DIAGNOSIS: RECURRENT MENINGIOMA WITH RHABDOMYOSARCOMATOUS DIFFERENTIATION
The presence of rhabdomyoblasts in central nervous system tumours is a well recognised but rare occurrence (e.g. (2), reviewed in (8)). When it does occur, the rhabdomyoblasts may be either the principle neoplastic cell (i.e. primary intracranial rhabdomyosarcoma) or they can be mixed with neuroglial cells (e.g. medullomyoblastoma, triton tumours and gliomyosarcoma). Our patient's case differs from both these groups of tumours as the rhabdomyoblasts are mixed with meningothelial cells.
There is one case report of a tumour that has some features in common with ours. Ferracini et al. (4) described a meningeal tumour composed almost entirely of rhabdomyoblasts. On retrospective analysis of all slides from this tumour, they found some areas with a meningothelial appearance on haematoxylin and eosin staining. These cells showed weak immunostaining for epithelial membrane antigen (EMA) consistent with meningothelial cells. They suggested that their tumour arose from a common precursor cell for meningothelial and muscle cells. However, in an accompanying commentary, Burger suggested that the areas of meningothelial whorls were in fact vascular (3). In contrast our patient's tumour consists of unambiguous meningothelial cells and rhabdomyoblasts based on morphological, immunohistochemical and ultrastructural grounds. Our case also differs as we found the rhabdomyoblasts developed in the recurrent tumour only.
Interestingly, a number of the previous cases of primary intracranial rhabdomyosarcoma have arisen from the meninges (e.g. 7, 10) and Smith et al.(10) have suggested that all primary intracranial rhabdomyosarcomas may arise from the meninges. Indeed, cranial skeletal muscle and meningothelial cells both arise from the neural crest. Furthermore, skeletal muscle cells have been identified in the meninges of patients without tumours (1, 5, 6, 9). However, the histogenesis of skeletal muscle cells in intracranial tumours remains controversial. In principle, the muscle cells may arise from the meningioma as a result of increasing anaplasia of the tumour. Alternatively, the tumour may arise from multipotent precursor cell capable of differentiating towards either cell type. Finally, the muscle cells may not be derived from the neoplastic cells at all. Instead they may have been driven to divide by factors released from the meningioma. In our patient the muscle cells appeared only in the recurrent tumour which would be most consistent with the former or latter explanations.
Contributed by Thomas S Jacques, Alan Valentine, Robert Bradford, James E McLaughlin