Children and young patients are usually affected, and no gender predominance is recognised. It accounts for around 2% (from 0.4-3.8%) of all primary intracranial tumors, and up to 10% of primary cerebral tumors in children.
Because of their rarity, large-scale, population-based studies focusing on epidemiology and outcomes are lacking.
Ganglioglioma is a benign slow-growing neoplasm that most frequently occurs at the supratentorial region. Nevertheless, there are occasional reports of ganglioglioma occurring in thebrainstem and spinal cord.
Molecular pathogenesis, risk factors for malignant progression, and their frequent association with drug-resistant focal seizures remain poorly understood. This contrasts recent progress in understanding the molecular-genetic basis and targeted treatment options in diffuse gliomas. The Neuropathology Task Force of the International League against Epilepsy examined available literature to identify common obstacles in diagnosis and research of LEAT. Analysis of 10 published tumour series from epilepsy surgery pointed to poor interrater agreement for the histopathology diagnosis. The Task Force tested this hypothesis using a web-based microscopy agreement study. In a series of 30 LEAT, 25 raters from 18 countries agreed in only 40% of cases. Highest discordance in microscopic diagnosis occurred between GG and DNT variants, when oligodendroglial-like cell patterns prevail, or ganglion cells were difficult to discriminate from pre-existing neurons. Suggesting new terminology or major histopathological criteria did not satisfactorily increase the yield of histopathology agreement in 4 consecutive trials. To this end, the Task Force applied the WHO 2016 strategy of integrating phenotype analysis with molecular-genetic data obtained from panel sequencing and 450k methylation arrays. This strategy was helpful to distinguish DNT from GG variants in all cases. The Task Force recommends, therefore, to further develop diagnostic panels for the integration of phenotype-genotype analysis in order to reliably classify the spectrum of LEAT, carefully characterize clinically meaningful entities and make better use of published literature 1).
Imaging findings mirror the various patterns of growth which these tumours may demonstrate and thus their appearance is very variable. Partially cystic mass with an enhancing mural nodule is seen in ~45% of cases. They may also simply present as a solid mass expanding the overlying gyrus. An infiltrating mass is uncommon and may reflect higher grade.
Findings are of a mass which is often non-specific. General features include:
iso- or hypodense
frequently calcified ~35%
bony remodelling or thinning can indicate the slow growing nature of the tumour
enhancement is seen in approximately 50% of cases (involving the solid non-calcified component)
Reported MR signal characteristics include:
T1 C+ (Gd) solid component variable contrast enhancement
T2 hyperintense solid component variable signal in the cystic component depending on the amount of proteinaceous material or the presence of blood products peritumoural FLAIR/T2 oedema is distinctly uncommon
T2* (GE/SWI) calcified areas (common) will show blooming signal loss
Main differential diagnosis is that of other cortical tumours, with helpful distinguishing features including:
contrast enhancement uncommon 'bubbly appearance' common
contrast enhancement prominent dural tail sign is often seen
Desmoplastic infantile astrocytoma and ganglioglioma
young children dural involvement prominent large often multiple lesions
If in the spinal cord consider:
The aim of a study was to evaluate whether ganglioglioma (GGL), dysembryoplastic neuroepithelial tumour (DNET) and FCD (focal cortical dysplasia) are distinguishable through diffusion tensor imaging. Additionally, it was investigated whether the diffusion measures differed in the perilesional (pNAWM) and in the contralateral normal appearing white matter (cNAWM). Six GGLs, eight DNETs and seven FCDs were included in this study. Quantitative diffusion measures, that is, axial, radial and mean diffusivity and fractional anisotropy, were determined in the lesion identified on isotropic T2 or FLAIR-weighted images and in pNAWM and cNAWM, respectively. DNET differed from FCD in mean diffusivity, and GGL from FCD in radial diffusivity. Both types of glioneuronal tumours were different from pNAWM in fractional anisotropy and radial diffusivity. For identifying the tumour edges, threshold values for tumour-free tissue were investigated with receiver operating characteristic analyses: tumour could be separated from pNAWM at a threshold ≤ 0.32 (fractional anisotropy) or ≥ 0.56 (radial diffusivity) *10-3 mm2/s (area under the curve 0.995 and 0.990 respectively). While diffusion parameters of FCDs differed from cNAWM (radial diffusivity (*10-3 mm/s2): 0.74 ± 0.19 vs. 0.43 ± 0.05; corrected p-value < 0.001), the pNAWM could not be differentiated from the FCD 2).
Advances in the immunohistochemical detection of neuron-specific and neuronal-associated antigens have resulted in the discovery of neuronal elements in certain primary human brain tumors. The results have been not only to expand what neuropathologists commonly recognize as gangliogliomas, including the tumors now known as glioneurocytic tumor with neuropil rosettes and papillary ganglioneuroma, but also to expand the spectrum of tumor types to now include tumors such as central neurocytoma, dysembryoplastic neuroepithelial tumor, and desmoplastic infantile ganglioglioma.
At this stage, no criteria for WHO II gangliogliomas have been established 7).
Gangliogliomas, as their name suggests, are composed of two cell populations:
ganglion cells (large mature neuronal elements): ganglio- neoplastic glial element: -glioma primarily astrocytic, although oligodendroglial or pilocytic astroctytoma components are also enountered 9 The proportion of each component varies widely, and it is the grade of the glial component that determines biological behaviour.
Dedifferentiation into high-grade tumours does occasionally occur, and it is usually the glial component (into a GBM). Only rarely is it the neuronal component (into a neuroblastoma).
They are closely related to both gangliocytomas (which contain only the mature neural ganglion cellular component) and ganglioneurocytoma (which also have small mature neoplastic neurones).
Neuronal origin is demonstrated by positivity to neuronal markers:
Neurofilament protein: positive
Chromogranin-A: positive (usually negative in normal neurones)
CD34: positive in 70-80%
The glial component may also show cytoplasmic positivity for GFAP.
Ganglioglioma and pleomorphic xanthoastrocytoma were the histologic types with the strongest association with CD34 positivity with an odds ratio of 9.2 and 10.4, respectively, compared with dysembryoplastic neuroepithelial tumors in Low-Grade Epilepsy-Associated Tumors 8).
BRAFV600 mutations are frequently found in several glioma subtypes, including pleomorphic xanthoastrocytoma (PXA) and ganglioglioma and much less commonly in glioblastoma.
Gross total resection is achieved in the majority of cases.
If only incomplete resection is achievable, or tumour recurrence occurs then radiotherapy may be of some benefit.
In the surgical treatment of temporal lobe epilepsy with mesial temporal lobe tumor, whether to remove the hippocampus aiming for a better seizure outcome in addition to removing the tumor is a dilemma. Two pediatric cases treated successfully with tumor removal alone are presented.
The first case was an 11-year-old girl with a ganglioglioma in the left uncus, and the second case was a 9-year-old girl with a pleomorphic xanthoastrocytoma in the left parahippocampal gyrus. In both cases, the hippocampus was not invaded, merely compressed by the tumor. Tumor removal was performed under intraoperative electrocorticography (ECoG) monitoring. After tumor removal, abnormal discharges remained at the hippocampus and adjacent temporal cortices, but further surgical interventions were not performed. The seizures disappeared completely in both cases.
When we must decide whether to remove the hippocampus, the side of the lesion, the severity and chronicity of the seizures, and the presence of invasion to the hippocampus are the factors that should be considered. Abnormal discharges on ECoG at the hippocampus or adjacent cortices are one of the factors related to epileptogenicity, but it is simply a result of interictal irritation, and it is not an absolute indication for additional surgical intervention 9).
Gangliogliomas and ganglioneuromas are slow growing benign tumors.
Local resection is the treatment of choice and determines prognosis. In the brain, where a reasonable resection margin can be achieved, the prognosis is good, with recurrence-free survival reported to be 97% at 7.5-year follow-up 10).
In contrast, in the spinal cord where complete resection is often not possible without devastating deficits, local recurrence is very common.
Although the majority of patients have an excellent prognosis, infants and patients with brainstem tumors have worse survival rates.
27 patients with drug resistant epilepsy and brain tumor, aged up to 19 years at the time of surgery, were studied between 1996 and 2013 and followed up for at least one year. The mean interval between the onset of seizures and the diagnosis of the tumor was 3.6 years, and from diagnosis to the surgery, 18 months. The location of the tumor was in the temporal lobe in 16, with ganglioglioma and dysembryoplastic neuroepithelial tumors being the most frequent. Among the patients, 92.5% and 90.4% were seizure-free in the first and fifth year after surgery, respectively. Twelve of 16 children were successful in becoming drug-free, with complete withdrawal by 3.2 years. Surgery proved to be potentially curative and safe in these cases, suggesting that the tumor diagnosis and surgery cannot be postponed 11).
Thirty-seven children were identified, with a median age at presentation of 8.2 years and median follow-up of 38.0 months. Eighteen tumors (48.6%) were typical and 19 (51.4%) were atypical. All typical lesions presented with seizures, whereas no atypical lesions did so. Sixteen (88.9%) typical lesions were located in the temporal lobe. In the atypical group, tumor location was variable, including 11 (57.9%) in the brainstem. Death during follow-up was statistically more common in the atypical group (31.6% vs 0%, p = 0.02). Gross-total resection (GTR) was achieved for 15 of 16 typical tumors (93.8%), compared with 3 atypical tumors (15.8%, p < 0.0001). Presentation with seizure or non-brainstem location were each associated with survival (p = 0.02 and 0.004, respectively). The presence of mutation in BRAF exon 15 did not differ between the 2 groups. Pediatric GG with typical imaging features is associated with excellent rates of GTR and overall survival. Atypical GG is commonly encountered, less amenable to GTR, and associated with a worse outcome. This may relate to anatomical or biological characteristics and merits further investigation 12).
348 children with low-grade GGs diagnosed from 2004 to 2010, with a median follow-up of 37 months. Tumors were more prevalent in males (n = 208, 59.8%) than females (n = 140, 40.2%) (P < .001). Almost 63% occurred in children >10 years, whereas only 3.5% were found in those <1 year old. Approximately 50% were located in the temporal lobes, and only 3.7% and 3.5% were located in the brainstem and spinal cord, respectively. Surgery was performed on 91.6% of cases, with gross total resection achieved in 68.3%. Radiation was used in 3.2%. Young age (<1 year) and brainstem location were associated with worse overall survival.
Low-grade GGs occur in older children with a male preference 13).
Rosselló et al. described a progression from ganglioglioma to this composite anaplastic entity after 32 months of follow-up, with apparently nontumoral parenchyma separating the 2 components. Polymerase chain reaction showed a wild-type BRAF gene. Seven months after concomitant chemoradiotherapy, radiologic progression led to a second line of chemotherapy, and a third line of chemotherapy was initiated after a subsequent progression at 11 months.
This case may add some evidence in favor of the glioneuronal maldevelopment hypothesis to explain the oncogenesis of these neuroepithelial tumors 14).
report a rare case of the craniocervical ganglioglioma. A 3.5-year-old male, presented with severe progressive quadriparesis, gait disturbance, and sphincter deficit. Physical examination demonstrated the quadriparesis, associated with positive Hoffman, Babinski, and clonus signs, and increased respond of deep tendon reflexes. Magnetic resonance imaging (MRI) demonstrated an ill-defined mass within medulla and upper cervical spinal cord, which was hypo to iso signal on T1, heterogeneous iso to hypersignal on T2 and demonstrated marked bright enhancement on T1 with gadolinium (Gad) injection. On surgery, the mass had a soft texture, ill-defined border, and grey to brown appearance. According to the frozen section report, and due to the absence of the tumour-neural parenchymal interference, only decompression of the tumour and expansile duraplasty were performed. The histopathology revealed ganglioglioma. On last follow-up 14 months after surgery, the patient was asymptomatic and neurological status was improved. The craniocervical MRI demonstrated the tumour that did not grow. Although it is rare, the ganglioglioma should be in the differentiated diagnoses of tumours with compatible clinical and radiologic features even in the unusual locations, especially in the pediatric and young patients. Safety surgical resection should be considered in these patients, whenever possible. In the case of partial resection, that is common in the tumours located within functionally critical structures, long close follow-up rather than radiation therapy is required 15).
Magnetic resonance imaging of a 67-yr-old woman presenting with dysphasia revealed a noncontrast enhancing left-sided lesion in the frontal and parietal pars opercularis. Due to the location of the lesion, nTMS was used to chart both primary motor and language cortex, utilizing this information to plan a safe SB trajectory and sampling area according to the initial work-up recommendations from the multidisciplinary neuro-oncology board. The SB was uneventful, with histology revealing a ganglioglioma, WHO I. The patient was discharged the following day, having declined to proceed with tumor resection (awake surgery) due to the non-negligible risk of morbidity. Upon 1- and 3-mo follow-up, she showed no signs of any procedure-related deficits.
nTMS can be implemented to aid with the planning of a stereotactic biopsy procedure in eloquent areas of the brain, and should be considered part of the neurosurgical armamentarium 16).