Bisphosphonate-loaded

Bone Cement as a Local

Adjuvant Therapy for
Giant Cell Tumor of Bone
A 1 to 12-Year Follow-up Study
David D. Greenberg, MD* and Francis Y. Lee, MD, PhD†
 • Background:

Abstract
 Nontargeted adjuvant therapies  have been applied
locally after curettage of giant cell tumors (GCT).
 Systemic bisphosphonates (BP) and denosumab 
osteoclast-targeting therapies
 Downside: Systemic side effects.
 We examined the effects of an alternative intraoperative
 local delivery of BP on GCTs.
• Material and Methods:
 17 patients with GCTs  extended surgical curettage
(high-speed burring + traditional adjuvant therapy + BP-
loaded polymethylmethacrylate bone cement)
 Clinical data and follow-up radiographs  investigate
local recurrence (LR) rate and complications
 • Result:

Abstract
 6 males and 11 females (mean age 33.7 y) without
pulmonary metastases.
 Follow-up ranged from 1 to 12 years.
 1 LR during the follow-up period for an LR rate of 5.9%.
 Mean final Musculoskeletal Tumor Society (MSTS) score
was 29.
 No systemic or localized avascular necrosis or atypical
fractures related to BPs noted
• Conclusion:
 BP-loaded polymethylmethacrylate is a targeted local adjuvant
therapy that is feasible, safe, and may reduce LRs while
alleviating the risk of systemic side effects.
Background
• Giant cell tumors (GCT) of bone  5% of all primary bone tumors
 Benign neoplasm with potential for aggressive local invasion and destruction.
 Commonly occurs in the epiphysis and metaphysis of the distal femur, proximal tibia, proximal
humerus, and distal radius  joints for load-bearing and functional activities of daily living.

• Optimal treatment  controversies

 Simple curettage alone  local recurrence (LR) rates 10% to 65%
 Aggressive en bloc resection + endoprosthethic reconstruction  lower LR but mobility and
functionality are often sacrificed.
 Aggressive curettage + bone cement filling  LR rates 10-50%

• Local adjuvants (argon beam coagulation, phenol, ethanol, hydrogen peroxide, liquid nitrogen, etc.) are
often utilized after intralesional curettage
 May cause tissue necrosis, cold damage, or chemical burn in the surrounding normal tissues.
 A recent systemic review and meta-analysis questions the value and need for these surgical adjuvants when
meticulous tumor removal is performed
 There is a need for alternatives to decrease LR of GCT of bone
Background
• Neoplastic GCT stromal cells produce RANKL, cytokines, and chemokines.
 RANKL  ↑ formation of multinucleated osteoclast-like giant cells  lytic destruction of the bone
and progression of GCT

• Targeted therapies such as BPs and denosumab have been used for treatment of GCT.
 Bisphosphonates (BP)  have been used to prevent osteoclastic bone resorption by inducing
osteoclastic cell death + induces apoptosis of proliferating GCT stromal cells.
 Systemic BP  lower rates of recurrence
 Denosumab, a RANKL inhibitor  block the formation of multinucleated giant cells.
 Short-term follow-up results of a phase II clinical trial of systemic denosumab  promising efficacy in terms of
disease progression
 Long-term follow-up evaluation or results after cessation of denosumab treatment are not known.
 Use of BP and denosumab  systemic side effects such as osteonecrosis of the jaw, atypical fractures,
and hypocalcemia.
 Cost of administering denosumab is very expensive
Background
• PMMA bone cement impregnated with zoledronic acid releases biologically
active zoledronate (biphosphonate) that is cytotoxic to stromal GCT cells.
 The use of BP is preferred because denosumab may not tolerate the heat
generated by PMMA
 Local delivery of these antiresorptive drugs may be an option that can be
easily incorporated into current surgical paradigms without concern of
systemic side effects.
• Our hypotheses: local delivery of BP as a surgical adjuvant is
technically feasible, well-tolerated by patients, and may potentially
decrease LR of GCT of bone.
Materials and Methods
• A multicenter proof-of-principle study was performed between February 2002
and May 2018 to investigate extremity GCT of bone surgically treated with
the addition of local BP.
• Inclusion criteria included:
 (1) benign GCT of bone
 (2) lesion located in an extremity
 (3) lesion amenable to intralesional curettage defined as having at least 1 intact column of
bone after removal.

• Exclusion criteria included:

 (1) nonextremity location
 (2) previous systemic therapy with BP or denosumab
 (3) lesion too extensive for intralesional treatment, either due to bone loss, joint invasion, or
large soft tissue component.
Materials and Methods
• All lesions were histologically confirmed before definitive treatment.
• Preoperative imaging, including radiographs of the lesion, a chest radiograph
or chest computed tomography, and magnetic resonance imaging of the
affected area, was obtained.
• Radiographs were graded according to the Campanacci staging system:
 Grade I tumors  well marginated border of a thin rim of mature bone and intact cortex
 Grade II tumors  relatively well-defined margin but no radiopaque rim; the combined
cortex and rim of reactive bone is rather thin and moderately expanded but still present. A
 Grade III tumor has indistinct borders and cortical destruction
Materials and Methods
• Surgical methods:
 Extensive curettage of the lesion to remove macroscopic tumor
 Highspeed burring of the residual cavity
 Adjuvant treatment to the residual cavity
 Argon beam coagulation combined with full-strength hydrogen peroxide
 Packing of the cavity with either:
 PMMA bone cement alone
or
 bone cement with subchondral allograft bone graft
 The 4 mg/100 mL of zoledronic acid (Zometa) was added to each bag of bone cement in the
operating room and mixed before packing the cavity.
 Of note, pamidronate was used in 2 patients at the beginning of the study before zoledronic acid
becoming popular in the United States after FDA approval in 2002.
 Elution and stability of pamidronate and zoledronate within the PMMA bone cement along with
side effects profiles are similar.
Materials and Methods
• Postoperative:
 Follow-ups consisted of clinical visits and radiographs.
 Clinical visits  2 weeks post-op, 6 weeks post-op, and then every 3 months for the first 2, every 6 months for years 3
through 5, and then annually thereafter.
 The patient’s surgical wound was inspected at each clinical visit for wound complications.
 At each follow-up, except for the 2-week visit, radiographs of the affected area and a chest radiograph were obtained.

• Criteria used for evidence of LR:

 (1) > 5 mm of lysis around the bone cement interface
 (2) extension of a lytic zone plus recurring/ progressing pain in the involved area; or
 (3) appearance of a soft tissue mass near the previously operated area, as noted by palpation and confirmed by subsequent imaging.
 Suspicion of recurrence  histologically confirmed and treated according to the surgeon.

• The Musculoskeletal Tumor Society (MSTS) Score  to assess functional results.

 is a validated and well-accepted functional scoring system used in orthopedic oncology research.
 a point system (0 to 5) scoring 6 categories: pain, function, emotional acceptance, use of supports, walking ability, and
gait.
 The summation of these values = functional score, higher scores indicating better function.
 Score grading:
 Excellent (30 to 24 scores), good (23 to 18 scores), fair (17 to 12 scores), or poor (<12 scores).
Results
• Six men and 11 women were included in this series and the combined average age was 33.7 years
(range, 15 to 51 y).
• Patient length of follow-up ranged from
12 months to 12 years (mean, 52 mo; median, 34 mo)
Results
• Tumor Characteristics:
 Locations :
 distal femur in 7 patients
 distal radius in 3 patients
 proximal tibia in 3 patients
 proximal humerus in 2 patients
 clavicle in 1 patient
 first metatarsal in 1 patient.
 Two patients presented with a pathologic fracture
 There were 16 primary GCTs and 1 recurrent GCT.
 The index procedure of 1 recurrent case did not include local BP use.
 Grading:
 7 GCTs were classified as grade 2
 10 GCTs were classified as grade 3.
 None of the patients have pulmonary metastases at diagnosis or developed metastases during follow-
up.
Imaging of GCT (pre-op, post-op, and follow-
up) from 2 cases
Results
• Tumor Recurrences
 There was a case of LR
during the follow-up
period (LR rate: 5.9%).
 26-year-old man (case 14)
with GCT of the distal femur
that was treated with 2
batches of Simplex P bone
cement with 8 mg of
zoledronic acid added.
 The LR was noted 5
months after index surgery
and was treated with a
distal femur replacement.
Results
• Complications
 1 patient developed a postoperative hematoma requiring irrigation and debridement.
 This patient healed uneventfully after the incision and drainage and had no additional
complications at latest follow-up
 1 patient (case 1) underwent removal of hardware at 2 years and 8 months after her index
surgery.
 There were no cases of remote side effects such as osteonecrosis of the jaw or atypical femur
fractures related to BP use

• MSTS Outcome Scores

 The overall mean MSTS Score at the most recent follow-up was 29 (range, 24 to 30)
indicating an “excellent” qualitative result.
 The MSTS Scoring System has been validated to reliably assess patients’ post surgery
outcomes across the above-listed domains and has shown low interobserver variability.
Discussion
• Histologically GCT is characterized by 3 cell types:
 (1) multinucleated osteoclast-like giant cells;
 (2) GCT stromal cells  neoplastic component of GCT + overproduce RANKL
 (3) mononuclear cells (osteoclast precursor cells)
 Recent literature  neoplastic stromal cells recruit osteoclast precursors results in paracrine
osteoclastogenesis
 In addition, the recruited osteoclast precursors may feed an autocrine loop by production of
RANKL and RANK  excessive osteoclastogenesis, bone destruction, and pathologic fractures.

• This study  locally delivering BP-eluting PMMA bone cement to the tumor
cavity specifically targets the ongoing pathologic osteoclastogenesis
 The LR rate of 5.9% in this feasibility study compared with the LR of ∼25% to 65%
reported in older series  targeted local adjuvant therapy may provide additional benefit
Discussion
• Limitations of this study:
 Relatively low level of evidence  small number of patients and lack of a
control group
 Short length of follow-up for some of the patients.
 Decision to allow traditional local surgical adjuvants  potential bias

• However, this series was intended as a feasibility study and not

meant to definitively prove the superiority of BP-loaded bone
cement
 This series show  targeted local therapy with BP can be performed safely,
is well-tolerated, and may potentially improve the rate of LR
Discussion
• Orthopedicsurgeons have tried many different options to prevent
LRs with intralesional treatment:
 Extensive curettage and high-speed burring combined with chemical or
physical agents such as PMMA bone cement, argon beam coagulation, liquid
nitrogen, phenol, and hydrogen peroxide.
 PMMA bone cement combines an adjuvant exothermic process with a cavity-filling
reconstruction.
 Alternatives: allograft or synthetic bone graft  for cavity-filling.
 The literature is divided regarding outcomes, with some authors reporting a lower risk
of LR with the use of PMMA
 No clearly superior traditional adjuvant therapy or cavity reconstruction has been
defined and LR rates remain relatively high and stagnant.
Discussion
• The pathologic bone loss seen in primary bone tumors often shares a common mechanism of
excessive osteoclastogenesis
 Previous translational research  zoledronate and pamidronate have proapoptotic effects of on GCT
cells
 we believe the use of BP-loaded PMMA bone cement creates a targeted, safe, and feasible solution that is
attractive to orthopedic surgeons and their patients with resectable cases of GCT.

• BP-loaded PMMA provides mechanical stability and a vehicle for continued BP delivery at
the tumor-cavity margin.
 Current clinical data  BPs are reliable inducers of osteoclast and GCT stromal cell apoptosis
 The current results provide clinical support for these previous translational research and concepts.

• A case report  local BP in a patient with GCT of the proximal tibia.

 Two months following  curettage and bone grafting.
 Sections obtained during GCT resection for histology revealed massive cellular necrosis of both the stromal
cells and osteoclast-like giant cells.
 This shows the promising efficacy of locally delivered BPs and their effect on the multiple cell lineages present
in GCT.
Discussion
• The authors hope that the current proposal serves as a stepping stone to fill the
gap between conventional surgical reconstruction and molecular biologically
targeted therapy
• BP-loaded PMMA cement is a simple modification and does not require
additional treatments at separate occasions.
• This results show a new adjuvant therapeutic concept for GCTs is feasible,
safe, and potentially effective.
• A multicenter randomized, controlled clinical trial has recently been initiated
 definitive data regarding the effect on LR of the use of locally delivered BP
in GCT patients.
THANK YOU