HISTORY: 10 year old female spayed Japanese Chin developed a mass on the right zygomatic arch. Radiographs showed regions of bone lysis and proliferation and invasion of the right temporalis muscle.
The zygomatic bone is effaced by an expansile locally infiltrative, poorly cellular neoplasm of mesenchymal cells arranged in small whorls and streams in an abundant sclerotic collagenous stroma with ossification of the stroma and remodeling forming marrow spaces in the central regions. Neoplastic cells have indistinct cell borders, scant cytoplasm, and are fusiform. Nuclei are flattened, ovoid, and often hyperchromatic, with some at the periphery having finely stippled chromatin. At the leading edge of the tumor the reactive bone forms a shell around the tumor, and is being lysed as the tumor expands and infiltrates. In some areas the neoplasm infiltrates and expands the surrounding skeletal muscle, and forms small whorls which advance out from the main mass.
Low power view of zygomatic arch tumor, showing some connection to the bone in the lower left corner, and the extension into the muscle into the upper left corner. The mass is haphazardly ossified, and interspersed with dense hypocellular fibrous tissue.
High power magnification at one edge of the tumor where it was less ossified, showing a hypocellular whorling mass of fibrous tissue eroding the surrounding reactive bone to the right of the image. This shell of reactive bone is indicative of a moderately growing mass. Slower growing tumors may have lamellar bone at the periphery while more rapidly growing tumors efface any reactive bone.
Moderate magnification of the center of the mass, in a region where the fibrous tumor is effacing islands of reactive woven bone. The woven bone is bordered by flattened to plump osteoblasts and loose connective tissue, but in some areas the fibroblastic tumor erodes them. In the center there is a whorl of fibrous tumor with a small area of mineralization. This is deceptive, as it can be interpreted as bone produced by the tumor. But in this case, this is simply mineralization of a dense collagenous stroma, which can occur in the right microenvironment.
High magnification view of a region of mineralizing tumor stroma. Notice the blending of collagen fibers of the tumor with the surrounding bone, and in the center the slight stippling of basophilic stroma indicating small areas of mineral deposition that are independent of osteoblasts.
Right zygomatic arch: Low grade sclerosing fibrosarcoma with osseous metaplasia
This is an uncommon tumor which most likely arose from either the periosteum or medullary cavity of the zygomatic bone. Differentials included were benign fibro-osseous tumors such as ossifying fibroma, or fibrous dysplasia, however the bone produced seemed to be eroded by the tumor stroma as it advanced, and did not seem to be produced by it. The central regions of osseous metaplasia were formed by calcification of the haphazard collagenous fibers produced by the tumor, which can be seen using polarized light microscopy. As the tumor matrix mineralized it then became fair game for the resident osteoclasts and osteoblasts to begin laying down reactive woven osteoid thereby forming the ossified portions of the tumor. The remodeling that has occurred is a factor of the prolonged time course and slow growth rate of this tumor.
A google search for the term sclerosing fibrosarcoma will find a medical diagnosis of “sclerosing epithelioid fibrosarcoma”, which has a particular histological appearance that this tumor did not share. The term “sclerosing fibrosarcoma” is more a descriptive term, and one given to me Dr Roy Pool from Texas A&M when I consulted him on this tumor. Perhaps this may become a distinctive entity in veterinary medicine as more cases are described and recognized. The biological behavior appears to be less aggressive than other fibrosarcomas of bone, and should be differentiated from the highly aggressive entity of the “histologically low-grade, biologically high-grade fibrosarcoma” of the mandible and maxilla.
The closest human analog to this tumor is the desmoplastic fibroma, but again the histological appearance is distinctive. The word “desmoplastic” refers to the abundnat collagen fiber content in this tumor which resembles connective tissues and shows no mineralization (desmos: A bond, fastening or chain; ie desmitis -inflammation of a tendon). The tumor in this bone does not share this type of collagenous fiber arrangement.
Initially, prior to histopathology, I was expecting a multilobular tumor of bone (MLTB; Multilobular osteochondrosarcoma) given the location. These tumors appear to be associated with bone symphyses and suture lines (skull, hard palate, maxilla, mandible, zygomatic bone). The microscopic architecture is usually characteristic, however it is said that as these become high grade this architecture is lost. Differentiating a high grade MLTB from an osteo- or chondrosarcoma can be difficult, but in my experience there are regions of spindle cell sarcomatous sheets and bundles that are not seen in chondrosarcoma, and there is no convincing bone produced. This probably reflects the fibro-osseous origin of the tumor. Nevertheless the tumor in this case is very low grade, and had no features in common with a MLTB.
If anyone has seen a similar tumor in the skull or long bones I would appreciate hearing about it.
Desmoplastic fibroma of bone: A report of 8 cases and review of the literature. J Bone Joint Surg Am.