DERMATOFIBROSARCOMA PROTUBERANS (DFSP)
This case is of unusual interest in that this 35 year old female had another DFSP removed from her upper back (near a shoulder) several years ago. The current lesion is from the lower back (near a buttock). I was unable to find another example of a patient having more than one of these tumors in my literature search.

Tumors of this type are prone to recur, mostly because the initial excision is often inadequate. This is related to the pattern of infiltration which may be asymmetrical, and which is characterized by infiltration of the subcutis by delicate strands of tumor that extend far beyond the main mass of the tumor. Metastatic behavior is rare.

There has been emphasis placed on the diagnostic importance of cellular storiform figures within the dermis. While this characteristic is present in most examples of DFSP, there are occasional cases wherein the dermal component is not strikingly cellular, and storiform (starburst) figures may be difficult to find. The characteristic pattern of infiltration of the subcutis, however, is present, and this greatly facilitates making the diagnosis. This asymmetrical, patchy pattern of infiltration has been variously described as 'Swiss cheese-like', 'honeycomb', and 'lacelike'.  Since cellular storiform figures can be found in the dermis in dermatofibromas and since dermatofibromas may extend into the subcutis, the importance of the recognition of the characteristic pattern of infiltration of the subcutis by a DFSP cannot be overly stressed. The spindle cells in DFSP are monotonous in appearance, and cytologic atypia is minimal or absent. Mitoses may be few in number or not demonstrated at all in a given biopsy. Epidermal hyperplasia may or may not be present. Foamy histiocytes and/or hemosiderin deposition are rarely seen in a DFSP, and their presence should raise the possibility that the lesion is more likely a dermatofibroma that is cellular.

The spindle cells are CD-34 positive. Factor XIIIa positive spindle cells can sometimes be found scattered within a DFSP, but the basic tumor cells are factor XIIIa negative. These findings contrast with the factor XIIIa positive spindle cells and CD-34 negative spindle cells in dermatofibromas.

Terminology note: It can be argued that DFSP's are not composed of ordinary fibrohistiocytic cells but of cells somehow related to cells of neural origin or neural association. Points in favor of this argument include the lack of foamy histiocytes, the demonstration of basal laminae by electron microscopy,  the CD-34 positivity of the tumor cells, and the finding of melanosomes in some of these (Bednar tumors). This concept of neural origin plus the lack of cytologic atypia exclude the rational inclusion of this tumor with malignant fibrous histiocytomas.
 
Scan power view. The pattern of infiltration of adipose tissue that is seen in this picture should immediately alert the observer to the likelihood of this being a DFSP. This pattern of spindle cells can be seen in fibrous hamartomas of infancy, and, rarely, in congenital nevi and in diffuse neurofibromas.
Low power view of the top of the specimen. Minimal epidermal hyperplasia is present. The degree of cellularity that is seen in this field, particularly in the superficial dermis, is less than that found in many dermatofibromas. These spindle cells were CD-34 positive (I had this study done for sake of completeness on this case).
High power view of above. There is no more atypia of these cells than that which is seen in the fibrohistiocytic cells of most dermatofibromas. The stroma is peculiar and not of ordinary reticular dermis type. There can be stromal variation in dermatofibromas, so this is not diagnostic.
Low power view of the base of the specimen. The subcutaneous component is quite cellular, and delicate, interlacing streamers of spindle cells infiltrate the adipose tissue peripheral to the main tumor mass. A few factor XIIIa positive cells were scattered among the almost universally positive CD-34 positive cells in this location.
High power view from the area of the junction of the middle and deep part of the specimen. This was the only field that I could find that might be interpreted as a cellular storiform figure.
High power view of some of the deep part of the specimen. This shows the degree of cellularity and lack of significant cytologic atypia of the spindle cells in the subcutis.

Click on your browser's 'Back' button to return to the previous page.