Solitary fibrous tumor.

Abstract

To the Editor: We read with interest the article “Solitary Fibrous Tumor of the Vagina” by Vadmal and Pellegrini (1). We recently saw a case that histologically was similar, and which also arose in the vagina (Fig. 1). The tumor was composed of a circumscribed nodule containing plump spindle cells, which as in the present case showed mainly a haphazard distribution with a prominent vasculature. However, in addition, in some areas there was more abundant hyalinized matrix material between cells (Fig. 2). A solitary fibrous tumor was high in our differential diagnosis. Thus we did immunohistochemical stains for CD-34 (Fig. 2), bcl-2, Vimentin, and CD-99, which were all positive. We also did immunohistochemical stains for α-smooth muscle actin, S-100 protein, EMA, A1/AE, CK7, and CK20. α-Smooth muscle actin and S-100 protein were negative; however A1/AE was diffusely positive (Fig. 2), CK 7 showed scattered positivity, and EMA showed focal positivity, although CK 20 was negative. This positive staining for cytokeratin occurred even though there were no areas that showed epithelial features histologically. From all this information we made the diagnosis of Spindle Cell Epithelioma of the Vagina (SCEV) or “Mixed Tumor of the Vagina”. These tumors are felt to be benign tumors that may recur if incompletely excised, but that to date have shown no metastatic spread (2–4). Immunohistochemical stains for estrogen and progesterone receptors were later performed, and they also showed scattered positivity (Fig. 3).FIG. 1.: Low power view showing a circumscribed nodule composed of plump spindle cells with a prominent vasculature (Hematoxylin and eosin x15).FIG. 2.: Immunohistochemical staining for CD34 showing membrane staining of the plump spindle cells. In this area there is increased hyalinized stromal matrix material between the tumor cells (x400).FIG. 3.: Immunohistochemical staining for cytokeratin AE1/AE3 showing diffuse staining of tumor cells (x100).Although the majority of SCEVs show scattered to focal areas with features suggestive of epithelial differentiation, they may show no such areas and appear to be mesenchymal tumors (2). In addition, unlike most other mixed tumors that have been described, they do not appear to arise from myoepithelial cells, and thus they do not express α-smooth muscle actin or S-100 protein (2). Although it has been previously proposed that these tumors may arise from progenitor cells, immunohistochemical stains for CD34, bcl-2, and CD99 have previously not been performed on these rare tumors (2–4). Vadmal and Pellegrini may have performed immunohistochemical stains for cytokeratins and/or EMA; however, these immunohistochemical stains were not reported in the manuscript, and the immunohistochemical stains that were reported would not distinguish solitary fibrous tumor from our case of SCEV (1). Thus, although we do not question the diagnosis of Vadmal and Pellegrini, we do believe that SCEV should be included in the differential diagnosis, and the presence or absence of positive staining for cytokeratin can be used to distinguish these entities in cases where they can not be definitively distinguished morphologically. Kathleen J. Smith COL, MC, USA Henry G. Skelton M.D.