A chondroblastoma is a rare, usually benign, tumor of bone that accounts for approximately 1% of all bone tumors. In 1931, Codman classified it as a chondromatous variant of giant cell tumors, when he described these lesions in the proximal humerus.[1] A decade later, Jaffe and Lichtenstein renamed the Codman tumor a benign chondroblastoma to emphasize the chondroblastic genesis of the lesion and to distinguish it from the classic giant cell tumor of bone.[2]
Although the exact etiology of chondroblastoma remains uncertain, the presentation, appropriate evaluation, and treatment of patients with the condition have been well described. (See Presentation, Workup, and Treatment.)
Various theories have been proposed concerning the pathogenesis of chondroblastomas. Mii et al described the results of ultrastructural examination of chondroblastomas,[3] demonstrating subcellular, calcium-containing precipitates similar to those seen in chondrocytes. On the basis of these findings, the authors concluded that the tumors are of chondrogenic origin. Aigner et al, however, noted the presence of osteoid matrix–containing type I collagen and the absence of true cartilage matrix production.[4] They considered the term chondroblastoma to be a misnomer and believed that the tumor should be reclassified as a bone-forming neoplasm.
Brien et al compared the characteristics of chondroblastoma of bone to chondroblastoma of soft tissue, giant cell tumor of the tendon sheath (GCTTS), and pigmented villonodular synovitis (PVNS).[5] On examination of about 15 examples of GCTTS and PVNS, large areas of chondroid differentiation were noted that could not be distinguished from chondroblastoma of bone by either histologic or electron microscopic features. The researchers theorized that chondroblastoma of bone stems from an intraosseous proliferation of tendon sheath cells that have a predilection for chondroid formation.
Chondroblastomas typically occur in the epiphyses of tubular long bones. The distal femoral and proximal tibial epiphyses are most frequently involved, followed by the proximal humerus, where approximately 18% of chondroblastomas appear.[6]
No risk factors are known for chondroblastoma. There have been reports of abnormalities in chromosomes 5 and 8, as well as of p53 mutations, in patients with chondroblastoma.[7] Sjögren et al performed cytogenetic analysis of benign and malignant cartilage tumors, and while they observed no consistent karyotypic abnormalities, there were recurrent breakpoints seen at 2q35, 3q21-23, and 18q21.[8]
In the United States, chondroblastoma accounts for approximately 1% of all bone tumors. The international incidence is not reported in the current literature.
Approximately 92% of patients presenting with chondroblastoma are younger than 30 years. However, chondroblastomas have been reported to arise in patients as young as 2 years and as old as 83 years. In several large series, most patients were diagnosed in the second decade of life. The male-to-female ratio is 2:1 in most series. No racial predilection is recognized.
Patients with benign chondroblastoma may limit activities due to pain. Malignant chondroblastomas, which may occur many years after the original lesion (even in the absence of radiation), are extremely rare and are associated with a dismal prognosis.
Local recurrence in long-bone lesions is approximately 10% and is higher for chondroblastomas arising in flat bones, especially those lesions arising in the vicinity of the triradiate cartilage. Average time to recurrence is 34 months after initial treatment. Most authors have not reported any significant difference in recurrence rates for tumors, regardless of the age or sex of the patient, size of the lesion, amount of calcification or vascular invasion seen on histologic examination, duration of follow-up, or method of treatment.
Springfield attributed a higher recurrence rate in patients with open physeal plates to a less aggressive curettage performed in an effort to avoid future growth arrest.[9] Recurrences may be treated with repeat curettage, with or without bone graft or cementation, and with marginal excision of any soft-tissue component.[10]
Whereas most chondroblastomas are small, well-marginated lesions that are successfully treated with intralesional curettage, a small subset of chondroblastomas behave in a much more aggressive fashion. Some of these tumors retain their benign microscopic features but nonetheless become very large or have the capability of metastasizing to the lungs and soft tissues.
Metastases may be synchronous or metachronous, occurring concurrently with the primary bone tumor or up to 33 years later. Metastases can occur even without surgical manipulation or local recurrence of the primary tumor. These more aggressive lesions may be treated with en-bloc resection and reconstruction where intralesional curettage would leave a large, bony defect. Pulmonary implants or soft-tissue metastases should be resected, especially if they are progressive.
Another rare subset of chondroblastomas may become frankly malignant even though no prior radiation therapy was used. Kyriakos et al used the term malignant chondroblastoma to describe tumors that continue to grow or disseminate, not just those that metastasize.[11] Malignant transformation typically occurs many (usually >10) years after treatment of the initial benign lesion. Pulmonary metastases may develop along with the malignant bony lesion.
Microscopic examination of the malignant bone lesion shows features similar to the original lesion (along with other areas with nuclear pleomorphism), abundant and abnormal mitotic figures, tumor necrosis, and intravascular thrombi. Ostrowski et al reported a patient with malignant transformation of a recurrent pelvic chondroblastoma with a p53 mutation.[12] Frankly malignant chondroblastoma tends to be resistant to surgery, radiation, and chemotherapy, and patients with these tumors have had dismal prognoses.
A retrospective study by Farfalli et al focused primarily on long-term joint status and functional outcomes (rather than oncologic outcomes) after curettage for epiphyseal chondroblastoma.[13] The investigators found that aggressive curettage of epiphyseal chondroblastoma frequently led to osteoarthritis and that tumors in the proximal femur appeared particularly likely to be associated with secondary osteoarthritis and prosthetic replacement.
Pain is the most common presenting symptom. It typically is mild and gradually progressive and initially may be attributed to a minor injury. If the lesion is juxta-articular, the patient may complain of joint swelling or diminished range of motion. Usually, constitutional symptoms are lacking. In their series of 70 patients, Turcotte et al found the average duration of symptoms in patients with chondroblastoma to be 20 months.[14]
The physical examination is remarkable for localized tenderness in most patients. Soft-tissue swelling, mass, or joint effusion is present in about 20% of cases. Muscular atrophy or decreased joint motion is less common.
Adequate regional radiographs should be obtained in patients with suspected chondroblastoma. (See the images below.)
View Image
Radiograph of epiphyseal lesion (hip).
View Image
Radiograph demonstrating tumor on both sides of physis (humerus).
A bone scan (see the image below) may be helpful in assessing a lesion's biologic activity; it can also aid in locating other regions of disease involvement that may result from entities other than a chondroblastoma or from the rare occurrence of multifocal chondroblastomas.
View Image
Bone scan.
Magnetic resonance imaging (MRI; see the image below) should be conducted if any question exists about the diagnosis or lesion boundaries, including perilesional soft-tissue extension or edema.
View Image
Magnetic resonance image of a hip showing lobular pattern of chondroblastoma.
Computed tomography (CT) may be helpful in defining the extent of the lesion within the bone, especially if the bone is in a more anatomically complex area, such as the hand, foot, or spine.
Benign pulmonary metastases occasionally may occur with chondroblastoma; therefore, a chest radiograph should be obtained. A CT scan of the chest may be used to further evaluate any suspicious areas. These pulmonary lesions are resectable and curable.
If malignancy is not suspected on the basis of clinical and radiographic features, tissue may be obtained for diagnosis at the time of surgical treatment of the lesion, and no preceding biopsy is needed. However, histologic confirmation by frozen section should be sought before definitive treatment is initiated. If the lesion appears atypical on appropriate imaging studies, a needle or incisional biopsy should be performed before definitive treatment is undertaken.
Chondroblastomas are lobulated tumors that consist of grayish-pink soft tissue intermixed with bluish chondroid tissue and calcifications. The lesions may have many hemorrhagic cystic areas.[16]
The tumors are composed of sheets of neoplastic mononuclear chondroblasts with eosinophilic cytoplasm and grooved nuclei (see the image below). Interspersed among the mononuclear cells are osteoclastlike giant cells. The chondroid matrix typically is pink; on rare occasions, the basophilic matrix seen in hyaline cartilage is present. Some chondroblastomas may have a spindle-cell component, which represents either spindle-shaped mononuclear cells or reparative cells of fibroblastic origin.
View Image
Histology of chondroblastoma.
One of the most characteristic findings in the histologic examination of chondroblastomas is linear deposition of calcification surrounding individual chondroblasts, creating a chicken-wire pattern (see the image below). Calcification may be so extensive that the chondrocytes in the area are not viable. In sections with well-preserved chondroblasts, mitoses may be seen, but atypical mitoses are not present in benign chondroblastoma.
Cystic changes within chondroblastomas are common. Some represent secondary aneurysmal bone cysts, which are found in 20-25% of all patients with chondroblastomas. Other cysts are filled with serous fluid and are divided into unilocular or multilocular spaces. When these other cysts are present, the tumors are termed cystic chondroblastomas. Initially, there was concern that cystic chondroblastomas had a much higher recurrence rate than typical chondroblastomas, but later reports did not confirm this theory.
Less frequently seen histologic findings are cellular atypia with enlargement and irregularity of chondroblast nuclei (occurring in 30% of cases); hemosiderin (in 25% of cases); surrounding cortical and soft-tissue permeation (in 5% of cases); myxoid areas (in 2% of cases); and vascular invasion (in 1% of cases).
Immunostaining occasionally can be helpful in confirming the diagnosis of chondroblastoma.[17] S-100 protein is strongly positive in the mononuclear cells, though it is absent in multinucleated giant cells and is present only focally in tumors with a large cystic component. Chondroblastomas are also positive for vimentin. Reticulin stain reveals a honeycomb pattern.
Percutaneous radiofrequency ablation (RFA) may be an alternative to surgery for the treatment of certain chondroblastomas,[18] but according to Rybak et al, larger lesions that are under weightbearing surfaces should be approached with caution because of an increased risk of articular collapse and recurrence.[19]
Radiation therapy has been employed in the treatment of chondroblastoma but has essentially no current role in its treatment.[20, 18] Chemotherapy has not been reported in the condition's treatment.
In search of a targeted therapy for patients with disseminated chondroblastoma or those in need of medical management, Yang et al reported positive effects of targeted inhibition of mTOR (mammalian target of rapamycin) and HIF (hypoxia-inducible factor) pathways in benchtop work with chondroblastoma. Combination treatment of low-dose rapamycin, FM19G11, and leucine deprivation were inhibitory on the chondroblastoma cell line examined.[21]
Pain medications should be administered as needed.
No evidence suggests that chondroblastoma resolves spontaneously; accordingly, surgical treatment is generally indicated. The most common surgical procedure used for chondroblastoma is curettage, with or without autograft or allograft bone grafting.[22, 23] Other options, used less frequently, include the following:
Substitution of polymethylmethacrylate or fat implantation for bone graft
Chemical cauterization (with phenol) of the curetted lesion
Liquid nitrogen cryotherapy
Marginal resection
Wide resection
Large or recurrent chondroblastomas should be managed by an orthopedic oncologist.
Surgical curettage vs radiofrequency ablation
Although open surgical curettage remains the most commonly used treatment, an emerging pool of literature continues to support the safety and efficacy of treatment with RFA.
Studies of longer-term follow-up after conventional curettage and grafting have generally reported excellent results with a low complication rate. At a mean 8-year follow-up of 24 patients treated by curettage and bone grafting, a recurrence rate of only 4% (1/24) was reported by Lehner et al, and 88% of patients achieved good or excellent results.[24] In a report of 14 patients using extended intralesional curettage with high-speed burring, intralesional cryotherapy, and autogenous bone grafting, Mashhour et al noted that the recurrence rate was low (1/14) after mean 4-year follow-up, but two patients experienced growth arrest.[25]
For balance, however, in a large series of 87 purely pediatric chondroblastoma cases, the recurrence rate after curettage and grafting was 32%.[26] Risk factors for recurrence included epiphyseal location (contrasted with metaphyseal, apophyseal, and combined metaphyseal-epiphyseal), proximal femoral lesions, and tarsal lesions. In 63% of the patients, treatment consisted of intralesional curettage with autogenous bone grafting; functional outcome was good for 68.5% of the patients; and 32% of the lesions recurred.[6]
The difficulty of treating femoral head lesions led two authors to publish papers examining that site in particular. In a series of 10 patients with femoral head lesions, a direct approach to the lesion through the femoral neck was favored over curettage through a drill hole within the femoral neck.[27] A trap-door technique was also reported as a successful salvage technique in that series. For a large femoral head defect, a vascularized fibular graft was successfully used to reconstruct the defect created by open surgical treatment.[28]
RFA for chondroblastoma has been described in several reports, but follow-up has been shorter than for surgical treatment, and caution is recommended for cases in which the lesions are larger than 2.5 cm and when there is no subchondral bone support. Nevertheless, proponents favor RFA for smaller lesions with intact subchondral bone and difficult-to-access lesions.[29] However, many of the reports have fewer than 10 patients, follow-up remains less than that of conventional curettage and grafting, and complications in up to 20% have been reported.[30, 31] Complications of RFA for chondroblastoma have included subchondral fracture, chondrolysis, persistent pain, and need for repeat RFA.
In a study of RFA for chondroblastoma by Rybak et al,[19] 12 of 14 patients available for follow-up (median, 41 months) reported complete relief of symptoms without the need for medications, and all returned to previous activities. One patient, who had the largest lesion, required surgery because of articular collapse in the area of treatment; another required surgical treatment because of mechanical problems. The authors concluded that percutaneous RFA is an alternative to surgery for selected chondroblastomas but that larger lesions under weightbearing surfaces must be approached with caution because of an increased risk of articular collapse and recurrence.
Xie et al carried out a retrospective study of 25 consecutive patients treated with RFA over a period of approximately 7 years.[32] Patients were assessed after 1 month, then every 3-6 months, and then yearly for up to 3 years. Recovery was monitored with serial magnetic resonance imaging (MRI), and functional outcome was quantified with the Musculoskeletal Tumour Society Score (MSTS). The authors found RFA to an effective alternative to surgery in the management of chondroblastoma and suggested that it should be considered as a first-line treatment.
In addition to recurrence, many complications can occur after treatment of chondroblastomas, including the following:
Infection
Development of degenerative joint changes
Fracture through the lesion
Failure of osteoarticular allografts, if used
Premature physeal closure and subsequent limb-length discrepancy or angular deformity of the limb
Malignant transformation or development of a postradiation sarcoma as late as 18 years after diagnosis (in rare cases in which radiation therapy is used)
Unless the lesion is particularly large and creates a risk of pathologic fracture, patients may participate in activity as tolerated. If an en bloc excision is performed, the patient's activity may be limited to protect the reconstruction.
In view of the 10% risk of local recurrence, patients should be monitored for at least several years. Monitor patients with open physes at the time of treatment for premature physeal closure.
At follow-up, patients should be evaluated with a thorough history and physical examination and with appropriate radiographs.
Timothy A Damron, MD, David G Murray Endowed Professor, Department of Orthopedic Surgery, Professor, Orthopedic Oncology and Adult Reconstruction, Vice Chair, Department of Orthopedics, State University of New York Upstate Medical University at Syracuse
Disclosure: Received research grant from: National Institutes of Health NIAMS; Orthopaedic Research and Education Foundation; Stryker; Cempra; Wright Medical<br/>Received income in an amount equal to or greater than $250 from: Stryker, Inc (Educational travel to Stryker sponsored meetings)<br/>Received royalty from Lippincott, Williams, and Wilkins for editing/writing textbook; Received grant/research funds from Genentech for clinical research; Received grant/research funds from Orthovita for clinical research; Received grant/research funds from National Institutes of Health for clinical research; Received royalty from UpToDate for update preparation author; Received grant/research funds from Wright Medical, Inc. for clinical research.
Specialty Editors
Francisco Talavera, PharmD, PhD, Adjunct Assistant Professor, University of Nebraska Medical Center College of Pharmacy; Editor-in-Chief, Medscape Drug Reference
Disclosure: Received salary from Medscape for employment. for: Medscape.
Chief Editor
Omohodion (Odion) Binitie, MD, Medical Director, Assistant Member, Department of Sarcoma, Section Head, Orthopedics, Adolescent/Young Adult and Pediatric Orthopedic Oncology, Medical Director, Physical Therapy, Speech Therapy, and Rehabilitation Services, Assistant Fellowship Program Director, Musculoskeletal Oncology, Moffitt Cancer Center; Assistant Professor, Department of Oncologic Sciences, Department of Ortho and Sports Medicine, University of South Florida Morsani College of Medicine
Disclosure: Nothing to disclose.
Additional Contributors
Howard A Chansky, MD, Associate Professor, Department of Orthopedics and Sports Medicine, University of Washington Medical Center
Disclosure: Nothing to disclose.
Acknowledgements
Hannah D Morgan, MD Consulting Staff, Connecticut Orthopaedic Specialists
Hannah D Morgan, MD is a member of the following medical societies: American Academy of Orthopaedic Surgeons and American Medical Association
){kind=link}
){kind=link}
){kind=link}
){kind=link}
){kind=link}
){kind=link}