Abstract
Giant cell tumor of bone (GCTB) consists of a mixture of neoplastic mononuclear cells and non-neoplastic cells, including polynuclear giant cells. Recently, with the spread of the immunohistochemical staining marker H3.3 G34W corresponding to specific genetic abnormalities, the histological diversity of GCTB has been recognized. GCTB without giant cells is uncommon, although it has also been reported previously. Herein, we describe a 45-year-old man with GCTB without giant cells who was successfully diagnosed using H3.3 G34W immunohistochemistry. Other unusual findings in GCTB that were identified in this patient include bone and osteoid formation with a long clinical course of 13 years. We also compared the histological findings of the current patient to those who received denosumab therapy.
Introduction
Giant cell tumor of bone (GCTB) is a locally aggressive and rarely metastasizing bone tumor with a peak incidence in the second to fourth decades of life. 1 Histologically, these tumors are generally composed of neoplastic mononuclear cells admixed with a large number of non-neoplastic osteoclast-like giant cells. 1 Recently, Behjati et al 2 identified a driver mutation in the H3 histone family member 3A gene (H3F3A), which encodes histone H3.3. This mutation is found in 96% of GCTB patients; however, it is not in other types of tumors. Approximately 90% of H3F3A mutations are H3.3 p.Gly34Trp missense mutations,2–6 which can be recognized by specific monoclonal antibodies. H3.3 G34W immunohistochemistry is useful in differentiating GCTB from other giant cell-rich lesions with excellent sensitivity and specificity, and it facilitates the recognition of GCTB with various histological features.7–9
H3.3 G34W immunohistochemistry also allows for the recognition of neoplastic mononuclear tumor cells on glass slides in daily practice. With the use of H3.3 G34W immunohistochemistry, GCTB without multinucleated giant cells has recently been reported. 10 GCTB is now known to have greater morphological diversity than previously understood.
In this report, we describe a patient with bone-forming GCTB without giant cells. This is the first case report of long-term follow-up without treatment for this type of GCTB. In addition, we evaluate the histological similarities to GCTB after denosumab therapy.
Case Presentation
Case Description
A 45-year-old man presented with pain in the left knee, which was first noted at the age of 32 years and was followed up for 13 years. His first x-ray examination revealed a well-defined osteolytic lesion with a sclerotic margin measuring 25 mm in diameter on the lateral side of the distal left femur (Figure 1A). The initial radiographic findings suggested a benign lesion, such as an aneurysmal bone cyst or non-ossifying fibroma. A recent follow-up x-ray showed a slight enlargement of the lesion (Figure 1B). Magnetic resonance imaging showed an ill-defined heterogeneous high-signal intensity mass with a low-signal intensity rim on T2-weighted imaging (Figure 1C). Slight peritumoral bone marrow edema was also noted. Positron emission tomography-computed tomography scanning revealed moderate fluorodeoxyglucose uptake (maximum standardized uptake value = 4.9) of an osteolytic lesion with a sclerotic rim (Figure 1D), which suggested chondrosarcoma as a differential diagnosis. The serum tartrate-resistant acid phosphatase 5b (TRACP-5b) level was low at 298 mU/dL. The biopsied specimen at the age of 45 years was entirely composed of xanthogranulomatous changes. Although the classic histological characteristics of GCTB, such as numerous giant cells, were not observed, H3.3 G34W-positive cells were immunohistochemically scattered, suggesting the diagnosis of GCTB. Given the high risk of fracture, total knee arthroplasty was performed 3 months later without neoadjuvant or adjuvant treatment with denosumab. The patient was free from recurrence and metastasis after 24 months of follow-up. Written informed consent was obtained from the patient for publication of this case report, with the preservation of the patient's anonymity.

Radiological findings. (A) The conventional radiograph at the time of his initial complaint shows a well-defined osteolytic lesion with a sclerotic margin in the lateral side of the distal left femur. (B) The latest conventional radiograph before the surgery shows only slight enlargement of the lesion. (C) Magnetic resonance imaging (T2-weighted imaging) shows ill-defined heterogeneous high-signal intensity mass with low-signal intensity rim and slight peritumoral bone marrow edema. (D) Positron emission tomography-computed tomography scanning revealed moderate fluorodeoxyglucose uptake (maximum standardized uptake value = 4.9) to an osteolytic lesion.
Macroscopic, Microscopic, and Immunohistochemical Findings
Grossly, a 40 × 40 × 35 mm, yellow, white, and light brown solid mass was observed at the lateral posterior end of the distal metaphysis (Figure 2). The entire lesion was sent for histological examination.

Macroscopic findings. Macroscopically, a 40 × 40 × 35 mm solid mass was observed in the distal end of the left femur. The cut surface was yellow, white, and light brown. The border of the lesion was partly unclear at the lateral posterior end of the distal metaphysis.
Histologically, there was a proliferation of spindle-shaped mononuclear cells with oval-shaped nuclei without atypia (Figure 3A). Clusters of foamy cells and fibrotic areas were interspersed (Figure 3B). Mononuclear cells partially formed osteoids and bones and encircled them (Figure 3C). Hyalinizing sclerotic areas were also observed (Figure 3D). Osteoclastic giant cells were not observed.

Microscopic findings. Histologically, several patterns were heterogeneously mixed within the tumor. (A) The tumor tissue consisted of spindle-shaped mononuclear cells without atypia. No giant cells were observed (inset, high-power view). (B) Clusters of foamy histiocytes mixed with mononuclear cells. (C) Mononuclear cells partly form woven bone or osteoid and surround them. (D) In some areas, the tumor tissue predominantly consisted of hyalinizing sclerotic changes.
Immunohistochemistry revealed that H3.3 G34W was positive in approximately 50% of the spindle-shaped mononuclear cells (Figure 4A). RUNX2 was also positive in most mononuclear cells (Figure 4B). NFATc1-positive cells were nearly absent, except for a few cells (Figure 4C). The Ki-67 proliferation index was approximately 3% (Figure 4D).

Immunohistochemical findings. (A) H3.3 G34W immunohistochemistry was positive for about 50% of spindle-shaped mononuclear cells. (B) RUNX2 was positive for most of neoplastic mononuclear cells. (C) Nuclear positivity of NFATc1 staining was seen in only a few cells. (D) The Ki-67 proliferation index of neoplastic cells was approximately 3%.
Discussion
In this report, we describe a patient with bone-forming GCTB without giant cells that had an indolent clinical course. To the best of our knowledge, this is the first report of long-term follow-up in a patient with GCTB.
A typical GCTB is histologically dominated by osteoclast-like giant cells, between which mononuclear cells are embedded. Because of its histology, a GCTB was initially thought to be an osteoclast tumor. 11 However, several studies that used tissue cultures suggested that the mononuclear cells are truly neoplastic, and the osteoclastic cells are merely recruited by the neoplastic cells, similar to other forms of cancer.1,12 The discovery of a distinct driver mutation of H3F3A revealed the neoplastic features of mononuclear tumor cells, thus providing evidence for the findings of tissue culture. 2 Nowadays, mononuclear tumor cells can be recognized in clinical practice through H3.3 G34W immunohistochemistry. It has become clear that GCTBs have more diverse histological appearances than previously thought.
In the present study, the tumor was histologically unusual in many ways. Notably, no giant cells were found in the tumors. The diagnosis of histologically atypical tumors requires comprehensive judgment and must be made after considering clinical, radiological, and histological findings. Before the discovery of the H3F3A mutation, this tumor would have been judged as fibrous histiocytoma of the bone. 13 However, the clinical and radiological suggestions of GCTB, and the fact that GCTB sometimes shows a secondary fibrohistiocytic reaction, led us to perform H3.3 G34W immunohistochemistry which enabled us to accurately diagnose this tumor as GCTB. Recently, three GCTBs without giant cells have been reported, and our findings also support the utility of H3.3 G34W immunohistochemistry. 10
Another notable histological finding in the current patient is that osteoid and bone formation was focally recognized in some areas. GCTB rarely exhibits bone formation although it sometimes shows secondary changes such as hemorrhage, aneurysmal bone cyst-like changes, and fibrohistiocytic reactions. 1 However, in this patient, bone and osteoid formation was found multifocally and surrounded by mononuclear tumor cells, suggesting osteogenesis by tumor cells. The finding of the proliferation of spindle cells with bone formation implies a histological resemblance to GCTB after administration of denosumab, as described in a previous paper. 14 Immunohistochemically, RUNX2 and H3.3 G34W were positive for osteogenic mononuclear tumor cells, and nuclear NFATc1-positive cells were almost absent. These immunohistochemical findings are also similar to those observed after the administration of denosumab.1,14 From these results, it is suspected that a similar mechanism, which disturbs the interaction between tumor cells and osteoclastic cells, mediated by the RANK-RANKL axis, worked in this tumor. RANK-RANKL axis mediation may have resulted in the induction of tumor cell osteogenesis and inhibition of NFATc1-positive giant cell recruitment.
The long-term course was also specific in the present patient. Although the clinical behavior of GCTB is variable, it is unusual to observe no rapid tumor growth or exacerbation of symptoms for almost 13 years. The low value of TRACP-5b, which is used as a marker of bone resorption and osteoclast activity, also suggests low tumor activity in this patient. TRACP-5b is usually high in GCTBs and has been reported as a useful marker for the diagnosis and recurrence of GCTB. 15 This long clinical course might be attributed to the suppression of osteolysis by a similar change as in denosumab-treated tumors, although it was not clear when this change occurred. In the aforementioned report of three patients with GCTB without giant cells, one patient showed marked osteoid or bone formation and no increase in the size of the lesion within a follow-up period of 12 months. 10 It is suggested that this type of GCTB with bone formation and no giant cells may have a long clinical course.
In summary, we report a patient with GCTB showing unusual histology with bone and osteoid formation and no multinuclear giant cells. Diagnosis based only on imaging and histological findings was challenging; however, immunostaining of H3.3 G34W was useful. GCTB without giant cells is thought to have changes similar to those that received denosumab treatment and may take a long-term course even without treatment.
Footnotes
Acknowledgments
The authors thank Yoko Anzai, Eri Uno, Yukiko Nishio, and Takayuki Akagi for their technical assistance.
Author Contributions
HT: Writing – original draft, Investigation, and Visualization. IK: Conceptualization, Investigation, Writing – Review, and Editing. JI: Investigation, Writing – Review, and Editing. YK: Resources, Writing – Review, and Editing. TO: Resources, Writing – review and editing, Visualization. MT: Resources, Writing – Review, and Editing. SF: Investigation, Writing – review and editing, Supervision.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Japan Society for the Promotion of Science, (grant number 16K19085).
Ethical Approval
Yokohama City University Institutional Review Board No. A170525013
Informed Consent
Not applicable
Trial Registration
Not applicable
