This article reviews the variation in imaging, histopathology, clinical course, and management seen with central neurocytomas (CNs). CNs have often been misdiagnosed as oligodendrogliomas and ependymomas; however, synaptophysin positivity can correctly diagnose these neurocytic neoplasms. Atypical CNs, an important variant first described in 1997, are marked by increased proliferative potential and associated with worse clinical outcomes in terms of long-term survival and local tumor control. Complete surgical resection is the cornerstone of therapy, and postoperative radiation is recommended in the setting of residual disease. Other less aggressive variants of central neurocytomas, including liponeurocytomas, ganglioneurocytomas, and pigmented neurocytomas, are also discussed.
Key points
- •
Central neurocytomas (CNs) can be variable in terms of location, histology, and imaging, and can be morphologically similar to oligodendrogliomas, ependymomas, pineocytomas, and neuroblastomas.
- •
CNs are seen in the ventricular system and extraventricular locations, and rarely in usual locations such as the spinal cord, retina, and pons.
- •
Atypical central neurocytomas are defined as having a high proliferative index (MIB-1 index >2%) with or without vascular proliferation, increased mitotic figures, and necrosis.
- •
Atypical CNs have higher recurrence and lower disease-free survival rates.
- •
Other rare histopathologic variants of CNs include ganglioneurocytoma, liponeurocytoma, myoneurocytoma, and pigmented neurocytoma, which have a similar clinical course to classical CNs and need to be differentiated from other primary central nervous system tumors.
Introduction
Central neurocytomas (CNs) are benign intraventricular tumors that occur most often in young adults and constitute 0.1% to 0.5% of all primary brain tumors. First described by Hassoun and colleagues in 1982, CNs were believed to have a cellular origin from the septum pellucidum. Recent investigations suggest that they derive from neuroglial precursor cells in the subependymal plate of the lateral ventricles and circumventricular organs with characteristics of dual differentiation along glial or neuronal lines (predominates). The neurocytes composing these neurocytomas are similar to cells in the dentate layer of the hippocampus or internal granular layer of the cerebellum. Their neuroectodermal origin allows them to differentiate along neurocytic or glial lineage.
Although primarily described as an intraventricular tumor, numerous cases of extraventricular neurocytoma (EVN) locations have been reported, namely cortex, cerebellum, olfactory bulb, insula, thalamus, pineal gland, retina and spinal cord. Under light microscopy, tumor cells have a honeycomb-like arrangement, small round or oval nuclei, scant cytoplasm, and a perinuclear halo, all of which can resemble oligodendroglioma, and large fibrillary areas resembling pineocytoma. Straight-line arrangement of cells and pseudorossette formation can give the appearance of clear cell ependymoma in some samples. CNs were probably historically misdiagnosed because of these similarities.
Determining the neuronal origin of tumor cells is crucial to definitively diagnose CNs. This assessment is accomplished using immunohistochemistry for synaptophysin or neuron-specific enolase and electron microscopy.
Although classic CN is present in the ventricles, occurs in young adults, and has uniform sheets of cells with neuronal differentiation, its clinical presentation, anatomic location, and cytopathologic characteristics and outcomes can vary. This review highlights the extent of variation and atypia seen in CN cases and how these factors could potentially affect their management.
Variation in location: neurocytomas
CNs are primarily intraventricular tumors (lateral ventricle, third ventricle, or near the foramen of Monro) that present with increased intracranial pressure and symptoms secondary to obstructive hydrocephalus. Rare variants, however, can occur in various sites of the central nervous system (CNS). EVNs are a well-known atypical subset of neurocytomas, and can be present in both supratentorial and infratentorial cortical locations. Cortical neurocytomas have been reported in frontal, parietal, temporal, and occipital lobes and in the insula. In a series of 6 patients, Sgouros and colleagues reported 2 cases of neurocytoma arising from the thalamus and extending to the ventricles. In a report on radiologic features of these tumors, Ng and colleagues introduced a case arising from the pineal gland. Rarely, EVNs have been reported to arise from the spinal cord. These variations in location are detailed elsewhere in this issue. EVNs can also occur in the cerebellum.
Introduction
Central neurocytomas (CNs) are benign intraventricular tumors that occur most often in young adults and constitute 0.1% to 0.5% of all primary brain tumors. First described by Hassoun and colleagues in 1982, CNs were believed to have a cellular origin from the septum pellucidum. Recent investigations suggest that they derive from neuroglial precursor cells in the subependymal plate of the lateral ventricles and circumventricular organs with characteristics of dual differentiation along glial or neuronal lines (predominates). The neurocytes composing these neurocytomas are similar to cells in the dentate layer of the hippocampus or internal granular layer of the cerebellum. Their neuroectodermal origin allows them to differentiate along neurocytic or glial lineage.
Although primarily described as an intraventricular tumor, numerous cases of extraventricular neurocytoma (EVN) locations have been reported, namely cortex, cerebellum, olfactory bulb, insula, thalamus, pineal gland, retina and spinal cord. Under light microscopy, tumor cells have a honeycomb-like arrangement, small round or oval nuclei, scant cytoplasm, and a perinuclear halo, all of which can resemble oligodendroglioma, and large fibrillary areas resembling pineocytoma. Straight-line arrangement of cells and pseudorossette formation can give the appearance of clear cell ependymoma in some samples. CNs were probably historically misdiagnosed because of these similarities.
Determining the neuronal origin of tumor cells is crucial to definitively diagnose CNs. This assessment is accomplished using immunohistochemistry for synaptophysin or neuron-specific enolase and electron microscopy.
Although classic CN is present in the ventricles, occurs in young adults, and has uniform sheets of cells with neuronal differentiation, its clinical presentation, anatomic location, and cytopathologic characteristics and outcomes can vary. This review highlights the extent of variation and atypia seen in CN cases and how these factors could potentially affect their management.
Variation in location: neurocytomas
CNs are primarily intraventricular tumors (lateral ventricle, third ventricle, or near the foramen of Monro) that present with increased intracranial pressure and symptoms secondary to obstructive hydrocephalus. Rare variants, however, can occur in various sites of the central nervous system (CNS). EVNs are a well-known atypical subset of neurocytomas, and can be present in both supratentorial and infratentorial cortical locations. Cortical neurocytomas have been reported in frontal, parietal, temporal, and occipital lobes and in the insula. In a series of 6 patients, Sgouros and colleagues reported 2 cases of neurocytoma arising from the thalamus and extending to the ventricles. In a report on radiologic features of these tumors, Ng and colleagues introduced a case arising from the pineal gland. Rarely, EVNs have been reported to arise from the spinal cord. These variations in location are detailed elsewhere in this issue. EVNs can also occur in the cerebellum.
Histopathologic variation
Tumors Mimicking Neurocytomas: Neurocytic Features
CNs overlap morphologically with oligodendrogliomas, neuroblastomas, pineocytomas, ependymomas, and dysembryoplastic neuroepithelial tumors (DNETs) ( Figs. 1–3 ). These tumors all share histology with sheets of uniform round cells. True CNs, however, are neuroepithelial tumors with neuronal differentiation, thus differentiating them from oligodendrogliomas, neuroblastomas, pineocytomas, and ependymomas, which have oligodendroglial, neuroblastic, pineocytic, and ependymal differentiation pathways, respectively. Neuronal markers, such as synaptophysin (predominantly in fibrillary zones and perivascular nuclei-free cuffs) and NeuN (nuclear staining), are specific for CNs in almost all cases. Rarely, variants may be encountered with staining for multiple cell lines, including neurocytes, oligodendrocytes, and neuroblasts. Other rare neurocytoma variants may demonstrate a pseudopapillary pattern with hyalinized vascular cores but maintain synaptophysin positivity. Hassoun and colleagues considered these overlaps with oligodendrogliomas, pineocytomas, and cerebral neuroblastomas in his landmark electron microscopic description of CNs. Electron micrographs can thus be used where immunohistochemistry is inconclusive in separating these closely related entities. Typical CNs will show round nuclei, distinct nucleolus with dispersed chromatin, cytoplasm with lamellated appearance, and numerous cellular processes. A prominent glycine and choline peak with magnetic resonance spectroscopy may be used to differentiate CNs from oligodendrogliomas and other morphologically similar tumors.
Atypical and malignant features
Since the original description of CNs in 1982, several different variants have been reported, with the atypical CN as the best described category. Typical CNs are slow-growing and homogenous tumors with calcifications, pseudorosette formation, and benign histologic appearance. The MIB-1 labeling index, a proliferative cell marker, is usually less than 2%. However atypical and anaplastic features may be observed in up to 20% of CN cases. These anaplastic features include elevated abnormal mitotic figures, microvascular proliferation, and focal necrosis. von Deimling and colleagues studied the differentiation patterns in 11 CN cases and found 2 malignant variants demonstrating vascular proliferation and marked mitotic figures. Similarly, Yasargil and colleagues observed 2 similar malignant variants in his series with vascular proliferation and necrosis. Earlier on, some authors used the terms anaplastic or malignant to describe neurocytomas with features of vascular proliferation, necrosis or mitotic activity. However, an MIB-1 index of greater than 2% is more consistently used to describe atypical neurocytomas. These cases were subsequently found to have a shorter recurrence-free interval, which may be related to dissemination and poor outcome.
Soylemezoglu and colleagues analyzed CN samples in 36 patients using the MIB-1 labeling index, and found a wide variation in their proliferative potential. In their study, 39% of samples with an MIB-1 index greater than 2% had a correlation with vascular proliferation. Furthermore, a 63% relapse rate was found in these atypical cases versus a 22% relapse rate in cases with an MIB-1 index less than 2%. Since this first detailed description of atypical CNs in 1997, numerous case reports and small case series have described similar increased MIB-1 indices and atypical histopathologic features. Proliferation potential is a more useful predictor of outcomes based on several studies, rather than just histologic atypia. Atypical neurocytomas have a worse prognosis in terms of local tumor control and long-term survival. Mackenzie also observed the lack of correlation between histologic atypia and MIB-1 index, and therefore proposed the term proliferating neurocytoma , because only the MIB-1 index was found to correlate with the clinical behavior of the tumor. Furthermore, she suggested that an MIB-1 index greater than 2% alone without histologic atypia could suffice in making these tumors atypical grade II. Sharma and colleagues also found no correlation between clinical outcomes and histologic atypia in their group of 20 patients. However, no correlation of outcomes with proliferation potential was seen either, even though only 1 patient in their series had an MIB-1 index greater than 2%.
The guidelines for this spectrum of atypia and proliferation observed in CNs are not well defined. Based on the latest World Health Organization (WHO) classification of CNs, they are still classified as WHO grade II in presence of atypia and increased proliferative index, similar to classic CNs. In a meta-analysis of 438 patients with CNs, Rades and colleagues defined atypical CNs with MIB-1 index greater than 3% as the prognostic cutoff for local control and survival. This study found the overall 10-year survival rate to be 97% in patients with typical CN versus 70% in those with atypical CN, whereas the local control rates at 10 years were found to be 72% and 46%, respectively. Therefore, based on this meta-analysis, patients with typical neurocytomas had superior 5- and 10-year survival rates and local tumor control.
These atypical neurocytomas are not neuroblastomas but rather aggressive forms of CN. Atypical CNs have cellular morphology, which is otherwise similar to classic CNs, and do not generally contain undifferentiated neuroblasts. The synaptic processes are well developed in atypical CNs in comparison with neuroblastomas. Favereaux and colleagues described 2 atypical CNs with an MIB-1 index of greater than 5.2% and both recurred. They suggested that atypical CNs are an intermediate spectrum category between classical CNs and cerebral neuroblastomas, which has also been suggested by other series ( Fig. 4 ). Significant controversy continues regarding the nomenclature for some malignant neurocytomas with multiple cell populations, and sometimes diagnosis can be challenging, as demonstrated by Vallat-Decouvelaere and colleagues in 2 cases of EVNs.
