|Current issue Archive About the journal Abstracting and indexing Subscription Contact Instructions for authors||
Giant cell ependymoma of the spinal cord and fourth ventricle coexisting with syringomyelia
Grażyna M. Szpak, Eliza Lewandowska, Bogna Schmidt-Sidor, Elżbieta Pasennik, Joanna Modzelewska, Tomasz Stępień, Grzegorz Zdaniuk, Jerzy Kulczycki, Teresa Wierzba-Bobrowicz
Folia Neuropathol 2008; 46 (3): 220-231
- 07szpak.pdf [0.89 MB]
The World Health Organisation (WHO), apart from four ependymoma patterns (classic or conventional, anaplastic, myxopapillary and subependymoma) and its four histopathological subtypes (cellular, papillary, clear cell and tanycytic) recognized several “other rare patterns with variable differentiation” including an exceptional rare variant termed “giant cell ependymoma” (GCE) . According to the WHO grading system, the pathological spectrum of ependymoma might be very heterogenous. Heterogeneity of the neoplastic cell population within the same tumour is also a characteristic feature of other exclusive types and variants of ependymoma [7,22].
Participation of giant multinucleated cells in ependymoma cell arrangement is regarded as a GCE hallmark. To distinguish GCE from other tumours which exhibit giant cell participation, perivascular pseudorosettes and coexisting areas of typical ependymoma and giant cells should be identified in light microscopy. In uncertain cases, electron microscopy remains a gold diagnostic standard [4,7,11]. Detailed histological, immunohistochemical and ultrastructural examination were performed to evaluate the still rather controversial differentiation and malignancy of GCE [1,14]. Recently reported molecular studies of spinal and intracranial ependymomas might help establish their classification .
To date, only a few cases of giant cell ependymoma have been reported [1,3,4,6,9,15,24,27,32]. The case presented in this paper is the first such widespread GCE co-existing with syringomyelia.
A 28-year-old man was admitted to the Neurological Department, Institute of Psychiatry and Neurology, with tetraplegia and signs of increased intracranial pressure, eight months after surgical spinal cervical decompression without tetraplegia improvement. Twenty-four months before entry, increasing neck pain emerged as the first sign of the disease. Twelve months later it was followed by progressive weakness of upper and lower extremities. On admission magnetic resonance imaging (MRI) revealed widespread fourth ventricle and intraspinal tumour (Fig. 1A) associated with syringomyelia. The patient died suddenly with signs of respiratory insufficiency.
Material and Methods
For histological and immunohistochemical (IHC) examination, samples obtained from the brain autopsy material were fixed in 4% formaldehyde buffered to pH 7.4 and embedded in paraffin. H&E, PTAH, PAS, Klüver-Barrera stainings and IHC reactions according to the labelled streptavidin-biotin complex method with DAB as chromogen were performed in 5 µm sections using antibodies to GFAP, CD68, vimentin, EMA, Ki67, PCNA, NSE, synaptophysin and cytokeratin (all antibodies from Dako).
For electron microscopic examination, fragments of tumour were taken from the paraffin block. After deparaffinizing and washing in water, the material was processed routinely for ultrastructural examination. Ultrathin sections were stained with uranyl acetate and lead citrate, and examined using an Opton 109 DPS electron microscope. For electron microscopic examinations, the samples were retrieved from the material primarily fixed in formaldehyde and embedded in paraffin; therefore, their quality is limited.
The brain autopsy revealed supratentorially moderate hydrocephalus, whereas a pearl-grey tumour loosely filling the lower half of the fourth ventricle and extending intraparenchymally to segments C3-C4 of the spinal cord was found infratentorially. There were uni- and multilocular syrinx cavities along the cervical and thoracic spinal cord (from C4-C5 to Th6) below the lower part of the tumour tissue (Fig. 1B-D).
Histological and immunohistochemical findings
PTAH staining revealed that the upper part of the tumour loosely filling the lower half of the fourth ventricle was separated from both the ventricle lumen and the medulla oblongata by capsule-like glial fibrillar fascicles. Tumour cystic masses compressed the floor of the fourth ventricle, tightening foramen of Luschka regions (Fig. 2A).
Sharply demarcated cervical nodular parts divided by numerous clefts, not infiltrating the medullar parenchyma, extended intraparenchymally into the glial stem (Fig. 2B-C). Strong PTAH and GFAP-positive structure of the glial stem was revealed in segment C4 of the spinal cord below the lower part of the intramedullary tumour and above the syrinxes (Fig. 2C-D). Multifocal ischaemic necrotic fields in resorptive stage with CD68-positive macrophages surrounding thick and hyalinised vascular walls were dispersed within the glial stem (Fig. 2Dd). The widespread syrinxes, extending below the glial stem, were not connected with the central canal in their course along the cervical and thoracic spinal cord (Fig. 2E). Their walls were composed of loose or more dense glial fibres and only a few astroglial cells (Fig. 2E-F). Swelling of myelin sheets and neuron loss were found adjacent to as well as below the tumour masses and syrinx cavities. Their cell infiltration was not found.
In all intraventricular and medullary tumour specimens, perivascular pseudorosettes, mainly composed of elongated medium-size cells of ependymoma, predominated over pseudorosettes made of monster giant cells, representing the heterogeneous cell population.
Some pseudorosettes of the discohesive pattern formed papillae-like structures containing GFAP-positive and GFAP-negative neoplastic cells (Fig. 3A-B). Elongated medium-size and smaller cells with eosinophilic cytoplasm and round or oval-shaped nuclei radiating toward the blood vessels formed the majority of pseudorosettes (Fig. 3C).
The same elongated medium-size cells with nuclear hyperchromasia, radiating toward the lumen, formed a lining of numerous clefts or channels (Fig. 3D) and a few ependymal true rosettes (Fig. 3E). Their formation from cells of the spinal cord central canal was revealed in the specimens of the cervical glial stem (Fig. 3F). The neoplastic cells, including giant cells, were cytokeratin, EMA, synaptophysin and NSE-negative.
In some pseudorosettes and cell clusters on the periphery of all nodular segments, pleomorphic round or multiform, frequently multinuclear, giant cells with bizarre and hyperchromatic nuclei were seen (Fig. 4A-C). Their intranuclear inclusion-like bodies or “pseudoinclusions”, probably result in intranuclear cytoplasm invaginations were eosinophilic and GFAP-positive or negative similar cytoplasm-like reaction respectively (Fig. 4D-F). Mitotic activity and Ki67 index of all tumour segments was low in both small and giant cells. A few mitoses mainly in medium-size cells were found (Fig. 3C).
The moderate tumour vascularity contained thin-wall central vessels of pseudorosettes and of fibrillar fascicles between lobulated nodular structures. Focal necroses did not reveal pseudopalisade features. There were multifocal ischaemic necroses either in coagulative or in resorptive stages around vessels with at different times vascular thrombi or hyalinized walls in ependymoma and in the glial stem (Fig. 5A-B). Microvascular glomerular proliferation was identified only beside ependymoma tissue on the tumour periphery and in the glial stem (Fig. 5C). Fibrillar structures of the glial stem composed of elongated neoplastic cells, resembling pilocytic astrocytoma with Rosenthal fibres, were revealed in segment C4 of the spinal cord (Fig. 5C). Syrinx cavities were observed along the cervical and thoracic segments of the spinal cord below ischaemic focal necroses of the glial spinal stem tissue (Fig. 2E-F).
Electron microscopy revealed numerous blepharoplasts (ciliary basal bodies) in ependymoma cells. These blepharoplasts showed morphological alterations. The structure of their 9 typical peripherally-placed microtubule doublets was often unrecognizable (Fig. 6A-B). Peripheral microtubules were visible only in some blepharoplasts (Fig. 6C-D). Junctional complexes between adjacent cells were long and tortuous (Fig. 7). Giant cells frequently showed bizarre nuclei with characteristic invaginations of the cytoplasm, which were separated from the chromatin by nuclear membrane (Fig. 8A-C). Some of the giant cells were multinucleated (Fig. 9).
This paper presents a case of widespread intramedullary ependymoma arising from the central canal spinal cord that extended along the cervical segments from C3/C4 to the lower half of the fourth ventricle, with coexisting syringomyelia of cervical and thoracic segments from C5 to Th6. This slow-growing, low-grade malignant tumour with degenerative cystic, vascular and ischaemic changes exhibited unusual polymorphic morphology.
All evaluated tumour segments featured the coexistence of different types of tumour cells forming arrangements characteristic of ependymoma. Clefts or channels, true ependymal rosettes and numerous perivascular pseudorosettes composed of medium-size elongated cells evidenced the ependymal origin, forming areas of classic low-grade ependymoma (WHO grade II). Pseudorosettes of the classic type of ependymomal cells predominated over clusters of pseudorosettes composed of monster and multinuclear pleomorphic cells, which were located particularly on the tumour periphery. In addition, in the region between the intramedullary tumour and syrinxes, the picture mimicked pilocytic astrocytoma.
It has been shown that the localization and histological classification of ependymoma are a significant predictor of clinical outcome [12,18,29]. In adults, ependymomas develop most frequently in the spinal cord. In our study, widespread tumour was both intramedullary and intracranial, with secondary multi-segmental syringomyelia. There are several pathomechanisms of syringomyelia. Comparable examination of the most upper and lower part of tumour in our case suggest that intramedullary ependymoma spreading up along the cervical spinal cord to the fourth ventricle pressed on Luschka’s foramens of the medulla oblongata. Secondary hydrocephalus was due to the impairment of free flow of the cerebrospinal fluid (CSF) . Progressive growth of the spinal cord tumour and changes of vascular density are accompanied by intramedullary pressure and degenerative changes with hyalinosis of vessel walls [5,25]. Secondary compression of long spinal tracts, neurons and vessel walls leads to blood flow changes, degenerative and ischaemic necrotic changes in the tumour and glial stem below the tumour, which was also revealed in our case.
Microcirculation impairment supports the hypothesis that highly extended syrinxes might result from disruption of the blood-brain barrier followed by the development of degenerative cysts and tumour-ischaemic-related syringomyelia secondary to ependymoma [19,20,25,28]. The intramedullary part of the ependymoma in our case was not capsulated, but sharply demarcated from adjacent parenchyma by fibrillar glial fascicles, similarly to other reported cases . A possibility of varied cell differentiation in the same tumour is now well known [7,22,23]. According to the WHO grading system, the pathological spectrum of ependymomas might be very wide, including in rare cases a giant cell component . In the first description of giant cell variant of ependymoma, entire participation of bizarre multinuclear cells was reported in two cases located in the region of the filum terminale. The first one was uniformly composed of pleomorphic giant cells, whereas in the second case foci of giant cells were revealed in
a myxopapillary ependymoma . In the majority of subsequently reported GCEs, focal clusters of giant cells participated with classic ependymoma cells arrangement .
Additionally, some authors have described a few cases of unusual variant ependymoma (not GCE?) in which giant cells participated with ependymoma cells of varied type [11,21,23]. Giant pleomorphic tumour cells are a characteristic feature of other glial tumours, including giant cell astrocytoma, pleomorphic xanthoastrocytoma and glioblastoma [30,31]. Specialized ependymal differentiation was also reported in different types of gliomas [14,29,31]. Our ultrastructural examination revealed numerous junctional complexes between adjoining cells and blepharoplasts with changed morphology, confirming ependymal character of the tumour despite the limits of formalin-fixed and paraffin-embedded material [14,16,17].
To date, GCE cases which exhibited focal or entire participation of giant cells coexisting with conventional ependymoma were reported along with lower or higher grade malignancy and anaplasia findings. In our case of GCE, wide monomorphic areas of conventional ependymoma predominated over clusters of pleomorphic giant cells, which were characterized by an unusually low Ki67 index. GFAP immunopositivity of intranuclear inclusions was shown in those cells, which suggests they could arise in connection with cytoplasmic intranuclear invagination. It was evident also ultrastructurally [16,17]. The cytoplasmic intranuclear “pseudoinclusions” reported mainly in malignant gliomas with giant cells were revealed sporadically in material of intramedullary or ventricular ependymoma (2 of 21 biopsy cases), and recently in one case of GCE and one case of intramedullary clear cell variant of ependymoma [1,10,13,26].
Spinal cord ependymomas of classic type are usually classified among slow-growing WHO-II grade tumours characterized by long-term survival [29,30]. The issue of grading of giant-cell variant ependymoma and the origin of specialized ependymal differentiation of multinucleated giant cells is still rather controversial. It is emphasized that these tumours are unusual variants of ependymoma whose architectural pattern is sufficiently distinctive to be recognized in H&E stains [7,11].
In the presented case, a low index, lack of vascular glomerular endothelial proliferation and pseudopalisading necroses of the tumour were revealed. Despite nuclear hyperchromasia, the pleomorphic giant cells and intranuclear pseudoinclusions in their bizarre nuclei, there was a lack of other features of anaplasia. Therefore, the WHO criteria for anaplastic (grade III) ependymoma were not fulfilled [5,6,29,30].
To summarize, our histological, immunohistochemical and ultrastructural findings were consistent with the diagnosis of a low-grade (WHO grade II) giant cell ependymoma exhibiting ischaemic and degenerative changes coexistent with syringomyelia. To date, similar lower grade malignancy of this unusual variant of ependymoma has been reported in only a few cases [6,32].
The authors thank Prof. Wielis³aw Papierz, Chair and Department of Pathomorphology, Medical University of £ódź, for his neuropathological consultation.
1. Adamek D, Dec M, Sobol G, Urbanowicz B, Jaworski M. Giant cell ependymoma: a case report. Clin Neurol Neurosurg 2008; 110: 176-181.
2. Biernat W, Zawrocki A. Molecular alterations in ependymomas. Folia Neuropathol 2007; 45: 155-163.
3. Brown DF, Chason DP, Schwartz LF, Coimbra CP, Rushing EJ. Supratentorial giant cell ependymoma: a case report. Mod Pathol 1998; 11: 398-403.
4. Cooper PB, Katus M, Geyer D, Smirniotopoulos JG, Sandberg GD, Rushing EJ. Rare giant cell ependymoma in an octogenarian. Case report and review of the literature. J Neurosurg 2006; 105: 908-911.
5. Duda-Szymañska J, Papierz W. Morphological analysis of vascular density in ependymomas. Folia Neuropathol 2007; 45: 115-119.
6. Fourney DR, Siadali A, Bruner JM, Gokaslan ZL, Rhines LD. Giant cell ependymoma of the spinal cord. Case report and review of the literature. J Neurosurg 2004; 100: 75-79.
7. Hirato J, Nakazato Y, Iijima M, Yokoo H, Sasaki A, Yokota M, Ono N,
Hirato M, Inoue H. An unusual variant of ependymoma with extensive tumor cell vacuolisation. Acta Neuropathol 1997; 93: 310-316.
8. Jean WC, Abdel Azis KM, Keller JT, van Loveren HR. Subtonsillar approach to the foramen of Luschka: an anatomic and clinical study. Neurosurgery 2003; 52: 860-866.
9. Jeon YK, Jung H-W, Park S-H. Infratentorial giant cell ependymoma: a rare variant of ependymoma. Pathol Res Pract 2004; 200: 717-725.
10. Kim NR, Chung DH, Lee SK, Ha SY. Intramedullary clear cell ependymoma in the thoracic spinal cord: a case with crush smear and ultrastructural findings. J Korean Med Sci 2007; 22: S149-S153.
11. Kleinman GM, Zagzag D, Miller DC. Epithelioid ependymoma:
a nev variant of ependymoma: report of three cases. Neurosurg 2003; 53: 743-747.
12. Korshunov A, Golanov A, Timirgaz V. Immunohistochemical markers for intracranial ependymoma recurrence. An analysis of 88 cases. J Neurol Sci 2000; 177: 72-82.
13. Kumar PV. Nuclear grooves in ependymoma. Cytologic study of 21 cases. Acta Cytol 1997; 41: 1726-1731.
14. Lehman NL. Central nervous system tumors with ependymal features: a broadened spectrum of primarily ependymal differentation? J Neuropathol Exp Neurol 2008; 67: 177-188.
15. Li J, Lopez J, Powell SZ, Fuller GN, Coons S. Giant cell ependymoma: two new cases and review of the literature. J Neuropathol Exp Neurol 2007; 66: 445.
16. Liberski PP. Electron microscopy in diagnosis of tumors of the nervous system. Folia Neuropathol 1999; 37: 123-127.
17. Liberski PP. The ultrastructure of ependymoma: personal experience and the review of the literature. Folia Neuropathol 1996; 4: 212-220.
18. Matyja E, Nagañska E, Z¹bek M, Koziara H. Myxopapillary ependymoma of the lateral ventricle with local recurrences: histopatological and ultrastructural analysis of a case. Folia Neuropathol 2003; 41: 51-57.
19. Milhorat TH, Capocelli AL Jr, Kotzen RM, Bolognese P, Heger IM, Cottrel JE. Intramedullary pressure in syringomyelia: clinical and pathophysiological correlates of syrinx distension. Neurosurgery 1997; 41: 1102-1110.
20. Milhorat TH, Capocelli AL Jr, Anzil AP, Kotzen RM, Milhorat RH. Pathological basis of spinal cord cavitation in syringomyelia: analysis of 105 autopsy cases. J Neurosurg 1995; 82: 802-812.
21. Moritani S, Kushima R, Bamba M, Kobayashi TK, Oka H, Fujimoto M, Hattori T, Okabe H. Highly anaplastic extraventricular ependymoma arising in an adult, mimicking metastatic adenocarcinoma with heavy stromal inflammation and emperiporesis. Pathol Int 2003; 53: 539-546.
22. Nagañska E, Matyja E, Z¹bekM, Jagielski J. Disseminated spinal and cerebral ependymoma with unusual histological pattern: clinicopathological study of a case with retrograde tumor spread. Folia Neuropathol 2000; 38: 135-141.
23. Pal P, Fernandes H, Elliso DW. Correspondence: reader submitted case report; edited by Hamilton RL. Brain Pathol 2005; 15: 367-368.
24. Pimentel J, Kepes JJ, Moura Nunes JF, Bentes C, Miguens J, Antunes JL. Supratentorial giant cell ependymoma. Clin Neuropathol 2001; 20: 31-37.
25. Preusser M, Wolfsberger S, Haberler C, Breitschopf H, Czech T, Slavc I, Harris AL, Acker T, Budka H, Hainfellner JA. Vascularisation and expression of hypoxia-related tissue factors in intracranial ependymoma and their impact on patient survival. Acta Neuropathol 2005; 109: 211-216.
26. Robertson DM, MacLean JD. Nuclear inclusions in malignant gliomas. Arch Neurol 1965; 13: 287-296.
27. Sangoi A, Lim M, Dulai M, Vogel H, Changs S. Suprasellar giant cell ependymoma: a rare neoplasm in unique location. Hum Pathol 2008. Abstract. Article in press.
28. Sarikaya S, Acikgöz B, Tekkök IH, Güngen YY. Conus ependymoma with holocord syringohydromyelia and syringobulbia. J Clin Neurosci 2007; 14: 901-904.
29. Smith C, Ironside JW. Diagnosis and pathogenesis of gliomas. Current Diagnostic Pathol 2007; 13: 180-192
30. WHO classification of tumours of the central nervous system. Louis DN, Ohgaki H, Wiestler OD, Cavenee WK (eds.). IARC, Lyon 2007.
31. You H, Kim YI, Im SY, Suh-Kim H, Paek SH, Park SH, Kim DG, Jung HW. Immunohistochemical study of central neurocytoma, subependymoma, and subependymal giant cell astrocytoma.
J Neurooncol 2005; 74: 1-8.
32. Zec N, De Girolami U, Schofield DE, Scott RM, Anthony DC. Giant cell ependymoma of the filum terminale. A report of two cases. Am J Surg Pathol 1996; 20: 1091-1101.
Copyright: © 2008 Mossakowski Medical Research Centre Polish Academy of Sciences and the Polish Association of Neuropathologists. This is an Open Access article distributed under the terms of the Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International (CC BY-NC-SA 4.0) License (http://creativecommons.org/licenses/by-nc-sa/4.0/), allowing third parties to copy and redistribute the material in any medium or format and to remix, transform, and build upon the material, provided the original work is properly cited and states its license.
pod redakcją Barbary Steinborn
Liczba stron 224
prof. dr hab. med. Marek Jarema
Liczba stron 168
Tłumacz: Aneta Szaraniec
Redakcja merytoryczna: dr psychiatrii FILIP RYBAKOWSKI
Liczba stron: 352
Oprawa: miękka ze skrzydełkami
prof. dr hab. n. med. Joanna Hauser, dr n. med. Monika Dmitrzak-Węglarz
Liczba stron 168