MPE is a glial tumour occurring almost exclusively in the region of the cauda equina and considered to be one of the most frequent primary tumours to occur in this location [3
]. MPE was first described as a separate entity by Kernohan in 1931 [11
] with isolated case reports subsequently appearing in the literature [[12
]]. The clinical presentation depends on the location of the tumour. The majority of cauda equina and filum terminale tumours present with low back pain due to nerve root compression similar to our cases. Lower limb weakness and sphincter dysfunction are the two other common clinical manifestations. Intramedullary MPE arise from the ependymal lined cells of the filum terminale. The proposed histogenesis of extramedullary MPE in this location is the presence of ependymal rest of the neural tube during canalization and retrogressive differentiation [14
] or from ectopic ependymal cell [16
]. Recent study has suggested that radial glia is the cell of origin of ependymoma [17
The classic morphology is easily recognizable, comprising multiple papillae covered by flattened to cuboidal cells embedded in a myxoid stroma and forming pseudorosettes. However, there are cases where the tumour obtains a solid growth pattern with aggregates of cells with "epithelial morphology" which in addition may show clearing of the cytoplasm. In such instances, metastatic carcinomas of renal origin and chordoma have to be ruled out by adequate clinical history and immunohistochemical stains.
It is well known that MPE stain positive with GFAP and S-100 protein. Cytokeratin positivity in MPE has been a subject of controversy with few cases reported showing positive cytokeratin expression in MPE [5
]. Both our cases were positive for CAM5.2, AE1AE3, EMA and CK7 but negative for CK20.
Morphological variations in MPE can resemble metastatic carcinoma. An erroneous diagnosis may subject the patients to unnecessary metastatic work-up and additional adjuvant therapy. In addition, one should also consider the psychological implication of diagnosing a carcinoma. MPE has a better prognosis with tendency for late recurrence, except for some cases with aggressive behavior and seeding to the CNS [21
]. MPE with clear cell changes and positive staining for S-100 protein, keratins and EMA can be misdiagnosed as chordoma, the latter being positive for Cytokeratin markers and S-100 protein but negative for GFAP. Although, metastatic carcinomas are positive for cytokeratins, these are consistently negative for GFAP.
MPE can be positive for Cytokeratins (CAM 5.2, AE1 AE3, and CK7) and focally for EMA. Given the similarities in morphology between metastatic carcinoma, chordoma and MPE, the diagnosis of MPE should be made in conjunction with clinical history including tumour location, immunohistochemical profile (co-expression of GFAP, S-100 protein and epithelial markers in MPE) and radiological findings. The recommended panel should include Cytokeratins (positive in MPE, metastatic carcinoma and chordoma), GFAP (positive in MPE and negative in metastatic carcinoma and chordoma), and S-100 protein (positive in MPE, chordoma but negative in most metastatic carcinomas).
The majority of MPE are slow growing gliomas with a tendency for local recurrence. Prognosis depends on complete surgical resection of the tumor. The risk of metastasis is very low and rare. The overall prognosis of MPE is much better than metastatic carcinoma and the latter has to be ruled out as they can have similar morphological features.
In conclusion, we reemphasize the importance of using a panel of immunohistochemical stains to differentiate between MPE and other tumour entities to avoid misdiagnosis.