Giant Cell Tumor of Bone

  1. Overview and Epidemiology
  2. Pathology and Histology
  3. RANK/RANKL/OPG Signaling Pathway
  4. Clinical Presentation and Locations
  5. Imaging Features
  6. Campanacci Grading System
  7. Biopsy and Diagnosis
  8. Surgical Treatment
  9. Denosumab: Anti-RANKL Therapy
  10. Malignant GCT and Pulmonary Metastases
  11. Key Research Papers
  12. PubMed Topic Searches
  13. Connections
  14. Featured Videos

Overview and Epidemiology

Giant cell tumor of bone (GCT) is a locally aggressive primary bone tumor that occupies a unique borderline category in oncology: it is classified as benign yet carries meaningful malignant potential. Roughly 1–2% of cases undergo frank malignant transformation, and approximately 5–10% of patients develop distant metastases — most commonly to the lungs — even without histological atypia in the primary tumor. This paradox, a tumor that looks benign under the microscope but can spread to the chest, defines much of the clinical challenge of GCT.

GCT accounts for approximately 5% of all primary bone tumors worldwide, making it the most common benign bone tumor in adults after giant cell reparative granuloma and fibrous dysplasia in some series. Incidence figures from population-based registries cluster around 1–1.5 cases per million per year. The tumor shows a striking predilection for skeletally mature young adults, with peak incidence between ages 20 and 40. Before the growth plates (physes) fuse, GCT is distinctly rare — the epiphyseal location the tumor characteristically occupies is not accessible until skeletal maturity. Conversely, GCT becomes uncommon after age 55, distinguishing it epidemiologically from the metastatic carcinomas that dominate bone tumor practice in older patients.

A modest female predominance is consistently reported across epidemiological series, with a female-to-male ratio of approximately 1.5:1. The reasons for this sex predilection are not fully understood but may relate to hormonal influences on RANKL signaling, the same pathway that drives osteoclast-mediated bone resorption in postmenopausal osteoporosis.

An important, albeit uncommon, association exists between GCT and Paget's disease of bone. Secondary GCT arising in pagetic bone typically presents in older patients and carries a higher risk of malignant transformation than the primary sporadic form. This population should be distinguished from the typical young-adult presentation when counseling patients and planning treatment.

In China and other parts of East Asia, GCT is diagnosed at higher rates relative to other primary bone tumors than in Western countries — some series report GCT comprising up to 20% of all primary bone tumors in Chinese registries. Whether this reflects a true biological difference or referral and diagnostic patterns remains under study.

Back to Table of Contents

Pathology and Histology

The hallmark of GCT under the microscope is a strikingly uniform mixture of three distinct cell populations. Understanding which population is truly neoplastic — and which cells are reactive bystanders — transformed the field and directly led to the development of denosumab as targeted therapy.

The Three Cell Populations

1. Multinucleated osteoclast-like giant cells — These are the namesake cells that give the tumor its appearance. Each giant cell may contain 50 to 100 nuclei. Critically, those nuclei are morphologically identical to the nuclei of the surrounding mononuclear stromal cells. This identity is a key histological clue: it means the giant cells were not independently malignant but were assembled from mononuclear precursors under the influence of RANKL signaling. The giant cells are the agents of bone destruction, carrying out osteoclast-like resorptive function at the tumor margin.

2. Mononuclear stromal cells — These spindle-to-ovoid cells are the true neoplastic population of GCT. They drive tumor behavior, proliferate, express RANKL at very high levels, and harbor the characteristic H3F3A driver mutations seen in over 90% of GCTs. Their mesenchymal stem cell–like features (expression of CD68, CD44, vimentin; some osteoblastic differentiation capacity) explain why the tumor arises in metaphyseal-epiphyseal bone. Under the microscope, they form the background matrix of the tumor, within which giant cells are uniformly scattered.

3. Monocyte/macrophage precursor cells — A population of mononuclear cells with monocytic phenotype (CD14+, CD68+) is recruited into the tumor by stromal cell–derived chemokines. These cells respond to RANKL, fuse with one another under its direction, and mature into the multinucleated giant cells. They are reactive, not neoplastic — removing RANKL stimulus (as denosumab does) causes the giant cells to disappear.

Histological Distribution

In conventional GCT, giant cells are evenly distributed throughout the tumor without clustering. This even distribution is an important distinguishing feature from other giant cell–rich lesions (such as brown tumors, where giant cells can cluster, or chondroblastoma, where they tend to cluster around chondroid islands).

There is no osteoid production by tumor cells (distinguishing GCT from osteosarcoma), no cartilage matrix (excluding chondroblastoma), and no malignant nuclear cytology in conventional GCT. The stromal cell nuclei are uniform, vesicular, with inconspicuous nucleoli. Mitotic figures may be present but are not atypical.

H3.3 Mutations

A landmark discovery published in 2013 by Behjati and colleagues identified somatic mutations in the H3F3A gene (encoding histone H3.3) in the neoplastic stromal cells of over 90% of GCTs. The specific substitution, most commonly G34W or G34L (glycine to tryptophan or leucine at position 34), is highly specific to GCT and is not found in other giant cell–rich tumors. This mutation alters epigenetic regulation of gene expression in the stromal cells and is now used diagnostically — particularly to distinguish GCT from other giant cell–containing lesions that can look histologically similar. H3F3A mutation testing by immunohistochemistry (anti-H3.3 G34W antibody) has high sensitivity and specificity and is increasingly incorporated into standard pathological workup.

A secondary aneurysmal bone cyst (ABC) component is identified in approximately 14% of GCTs. When present, blood-filled spaces and fibrous septa with osteoclast-type giant cells overlie the conventional GCT architecture. This combination can make diagnosis on limited biopsy material challenging.

Back to Table of Contents

RANK/RANKL/OPG Signaling Pathway

The biology of GCT is inseparable from the RANK/RANKL/OPG axis — the master regulatory pathway of osteoclast biology that was hijacked by tumor stromal cells to drive bone destruction. Understanding this pathway explains the tumor's behavior, its radiographic appearance, and the mechanism of its most important medical therapy.

Normal Physiology

RANKL (receptor activator of NF-κB ligand, also called TNFSF11) is a membrane-bound or soluble cytokine produced primarily by osteoblasts and stromal cells in the bone marrow. It binds to its receptor, RANK, expressed on osteoclast precursors. RANK-RANKL binding activates NF-κB and downstream signaling cascades that drive the differentiation of mononuclear osteoclast precursors into mature, multinucleated, bone-resorbing osteoclasts. This is the physiological mechanism underlying normal bone remodeling.

OPG (osteoprotegerin, TNFRSF11B) is a soluble decoy receptor produced by osteoblasts that competes with RANK for RANKL binding. OPG neutralizes RANKL before it can reach osteoclast precursors, acting as a brake on bone resorption. The ratio of RANKL to OPG in the bone microenvironment determines the net rate of osteoclast activity and, therefore, bone resorption versus formation.

Hijacked in GCT

In GCT, the neoplastic stromal cells express RANKL at levels far exceeding those seen in normal bone. Concurrently, OPG expression is relatively low, tipping the RANKL:OPG ratio dramatically toward unchecked osteoclastogenesis. The result is a tumor that continuously recruits circulating monocyte precursors, fuses them into osteoclast-like giant cells, and directs relentless bone resorption — producing the expanding lytic cavity seen on imaging.

This mechanism explains several clinical features of GCT. The tumor grows by eating bone, not by producing matrix. It tends to thin the cortex rather than penetrate it until advanced (grades II–III). The subchondral bone is attacked last, as the articular cartilage provides some protection, but eventually cortical breakthrough occurs in aggressive lesions.

Denosumab: Therapeutic Interruption of RANKL

Denosumab is a fully human IgG2 monoclonal antibody that binds RANKL with very high affinity and specificity, neutralizing it before it can engage RANK on osteoclast precursors. With RANKL suppressed, the recruitment and fusion of osteoclast precursors ceases. The existing multinucleated giant cells are no longer sustained — they undergo apoptosis or lose resorptive function. Within weeks to months of starting denosumab, the giant cell population in the tumor dramatically diminishes or disappears on repeat biopsy, and new bone shell formation (sclerosis) occurs at the tumor margins, visible on imaging.

Importantly, denosumab acts on the reactive giant cells, not on the underlying neoplastic stromal cells. The stromal cells persist, their H3F3A mutation is unchanged, and the RANKL they produce remains ready to recruit new giant cells if denosumab is discontinued. This explains the clinical phenomenon of rebound — tumor regrowth and osteolysis after stopping denosumab — and underscores that denosumab is a bridge to surgery, not a cure.

Back to Table of Contents

Clinical Presentation and Locations

Typical Symptoms

Patients with GCT most commonly present with gradually progressive joint pain. The onset is insidious — many patients report months to years of aching discomfort before seeking evaluation. The pain is typically activity-related early on, becoming more persistent as the tumor expands. Local swelling and decreased range of motion of the adjacent joint are common accompanying findings, reflecting the epiphyseal location near the joint space. Constitutional symptoms (fever, night sweats, weight loss) are absent in conventional GCT and, when present, should raise suspicion for malignancy or an alternative diagnosis.

Pathological fracture occurs in 5–12% of patients at presentation. This rate is higher for grade III lesions, where cortical destruction has rendered the bone structurally incompetent. Fracture significantly complicates surgical planning and may necessitate prior denosumab therapy to consolidate the tumor before definitive resection.

Laboratory tests are typically normal in GCT. Serum alkaline phosphatase (ALP) and lactate dehydrogenase (LDH) — which may be elevated in osteosarcoma and Ewing sarcoma, respectively — are usually within normal limits. Calcium and PTH should be checked to exclude hyperparathyroidism with brown tumors (which are histologically indistinguishable from GCT). Occasional incidental discovery of GCT on imaging obtained for an unrelated indication occurs, particularly for smaller grade I lesions.

Anatomical Locations

The epiphysis of long bones after skeletal maturity is the defining anatomical home of GCT. The tumor almost invariably arises in or involves the epiphysis, placing it within a few millimeters of the articular cartilage. Extension across the open or fused physis into the metaphysis occurs in approximately 70% of cases — a fact that distinguishes GCT from chondroblastoma, which remains epiphyseal.

The knee region dominates the distribution:

Additional locations include the proximal fibula, proximal humerus, sacrum, pelvis, and spine (vertebral body). Sacral and pelvic GCTs present unique surgical challenges due to anatomical complexity and proximity to neurovascular structures. Spinal GCT most commonly involves the vertebral body and may cause neurological compromise. Flat bones (skull, sternum, clavicle) are rarely affected.

GCT is characteristically eccentric within the epiphysis — it arises off-center within the bone cross-section, tending toward one cortex. The tumor extends to touch the subchondral bone (within a few millimeters of articular cartilage) in the great majority of cases, making joint-preserving surgery a constant consideration.

Back to Table of Contents

Imaging Features

The imaging evaluation of GCT follows a standard multimodality approach, with each modality contributing unique information to surgical planning and staging.

Plain Radiography

The plain X-ray is the starting point. Classic GCT appears as an eccentric lytic lesion in the epiphysis extending to the subchondral bone. Key radiographic features include:

Computed Tomography (CT)

CT excels at delineating cortical integrity — detecting subtle cortical breakthrough not visible on plain films. It identifies internal calcification (if any), maps the tumor's relationship to adjacent structures, and aids surgical planning. CT is essential for detecting pulmonary metastases: chest CT should be performed at baseline for all GCTs, given the ~1–4% risk of lung involvement.

Magnetic Resonance Imaging (MRI)

MRI is the definitive modality for soft tissue characterization. It provides superior delineation of:

On MRI, GCT is typically intermediate-to-low signal on T1 (reflecting cellular solid tissue) and heterogeneous on T2, often with areas of low T2 signal corresponding to hemosiderin deposition from prior microhemorrhage. Avid enhancement with gadolinium reflects the tumor's vascularity.

Nuclear Medicine and PET

Bone scan (Tc-99m) shows increased uptake in GCT but adds little specificity. PET-CT is not routinely used for benign GCT but reports of high FDG avidity exist — this can cause diagnostic confusion with high-grade sarcoma. PET-CT may be used in the metastatic workup of aggressive or recurrent GCT. After denosumab therapy, decreased FDG uptake and increasing sclerosis on CT indicate treatment response.

Back to Table of Contents

Campanacci Grading System

The Campanacci grading system, originally described by the Italian orthopaedic oncologist Marcello Campanacci, stratifies GCT by radiographic aggressiveness and correlates with recurrence risk and surgical decision-making. It remains the most widely used grading system for GCT worldwide.

Grade I — Latent (Inactive)

Radiographic appearance: Well-defined, sharply marginated lytic lesion with a thick cortical shell that is intact and shows no expansion. The margin is clearly defined with a smooth interface between tumor and surrounding bone.

Clinical behavior: Slow growth, minimal risk of pathological fracture. May be discovered incidentally. Lowest recurrence risk after surgery.

Grade II — Active

Radiographic appearance: Moderately defined margin with a thin cortical shell showing slight expansion. The border remains definable but is less sharp. No periosteal reaction. This is the most common presentation at diagnosis.

Clinical behavior: Moderate growth, symptomatic. Majority of GCTs are grade II. Recurrence rates after extended curettage run 15–25%.

Grade III — Aggressive

Radiographic appearance: Ill-defined margins with cortical destruction (full cortical breakthrough) and extraosseous soft tissue extension. Periosteal reaction may be present, reflecting reactive bone formation at the advancing tumor front.

Clinical behavior: Rapid growth, high fracture risk. More likely to be associated with pulmonary metastases. Strongest indication for preoperative denosumab to downsize the tumor and allow limb-salvage resection. Recurrence rates after curettage alone approach 40–50%, justifying consideration of wide resection at certain anatomical sites.

Clinical Application

Grading directs surgical planning: grades I–II favor intralesional curettage with adjuvant treatment; grade III raises the threshold for considering en bloc wide resection, especially at expendable sites (fibula, coccyx, sacral ala) or after denosumab-assisted downsizing at critical sites (sacrum, spine). Grade III at the distal radius is frequently managed by wide resection and reconstruction given the anatomically constrained space and inability to perform adequate curettage. The Campanacci grade also informs the indication for and duration of denosumab therapy.

Back to Table of Contents

Biopsy and Diagnosis

Tissue sampling is required before definitive treatment in all cases. The goals of biopsy are to confirm the diagnosis of GCT, exclude malignant mimics, and provide material for ancillary molecular testing.

Biopsy Technique

Core needle biopsy is the preferred technique at most centers. It provides adequate tissue for histology, immunohistochemistry, and molecular testing (including H3F3A mutation) while minimizing contamination of the surgical field. Image guidance (CT or fluoroscopy) is used to target the solid, non-necrotic portions of the tumor and avoid the secondary ABC component (blood-filled spaces yield only blood and fibrin, not diagnostic tissue). The biopsy tract must be positioned so it can be excised en bloc with the specimen if the case later requires wide resection — tract contamination is a real oncological risk.

Differential Diagnosis: Giant Cell–Rich Lesions

Multiple bone lesions contain multinucleated giant cells, making histological diagnosis alone insufficient. The differential requires clinical, radiographic, and molecular correlation:

Ancillary Testing

H3F3A immunohistochemistry (anti-H3.3 G34W antibody) has approximately 88% sensitivity and over 95% specificity for GCT among giant cell–rich tumors. It is particularly valuable when biopsy material is limited or when the histological picture is ambiguous. Negative staining does not exclude GCT (a minority of GCTs have G34L or other variants not detected by the G34W antibody) but should trigger search for alternative diagnoses. Molecular sequencing of the H3F3A locus offers the highest sensitivity for cases where immunohistochemistry is equivocal.

Back to Table of Contents

Surgical Treatment

Surgery remains the cornerstone of GCT treatment for the majority of patients. The operative strategy is determined by Campanacci grade, location, tumor size, and the availability and extent of denosumab pretreatment.

Intralesional Curettage with Adjuvant Treatment

Extended intralesional curettage is the standard of care for grades I and II GCT at anatomically accessible sites. The surgeon enters the tumor cavity, removes all grossly visible tumor with curettes, and then applies adjuvant physical and chemical measures to treat the residual microscopic disease on the cavity walls:

After curettage and adjuvant treatment, the cavity must be filled to restore mechanical stability:

Recurrence rates with extended curettage plus adjuvants have been reduced to 15–25% in modern series, down from 40–50% with simple curettage alone. Recurrences are typically detected within 2–3 years of surgery on routine follow-up imaging and are usually amenable to repeat curettage, though each recurrence carries increased risk for the next.

Wide En Bloc Resection

Wide resection — removing the tumor with a continuous cuff of surrounding normal tissue — is reserved for specific clinical scenarios:

Reconstruction after wide resection depends on the site. Distal radius reconstruction options include vascularized or non-vascularized fibular autograft (with or without wrist arthrodesis) or prosthetic replacement. Sacral and pelvic GCTs may require combined anterior-posterior approaches, selective arterial embolization to reduce intraoperative blood loss, and complex reconstruction with pelvic rings or cage implants.

Embolization

Selective arterial embolization is used as an adjunct or occasionally as primary palliation for sacral and pelvic GCTs where surgical access is limited. Serial embolization sessions (every 6–8 weeks) can achieve significant tumor devascularization and shrinkage, and may be combined with denosumab for maximal preoperative tumor reduction. Embolization alone rarely achieves cure but can temporize symptoms and buy time for other interventions.

Back to Table of Contents

Denosumab: Anti-RANKL Therapy

Denosumab (Xgeva, Prolia) transformed the management of GCT from a purely surgical problem into one where biology-directed therapy could change what surgery was possible — or whether surgery was needed at all. Its approval by the FDA in 2013 for GCT unresponsive to surgery or where surgery is likely to result in severe morbidity was based on compelling phase 2 data demonstrating dramatic histological and radiographic responses.

Dosing Regimen

The approved dosing for GCT is 120 mg subcutaneous every 4 weeks, with additional loading doses on days 8 and 15 of the first month to achieve rapid steady-state RANKL suppression. This loading strategy exploits the pharmacokinetics of the antibody to eliminate the initial lag before full osteoclast suppression. Subcutaneous injection into the thigh, abdomen, or upper arm. No dose adjustment required for mild-to-moderate renal impairment, unlike bisphosphonates (an advantage in younger patients).

Clinical Response

In the pivotal Thomas et al. phase 2 trial, 35 of 37 patients (96%) with unresectable or metastatic GCT showed tumor response — defined as no new lesions, no progression of existing target lesions, and at least one of: elimination of giant cells on histology, new bone formation on CT, or no progression on plain radiographs by 25+ weeks. Response typically begins within 4–8 weeks of the first dose, with significant radiographic changes (rim sclerosis, trabecular consolidation) evident by 3 months.

Histologically, the tumor after denosumab shows dramatic reduction or complete absence of osteoclast-like giant cells, replaced by a background of fibroblast-like stromal cells laid down in a more organized collagen matrix. New woven and lamellar bone is deposited at the periphery. The appearance can mimic a low-grade osteosarcoma (fibrous matrix, new bone, few giant cells) — a phenomenon that has led to diagnostic errors on re-biopsy after denosumab, and pathologists must be informed when reviewing post-denosumab specimens.

Indications and Surgical Timing

Denosumab is indicated in GCT for:

The optimal duration of preoperative denosumab is typically 3–6 months — long enough to achieve maximum tumor shrinkage and new bone shell formation, but with surgery performed while still on denosumab (or shortly after the last dose) to minimize rebound osteolysis. Discontinuation of denosumab without definitive surgery almost invariably results in tumor regrowth within months, with potential for more aggressive behavior in some cases. Several reports document rapid tumor progression after abrupt denosumab cessation, including cases of apparent malignant transformation of stromal cells deprived of their giant cell "chaperones."

Adverse Effects and Monitoring

Bisphosphonates as Adjuvant Therapy

Zoledronic acid (zoledronate) has been studied as an adjuvant to surgery for GCT based on its ability to inhibit osteoclast function through a different mechanism (inhibition of farnesyl pyrophosphate synthase in the mevalonate pathway). Small series and retrospective studies suggest reduction in local recurrence rates when zoledronic acid is given perioperatively with curettage. It has not displaced denosumab as the preferred biological agent but represents a lower-cost option in resource-limited settings.

Back to Table of Contents

Malignant GCT and Pulmonary Metastases

Pulmonary Metastases

Pulmonary metastases from GCT occupy a biologically unique position in oncology — they are called "benign pulmonary metastases" because they are histologically identical to the primary tumor, containing the same three-cell population with conventional GCT morphology and no cytological malignancy. Yet they demonstrably represent hematogenous dissemination and can grow, though many grow very slowly over years and some undergo spontaneous regression.

The reported incidence in large series is 1–4% of all GCTs. Risk factors for pulmonary metastasis include:

Management of pulmonary metastases:

Baseline chest CT at diagnosis, with follow-up CT at 6-month intervals for the first 2–3 years and annually thereafter, is standard surveillance practice.

Malignant Transformation

Approximately 1–2% of GCTs undergo frank malignant transformation, resulting in a high-grade sarcoma with conventional GCT elements. Two forms are recognized:

Primary malignant GCT (de novo) is exceedingly rare — it presents as a tumor with areas of conventional GCT morphology co-existing with frank sarcomatous regions (osteosarcoma, fibrosarcoma, or undifferentiated pleomorphic sarcoma pattern). The sarcomatous component must be identified to distinguish this from conventional GCT with reactive osteoid deposition or from osteosarcoma with giant cells. Prognosis is poor, similar to high-grade osteosarcoma, with multimodal therapy (surgery + chemotherapy) required.

Secondary malignant GCT (radiation-induced) has historically been the more common form, arising in GCT sites previously treated with radiation therapy. Radiation was once used for surgically inaccessible GCTs (sacral, spinal) before denosumab was available. The latency from irradiation to sarcoma development is typically 5–10 years, and the risk at 10 years is estimated at approximately 10%. This unacceptably high rate of radiation-induced sarcoma is the primary reason radiation is now largely abandoned for conventional GCT. Denosumab has supplanted radiation for inoperable sites, eliminating this long-term complication in appropriately managed patients.

Spontaneous malignant transformation without prior radiation is rare and its biological determinants are incompletely understood. Follow-up imaging at regular intervals (every 6 months for 2 years, then annually to year 5) is standard practice and will detect the majority of local recurrences and, with chest CT, pulmonary metastases at a stage amenable to curative-intent intervention.

Back to Table of Contents

Key Research Papers

  1. Thomas DM et al. Denosumab in patients with giant-cell tumour of bone: an open-label phase 2 study. Lancet Oncology 2010; 11:275–280. PMID 20362507
  2. Behjati S et al. Distinct H3F3A and H3F3B driver mutations define chondroblastoma and giant cell tumor of bone. Nature Genetics 2013; 45:1479–1482. PMID 23832583
  3. Campanacci M et al. Giant-cell tumor of bone. Journal of Bone and Joint Surgery (Am) 1987; 69:106–114. PMID 3559675
  4. Rutkowski P et al. Denosumab treatment of inoperable or locally advanced giant cell tumor of bone. European Journal of Cancer 2015; 51:1030–1038. PMID 25936224
  5. van der Heijden L et al. Outcome of giant cell tumour of bone after intralesional treatment: a meta-analysis of 1301 patients. European Journal of Cancer 2013; 49:3375–3385. PMID 23968735
  6. Errani C et al. Giant cell tumor of the appendicular skeleton. Clinical Orthopaedics and Related Research 2011; 469:3239–3248. PMID 21327475
  7. Clayer M. Injectable form of calcium sulphate as adjuvant treatment of aneurysmal bone cysts or giant cell tumour of bone. ANZ Journal of Surgery 2013; 83:336–339. PMID 23825016
  8. Beebe-Dimmer JL et al. Giant cell tumor of bone: findings from the SEER database, 1975–2013. Journal of Surgical Oncology 2017; 115:261–268. PMID 28112422
  9. Balke M et al. RANK expression as a prognostic marker in osteosarcoma. European Journal of Cancer 2008; 44:1715–1722. PMID 18178418
  10. Martin SE et al. Giant cell tumor of bone with pulmonary metastases: clinical and immunohistochemical correlation. American Journal of Clinical Pathology 2012; 137:960–966. PMID 22706856
  11. Boriani S et al. Giant cell tumor of the spine: a review of 17 cases. Spine 2012; 37:E285–292. PMID 22561825
  12. Turcotte RE et al. Giant cell tumor of the sacrum. Clinical Orthopaedics and Related Research 2002; 397:222–235. PMID 12218476

Back to Table of Contents

PubMed Topic Searches

For current research and clinical trial data, search PubMed directly:

  1. Giant cell tumor bone denosumab RANKL treatment
  2. Giant cell tumor bone intralesional curettage recurrence
  3. Giant cell tumor bone H3F3A mutation diagnosis
  4. Giant cell tumor bone pulmonary metastases outcomes

Back to Table of Contents

Connections

Back to Table of Contents