D-2-HG inhibits the α-KG dependent dioxygenase family including ten-eleven translocation (TET) DNA modifying enzymes and jumonji C domain containing (JmjC) histone lysine demethylase (KDMs) (Xu et al., 2011; Chowdhury et al., 2011). This leads to increased methylation of histone (Kohli and Zhang, 2013) and O⁶-methylguanine-DNA methyltransferase (MGMT) promoter methylation (Turcan et al., 2012). Resulting epigenetic modification in the form of elevated H3K4 trimethylation (H3K4me3) (Gunn et al., 2023) at the telomerase reverse transcriptase (TERT) promoter facilitates transcription factors like Myc-Max to bind and activate TERT expression by promoting a more open chromatin structure, ultimately contributing to increased telomerase activity and cellular immortalization (Han et al., 2020). TERT is the catalytic subunit of telomerase and plays a pivotal role in telomere maintenance. Aberrant TERT methylation patterns are associated with inadequate treatment response, increased risk of recurrence and poor prognosis in various cancers. The mechanisms of gliomagenesis through hypermethylation are still being explored but likely involve inappropriate methylation silencing of tumor suppressor genes such as cyclin-dependent kinase inhibitor 2A/B (CDKN2A/B), retinoblastoma-associated protein (RB), RASSF1A (Wolter et al., 2001; Alonso et al., 2003; Kuo et al., 2013; Unruh et al., 2019) and aberrant activation of pro-malignant genes like epidermal growth factor receptor (EGFR), tissue factor (F3), prominin 1 (PROM1) (Unruh et al., 2019; de Botton et al., 2023; Losman and Kaelin, 2013; Bhavya et al., 2020; Unruh et al., 2019) and the oncogene platelet-derived growth factor receptor alpha (PDGFRA) (Flavahan et al., 2016). However, the epigenetic state is fluid and IDH mutant gliomas can lose much of their DNA methylation during tumor progression (Mazor et al., 2015; de Souza et al., 2018; Ceccarelli et al., 2016). Genetically similar stem-like cell populations have been identified in IDH-mutant gliomas (Venteicher et al., 2017; Tirosh et al., 2016) supporting the notion that epigenetic reprogramming induced by IDH mutations contributes to a differentiation block (Lu et al., 2012).

D-2-HG produced by IDH mutant glioma cells is actively exported into the tumor microenvironment (Linninger et al., 2018) where it suppresses antitumor T-cell function (Bunse et al., 2018), thereby promoting intratumoral immunosuppression (Bunse et al., 2018; Notarangelo et al., 2022; Mellinghoff et al., 2023). IDH-mutant gliomas exhibit reduced infiltration by tumor-infiltrating lymphocytes compared to their IDH wildtype counterparts. Epigenetic silencing of NKG2D ligands helps escaping surveillance by natural killer cells (Zhang et al., 2016). Also, 2-D-HG affects dendritic cell maturation and antigen presentation by innate lymphoid cells adding to reduced capacity to activate T cells (Friedrich et al., 2023). Other effects of elevated D-2-HG levels include promotion of angiogenesis via Vascular Endothelial Growth Factor Receptor VEGFR2 signaling, increased matrix metalloproteinase (MMP2) activity (Seok et al., 2019) and transformation of human astrocytes (Koivunen et al., 2012; Johannessen et al., 2016).

By interfering with normal cellular metabolic pathways, 2-D-HG depletes α-KG and NADPH from the Krebs cycle leading to disruption to redox balance (Badur et al., 2018) as evidenced by a ¹³C metabolic flux analysis (Grassian et al., 2014). Resulting accumulating oxidative damage is a hallmark of IDH mutant cancer biology (Han et al., 2020). The Krebs cycle is adjusted to compensate for these metabolic pathways and the loss of α-KG (Borodovsky et al., 2012) with the recruitment of several alternate non-Krebs-cycle carbohydrate sources (Maus and Peters, 2017; Ohka et al., 2014) including increased dependence on glutaminolysis for glutamate production (McBrayer et al., 2018; Seltzer et al., 2010). Glutamate dehydrogenase 2 is responsible for converting glutamate into α-KG and it is expressed at high levels in the brain. This enzyme mitigates the metabolic vulnerabilities associated with IDH mutations and IDH mutated glioma cells demonstrate increased sensitivity to glutaminase inhibition confirming the role of glutaminolysis as a key compensatory pathway in metabolic homoeostasis maintenance (Seltzer et al., 2010). However, this ‘glutamine addiction’ is not specific to IDH mutant gliomas and glutamine catabolism occurs a major metabolic change in many types of cancers (Lukey et al., 2013). This metabolic pathway is an attractive therapeutic target with various approaches being employed or under investigation to exploit the dependence of cancer cells on glutamine. These include depleting cancer cell glutamine supply (L-asparaginase in acute lymphoblastic leukemia), inhibiting cancer cell glutamine uptake (inhibition of the c-Myc-regulated transporter SLC1A5), use of glutamine mimetics (6-diazo-5-oxo-L-norleucine, acivicin, azaserine) and use of selective glutamines inhibitors (968 and BPTES) (Lukey et al., 2013). GLS inhibitor telaglenastat has shown promising results in preclinical studies and is currently being investigated in a clinical trial for patients with IDH mutant gliomas in combination with radiation therapy and temozolomide (NCT 2018). Due to its structural and functional similarity with glutamate, 2-D-HG contributes to seizures in patients with glioma (Reitman et al., 2011; Ruda et al., 2024).

IDH mutant glioma exhibit increased reliance on nicotinamide phosphoribotransferase (NAMPT) for nicotinamide adenine dinucleotide (NAD) biosynthesis (Tateishi et al., 2015). NAD depletion via concomitant NAMPT inhibition has been identified as a metabolic susceptibility of IDH1 mutant cancers and results in cytotoxicity triggered by autophagy (Tateishi et al., 2015). 2-D-HG also directly stimulates EgIN prolyl hydroxylase which in turn causes reduced HIF1α activity (Koivunen et al., 2012). HIF1α acts as a tumor suppressor in mouse models and its level are low in IDH mutant glioma (Blouw et al., 2003; Zagzag et al., 2000).

Ribosomal RNA (rRNA) 2’O-ribose methylation distinguishes IDH wildtype and mutant HGG. IDH wild-type glioblastomas display the most prominent defects in rRNA epitranscriptomics. The observed reduction in rRNA 2′-O-methylation (2′Ome) levels may be an indirect consequence of increased rRNA transcription linked to the heightened proliferative activity of glioblastoma cells. This upregulation in rRNA synthesis could potentially exceed the capacity of the rRNA 2′Ome modification machinery, leading to insufficient methylation. IDH mutant astrocytomas and oligodendrogliomas demonstrate elevated expression of ribosome biogenesis (RiBi) factors indicating that RiBi activity does not correlate directly with proliferative rate in HGG. These findings suggest that changes in rRNA epitranscriptomic modifications in HGG cannot be solely attributed to differences in ribosome biogenesis. IDH mutant glioma cells exhibit higher cytotoxicity to RNA polymerase I inhibitors, highlighting potential therapeutic implications (Paraqindes et al., 2023).

TP53 and alpha thalassemia X-linked intellectual disability (ATRX): Class-defining loss of function mutations in TP53 and ATRX genes are seen in approximately 90 and 70% of astrocytomas, respectively, (Cancer Genome Atlas Research Network, 2015; Jiao et al., 2012). ATRX mutations are associated with epigenomic dysregulation and telomerase dysfunction (Clynes et al., 2013) resulting in alternative lengthening of telomerase (Heaphy et al., 2011). Astrocytomas with Li-Fraumeni Syndrome/Germline TP53 mutations show remarkably selective occurrence of IDH1:c.394C > T p. R132C mutations (Watanabe et al., 2009).

CDKN2A/B: CDKN2A/B are tumor suppressor genes and their loss of expression contributes to malignant tumor progression by disrupting cell cycle regulation and enhancing cell proliferation (Sharpless, 2005). IDH mutant astrocytomas with CDKN2A/B homozygous deletions have significantly shorter survival (Yeo et al., 2023; Brat et al., 2020; Appay et al., 2019; Alentorn et al., 2015). Notably, the presence of CDKN2A/B loss denotes WHO Grade 4 disease independent of histopathological features (Louis et al., 2021). Similarly, hemizygous loss of CDKN2A/B is associated with reduced survival in IDH1/2 mutant astrocytomas (Ghosh et al., 2025). Both CDKN2A homozygous and hemizygous deletions are enriched in post-treatment, recurrent IDH mutant astrocytomas (Kocakavuk et al., 2023).

Other genetic alterations: Intrachromosomal and extrachromosomal focal amplifications drive tumor growth and are associated with negative prognosis in many cancers (Yi et al., 2022). Amplifications of CDK4/6, CCND2, PDGFRA, MYCN as well as RB1 mutation/homozygous deletion, MET alterations, PIK3R1/PIK3CA mutations, NOTCH1 mutations and TCF12 mutations are also associated with tumor progression & shorter survival in IDH mutant astrocytomas (Brat et al., 2020; Li et al., 2019; Broderick et al., 2004; Labreche et al., 2015; Kamoun et al., 2016; Aoki et al., 2018).

TERT: Oligodendrogliomas have loss of one copy of the entire short arm of chromosome 1 (1p) along with one copy of the long arm of chromosome 19 (19q) by definition and may harbor TERT activating promoter mutations (Miller et al., 2017; Koelsche et al., 2013; Lee et al., 2018) which are early and clonal events in gliomagenesis (Suzuki et al., 2015; Exner et al., 2019). TERT mutations present in the vast majority of oligodendroglioma and IDH wildtype glioblastoma are mutually exclusive with ATRX mutations and alternative lengthening of telomerase seen in astrocytoma (Cancer Genome Atlas Research Network, 2015; Eckel-Passow et al., 2015). TERT mutations lead to telomerase stabilization, cellular immortalization and proliferation (Hafezi and Perez Bercoff, 2020) and are characteristically absent in adolescents (Lee et al., 2018).

CIC: The CIC gene encoding the Protein capicua homolog is localized to chromosome 19q and the remaining allele is inactivated in 70% tumors (Eckel-Passow et al., 2015; Bettegowda et al., 2011). CIC represses receptor tyrosine kinase/MAPK pathway targets by its interaction with chromatin regulators (Wong and Yip, 2020) and its deletion promotes neural stem cell proliferation indicating a direct role in gliomagenesis (Yang et al., 2017).

CDKN2A/B: Similar to astrocytoma, CDKN2A/B homozygous deletion is an adverse prognostic factor in grade 3 oligodendroglioma (Fallon et al., 2004).

Other aberrations include FUBP1 somatic mutations in 20–30% (Sahm et al., 2012), combined CIC and FUBP1 mutations (Chan et al., 2014), NOTCH1 mutations in 15% and less frequent other NOTCH pathway mutations (Suzuki et al., 2015), SETD2, PIK3CA and SWI/SNF chromatin remodeling complex mutations (Cancer Genome Atlas Research Network, 2015). Patients with tumors harboring CIC or combined CIC and FUBP1 mutations as well as NOTCH1 mutations have shorter time to recurrence (Suzuki et al., 2015; Chan et al., 2014; Gleize et al., 2015). Other co-existing genetic alterations associated with tumor progression and shorter survival in oligodendroglioma include homozygous PIK3CA mutation (Broderick et al., 2004; Tateishi et al., 2019), TCF12 mutation (Labreche et al., 2015) and increased MYC signaling (Kamoun et al., 2016).

The DNA methylation profiling clearly distinguishes IDH mutant gliomas from their wild type counterparts by the glioma CpG island methylator phenotype (G-CIMP) (Turcan et al., 2012; Wiestler et al., 2014; Noushmehr et al., 2010). G-CIMP-high (IDH mutant) and G-CIMP-low (IDH wild type) signatures are determined based on the extent of hypermethylation and the latter confers a worse prognosis (Ceccarelli et al., 2016).

DNA methylation profiling is a method used to differentiate cancer cells from normal tissue. Consistent with the finding of a shared histogenesis and developmental hierarchy, DNA methylation also shows high similarity between three groups compared to IDH wildtype glioma and IDH mutant non-glioma tumors (Venteicher et al., 2017). The current version of the DNA methylation-based CNS tumor classification system separates IDH mutant gliomas into three groups per 2021 WHO CNS tumor classification (Capper et al., 2018).

These three groups are closely related and classed under the “IDH glioma” methylation class family. Due to the strong association between the DNA methylation class and the presence of an IDH mutation, the methylation class is considered a proof of IDH mutational status (Capper et al., 2018) Methylation classes A IDH, A IDH HG and O IDH show a high overlap with methylation groups LGm1, LGm2 and LGm3, respectively, as described in The Cancer Genome Atlas (TCGA) dataset of Ceccarelli et al. in 2016 (Ceccarelli et al., 2016). The distinction between subclass-A IDH and subclass-A IDH HG has prognostic implications as confirmed in independent datasets (Kling et al., 2024; Tesileanu et al., 2021).

Infratentorial astrocytoma form a distinct methylation cluster and a discrete subgroup within IDH-mutant astrocytoma (Banan et al., 2020). These infratentorial tumors have high frequency of non-canonical IDH mutations thereby failing to be detected by immunohistochemistry (IHC). Incidence of ATRX mutations is also much lower in these tumors. Patients with infratentorial IDH mutant astrocytomas have significantly better clinical outcomes compared to those with diffuse midline gliomas harboring the H3K27M mutation, but significantly worse outcomes than patients with supratentorial IDH mutant astrocytomas (Banan et al., 2020).

A recent analysis of 98 tumors revealed that the current WHO grading criteria for IDH mutant astrocytomas had limited prognostic utility. In contrast, stratification based on DNA methylation profiles provided a more accurate prediction of overall survival. CNS WHO grade 3 patients were separated into low- and high-grade IDH-mutant astrocytoma groups based on DNA methylation profiling, eliminating the necessity for an intermediate histologic grade (Galbraith et al., 2024). On the other hand, Deep learning from histoPathoLOgy and methYlation (DEPLOY), a deep learning model classifying CNS tumors, was trained to predict tumor types including IDH mutant gliomas from hematoxylin and eosin (H&E) sections, (Hoang et al., 2024). DEPLOY comprises three distinct computational modules: (1) a “direct model” that classifies CNS tumors directly from histopathology slides; (2) an “indirect model” that first predicts DNA methylation beta values from slide images, which are then used to infer tumor classification; and (3) a “demographic model” that predicts tumor types directly from routinely collected patient demographic data. DEPLOY accurately predicted DNA methylation beta values from histopathological images. Utilizing a ten-class classification model trained on an internal cohort of 1,796 patients, DEPLOY predicted tumor categories in three independent external test datasets comprising a total of 2,156 patients. On high-confidence predictions, the model achieves an overall accuracy of 95% and a balanced accuracy of 91%. These findings highlight the potential of DEPLOY as a clinical decision-support tool to assist pathologists in the rapid and accurate diagnosis of CNS tumors potentially obsoleting the need for molecular studies.

The current standard of care for patients with IDH mutant Grade 3 astrocytoma is maximal safe surgical resection followed by involved field RT and maintenance TMZ (Mohile et al., 2022; Weller et al., 2021; Wick et al., 2009; Wick et al., 2016). CATNON (EORTC study 26,053–22,054) results showed survival benefit from adjuvant TMZ in adults with IDH mutant anaplastic astrocytoma (van den Bent et al., 2017; van den Bent et al., 2021). Korean Society of Neuro-Oncology Group study KNOG-1101 showed improved survival in adults with non codeleted anaplastic gliomas treated with concurrent and adjuvant TMZ with RT compared to only RT. However, there was no survival difference in a subgroup analysis of patients with IDH1 mutation which was likely due to the small sample size (Hwang et al., 2020).

The current standard of care for patients with IDH mutant Grade 3 oligodendroglioma is maximal safe surgical resection followed by involved field RT and maintenance PCV polychemotherapy (Mohile et al., 2022; Weller et al., 2021; Cairncross et al., 2013; van den Bent et al., 2013). EORTC Brain Tumour Group Study 26,951 was a randomized trial comparing efficacy of RT alone and RT followed by six cycles of PCV chemotherapy in adults with anaplastic oligodendroglioma. Adjuvant PCV was more beneficial in patients with 1p/19q codeleted tumors and IDH mutation was of prognostic significance as well (van den Bent et al., 2013). Radiation Therapy Oncology Group (RTOG) Study 9,402 compared intensive PCV chemotherapy before RT to RT alone. Chemoradiotherapy was shown to be a highly effective therapy with survival benefit over RT alone in patients with 1p/19q codeleted oligodendroglioma (Cairncross et al., 2013). There are ongoing phase 3 clinical trials trying to identify superior postsurgical treatment options and improve brain function and quality of life in patients with grade 2/3 oligodendroglioma (NCT, 2009, 2015; Wick et al., 2022).

The current standard of care for patients with IDH mutant WHO Grade 4 astrocytoma is either RT + adjuvant TMZ (similar to treatment for IDH mutant Grade 3 astrocytoma) or concurrent and adjuvant TMZ with RT (similar to treatment for IDH wildtype glioblastoma WHO Grade 4) (Mohile et al., 2022; Weller et al., 2021).

Therapy with the alkylating chemotherapy drug TMZ leads to acquired defects in DNA mismatch repair genes and a hypermutant phenotype in approximately 60% of oligodendrogliomas and 30% of astrocytomas (Eckel-Passow et al., 2015; Johnson et al., 2014; Bell et al., 2015). MGMT promoter methylated gliomas are more commonly associated with TMZ-induced hypermutation (Mathur et al., 2020). These tumors are more aggressive and resistant to further therapy with alkylator agents on account of mismatch repair (MMR) pathway mutations (Touat et al., 2020; Leelatian et al., 2021). However, the benefit of TMZ therapy with RT outweigh the risks and leads to improved survival compared to RT alone (van den Bent et al., 2021).

Phase 1 trials of ivosidenib and vorasidenib confirmed safety and activity in patients with nonenhancing IDH mutant glioma (Mellinghoff et al., 2020; Mellinghoff et al., 2021). These early trials showed efficacy of IDH inhibitors in nonenhancing tumors regardless of grade suggesting role of additional molecular alteration in advanced stages of disease (Ruda et al., 2024). Both agents were compared in a randomized phase 1 perioperative clinical trial of predominantly grade 2 IDH mutant glioma patients but included 6 patients with grade 3 tumors. Although, both drugs were found to be safe, vorasidenib had a better tumor penetration and was selected for the subsequent INDIGO clinical trial (Mellinghoff et al., 2023). INDIGO was a decisive phase 3, double-blinded trial which investigated the role of vorasidenib in adults with residual or recurrent grade 2 IDH mutant gliomas (Mellinghoff et al., 2023). Of 331 patients with no prior therapy other than surgery, 168 were randomized to receive vorasidenib 40 mg once a day. Patients randomized to the placebo arm were allowed to crossover to vorasidenib upon disease progression. At a median follow-up of 14.2 months, the primary endpoint of imaging-based PFS was significantly better for the vorasidenib group compared to the placebo group (hazard ratio 0.39, median 27.7 vs. 11.1 months, p < 0.001). The key secondary endpoint of time to next intervention was significantly improved in the vorasidenib group versus the placebo group (hazard ratio 0.26, p < 0.001). The safety profile of vorasidenib revealed low-grade toxicities with low numbers of serious adverse events and treatment discontinuations. INDIGO trial established a new treatment paradigm for a disease with largely “watch and wait” strategy, offering patients molecularly targeted treatment option. However, it is unclear if the PFS advantage will impact the OS as it will take years to tease out this information. The survival data will also be difficult to interpret due to the crossover design of the trial and as many as 32% of patients crossing over from the placebo arm to the vorasidenib arm. It is also unclear how efficacious vorasidenib is compared to the standard care of RT + PCV or RT + TMZ.

Newer IDH 1 inhibitors provide long-term responses in enhancing tumors and HGG in contrast to vorasidenib and ivosidenib (Ruda et al., 2024). D-2-HG inhibition and antitumor activity were seen in phase 1 clinical trials of IDH1 inhibitors Bay 1,436,032 and DS-1001 (Wick et al., 2021; Natsume et al., 2023). IDH1 inhibitor olutasidenib (FT-2102) was investigated in a nonrandomized phase 1b/2 clinical trial enrolling patients with relapsed or refractory IDH1 mutant glioma, 66% patients of which had HGG. This patient cohort was heavily pretreated with all patients receiving RT and 92% systemic therapy and surgery. Olutasidenib was well tolerated and showed antitumor activity with a 12-month PFS of 20.8% (de la Fuente et al., 2023). IDH1/2 inhibitors LY3410738 and HMPL-306 are being investigated in patients with IDH mutant solid tumors and gliomas (NCT, 2021, 2020).

Overall, IDH inhibitors have shown promising results with INDIGO trial results leading to the Food and Drug Administration (FDA) approval of vorasidenib in August 2024 for 12 years of older patients with grade 2 IDH1/2 mutant glioma following surgery. This is the first targeted therapy approved for IDH mutant glioma since the discovery of IDH mutations in 2008. Further clinical trials elucidating activity of IDH inhibitors either alone or in combination with chemoradiotherapy or other targeted therapy are necessary with a potential to advance survival and quality of life for patients with IDH mutant gliomas.

IDH1 R132H are frequent driver mutations encoding a shared clonal neopeptide recognized by mutation-specific T-helper cells. IDH1 R132H is expressed only in tumor and not in normal cells. These properties make IDH1 R132H an ideal candidate for a cancer vaccine target (Platten et al., 2021). An IDH1 R132H specific peptide vaccine was evaluated in a phase 1 clinical trial and was administered to 33 patients with grade 3/4 IDH mutant astrocytoma. It was found to be safe with adverse events restricted to grade. The vaccine administration was associated with a high number of pseudoprogressions. 3-year PFS was 64%, OS was 84% and immunity was induced in 93.3% patients (Platten et al., 2021). The successor trial, AMPLIFY-NEOVAC, was recently completed and randomized patients with relapsed IDH1 R132H positive glioma to receive immune checkpoint inhibitor (ICI) avelumab alone, IDH-vac alone or the combination of both. This was the first ICI clinical trial in the neoadjuvant setting and aims to comprehensively assess biologic responses to peptide vaccine and/or ICI (Bunse et al., 2022).

High levels D-2-HG result in epigenetic dysregulation leading to tumorigenesis and this is a potential pathway to exploit for the treatment of IDH mutant tumors. 5-azacytidine and decitabine are DNA methyltransferase inhibitors which induce DNA hypomethylation. This translates to reactivation of tumor suppressor genes and downregulation of oncogene expression. This mechanism as a potential option for therapy is being investigated in the preclinical and clinical settings (Alshiekh Nasany and de la Fuente, 2023).

IDH mutant cells escape radiation and chemotherapy induced injury through PARP mediated DNA repair (Sulkowski et al., 2017). The OLAGLI trial showed some activity of olaparib monotherapy (Ducray et al., 2021); however, veliparib in combination with RT and TMZ failed to improve outcome in pediatric patients with IDH mutant gliomas (Karajannis et al., 2025). PARP inhibitors such as niraparib (NCT, 2021), olaparib (Ducray et al., 2021; NCT, 2019b), BGB290 (pamiparib) (NCT, 2019c) are being investigated as single agents or in combination.

D-2-HG induced DNA hypermethylation silences programmed cell death-1 (PD-1) and its ligand (PDL-1) and this presents as a potential therapeutic option for patients with IDH mutant gliomas (Wang et al., 2016). Pembrolizumab and nivolumab are being investigated in combination with IDH inhibitors for patients with IDH mutant solid tumors including HGG (NCT, 2021, 2022, 2019).

Incorporating CDKN2A/B status and gene amplifications with histopathologic grade improves the accuracy of overall survival prediction (Ghosh et al., 2025). Both homozygous and hemizygous losses of CDKN2A/B, as well as focal gene amplifications, are linked to poorer overall survival, independent of tumor grade (Korshunov et al., 2019). Three prognostically distinct groups are identified based on CDKN2A/B loss, focal gene amplifications and WHO tumor grade as shown in Figure 7.

Low risk patients with grade 3 tumors with intact CDKN2A/B and no focal amplifications have the best outcome with median OS of 205.7 months. Intermediate risk patients with grade 3 astrocytomas and CDKN2A/B hemizygous loss, intact CDKN2A/B but focal amplifications and grade 4 patients with intact CDKN2A/B and no focal amplifications have median OS of 80.4, 88.7 and 91.5 months, respectively. High risk patients with grade 4 tumors with intact CDKN2A/B and focal amplifications (55.9 months) or with loss of CDKN2A/B (hemizygous 31.9 months, homozygous 32.5 months) have the worst median OS (Ghosh et al., 2025; Weller et al., 2024; von Deimling et al., 2018). These results need further validation as another recent study of 334 patients with Grade 2/3 IDH mutant glioma showed no prognostic association of hemizygous CDKN2A/B deletions (Ippen et al., 2025).

CDKN2A homozygous loss is associated with adverse prognosis in grade 3 oligodendroglioma (Figarella-Branger et al., 2025). In the French POLA cohort, the prognosis of such patients was close to or even worse than that of Grade 4 IDH mutant astrocytoma. In view of this dismal outcome, it is suggested to group these patients as Grade 4 oligodendroglioma (Appay et al., 2019).