Abstract
Malignant transformation of giant cell tumour of the bone is extremely rare. In addition, bone transformation in giant cell tumour may occur in different phases. With conventional X-rays, CT scans or MRIs, it may be challenging to distinguish among different phases of bone transformation, normal bone, soft tissue disease and bone disease (benign vs malignant lesions) and changes in multiple organs such as lung, liver and lymph nodes unless every lesion is biopsied, which is not practical. Molecular imaging with different isotopes (Tc-99m phosphonate, 2-deoxy-2-(18F)fluoro-d-glucose and sodium fluoride-18) may help to better characterise the disease. We hypothesised that molecular imaging could offer qualitative and quantitative characterisation of all stages of bone formation, destruction, reactivity or neoplasia in a patient with giant cell tumour of the bone, and we present the first case of molecular imaging where bone formation was seen in multiple soft tissues, such as lungs, muscles, lymph nodes and liver.
Keywords: cancer intervention, carcinogenesis, screening (oncology), radiology
Background
Giant cell tumour of bone is a rare, benign neoplasm that is locally aggressive. Histopathologically, it is characterised by many osteoclastic giant cells that are induced by neoplastic mononuclear cells. Malignant transformation of giant cell tumour of the bone, which could be primary or secondary neoplastic transformation and sarcomatous change, is an extremely rare event.1–3 Receptor activator of nuclear factor kappa B ligand (RANKL) seems to be crucial in the pathogenesis of giant cell tumour, and the RANKL inhibitor denosumab has shown activity in locally aggressive giant cell tumours.4–6 After malignant transformation, however, treatment is challenging given the aggressive nature of the disease. In many cases, it behaves like osteosarcoma. After surgery and radiation therapy, aggressive malignant transformed giant cell tumours are treated with chemotherapeutic agents as for osteosarcoma.
Bone transformation in malignant giant cell tumour may be in different phases. With conventional X-rays, CT scans or MRIs, it may be challenging to distinguish among different phases of bone transformation, normal bone, soft tissue disease, bone disease (benign vs malignant lesions) and changes in organs other than bone (such as lung, liver and lymph nodes) unless every lesion is biopsied, which is not practical.
Molecular imaging with different isotopes may provide more insight in characterising the disease. Tc-99m phosphonate compounds used for commonly called ‘bone scintigraphy’ are avidly taken up by osteoblastic cells, thereby showing areas of new bone formation. 2-Deoxy-2-(18F)fluoro-d-glucose (18F-FDG), which is the most common agent used in positron emission tomography (PET), is an analogue of glucose, and it provides valuable functional information based on the increased glucose uptake and glycolysis of cancer cells and other metabolic abnormalities. Bone uptake of sodium fluoride-18 (Na18F) reflects bone remodelling, and this uptake is a portion of the bone matrix mineralisation. 18F− is traded for OH− in the bone. This hydroxy–apatite bone matrix is changed into fluoroapatite. Because increased uptake of Na18F reflects bone reaction to cancer tissue, a positive finding with Na18F PET/CT may be seen in both benign and malignant bone conditions. In the soft tissues, bone formation may occur as well. Together these radioactive isotopes may provide information complementary to each other in imaging.
We hypothesised that molecular imaging (with Tc-99m MDP, sodium fluoride-18-PET/CT and fluoro-18-deoxyglucose PET/CT) could offer dynamic qualitative and quantitative characterisation of all stages of bone formation, destruction, reactivity or neoplasia in a patient with giant cell tumour of the bone. Here, we present the first case of molecular imaging where bone formation was seen in multiple soft tissues, including the lungs, muscles, lymph nodes and liver.
Case presentation
A woman in her 30s was in her usual state of health when she noted lower back pain. Workup with CT scans and MRI of the spine revealed tumour at Th 8-Th 9, which proved to be sarcoma with osteoid components following resection. Two years later, she noted left rib cage pain ultimately diagnosed as recurrent tumour in the surgical bed and underwent a second surgical resection. During follow-up a year later, regrowth of the tumour was noted in the thoracic spine. She was evaluated by neurosurgery and underwent a third resection of the thoracic spinal tumour, which included posterior osteotomy for en bloc resection of the Th 9 vertebral body, thoracoplasty of Th 9 and Th 10 and posterior spinal fusion of Th 5-Th 11. Spinal instrumentation was removed, and the spinal cord was decompressed. Anterior spinal reconstruction was performed using a Stryker expandable cage. The patient was hospitalised for about 3 weeks, during which time she experienced left leg-wide weakness postprocedure and subsequently recovered. Pathology revealed grade 2/3 osteosarcoma with vertebral lymphatic invasion with intravascular osteosarcoma seen. Ninth and 10th posterior rib margins were positive. Postsurgical CT of the chest revealed postoperative changes in the thoracic spine as well as a 2 mm nodule in the left lung. Postoperatively, she received radiation to the distal thoracic spine to a dose of 60 Gy in 30 fractions. Subsequently, she commenced adjuvant chemotherapy with cisplatin plus doxorubicin for a total of six cycles. She was observed for approximately 16 months and eventually had disease progression. At the time of progression, she received ifosfamide plus etoposide for a total of four cycles. She was also enrolled on a clinical trial, receiving insulin-like growth factor 1 receptor antibody for 14 months. Due to progression, she was switched to gemcitabine plus docetaxel for four cycles and then gemcitabine alone for another five cycles. On progression, she was started on sorafenib for a total of approximately 8 months. She experienced numbness in the back and right upper extremity, and MRI of the cervical spine showed multifocal metastatic disease involving vertebral bodies and the base of the pedicle articular facet junction with lateral recess stenosis of right C5-C6 and right C7-Th 1. MRI of the lumbar spine revealed relatively stable-appearing metastatic disease at L1 extending into the lateral right-sided paravertebral at L1. Extensive metastatic lesions were noted involving the left S1 and S2 with encasement of the left exiting S1 traversing the S2 nerve root. Pain in the right upper extremity was treated with steroids, which helped ease the pain somewhat over the next several weeks. The patient then developed constant severe pain within her left shoulder blade. The pain radiated slightly to her neck and she noted numbness of the back of her arm from the shoulder to the elbow. CT of the cervical spine revealed sclerotic lesions in the region of the clivus, right occipital condyles, C2-C4, C5-C6 and C7. There was a lytic focus to the right transverse process of Th1 with soft tissue mineralisation noted. Paravertebral foraminal and foramen transversarium mineralisation on the right at C5 through C6 with right epidural extension was noted, as well as significant foraminal stenosis in the right C5-C6 and Th 1 through Th 2. Foraminal encroachment at C7-Th 1 was also noted. The patient received a total of 37.5 Gy in 15 fractions of 2.5 Gy with a 6 MV photon dose to the 98% isodose line; the region treated was C4-Th 2. She underwent hemiarthroplasty of the right femoral neck for impending pathological fracture. She then received a total of 27.5 Gy in five fractions of 5.5 Gy with a 6 MV photon dose to the 100% isodose line to the C-spine and then a total of 13 days of radiotherapy to both the right shoulder as well as the L-spine and pelvis. The right shoulder treatment was to a dose of 30 Gy in 10 fractions of 3 Gy each with a 6 MV photon dose to the 98% isodense line. The spine and pelvic field were treated with a total of 30 Gy in 10 fractions of 3 Gy each with 15 mV photons to the 100% isodose line. Following this was again treated for total of 22 days of C4 through Th2 with a total of 37.5 Gy in 15 fractions of 2.5 Gy each with a 6 MV photon dose to the 98% isodose line. The patient also received radiation to the right cavernous sinus as well. Throughout this entire course she was started on the RANKL inhibitor denosumab and continued on that.
Because of further progressive disease, she was referred to our institution for a second opinion and enrolment into a radiopharmaceutical trial with the alpha emitter radium-223.7 8 At that time, the pathology findings were confirmed and molecular imaging was performed to better characterise this rare disease.
Investigations
The extent of her disease prior to therapy was evaluated by bone scintigraphy with Tc-99m methylene diphosphonate (MDP) single-photon emission computed tomography (SPECT), sodium fluoride-18-PET/CT and fluoro-18-deoxyglucose PET/CT. The whole body images on consecutive days are shown in figure 1.
We also evaluated the serum bone turnover markers: osteocalcin, β-crosslaps or C-terminal telopeptide, alkaline phosphatase and bone alkaline phosphatase.
Tumour burden was assessed by lactate dehydrogenase (LDH) Other tumour markers were obtained at baseline: CA 125, CA 15–3 and carcino embryonic antigen (CEA).
Histopathological review.
Figure 1.
Whole body molecular imaging studies with different techniques. Bone scintigraphy with whole body planar images: (A) anterior-poster (AP) view, (B) posterior-anterior (P/A) view, (C) Na18F-PET maximum intensity projection (MIP) images and (D) 18FDG-PET MIP images. The figure indicates that Na18F-PET (C) shows more lesions than does bone scintigraphy (A and B). Na18F, sodium fluoride-18; PET, positron emission tomography.
Differential diagnosis
The differential diagnosis included giant cell tumour of bone, osteosarcoma and malignant transformation of giant cell tumour of bone.
Results
All whole body planar and maximum intensity projection images showed multiple foci of increased skeletal tracer activity, consistent with diffusely metastatic malignant giant cell tumour of the bone. The dominant radiotracer activity was seen in the posterior right chest wall and in the sacrum and iliac bones, with additional sites of activity throughout the axial and proximal appendicular skeleton (figure 1A–D). The three tracers demonstrated different distributions. There were extensive skeletal metastases on Na18F-PET/CT, with multiple foci in the skull, in the cervical, thoracic and lumbar spine, multiple bilateral ribs, bilateral scapulae, clavicles, sternum, proximal humeri, extensive multifocal involvement of the pelvis and bilateral femurs.
SPECT-CT imaging of the thorax confirmed radiotracer activity in multiple sites of osteoblastic metastatic disease. There were postsurgical changes in the thoracic spine with osteosynthetic material. Extensive recurrent tumour was seen around the surgical bed, including disease extending into the left posterior chest wall at the level of the eighth rib (figure 2A, D and E). There were enlarged bilateral pulmonary nodules consistent with metastases. Some of these were ossified or partially ossified and showed fairly substantial radiotracer activity, whereas the non-calcified showed very minimal activity.
Figure 2.
Molecular imaging of lung metastases with different techniques at the same level. Bone scintigraphy with SPECT fusion image with CT (A), Na18F-PET images (B), Na18F-PET fusion image with CT (C), CT with skeletal window from the lungs (D), CT with soft tissue window from the lungs (E), 18FDG-PET images (F) and 18FDG-PET fusion image with CT (G). Na18F, sodium fluoride-18; PET, positron emission tomography; SPECT, single-photon emission computed tomography.14
On FDG-PET, there were numerous bilateral pulmonary nodules, the largest of which was visibly FDG avid, consistent with extensive pleural metastases. These were 1.5–3 cm in size, and the maximum standardised uptake value (SUVmax) varied from 6.8 to 12.3 (figure 2F,G). Note that these bilateral lesions were not bone-producing bone as seen in figure 2A (bone SPECT-CT) and figure 2B,C (Na18F-PET).
Many of these bilateral pulmonary nodules consistent with pulmonary metastases were without associated Na18F uptake, although there were still multiple right lower lobe pulmonary nodules that demonstrated associated uptake. The most intensely sodium fluoride-avid calcified pulmonary nodule demonstrated an SUVmax of 17.8.
In the skeleton, there were multiple FDG-avid skeletal metastases, but many lesions, such as those of the skull and cervical spine, were better seen on the F-18 sodium fluoride PET/CT (figure 1A–B). The lesions were predominantly sclerotic, and many had expansions to soft tissues, as also seen in figure 2D,E, where the lesion at the level of Th8 on the left involved adjacent soft tissues at the posterior chest wall (figure 2D; SUV 13.9). Many of the skeletal metastases were expansive and involved adjacent soft tissues as seen in Na18F-PET images, for example, an expansion 3 cm lesion at the medial aspect of the right scapula (figure 3A, SUVmax 24.7). A large, calcified, lobulated mass at the level of Th8 extended to the left involving the paravertebral soft tissues and soft tissues adjacent to the left eighth rib (figure 2A–G, SUVmax 26.3). Additional representative examples include a large lesion at L1 with photopenia centrally and peripheral intense uptake (figure 3B, SUVmax 38.9), an exophytic lesion in the left iliac bone (figure 3C, SUV 37.5) and a lesion in the left and right intramedullary femur (figure 3D, SUV up to 31.8). Focal uptake was noted corresponding to a soft tissue lesion that might have involved the right thigh musculature (figure 3D; SUV 17.5).
Figure 3.
Na18F-PET images fusion images with CT. Demonstrates soft tissue expansion above the right scapula (arrow) (A), photopenic middle in Th 8 vertebra and invasion to paraspinal spaces (arrow) (B), soft tissue expansion to muscles inside the left ileum/musculus iliacus (arrow) (C) and invasion to upper thigh muscle musculus adductor longus (arrow) and a lesion in the left and right intramedullary femur (yellow arrow) (D). Na18F, sodium fluoride-18; PET, positron emission tomography.
Interestingly, a low-attenuation lesion in the lateral part of the right hepatic lobe demonstrated a faint sodium fluoride uptake in the liver (figure 4A,B, fusion and PET images: arrows, respectively). This lesion was FDG avid (SUV 15.2). Because this lesion was weakly calcified, it was seen on NaF-PET (figure 4A,B). Increased Na18F uptake was also noted corresponding to a 1.8×1.5 cm right external iliac lymph node (figure 4C, SUV 9.2), which also was FDG avid (figure 4D, SUV 9.3).
Figure 4.
Molecular imaging showing unusual extraosseous sites of metastases Na18F-PET images in the demonstrate a faint uptake in a small liver metastasis fusion image with CT (A) and Na18F-PET image (B). The right external iliac lymph node is visualised by both Na18F-PET (C) and 18FDG-PET (D). Na18F, sodium fluoride-18; PET, positron emission tomography.
The volume of whole skeleton in the imaged area was 4830 cm3, as measured on CT (Houndsfield unit (HU)>200). Similarly, the volume of sclerotic bone was 1915 cm3 (HU >600); that is, 39.6% of the bone was sclerotic. In this same location (as defined by CT), the total Sodium Fluoride (NaF) amount (concentration×volume) was 1030 g. In this skeletal volume (4830 cm3), the pathological Na18F volume (TFI10, when defining the SUVmax >10) was 445 cm3 and the maximum Na18F uptake was 45.6 (mean 14.7).
In the sclerotic skeletal volume (1915 cm3), the pathological Na18F volume (TFI10, when SUVmax >10) was 91.3 cm3 and the maximum Na18F uptake of 45.6 was in this volume, but the mean value of 13.6 was actually lower than the average in the whole skeleton (14.7). Thus, only one-fifth (20.5%) of the skeletal disease burden was located in the sclerotic bone, as seen on CT.
The total lesion glycolysis was 5790 g in the total skeletal volume (4830 cm3).
In the total skeletal volume (4830 cm3), the pathological 18FDG volume (SUVmax >2.5) was 2230 cm3 and the maximum 18FDG uptake was 14.8 (mean 1.6). In the sclerotic skeletal volume (1915 cm3), the pathological 18FDG volume (SUVmax >2.5) was 780 cm3 and the maximum 18FDG uptake was 12.9 (mean 1.6). The highest uptake of 18FDG was located outside the sclerotic area, but still 35.0% of 18FDG disease burden in the skeleton was located in the sclerotic area.
In the soft tissues, there were many muscular expansions, for example, muscular expansion in 3D was extremely fluoride avid (SUVmax 37.5), and the tissues also showed FDG uptake (SUVmax 7.8). However, in the pathological NaF lesion volume of 190 cm3, the mean 18FDG uptake was only 1.7, meaning that most of the Na18F avid tumour tissue was actually 18FDG negative.
In the lungs, there were large differences between tracers, because a large part of the lesions, approximately 80%, were Na18F negative but almost all were FDG positive. The total lesion glycolysis in the lung metastases was 6580 g, whereas the total NaF burden in the lungs was 1340 g. Additionally, liver and lymph nodes were visualised with both tracers. Because there were only single solitary lesions, no further conclusions can be drawn.
Cumulative dose of radiation with NaF-PET and FDG PET/CT
With Na18F, the effective radiation dose is 0.024 mSv/MBq and with 18FDG 0.019 mSv/MBq. By taking account administered activities and CT studies, the total effective dose was 11 mSv +11 mSv, that is, 22 mSv.
Multiplanar MRI of the cervical, thoracic and lumbar spine with and without intravenous contrast using multisequence parameters was also performed. Epidural tumour was prominent at Th 7, occupying the left side of the spinal canal and displacing the cord to the right, producing mild/moderate cord compression. There was diffuse metastatic destruction of the Th 7 vertebral body extending into the inferior aspect of Th 6 and the origin of the left seventh rib. There was also paraspinal extension extending laterally and anteriorly at the level of Th 7, which correlated with the extensive metastatic activity seen on PET/CT studies (figure 2A–G).
A large degree of epidural tumour was also visualised at L5 and in the upper sacrum. Tumour nearly completely filled the left L5/S1 neural foramen. Epidural tumour additionally entered the right L1/2 neural foramen, abutting the right L1 nerve root. There was marked involvement of the left sacral ala with filling of the lower spinal canal with metastatic epidural involvement and filling of the canal with static disease.
Scattered osseous metastases were seen on MRI. The lesion within the Th 12 vertebral body measured 2.5 cm and the lesion within the L4 vertebral body measured 2.6 cm.
Numerous pulmonary metastases were visualised. The bone turnover and tumour markers were also obtained (table 1).
Table 1.
Baseline characteristics of the serum turnover markers and molecular imaging parameters with respect to the activity in the scans
| Parameter | Value (normal range) |
| Alkaline phosphatase | 93 IU/L (38–126) |
| Bone alkaline phosphatase | 8.1 µg/L (<22) |
| β-crosslaps or C-terminal telopeptide | 101 pg/mL (25–573) |
| Osteocalcin | 4 ng/mL (9–42) |
| Calcium | 9.3 mg/dL (8.4–10.2) |
| Serum LDH | 618 IU/L (313–618) |
| Albumin | 3.6 g/dL (3.5–4.7) |
| CEA | 2.1 ng/mL (0.0–3.0) |
| CA 19–9 | 2.2 U/mL (0.0–35.0) |
| CA 15–3 | 20.5 U/mL (0.0–25.0) |
| CA 125 | 14.4 U/mL (0.0–35.0) |
| Total skeletal volume | 4830 cm3 |
| Pathological skeletal NaF volume | 1030 cm3 |
| Pathological skeletal FDG volume | 2230 cm3 |
| Total sclerotic bone volume | 1915 cm3 |
| Pathological sclerotic bone NaF volume | 445 cm3 |
| Pathological sclerotic bone FDG volume | 780 cm3 |
| Total lesional glycolysis in lung metastases | 6580 gm |
| Total NaF burden in lung metastases | 1340 gm |
In the majority of sections, histopathology showed a cellular tumour composed of atypical spindle cells with scattered mitotic figures, and focally, there were areas of more conventional giant cell tumour of bone with multinucleated giant cells and mononuclear cells (figure 5, H&E, 200×). This was consistent with malignant transformation of giant cell tumour of the bone.
Figure 5.
Histopathology images of the giant cell tumour with malignant transformation. (A) Most sections revealed a cellular tumour composed of atypical spindle cells with scattered mitotic figures (H&E, 200×). (B) Focally, there were areas of more conventional giant cell tumour of bone with multinucleated giant cells and mononuclear cells (H&E, 200×).
Outcome and follow-up
We have described the comprehensive molecular imaging of giant cell tumour of bone.
Discussion
Malignant transformation of giant cell tumour of the bone is an extremely rare disease that has been defined mainly by pathological features. A review of literature shows cases of giant cell tumour that have undergone malignant transformation to osteosarcoma are being reported.9–11 Our case would add to these growing body of cases. In all these cases, however, it is unclear if denosumab played a role in malignant transformation, but all the more it underscores the importance of close-follow up of patients with giant cell tumour of the bone on denosumab. Previous studies have imaged this tumour using MRIs, which may provide clues into some of the disease biology. Bone sarcomas, including giant cell tumour, that take up FDG-PET have been imaged by FDG-PET, and their characteristics have been reported.12 13 The radiographic differential diagnoses of giant cell tumour of bone are osteosarcoma, malignant transformation of giant cell tumour of bone, Ewing’s sarcoma, chondrosarcoma and any of the tumours that can metastasise to the bone.
Conventional imaging may not be able to characterise transformed giant cell tumour in the bone comprehensively. Here we have shown that PET tracers NaF and FDG have different behaviour in this disease, and we have also shown that the tumour is heterogeneous and has a complex biology, being characterised by both bone mineralisation and active glycolysis. Different tumour localisations demonstrated different amounts of bone mineralisation, and we found that such mineralisation occurred in many soft tissue types simultaneously. Soft tissue invasion was evident in local adjacent musculature in the thoracic, pelvic and lower extremity areas (figures 2 and 3). Clear locoregional spread could be demonstrated to pelvic lymph nodes. Distant metastases were seen in lungs, liver and pelvic lymph nodes, and this finding has not been previously noted in the literature. FDG-PET showed lesions in soft tissues as well but did not indicate whether bone formation was present or not. This can clearly be seen by comparing the FDG-PET images with those of NaF-PET. However, NaF-PET demonstrated more abundant skeletal lesions than did FDG-PET (figure 1). This allows us to decide when bone-targeted therapies are useful in bone sarcomas or other similar conditions when there is bone formation in soft tissues. In addition to really unravel the different stages of bone formation, destruction, reactivity or neoplasia, we would need a biopsy at different stages, and we did not perform this given the patients’ clinical condition. With one case, we cannot determine the prognostic value for using multimodality imaging for assessment of this rare tumour. However, a multimodal imaging technique could be helpful in go versus no go decisions for these expensive therapies such as Radium 223. Moreover, it is important to know the response to a given therapy for patients’ clinical benefit. There is a need for further prospective studies to better correlate the imaging findings with biopsy, and bone turnover markers to aid in the differential diagnosis of neoplasia versus stages of bone reactivity.
This information also helps us to understand why patients treated with bone-targeting substances do not always respond or why only some lesions respond. There are data using NaF-PET to assess response to therapy in prostate cancer post-Radium 223 therapy. However, there are no data in this rare disease. Tc-99m phosphonate can pick bone metastases and may not pick soft tissue disease. FDG-PET may pick soft tissue disease and may not be the best test for bone disease. Sodium fluoride is similar to Tc-99m phosphonate that it can better pick bone disease than soft tissue disease. However, NaF-PET scan can both qualitatively and quantitatively image bone tumours compared with just quantitative with bone scans.
The uptake of the radioactive isotopes showed similarities between Tc-99 MDP and NaF-PET and many differences in the FDG-PET uptake and among the different scans. All three scans are complementary to each other and would be needed to characterise the diverse lesions. Characterisation and follow-up using the three scans with various treatment options for osteosarcoma and other bone sarcomas may show differential responses to therapy in the future.
We have presented the first case of comprehensive characterisation of malignant giant cell tumour. This provides a ‘snap shot’ of this aggressive cancer at a moment of time. Further studies are warranted with these different radiotracers in larger numbers of patients with giant cell tumour of the bone to better define the disease biology.
Learning points.
Malignant transformation of giant cell tumour is an extremely rare event.
Molecular imaging provides insights into the diversity of the disease.
Tc-99m phosphonate can pick bone metastases and may not pick soft tissue disease.
FDG-PET may pick soft tissue disease and may not be the best test for bone disease.
Sodium fluoride positron emission tomography (NaF-PET)/CT is similar to Tc-99m phosphonate that it can better pick bone disease than soft tissue disease.
NaF-PET and FDG-PET offer complementary information to better characterise neoplastic bone formation seen in multiple soft tissues, such as lungs, muscles, lymph nodes, and liver.
Footnotes
Twitter: @VivekSubbiah
Contributors: KK: discuss planning, reporting, conception and design, acquisition of data or analysis and interpretation of data. W-LW: discuss planning, acquisition of data or analysis and interpretation of data. VS: discuss planning, conduct, reporting, conception and design, acquisition of data or analysis and interpretation of data.
Funding: This study was funded by Shanon Wilkes Research Foundation. The University of Texas MD Anderson Cancer Center is funded by NIH/NCI Cancer CenterSupport Grant P30 CA016672.
Competing interests: None declared.
Provenance and peer review: Not commissioned; externally peer reviewed.
Patient consent for publication: Obtained
References
- 1. Marui T, Yamamoto T, Yoshihara H, et al. De novo malignant transformation of giant cell tumor of bone. Skeletal Radiol 2001;30:104–8. 10.1007/s002560000305 [DOI] [PubMed] [Google Scholar]
- 2. Oda Y, Sakamoto A, Saito T, et al. Secondary malignant giant-cell tumour of bone: molecular abnormalities of p53 and H-ras gene correlated with malignant transformation. Histopathology 2001;39:629–37. 10.1046/j.1365-2559.2001.01275.x [DOI] [PubMed] [Google Scholar]
- 3. Mori Y, Tsuchiya H, Karita M, et al. Malignant transformation of a giant cell tumor 25 years after initial treatment. Clin Orthop Relat Res 2000;381:185–91. 10.1097/00003086-200012000-00022 [DOI] [PubMed] [Google Scholar]
- 4. Thomas D, Henshaw R, Skubitz K, et al. Denosumab in patients with giant-cell tumour of bone: an open-label, phase 2 study. Lancet Oncol 2010;11:275–80. 10.1016/S1470-2045(10)70010-3 [DOI] [PubMed] [Google Scholar]
- 5. Ueda T, Morioka H, Nishida Y, et al. Objective tumor response to denosumab in patients with giant cell tumor of bone: a multicenter phase II trial. Ann Oncol 2015;26:2149–54. 10.1093/annonc/mdv307 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Branstetter DG, Nelson SD, Manivel JC, et al. Denosumab induces tumor reduction and bone formation in patients with giant-cell tumor of bone. Clin Cancer Res 2012;18:4415–24. 10.1158/1078-0432.CCR-12-0578 [DOI] [PubMed] [Google Scholar]
- 7. Subbiah V, Anderson P, Rohren E. Alpha emitter radium 223 in high-risk osteosarcoma: first clinical evidence of response and blood-brain barrier penetration. JAMA Oncol 2015;1:253–5. 10.1001/jamaoncol.2014.289 [DOI] [PubMed] [Google Scholar]
- 8. Subbiah V, Anderson PM, Kairemo K, et al. Alpha particle Radium 223 dichloride in high-risk osteosarcoma: a phase I dose escalation trial. Clin Cancer Res 2019:clincanres.3964.2018 10.1158/1078-0432.CCR-18-3964 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Tsukamoto S, Righi A, Vanel D, et al. Development of high-grade osteosarcoma in a patient with recurrent giant cell tumor of the ischium while receiving treatment with denosumab. Jpn J Clin Oncol 2017;47:1090–6. 10.1093/jjco/hyx112 [DOI] [PubMed] [Google Scholar]
- 10. Park A, Cipriano CA, Hill K, et al. Malignant Transformation of a Giant Cell Tumor of Bone Treated with Denosumab: A Case Report. JBJS Case Connect 2016;6:e78 10.2106/JBJS.CC.16.00024 [DOI] [PubMed] [Google Scholar]
- 11. Aponte-Tinao LA, Piuzzi NS, Roitman P, et al. A high-grade sarcoma arising in a patient with recurrent benign giant cell tumor of the proximal tibia while receiving treatment with denosumab. Clin Orthop Relat Res 2015;473:3050–5. 10.1007/s11999-015-4249-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. O’Connor W, Quintana M, Smith S, et al. The hypermetabolic giant: 18F-FDG avid giant cell tumor identified on PET-CT. J Radiol Case Rep 2014;8:27–38. 10.3941/jrcr.v8i6.1328 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Costelloe CM, Chuang HH, Madewell JE. FDG PET/CT of primary bone tumors. AJR Am J Roentgenol 2014;202:W521–31. 10.2214/AJR.13.11833 [DOI] [PubMed] [Google Scholar]
- 14. Kairemo K, Rohren EM, Anderson PM, et al. Development of sodium fluoride PET response criteria for solid tumours (NAFCIST) in a clinical trial of radium-223 in osteosarcoma: from RECIST to PERCIST to NAFCIST. ESMO Open 2019;4:e000439 10.1136/esmoopen-2018-000439 [DOI] [PMC free article] [PubMed] [Google Scholar]