveniently followed in clinical labs with the provision of good reproducibility. It exhibits an analytic sensitivity of detecting 5% mutant alleles or less, is quick as well as cost-effective, and displays 98%-100% concordance in genotype scoring with Sanger sequencing results.⁶⁷ The RT-PCR is increasingly being implemented over or alongside FISH in a diagnostic setting especially for assessment of KIAA1549-BRAF fusion.⁷³ It is fast, with a sensitivity and specificity of >90% on FFPE tissue, delivers direct interpretation, and can detect the fusion of distinct exons.⁷³ The only drawback is that it can target only one or a few particular alterations per reaction and thus is a bit cumbersome technique.⁷⁴

NanoString nCounter system is an FDA-approved platform capable of multiplexing large numbers of genes, needs no library preparation, is functional with both frozen and FFPE preserved samples, and has shown compatibility with even degraded RNA samples.^75,76 It is preferable as it exempts from the need for complex result interpretation, permits a single sample comparative analysis, and is very strict with its cutoffs. It takes 48 hours to complete and costs significantly less than other methods of fusion detection.^8,77 Ryall et al have shown the robust performance of NanoString assay for detection of fusions in PLGGs including the rare NTRK and CRAF fusions. The NanoString assay showed a sensitivity and specificity of 97% and 98%, respectively.⁸ However, input needs are quite high (200-500 ng) and can be executed in a fixed number of batches.⁸ In another study on infant gliomas, fusion targets for ALK, ROS, and NTRK were also included in the NanoString panel and showed similar results.⁵⁷

Next-generation sequencing platform (NGS) is the most advanced technology because of its enhanced analytic sensitivity to detect specimens with 1%-2% mutations, comprehensive genomic coverage limit of detection, large data storage facilities, fast turnaround time for high sample volumes, and ability to sequence hundreds to thousands of genes or gene regions simultaneously.^13,78 Several institutions are working to incorporate next-generation sequencing into the care of children with brain tumors. Four recent studies using targeted sequencing highlighted the potential benefits of incorporating NGS into clinical “standard of care” for pediatric glioma patients.^79,80,81,82 The data can be utilized to refine tumor diagnosis, allow for patient-specific risk stratification, and identification of therapeutic targets.⁷⁸ The integration of expanded NGS in clinical practice however needs ensuring adequate tissue sampling for both histopathologic and molecular analyses and utilizing tissue preservation protocols. Most testing requires 2-4 weeks for the completion of results. Further, the availability of computational resources and technical expertise to analyses and clinically interpret the data poses additional hurdles. Ethical guidelines have to be followed when germline mutations like NF1 unravel in about 10%-15% of low-grade gliomas.¹²

DNA methylation profiling has emerged as a powerful diagnostic tool for the explicit determination of CNS tumors with differing morphological or molecular aspects and has been implemented in several diagnostic institutions worldwide. DNA methylation profiling has still not been accredited by WHO for diagnostic relevance in PLGGs, nevertheless, its ambit as a potential classification tool is evident. This particular technique has provided clarity in distinguishing several low-grade gliomas and glioneuronal tumors, known to showcase overlapping features like GGs, PAs, and PXAs.⁸³ For instance, PAs contain a distinct hypomethylation signature that lacks in DAs and other brain tumors.⁸⁴ Fukuoka et al suggested that histopathological and genetic analysis may play a preliminary role in classification of most PLGGs.⁸⁵ However, methylation arrays still hold a pertinent role as a classifier in exceptional cases wherein histopathology and economical molecular platforms are inept in establishing a diagnosis.⁸⁶ This method can be practiced with FFPE, low input bisulfite DNA (20-50 ng), however, low pH and high temperature needed in the conversion process can cause 90% degradation of DNA.^12,86 The technique is not feasible in case of inadequate biopsy tissue or poor DNA yields.⁸⁷ It is subjected to batch effect, requires expensive equipment and trained manpower.⁸³ Table 1 summarizes the frequency and significance of various molecular alterations in PLGGs. An overview of various platforms used for the assessment of these alterations is also provided.

Table 1.

Summary of Platforms Used in Various Studies for Analysis of Molecular Alterations in PLGG

Molecular Alteration Significance	Frequency8,11,12,15,18,23,27,28,29,33,39,46	Significance12,20,21,24,25,33,39,45,46,50,51	Platforms
BRAF V600E mutation	PXAs (70%), GGs (18%-33%), DNTs (20%-25%), PA (9%)	• Diagnosis and prognosis (worse OS and PFS) in PXAs	(1)IHC15,62 • excellent sensitivity and specificity (>95%) • Spring Bioscience/FDA approved Ventana, VE1 antibody • Seen in neuronal cells in GG
		• BRAF inhibitors and its targeted response documented in V600E-mutant PXAs, brainstem GGs, and pediatric GBMs	(2)Sanger sequencing66,70 • gold standard in diagnostic laboratories • cost-efficient • can detect rare variants of BRAF V600 • unable to detect mutations with <10%-20% mutant alleles
			(3)RT-PCR73 • first companion BRAF test approved by FDA • Cobas 4800 BRAF V600 mutation test (RocheMolecular Diagnostics) • allele-specific quantitative PCR (ASQ-PCR) • 98%-100% concordance in genotype scoring with Sanger sequencing results
KIAA1549-BRAF fusion	PA (70%-80%), DA (5%-10%), Rosette forming glioneuronal tumor (20%-30%)	• Prognosis in PA • Clinical trials for BRAF and MAPK inhibitors	(1)RT-PCR67,73 • 97% sensitivity and 91% specificity • identical results in RT-PCR assay, 87% of patients with multiple biopsies • cannot detect multiple fusion events
			(2)ddPCR27 • identify trace (850 pg) amounts of genetic material • 100% sensitivity and sensitivity
			(3)NanoString assay8 sensitivity and specificity of 97% and 98%
			(4)FISH67,68,73 • cutoff: minimum 10%-15% of tumor cells • difficult to interpret due to close proximity of KIAA1549 and BRAF genes • amplification of 7q34 region can complicate interpretation
MYB alterations (MYB amplification, MYB 3′-truncating fusion)	Pediatric DAs (41%).	• Diagnosis • Prognosis (good)	(1)FISH37,39 • Homebrew probes RP11-63K22 (5′ to MYB) and RP11-170P19 (3′ to MYB) that map to 6q23.3 enable observation of different MYB statuses like disomy, MYB rearrangement, or 3′ MYB deletion • sensitivity (96%) and specificity (89%) in breast ACC (2)IHC37 • Positive MYB IHC in diffuse LGGs (60%), PA (41%), and HGGs (19%), but abnormalities at the genomic level were only a feature of diffuse gliomas
MYB-QKI fusion	Angiocentric glioma (87%)	• Diagnosis • Prognosis (good)	(1)FISH46 • Qaddoumi et al used iFISH profiles to check MYB-QKI fusion by using “break-apart” probe sets and depicted rearrangement/deletion of QKI
MYBL1 alterations	DAs (6%-14%)	• Prognosis (good)	(1)RT-PCR45 • detect MYBL1-trunc1 and MYBL1-trunc2 in samples of DAs • requirement of fusion-specific primers results in inability to detect rearrangements involving unknown fusion partners • formalin-fixed, paraffin-embedded tissues give false-negative results might be due to highly degraded RNA availability (2)FISH45 • Homebrew probes RP11-110J18 (5′ to MYBL1) and RP11-707M3 (3′ to MYBL1) that map to8q13.1 • MYBL1 centromeric breakpoint can be interpreted as duplication of 1 allele in more than 60% of nuclei • Confirmatory tool rather than screening tool
FGFR1 mutations	DNET (20%-30%, midline), PA (3%-5%), DA (2%-5%), oligodendroglioma (10%-20%), desmoplastic infantile astrocytoma and ganglioglioma (5%-10%), Rosette forming glioneuronal tumor (20%-30%)	• Poor prognosis in PAs • Ongoing clinical trials of FGFR inhibitors in brain tumors	(1)Sanger sequencing46 • Most common hotspot mutations: p.N546K and p.K656E (2)IHC47 • Phospho-ERK antibody CS-4376 (1:200) optimized in Ventana machine
FGFR1-TACC fusions	PA (3%-5%), oligodendroglioma (3%-5%), DNET (10%-15%)	Prognosis unclear in Pas	NanoString8 • FDA approved • multiplexing large numbers of genes • works with both frozen and FFPE • single-sample comparative analysis • very strict with cutoffs • sensitivity and specificity of >95% • input needs are quite high (200-500 ng)
			Targeted NGS78–81 • detects specimens with 1%-2% mutations • 2-4 weeks for completion of results • sequence hundreds to thousands of genes simultaneously • needs ensuring adequate tissue sampling • hurdles computational resources and technical expertise
FGFR1-tyrosine kinase domain duplication (FGFR1-TKDD)	Oligodendroglioma (10%-20%), DNET (20%-30%)
FGFR2 fusion (FGFR2-CTNNA3), FGFR3	Polymorphous low-grade neuroepithelial tumor of the young (PLNTY) (30%-40%)
Fusions	Multinodular and vacuolating neuronal tumor (MVNT) (3%-5%), FGFR2 fusion
ALK/ROS fusions	Isolated case reports in gliomas

Abbreviations: ACC, adenoid cystic carcinoma; DA, diffuse astrocytoma; ddPCR, droplet digital PCR; DNET, dysembryoplastic neuroepithelial tumor; DNT, dysembryoplastic epithelial tumor; FFPE, formalin-fixed paraffin-embedded; FGFR, fibroblast growth factor receptor; FISH, fluorescent in-situ hybridization; GBM, glioblastoma; GGs, gangliogliomas; HGGs, high-grade gliomas; IHC, immunohistochemistry; LGGs, low-grade gliomas; MAPK, mitogen-activated protein kinase; NGS, next-generation sequencing; OS, overall survival; PA, pilocytic astrocytoma; PFS, progression-free survival; PLGGs, pediatric low-grade gliomas; PXAs, pleomorphic xanthoastrocytomas; RT-PCR, reverse transcription polymerase chain reaction.

Conclusions

Current neuro-oncological practice is increasingly dependent on the integration of histopathologic data and genetic alterations. The availability of targeted inhibitors for most of the molecular alterations in PLGGs makes precision diagnostics need of the hour. It is not only crucial to detect the common molecular alterations but also the rare genetic mutations and fusions for proper enrollment in clinical trials. However, it is not possible for all laboratories especially in a developing country to utilize high-throughput expensive techniques in all cases. IHC, Sanger sequencing, FISH, and loss of heterozygosity (LOH) analysis are established techniques that can provide very valuable molecular information. Each of these methods, however, have their own merits and demerits. A laboratory may need to combine several techniques considering test cost and turnaround time to get the complete package of the required molecular information. This paper has made an effort to prime and discuss various platforms that can be applied for detecting the major alterations in PLGGs. Yet we have delineated the most robust, reliable, and cost-effective strategy for implementation in diagnostic settings. Any technique or strategy can be adopted by diagnostic laboratories as per their respective convenience and requirement to bring about the age of “integrated diagnosis.”

Funding

None declared.

Conflict of interest statement. None declared.