Solitary Fibrous Tumor of the Skin

To the Editor: We read with great interest the article by Cowper et al. (1) regarding solitary fibrous tumors of the skin. In their article, the authors presented a series of three cases of solitary fibrous tumor that arose in cutaneous sites. Two of the tumors were located on the posterior neck, and the third on the occipital region. The tumors involved the dermis as well as the subcutaneous adipose tissue, and all were composed of cytologically bland spindle-shaped cells arranged in a variety of patterns with thickened, “ropey” collagen bundles intermingled among the spindle-shaped cells. This histologic description, with some of the photomicrographs, particularly Figure 3, is quite reminiscent of spindle cell lipoma, a benign lipomatous neoplasm that characteristically involves the posterior neck, upper shoulders, or back of middle-aged men (2). Although spindle cell lipoma is typically centered in the subcutaneous tissue, there may be dermal involvement. The constituent cells are cytologically bland and quite similar to those seen in solitary fibrous tumor (from any site) and the ropey collagen bundles described by the authors are probably more characteristic of spindle cell lipoma than they are of solitary fibrous tumor. In fact, the collagen in solitary fibrous tumor is often characteristically “wire-like” and has a different quality than that seen in spindle cell lipoma. Immunohistochemistry is of little use in these two tumors, because both are characteristically CD34 (3) and bcl-2-positive (4). Based upon the available data, we feel the cases described by Cowper et al. have features of both solitary fibrous tumor and spindle cell lipoma. Certainly, Figure 3 is indistinguishable from a spindle cell lipoma. It is interesting that a prior study has even raised the possibility that spindle cell lipoma may represent an incipient solitary fibrous tumor (5). Like many other lipomatous neoplasms, spindle cell lipoma has a characteristic cytogenetic aberration, that is, deletions of both 13q and 16q (6). There are very few cytogenetic data regarding solitary fibrous tumor, and it would be interesting to determine whether the latter has similar cytogenetic aberrations as the former. Until such data are available, the histogenetic relationship between these two neoplasms remains in question. Jessica E. Sigel, M.D. John R. Goldblum, M.D.