Cutaneous Mesenchymal Tumors and Non-Mesenchymal Mimics:
An Update and Approach to Diagnosis
Case 5 -
Dermatofibrosarcoma Protuberans with Fibrosarcomatous Change
Steven D. Billings, M.D.
Andrew L. Folpe, M.D.
A 32 year-old man presented with a firm nodule on the trunk. An excisional biopsy was obtained.
The biopsy showed relatively normal epidermis overlying a circumscribed, moderate to
markedly cellular dermal and subcutaneous tumor. The majority of the tumor had a distinctly fascicular
growth pattern that in areas had a herringbone pattern. In this area the tumor showed increased
cellularity, cytologic atypism, and numerous mitotic figures. At the periphery, the tumor had a more
storiform pattern, was less cellular, and was composed of uniform, slender spindled tumor cells.
Entrapment of individual fat cells was seen in the storiform area.
Diagnosis: Dermatofibrosarcoma protuberans with fibrosarcomatous change
Clinical and Histopathological Features
Dermatofibrosarcoma protuberans (DFSP) most commonly presents in early to mid adult life. However, a
wide age range may be affected, and cases in children and even infants are well documented. Typically
DFSP arises in the trunk or proximal extremities. The initial presentation is as a plaque that may
subsequently develop into a nodular or multinodular mass. The plaque stage may clinically cause
confusion with sclerosing dermatoses such as morphea or scleroderma.
In its classic form, DFSP is a relatively stereotypical lesion, consisting of an exquisitely storiform
proliferation of monotonous, slender spindled cells with dark nuclei and lightly staining cytoplasm,
typically showing diffuse infiltration of the subcutaneous adipose tissue ("honeycomb pattern"). DFSP
usually displays little heterogeneity within a given lesion, lacking the admixed cell populations of
siderophages and foamy macrophages seen in its principal morphologic mimic, dermatofibroma/ benign
fibrous histiocytoma. However, small subsets of DFSP may deviate morphologically from this classic
appearance, and may show such features as GCFB-like areas, fibrosarcomatous change, myxoid change and
myoid nodule formation. Prominent myxoid change may cause significant diagnostic difficulty, inasmuch as
the accumulation of ground substance can efface the characteristic storiform pattern, cause the tumor
cells to take on a more stellate appearance, and accentuate the usually inapparent, but surprising
complex, underlying capillary vasculature, reminiscent of that seen in myxoid liposarcoma. Myoid nodules
are characterized by a smooth muscle or myofibroblastic proliferation surrounding vessels within the
tumor. This appears to be a reactive process.
The Bednar tumor, or pigmented DFSP is now recognized as a morphologic variant of DFSP. Pigmented
DFSP closely resembles ordinary DFSP, but is remarkable for the presence of pigmented dendritic cells
present in varying amounts within the tumor. Ultrastructurally, these dendritic cells contain mature
melanosomes. The origin of the pigmented cells remains controversial. Whereas some have proposed that
these tumors are colonized by melanin containing cell, others suggest that the tumor synthesizes
melanosomes. Regardless of the origin of these melanosomes, pigmented DFSP is clearly closely related to
DFSP, as evidenced by their similar clinical presentations, the presence of areas identical to ordinary
DFSP within pigmented DFSP, the occurrence of lesions with composite features of pigmented DFSP and GCFB,
and the presence in pigmented DFSP of the same genetic alterations as in ordinary DFSP.
A more controversial histologic finding is the presence of sarcomatous change in DFSP. Although these
areas may resemble malignant fibrous histiocytoma or myxofibrosarcoma, they most commonly resemble
ordinary fibrosarcoma. Such fibrosarcomatous areas have a fascicular herringbone pattern, and have more
prominent nuclear atypia and a higher mitotic rate compared with conventional DFSP. Fibrosarcomatous
transformation in DFSP appears to be related to mutations in the p53 pathway. Some authors have
suggested that DFSP-FS may have a higher rate of local recurrence and distant metastasis. The true
impact of fibrosarcomatous change on the biologic behavior of DFSP remains uncertain. Although
fibrosarcomatous change may signify a potentially more aggressive course, patients treated with an
adequately wide excision appear to have clinical outcomes similar or identical to those with conventional
There have been recent advances in the therapy of DFSP, which traditionally has a local recurrence
rate as high as 20% even with wide excision (3 cm margins). Mohs micrographic surgery appears to offer a
better approach with local recurrence rates between 2-7%. Dermatofibrosarcoma protuberans may grow in a
very asymmetric fashion and the Mohs technique can compensate for this phenomenon allowing complete
excision where a traditional wide local excision would leave residual tumor.
In 1990 the presence of a supernumerary ring chromosome in DFSP was first reported. Subsequently it
was discovered that the this giant chromosome contained an occult translocation, t (17;22), resulting in
a gene fusion whereby the platelet derived growth factor β (PDGFβ) gene is placed under the
control of the collagen type 1 α1 (COL1A1) gene promoter. It has been proposed that this fusion
gene product promotes autocrine stimulation of tumor growth in DFSP. It is now clear that almost all
DFSP contain this specific cytogenetic abnormality. As is discussed below, an identical translocation
has been identified in GCFB. Knowledge of the cytogenetic abnormality of DFSP has lead to potential
chemotherapeutic treatment. Imatinib mesylate (STI-571) selectively inhibits specific tyrosine kinases
including platelet-derived growth factor- β, and there is both in vitro and clinical evidence to
support a role for this drug in the treatment of DFSP in selected cases.
The differential diagnosis of DFSP primarily centers on dermatofibroma (DF). In many respects, this
differential diagnosis is less problematic than is traditionally thought provided an adequate biopsy.
Often times these lesions can be differentiated at scanning magnification. DFSP has a very infiltrative
pattern of growth that often permeates the subcutaneous fat while DF is a relatively circumscribed tumor
that rarely and only superficially involves the subcutis. Another scanning magnification feature is the
presence of hyperplasia and hyperpigmentation of the epidermis often seen overlying DF; this epidermal
change is much less often seen in DFSP. In DF the tumor cells are arranged in haphazardly arranged short
fascicles with a more heterogeneous population of cells that varies from plump spindle cells to
histiocytic cells and often with secondary changes including siderophages and inflammatory cells. This
is in contrast to the more uniform DFSP.
In superficial biopsies, it may still be difficult to distinguish DFSP from DF. In such cases,
immunostains for CD34 and Factor XIIIa may be helpful. When interpreting CD34 and Factor XIIIa
immunostains in this differential diagnosis, it is important to remember that both markers may be present
in both DFSP and DF. However, DFSP and DF have different patterns of expression of CD34 and Factor
XIIIa. DFSP shows uniform, strong expression of CD34 in essentially all of the lesional cells,
occasionally with a reactive cuff of Factor XIIIa-positive cells, whereas DF typically shows a more mixed
population of CD34 and Factor XIIIa-positive cells, admixed with S100 protein-positive Langerhans cells
and CD68-positive histiocytes. Pigmented DFSP may on occasion be confused with a true melanocytic tumor,
such as cellular blue nevus or melanoma. Unlike true melanocytic tumors, pigmented DFSP express CD34 and
are negative for S100 protein and melanocytic markers (e.g., HMB45, Melan A). Apolipoprotein D (ApoD)
has been relatively recently proposed as another useful positive marker of DFSP, in a single study. It
is unclear what advantages, if any, this marker has over CD34, and caution is merited in its potential
use in this differential diagnosis.
Giant Cell Fibroblastoma
It is now clear that GCFB (GCFB) is the juvenile form of DFSP. This was first proposed by Shmookler
et al in 1989, who detailed cases of GCFB with both histologic and clinical overlap with DFSP. This
relationship has been further supported by the description of tumors with composite features of DFSP and
GCFB, by their identical immunophenotype, and by cases of GCFB that have recurred as DFSP and vice
versa. Giant cell fibroblastoma also harbors the same cytogenetic abnormality as DFSP.
Clinical and Histopathological Features
GCFB usually occurs in a much younger patient population than does DFSP, with the majority of patients
in the first decade of life. Cases in adults are uncommon, although it should be noted that GCFB may
rarely be seen in older adults and that DFSP may occur in infants. Like DFSP, GCFB typically presents as
dermal or subcutaneous mass, which most frequently involves the trunk, thigh or inguinal region. The
clinical behavior of GCFB is essentially the same as DFSP, with frequent local recurrences, but only
extremely rare metastases. Fibrosarcomatous change is, however, extraordinarily rare in classic GCFB.
Histologically, GCFB shows varying cellularity and is composed of spindled to stellate cells with mild to
moderate nuclear pleomorphism loosely arranged in an abundant myxoid to hyalinized stroma. Typically
these spindled cells infiltrate around adnexal structures and though fat in an identical fashion to DFSP.
Most characteristically, this stroma displays prominent cracking artifact, with the formation of
pseudovascular spaces lined by a discontinuous layer of enlarged, multinucleated tumor giant cells. The
presence of such cells may result in an incorrect diagnosis of a pleomorphic sarcoma, if one is unaware
of the appearance of GCFB.
The differential diagnosis of GCFB usually centers on sarcomas such as myxofibrosarcoma (myxoid
malignant fibrous histiocytoma). Unlike GCFB, myxofibrosarcoma typically presents as a more deeply
seated mass in older adults, and is histologically characterized by multinodular growth, arborizing
thick-walled blood vessels, and solid nodules of pleomorphic, mitotically active cells. Although an
atypical DF (dermatofibroma with monster cells) might also be considered in the differential diagnosis of
GCFB, the absence of the distinctive myxohyaline matrix with cracking artifact should allow this
distinction without undue trouble. Probably the most difficult entity to distinguish from GCFB is the
fibrous hamartoma of infancy, a tumor that occurs in a very similar age group, may contain both
infiltrative monomorphic spindled cells as well as myxohyaline matrix, and may be CD34-positive. Unlike
GCFB, however, fibrous hamartoma of infancy is does not contain multinucleated giant cells, and is a
triphasic tumor, with nodules of primitive mesenchymal cells, fascicles of mature-appearing fibroblastic
cells, and fat.
- Pappo AS, Rao BN, Cain A, Bodner S, Pratt CB. Dermatofibrosarcoma protuberans: the pediatric
experience at St. Jude Children's Research Hospital. Pediatric Hematology & Oncology
- Sigel JE, Bergfeld WF, Goldblum JR. A morphologic study of dermatofibrosarcoma protuberans:
expansion of a histologic profile. Journal of Cutaneous Pathology 2000;27(4):159-63.
- Fletcher CD, Evans BJ, MacArtney JC, Smith N, Wilson Jones E, McKee PH. Dermatofibrosarcoma
protuberans: a clinicopathological and immunohistochemical study with a review of the literature.
- Orlandi A, Bianchi L, Spagnoli LG. Myxoid dermatofibrosarcoma protuberans: morphological,
ultrastructural and immunohistochemical features. Journal of Cutaneous Pathology 1998;25(7):386-93.
- Connelly JH, Evans HL. Dermatofibrosarcoma protuberans. A clinicopathologic review with emphasis on
fibrosarcomatous areas. American Journal of Surgical Pathology 1992;16(10):921-5.
- Goldblum JR, Reith JD, Weiss SW. Sarcomas arising in dermatofibrosarcoma protuberans: a reappraisal
of biologic behavior in eighteen cases treated by wide local excision with extended clinical follow up.
Am J Surg Pathol 2000;24(8):1125-30.
- Mentzel T, Beham A, Katenkamp D, Dei Tos AP, Fletcher CD. Fibrosarcomatous ("high-grade")
dermatofibrosarcoma protuberans: clinicopathologic and immunohistochemical study of a series of 41 cases
with emphasis on prognostic significance. American Journal of Surgical Pathology 1998;22(5):576-87.
- Bowne WB, Antonescu CR, Leung DH, Katz SC, Hawkins WG, Woodruff JM, et al. Dermatofibrosarcoma
protuberans: A clinicopathologic analysis of patients treated and followed at a single institution.
- Gloster HM, Jr., Harris KR, Roenigk RK. A comparison between Mohs micrographic surgery and wide
surgical excision for the treatment of dermatofibrosarcoma protuberans. Journal of the American Academy
of Dermatology 1996;35(1):82-7.
- Fletcher CD, Theaker JM, Flanagan A, Krausz T. Pigmented dermatofibrosarcoma protuberans (Bednar
tumour): melanocytic colonization or neuroectodermal differentiation? A clinicopathological and
immunohistochemical study. Histopathology 1988;13(6):631-43.
- Sirvent N, Maire G, Pedeutour F. Genetics of dermatofibrosarcoma protuberans family of tumors: from
ring chromosomes to tyrosine kinase inhibitor treatment. Genes, Chromosomes & Cancer
- Rubin BP, Schuetze SM, Eary JF, Norwood TH, Mirza S, Conrad EU, et al. Molecular targeting of
platelet-derived growth factor B by imatinib mesylate in a patient with metastatic dermatofibrosarcoma
protuberans.[comment]. Journal of Clinical Oncology 2002;20(17):3586-91.
- Goldblum JR, Tuthill RJ. CD34 and factor-XIIIa immunoreactivity in dermatofibrosarcoma protuberans
and dermatofibroma. American Journal of Dermatopathology 1997;19(2):147-53.
- Shmookler BM, Enzinger FM, Weiss SW. Giant cell fibroblastoma. A juvenile form of
dermatofibrosarcoma protuberans. Cancer 1989;64(10):2154-61.
- Beham A, Fletcher CD. Dermatofibrosarcoma protuberans with areas resembling giant cell
fibroblastoma: report of two cases. Histopathology 1990;17(2):165-7.
- Michal M, Zamecnik M. Giant cell fibroblastoma with a dermatofibrosarcoma protuberans component.
American Journal of Dermatopathology 1992;14(6):549-52.
- Cin PD, Sciot R, de Wever I, Brock P, Casteels-Van Daele M, Van Damme B, et al. Cytogenetic and
immunohistochemical evidence that giant cell fibroblastoma is related to dermatofibrosarcoma protuberans.
Genes, Chromosomes & Cancer 1996;15(1):73-5.