Secondary Logo

Journal Logo

It's a Dermatofibroma, CD34 is Irrelevant!

Zelger, Bernhard M.D.

The American Journal of Dermatopathology: October 2002 - Volume 24 - Issue 5 - p 453-454
Letters To The Editor

Department of Dermatology

University of Innsbruck

Innsbruck, Austria

To the Editor:

The concept of exchanging ideas about diagnoses of difficult and challenging dermatohistopathologic cases by experts via the column of an international journal is really splendid and should animate to open discussion. A spindle cell CD34-positive dermal proliferation was the focus of interest in the most recent issue of the Section “Cases in Consultation,” edited by Luis Requena (1) and the opinions of the consulted experts included a generic designation of dermal myxoid fibroma with the descriptor of CD34 positivity by Jan Pitha; a benign neoplasm of the follicle-associated mesenchyme such as trichodiscoma or trichogenic myxoma by Bruce R. Smoller; a likely benign and an as-yet-undefined spindle cell tumor of periadnexal CD34-positive dendritic cells associated with a subtle induction of epithelial strands, cannot exclude early dermatofibrosarcoma protuberans, by Steve Somach; and, finally, a perineurioma or “adventitial dermal dendrocytoma,” cannot exclude early dermatofibrosarcoma protuberans, by Tim McCalmont. It seems clear to us that there are obvious problems with classification of this lesion.

The most important question to solve this diagnostic quandary is: what is the basic pathologic process? My answer from interpretation of the scanning magnification where one sees a mid-to-deep dermal, ill-defined plaque with an increase of collagen, is a fibrosing dermatitis. The lesion is not a malignant neoplasm, in particular not a dermatofibrosarcoma protuberans, as the growth appearance is homogenous, the periadnexal connective tissue is spared and not destructed and there are no lace-like extensions into the subcutis. The lesion is also not a benign neoplasm, like a trichogenic myxoma or a perineurioma, and not a hamartoma or malformation, like a trichodiscoma (or fibrofolliculoma), as all these lesions are usually well-circumscribed, both clinically and histologically. Fibrocytes in a fibrosing dermatitis (2) that are mostly spindly, sometimes epithelioid (e.g., in epithelioid cell histiocytoma, granular cell and clear cell dermatofibroma), with variable coarse collagen bundles in haphazard array mostly in the reticular dermis, sometimes also in the papillary dermis (e.g. in atrophic dermatofibroma and epithelioid cell histiocytoma) and/or subcutaneous tissue (in deep penetrating dermatofibroma), indicate a dermatofibroma. Such lesions are usually well-vascularized, as in the present case, reveal variable presence of lymphocytes (which admittedly is minute), and show epidermal, melanocytic, and adnexal hyperplasia visible at scanning magnification: the epidermis is broadened, the adnexal epithelium is increased, more ramified, and fenestrated (somehow reminiscent of eccrine syringofibroadenoma), and both, epidermis and adnexa, are more densely stained than comparable epithelia lateral to the lesion.

In short, this is a dermatofibroma, and it does not matter what immunohistochemistry shows. Over the years we have collected more than 3000 dermatofibromas; in those cases where immunohistochemistry was done reactivity for CD34 is conventionally seen in the surroundings or at the periphery of the lesion, but in a small number of cases we have seen diffuse reactivity of an otherwise typical dermatofibroma.

The situation seems similar to dermal plaque-like fibromatosis (3) or dermatomyofibroma (4), which characteristically are dermal plaques with horizontal orientation of fibrocytes or myofibrocytes, spare the adnexal structures, and are positive for smooth muscle actin. While we have seen such cases, which clinically usually are large plaques around the shoulder girdle of young females, we have also observed identical lesions without reactivity for smooth muscle actin (5) and similar smooth muscle actin positive lesions with “overlap” to more conventional dermatofibromas. According to the combination of conventional and unconventional findings within a single lesion we have suggested the term combined dermatofibromas (6). Recognition of combined dermatofibroma allows the histopathologist to apply a confident benign label to unusual lesions that might otherwise elude diagnosis, or tempt him or her into describing “new” entities, or, occasionally, to make a misdiagnosis of malignancy. The situation of the case under discussion seems similar to us. The basic pathologic process is fibrosing dermatitis of the dermatofibroma type despite CD34 reactivity!

Bernhard Zelger, M.D.

Back to Top | Article Outline


1. Pitha J, Smoller BR, Somach S, et al. A spindled cell CD34+ dermal proliferation. Am J Dermatopathol 2002; 24:85–8.
2. Ackerman AB. Dermatofibroma. In: Ackerman AB, Chongchitnant N, Sanchez J, et al., eds. Histologic diagnosis of inflammatory skin disorders: an algorithmic method based on pattern analysis, 2nd ed. Baltimore: Williams & Wilkins, 1997:279–88.
3. Hügel H. Die plaqueförmige dermale Fibromatose. Hautarzt 1991; 42:223–6
4. Kamino H, Reddy VB, Gero M, et al. Dermatomyofibroma–a benign cutaneous, plaque-like proliferation of fibroblasts and myofibroblasts in young adults. J Cut Pathol 1992; 19:85–93.
5. Zelger BG, Zelger B. Dermatofibrome: ein klinisch-pathologisches Klassifikationsschema. Pathologe 1998; 19:412–9.
6. Zelger BG, Sidoroff A, Zelger B. Combined dermatofibroma: coexistence of two or more variant patterns in a single lesion. Histopathology 2000; 36:529–39.
© 2002 Lippincott Williams & Wilkins, Inc.