Current Treatment Options in Oncology (2014) 15:507–518

DOI 10.1007/s11864-014-0289-1

Sarcoma (SH Okuno, Section Editor)

Giant Cell Tumor of Bone:

Current Treatment Options
Keith M. Skubitz, MD
Address
Department of Medicine, University of Minnesota Medical School, and the
Masonic Cancer Center, Box 286University Hospital, Minneapolis, MN, USA
Email: [email protected]

Published online: 2 May 2014

* Springer Science+Business Media New York 2014

Keywords Giant cell tumor I GCTB I Bone I Denosumab I Bisphosphonate I Surgery I Radiation therapy I
Chemotherapy I Paracrine signaling I RANK I RANKL I Monoclonal antibody I Osteoclast

Opinion statement
Giant cell tumor of bone (GCTB) comprises up to 20 % of benign bone tumors in the US. GCTB
are typically locally aggressive, but metastasize to the lung in ~5 % of cases. Malignant trans-
formation occurs in a small percentage of cases, usually following radiation therapy.
Historically, GCTB have been treated primarily with surgery. When the morbidity of surgery
would be excessive, radiation therapy may achieve local control. In most cases the primary
driver of the malignant cell appears to be a mutation in H3F3A leading to a substitution of
Gly34 to either Trp or Leu in Histone H3.3. This change presumably alters the methylation
of the protein, and thus, its effect on gene expression. The malignant stromal cells of GCTB
secrete RANKL, which recruits osteoclast precursors to the tumor and stimulates their differ-
entiation to osteoclasts. The elucidation of the biology of GCTB led to trials of the anti-RANKL
monoclonal antibody denosumab in this disease, with a clear demonstration of beneficial clin-
ical effect. Surgery remains the primary treatment of localized GCTB. When surgery is not pos-
sible or would be associated with excessive morbidity, denosumab is a good treatment option.
The optimal length of treatment and schedule of denosumab is unknown, but recurrences after
apparent complete responses have been observed after stopping denosumab, and long-term
follow-up of denosumab treatment may reveal unrecognized effects. The role of denosumab in
the preoperative or adjuvant setting will require clinical trials. In some cases local radiation
therapy may be useful, although long term effects should be considered.

Introduction
Giant cell tumor of bone (GCTB), also known as and malignant GCTB are recognized. Transformation to
osteoclastoma, seems to have been first described by a high–grade sarcoma occurs rarely spontaneously, and
Cooper and Travers in 1818 [1], and more formally by with higher frequency after radiation therapy. The tumor
Bloodgood in 1912 [2]. GCTB comprises up to 20 % of is composed of both malignant stromal tumor cells, as
benign bone tumors in the US. GCTB typically occurs at well as osteoclasts and osteoclast precursor cells recruited
the ends of long bones and is locally aggressive, osteolytic, by RANKL secreted by the tumor cells (Fig. 1).
and often associated with pain, but metastases occur in Mutations in histone H3.3 are the most likely driv-
only about 5 % of cases [3–6, 7•, 8–12]. Both benign er mutation in this disease. The malignant stromal cells
508 Sarcoma (SH Okuno, Section Editor)

Fig. 1. Top, low power view of GCTB showing multiple giant cells
interspersed among malignant stromal cells and mononuclear
osteoclast precursors. Bottom, higher power view showing mul-
tinucleate osteoclast (giant) cells. The author holds the copyright
to these images.

of GCTB secrete RANKL, which recruits osteoclasts to the ized disease, surgery is the standard therapy. Strong clinical
tumor, forming the rationale for treatment with evidence supports the use of denosumab in advanced or
denosumab, a monoclonal antibody to RANKL. metastatic disease. Osteoclast like giant cells can be present
Changes in the malignant stromal cells following treat- in many other conditions, including reactive conditions
ment with denosumab suggests a paracrine loop between and other benign and malignant tumors [13], and only
the malignant cells and the osteoclasts. For primary local- GCTB is considered here.

Epidemiology
GCTB most commonly occurs in the epiphyses of long bones, but may occur
in other bones, and on rare occasions may be multi-centric. GCTB usually
occurs in skeletally mature patients between 20 and 40 years old, comprising
up to 20 % of benign bone tumors in the US [9, 10, 12]. The rate of GCTB is
slightly higher in women than men (1.5 to 1 ratio) [14], and may be higher
in China than the US [7•]. Familial clustering of GCTB and Paget’s disease
has been reported [15, 16]. In most cases the primary driver of the malignant
cell appears to be a mutation in H3F3A leading to a substitution of Gly34
to either Trp or Leu in Histone H3.3 [17••].

Pathogenesis
Bone undergoes constant remodeling by osteoclasts that dissolve bone, and
osteoblasts that form new bone. The osteoclasts develop from circulating
 Giant Cell Tumor of Bone Skubitz 509

osteoclast progenitor cells, first identified by the pioneering work of Walker

[18, 19]. This work provided the first demonstration of the existence of stem
cells in blood for nonhematopoietic tissue, and also paved the way for a
better understanding of the biology of GCTB. There is cross-talk between
osteoblasts and osteoclasts during normal bone physiology, with osteoblasts
producing receptor activator of nuclear factor kappa B ligand (RANKL)
(Fig. 2). Osteoclasts and their progenitors are dependent on RANKL and in its
absence undergo apoptosis. During normal bone remodeling, osteoclasts in
the cutting cone move through removing old bone, and are closely followed
by osteoblasts laying down new bone, suggesting signaling from the osteo-
clast to the osteoblast as well. GCTB express genes expressed by osteoclasts,
but also high levels of RANKL, that is produced by the stromal tumor cells,
which have characteristics of osteoblasts [20–25]. These observations led to
the hypothesis that elimination of osteoclasts by removal of RANKL could
both eliminate osteoclasts in the tumor, and also unknown factors produced
by the osteoclast that might have trophic effects on the stromal tumor cells
(Fig. 2). Consistent with this hypothesis, in vitro growth of isolated GCTB
stromal cells results in rapid loss of RANKL expression, suggesting that
RANKL expression is at least partly dependent on factors produced by other
cells [23]. Since a monoclonal antibody to RANKL was developed for use in
osteoporosis and cancer metastatic to bones, a drug was available to test this
hypothesis [11]. Clinical trials of denosumab in GCTB found that
denosumab treatment resulted in a marked reduction in giant cells, as ex-
pected, but also led to replacement of the spindle shaped dense stroma with
a less cellular stroma with new osteoid formation, providing further support
for this pathophysiology [26].
The transforming event in GCTB appears to be a mutation in H3F3A; ge-
nomic analysis has found mutations in H3F3A in 49 of 53 cases of GCTB
[17••]. All mutations involved glycine 54, with 48 encoding Gly34Trp and 1
encoding Gly34Leu alterations. In the same study, mutations in H3F3A or
H3F3B, that reside on different chromosomes but encode identical histone
H3.3 proteins, were found in chondroblastoma. In the chondroblastomas,
that like GCTB typically arise in the epiphysis, histone H3.3 mutations were
found in 73 of 77 cases. All 73 mutations involved lysine 36, with 68 in
H3F3B and 5 in H3F3A. In GCTB, these histone H3.3 mutations were found
exclusively in the stromal cell population, and not in the osteoclasts or os-
teoclast precursors [17••]. In contrast, histone H3.3 mutations were only
found in 2 of 103 osteosarcomas (1 in HF3A and 1 in H3F3B), and 1 of 75
conventional chondrosarcomas. H3F3A mutations are also commonly found
in pediatric brain tumors, but rarely in other cancers [17••]. The association
of each tumor type with a specific histone H3.3 mutation is not understood,
but it is possible that specific histone H3.3 mutations could result in a
growth advantage for specific cell lineages [17••].

Presentation
GCTB patients typically present with pain at the site of tumor, typically at the
end of a long bone. Tumors are usually radiolucent without sclerotic mar-
510 Sarcoma (SH Okuno, Section Editor)

Fig. 2. Panel A, the GCTB stromal cells produce RANKL that

recruits monocytic osteoclast precursors from blood to the
tumor, and stimulates differentiation into multinucleate os-
teoclast (“giant”) cells. Panel B, during bone remodeling
osteoclasts move through bone, followed closely by osteo-
blasts forming new bone. This suggests there may an inter-
action between osteoclasts and osteoblasts, possibly in which
osteoclasts produce a factor or factors that have a trophic
effect on osteoblasts. Panel C, it is possible that osteoclasts
produce a factor or factors that have a trophic effect on the
malignant stromal cells of GCTB. Removing RANKL with
denosumab would eliminate the osteoclasts, and concomi-
tantly any factors produced by the osteoclasts. Elimination of
osteoclast products could have an effect on the malignant
GCTB stromal cells. The author holds the copyright to these
images.

gins. Pathologic fractures are common, and axial tumors may present with
neurologic symptoms. Up to 5 % of patients present with metastatic disease,
usually in the lung.

Diagnosis
GCTB can often be suspected based on imaging with plain films, CT, or MRI, but
diagnosis requires a biopsy. In many cases, definitive surgery can be performed
at the time of the biopsy. Although the vast majority of GCTB have benign
histologic features, a small percentage present as a malignant GCTB with a more
aggressive cytologic appearance. Because osteoclast like giant cells can be present
in many other conditions, including reactive conditions and other benign and
malignant tumors [13], evaluation should be performed by a pathologist ex-
 Giant Cell Tumor of Bone Skubitz 511

perienced in this field to exclude other diagnoses such as giant cell rich osteo-
sarcoma, brown tumor of hyperparathyroidism, etc. Staging requires chest im-
aging with either a noncontrast chest CT or x-ray.
Prognosis
En block excision of the primary tumor has a recurrence rate of G20 %
whereas intralesional curettage has been reported to have much higher re-
currence rates. Metastatic disease does not usually respond well to chemo-
therapy and may cause death, although in many cases repeated surgical
excision may be beneficial [27].
Treatment
Depending on the extent of disease, the primary treatment of GCTB is sur-
gery. When surgery is not possible or would be associated with unacceptable
toxicity, treatment with denosumab or radiation therapy may be useful.

Surgical management of localized disease

For localized primary disease, surgery is standard treatment. En block exci-
sion of the primary tumor has a recurrence rate of G20 % whereas
intralesional curettage has been reported to result in recurrence rates up to
40 %–50 % in some series [8, 28]. In addition to intralesional curettage,
the cavity can be treated with various agents, such as phenol or cryotherapy
[3, 29–32] to try to decrease recurrence. The heat generated by polymeriza-
tion of methyl methacrylate cement may have a local antitumor effect as well
[6, 28, 33]. The risk of recurrence after primary surgery correlates with the sur-
gical treatment, and to a lesser extent with the Campanacci staging system
[8], where stage I tumors do not involve the bone cortex, stage II tumors im-
pinge on but do not penetrate the cortex, and stage III tumors extend to soft
tissue. Embolization may also be useful in some cases [34, 35].
Surgical management of metastatic disease
Metastatic disease in GCTB is uncommon and usually involves the lung.
Many of these patients can achieve long term remission or cure by resec-
tion of lung metastases, depending on the number of lesions and rate of
growth.

Adjunctive therapies
Pharmacologic
Denosumab
A phase II study of denosumab, a fully human monoclonal antibody to
RANKL, demonstrated a tumor response in 30/35 patients with GCTB [26].
Treatment resulted in the near complete elimination of giant cells in post-
treatment specimens relative to baseline, and post-treatment samples showed
512 Sarcoma (SH Okuno, Section Editor)

a replacement of the spindle-shaped dense stroma with a less cellular stroma

with new osteoid formation lined by RANKL-expressing cells [26]. The
changes in the immature state of the neoplastic stromal cells associated with
elimination of osteoclasts, provided confirmation for the hypothesis that the
neoplastic cells receive growth/differentiation signals from the osteoclasts. In
addition, clinical benefit as manifest by reduced pain and increased function
were observed [26]. Formal analysis by histology, of samples from pa-
tients in the initial study demonstrated both loss of giant cells, but also
in many cases a decrease in proliferative, densely cellular tumor stromal
cells, and replacement with nonproliferative, differentiated densely new
woven bone [26, 36•].
These striking results lead to a larger multinational phase II trial. Since
RECIST criteria are of limited use in bone lesions, lack of progression, and
loss of giant cells determined by histology were measures of efficacy in the
first trial of denosumab in GCTB [26], whereas EORTC and inverse Choi
criteria were also used in the second trial. An interim analysis of the second
phase II trial found 163/169 patients with “surgically unsalvageable” GCTB
had no disease progression at a median follow-up of 13 months [37••]. In
a second cohort of patients with surgically salvageable disease in whom
surgery was felt to be associated with severe morbidity, seventy-four out
of one-hundred patients had no surgery and 16/26 who had surgery had
a less morbid procedure than initially planned at a median follow-up of
9.2 months [37••]. This first interim analysis found a response rate of
~25 % by modified RECIST, and ~75 % by EORTC or inverse Choi criteria
[37••]. In addition to objective measures of tumor response, clinically rel-
evant pain reduction was reported by ~30 % of patients during the first
week of treatment, and by ~50 % of patients at each study visit from
months 2–30 [38•].
Serious adverse events were reported in 9 % of patients in this trial, and
denosumab was discontinued in 5 % of patients in this trial because of tox-
icity. Adverse events were as expected from larger trials of denosumab in met-
astatic cancer [39, 40, 41•, 42]. Serious adverse events leading to stopping
treatment included osteonecrosis of the jaw (ONJ) (1 %), hypocalcemia
(none serious) (5 %), hypophosphatemia (3 %), serious infection (2 %),
and new primary malignancy (1 %); the most common grade 3–4 adverse
events included hypophosphatemia (3 %), anemia, back pain, and pain in
the extremities, each occurring in 1 % of patients [37••].
Denosumab is a good option for patients with unresectable disease or in
patients in whom surgery would be a very morbid procedure. However, in
these cases tumor control may require lifelong denosumab, and there is lim-
ited experience with long term treatment. The optimal schedule if long term
treatment is needed remains to be defined. In addition, patients should not
become pregnant while on denosumab. Data at present do not currently sug-
gest an increased risk of malignant transformation.
Use of denosumab in young patients whose growth plates have not yet
closed could result in changes similar to osteopetrosis. One case report of a
10-year-old girl with metastatic GCTB who responded well to denosumab,
noted the development of dense metaphyseal bands on x-ray and general-
ized increased bone mineral density, raising concern about long term ef-
fects in prepubertal children, although her growth velocity remained
 Giant Cell Tumor of Bone Skubitz 513

normal for her age and gender [43]. An earlier report described osteopetrosis in-
duced by treatment of a peripubertal child with bisphosphonates [44].

Bisphosphonates
Bisphosphonates have also been used in GCTB, although reports have been
limited to retrospective and uncontrolled series, and case reports [45, 46,
47•, 48]. One case report clearly demonstrated a response of a GCTB in
the cervical spine to i.v. zoledronic acid over a 3-year period [47•]. A ran-
domized phase II trial is ongoing to determine if adjuvant zoledronic acid
improves the 2 year recurrence rate of “high risk” GCTB compared with stan-
dard care (NCT00889590).
Questions remain about the optimal approach to unresectable or meta-
static GCTB. Normal bone is constantly undergoing resorption and replace-
ment, a process regulated by coupling of osteoclast and osteoblast activity.
Suppression of bone resorption is followed by an inhibition of bone produc-
tion within a few months by an unknown mechanism, presumably because
of signaling between osteoclasts and osteoblasts [49]. Both the production of
unknown factors by the osteoclasts and release of growth factors from the
bone matrix, which is known to contain osteoblast growth factors, may be
involved on osteoclast signaling to the osteoblast [49]. The Wnt/beta-catenin
signaling pathway plays an important role in regulating bone formation [49–
51]. For example, the Wnt pathway regulates osteoblast differentiation, mat-
uration, and number, and this signaling is inhibited by factors, including
sclerostin and Dickkopf-1 (DKK1), that bind Wnt co-receptors [49–51].
Sclerostin is primarily expressed in osteocytes and expression is decreased
by mechanical loading or parathyroid hormone, which may promote bone
formation [49, 52, 53]. Although DKK1 is widely expressed in embryonic
mice, it is expressed primarily in osteoblasts and maturing osteocytes in
adults [49, 54], although it is expressed in some pathologic states such as my-
eloma and rheumatoid arthritis [49]. Denosumab treatment of postmeno-
pausal osteoporosis resulted in increased serum sclerostin and decreased
DKK1 [49]. Chronic bisphosphonate treatment has also been reported to in-
crease serum sclerostin [49, 55]. Thus, the decrease in bone formation after
some months of bisphosphonate treatment could in part be because of an
increase in sclerostin [49].
In contrast to sclerostin, denosumab treatment resulted in a decrease in se-
rum DKK1 whereas bisphosphonate treatment did not [49]. Bisphosphonates
act mostly on mature osteoclasts as they degrade matrix, and leads to an in-
crease in osteoclast precursors [49, 56], whereas denosumab treatment also
blocks recruitment and differentiation of osteoclast precursors. It has been hy-
pothesized that with denosumab, the increase in sclerostin would tend to de-
crease bone formation, whereas the decrease in DKK1 may limit this effect on
bone formation [49].
An important toxicity of bisphosphonates is osteonecrosis of the jaw
(ONJ), and this is also a toxicity of denosumab [40, 42, 57•]. The risk of
ONJ is time dependent. In larger trials of denosumab in other malignan-
cies the risk of ONJ was about ~1 % at 1 year and ~4 % after 3 years
[57•]. ONJ was much more frequent with bisphosphonates before rou-
tine preventive approaches were used. Finally, given the reported lym-
514 Sarcoma (SH Okuno, Section Editor)

phoid defects in mice that lack RANKL [58], other long term toxicities of
denosumab may exist.
Thus, even though the results of trials of denosumab in GCTB are impres-
sive, consideration should be given to a randomized trial of denosumab,
which is easier to give and has no recognized nephrotoxicity, vs a bisphos-
phonate, which is currently cheaper, in select GCTB cases. A randomized trial
of denosumab compared with bisphosphonates that allows cross-over would
answer the question of which approach would be more effective for GCTB
not amenable to surgical treatment.

Cytotoxic chemotherapy
A number of reports have described the use of cytotoxic chemotherapy and
interferon in GCTB [59–63]. Given the typically benign nature of GCTB
and the toxicities of these agents, this approach has been replaced by the
use of denosumab in all but rare cases.

Radiation therapy
There is limited data on the use of radiation therapy in GCTB, but in select
cases it may be useful. Although malignant transformation of GCTB can oc-
cur with reported rates of 1.5 %–15 %, most are associated with previous ra-
diation therapy [64]. In 1 series, 26/407 GCTBs were malignant (19 after
initial treatment and 7 at diagnosis); of the 19 malignant GCTB in this series,
18 (~95 %) had received radiation therapy 4–39 years earlier [64]. Rates of
malignant transformation after radiation therapy of between 7 %–33 % have
been reported [64], and the observed rates could be influenced by length of
follow-up. More recent reports using more modern techniques confirm that
radiation induced sarcoma remains an issue [65]. In a study of 25 patients
between 1956–2000, 2 developed osteosarcoma at 11–12 years, with a me-
dian follow-up of 8.8 years (0.67–34 years), and a disease free survival rate of
58 % [65]. A retrospective review of 26 tumors in 24 patients between 1972–
1996 reported local control in 20/26 cases and 1 radiation induced sarcoma
22 years after treatment [66]. In a series of patients deemed poor candidates
for surgery and treated with megavoltage radiotherapy between 1985–2007
with a median follow-up time of 58 months, local control was achieved in
65/77 and 12/77 progressed in the radiated field; 2 developed malignant
transformation [67]. In another recent study of 34 patients treated with
megavoltage therapy (13 after gross total surgical resection) between
1973–2008, and a median follow-up of 16.8 years, local control was report-
ed in ~85 % at 5 years, with 3 developing lung metastases and 1 malignant
transformation [68]. It is possible that IMRT may offer an advantage in some
cases; 1 report observed local control in 4/5 cases at a median follow-up of
46 months [69].
In addition to the effect of follow-up time, many reports of malignant
transformation involve relatively small numbers of patients. For instance,
if a study of 34 patients with “adequate” follow-up had an observed event
rate of 1/34 or 3 %, the 95 % confidence interval (CI) of the rate would
be ~G1 %–9 %, and with an event rate of 2/34 or 5 % the 95 % CI would
be ~G1 %–14 %. In a study of 77 patients with an event rate of 2/77 or
 Giant Cell Tumor of Bone Skubitz 515

2.5 %, the 95 % CI would be ~G1 %–6 %, and with an event rate of 3/77 or
4 %, the 95 % CI would be ~G1 %–8 %.

Diet and lifestyle

There is no evidence to support any role for diet and lifestyle changes in the
treatment of GCTB.

Compliance with Ethics Guidelines

Conflict of Interest
Keith Skubitz has been a consultant for Amgen, Ariad/Merck, Novartis, Onyxx, Johnson & Johnson, Pfizer/
Schering-Plough, Systems Medicine, and Seattle Genetics; owns publicly traded stock in Johnson & John-
son; has received research funding from Amgen, Novartis, GSK, Ariad/Merck, Celgene, Cell Therapeutics,
Systems Medicine, Infinity, Schering-Plough, Bayer, Pfizer, and Daiichi; and provided expert testimony
on the role of bisphosphonates in osteonecrosis of the jaw.

Human and Animal Rights and Informed Consent

This article does not contain any studies with human or animal subjects performed by any of the authors
other than reviews of other studies.

References and Recommended Reading

Papers of particular interest, published recently, have been
highlighted as:
• Of importance
•• Of major importance
1. Cooper A, Travers B. On exostosis. In: Surgical essays. followed for a median time of 5 years. Acta Orthop.
London: Cox and Sons; 1818. p. 167–224. 2008;79:86–93.
2. Bloodgood II JC. The conservative treatment of giant- 7.• Niu X, Zhang Q, Hao L, Ding Y, Li Y, Xu H, et al.
cell sarcoma, with the study of bone transplantation. Giant cell tumor of the extremity: retrospective
Ann Surg. 1912;56:210–39. analysis of 621 Chinese patients from one institu-
3. Balke M, Schremper L, Gebert C, Ahrens H, tion. J Bone Joint Surg Am. 2012;94:461–7.
Streitbuerger A, Koehler G, et al. Giant cell tumor of This retrospective study of 621 patients provides epidemio-
bone: treatment and outcome of 214 cases. J Cancer logic and outcome data on a large number of Chinese pa-
Res Clin Oncol. 2008;134:969–78. tients
4. Errani C, Ruggieri P, Asenzio MA, Toscano A, 8. Campanacci M, Baldini N, Boriani S, Sudanese A.
Colangeli S, Rimondi E, et al. Giant cell tumor of the Giant-cell tumor of bone. J Bone Joint Surg Am.
extremity: a review of 349 cases from a single insti- 1987;69:106–14.
tution. Cancer Treat Res. 2010;36:1–7.
9. Raskin KA, Schwab JH, Mankin HJ, Springfield DS,
5. Gupta R, Seethalakshmi V, Jambhekar NA, Hornicek FJ. Giant cell tumor of bone. J Am Acad
Prabhudesai S, Merchant N, Puri A, et al. Clinico-
Orthop Surg. 2013;21:118–26.
pathologic profile of 470 giant cell tumors of bone
from a cancer hospital in western India. Ann Diagn 10. Szendroi M. Giant-cell tumor of bone. J Bone Joint
Pathol. 2008;12:239–48. Surg (Br). 2004;86:5–12.
6. Kivioja AH, Blomqvist C, Hietaniemi K, Trovik C, 11. Thomas DM, Skubitz KM. Giant cell tumor of bone.
Walloe A, Bauer HC, et al. Cement is recommended Curr Opin Oncol. 2009;21:338–44.
in intralesional surgery of giant cell tumors: a Scan- 12. WHO. Pathology and genetics of tumors of soft tis-
dinavian sarcoma group study of 294 patients sue and bone. Lyon: IARC Press; 2002.
516 Sarcoma (SH Okuno, Section Editor)

13. Skubitz KM, Manivel JC. Giant cell tumor of the nuclear factor kappa b) and rank ligand are expressed
uterus: case report and response to chemotherapy. in giant cell tumors of bone. Am J Clin Pathol.
BMC Cancer. 2007;7. 2002;117:210–6.
14. Zheng MH, Robbins P, Xu J, Huang L, Wood DJ, 25. Skubitz KM, Cheng EY, Clohisy DR, Thompson RC,
Papadimitriou JM. The histogenesis of giant cell tu- Skubitz AP. Gene expression in giant-cell tumors. J
mor of bone: a model of interaction between neo- Lab Clin Med. 2004;144:193–200.
plastic cells and osteoclasts. Histol Histopathol. 26. Thomas D, Henshaw R, Skubitz K, Chawla S,
2001;16:297–307. Staddon A, Blay JY, et al. Denosumab in patients
15. Jacobs TP, Michelsen J, Polay JS, D'Adamo AC, with giant-cell tumor of bone: an open-label, phase 2
Canfield RE. Giant cell tumor in Paget's disease of study. Lancet Oncol. 2010;11:275–80.
bone: familial and geographic clustering. Cancer. 27. Siebenrock KA, Unni KK, Rock MG. Giant-cell tumor
1979;44:742–7. of bone metastasizing to the lungs. A long-term fol-
16. Rendina D, Mossetti G, Soscia E, Sirignano C, low-up. J Bone Joint Surg (Br). 1998;80:43–7.
Insabato L, Viceconti R, et al. Giant cell tumor and 28. Becker WT, Dohle J, Bernd L, Braun A, Cserhati M,
Paget's disease of bone in one family: geographic Enderle A, et al. Local recurrence of giant cell tumor
clustering. Clin Orthop Relat Res. 2004;421:218–24. of bone after intralesional treatment with and with-
17.•• Behjati S, Tarpey PS, Presneau N, Scheipl S, Pillay N, out adjuvant therapy. J Bone Joint Surg Am.
Van Loo P, et al. Distinct h3f3a and h3f3b driver 2008;90:1060–7.
mutations define chondroblastoma and giant cell 29. Durr HR, Maier M, Jansson V, Baur A, Refior HJ.
tumor of bone. Nat Genet. 2013;45:1479–82. Phenol as an adjuvant for local control in the treat-
This important paper identifies the likely driver mutations ment of giant cell tumor of the bone. Eur J Surg
that result in GCTB and chondroblastoma. The transforming Oncol. 1999;25:610–8.
event in GCTB appears to be a mutation in H3F3A (found in 30. Malawer MM, Bickels J, Meller I, Buch RG, Henshaw
49 of 53 cases). In the same study, mutations in H3F3A or RM, Kollender Y. Cryosurgery in the treatment of
H3F3B, that reside on different chromosomes but encode giant cell tumor. A long-term follow-up study. Clin
identical histone H3.3 proteins, were found in Orthop Relat Res. 1999;359:176–88.
chondroblastoma (histone H3.3 mutations were found in 73 31. Zhen W, Yaotian H, Songjian L, Ge L, Qingliang W.
of 77 cases, with 68 in H3F3B and 5 in H3F3A) Giant-cell tumor of bone. The long-term results of
18. Walker DG. Congenital osteopetrosis in mice cured treatment by curettage and bone graft. J Bone Joint
by parabiotic union with normal siblings. Endocri- Surg (Br). 2004;86:212–6.
nology. 1972;91:916–20. 32. Zwolak P, Manivel JC, Jasinski P, Kirstein MN, Dudek
19. Walker DG. Control of bone resorption by hemato- AZ, Fisher J, et al. Cytotoxic effect of zoledronic acid-
poietic tissue. The induction and reversal of con- loaded bone cement on giant cell tumor, multiple
genital osteopetrosis in mice through use of bone myeloma, and renal cell carcinoma cell lines. J Bone
marrow and splenic transplants. J Exp Med. Joint Surg Am. 2010;92:162–8.
1975;142:651–63. 33. Klenke FM, Wenger DE, Inwards CY, Rose PS, Sim
20. Atkins GJ, Haynes DR, Graves SE, Evdokiou A, Hay S, FH. Recurrent giant cell tumor of long bones: anal-
Bouralexis S, et al. Expression of osteoclast differen- ysis of surgical management. Clin Orthop Relat Res.
tiation signals by stromal elements of giant cell tu- 2011;469:1181–7.
mors. J Bone Miner Res. 2000;15:640–9. 34. Hosalkar HS, Jones KJ, King JJ, Lackman RD. Serial
21. Goldring SR, Roelke MS, Petrison KK, Bhan AK. arterial embolization for large sacral giant-cell tu-
Human giant cell tumors of bone identification and mors: mid- to long-term results. Spine.
characterization of cell types. J Clin Invest. 2007;32:1107–15.
1987;79:483–91. 35. Lin PP, Guzel VB, Moura MF, Wallace S, Benjamin
22. Huang L, Xu J, Wood DJ, Zheng MH. Gene expres- RS, Weber KL, et al. Long-term follow-up of patients
sion of osteoprotegerin ligand, osteoprotegerin, and with giant cell tumor of the sacrum treated with se-
receptor activator of nf-kappab in giant cell tumor of lective arterial embolization. Cancer. 2002;95:1317–
bone: possible involvement in tumor cell-induced 25.
osteoclast-like cell formation. Am J Pathol. 36.• Branstetter DG, Nelson SD, Manivel JC, Blay JY,
2000;156:761–7. Chawla S, Thomas DM, et al. Denosumab induces
23. Morgan T, Atkins GJ, Trivett MK, Johnson SA, tumor reduction and bone formation in patients
Kansara M, Schlicht SL, et al. Molecular profiling of with giant-cell tumor of bone. Clin Cancer Res.
giant cell tumor of bone and the osteoclastic locali- 2012;18:4415–24.
zation of ligand for receptor activator of nuclear This study describes pathologic changes in GCTB after
factor kappab. Am J Pathol. 2005;167:117–28. treatment with denosumab
24. Roux S, Amazit L, Meduri G, Guiochon-Mantel A, 37.•• Chawla S, Henshaw R, Seeger L, Choy E, Blay JY,
Milgrom E, Mariette X. Rank (receptor activator of Ferrari S, et al. Safety and efficacy of denosumab for
 Giant Cell Tumor of Bone Skubitz 517

adults and skeletally mature adolescents with giant cervical spine with bisphosphonate as single therapy.
cell tumor of bone: interim analysis of an open-label, Spine. 2012;37:E396–9. This case report clearly
parallel-group, phase 2 study. Lancet Oncol. demonstrates a response of GCTB to zoledronic acid
2013;14:901–8. over a 3-year period.
This study provides data on the use of denosumab in a large This case report clearly demonstrates a response of GCTB to
number of patients with GCTB from a phase II trial. The data zoledronic acid over a 3-year period
confirm the results of the earlier proof-of-concept trial, and 48. Tse LF, Wong KC, Kumta SM, Huang L, Chow TC,
document the efficacy of denosumab in GCTB Griffith JF. Bisphosphonates reduce local recurrence
38.• Broto JM, Cleeland C, Glare P, Engellau J, Skubitz K, in extremity giant cell tumor of bone: a case-control
Blum R, et al. Effects of denosumab on pain and study. Bone. 2008;42:68–73.
analgesic use in giant cell tumor of bone: interim 49. Gatti D, Viapiana O, Fracassi E, Idolazzi L, Dartizio
results from a phase II study. Acta Oncol. 2014; [In C, Povino MR, et al. Sclerostin and dkk1 in post-
press]. menopausal osteoporosis treated with denosumab. J
This study describes the effect of denosumab treatment on Bone Miner Res. 2012;27:2259–63.
pain in GCTB in a large number of patients from a phase II 50. Baron R, Rawadi G. Targeting the wnt/beta-catenin
trial, documenting rapid and prolonged improvement in pathway to regulate bone formation in the adult
pain. skeleton. Endocrinology. 2007;148:2635–43.
39. Fizazi K, Carducci M, Smith M, Damiao R, Brown J, 51. Ott SM. Sclerostin and wnt signaling—the pathway
Karsh L, et al. Denosumab versus zoledronic acid for to bone strength. J Clin Endocrinol Metab.
treatment of bone metastases in men with castration- 2005;90:6741–3.
resistant prostate cancer: a randomized, double-blind
study. Lancet. 2011;377:813–22. 52. Bellido T, Ali AA, Gubrij I, Plotkin LI, Fu Q, O'Brien CA,
et al. Chronic elevation of parathyroid hormone in
40. Henry DH, Costa L, Goldwasser F, Hirsh V, Hungria mice reduces expression of sclerostin by osteocytes: a
V, Prausova J, et al. Randomized, double-blind study novel mechanism for hormonal control of osteo-
of denosumab versus zoledronic acid in the treat- blastogenesis. Endocrinology. 2005;146:4577–83.
ment of bone metastases in patients with advanced
cancer (excluding breast and prostate cancer) or 53. Robling AG, Niziolek PJ, Baldridge LA, Condon KW,
multiple myeloma. J Clin Oncol. 2011;29:1125–32. Allen MR, Alam I, et al. Mechanical stimulation of
bone in vivo reduces osteocyte expression of sost/
41.• Lipton A, Fizazi K, Stopeck AT, Henry DH, Brown JE, sclerostin. J Biol Chem. 2008;283:5866–75.
Yardley DA, et al. Superiority of denosumab to zo-
ledronic acid for prevention of skeletal-related 54. Li J, Sarosi I, Yan XQ, Morony S, Capparelli C, Tan
events: a combined analysis of 3 pivotal, random- HL, et al. Rank is the intrinsic hematopoietic cell
ized, phase 3 trials. Eur J Cancer. 2012;48:3082–92. surface receptor that controls osteoclastogenesis and
This analysis from 3 large randomized trials provides data on regulation of bone mass and calcium metabolism.
skeletal events and adverse events with denosumab com- Proc Natl Acad Sci U S A. 2000;97:1566–71.
pared with zoledronic acid in a large number of patients 55. Gatti D, Viapiana O, Idolazzi L, Fracassi E, Rossini M,
42. Stopeck AT, Lipton A, Body JJ, Steger GG, Tonkin K, Adami S. The waning of teriparatide effect on bone
de Boer RH, et al. Denosumab compared with zole- formation markers in postmenopausal osteoporosis
dronic acid for the treatment of bone metastases in is associated with increasing serum levels of dkk1. J
patients with advanced breast cancer: a randomized, Clin Endocrinol Metab. 2011;96:1555–9.
double-blind study. J Clin Oncol. 2010;28:5132–9. 56. Weinstein RS, Roberson PK, Manolagas SC. Giant
43. Karras NA, Polgreen LE, Ogilvie C, Manivel JC, osteoclast formation and long-term oral bisphos-
Skubitz KM, Lipsitz E. Denosumab treatment of phonate therapy. N Engl J Med. 2009;360:53–62.
metastatic giant-cell tumor of bone in a 10-year-old 57.• Smith MR, Saad F, Coleman R, Shore N, Fizazi K,
girl. J Clin Oncol. 2013;31:e200–2. Tombal B, et al. Denosumab and bone-metastasis-
44. Whyte MP, Wenkert D, Clements KL, McAlister WH, free survival in men with castration-resistant prostate
Mumm. Bisphosphonate-induced osteopetrosis. N cancer: results of a phase 3, randomized, placebo-
Engl J Med. 2003;349:457–63. controlled trial. Lancet. 2012;379:39–46.
45. Balke M, Campanacci L, Gebert C, Picci P, Gibbons This large randomized trial of denosumab provides data on
M, Taylor R, et al. Bisphosphonate treatment of ag- control of skeletal events and also adverse events
gressive primary, recurrent and metastatic giant cell 58. Kim N, Odgren PR, Kim DK, Marks Jr SC, Choi Y.
tumor of bone. BMC Cancer. 2010;10. Diverse roles of the tumor necrosis factor family
46. Chaudhary P, Khadim H, Gajra A, Damron T, Shah member trance in skeletal physiology revealed by
C. Bisphosphonate therapy is effective in the treat- trance deficiency and partial rescue by a lymphocyte-
ment of sacral giant cell tumor. Onkologie. expressed trance transgene. Proc Natl Acad Sci U S A.
2011;34:702–4. 2000;97:10905–10.
47.• Gille O, Oliveira Bde A, Guerin P, Lepreux S, Richez 59. Dominkus M, Ruggieri P, Bertoni F, Briccoli A, Picci
C, Vital JM. Regression of giant cell tumor of the P, Rocca M, et al. Histologically verified lung metas-
518 Sarcoma (SH Okuno, Section Editor)

tases in benign giant cell tumors—14 cases from a nineteen patients. J Bone Joint Surg Am.
single institution. Int Orthop. 2006;30:499–504. 1986;68:1073–9.
60. Kaiser U, Neumann K, Havemann K. Generalised 65. Caudell JJ, Ballo MT, Zagars GK, Lewis VO, Weber KL,
giant-cell tumor of bone: successful treatment of Lin PP, et al. Radiotherapy in the management of
pulmonary metastases with interferon alpha, a giant cell tumor of bone. Int J Radiat Oncol Biol
case report. J Cancer Res Clin Oncol. Phys. 2003;571:158–65.
1993;119:301–3. 66. Feigenberg SJ, Marcus Jr RB, Zlotecki RA, Scarbor-
61. Maloney WJ, Vaughan LM, Jones HH, Ross J, Nagel ough MT, Berrey BH, Enneking WF. Radiation ther-
DA. Benign metastasizing giant-cell tumor of bone. apy for giant cell tumors of bone. Clin Orthop Relat
Report of three cases and review of the literature. Clin Res. 2003;411:207–16.
Orthop Relat Res. 1989;243:208–15. 67. Ruka W, Rutkowski P, Morysinski T, Nowecki Z,
62. Osaka S, Toriyama M, Taira K, Sano S, Saotome K. Zdzienicki M, Makula D, et al. The megavoltage ra-
Analysis of giant cell tumor of bone with pul- diation therapy in treatment of patients with ad-
monary metastases. Clin Orthop Relat Res. vanced or difficult giant cell tumors of bone. Int J
1997;335:253–61. Radiat Oncol Biol Phys. 2010;78:494–8.
63. Stewart DJ, Belanger R, Benjamin RS. Prolonged 68. Shi W, Indelicato DJ, Reith J, Smith KB, Morris CG,
disease-free survival following surgical debulking Scarborough MT, et al. Radiotherapy in the man-
and high-dose cisplatin/doxorubicin in a patient agement of giant cell tumor of bone. Am J Clin
with bulky metastases from giant cell tumor of bone Oncol. 2013;36:505–8.
refractory to "standard" chemotherapy. Am J Clin 69. Roeder F, Timke C, Zwicker F, Thieke C, Bischof M,
Oncol. 1995;18:144–8. Debus J, et al. Intensity modulated radiotherapy
64. Rock MG, Sim FH, Unni KK, Witrak GA, Frassica FJ, (IMRT) in benign giant cell tumors—a single insti-
Schray MF, et al. Secondary malignant giant-cell tu- tution case series and a short review of the literature.
mor of bone. Clinicopathological assessment of Radiat Oncol. 2010;5:18.