You are on page 1of 8

Clin Orthop Relat Res (2018) 476:1783-1790

DOI 10.1007/s11999.0000000000000104

2017 International Society of Limb Salvage Proceedings

Preoperative Denosumab With Curettage and Cryotherapy in

Giant Cell Tumor of Bone: Is There an Increased Risk of
Local Recurrence?
Guido Scoccianti MD, Francesca Totti MD, Maurizio Scorianz MD, Giacomo Baldi MD, Giuliana Roselli MD,
Giovanni Beltrami MD, Alessandro Franchi MD, Rodolfo Capanna MD, Domenico Andrea Campanacci MD

Received: 6 August 2017 / revised: 9 October 2017 / Accepted: 22 November 2017 / Published online: 9 February
2018 Copyright © 2018 by the Association of Bone and Joint Surgeons

Background Denosumab is a monoclonal RANKL anti-body, (curettage) rather than a resection. However, few studies
which was originally introduced for the treatment of are available evaluating the benefits and risks of
osteoporosis and bone metastases from solid tumors, but more denosumab for the latter indication.
recently has been used for treatment of giant cell tumor of bone Questions/purposes (1) Does preoperative treatment
(GCTB). In GCTB, denosumab has been used as a sin-gle with denosumab reduce the risk of local recurrence in
agent in patients with inoperable tumors; it also has been used patients treated for GCTB? (2) Are there adverse effects
before surgery in some patients with the aim to down-stage the of short-term denosumab use before surgery and, if so,
tumor to facilitate a joint-preserving procedure what are they?

One of the authors (RC) received personal fees from Waldemar Link (Hamburg, Germany) and grants from Adler Ortho SRL
(Milan, Italy), outside the submitted work.
All ICMJE Conflict of Interest Forms for authors and Clinical Orthopaedics and Related Research® editors and board
members are on file with the publication and can be viewed on request.
Clinical Orthopaedics and Related Research® neither advocates nor endorses the use of any treatment, drug, or device. Readers are
encouraged to always seek additional information, including FDA approval status, of any drug or device before clinical use.
Each author certifies that his or her institution approved or waived approval for the human protocol for this investigation and
that all investigations were conducted in conformity with ethical principles of research.
This work was performed at the Department of Orthopaedic Oncology and Reconstructive Surgery, Azienda Ospedaliero
Universitaria Careggi, Firenze, Italy.

G. Scoccianti, F. Totti, M. Scorianz, D. A. Campanacci, Department of Orthopaedic Oncology and Reconstructive

Surgery, Azienda Ospedaliero Universitaria Careggi, Firenze, Italy

G. Baldi, Department of Medical Oncology, Hospital of Prato, Prato, Italy

G. Roselli, Department of Radiology, Azienda Ospedaliero Universitaria Careggi, Firenze, Italy

G. Beltrami, Department of Paediatric Orthopaedic Oncology, Azienda Ospedaliero Universitaria Careggi, Firenze, Italy

A. Franchi, Department of Translational Research and New Technologies, Department of Anatomic Pathology, Azienda
Ospedaliero Universitaria Pisana, Pisa, Italy

R. Capanna, Department of Orthopaedics and Traumatology, Azienda Ospedaliero Universitaria Pisana, Pisa, Italy

G. Scoccianti (✉), Department of Orthopaedic Oncology and Reconstructive Surgery, Azienda Ospedaliero Universitaria
Careggi, Largo Brambilla 1, 50134, Firenze, Italy, email:

Copyright 2018 by the Association of Bone and Joint Surgeons. Unauthorized reproduction of this article is prohibited.
1784 Scoccianti et al. Clinical Orthopaedics and Related Research®

Methods All patients with a diagnosis of GCTB surgically presentation. We did not observe any adverse effects with
treated at our institution from June 2009 to June 2016 with denosumab, but we caution readers that this study was
curettage and cryotherapy were retrospectively evaluated to underpowered to detect even relatively common compli-
compare patients treated with curettage alone versus cations and relatively large differences in the risk of local
patients treated with curettage after preoperative therapy recurrence. Future studies should evaluate denosumab
with denosumab. During that period, we treated 97 patients prospectively; given the relative rarity of this tumor, we
for GCTB; 30 patients were excluded because they received suspect multicenter studies are needed to achieve this.
a resection; 34 patients were excluded because they Level of Evidence Level III, therapeutic study.
received curettage without cryotherapy. Of the remaining
33 patients, four were excluded because they received
denosumab only after surgery, one because she received
zoledronic acid, one because she received a curettage after Introduction
her refusal of a re-section that was the advised procedure,
two because they were lost to followup early, and four Denosumab, a monoclonal antibody that acts as a receptor
because they were treated for recurrence rather than a new inhibitor of nuclear factor-kb ligand (RANKL), origi-nally
diagnosis of GCTB. The remaining 21 patients were was used to treat osteoporosis and bone metastases from
included. Twelve lesions had been treated with surgery after solid tumors. Because it acts to reduce the formation and
denosumab and nine with surgery alone. During the study survival of osteoclasts, denosumab has more recently been
period, we preferentially used denosumab for the more introduced in the treatment of patients with giant cell tumor
aggressive-looking lesions. After curettage, cryotherapy of of bone (GCTB) [4, 21, 22]. Because of its relative novelty,
the residual bone walls was performed with argon in this setting, denosumab generally is used (1) as the only
cryoprobes to -150° C after pouring gel into the cavity, and treatment in patients with locally advanced in-operable
we then used cement (17 patients) or morcellized allograft tumors or in those with metastatic tumors; (2) as
(four patients). Tumors were Campa-nacci Grade 3 in eight preoperative treatment to facilitate surgery; or (3) in cer-tain
of 12 patients in the denosumab group and in two of nine patients with large juxtaarticular tumors to make
patients in the surgery-only group (p = 0.08), but the extent intralesional therapy (like curettage) possible instead of
of epiphyseal juxtaarticular bone involvement was not osteoarticular resection [8, 15, 17, 19, 21, 23].
different between the groups with the numbers available. After the first proposals of the use of denosumab in
Median followup was 39 months (range, 14-55 months) in GCTB [4, 21, 22], many articles were published ad-
the denosumab group and 27 months (range, 18-92 months) dressing the histopathologic effects of denosumab sug-
in the surgery-only group. We used chart review to record gesting that denosumab does not kill the neoplastic
the proportion of patients in each treatment group who had stromal cells but rather inhibits their activity, forcing
a local recurrence and to tally adverse events. them to a quiescent status. By contrast, giant cells, which
Results With the numbers available, there was no difference in are not the real tumor cells, disappear and bone regrowth
the proportion of patients experiencing a recurrence (five of 12 is made possible around and also inside the lesion area
in the denosumab group and one of nine in the surgery-only [5, 11, 13, 14, 18, 20, 25]. However, few studies are
group; p = 0.18). We found no adverse effects associated with available concerning clinical results of this treatment in
denosumab either during or after treatment; specifically, we terms of recurrence rate [8, 15, 17, 23], especially when
found no alterations in electrolyte levels, blood count, or liver denosumab is used as neoadjuvant therapy before sur-
and renal function parameters. In this small series, no patient gical intralesional procedures.
has developed osteonecrosis of the jaw. Therefore, we decided to evaluate our series of patients
Conclusions In this small series, use of denosumab before treated with preoperative denosumab and surgical curettage
surgery for GCTB appeared to allow the reforming of a to try to answer the following questions: (1) Does
bone peripheral rim around the tumor, perhaps facilitating preoperative treatment with denosumab reduce the risk of
curettage rather than osteoarticular resection in some local recurrence? (2) Are there adverse effects of short-term
patients. However, we did not observe a decrease in the risk denosumab use before surgery and, if so, what are they?
of local recurrence with the use of denosumab, suggesting
that it may not decrease the aggressiveness of the disease;
according to our preliminary results, we cannot exclude that
the rate of local recurrence could be even higher after Patients and Methods
curettage in denosumab-treated patients than in nontreated
patients, and until or unless larger studies demonstrate such Between June 2009 and June 2016, we treated 97 patients
a reduction, primary intralesional surgery without deno- for GCTB at one institution. For this study, we wished to
sumab seems more prudent when curettage is feasible at focus on more aggressive and/or juxtaarticular lesions,

Copyright 2018 by the Association of Bone and Joint Surgeons. Unauthorized reproduction of this article is prohibited.
Volume 476, Number 9 Denosumab in Giant Cell Tumor of Bone 1785

which we treated with curettage and cryotherapy and which the subchondral bone, in whom the surgeon was
made up 34% (33 of 97) of the patients with GCTB we concerned about risks to the joint surface with curettage.
treated during that time. Starting in 2012, we began to use No definite criteria exist to guide between resection
denosumab before surgery in patients who we felt were at and curettage for patients with GCTB. Therefore, an
particularly high risk of recurrence. The current study element of subjectivity necessarily remains affecting
sought to evaluate the efficacy of that treatment, comparing surgeon choice and the readers must be aware of this
the group of patients treated only with surgery with the when analyzing and comparing the results we present.
group that received preoperative denosumab. In the 97 patients of the whole series, we used deno-
Because the aim of our study was to evaluate the results sumab preoperatively in 23 patients; 16 of them received
in terms of local recurrences of joint-sparing surgery, we a curettage, one received excision of a soft tissue re-
excluded 30 patients because they were treated with a re- currence, and six received a resection despite denosumab
section procedure (24 without preoperative denosumab and treatment. In the patients who were treated with a
six with preoperative denosumab). Thirty-four patients resection, the lesions involved the proximal fibula in
received a curettage, but they were excluded because they three patients and the proximal radius, distal femur, and
did not receive intraoperative cryotherapy. Four more proximal tibia in one patient each.
patients were excluded because they had received deno- All the patients were evaluated before surgery with
sumab only after surgery and one because she had received radiographs of the GCTB lesion, MRI with gadolinium,
zoledronic acid. Of the remaining 28 patients, 24 presented and CT scanning of the lesion and chest. In patients pre-
with a primary lesion and four with a recurrent lesion. operatively treated with denosumab, restaging with local
Because we decided to exclude also the patients treated MRI and CT was performed after the end of neoadjuvant
because of recurrent lesions, these four patients also were therapy, before the surgical procedure. All the tumors
excluded. Two more patients (both treated only with sur- were active Stage 2 or aggressive Stage 3 lesions
gery and cryotherapy) were excluded because they were lost according to the Campanacci classification [7].
to followup soon after surgery. One more patient (a young Along with the use of Campanacci grading, with the aim
woman affected by aggressive GCTB of the distal radius) to compare the two groups of patients including different
was excluded because, after denosumab, the CT tumoral sites, ranging from the femur to segments as small
modifications in tumor characteristics were considered not as a finger phalanx, we decided to evaluate the proportion of
sufficient to perform a curettage confidently and it was epiphyseal bone involvement at presentation, setting as an
decided that the patient should undergo resection of the evaluation parameter the ratio between the maximum di-
distal radius, but the patient refused this kind of surgery and ameter of the lesion and the maximum diameter of the
asked for joint-preserving surgery. In this patient, local epiphyseal bone at the level of subchondral bone or at the
recurrence was particularly likely to occur because curet- level of the lesion most near to the subchondral bone mea-
tage could not guarantee adequate tumor removal and we sured on axial two-dimensional CT scan images. For axial
therefore excluded the patient from the series to avoid bias or girdle lesions (sacrum, pelvis), the evaluation was
in the rate of local recurrences. conducted similarly, defining as diameter of the involved
Twenty-one patients remained for evaluation: 12 patients bone the AP length of the sacral wing and the maximum
had surgery after receiving denosumab and nine patients were width of the iliac wing at the involved level. No volumetric
treated with surgery alone. Cryotherapy was used in evaluation of the tumor was accomplished to analyze tumor
juxtaarticular lesions with involvement of subchondral bone size mod-ifications before and after denosumab treatment,
when “aggressive” surgical curettage was not possible without because it was not the purpose of the study. According to
damaging the joint or it was considered at high risk for being this param-eter, all patients were subdivided into three
inadequate; cryotherapy was adopted also in non-juxtaarticular groups: 76% to 100% involvement, 50% to 75%
lesions in case of wide involvement of the bone segment both involvement, or < 50% involvement.
for its adjuvant effect and to decrease bleeding in locations Denosumab treatment was administered at the
unfit for tourniquet use such as the pelvis or shoulder girdle. medical oncology department by a medical oncologist
The decision to use denosumab during that period for this (G. Baldi) according to the following schedule: 120-mg
indication was generally driven by the ex-tension and subcutaneous weekly administration for 3 weeks and
aggressiveness of the lesion on the basis of the imaging then 120 mg sub-cutaneous monthly for 3 months. If the
findings (cortical erosion, soft tissue involvement, subchondral CT evaluation at 3 months showed a peripheral bone rim
bone involvement). We strongly considered denosumab still inadequate to permit aggressive curettage, the
treatment for patients at definite risk for a re-section procedure schedule was prolonged for 3 more months to try to
because of extraosseous expansion of the tumor and only a thin obtain a better response. Nine of 12 patients received
or even interrupted peripheral bone rim around the tumor and denosumab also in the adjuvant setting (same dosage of
patients with a lesion widely involving monthly administration for 6 months) as per protocol.

Copyright 2018 by the Association of Bone and Joint Surgeons. Unauthorized reproduction of this article is prohibited.
1786 Scoccianti et al. Clinical Orthopaedics and Related Research®

Intraoperative cryotherapy was administered with cry- time of data gathering and evaluation, the study was ret-
oprobes reaching -150° C (Endocare Cryocare Systems; rospective and no objection/exception was formulated by
HealthTronics, Austin, TX, USA) after filling the void the local institutional review board. The study was
resulting from curettage with a sterile gel. Number and performed in accordance with the 1975 Declaration of
size of the cryoprobes used and number of cycles of Helsinki.
freezing (two to three) varied according to size of the
lesion and its localization (presence of nearby nerves,
vessels, or skin at risk for damage). Filling of the cavity Statistical Analysis
after curettage was performed with cement in most cases
(17) or morcellized bone allografts (four). Differences between groups in terms of clinical and surgical
All the histopathologic specimens were evaluated by an characteristics were assessed using Fisher’s exact test. The
expert pathologist in bone and soft tissue diseases at our statistical analysis was performed using R software [16]. A
pathology department (AF). Radiologic imaging was p value of # 0.05 was considered statistically significant.
evaluated by a radiologist (GR) and an orthopaedic surgeon The median age at presentation was 42 years (range,
(DAC, GS) dedicated to bone and soft tissue oncology. 17-66 years). The most frequent site of the tumor was the
After surgery, followup was performed with clinical distal femur (seven patients) followed by the distal tibia
and local radiographic evaluation after 1 month and then (four). The other tumors were located in the distal radius
every 3 months for the first year, every 4 months for the and sacrum (two patients at each site) and proximal
second year, every 6 months for the third and fourth humerus, distal humerus, finger phalanx, iliac wing,
years, and annually thereafter. In case of suspicion for proximal tibia, and patella (one at each site) (Table 1).
local re-currence, an MRI with gadolinium was used for Median followup was 39 months (range, 14-55
further evaluation. If the suspicion was confirmed by months) in the denosumab group and 27 months (range,
MRI, a CT-guided needle biopsy was performed. A chest 18-92 months) in the surgery-only group.
radiograph was performed once a year. All of the patients treated with denosumab showed
The patients were followed during denosumab treatment new bone formation around and partially inside the
and during followup by a medical oncologist (G. Baldi) and lesion at the radiologic posttreatment evaluation. This
an orthopaedic surgeon (GS, G. Beltrami) dedicated to or- allowed us to perform a curettage procedure more
thopaedic oncology. During preoperative and postoperative confidently. In all nine lesions treated with only surgery,
treatment, clinical evaluation was performed by a medical curettage was considered feasible already at presentation.
oncologist (G. Baldi) before every administration of deno- Tumors in the denosumab group were of a higher grade
sumab to detect any possible adverse effect in the period than in the surgery-only group (Campanacci Grade 3 in
between the two administrations. Adverse events and labo- eight of 12 patients in the denosumab group and two of nine
ratory abnormalities were assessed according to Common in the surgery-only group; p = 0.08). However, they were
Terminology Criteria for Adverse Events (CTCAE; version not different in terms of epiphyseal juxtaarticular bone in-
3.0), which was in current use at the time of study start. volvement; in terms of this measure, dividing them into
Standard assessments of blood count, liver function, renal those with > 75%, 50% to 75%, and < 50%, the results were
function, and electrolytes including calcium were done eight, four, and zero patients, respectively, in the
weekly in the induction phase and then monthly in the denosumab group and two, five, and two patients,
maintenance phase. Moreover, every patient underwent at respectively, in the surgery-only group (p = 0.12).
baseline a dental radiograph and a dental visit to exclude
risk factors for osteonecrosis of the jaw; then the patients
un-derwent a dental visit every 3 months during treatment.
All patients received daily supplemental doses of cal- Results
cium and vitamin D.
We used chart review to record the proportion of patients in Local Recurrence
each treatment group who had a local recurrence and to tally With the numbers available, there was no difference in the
adverse events. In terms of adverse events, we specifi-cally proportion of patients experiencing a recurrence (five of 12
looked for abnormalities in calcium and other electro-lytes in the denosumab group and one of nine in the surgery-only
metabolism or in blood count, liver or renal dysfunction group; p = 0.18; Table 2). The recurrences in the denosu-
occurrence, osteonecrosis of the jaw, or any other unexpected mab group occurred at a median of 23 months (range, 7-54
adverse events observed during denosumab treatment. months); one of these patients presented three subsequent
At the time of treatment, for each patient receiving recurrences and also pulmonary metastases. One patient in
denosumab, we requested and obtained authorization by the the group treated only with surgery and cryotherapy de-
Regional Therapeutic Commission of Tuscany. At the veloped a local recurrence at 14 months from surgery. All

Copyright 2018 by the Association of Bone and Joint Surgeons. Unauthorized reproduction of this article is prohibited.
Volume 476, Number 9 Denosumab in Giant Cell Tumor of Bone 1787

Table 1. Patient demographics and surgical details of the two groups of patients included in the study
Preoperative denosumab + curettage Only curettage and cryotherapy
Patient characteristics and cryotherapy (denosumab group) (surgery-only group)
Number 12 9
Median age (years; range) 30 (17-66) 47 (20-55)
Sex F 7, M 5 F 3, M 6
Site Lower limb 7 Lower limb 7
Upper limb 3 Upper limb 1
Pelvis/sacrum 2 Pelvis/sacrum 1
Followup (months; median/range) 39 (14-55) 27 (12-92)
Cement 8 9
Bone graft 4 0
Plate fixation 6/12 4/9
F = female
M = male.

the recurrences were histopathologically confirmed. The site renal function abnormalities. No patient thus far has de-
of the tumor in patients affected by local recurrence was the veloped signs of osteonecrosis of the jaw. No malignant
iliac wing, distal femur, patella (multiple recur-rences and neoplastic transformation has been observed in the
pulmonary metastases), proximal tibia, distal tibia, and evalu-ation of the specimens either from pretreated
finger phalanx in one patient each. All six patients tumors or from recurrences.
underwent revision surgery. Two patients underwent a re-
section procedure with reconstruction (iliac wing peri-
acetabular region with a custom-made prosthesis; finger
phalanx with a composite prosthesis using an autologous Discussion
bone graft from the iliac crest). Four patients were again
treated with curettage. The patient affected by pulmonary The use of denosumab for GCTB in the preoperative set-
metastases was treated with surgery for the local recurrence ting was greeted with much interest by the orthopaedic
and was restarted on denosumab, which was recently dis- community in the hopes of avoiding some osteoarticular
continued, and she is now under strict monitoring. resections and complex reconstructive procedures in young
patients. Although the use of denosumab in GCTB was
introduced almost 10 years ago [22], few clinical data about
Adverse Effects its results are yet available [8, 15, 17, 19, 23]. Prior work
suggests that denosumab does not eliminate tumor cells but
No adverse effect of the use of denosumab was detected only suppresses their activity [11, 13, 14]. As such, the
either during or after treatment. All the patients completed issue of local recurrence risk seems important, considering
the treatment schedule without interruption. No alterations that the surgeon operating on a patient with GCTB pre-
in serum levels of calcium or other electrolytes could be treated with denosumab now faces a different looking tu-
found nor did we observe any hematologic or liver and mor than what surgeons have grown accustomed to seeing.

Table 2. Results in terms of local recurrence

Preoperative denosumab + curettage Only curettage and cryotherapy
Results(local recurrence) and cryotherapy (denosumab group) (surgery-only group)
Total number of patients 12 9
Patients presenting a local recurrence 5 1
Time from surgery to local recurrence 7-54 (mean 26) 14
Site of local recurrence Lower limb 3 Lower limb 1
Upper limb 1 Upper limb 0
Pelvis/sacrum 1 Pelvis/sacrum 0
Treatment of local recurrence Resection 2 Curettage 3 Curettage 1

Copyright 2018 by the Association of Bone and Joint Surgeons. Unauthorized reproduction of this article is prohibited.
1788 Scoccianti et al. Clinical Orthopaedics and Related Research®

Specifically, these denosumab-treated lesions often have a series of patients with GCTB treated without denosumab
sclerosis in the lesions and a peripheral rim of bone be- but with intralesional surgery and local adjuvants, which
tween tumor tissue and remaining healthy bone, which may have estimated the risk to be between 14% and 19% [2, 9,
make it harder for surgeons to appreciate or adjuvant 10, 12]. We note that those studies included all patients,
therapies to extend the margins of the curettage. Perhaps whereas ours included only the most aggressive lesions,
because of this, it is possible that the risk of recurrence after which, as a result of their characteristics, were selected to
treatment with denosumab is greater than patients not thus receive preoperative denosumab. This makes comparing the
treated. Although we could not prove this to be the case recurrence risk difficult, and we believe this is important
(because of relatively small numbers and relatively short when comparing these results. Two other papers about local
followup), our data still suggest it may be so, and future control after curettage with preoperative denosumab found a
studies are needed to confirm this supposition. recurrence risk of 15% and 20% [17, 19], but mean fol-
The major limitation on this study was selection bias. lowup in those reports was short, at 13 and 18 months,
Because this was a retrospective study, randomization was respectively; we note two of six recurrences in our series
impossible; moreover, because we recently have been occurred > 2 years after surgery (37 and 54 months) and
treating more aggressive lesions with denosumab, a case- another occurred at 23 months. In terms of the timing of
control study with contemporaneous controls was not recurrences, our data tentatively suggest that recurrences in
possible. Instead, we tried to identify a suitable control patients with giant cell tumors treated with denosumab
group by limiting the study to patients treated with cryo- might occur later than usually observed in historical series
therapy; these patients’ lesions tended to be more aggres- [3] and therefore the recurrence rates reported by those
sive lesions among those we treated. However, as other authors could be underestimated.
determined by Campanacci grading, selection bias toward In our series, local recurrences occurred after denosu-
the most aggressive tumors being treated with denosumab mab treatment not only in patients who were at
remained; however, in terms of the extent of epiphyseal presentation particularly at risk for resection surgery
involvement, there was no difference between the groups, (Campanacci Grade 3 lesions with > 75% involvement
suggesting the two groups may indeed have been reason- of the epiphysis at the juxtaarticular level; four of the five
ably comparable, at least with respect to that surrogate for recurrences in the denosumab group), but also in one
aggressiveness. Another limitation was the very small patient with a less aggressive-looking lesion (one of the
sample size. With the numbers available, we did not see an five recurrences in that group). In fact, it is possible that
increased risk of local recurrence in the denosumab group, by using denosumab—which seems to cause sclerosis in
and we could not demonstrate either a benefit of denosu- the lesions and to create a peripheral rim of bone around
mab on local recurrence control nor an increased risk of them—may make it more difficult for surgeons to
recurrence in patients pretreated with denosumab. None- appreciate or adju-vant therapies to extend the margins,
theless, the crude incidence of local recurrence in perhaps making it more likely that local recurrence might
denosumab-treated patients was more than three times occur. We suggest that surgeons pay special attention to
higher than in nontreated patients, even if this was not a aggressive curettage after use of denosumab in patients
statistical difference given the small sample size. The low with GCTB. Even so, the reduction in tumor size caused
statistical power should cause the reader to interpret our no by denosumab and above all the restoration of a
difference findings cautiously; it is possible that larger peripheral bone rim around the tumor can make curettage
studies could demonstrate that indeed there are differences more attractive than resection [19, 23, 24], which may be
in the risk of local recurrence with or without the use of a risk worth taking in young patients to preserve
denosumab that were missed. Likewise, this study is se- function. In this clinical setting, an increased risk of local
verely underpowered also to detect even relatively com- recurrence can be consciously ac-cepted to try to save
mon complications and adverse events and therefore we the joint and to avoid complex osteoarticular
cannot draw any inferences about safety. Even so, we offer reconstructions in young patients (Fig. 1A-H).
the tentative conclusion that denosumab does not prevent In our very small series, we did not observe any com-
local recurrences, which can occur as often or even more plications associated with denosumab, but we caution
often than in nontreated patients. Our purpose in sharing readers that even common complications may not be
these findings was to share our concerns that the pre- detected when so few patients are evaluated. Because of
operative use of denosumab may be associated with an that, we can draw no inferences about the safety of deno-
increased risk of local recurrences of GCTB, even if we sumab in this setting, and in other series, several adverse
could not prove it here. This concern warrants further in- effects have been reported [8, 19, 24], including arthralgia,
quiry by future, larger studies. fatigue, pain in the extremity, headache, nausea, back pain,
The risk of local recurrence we observed after denosu- hypophosphatemia, hypocalcemia, osteonecrosis of the jaw,
mab treatment (41%) is higher than that seen in prior large osteitis, fracture, pneumothorax (in patients affected

Copyright 2018 by the Association of Bone and Joint Surgeons. Unauthorized reproduction of this article is prohibited.
Volume 476, Number 9 Denosumab in Giant Cell Tumor of Bone 1789

Fig. 1 A-H Plain radiographs showing a Grade 3 GCTB of the distal tibia in a 19-year-old woman (A-B). Radiographic (C-D)
appearance after 6 months’ treatment with denosumab: new bone formation is evident at the periphery and within the lesion.
The patient underwent intralesional curettage and cryotherapy was used as a local adjuvant; the intraoperative photographs
show the freezing process with three probes inside the cavity filled with sterile gel (E). The bony defect was packed with cement
and allogeneic bone grafts in contact with subchondral bone. Plain radiographs at 2-year followup showing no evidence of local
recurrence (F). Fifty-four months after primary surgery, coronal MRI shows local recurrence in the distal fibula (G). The patient
underwent surgery with cement removal, curettage, and new cementation in both the distal tibia and fibula (H).

by pulmonary metastases) and even other neoplasms, my- interfere with the ability of surgeons to appreciate or ad-
eloproliferative disorders, and malignant transformation of juvant therapies to extend the margins, perhaps increasing
GCTB. Our series used a shorter course of denosumab than the risk of local recurrence. On the basis of this consider-
some other series [8, 19, 24], and this could account for the ation, at present, when the characteristics of a lesion make a
absence of complications in our patients; however, again, curettage procedure feasible at presentation, a primary
we caution readers that the small size of our treatment intralesional surgical procedure without denosumab treat-
group also could explain the lack of observed ment appears to be the more prudent choice. However,
complications. Rekhi et al. [17], who reported the results of future, larger studies are needed to confirm or refute this
a series of patients treated with a short course of premise. In addition, because we have observed some late
neoadjuvant deno-sumab, did not report adverse events recurrences among patients pretreated with denosumab, we
either, but in their article, this issue was not clearly dealt recommend continued clinical and radiographic surveil-
with. Further mon-itoring is mandatory, particularly for the lance of these patients.
reported occur-rence of malignant transformation of the
tumor after treatment with denosumab [1, 6, 19].
Although denosumab may decrease the size or extent of References
these lesions before surgical treatment and may result in the
formation of a peripheral rim of bone around the lesions, we 1. Aponte-Tinao LA, Piuzzi NS, Roitman P, Farfalli GL. A high-
remain concerned that it does not decrease the aggres- grade sarcoma arising in a patient with recurrent benign giant
cell tumor of the proximal tibia while receiving treatment with
siveness of the disease. The result of this might be that denosumab. Clin Orthop Relat Res. 2015;473:3050–3055.
curettage even with adjuvants may not eradicate all of the 2. Arbeitsgemeinschaft Knochentumoren, Becker WT, Dohle J,
tumor, and in fact by causing sclerosis in the lesion, it may Bernd L, Braun A, Cserhati M, Enderle A, Hovy L, Matejovsky

Copyright 2018 by the Association of Bone and Joint Surgeons. Unauthorized reproduction of this article is prohibited.
1790 Scoccianti et al. Clinical Orthopaedics and Related Research®

Z, Szendroi M, Trieb K, Tunn PU. J Bone Joint Surg Am. 14. Mak IW, Evaniew N, Popovic S, Tozer R, Ghert M. A trans-
2008; 90:1060–1067. lational study of the neoplastic cells of giant cell tumor of
3. Athanasou NA, Bansal M, Forsyth R, Reid RP, Sapi Z. Giant bone following neoadjuvant denosumab. J Bone Joint Surg
cell tumor of bone. In: Fletcher CD, Bridge JA, Hogendoorn Am. 2014; 96:e127.
PCW, Mertens F, eds. World Health Organization (WHO) 15. Muller¨ DA, Beltrami G, Scoccianti G, Campanacci DA,
Classifica-tion of Tumors of Soft Tissue and Bone. 4th ed. Franchi A, Capanna R. Risks and benefits of combining
Lyon, France: IARC Press; 2013:321–324. denosumab and surgery in giant cell tumor of bone–a case
4. Balke M, Hardes J. Denosumab: a breakthrough in treatment of
series. World J Surg Oncol. 2016;14:281.
giant-cell tumour of bone? Lancet Oncol. 2010;11:218–219.
16. R Core Team. R: A Language and Environment for Statistical
5. Branstetter DG, Nelson SD, Manivel JC, Blay JY, Chawla S,
Computing. Vienna, Austria; 2014. Available at:
Thomas DM, Jun S, Jacobs I. Denosumab induces tumor re- Accessed October 4, 2017.
duction and bone formation in patients with giant-cell tumor of
bone. Clin Cancer Res. 2012;18:4415–4424. 17. Rekhi B, Verma V, Gulia A, Jambhekar NA, Desai S, Juvekar
SL, Bajpai J, Puri A. Clinicopathological features of a series of
6. Broehm CJ, Garbrecht EL, Wood J, Bocklage T. Two cases of
sarcoma arising in giant cell tumor of bone treated with deno- 27 cases of post-denosumab treated giant cell tumors of bones:
sumab. Case Rep Med. 2015;2015:767198. a single institutional experience at a tertiary cancer referral
7. Campanacci M, Baldini N, Boriani S, Sudanese A. Giant-cell centre, India. Pathol Oncol Res. 2017;23:157–164.
tumor of bone. J Bone Joint Surg Am. 1987;69:106–114. 18. Roitman PD, Jauk F, Farfalli GL, Albergo JI, Aponte-Tinao LA.
Denosumab-treated giant cell tumor of bone. Its histologic spec-trum
8. Chawla S, Henshaw R, Seeger L, Choy E, Blay JY, Ferrari S,
Kroep J, Grimer R, Reichardt P, Rutkowski P, Schuetze S, and potential diagnostic pitfalls. Hum Pathol. 2017;63:89–97.
Skubitz K, Staddon A, Thomas D, Qian Y, Jacobs I. Safety 19. Rutkowski P, Ferrari S, Grimer RJ, Stalley PD, Dijkstra SP,
and efficacy of denosumab for adults and skeletally mature Pienkowski A, Vaz G, Wunder JS, Seeger LL, Feng A,
adolescents with giant cell tumour of bone: interim analysis of Roberts ZJ, Bach BA. Surgical downstaging in an open-label
an open-label, parallel-group, phase 2 study. Lancet Oncol. phase II trial of denosumab in patients with giant cell tumor of
2013; 14:901–908. bone. Ann Surg Oncol. 2015;22:2860–2868.
9. Errani C, Ruggieri P, Asenzio MA, Toscano A, Colangeli S, 20. Steensma MR, Tyler WK, Shaber AG, Goldring SR, Ross FP,
Rimondi E, Rossi G, Longhi A, Mercuri M. Giant cell tumor of Williams BO, Healey JH, Purdue PE. Targeting the giant cell
the extremity: a review of 349 cases from a single institution. tumor stromal cell: functional characterization and a novel
Cancer Treat Rev. 2010;36:1–7. therapeutic strategy. PLoS One. 2013;26:e69101.
10. Gaston CL, Bhumbra R, Watanuki M, Abudu AT, Carter SR, 21. Thomas D, Henshaw R, Skubitz K, Chawla S, Staddon A,
Jeys LM, Tillman RM, Grimer RJ. Does the addition of ce- Blay JY, Roudier M, Smith J, Ye Z, Sohn W, Dansey R, Jun S.
ment improve the rate of local recurrence after curettage of Denosumab in patients with giant-cell tumour of bone: an
giant cell tumours in bone? J Bone Joint Surg Br. 2011;93: open-label, phase 2 study. Lancet Oncol. 2010;11:275–280.
22. Thomas D, Skubitz K. Giant cell tumour of bone. Curr Opin
11. Girolami I, Mancini I, Simoni A, Baldi GG, Simi L, Campanacci
Oncol. 2009;21:338–344.
D, Beltrami G, Scoccianti G, D’Arienzo A, Capanna R, Franchi A. 23. Traub F, Singh J, Dickson BC, Leung S, Mohankumar R,
Denosumab treated giant cell tumour of bone: a morphologi-cal,
Blackstein ME, Razak AR, Griffin AM, Ferguson PC, Wunder JS.
immunohistochemical and molecular analysis of a series. J Clin
Efficacy of denosumab in joint preservation for patients with giant
Pathol. 2016;69:240–247.
cell tumour of the bone. Eur J Cancer. 2016;59:1–12.
12. Kivioja AH, Blomqvist C, Hietaniemi K, Trovik C, Walloe A,
24. Ueda T, Morioka H, Nishida Y, Kakunaga S, Tsuchiya H, Matsu-
Bauer HC, Jorgensen PH, Bergh P, Folleras˚ G. Cement is rec-
moto Y, Asami Y, Inoue T, Yoneda T. Objective tumor re-sponse
ommended in intralesional surgery of giant cell tumors: a
Scan-dinavian Sarcoma Group study of 294 patients followed to denosumab in patients with giant cell tumor of bone: a
for a median time of 5 years. Acta Orthop. 2008;79:86–93. multicenter phase II trial. Ann Oncol. 2015;26:2149–2154.
13. Lau CP, Huang L, Wong KC, Kumta SM. Comparison of the 25. Wojcik J, Rosenberg AE, Bredella MA, Choy E, Hornicek FJ,
anti-tumor effects of denosumab and zoledronic acid on the Nielsen GP, Deshpande V. Denosumab-treated giant cell
neoplastic stromal cells of giant cell tumor of bone. Connect tumor of bone exhibits morphologic overlap with malignant
Tissue Res. 2013;54:439–449. giant cell tumor of bone. Am J Surg Pathol. 2016;40:72–80.

Copyright 2018 by the Association of Bone and Joint Surgeons. Unauthorized reproduction of this article is prohibited.