GROSS AND MICROSCOPIC FINDINGS:
The findings were very similar in both cases and are described together. Grossly, the lesions were indistinguishable from a meningioma, with an extraaxial firm solid mass showing broad dural attachment (Fig. 3, case 2). Microscopic examination revealed mixed inflammatory infiltrates in a dense collagenous background. The infiltrates consisted of plasma cells, lymphocytes and large histiocytic cells. Plasma cells were numerous, and often arranged along small blood vessels (Fig. 4, case 1). They showed no nuclear atypia or mitotic activity. Numerous Russell bodies were noted. Immunohistochemical stains for lambda and kappa chains revealed a mixed staining pattern, confirming polyclonal nature of the plasma cell infiltrate. Mature appearing small lymphocytes were also present, scattered throughout, and occasionally forming lymphoid aggregates without germinal centers. The third component of the infiltrate consisted of distinctive large histiocytic cells, which were interspersed between other cell types, but often formed large sheets of cells, which at low power appeared as pale areas. These cells had large vesicular empty nuclei, usually without nucleoli, and abundant, but poorly defined, pale cytoplasm (Fig. 5, case 2). Occasional binucleated cells were noted, as were some histiocytic cells with somewhat shrunken eosinophilic cytoplasm. There was no evidence of anaplasia or mitotic activity. On careful search, emperipolesis could be identified in H&E-stained sections, consisting of histiocytic cells with viable-appearing lymphocytes or plasma cells engulfed within the cytoplasm (Fig. 6, case 1).
By immunohistochemistry, the histiocytic cells were positive for S100 protein (Fig. 7, case 2), as well as monocyte/macrophage markers, such as CD68, and HAM56, and for vimentin (Fig. 8, case 1), and were negative for EMA, LCA, GFAP, synaptophysin and neurofilament protein. Unlike histiocytic cells of Langerhans histiocytosis, they were negative for CD1a. Emperipolesis was dramatically highlighted by immunostains for S100 protein, and for vimentin (Fig. 8, case 1), which outlines the intracytoplasmic lymphocytes within the histiocytic cells. These stains outlined the cytoplasm of the large histiocytic cells, thus better visualizing small cells within. Immunohistochemical study for LMP-1 antigen of Epstein-Barr virus was done in case 1, and was negative. No meningothelial component was identified. Immunostaining for EMA was negative, except for expected staining of plasma cells. Special stains for microorganisms were negative. The interface between the brain and inflammatory mass was mostly sharp and well defined, but focal intraparenchymal extension of the inflammatory infiltrate along the perivascular spaces was noted in case 1, and was more prominent in case 2.
Electron microscopic study was performed in case 2. The large histiocytic cells showed a paucity of organelles and extensive cell processes, without basal lamina. No Birbeck granules were identified.