The classification of central nervous system (CNS) tumors is evolving rapidly, driven by advances in molecular pathology and the widespread adoption of next-generation sequencing (NGS). The 2021 WHO Classification integrated molecular criteria into diagnostics, expanding the spectrum to over 100 distinct primary brain tumors (1). Although this shift improved diagnostic precision, it has unmasked previously unknown genetic alterations, now categorized as not otherwise specified (NOS) or not elsewhere classified (NEC), whose clinical implications remain under investigation. Epidemiological data from the Central Brain Tumor Registry of the United States (CBTRUS) estimate that these categories account for approximately 11% of gliomas. However, this prevalence is considered provisional and is expected to decline as the implementation of molecular techniques enables reclassification into distinct subtypes (2).

The inherent vulnerability of the EWSR1 gene to chromosomal breakage and translocation often results in its involvement in fusion events with different molecular partners, contributing to the development of diverse cancers (7). The EWSR1::PATZ1 gene fusion was initially described in sarcomas, with only four published series documenting approximately 30 cases to date. Most reported tumors arise in the thoracoabdominal region, exhibit round or spindle-cell low-grade or high-grade morphology, demonstrate polyphenotypic (ambiguous – neuro and muscular, specially rhabdomyoblastic) immunoprofile, and mainly display aggressive clinical behavior with poor response to conventional chemotherapy (3).

Since its initial identification in a CNS tumor in 2016, the EWSR1::PATZ1 fusion has been reported across diverse primary CNS tumors, including NEpT, ependymoma, ganglioglioma, pleomorphic xanthoastrocytoma, low-grade glial/glioneuronal tumors, astroblastoma and glioblastoma (5). Nonetheless, consensus on the optimal diagnostic criteria or management strategies for these tumors remains elusive, underscoring a critical gap in the literature. PATZ1 fusion-driven tumors are associated with upregulated anti-apoptotic pathways, and elucidating these mechanisms is critical for the development of novel strategies (4). Herein, we present two rare cases of primary CNS NEpTs harboring EWSR1::PATZ1 fusions, a unique clinical course, and co-existing previously undescribed genetic and methylation alterations in this tumor type.

NEpTs harboring EWSR1::PATZ1 fusions represent an exceptionally rare subgroup of primary brain neoplasms, currently unrecognized by the WHO classification system. Our case series is unique for several reasons. First, we describe two young patients (male and female) presenting with EWSR1::PATZ1 fusion tumors in distinct neuroanatomical regions and exhibiting heterogeneous histopathological and molecular profiles. Second, we provide comprehensive methylation analysis, a feature underexplored in prior reports. Third, we identified a co-occurring MUTYH mutation, a finding not previously documented in this context, that cannot be determined as somatic or germline based on the molecular tests. Fourth, the therapeutic approaches varied between the two cases, providing valuable insights to inform future treatment recommendations for this entity. Fifth, both cases exhibited an intermediate prognosis compared to GBM, reinforcing the potential for prognostic stratification through the identification of this fusion. Finally, our cases, together with the insights obtained from the literature review, may provide a basis for the formal inclusion of these tumors in the WHO classification.

In this case series, we highlight the diagnostic challenges posed by EWSR1::PATZ1-fused tumors. Regarding the initial CNS imaging workup, our findings align with previous reports. For example, a study of seven cases analyzing cranial MRI features identified strong contrast enhancement, intralesional hemorrhage, cystic components, intraventricular extension (often associated with hydrocephalus), and calcifications as consistent imaging characteristics of these tumors (6). Both Case 1 and Case 2 exhibited key elements of these criteria, with marked contrast enhancement and a cystic component being shared imaging characteristics.

From a histopathological standpoint, Case 1 partially matched a pattern reported in a review of 14 literature cases and a pediatric tumor case report, featuring small round high-grade poorly differentiated cells with relatively scant eosinophilic cytoplasm, arranged in vascular pseudorosettes with significant stromal vascularization (7, 8). Case 2 exhibited notable diagnostic discordance. It was initially classified as an extraventricular neurocytoma by an external service based on its low-to-intermediate grade, small, monomorphic epithelioid cells with clear cell morphology – a diagnosis not previously associated with EWSR1::PATZ1 fusions (5). However, upon reevaluation of the specimen from the recurrence, most of the characteristics were also present, supporting the diagnostic impression of NEpT. Notably, Ki-67 index diverged markedly (70% in high-grade Case 1 vs. 2-3% in low-grade Case 2) and the clinical impact of this finding is still unknown. While molecular diagnostics are indispensable in neuro-oncology, our data demonstrate that combined assessment through histopathology, immunohistochemistry, and cranial imaging can inform the diagnostic hypothesis of neuroepithelial tumors (NEpTs), especially in low-resource environments. Neuropathologist-driven evaluation enhances this strategy by mitigating morphological variability and promoting cost-effective use of molecular assays through targeted case selection for genomic and epigenomic profiling.

EWSR1 and PATZ1 are located in close proximity on chromosome 22. The EWSR1::PATZ1 fusion arises from an intrachromosomal rearrangement between EWSR1 (22q12.2) and PATZ1 (22q12.2), resulting in an in-frame fusion that combines the N-terminal domain of EWSR1 with the zinc finger domain of PATZ1 (9). This molecular complexity justifies the necessity of NGS, particularly in young IDH-wildtype patients, to identify oncogenic drivers and actionable therapeutic targets (10). In Case 1, it is noteworthy that other hallmark characteristics of GBM, such as EGFR amplification, TERT promoter mutations, or gains of chromosome 7 with loss of chromosome 10, as well as other driver alterations in high-grade CNS neoplasms, were absent. In Case 2, the lack of BRAF, FGFR1 or MYB/MYBL alterations – genes commonly implicated as sole drivers in low-grade NEpT – underscores EWSR1::PATZ1 fusion role as the primary driver in these tumors (11).

The role of co-mutations also warrants further investigation. In sarcomas, CDKN2A alterations, reported in 71% of EWSR1::PATZ1 fusion cases in a series of 11 patients, are associated with aggressive phenotypes and poorer outcomes (12). In Case 1, we identified a heterozygous MUTYH mutation (c.1103G>A, p.(Gly368Asp), G368D), whose somatic or germline status could not be determined by molecular testing, representing a novel finding in NEpT with EWSR1::PATZ1 fusion. While germline MUTYH mutations are linked to diffuse midline pediatric gliomas, their somatic role in CNS carcinogenesis is yet to be fully elucidated. Interestingly, a single case of atypical central neurocytoma with EWSR1::ATF1 fusion and MUTYH G382D mutation (detected via NGS) achieved ~3 years of PFS with conventional glioma therapy in a 13-year-old patient (13).

The MUTYH c.1103G*–>A (p.Gly368Asp) variant is a missense mutation associated with partial or complete loss of glycosylase activity, impairing the removal of mispaired adenines opposite 8-oxoguanine (8-oxoG) during base excision repair (BER). This deficiency compromises the cellular defense against G:C to T:A transversions, particularly under oxidative stress. In gliomas, temozolomide (TMZ) is a widely used alkylating agent that induces DNA damage primarily through methylation of guanine at the O₆ position and promotes the generation of reactive oxygen species (ROS), further contributing to the accumulation of oxidative substrates such as 8-oxoG. In the absence of functional MUTYH, these oxidative lesions remain unrepaired, leading to replication errors, accumulation of somatic mutations, and progressive genomic instability. Over time, this process may favor the selection of resistant tumor subclones and contribute to the development of a hypermutated phenotype. This mutational profile has been reported in recurrent gliomas and is frequently associated with secondary alterations in mismatch repair genes, such as MSH6 or MSH2, ultimately correlating with increased tumor aggressiveness and reduced responsiveness to standard therapies, including the Stupp protocol (14, 15).

Notably, Siegfried et al. analyzed methylation profiles in 7 of 40 BRAFV600-negative gangliogliomas and unexpectedly identified EWSR1::PATZ1 fusions during molecular evaluation of a putative papillary ganglioglioma for SLC44A1 (solute carrier family 44, member 1)::PRKCA (protein kinase C alpha) fusions. Their analysis revealed that these fusions define a distinct, well-differentiated methylation cluster, clearly segregated from other glioneuronal tumors. This molecular distinctiveness – coupled with the tumor’s ventricular localization and glioneuronal morphology – provides strong support for classifying EWSR1::PATZ1-fused tumors as a unique clinicopathological entity (16).

The prognostic and therapeutic implications of MGMT promoter methylation remain unclear in NEpTs harboring EWSR1::PATZ1 fusions. In Case 1, the detection of MGMT promoter methylation in an IDH-wildtype, high-grade tumor prompted treatment with the Stupp protocol, a regimen typically used for GBM. The patient responded favorably to chemoradiotherapy, achieving a progression-free survival (PFS) of 14 months, exceeding the median PFS of 6.9 months reported in the phase III study by Stupp et al., Radiotherapy plus Concomitant and Adjuvant Temozolomide for Glioblastoma (PMID: 15758009) (17). In contrast, Case 2 experienced recurrence 8 months after the initial surgery. Key factors contributing to this outcome include the initial R1 resection and the absence of adjuvant therapy. Following a second R1 resection and subsequent adjuvant radiotherapy, the patient has remained disease-free for 11 months to date.

In our cohort, methylation profiling proved diagnostically decisive, aligning with findings from previous studies on sarcomas and CNS tumors (18, 19). Objective parameters, such as calibrated scores, may assist pathologists in the identification and stratification of these tumors, serving as a technical and reproducible benchmark for the development of a diagnostic algorithm for NEpTs. This molecular signature further substantiates the proposal to classify NEpTs as a distinct diagnostic entity. Its recognition as a core diagnostic element may inform future revisions of the WHO classification, supporting the broader integration of methylation profiling into CNS tumor diagnostics to enhance nosological precision. However, while methylation patterns are highly informative in distinguishing molecular subtypes, they may not fully capture tumor heterogeneity. Other epigenetic mechanisms, including histone modifications, chromatin remodeling, and non-coding RNA regulation, also influence tumor phenotypes. Epigenetic alterations evolve over time and across tumor regions, shaped by the microenvironment, selective therapy pressure, and clonal evolution. Therefore, relying solely on a single methylation snapshot for diagnosis may be insufficient, given the complexity of tumor biology (20, 21).

When comparing these results to those in low-grade gliomas, it is evident that the median PFS for grade 2 gliomas stands at 4 years, according to the RTOG 9802 (ClinicalTrials.gov, NCT00003375) study (22), whereas grade 3 gliomas demonstrate a median of 1.7 years, as shown in the RTOG 9402 (NCT00002569) (23), and 13.2 months, as documented in EORTC 26951 trial (NCT00002840) (24). Alhalabi et al. reported a median PFS of 144 months across 60 PATZ1-fused brain tumors but only 31 months (range: 6–144) for EWSR1::PATZ1 fusion cases confirmed by RNA sequencing (25). While longer follow-up is needed, our findings position these tumors prognostically between GBM and low-grade gliomas.

PATZ1 interacts with multiple genes implicated in apoptosis regulation, including DR5, PUMA, and BCL-2 family members, offering actionable therapeutic targets. Recent drug screening in a GBM model with MN1::PATZ1 fusion identified promising candidates: paclitaxel, D-actinomycin, volasertib (PLK1 inhibitor), navitoclax (BCL-2/BCL-XL/BCL-W inhibitor), and I-BET-151 (bromodomain inhibitor) (25, 26). The interplay between PATZ1, the tumor necrosis factor-related apoptosis-inducing ligand (TRAIL) pathway and TMZ offers a pivotal therapeutic axis. TRAIL, the natural ligand for apoptotic receptors, can trigger apoptosis independently of p53, which is mutated in up to 65% of GBM cases, partly contributing to TMZ resistance. TMZ may also act as a TRAIL “sensitizer”, overcoming resistance by upregulating death receptor expression and promoting significant caspase activation (27, 28). PATZ1 functions as a transcriptional repressor that confers resistance to apoptosis, which is the primary antitumor mechanism of TMZ. In summary, targeting PATZ1 could represent a novel therapeutic strategy to overcome TMZ resistance.

This report provides a comprehensive molecular and clinical characterization of high-grade and low-grade ultra-rare NEpTs harboring the EWSR1::PATZ1 fusion, highlighting the role of methylation profiling as a valuable diagnostic tool. We present early treatment outcomes with glioma-based therapies and identify the first documented MUTYH mutation in these tumors, which may provide valuable insights into CNS tumorigenesis. Our findings reinforce the role of EWSR1::PATZ1 as a driver alteration, defining a subset of tumors with distinct clinicopathological and molecular hallmarks, such as ventricular location, glioneuronal differentiation, and a unique methylation profile, with an intermediate prognosis. These characteristics strongly support its classification as an independent entity in future WHO revisions. Additionally, we encourage further research into targeted therapies that explore the downregulation of PATZ1 expression, understanding its potential to serve as a therapeutic target in these tumors.