You are in: eMedicine Specialties > Orthopedic Surgery > NEOPLASMS Giant Cell TumorArticle Last Updated: Nov 12, 2002AUTHOR AND EDITOR INFORMATIONSection 1 of 11
  Author: Valerae O Lewis, MD, Assistant Professor, Department of Orthopaedic Oncology, MD Anderson Cancer Center Valerae O Lewis is a member of the following medical societies: American Academy of Orthopaedic Surgeons Coauthor(s): Terrance Peabody, MD, Assistant Professor, Department of Surgery, Associate Professor of Surgery, Section of Orthopedic Surgery and Rehabilitative Medicine, University of Chicago; A Kevin Raymond, MD, Section Head of Orthopedic Pathology, Associate Professor, Department of Pathology, University of Texas MD Anderson Cancer Center Editors: Lynn A Crosby, MD, FACS, Chief of Shoulder Division, Professor, Department of Orthopedic Surgery, Wright State University School of Medicine; Francisco Talavera, PharmD, PhD, Senior Pharmacy Editor, eMedicine; Sean P Scully, MD, PhD, Professor, Department of Orthopedics, University of Miami; Dinesh Patel, MD, FACS, Associate Clinical Professor of Orthopedic Surgery, Harvard Medical School; Chief of Arthroscopic Surgery, Department of Orthopedic Surgery, Massachusetts General Hospital; Harris Gellman, MD, Consulting Surgeon, Broward Hand Center, Voluntary Clinical Professor of Orthopedic Surgery and Plastic Surgery, Departments of Orthopedic Surgery and Surgery, University of Miami School of Medicine Author and Editor Disclosure Synonyms and related keywords: GCT, osteoclastoma INTRODUCTIONSection 2 of 11
  Cooper and Travers first described giant cell tumor (GCT) of bone in 1818 (Cooper, 1818). GCTs of bone have been described as the most challenging benign bone tumors (McDonald, 1998). Although benign, GCTs show a tendency for significant bone destruction, local recurrence, and occasionally metastasis. The natural history of GCTs varies widely and can range from local bony destruction to local metastasis, metastasis to the lung, metastasis to lymph nodes (rare), or malignant transformation (rare) (Unni, 1996; Campanacci, 1987; McDonald, 1986; Goldenberg, 1970; McDonald, 1998; Cheng, 1997; Connell, 1998; Dahlin, 1985; Dahlin, 1970; Fitz, 1966; Kay, 1994; Kitano, 1999; Kreieberg, 1985; Mirra, 1982; Mnaymneh, 1964; Present, 1986; Riley, 1967; Shifrin, 1972). Approximately 3% of GCTs metastasize to the lung. The metastases appear as clusters of GCTs located within the lung (Connell, 1998; Bertoni, 1988; Maloney, 1989; Szyfelbein, 1979; Tubbs, 1992; van Hoeven, 1994). GCTs metastasis generally appear an average of 3-5 years after the initial diagnosis of the primary lesion. However, GCTs metastasis may not be detected for 10 years or longer (Kay, 1994; Maloney, 1989; Szyfelbein, 1979; Tubbs, 1992; Rock, 1984). The natural history of these lung metastases is as unpredictable as that of the primary disease (Cheng, 1997; Kay, 1994; Mirra, 1982; Bertoni, 1988; Rock, 1984; Nojima, 1994). Pulmonary metastases that spontaneously regress, remain stable, continuously grow slowly, or rapidly progress have been reported. FrequencyIn the United States and Europe, GCTs represent approximately 5% of all primary bone tumors and 21% of all benign bone tumors (Unni, 1996). In China, GCTs account for 20% of all primary bone tumors (Sung, 1982). A female predominance exists, with a female-to-male ratio of 1.3-1.5:1 (Unni, 1996; Frassica, 1993; Schajowicz, 1991). GCTs occur most commonly in the third decade of life; less than 5% of GCTs occur in patients who are skeletally immature (Unni, 1996; Campanacci, 1987; Kransdorf, 1992; Picci, 1983). In the Mayo Clinic series, 84% of the GCTs occurred in patients older than 19 years (Unni, 1996). PathophysiologySee Workup, Histologic Findings, below. ClinicalMost GCTs are located within the epiphyses of long bones but often extend into the metaphysis. In several published series, only 1.2% of GCTs involved the metaphysis or diaphysis without epiphyseal involvement (Bogumill, 1972; Campanacci, 1975; Fain, 1993; Peison, 1976; Sherman, 1961; Wilkerson, 1969). Approximately 50% of GCTs are located about the knee at the distal femur and proximal tibia, with the proximal humerus and distal radius representing the third and fourth most common sites, respectively (see Image 1). Most commonly, GCTs are solitary lesions; less than 1% are multicentric. Fewer than 60 multicentric lesions have been reported in the literature (Cummins, 1996; Hindman, 1994; Kadir, 1978; Kaufman, 1977; Madhuri, 1993; Peimer, 1980; Sanghvi, 1999; Sim, 1977). Multicentric involvement tends to be more clinically aggressive, and, unlike the solitary lesions, multicentric GCT has a propensity for the small bones of the hands and feet. Patients with multicentric lesions tend to be younger than those with lesions elsewhere. Pain is the most common presenting symptom. Swelling and deformity are associated with larger lesions. Soft-tissue extension is common. The incidence of pathologic fracture at presentation is 11-37% (Campanacci, 1975; Hudson, 1984; Enneking, 1983). INDICATIONSSection 3 of 11
The presence of tumor is the indication for surgery. See Treatment, below, for detailed information. CONTRAINDICATIONSSection 4 of 11
Radiation therapy and embolization generally are reserved for lesions or for patients in whom surgical treatment is not feasible. Radiation therapy has been proposed for patients who are not surgical candidates, for those whose tumors are in locations not amenable to operative treatment, and for those in whom a potential for significant morbidity from tumor relapse or subsequent surgery exists (Malone, 1995; Schwartz, 1989). WORKUPSection 5 of 11
Imaging Studies
Histologic FindingsOn gross inspection, these lesions are characteristically chocolate brown, soft, spongy, and friable. Yellowish-to-orange discoloration due to hemosiderin may be present. Cystic cavities within the tumor are common. Often, these cavities are blood filled (see Image 8). Examination of resected specimen reveals a variable degree of cortical expansion and disruption. Despite the cortical disruption, the periosteum rarely is breached (Unni, 1996) (see Image 9). Histologically, the lesions tend to be cellular (see Image 10). Although the multinucleated giant cell is the characteristic cell type, these lesions have a background network of stromal mononuclear cells. The mononuclear cells are plump and round, oval, or spindle shaped. They may have prominent mitotic activity, but cellular atypia is rare (see Image 11). The degree of mitotic activity has no prognostic significance. Multinucleated giant cells, as the name suggests, have numerous centrally located nuclei as opposed to the peripherally located nuclei of Langerhans-type giant cells seen in atypical infections (see Image 12). The nuclei tend to be compact and oval and contain prominent nucleoli. These are similar in appearance to those of the surrounding stromal cells, and the giant cell often appears tobe a syncytium of these stromal cells. Giant cells generally are distributed throughout the lesion. The concentration of multinucleated giant cells varies considerably from tumor to tumor. Some tumors have many multinucleated giant cells, whereas others have a few giant cells nestled in swirls of spindle-shaped stromal cells (see Image 13). The concentration of multinucleated giant cells is not related to the incidence of local recurrence or metastases. In some lesions, giant cells invade the small perforating vessels (see Image 14). This intravascular invasion can be found in approximately 5% of cases. This invasion, although appearing aggressive, is not correlated with the prognosis (Sanerkin, 1980). At histologic analysis, the differential diagnosis includes brown tumors of hyperparathyroidism; aneurysmal bone cysts; and, rarely, chondroblastoma, osteoblastoma, or osteosarcoma. In an attempt to relate the histologic features with the clinical course, several histologic grading systems have been developed. The earliest was devised by Jaffe et al in 1940. In grade I at the benign end of the spectrum, giant cells are numerous, mononuclear cells are rare, and mitotic activity is absent. In grade II, mononuclear stromal cells are numerous, and moderate atypia and mitotic activity is seen. In grade III, giant cells are few and small, atypia and pleomorphism are common, and mitotic activity is frequent. However, this grading system has no prognostic significance (Frassica, 1993; Schajowicz, 1991; Campanacci, 1987; Goldenberg, 1970). In an attempt to improve the prognostic relevance of the histologic grading system, several authors have modified the staging system of Jaffe et al (Sanerkin, 1980). Generally, these staging systems include sarcomatous lesions as grade III lesions. Unfortunately, these modified systems, like that of Jaffe et al, are of little value in predicting patient outcomes. TREATMENTSection 6 of 11
Medical therapyPulmonary metastases have been cited as the cause of death in 16-25% of reported cases (Kay, 1994; Maloney, 1989; Rock, 1984). The need for early detection and treatment of these metastases has been emphasized. Pulmonary metastases have been treated with wide resection, chemotherapy, radiation therapy, and interferon alpha. When possible, wide surgical resection is the treatment of choice (Goldenberg, 1970; Kay, 1994; Bertoni, 1988; Maloney, 1989; Mirra, 1982). When the pulmonary metastases cannot be completely surgically excised, adjuvant treatment, such as chemotherapy or radiation therapy, has been advocated. In addition, in situations when the metastases are unresectable, both chemotherapy and radiation have been used as solitary agents (Maloney, 1989; Kutchemeshgi, 1974; Ladanyi, 1989; Stargardter, 1971; Stewart, 1995; Vanel, 1983). At University of Texas MD Anderson Cancer Center, interferon has been used with promising results (Benjamin, 1999). Spontaneous malignant transformation of GCT is not uncommon. Malignant transformation has been defined as a sarcoma associated with a benign typical GCT at presentation or as a sarcoma arising at the site of a preexisting GCT (Unni, 1996; Anract, 1998 #80). Malignant transformations have resulted in osteosarcoma, fibrosarcoma, or malignant histiocytoma (Benjamin, 1999; Gitelis, 1989; Mori, 2000; Ortiz-Cruz, 1995). Periods of 4-40 years for malignant transformation have been reported (Rock, 1984; Gitelis, 1989; Mori, 2000; Hefti, 1992). Many authors have reported a strong association between radiation therapy and malignant transformation of the GCT (Unni, 1996; Campanacci, 1987; Goldenberg, 1970; Dahlin, 1970; Mori, 2000; Hefti, 1992; Rock, 1986). However, much of this information was derived during the era of orthovoltage radiation. Recent studies have examined the effect of megavoltage radiation and have shown it to be well tolerated and not associated with malignant transformation (Bennett, 1993; Chen, 1986; Malone, 1995; Nair, 1999; Schwartz, 1989). GCTs that have undergone malignant transformation are treated as sarcomas. Radiation therapy and embolization are generally reserved for lesions or for patients in whom surgical treatment is not feasible. Radiation therapy has been proposed for patients who are not surgical candidates, for those whose tumors are in locations not amenable to operative treatment, or for those in whom a potential for significant morbidity from tumor relapse or subsequent surgery exists (Malone, 1995; Schwartz, 1989). Although megavoltage radiation now is used recommendations regarding radiation dose and fractionation schedules vary in the literature. Dose recommendations range from 35-70 Gy (Bennett, 1993; Chen, 1986; Malone, 1995; Nair, 1999; Harwood, 1977). Recurrence rates in these series ranged from 10-15%, and malignant transformation was uncommon. However, long-term follow-up still is warranted. Surgical therapyIn the past, GCTs were treated with amputation or with wide resection and reconstruction. However, with the knowledge that GCT is a locally aggressive yet benign disease, the surgical treatment of GCTs is intralesional for most locations. Various treatment options are advocated in the current literature, including the following: (1) curettage, (2) curettage and bone grafting, (3) curettage and insertion of polymethylmethacrylate (PMMA), (4) cryotherapy after curettage of the cavity, (5) curettage and a chemical adjuvant (phenol, zinc chloride alcohol, and H2O2) prior to the insertion of PMMA or a bone graft, (6) primary resection, (7) radiation therapy, and (8) embolization of the feeding vessels (Sung, 1982; Campanacci, 1987; Goldenberg, 1970; Dahlin, 1985; Dahlin, 1970; Aboulafia, 1994; Carrasco, 1989; Carrasco, 1989; Clohisy, 1994; Eckardt, 1986; Gitelis, 1993; Kocher, 1998; Larsson, 1975; Marcove, 1982; Marcove, 1978; McCarthy, 1980; McDonald, 1986; Muscolo, 1993; Miller, 1990; O'Donnell, 1994; Persson, 1976; Waldram, 1990). Resection Although intralesional procedures remain the treatment of choice for most GCTs, wide en bloc resection offers the lowest recurrence rate and is recommended for lesions in certain locations. In the proximal fibula, wide resection without reconstruction is often performed. Similarly, GCTs of the distal radius often are resected and reconstructed with autograft or allograft (see Image 15). However, in the long bones, resection necessitates prosthetic or allograft reconstruction (Clohisy, 1994; Kocher, 1998; Mankin, 1982; Mankin, 1976). In grade III lesions, this may be the best surgical option, but resection and reconstruction is associated with considerable surgical and functional morbidity. Intralesional procedures Intralesional curettage and bone grafting is a limb-sparing option that is associated with good functional outcomes in most cases. However, simple curettage with or without bone graft has recurrence rates of 27-55% (McDonald, 1986; Goldenberg, 1970; Dahlin, 1970; Kreieberg, 1985; Eckardt, 1986; Gitelis, 1993; Waldram, 1990). The high risk of recurrence led several surgeons to replace bone graft packing of the lesion with PMMA packing (see Image 16) (Vidal, 1969). The heat given off by the hardening PMMA is thought to lead to thermal necrosis of the remaining tumor cells in the curetted cavity (Leeson, 1993; Mjoberg, 1984). The PMMA technique, compared with bone grafting, offers the advantages of lack of donor-site morbidity, an unlimited supply, immediate structural stability, low cost, and ease of use. In addition, the barium contained in the methylmethacrylate results in a radiopaque substance that sharply contrasts with the surrounding bone. Local recurrences are more readily apparent than in cases in which bone graft is used (see Image 17). The disadvantages of using cement include difficulty in removing it when revision is needed and the possibility that subchondral cement may predispose the joint to early degenerative osteoarthritis (Persson, 1976; Campanacci, 1990). The latter is a theory that remains to be proven (Wilkins, 1987; Willert, 1987; Quint, 1998; Quint, 1996). In fact, using a canine model, Frassica et al showed that subchondral PMMA did not cause joint degeneration. However, in a later study, Frassica and colleagues showed that subchondral bone grafts are superior to cement for restoration of the normal subchondral anatomy (Frassica, 1990). Several authors have added the technique of high-speed burring of the cavity after simple intralesional curettage. A large cortical window is necessary to expose the entire tumor and tumor cavity, allowing thorough curettage and burring of the cavity (see Image 18). This has been found to reduce the recurrence rates to 12-25% (Leeson, 1993; Blackley, 1999). The high-speed burr not only adds a thermal component to eradication of the tumor but also allows more thorough removal of the tumor. High-speed burring of the cavity then may be followed by a chemical or physical adjuvant and packing of the lesion with PMMA or a bone graft. Adjuvant therapies Adjuvant therapies, such as phenol, H2O2, or liquid nitrogen use and argon beam coagulation, all have advantages and disadvantages. However, they all offer a method for eradication of microscopic disease. Many authors suggest that phenol is an effective means of decreasing the recurrence rate of giant cell tumors (Gitelis, 1993; O'Donnell, 1994; Quint, 1998; Capanna, 1985; Durr, 1999; Rock, 1990). After curettage is performed and all perforations in the bone are sealed, phenol is poured into the cavity. This results in a cellular death at a depth of approximately 1-2 mm (Quint, 1998). The use of 5% phenol has been advocated. Recurrence rates with curettage and phenol and packing with PMMA or bone grafts are 5-17% (Gitelis, 1993; O'Donnell, 1994; Quint, 1998; Capanna, 1985; Durr, 1999; Rock, 1990). Phenol is systemically toxic. Preventing exposure to the surrounding tissues while at the same time allowing exposure to the entire curetted cavity is difficult. It can cause a serious chemical burn, and it is also readily absorbed through the skin and mucosa. The material has a hazardous effect on the nervous system, heart, kidneys, and liver. It damages the DNA, coagulates protein, and causes cellular necrosis (Quint, 1998; Quint, 1996; Blackley, 1999). Several authors have raised the concern of the rapid absorption of the phenol through cancellous bones (Quint, 1998; Blackley, 1999). Many authors advocate cryosurgery as an adjuvant (Aboulafia, 1994; Marcove, 1982; Marcove, 1978; Malawer, 1999; Malawer, 1991; Pogrel, 1995). Liquid nitrogen is a chemical reagent used in cryosurgery. In the direct-pour technique, after the curettage is performed and after all perforations in the bone are sealed, liquid nitrogen is poured through a stainless steel funnel into the cavity (Marcove, 1978) (see Image 19). The liquid nitrogen is left in the cavity until it all evaporates. The surrounding tissues are irrigated with warm sodium chloride solution in an attempt to prevent or minimize thermal injury to the surrounding tissues. The process is repeated 2-3 times, resulting in cellular death at a depth of approximately 1-2 cm (Grogan, 1984). The cavitary defect is then reconstructed with PMMA or bone grafts. Recurrence rates with cryosurgery have been reported to be 2-12% (Aboulafia, 1994; Marcove, 1982; Marcove, 1978; Malawer, 1999; Malawer, 1991; Pogrel, 1995). The disadvantages of cryosurgery include the need for wide exposure, the need to protect the soft tissues, skin necrosis, osteonecrosis, and fracture (Marcove, 1982; Blackley, 1999; Malawer, 1999). Fracture is the most commonly reported and gravest complication (Aboulafia, 1994; Marcove, 1982; Malawer, 1999; Malawer, 1991). Malawer et al noted that internal fixation with Steinmann pins and reconstruction of the cavitary defect with PMMA significantly reduced the incidence of fracture and suggested that all patients who undergo cryosurgery receive internal stabilization as well (Malawer, 1999) (see Image 20). Some authors, as an alternative to cryosurgery and phenol therapy, have advocated argon-beam coagulation. It lacks the application hazards identified with both phenol and liquid nitrogen. Thermal coagulation applied through a concentrated argon gas is used to paint the tumor cavity (see Image 21). The penetration is approximately 2-3 mm. Recurrence rates for this procedure when paired with PMMA have been reported at 7%. No acute complications were noted. Long-term follow-up is warranted to assess the effect of argon beam coagulation on joint and/or subchondral physiology and on the incidence of pathologic fracture. Summary A review of the literature reveals that adjuvant treatment, when paired with intralesional curettage, offers excellent recurrence-free survival. Successful treatment of GCTs depends more on the thoroughness of intralesional curettage than on the specific adjuvant employed. The adequacy of tumor removal is influenced by tumor location, associated fracture, soft-tissue extension, and an understanding of the functional consequences of resection. The specific adjuvant treatment used appears to be at the surgeon's discretion; each option has advantages and disadvantages. Follow-upAfter treatment, patients with GCT should be monitored with serial physical examinations and radiography of the involved site and of the chest. Relapses may be associated with new pain or swelling. Tumor recurrences have been noted many years after initial treatment, and long-term observation of at least 5 years is recommended. In summary, GCTs of bone are benign but locally aggressive primary bone tumors. Local control is most closely related to complete tumor removal. However, the functional consequences and good long-term results often dictate intralesional (curettage) procedures. COMPLICATIONSSection 7 of 11
Complications are discussed for each treatment modality in the Treatment section, above. OUTCOME AND PROGNOSISSection 8 of 11
The overall prognosis generally is good. However, pulmonary metastases have been cited as the cause of death in 16-25% of reported cases (Kay, 1994; Maloney, 1989; Rock, 1984). FUTURE AND CONTROVERSIESSection 9 of 11
See Treatment, above. MULTIMEDIASection 10 of 11
REFERENCESSection 11 of 11
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
