A case series of Diffuse Glioneuronal Tumours with Oligodendroglioma‐like features and Nuclear Clusters (DGONC)

In this study, we report three paediatric cases of Diffuse Glioneuronal Tumours with Oligodendroglioma‐like features and Nuclear Clusters (DGONC).


| 465
A CASE SERIES OF DIFFUSE GLIONEURONAL TUMOURS WITH OLIGODENDROGLIOMA-LIKE FEATURES AND NUCLEAR CLUSTERS (DGONC) loss of 19q (35%), but unlike well-characterised glioneuronal tumours, no other defining genetic alterations were identified.
We reviewed our existing paediatric CNS tumour cohorts to determine if we could identify additional cases of DGONC. We reviewed 123 cases with an undetermined final diagnosis originating from two previously described cohorts with existing methylation array data. 2 From these, we selected tumours that failed to classify using the most up-to-date version of the methylation classifier developed by the DKFZ 3 (MNPv11b6) and that had monosomy of chromosome 14. 1 We then performed unsupervised hierarchical clustering using the 10,000 most variably methylated probes for our candidate-DGONC samples together with a cohort of 18 DGONC reported in the original article, 1 along with tumours with a different confirmed molecular diagnosis. 2,3 Three samples reported were found to have methylation profiles with good similarity to Deng et al.'s published cases ( Figure 1A, labelled GOSH DGONC) and one patient was later identified as having a second tumour biopsy taken a year later, which also clustered with the reference DGONC cohort.
All patients with a proposed DGONC were male children and, at the time of diagnosis, aged between 11 and 12 years. Case DGONC_1 had been diagnosed as a "malignant glioneuronal tumour", while DGONC_2 and 3 had both been called "high-grade neuroepithelial tumour". These tumours presented as well-defined cortical or subcortical supratentorial masses, and as previously described did not display any specific lobar predisposition. From the available radiology ( Figure 1C and Figure S1), the tumours had no appreciable perilesional oedema, were hyperintense compared to cortex on T2 and FLAIR weighting and enhanced poorly with contrast if at all. They also demonstrated internal matrix calcification, with centrally low ADC values on diffusion-weighted images. Imaging for the relapsed case indicated residual tumour but did not show additional radiological features.
All patients are alive and had relatively favourable outcomes, although follow-up time following diagnosis varied between only 0.7 and 6.6 years. Following surgery, all patients received craniospinal radiation together with chemotherapy and are in complete remission. All patients received a gross total resection, the additional surgery for DGONC_1 was to remove persistent residual tumour following chemotherapy and there was no notable progression on radiology ( Figure S1). One patient had a previous diagnosis and family history of Autistic Spectrum disorder and Hyperactivity, the other two patients had no known underlying conditions. Morphologically all three cases had remarkably similar characteristics (see Figure 1D-I and Table S1). They were diffusely arranged tumours composed of round cells with perinuclear haloes, akin to oligodendrogliomas or neurocytic tumours. There were nuclear clusters, at least some of which were part of multinucleate cells.
Calcification, ganglion cells, apoptosis and foamy cells were commonly noted. Vascular proliferation was present in two of the four samples. The tumours were of moderate cellularity and the mitotic index ranged between 0.42 and 3.38 per mm 2 , with a focal maximum of 30% of cells staining positive for Ki67. The tumour cells did not express astrocytic markers (e.g. GFAP) but there was diffuse strong staining for neuronal markers Synaptophysin and NeuN, as well as the oligodendrocyte marker OLIG2. CD34 staining was present in the endothelium only. We did not observe histological differences between the primary and secondary DGONC_1 sample.
Apart from the methylation arrays, existing molecular workup did not identify recurrent molecular events (Table S1). Where data were available, no pathogenic variants were observed in the histone H3 genes, and no pathogenic variants or fusion events were detected in BRAF. Targeted panel sequencing on DNA and RNA was available for one case (a custom panel 4 and Illumina TruSight RNA Pan-Cancer fusion Panel, respectively) but failed to reveal any predicted driving molecular events. In addition to monosomy of chromosome 14, on review of the copy number plots, we noted a common focal loss of chromosome 1p (previously not reported) and loss of chromosome 3p.
In conclusion, we have independently identified three DGONCs with consistent histopathological, molecular and clinical features to those previously described. This study confirms that DGONC is a molecularly defined entity with a unique methylation profile   All authors reviewed and accepted the final manuscript.

E TH I C A L A PPROVA L
This study was covered under ethical approval granted by BRAIN UK tissue bank (REC: 14/SC/0098, references 16/007 and 17/007).

PEER R E V I E W
The peer review history for this article is available at https://publo ns.com/publo n/10.1111/nan.12680.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request. This article was peer-reviewed by independent, anonymous expert referees, and the authors (including TSJ and SMLP) had no role in either the editorial decision or the handling of the paper.

R E FE R E N C E S S U PP O RTI N G I N FO R M ATI O N
Additional supporting information may be found online in the Supporting Information section.

Fig S1
Table S1 How to cite this article: Pickles