 |
 |
 |
|
| home | CME | help | feedback | subscriptions | archive |
|
© 1997 The Journal of Bone and Joint Surgery, Inc. Late Malignant Transformation of a Benign Giant-Cell Tumor of Bone. A Case Report* , LEIDEN, THE NETHERLANDSInvestigation performed at Leiden University, Leiden
Malignant giant-cell tumors of bone are very rare. These tumors usually present as de novo lesions or, several years after treatment with curettage combined with radiation therapy, as malignant transformation of a benign giant-cell tumor. However, malignant transformation without adjuvant radiation therapy has also been reported1,2,5,8-12,14,16. We present a well documented case of late progression of a giant-cell tumor of bone to a malignant fibrous histiocytoma.
In September 1991, a forty-five-year-old man was seen because of pain on the medial side of the left knee, especially at night. Almost twenty years earlier, in March 1972, a benign osteolytic lesion, which had been radiographically and histologically consistent with a low-grade giant-cell tumor of bone (Figs. 1 and 2), had been removed with curettage from the medial portion of the epiphysis and metaphysis of the proximal aspect of the left tibia. The defect had been filled with bone taken from the left iliac crest. No radiation therapy had been given. Follow-up radiographs made in 1980 had demonstrated sclerosis at the original site of the tumor as a result of the bone-grafting. No signs of recurrent tumor had been noted. Postoperatively, the patient had been pain-free until August 1991. Radiographs made in September 1991 showed no signs of local recurrence.
The tissue that had been removed with curettage in 1972 had consisted of a stroma of mononuclear cells with round or oval nuclei. Numerous multinucleated giant cells were distributed randomly between the stromal cells. The level of mitotic activity was low (approximately one mitosis per ten high-power fields). In some areas, the stromal cells were more spindle-shaped and had a tendency to form bundles. A varying amount of fibrosis was present. Sometimes an admixture with a round-cell inflammatory infiltrate was found (Fig. 2). On the basis of the histopathological findings, a conventional benign giant-cell tumor of bone was diagnosed. The tumor was classified as grade II, according to the four-grade system of Mulder et al. Comparison of radiographs made in March 1992 with those made in 1980 and September 1991 demonstrated renewed osteolysis in the medial part of the proximal aspect of the tibia (Fig. 3). Magnetic resonance imaging demonstrated viable tissue within the lesion, cortical interruption, and soft-tissue involvement (Fig. 4). A biopsy with a Jamshidi trocar was performed18.
Histological examination revealed tissue fragments composed of markedly anisomorphic mesenchymal cells with a marked degree of nuclear hyperchromasia and numerous giant cells. A few multinucleated osteoclast-like giant cells were also found in these areas. In other areas, the cells were more spindle-shaped and were arranged in interwoven bundles; in some areas, a storiform pattern was noted. The spindle cells showed a moderate degree of pleomorphism. Generally, the level of mitotic activity was low; however, in some areas, it was as high as one mitotic figure per two to three high-power fields. Some of the tissue fragments contained fibrosis and foam cells. Areas of necrosis were also observed. No deposition of osseous or cartilaginous extracellular matrix was found. The tumor was classified as a malignant fibrous histiocytoma of bone (Figs. 5-A and 5-B).
There was no sign of associated regional lymphadenopathy, and computed tomography of the chest showed no metastases. Chemotherapy was given according to the scheme of the European Organization on Research and Treatment of Cancer (protocol number 80871); every three weeks, the patient received Adriamycin (doxorubicin) (twenty-five milligrams per square meter of skin per day for three days) and cisplatin (100 milligrams per square meter of skin for one day). After three courses of chemotherapy, an en bloc resection of the medial part of the proximal aspect of the tibia was performed. A ten-centimeter-long hemi-osteoarticular allograft from the proximal aspect of a donor tibia was obtained from the Leiden Bone Bank and used for anatomical reconstruction of the knee. Histological evaluation of the resection specimen showed morphological characteristics comparable with those demonstrated by the previous biopsy, with adjacent fields of necrosis, reactive fibrosis, and vascular proliferation. All margins of resection were free of tumor. The postoperative recovery was uneventful. The patient received another three courses of chemotherapy. After one year, the knee was pain-free, with an active range of motion from full extension to 130 degrees of flexion. A slight varus deformity (angulation) was noted. The patient was able to work as a mechanic without restrictions. Two years postoperatively, the varus deformity appeared to be progressing slowly. In June 1994, an open-wedge osteotomy of the left tibia was performed to restore the neutral axis of the knee. Six months after the open-wedge osteotomy and two and one-half years after the resection of the malignant fibrous histiocytoma-like giant-cell tumor, no metastatic disease was evident. Radiographs of the left tibia showed formation of callus, and the patient had regained a full range of motion of the knee. However, in March 1995, he was readmitted to our clinic because of pain in the right hip. Radiographs showed lytic destruction of the proximal aspect of the right femur, right acetabulum, and right lateral wing of the ilium. A bone scan showed high uptake on the right side of the pelvis as well as in the proximal aspect of the right femur, the mid-cervical vertebrae, the ninth and eleventh thoracic vertebrae, the proximal aspect of the left humerus, the second rib on the left, and the eighth rib on the right. These areas of increased uptake had not been present on a bone scan made in June 1992. A specimen from the proximal aspect of the right femur, obtained during a biopsy with a Jamshidi trocar, showed a storiform spindle-cell sarcoma (Fig. 6), consistent with metastases from the sarcoma resected three years earlier.
In the past, it has been reported that 30 per cent of all giant-cell tumors are malignant5,8. Currently, pathologists seldom diagnose malignant giant-cell tumor. Most such lesions are considered to be malignant fibrous histiocytoma or giant-cell-rich variants of either fibrosarcoma or osteosarcoma. In a series of 529 giant-cell tumors that we identified from the registry of The Netherlands Committee on Bone Tumors, we noted only one lesion that we considered to be a malignant giant-cell tumor; this diagnosis was supported by the radiographic features. Dissemination to distant sites, especially the lungs, in patients who have a benign giant-cell tumor of bone has been reported6,12. In the patient described in this paper, a grade-II giant-cell tumor was diagnosed in the proximal part of the tibia. The lesion had the typical features of a benign giant-cell tumor, without any histological changes that suggested malignancy. The lesion was curetted, but no radiation therapy or other adjuvant treatment was given. Twenty years later, the patient had a recurrent lesion with the features of a malignant fibrous histiocytoma. Extensive examination of the specimen from the en bloc resection did not reveal any regions with the typical features of the previously diagnosed benign giant-cell tumor. In the areas that had accumulations of multinucleated giant cells, the tissue was rather vascular and edematous and contained inflammatory cells. This finding suggests a proliferation of reactive tissue rather than the presence of tumor tissue, which raises the question of whether the lesion was a second primary tumor with the features of a malignant fibrous histiocytoma or a recurrent giant-cell tumor that had transformed into a malignant fibrous histiocytoma. The fact that the lesion developed at the site where the original giant-cell tumor had been curetted, and the fact that giant-cell tumors can have features of fibrohistiocytic differentiation, suggest a true recurrence rather than a second primary tumor. The relatively low level of activity in the original tumor, which had only one mitosis per ten high-power fields, could explain the slow recurrence of the present tumor. Another, highly unlikely, explanation is that the malignant fibrous histiocytoma developed in a bone-grafted area. This phenomenon has not been described before, to our knowledge, but it may be comparable with that of a malignant fibrous histiocytoma occurring within an area of a bone infarct3,4,7,13. In both instances, the reparative proliferative changes in the border of an area of dead bone serve as the nidus for the formation of a malignant tumor. Of the 529 giant-cell tumors that we identified from the registry, only the one described in the present report was a malignant lesion that had recurred without previous radiation therapy for the primary tumor. The case described by Kimball and Desanto differed from that of the patient described here; the grade-II giant-cell tumor in their patient revealed metastatic potential as early as three years after resection of the primary tumor. In our patient, a malignant phenotype was described only in relation to the metastases. Only after the specimens from the original tumor had been re-examined, and atypical areas with osteoid deposition were identified, was the lesion classified as borderline malignant. Of seventy-six patients reported on by Hutter et al., three had transformation to a malignant tumor, all within five years after operative therapy only. In a study of 195 giant-cell tumors, Dahlin et al. found malignant transformation in only two patients, seventeen months after operative treatment of a low-grade giant-cell tumor in one and fifteen years after operative treatment of a recurrent giant-cell tumor in the other. Rock et al. reported that of nineteen patients who had a secondary malignant giant-cell tumor of bone, one had not received radiation therapy in the course of treatment of a benign giant-cell tumor. That patient had a grade-III fibrosarcoma fifteen years after curettage and bone-grafting and, some time later, local recurrence. In a study of forty-one patients, Mnaymneh et al. reported fibrosarcomatous changes in two. In one of these patients, the changes were found four years after operative removal of a giant-cell tumor of the sacrum; in the other, they were found five years after operative removal of a similar tumor in the tibia. Schajowicz described the case of a twenty-one-year-old woman in whom, twenty-eight months after curettage and bone-grafting of a histologically benign-looking giant-cell tumor in the proximal aspect of the tibia, an osteosarcoma was diagnosed at the same site. Boriani et al., in a study of 327 patients, reported two malignant recurrences, among patients who had not received any radiation therapy, 1.5 and two years postoperatively. In all of the reports of very early malignant progression of a benign giant-cell tumor, it remains doubtful whether the tumor was not in fact malignant from the onset without being recognized as such or whether it already contained malignant areas that might have been overlooked as a result of an error in the sampling of the original specimen. In conclusion, the diagnosis of a malignant giant-cell tumor is made very rarely. Malignant transformation after operative treatment of a benign giant-cell tumor hardly ever occurs when radiation therapy has not been given. In the rare instances of malignant transformation, the recurrence usually takes place within five years. However, malignant recurrence of a benign giant-cell tumor is possible even two decades after curettage, as demonstrated in the case of the patient reported on here.
*No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article. No funds were received in support of this study.
  CiteULike  Connotea  Del.icio.us  Facebook  Technorati  Twitter What's this?
This article has been cited by other articles:
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Top
Introduction
Case Report
| home | CME | help | feedback | subscriptions | archive | site index |
|
Stanford University Libraries' HighWire Press (TM) assists in the publication of JBJS Online. |