Indian Journal of Pathology and Microbiology

: 2021  |  Volume : 64  |  Issue : 4  |  Page : 633--637

Tanycytic ependymoma: highlighting challenges in radio-pathological diagnosis

Pragya Jain1, Ravindra K Saran1, Daljit Singh2, Anita Jagetia2, AK Srivastava2, Hukum Singh2,  
1 Department of Pathology, GIPMER, Jawaharlal Nehru Marg, New Delhi, India
2 Department of Neurosurgery, GIPMER, Jawaharlal Nehru Marg, New Delhi, India

Correspondence Address:
Ravindra K Saran
Department of Pathology, GIPMER, Jawaharlal Nehru Marg, 64 Khamba, New Delhi


Background: Tanycytic ependymoma (TE) (WHO grade II) is a rare and morphologically distinct variant of ependymoma with only 77 cases reported worldwide so far. Variable clinical and radio-pathological features lead to misdiagnosis as WHO grade 1 tumors. On imaging, differentials of either schwannoma, meningioma, low-grade glial (like angiocentric glioma), or myxopapillary ependymoma are considered. In this study, we aim to discuss clinical, radiological, and pathological features of TE from our archives. Method: We report clinicopathological aspects of six cases of TE from archives of tertiary care center between 2016 and 2018. Detailed histological assessment in terms of adequate tissue sampling and immunohistochemistry was done for each case. Result: The patient's age ranged between 10 and 53 years with a slight male predilection. Intraspinal location was seen in two cases (intramedullary and extramedullary), three cases were cervicomedullary (intramedullary), and one was intracranial. One case was associated with neurofibromatosis type 2. Four cases mimicked as either schwannoma or low-grade glial tumor on squash smears. On imaging, ependymoma as differential was kept in only two cases and misclassified remaining either as low-grade glial or schwannoma. Discussion: In initial published reports, the spine is the most common site (50.4%) followed by intracranial (36.4%) and cervicomedullary (3.9%). They have also highlighted the challenges in diagnosing them intraoperatively and radiologically. Treatment is similar to conventional ependymoma if diagnosed accurately. A multidisciplinary approach with the integration of neurosurgeon, neuroradiologist, and neuropathologist is required for accurate diagnosis and better treatment of patients.

How to cite this article:
Jain P, Saran RK, Singh D, Jagetia A, Srivastava A K, Singh H. Tanycytic ependymoma: highlighting challenges in radio-pathological diagnosis.Indian J Pathol Microbiol 2021;64:633-637

How to cite this URL:
Jain P, Saran RK, Singh D, Jagetia A, Srivastava A K, Singh H. Tanycytic ependymoma: highlighting challenges in radio-pathological diagnosis. Indian J Pathol Microbiol [serial online] 2021 [cited 2021 Nov 27 ];64:633-637
Available from:

Full Text


After the introduction of the WHO 2016 classification of CNS tumors, a paradigm shift is observed from morphology to more comprehensive molecular diagnosis. But morphology and immunohistochemistry (IHC) remain the cornerstone in diagnosing and stratifying ependymomas. Recently, the Consortium to Inform Molecular and Practical Approaches to CNS Tumor Taxonomy (c-IMPACT NOW)-7 in their update has proposed the division of ependymoma based on the anatomic sites and gene profiling. They also suggested that morphological subtyping should be a distinct pattern in the histopathological description.[1]

Classical ependymoma is a WHO grade II tumor, morphologically segregated into a clear cell, papillary, and tanycytic ependymoma. Tanycytic ependymoma (TE) is a morphologically distinct, rare ependymoma bearing an indolent course. They differ from conventional ependymoma due to their cellular characteristics. It was first described by Freide and Pollock in 1978 and was included under ependymoma subtypes in WHO 2000.[2],[3] A limited number of cases of TE are reported so far worldwide. Kasper et al.[4] in their largest metanalysis pooled 51 case studies and a total of 77 cases of TE reported worldwide so far, thereby emphasizing its rarity. Limited literature is available on treatment guidelines and outcomes. Due to its occurrence at a variable location and masquerading nature (radiologically and histologically), diagnosing them remains a challenge. Neumann et al.[5] studied molecular characteristics among histopathological variants of ependymoma using DNA methylation and gene analysis. They observed that TE is just a morphologically distinct variant and methylation pattern match with other groups of ependymoma (i.e., myxopapillary, spinal, and posterior fossa group).

This study aims to discuss clinical, radiological, and pathological features of TE identified from the records of tertiary care hospitals in North India


This study was conducted in the Department of Pathology in collaboration with the Department of Neurosurgery and Radiology at tertiary care hospital. Cases diagnosed as TE between January 2016 and December 2018 were included. They were reviewed in light of clinical/radiological suspicion, squash smear diagnosis, and final diagnosis after IHC. The relevant details of the cases (age, gender, site, clinical and radiological suspicion, any other associated syndrome, and IHC) were noted on the prestructured proforma. Immunohistochemical analysis using GFAP, p53, ATRX, S-100, EMA, H3K27me3, vimentin, synaptophysin, and MIB was done on all cases. Follow-up of all the patients was taken by telephonic conversation. Surgical outcome was categorized as gross total resection (GTR) where the whole tumor was removed macroscopically and subtotal resection (STR) where limited resection was possible, and residual tumor was left. The surgical outcome was assessed on a 6-week postoperative MRI. Informed consent was obtained from all participants.


We reported 58 cases of ependymoma from our department between January 2016 and December 2018 of which six were diagnosed as TE. All the cases were reviewed by two pathologists. One case which was earlier diagnosed as TE was reclassified as angiocentric glioma and was removed from the study. The detailed summary and clinicopathological details of the cases are shown in [Table 1]. Age at presentation ranged from 10 to 53 years with slight male predominance (M: F = 2:1). Amongst them, three cases were located at cervicomedullary junction (all intramedullary). Two cases were spinal (intramedullary and extramedullary), and one was intracranial (frontotemporal) in location. Symptoms were according to the site of involvement. Tumor involving cervicomedullary junction presented with hemiparesis and sensory disturbances. Headache and weakness were the common findings in those involving intracranial location. Cases with spinal involvement presented with lower motor neuron type palsy. Association with neurofibromatosis -2 (NF-2) was seen in one case.{Table 1}

Radiological findings

Preoperative MRI was available for all cases. Imaging revealed a well-defined, isointense (5 cases) to a few hypointense (1 case) areas on T1-weighted image. T2-weighted images were hyperintense in five cases and isointense in one case. In a case with intracranial involvement, the lesion was solid cystic with contrast enhancement. Radiological suspicion of pilocytic astrocytoma was suggested in the case. Possibility of schwannoma was given in three cases (Two cervicomedullary and one spinal). A differential of astrocytoma along with ependymoma was given in two cases (spinal and one cervicomedullary junction, respectively). Preoperative radiology images of two patients (case 3 and case 6) are shown in [Figure 1]a, [Figure 1]b and [Figure 2]a, [Figure 2]b.{Figure 1}{Figure 2}

Pathological findings

Squash cytology was performed for all cases. The diagnosis of schwannoma was given in four cases (two intraspinal and two cervico-medullary). A low-grade glial tumor was given in one case due to intracranial involvement. Mixed morphology with the presence of perivascular rosettes with elongated cells was seen in two cases, and differential of ependymoma was given along with low-grade glial and schwannoma in those cases. [Figure 3]a and [Figure 3]b shows squash smear photomicrographs of case 6.{Figure 3}

Hematoxylin and eosin-stained slides of the paraffin-embedded tissue revealed elongated and spindled cells in the fibrillar background. Finely speckled chromatin and inconspicuous nucleoli were seen in all cases [Figure 4]a and [Figure 4]b. We observed two forms of TE as described by Kawano et al.[11] namely, pure (only spindle morphology and absence of perivascular rosettes0 and mixed (presence of perivascular rosettes). Only two cases had mixed morphology with the presence of typical perivascular rosettes which clinched the diagnosis in those cases [Figure 4]c. None of the cases showed anaplastic features. Immunohistochemically, all cases were GFAP, S100, and vimentin-positive. p53 and synaptophysin were negative. There was no loss of ATRX. H3K27me3 was done of three cases at cervicomedullary junction and was found to be positive denoting lack of mutation. The diagnosis of TE was made after demonstrating perinuclear or canalicular EMA positivity in all cases. MIB index was low, ranged from 2% to 5% suggestive low proliferative activity [Figure 5]a, [Figure 5]b, [Figure 5]c, [Figure 5]d.{Figure 4}{Figure 5}

Postoperative outcome

Minimum 1 year follow-up was taken for all cases. STR was done in four cases (three with cervico-medullary involvement and one with D9-L1 involvement). GTR was possible in the remaining two cases. Adjuvant radiotherapy was given in four cases (three cases in which STR was achievable and one with intracranial involvement). In cases, where STR was done, the patients developed neurological deficits. None of the cases had a recurrence.


TE (WHO grade II) mimics other WHO grade 1 tumors due to its deceptive radiology and morphology. The pathologist should always have a high index of suspicion in these cases for a better prognosis of patients. They arise from tanycytes of ependyma which are bipolar radial glial cells connecting ependyma and capillary wall forming a connection between cerebrospinal fluid, parenchyma, and vasculature. TE was first described by Friede and Pollock[2] in 1978 in two cases, one at the intracranial location and the other at the foramen of Monro. Both the lesions were initially suspected as a glial tumor Kasper et al.[4] conducted the largest metanalysis on TE from Harvard school of medicine in 2018. They included 51 articles to include 77 cases of TE defined so far. Many of them masqueraded as low-grade glial or schwannoma. To the best of our knowledge, very few cases are reported from India.[4],[6]

The spinal cord (50.6%) was the most common location, followed by intracranial (36.4%) and cervicomedullary (10.4%). Intramedullary (64%) is more common than extramedullary amongst tumor located in the spinal canal. With regard to the intracranial lesion, supratentorial location is more favored in comparison to infratentorial (71% vs 25%, respectively). We are reporting a mixed bag of TE located at cervicomedullary (3/6), spinal (2/6), and intracranial (1/6) locations. The mean age of presentation as reported in the literature is a wide range, ranging from 2 to 75 years.[4] Similar is our observation (10–53 years with slight male predominance). Kambe et al.[7] reported TE at the supratentorial cortical location in a 2-year-old male, presented with absence seizures. Other authors too have reported intracranial TE in pediatric patients[4] In our series, the patient with intracranial TE was in pediatric age group (10 years). The cases had a solid cystic presentation on imaging, and the possibility of pilocytic astrocytoma was kept on radiology. Isolated cases of spinal TE in association with neurofibromatosis type II have been reported in the past; however, the genetic status is unclear.[4],[8],[9] Tao XG et al.[10] have reported two cases of spinal cases of TE in 16/F and 43/F patients with associated NF-2. To the best of our knowledge, this is the first case of TE occurring at the cervicomedullary junction in a young patient of 18-year-old female with NF-2.

Histologically, TE presents as a fibrillary tumor having long bipolar processes. Perivascular rosettes of ependymoma are mostly absent but can be seen in some. Necrosis and mitosis are generally not seen. Kawano et al.[11] distinguished TE histologically into two forms “pure” and “mixed” forms. “Pure” form in which conventional features of ependymoma such as perivascular rosettes are not seen and “mixed” form in which showed other histological features of conventional ependymoma along with typical morphology of TE. “Mixed” histology was seen in 36.4% (28/77) in their metanalysis. Mixed histology was seen in two cases in our series in which differential of ependymoma (case 1 and case 4) was given, whereas the rest all had pure morphology in which other diagnoses such as pilocytic astrocytoma or schwannoma was given. In cases where “mixed” morphology is seen it become easier for neuropathologist to provide a differential of TE on squash smears, whereas “pure” TE frequently mimics other WHO grade I tumors such as schwannoma, low grade glial, or meningioma. Kasper et al.[4] in their metanalysis have summarized studies in which TE due to its pure morphology resembled pilocytic astrocytoma or schwannoma. In our series, the intracranial solid cystic lesion had a fibrillary background and the presence of an occasional eosinophilic granular body; thus, the possibility of pilocytic astrocytoma was suggested. Anaplasia generally is not the feature of TE; however, some authors have reported anaplastic features in spinal ependymoma.[4] We did not find any anaplasia in our series.

Immunohistochemically, TE is GFAP, S-100, and vimentin-positive. Dot-like/canalicular EMA staining pattern clinches the diagnosis of TE. MIB labeling index like all other grades II ependymoma is low. In our series, all the cases were GFAP and S100 positive and had a low labeling index. The diagnosis of TE was made after demonstrating dot/canalicular like the positivity of EMA. The ultrastructural analysis also aids in diagnosing TE and differentiating it from other entities. On electron microscopy, TE is slightly different from classical ependymoma as composed of slender bipolar cells with only one pole bearing microvillus, whereas the opposite pole abuts the basal surface.[12] In our series, we could not perform electron microscopy.

The radiology of TE is also not extensively studied in the literature with only a few studies.[13] Solely on radiology, diagnosing TE remains a challenge, especially due to potential grade 1 mimickers due to site and presentation. The predominant radiological features according to Tomek et al.[13] are contrast enhancement (94%), heterogenous enhancing pattern (50%), T1 isointensity (60%), and cystic component (53%). Frequent other patterns are also being reported such as dumbbell-shaped or cystic mass with the mural nodule. Presentation as solid cystic mass with mural nodule brings pilocytic astrocytoma or hemangioblastoma as the differential. When it presents as a dumbbell-shaped lesion or with a dural abutment, it is diagnosed as neurofibroma or meningioma. In our series, the possibility of pilocytic astrocytoma was given on radiology in one intracranial solid cystic lesion (Case 6). Schwannoma as differential was given in three spinal lesions. Observations made by our series highlighted the pitfalls in radiological diagnosis.

The possible differential of TE is other tumors with ependymal features such as angiocentric glioma and pilomyxoid astrocytoma. Angiocentric glioma is cerebro-cortical tumors with few features of ependymoma such as bipolar spindled cells, clustering around blood vessels simulating perivascular pseudo rosettes, and dot-like EMA positivity. They can be differentiated from TE by longitudinal angiocentric alignment of tumor cells. It also showed a parallel arrangement of cells along the vessels. Focal clusters of spindle cells giving the appearance of schwannoma. The dot-like EMA positivity was the cause of misinterpretation in one of our initial cases. Pilomyxoid astrocytoma is another glial tumor seen in children and involving the hypothalamus, cerebellum, and spinal cord. Due to its putative origin from radial glial cells and overlapping ultrastructural features with TE, they are also known as tanycytoma.[14] They can be differentiated from TE by myxoid background and dot and canalicular EMA positivity.

The prognosis of TE remains favorable with a good outcome; however, treatment remains debatable. Currently, there are no treatment guidelines available for TE, and is mostly based on experience with another ependymoma. GTR is the mainstay of treatment. Adjuvant radiotherapy is the optional therapy, mainly where STR only was the surgical outcome. Tao et al.[15] in their series of intracranial TE have reported a good overall prognosis and suggested that adjuvant radiotherapy be administered to all cases (including GTR and STR) to achieve a good outcome. In our series, GTR was possible in two cases, and STR was achievable in four cases. Radiotherapy was given in three cases (one with GTR and two with STR). One of the cases despite having GTR received adjuvant radiotherapy, whereas the other did not.


This series widens the clinicopathologic spectrum and highlights the common pitfall in the pathological the diagnosis of TE. Due to its deceptive presentation, it is imperative to design an integrated and multidisciplinary approach for diagnosis and management. There are no standardized treatment guidelines for this entity and are mostly governed by the judgment of the clinician. More awareness about pitfalls and a high index of suspicion will broaden the cohort of these patients for detailed follow-up and designing better treatment guidelines.

Declaration of patient consent

The authors certify that they have obtained all appropriate patient consent forms. In the form, the patientshave given their consent for their images and other clinical information to be reported in the journal. The patients understand that their names and initials will not be published and due efforts will be made to conceal their identity, but anonymity cannot be guaranteed.

Financial support and sponsorship


Conflicts of interest

There are no conflicts of interest.


1Ellison DW, Aldape KD, Capper D, Fouladi M, Gilbert MR, Gilbertson RJ, et al. cIMPACT-NOW update 7: Advancing the molecular classification of ependymal tumors. Brain Pathol 2020;30:863-6.
2Friede RL, Pollak A. The cytogenetic basis for classifying ependymomas. J Neuropathol Exp Neurol 1978;37:103-18.
3Kleihues P, Louis DN, Scheithauer BW, Rorke LB, Reifenberger G, Burger PC, et al. The WHO classification of tumors of the nervous system. J Neuropathol Exp Neurol 2002;61:215-29.
4Kasper EM, Ippen FM, Maragkos GA, Anderson MP, Rojas R, Mahadevan A. Tanycytic ependymoma of the brain stem, presentations of rare cystic disease variants and review of literature. J Neurosurg Sci 2018;62:78-88.
5Neumann JE, Spohn M, Obrecht D, Mynarek M, Thomas C, Hasselblat M, et al. Molecular characterization of histopathological ependymoma variants. Acta Neuropathol 2020;139:305-18.
6Parkhi M, Gupta K, Singh A, Salunke P. Giant cell-rich tanycytic ependymoma as intramedullary spinal mass. Neurol India 2019;67:1327-30.
7Kambe A, Kurosaki M, Watanabe T, Nakazato Y. Pediatric supratentorial cortical tanycytic ependymoma associated with absence seizures. Clin Neuropathol 2014;33:308-10.
8Lim BS, Park SQ, Chang UK, Kim MS. Spinal cord tanycytic ependymoma associated with neurofibromatosis type 2. J Clin Neurosci 2010;17:922-4.
9Kobata H, Kuroiwa T, Isono N, Nagasawa S, Ohta T, Tsutsumi A. Tanycytic ependymoma in association with neurofibromatosis type 2. Clin Neuropathol 2001;20:93-100.
10Tao XG, Hou ZG, Hao SY, Zhang JT, Liu BY. Two cases of spinal tanycytic ependymoma associated with neurofibromatosis type 2. Chin Med J (Engl) 2017;130:872-3.
11Kawano N, Yagishita S, Hara M, Tadokoro M. Pathologic features of ependymoma: Histologic patterns and a review of the literature. Neuropathology 1998;18:1-12.
12Mrak RE. The big eye in the 21st century: The role of electron microscopy in modern diagnostic neuropathology. J Neuropathol Exp Neurol 2002;61:1027-39.
13Tomek M, Jayajothi A, Brandner S, Jaunmuktane Z, Lee CH, Davagnanam I. Imaging features of spinal tanycytic ependymoma. Neuroradiol J 2016;29:61-5.
14Lehman NL. Central nervous system tumors with ependymal features: A broadened spectrum of primarily ependymal differentiation? J Neuropathol Exp Neurol 2008;67:177-88.
15Tao X, Dong J, Hou Z, Hao S, Zhang Q, Wu Z, et al. The clinical features and surgical outcomes of intracranial tanycytic ependymomas: A single-institutional experience. J Neurooncol 2017;134:339-47.