CIC/ATXN1-rearranged tumors in the central nervous system are mainly represented by sarcomas: A comprehensive clinicopathological and epigenetic series.

CIC fusions have been described in two different central nervous system (CNS) tumor entities. On one hand, fusions of CIC or ATXN1 genes belonging to the same complex of transcriptional repressors, were reported in the CIC-rearranged, sarcoma (SARC-CIC). The diagnosis of this tumor type, which was recently added to the World Health Organization (WHO) Classification of CNS tumors, is difficult mainly because the data concerning its histopathology (as compared to its soft tissue counterpart), immunoprofile, and clinical as well as radiological characteristics are scarce in the literature. On the other hand, a recent study, based on DNA-methylation profiling, has identified a novel high-grade neuroepithelial tumor characterized by recurrent CIC fusions (HGNET-CIC). The aim of this multicentric study was to characterize a cohort of 15 primary CNS tumors harboring a CIC or ATXN1 fusion in terms of clinical, radiological, histopathological, immunophenotypical, and epigenetic characteristics. According to the integrated diagnoses, 14/15 tumors corresponded to SARC-CIC, and only one to HGNET-CIC. The tumors showed similar clinical (mainly pediatric), radiological (mostly supratentorial, cystic, and contrast enhancing), immunophenotypical (common expression of glioneuronal markers), and genetic (similar spectrum of fusions) profiles but their histopathological appearance was clearly distinct. Moreover, we found a novel fusion transcript (CIC::EWSR1) in a SARC-CIC. Most DNA methylation profiles using the Heidelberg Brain Tumor Classifier (v12.8) annotated the samples to the methylation class "SARC-CIC" (9/14 tumors with available data). By using uniform manifold approximation and projection analysis, four other samples were classified as SARC-CIC and another clustered within the methylation class of HGNET-CIC. Our findings confirm that CNS CIC-fused tumors do not represent a single molecular tumor entity. Further analyses are needed to characterize HGNET-CIC in more detail. These results may help to refine the essential diagnostic criteria for SARC-CIC and their terminology (with a suggested consensual name of sarcoma, CIC/ATXN1-complex rearranged).

Expanding the clinicopathologic spectrum and genomic landscape of tumors with SMARCA2/4::CREM fusions.

CREB gene family (ATF1, CREB1, CREM) fusions with either EWSR1 or FUS gene partners drive the pathogenesis of a wide range of neoplasms, including various soft tissue tumors, intracranial myxoid mesenchymal tumors (IMMTs), hyalinizing clear cell carcinoma (HCCC), and rare mesotheliomas. Recently, a SMARCA2::CREM fusion was reported in one case each of IMMT and HCCC. In this study, we expand the clinicopathologic and molecular spectrum of these neoplasms by describing three additional cases with SMARCA2::CREM and one with a novel SMARCA4::CREM fusion, highlighting the recurrent potential of additional CREB gene fusion partners beyond FET family members. To evaluate if these fusions define a new pathologic entity, we performed a comprehensive genomic and methylation analysis and compared the results to other related tumors. Tumors occurred in children and young adults (median age 20 years) and spanned a broad anatomic distribution, including soft tissue, intracranial, head and neck, and prostatic urethra. Microscopically, the tumors shared an undifferentiated round to epithelioid cell phenotype and a hyalinized fibrous stroma. Immunohistochemically, a polyphenotypic profile was observed, with variable expression of SOX10, desmin, and/or epithelial markers. No targetable genomic alterations were found using panel-based DNA sequencing. By DNA methylation and transcriptomic analyses, tumors grouped closely to FET::CREB entities, but not with SMARCA4/SMARCB1-deficient tumors. High expression of CREM by immunohistochemistry was also documented in these tumors. Patients experienced local recurrence (n = 2), locoregional lymph node metastases (n = 2), and an isolated visceral metastasis (n = 1). Overall, our study suggests that SMARCA2/4::CREM fusions define a distinct group of neoplasms with round cell to epithelioid histology, a variable immunoprofile, and a definite risk of malignancy. Larger studies are needed to further explore the pathogenetic relationship with the FET::CREB family of tumors. © 2024 The Pathological Society of Great Britain and Ireland.

Post-zygotic mosaicism of SMARCB1 variants in patients with rhabdoid tumors: a not so rare condition exposing to successive tumors.

BACKGROUND: Rhabdoid tumors (RT) are aggressive, rare tumors predominantly affecting young children, characterized by bi-allelic SMARCB1 gene inactivation. While most SMARCB1 alterations are acquired de novo, a third of cases exhibit germline alterations, defining Rhabdoid Tumors Predisposition Syndrome (RTPS1). With increased sensitivity of next-generation sequencing (NGS), mosaicisms in genes linked to genetic diseases are more detectable. This study focuses on exploring SMARCB1 germline alterations, notably mosaicism in blood samples of children with RT and in parents, using a custom NGS panel. METHODS: A cohort of 280 children and 140 parents with germline analysis was studied. Germline DNA from 111 children with RT and 32 parents were re-analyzed with a custom NGS panel with 1,500X average depth targeting the SMARCB1 gene to identify intragenic variants not detected with conventional low-sensitivity methods. Follow-up data was obtained for 77 patients. RESULTS: Nine previously undetected mosaicism cases were identified, totaling 17/280 patients with a mosaic variant (6.1%) in the cohort, with variant allele frequencies between 0.9% and 33%, thus highlighting the prior underestimation of its prevalence. Follow-up data showed that 4 out of 7 survivors with mosaic variants developed distinct novel tumors, two sharing SMARCB1 alterations with the initial tumor, emphasizing the potential clinical impact of SMARCB1 mosaicism. CONCLUSIONS: The hitherto underestimated rate of SMARCB1 mosaicism in RT underscores the need for optimized genetic counseling and oncological monitoring. The findings have significant medical implications, considering the dire prognosis of RT.

Supra-tentorial Ependymomas with ZFTA Fusion, YAP1 Fusion, and Astroblastomas, MN1-altered: Characteristic Imaging Features.

PURPOSE: Supratentorial (ST) ependymoma subgroups are defined by two different fusions with different prognoses. Astroblastomas, MN1-altered, have ependymal-like histopathologic features and represent a differential diagnosis in children. We hypothesized that ZFTA-fused ependymoma and YAP1-fused ependymoma on the one hand, and astroblastoma, MN1-altered, on the other hand, show different MRI characteristics. METHODS: We retrospectively analyzed the preoperative imaging of 45 patients with ST ependymoma or astroblastoma between January 2000 and September 2020, blinded to histomolecular grouping. Several characteristics, such as location, tumor volume, calcifications, solid/cystic component, and signal enhancement or diffusion were evaluated. We compared imaging characteristics according to their molecular subtype (ZFTA-fused, YAP1-fused, and astroblastoma, MN1-altered). RESULTS: Thirty-nine patients were classified as having an ependymoma, 35 with a ZFTA fusion and four with a YAP1 fusion, and six as having an astroblastoma, MN1-altered. YAP1-fused ependymomas were more likely to involve at least 3 lobes than ZFTA-fused ependymomas. Astroblastomas were located in the frontal lobe in 100% of the tumors versus 49% of the ependymomas. Cerebral blood flow by arterial spin labeling was higher in astroblastomas than in ependymomas. There were no differences in the other characteristics between the molecular groups. All the tumors showed common features: intra-axial extra-ventricular tumors, very frequent contrast enhancement (39/43, 91%), a cystic/necrotic component (41/45, 91%), restricted diffusion (32/36, 89%), calcifications (15/18, 83%), and peri-tumoral edema (38/44, 86%). CONCLUSION: The distinction between ST ependymoma subtypes and astroblastomas can be guided by several imaging features. These tumors share common imaging features that may help to differentiate ST ependymomas and astroblastomas from other pediatric ST tumors.

Long-term survivors in 976 supratentorial glioblastoma, IDH-wildtype patients.

OBJECTIVE: Glioblastoma, isocitrate dehydrogenase (IDH)-wildtype is the most aggressive glioma with poor outcomes. The authors explored survival rates and factors associated with long-term survival in patients harboring a glioblastoma, IDH-wildtype. METHODS: In an observational, retrospective, single-center study, the authors examined the medical records of 976 adults newly diagnosed with supratentorial glioblastomas, IDH-wildtype between January 2000 and January 2021. They analyzed clinical-, imaging-, and treatment-related factors associated with 2-year and 5-year survival. RESULTS: The median overall survival was 11.2 months (12.2 months for patients included after 2005 and the introduction of standard combined chemoradiotherapy). The median progression-free survival was 9.4 months (10.0 months for patients included after 2005). Overall, 17.6% of patients reached a 2-year overall survival, while 2.2% of patients reached a 5-year overall survival. Furthermore, 6.6% of patients survived 2 years without progression, while 1.1% of patients survived 5 years without progression. Two factors that were consistently associated with 2-year and 5-year survival were first-line oncological treatment with standard combined chemoradiotherapy and methylated O6-methylguanine-DNA methyltransferase promoter. Other factors that were significantly associated with 2-year or 5-year survival were age at diagnosis ≤ 60 years, headaches or signs of raised intracranial pressure at diagnosis, cortical contact of contrast enhancement, no contrast enhancement crossing the midline on initial imaging, total or subtotal tumor resection, and a second line of oncological treatment at recurrence. Within 21 cases of 5-year survival, 18 were confirmed to be glioblastomas, IDH-wildtype, and 7 of the 5-year survivors (38.9%) had additional genetic alterations: 3 cases had an FGFR mutation or fusion, 3 cases had a PIK3CA mutation, 1 case had a PTPN11 mutation, and 1 case had a PMS2 mutation in the context of constitutional mismatch repair deficiency syndrome. CONCLUSIONS: Five-year overall survival in patients with glioblastoma, IDH-wildtype is extremely low. Predictors of a longer survival are mostly treatment factors, emphasizing the importance of a complete oncological treatment plan, when achievable. Glioblastoma, IDH-wildtype 5-year survivors could be screened for actionable targets in case of recurrence.

Medulloblastomas with ELP1 pathogenic variants: A weakly penetrant syndrome with a restricted spectrum in a limited age window.

Background: ELP1 pathogenic variants (PV) have been recently identified as the most frequent variants predisposing to Sonic Hedgehog (SHH) medulloblastomas (MB); however, guidelines are still lacking for genetic counseling in this new syndrome. Methods: We retrospectively reviewed clinical and genetic data of a French series of 29 ELP1-mutated MB. Results: All patients developed SHH-MB, with a biallelic inactivation of PTCH1 found in 24 tumors. Other recurrent alterations encompassed the TP53 pathway and activation of MYCN/MYCL signaling. The median age at diagnosis was 7.3 years (range: 3-14). ELP1-mutated MB behave as sporadic cases, with similar distribution within clinical and molecular risk groups and similar outcomes (5 y - OS = 86%); no unusual side effect of treatments was noticed. Remarkably, a germline ELP1 PV was identified in all patients with available constitutional DNA (n = 26); moreover, all tested familial trio (n = 11) revealed that the PVs were inherited. Two of the 26 index cases from the French series had a family history of MB; pedigrees from these patients and from 1 additional Dutch family suggested a weak penetrance. Apart from MB, no cancer was associated with ELP1 PVs; second tumors reported in 4 patients occurred within the irradiation fields, in the usual time-lapse for expected radiotherapy-induced neoplasms. Conclusions: The low penetrance, the "at risk' age window limited to childhood and the narrow tumor spectrum, question the actual benefit of genetic screening in these patients and their family. Our results suggest restricting ELP1 germline sequencing to patients with SHH-MB, depending on the parents" request.

Specific brain MRI features of constitutional mismatch repair deficiency syndrome in children with high-grade gliomas.

BACKGROUND: Children with constitutional mismatch repair deficiency (CMMRD) syndrome have an increased risk of high-grade gliomas (HGG), and brain imaging abnormalities. This study analyzes brain imaging features in CMMRD syndrome children versus those with HGG without CMMRD. METHODS: Retrospective comparative analysis of brain imaging in 30 CMMRD children (20 boys, median age eight years, 22 with HGG), seven with Lynch syndrome (7 HGG), 39 with type 1 neurofibromatosis (NF1) (four with HGG) and 50 with HGG without MMR or NF1 pathogenic variant ("no-predisposition" patients). RESULTS: HGG in CMMRD and Lynch patients were predominantly hemispheric (versus midline) compared to NF1 and no-predisposition patients (91% and 86%, vs 25% and 54%, p = 0.004). CMMRD-associated tumors often had ill-defined boundaries (p = 0.008). All CMMRD patients exhibited at least one developmental venous anomaly (DVA), versus 14%, 10%, and 6% of Lynch, NF1, and no-predisposition patients (p < 0.0001). Multiple DVAs were observed in 83% of CMMRD patients, one NF1 patient (3%), and never in other groups (p < 0.0001). Cavernomas were discovered in 21% of CMMRD patients, never in other groups (p = 0.01). NF1-like focal areas of high T2-FLAIR signal intensity (FASI) were more prevalent in CMMRD patients than in Lynch or no-predisposition patients (50%, vs 20% and 0%, respectively, p < 0.0001). Subcortical and ill-limited FASI, possibly involving the cortex, were specific to CMMRD (p < 0.0001) and did not evolve in 93% of patients (13/14). CONCLUSION: Diffuse hemispherically located HGG associated with multiple DVAs, cavernomas, and NF1-like or subcortical FASI strongly suggests CMMRD syndrome compared to children with HGG in other contexts. CLINICAL RELEVANCE STATEMENT: The radiologic suggestion of CMMRD syndrome when confronted with HGGs in children may prompt genetic testing. This can influence therapeutic plans. Therefore, imaging features could potentially be incorporated into CMMRD testing recommendations. KEY POINTS: Using imaging to detect CMMRD syndrome early may improve patient care. CMMRD features include: hemispheric HGG with multiple developmental venous anomalies and NF1-like or subcortical areas with high T2-FLAIR intensity. We propose novel imaging features to improve the identification of potential CMMRD patients.

Diffuse glioneuronal tumor with oligodendroglioma-like features and nuclear clusters (DGONC), new name and new problems: an illustration of one case with atypical morphology and biology.

A novel histomolecular tumor of the central nervous system (CNS), the "diffuse glioneuronal tumor with oligodendroglioma-like features and nuclear clusters (DGONC)," has recently been identified, based on a distinct DNA methylation profile and has been added to the 2021 World Health Organization Classification of CNS Tumors. This glioneuronal tumor mainly affects the supratentorial area in children and recurrently presents with a monosomy of chromosome 14. Herein, we report the case of a DNA-methylation based diagnosis of DGONC having atypical features, such as pseudo-rosettes and the absence of a chromosome 14 monosomy, thus rendering its diagnosis very challenging. Because of the wide variety of morphologies harbored by DGONC, a large range of differential diagnoses may be hypothesized from benign to malignant. Interestingly, the current case, like one previously reported, exhibited a co-expression of OLIG2, synaptophysin and SOX10, without GFAP immunopositivity. This particular immunophenotype seems to be a good indicator for a DGONC diagnosis. The classification of DGONC amongst glioneuronal or embryonal tumors is still debated. The clinical (a pediatric supratentorial tumor), morphological (from a benign oligodendroglioma-like tumor with microcalcifications and possible neuropil-like islands to a malignant embryonal tumor with a possible spongioblastic pattern), and immunohistochemical (co-expression of OLIG2 and synaptophsyin) profiles resemble CNS, neuroblastoma, FOXR2-activated and may potentially bring them together in a future classification. Further comprehensive studies are needed to conclude the cellular origin of DGONC and its prognosis.

A comprehensive histomolecular characterization of meningioangiomatosis: Further evidence for a precursor neoplastic lesion.

Meningioangiomatosis (MAM) remains a poorly understood lesion responsible for epileptic disease. In the past, MAM was primarily described in the context of neurofibromatosis type 2 before being mainly reported sporadically. Moreover, the malformative or tumoral nature is still debated. Because a subset of MAM are associated with meningiomas, some authors argue that MAM corresponds to an infiltration pattern of these tumors. For these reasons, MAM has not been added to the World Health Organization (WHO) Classification of Central Nervous System Tumors as a specific entity. In the present study, we characterized a series of pure MAM (n = 7) and MAM associated with meningiomas (n = 4) using histopathology, immunohistochemistry, genetic (fluorescent in situ and DNA sequencing analyses), and epigenetic (DNA-methylation profiling) data. We evidenced two distinct morphological patterns: MAM with a fibroblastic-like pattern having few lesional cells, and MAM with a more cellular pattern. A subset was associated with the genetic alterations previously reported in meningiomas (such as a KMT2C mutation and a hemizygous deletion of chromosome 22q including the NF2 gene). The DNA-methylation profile, using a t-distributed stochastic neighbor embedding analysis, evidenced that MAM (pure or associated with meningiomas) clustered in a separate group from pediatric meningiomas. The present results seem to suggest that MAM represents a neoplastic lesion and encourage the further study of similar additional series so that it may be included in a future WHO classification.

CNS tumors with PLAGL1-fusion: beyond ZFTA and YAP1 in the genetic spectrum of supratentorial ependymomas.

A novel methylation class, "neuroepithelial tumor, with PLAGL1 fusion" (NET-PLAGL1), has recently been described, based on epigenetic features, as a supratentorial pediatric brain tumor with recurrent histopathological features suggesting an ependymal differentiation. Because of the recent identification of this neoplastic entity, few histopathological, radiological and clinical data are available. Herein, we present a detailed series of nine cases of PLAGL1-fused supratentorial tumors, reclassified from a series of supratentorial ependymomas, non-ZFTA/non-YAP1 fusion-positive and subependymomas of the young. This study included extensive clinical, radiological, histopathological, ultrastructural, immunohistochemical, genetic and epigenetic (DNA methylation profiling) data for characterization. An important aim of this work was to evaluate the sensitivity and specificity of a novel fluorescent in situ hybridization (FISH) targeting the PLAGL1 gene. Using histopathology, immunohistochemistry and electron microscopy, we confirmed the ependymal differentiation of this new neoplastic entity. Indeed, the cases histopathologically presented as "mixed subependymomas-ependymomas" with well-circumscribed tumors exhibiting a diffuse immunoreactivity for GFAP, without expression of Olig2 or SOX10. Ultrastructurally, they also harbored features reminiscent of ependymal differentiation, such as cilia. Different gene partners were fused with PLAGL1: FOXO1, EWSR1 and for the first time MAML2. The PLAGL1 FISH presented a 100% sensitivity and specificity according to RNA sequencing and DNA methylation profiling results. This cohort of supratentorial PLAGL1-fused tumors highlights: 1/ the ependymal cell origin of this new neoplastic entity; 2/ benefit of looking for a PLAGL1 fusion in supratentorial cases of non-ZFTA/non-YAP1 ependymomas; and 3/ the usefulness of PLAGL1 FISH.

Radiographic growth rate as a predictor of aggressiveness of diffuse gliomas without 1p19q codeletion.

OBJECTIVE: The 2021 WHO classification of CNS tumors has refined the definition of adult-type diffuse gliomas without 1p19q codeletion. Nevertheless, the aggressiveness of gliomas is based exclusively on histomolecular criteria performed on a limited sample of the tumor. The authors aimed to assess whether the spontaneous radiographic tumor growth rate is associated with tumor aggressiveness and allows preoperative identification of malignancy grade of adult-type diffuse gliomas without 1p19q codeletion. METHODS: The authors retrospectively reviewed the records of adult patients harboring a newly diagnosed supratentorial diffuse glioma without 1p19q codeletion, with available preoperative MRI follow-up between January 2008 and April 2022. The spontaneous radiographic tumor growth rate was quantified by tumor volume segmentation and regression of the evolution of the mean tumor diameter over time and was compared with clinical, imaging, histomolecular, and survival data. RESULTS: Ninety-six patients were included. The spontaneous radiographic tumor growth rates (mean 17.8 ± 38.8 mm/year, range 0-243.5 mm/year) significantly varied according to IDH1/2 mutation (p < 0.001), grade of malignancy (p < 0.001), and presence of microvascular proliferation (p < 0.001). The spontaneous radiographic tumor growth rate allowed preoperative identification of high-grade cases: 100% of grade 3 and 4 IDH-mutant diffuse astrocytomas had a spontaneous radiographic tumor growth rate ≥ 8.0 mm/year, and 100% of IDH-wild-type glioblastomas had a spontaneous radiographic tumor growth rate ≥ 42.0 mm/year. A spontaneous radiographic growth rate ≥ 8.0 mm/year was an independent predictor of shorter progression-free (p = 0.014) and overall (p = 0.007) survival. A mitotic count threshold ≥ 4 mitoses was the optimal threshold for identifying aggressive IDH-mutant astrocytomas based on spontaneous radiographic tumor growth. CONCLUSIONS: The spontaneous radiographic tumor growth rates could be used as an additional tool to preoperatively screen tumor aggressiveness of adult-type diffuse gliomas without 1p19q codeletion.

Characteristics and Prognosis of Tumor-Related Epilepsy During Tumor Evolution in Patients With IDH Wild-Type Glioblastoma.

BACKGROUND AND OBJECTIVES: Tumor-related epilepsy is a well-known symptom of glioblastoma. However, the particular characteristics of epileptic seizures related to glioblastoma, isocitrate dehydrogenase (IDH)-wild-type is almost unexplored longitudinally during the whole course of the disease. We assessed tumor-related epilepsy and seizure control during tumor evolution and the prognostic significance of tumor-related epilepsy. METHODS: We performed an observational, retrospective single-center study at one tertiary referral neuro-oncology surgical center (2000-2020). We included adult patients treated for a newly diagnosed supratentorial glioblastoma, IDH-wild-type with available preoperative and postoperative MRI and with available epileptic seizure status at diagnosis. To determine factors associated with tumor-related epilepsy or seizure control, univariate analyses were performed using the χ2 or Fisher exact tests for categorical variables and the unpaired t test or Mann-Whitney rank-sum test for continuous variables. Predictors associated with tumor-related epilepsy and seizure control in unadjusted analysis were entered into backward stepwise logistic regression models. RESULTS: One thousand six patients were enrolled. The cumulative incidence of tumor-related epilepsy increased during tumor evolution (33.1% at diagnosis, 44.7% after oncologic treatment, 52.4% at progression, and 51.8% at the end-of-life phase) and is related to tumor features (cortex involvement, no necrosis, and small volume). Uncontrolled epileptic seizures increased during tumor evolution (20.1% at diagnosis, 32.0% after oncologic treatment, 46.7% at progression, and 41.1% at the end-of-life phase). Epileptic seizure control after oncologic treatment was related to seizure features (uncontrolled before oncologic treatment and focal-to-bilateral tonic-clonic seizures) and to the extent of resection. Epileptic seizure control at tumor progression was related to seizure features (presence at diagnosis and uncontrolled after oncologic treatment) and to the time to progression. Tumor-related epilepsy at diagnosis was a predictor of a longer overall survival (adjusted hazard ratio, 0.78; 95% CI 0.67-0.90; p < 0.001) independent of age, Karnofsky Performance Status score, tumor location and volume, extent of resection, standard combined chemoradiotherapy, levetiracetam use, and MGMT promoter methylation. DISCUSSION: The progression of tumor-related epilepsy with the evolution of glioblastoma, IDH-wild-type and the effects of surgery on seizure control argue for proper antiseizure medication and maximal safe resection. Tumor-related epilepsy is an independent predictor of a longer survival.

Efficacy and Safety of Carmustine Wafer Implantation After Ventricular Opening in Glioblastomas, Isocitrate Dehydrogenase-Wildtype, in Adults.

BACKGROUND AND OBJECTIVES: We assessed the impact of ventricular opening on postoperative complications and survival of carmustine wafer implantation during surgery of newly diagnosed supratentorial glioblastomas, isocitrate dehydrogenase (IDH)-wildtype in adults. METHODS: We performed an observational, retrospective, single-center cohort study at a tertiary surgical neuro-oncological center between January 2006 and December 2021. RESULTS: One hundred ninety-four patients who benefited from a first-line surgical resection with carmustine wafer implantation were included. Seventy patients (36.1%) had a ventricular opening. We showed that ventricular opening (1) did not increase overall postoperative complication rates (P = .201); (2) did not worsen the early postoperative Karnofsky Performance Status score (P = .068); (3) did not increase the time interval from surgery to adjuvant oncological treatment (P = .458); (4) did not affect the completion of the standard radiochemotherapy protocol (P = .164); (5) did not affect progression-free survival (P = .059); and (6) did not affect overall survival (P = .142). CONCLUSION: In this study, ventricular opening during first-line surgical resection did not affect the survival and postoperative complications after use of carmustine wafer implantation in adult patients with a newly diagnosed supratentorial glioblastoma, IDH-wildtype. This warrants a prospective and multicentric study to clearly assess the impact of the ventricular opening after carmustine wafer implantation in glioblastoma, IDH-wildtype.

Utility of combining OLIG2 and SOX10 IHC expression in CNS tumours: promising biomarkers for subtyping paediatric- and adult-type gliomas.

AIMS: The SOX10 transcription factor is important for the maturation of oligodendrocytes involved in central nervous system (CNS) myelination. Currently, very little information exists about its expression and potential use in CNS tumour diagnoses. The aim of our study was to characterize the expression of SOX10 in a large cohort of CNS tumours and to evaluate its potential use as a biomarker. METHODS: We performed immunohistochemistry (IHC) for SOX10 and OLIG2 in a series of 683 cases of adult- and paediatric-type CNS tumours from different subtypes. The nuclear immunostaining results for SOX10 and OLIG2 were scored as positive (≥10% positive tumour cells) or negative. RESULTS: OLIG2 and SOX10 were positive in diffuse midline gliomas (DMG), H3-mutant, and EZHIP-overexpressed. However, in all DMG, EGFR-mutant, SOX10 was constantly negative. In diffuse paediatric-type high-grade gliomas (HGG), all RTK1 cases were positive for both OLIG2 and SOX10. RTK2 cases were all negative for both OLIG2 and SOX10. MYCN cases variably expressed OLIG2 and were all immunonegative for SOX10. In glioblastoma, IDH-wildtype, OLIG2 was mostly positive, but SOX10 was variably expressed, depending on the epigenetic subtype. All circumscribed astrocytic gliomas were positive for both OLIG2 and SOX10 except pleomorphic xanthoastrocytomas, astroblastomas, MN1-altered, and subependymal giant cell astrocytomas. SOX10 was negative in ependymomas, meningiomas, pinealoblastomas, choroid plexus tumours, intracranial Ewing sarcomas, and embryonal tumours except neuroblastoma, FOXR2-activated. CONCLUSION: To conclude, SOX10 can be incorporated into the IHC panel routinely used by neuropathologists in the diagnostic algorithm of embryonal tumours and for the subtyping of paediatric and adult-type HGG.

CNS erythroblastic sarcoma: a potential emerging pediatric tumor type characterized by NFIA::RUNX1T1/3 fusions.

Erythroblastic sarcoma (ES) (previously called chloroma or granulocytic sarcoma) are rare hematological neoplams characterized by the proliferation of myeloid blasts at extramedullary sites, and primarily involve the skin and soft tissue of middle-aged adults. ES may be concomitant with or secondary to myeloid neoplasms (mostly acute myeloid leukemia (AML)) or in isolated cases (de novo) without infiltration of the bone marrow by blasts. ES share cytogenetic and molecular abnormalities with AML, including RUNX1T1 fusions. Some of these alterations seem to be correlated with particular sites of involvement. Herein, we report an isolated erythroblastic sarcoma with NFIA::RUNX1T1 located in the central nervous system (CNS) of a 3-year-old boy. Recently, two pediatric cases of CNS MS with complete molecular characterization have been documented. Like the current case, they concerned infants (2 and 3 years-old) presenting a brain tumor (pineal involvement) with leptomeningeal dissemination. Both cases also harbored a NFIA::RUNX1T3 fusion. ES constitutes a diagnostic challenge for neuropathologists because it does not express differentiation markers such as CD45, and may express CD99 which could be confused with CNS Ewing sarcoma. CD43 is the earliest pan-hematopoietic marker and CD45 is not expressed by erythroid lineage cells. E-cadherin (also a marker of erythroid precursors) and CD117 (expressed on the surface of erythroid lineage cells) constitute other immunhistochemical hallmarks of ES. The prognosis of patients with ES is similar to that of other patients with AML but de novo forms seem to have a poorer prognosis, like the current case. To conclude, pediatric ES with NFIA::RUNX1T1/3 fusions seem to have a tropism for the CNS and thus constitute a potential pitfall for neuropathologists. Due to the absence of circulating blasts and a DNA-methylation signature, the diagnosis must currently be made by highlighting the translocation and expression of erythroid markers.