Japanese Journal of Clinical Oncology

MIB-1 Labeling Index

Immunostaining for MIB-1 was performed by the same procedure as the other monoclonal antibodies above. The MIB-1 labeling index (LI) was determined as the percentage of Ki-67 positive cells based on a count of 1000 tumor cells (3,4). The LI of the present tumor was 8.5 (Fig. 4f).

DISCUSSION

We had to differentiate several spindle cell neoplasms in making a final diagnosis in the present case.

On routine microscopic examination, the neoplastic cells were all spindle shaped and showed interlacing bundle formation and a herringbone pattern, suggesting fibrosarcoma (5), but a hemangiopericytomatous pattern characteristic of solitary myofibroma and infantile myofibromatosis was not seen in the present tumor (6). The present tumor showed no significant dermal/epidermal and dermal changes, which precluded the possibility of fibrosarcomatous areas of dermatofibrosarcoma protuberans (DFSP) (7). Histochemically, reticulin fibers separating individual neoplastic cells were detected but boxing appearance diagnostic of leiomyosarcoma was not seen (8).

Table 1. Antibodies used for immunohistochemical staining

Antibody	Specificity	Source	Antigen retrieval
1A4	Alpha-smooth muscle actin (ASMA)	DAKO Japan, Kyoto	-
CALP	Calponin	DAKO Japan, Kyoto	Pronase
h-CD	High molecular weight caldesmon (h-CD)	DAKO Japan, Kyoto	Microwave
SMMS-1	Smooth muscle myosin, heavy chain (SM-MHC)	DAKO Japan, Kyoto	Boiling
D33	Desmin	DAKO Japan, Kyoto	-
DER-R-11	Desmin	Novocastra, Newcastle, UK	Microwave
V9	Vimentin	DAKO Japan, Kyoto	-
Polyclonal*	Factor-XIIIa	Calbiochem, La Jolla, CA	Pronase
JC/70A	CD31	DAKO Japan, Kyoto	Pronase
MY10	CD34	Nippon Becton Dickinson, Tokyo	Pronase
AE1/AE3	Cytokeratins (both high and low molecular weight)	Boehringer Mannheim Biochemica, Mannheim, Germany	Trypsin
E29	Epithelial membrane antigen (EMA)	DAKO Japan, Kyoto	Trypsin
CIV22	Type IV collagen	DAKO Japan, Kyoto	Pronase
Polyclonal*	Laminin	Seikagaku, Tokyo	Trypsin
Polyclonal*	S-100 protein	DAKO Japan, Kyoto	-
MIB-1	Ki-67	Immunotech, Marseille, France	Microwave

^*These three antibodies are rabbit polyclonal antibodies and all the others are mouse monoclonal antibodies.

We used several markers for smooth muscle cells and fibroblastic and myofibroblastic cells. Generally, ASMA, calponin and SM-MHC are positive in smooth muscle cells, myofibroblasts and myoepithelial cells (9). In contrast, h-CD is positive only in smooth muscle cells and a subset of myoepithelial cells in the galactophorous sinuses and generally negative for myofibroblasts and fibroblasts (1). It is said that the presence of desmin diffusely throughout a tumor is indicative of myoid differentiation (10). The presence of either ASMA or desmin focally should not necessarily be equated with myoid lineage but rather with myofibroblastic lineage (9,10). Factor-XIIIa is used as a marker of fibroblastic or myofibroblastic neoplasms (11). Basement membrane-associated molecules including type IV collagen and laminin are detected around each neoplastic cells of smooth muscle tumors (12,13) and those of neurogenic tumors (14). Fragments of basement membrane are also detected around myofibroblasts (15). In the present tumor, the neoplastic cells were negative for S-100 protein, the useful marker in diagnosis of schwannoma (14). Although more than 50% of the neoplastic cells were positive for CD34, the marker specific for endothelial cells, dendritic interstitial cells, peri-follicular stromal cells, SFT and DFSP (7,16,17), we discounted the possibility of SFT and DFSP. In contrast to the present tumor, SFT displays different growth patterns including a haphazard or vague storiform arrangement and a hemangiopericytomatous pattern (17). Fibrosarcoma areas of DFSP are indistinguishable from the present tumor, microscopically and immunohistochemically; in contrast, however, the present tumor showed no microscopic findings characteristic for DFSP (7). ASMA was positive in a few reported cases of fibrosarcomas, indicating myofibroblastic differentiation (18). EMA negativity in any of the neoplastic cells suggested that they were not epineural fibroblasts in origin (14). The results of the immunohistochemical study suggested that the neoplastic cells showed differentiation toward fibroblasts/myofibroblasts/dendritic interstitial cells. Although positive for CD34, the neoplastic cells were not endothelial, owing to their negativity for CD31, which is more specific for endothelial cells than CD34 (19). Ultrastructurally, it has already been elucidated that some neoplastic cells of fibrosarcomas showed the features of myofibroblasts (20,21). The MIB-1 LI ranged from 3.0 to 10.8 and from 5.3 to 27.6 for three ovarian fibrosarcomas and for five soft tissue fibrosarcomas, respectively (3,4). The MIB-1 LI of the present tumor was therefore consistent with malignancy rather than benignancy. The routine macroscopic, microscopic and immunohistochemical studies mentioned above indicate the diagnosis of well differentiated fibrosarcoma, adult type.

Fibrosarcoma generally involves foremost deeper structures, where it tends to originate from the intra- and intermuscular fibrous tissue, fascial envelopes, aponeuroses and tendons, namely subfascial involvement (5,22). Most of the smaller tumors tend to be well circumscribed and frequently are partly or completely encapsulated like the present tumor. Hence small, well circumscribed fibrosarcomas may be misleading and may result in an erroneous benign diagnosis and inadequate therapy (5,22). Fibrosarcomas arising from the subcutis excluding dermatofibrosarcoma protuberans are rare and tend to arise in tissues damaged by radiation, heat or scarring (5,22). The present tumor was freely movable and did not seemed to be fixed to the surrounding tissue, suggesting the subcutaneous tissue in origin not arising from the epicranial aponeurosis. The patient had no past history of radiation or scarring in the occipital region. It is suggested that the present tumor arose from the subcutaneous tissue and showed superficial growth, that is, upward growth to the overlying subcutaneous tissue without displaying involvement of aponeurosis and downward growth to the skull. Fibrosarcoma occurring in children [ge]10 years have a metastatic rate close to that of adult patients (i.e. 50%) (23,24). The patient therefore must be followed up carefully, although he has already had an additional local resection, the specimen of which contained no residual neoplastic cells.

Recently, a variety of disease entities have been proposed in the range of low-grade fibrosarcoma, such as sclerosing epithelioid fibrosarcomas (25) and low-grade fibromyxoid sarcomas (26) and myofibrosarcomas (27). Sclerosing epithelioid fibrosarcomas are characterized by a proliferation of fairly uniform, small, slightly angulated, round to ovoid epithelioid cells with prominent hyaline sclerosis (25), which were not detected in the present tumor. Low-grade fibromyxoid sarcomas are characterized by contrasting fibrous and myxoid areas of various size, a swirling, whorled growth pattern at least in part, low to moderate cellularity and bland, deceptively benign-appearing fibroblastic spindle cells with no or slight nuclear pleomorphism and few mitotic figures (26); however, the present tumor did not show myxoid change. Myofibrosarcomas usually arise in the breast and are characterized by spindle cells of moderate to large size, mostly arranged in large and irregular, straight or sinuous fascicles, focal nuclear pleomorphism (27); however, the fascicles were uniform and the neoplastic cells had uniform spindle cells without nuclear pleomorphism in the present tumor.

In conclusion, the present tumor was composed of monomorphic spindle cells showing interlacing bundle formation and a herringbone pattern, positive for the markers of fibroblasts or myofibroblasts. The neoplastic cells were also positive for CD34; however, we should diagnose the present tumor as well differentiated fibrosarcoma, on the basis of the routine microscopic findings.

Acknowledgments

The authors are grateful to Professor Masazumi Tsuneyoshi, Second Department of Pathology, Faculty of Medicine, Kyushu University for kind comments about the interpretation of microscopic findings in the present case, Ms Hisayo Yamasaki, Ms Miko Mitani, Mr Tadatoshi Tokaji and Mr Yoshihiro Hayashi, First Department of Pathology and Mr Masatoshi Shirota, Medical Research Center, Kochi Medical School for their excellent technical assistance.