denosumab induces tumor reduction and bone formation in ... · denosumab in reducing...

Cancer Therapy: Clinical

Denosumab Induces Tumor Reduction and Bone Formationin Patients with Giant-Cell Tumor of Bone

Daniel G. Branstetter1, Scott D. Nelson2, J. Carlos Manivel3, Jean-Yves Blay4, Sant Chawla5,David M. Thomas6, Susie Jun7, and Ira Jacobs7

AbstractPurpose:Giant-cell tumor of bone (GCTB) is a locally aggressive, benign osteolytic tumor in which bone

destruction is mediated by RANK ligand (RANKL). The RANKL inhibitor denosumab is being investigated

for treatment of GCTB. We describe histologic analyses of GCTB tumor samples from a phase II study of

denosumab in GCTB.

Experimental Design: Adult patients with recurrent or unresectable GCTB received subcutaneous

denosumab 120 mg every 4 weeks (with additional doses on days 8 and 15). The primary histologic

efficacy endpointwas the proportionof patientswhohad a90%ormore eliminationof giant cells from their

tumor. Baseline and on-study specimens were also evaluated for overall tumormorphology and expression

of RANK and RANKL.

Results: Baseline tumor samples were typically composed of densely cellular proliferative RANKL-

positive tumor stromal cells, RANK-positive rounded mononuclear cells, abundant RANK-positive tumor

giant cells, and areas of scant de novo osteoid matrix and woven bone. In on-study samples from 20 of 20

patients (100%), a decrease of 90%ormore in tumor giant cells and a reduction in tumor stromal cells were

observed. In these analyses, thirteen patients (65%) had an increased proportion of dense fibro-osseous

tissue and/or new woven bone, replacing areas of proliferative RANKL-positive stromal cells.

Conclusions: Denosumab treatment of patients with GCTB significantly reduced or eliminated RANK-

positive tumorgiant cells.Denosumabalso reduced the relative content ofproliferative, densely cellular tumor

stromal cells, replacing them with nonproliferative, differentiated, densely woven new bone. Denosumab

continues to be studied as a potential treatment for GCTB. Clin Cancer Res; 18(16); 4415–24.�2012 AACR.

IntroductionClinically, giant-cell tumor of bone (GCTB) is a benign,

locally aggressive, osteolytic tumor that causes significantbone destruction and has a predilection for the epiphyseal/metaphyseal region of long bones and spine (1). Histolog-ically, GCTB is composed of sheets of neoplastic ovoidmononuclear cells with high RANK ligand (RANKL) expres-sion, RANK-positive mononuclear cells of myeloid linage,and a randomly distributed population of large RANK-expressing osteoclast-like giant cells (2–13).Many research-ers have shown that the giant cell of GCTB is osteoclastic in

nature (4, 6, 10) and that the true neoplastic cells are ovoidcells displaying markers of mesenchymal stem cells thathave partially differentiated along the osteoblast lineage(14–17). These tumors are frequently described to containsmall areas of osteoid matrix deposition, woven bone, andoccasionally new bone, which may be reactive tissue at thetumor margin or formed de novo within the tumor(9, 18, 19). The hallmarks of this tumor are its aggressivelylytic behavior and the clear role that osteoclast-like tumorgiant cells play in the lytic process. Denosumab is a fullyhuman monoclonal antibody (mAb) that specifically inhi-bits normal and tumor-associated bone lysis by preventingRANKL-mediated formation and activation of multinucle-ated osteoclasts or giant cells from RANK-positive mono-nuclear preosteoclasts and macrophages (20, 21). Subcu-taneous administration of denosumab has been shown tosuppress bone destruction in patients with osteolytic bonedisease in multiple myeloma or bone metastases in breastand prostate cancer and other solid tumors (22–24). It washypothesized that denosumabwould inhibit the osteoclast-like giant cells in a similar manner.

An open-label, phase II, proof-of-concept clinical studywas conducted in patients with recurrent or primary unre-sectable GCTB to evaluate the efficacy of subcutaneous

Authors' Affiliations: 1Amgen Inc., Seattle, Washington; 2University ofCalifornia, Los Angeles, California; 3University of Minnesota, Minneapolis,Minnesota; 4University Claude Bernard Lyon I and Hopital Edouard Heriot,Lyon, Cedex, France; 5Sarcoma Oncology Center, Santa Monica, Califor-nia; 6Peter MacCallum Cancer Centre, East Melbourne, Victoria, Australia;and 7Amgen Inc., Thousand Oaks, California

Note: Supplementary data for this article are available at Clinical CancerResearch Online (http://clincancerres.aacrjournals.org/).

Corresponding Author: Ira Jacobs, Amgen Inc., 1 Amgen Center Drive,MS 38-2-A, Thousand Oaks, CA 91320. Phone: 805-447-9161; Fax: 805-375-9763; E-mail: [email protected]

doi: 10.1158/1078-0432.CCR-12-0578

�2012 American Association for Cancer Research.

ClinicalCancer

Research

www.aacrjournals.org 4415

on August 21, 2019. © 2012 American Association for Cancer Research. clincancerres.aacrjournals.org Downloaded from

Published OnlineFirst June 18, 2012; DOI: 10.1158/1078-0432.CCR-12-0578

http://clincancerres.aacrjournals.org/

denosumab in reducing tumor-associated bone lysis. Thesafety and efficacy of denosumab in this study have beendescribed previously (20). This article describes the histo-logic effects of denosumab treatment in GCTB in a subset of20 patients for whom histologic data were available frombaseline and on-study biopsies or resections in the phase IIstudy.

Materials and MethodsPhase II clinical study

The clinical study of denosumab in GCTB has beendescribed previously (20). Briefly, in this open-label, sin-gle-group study, 37 patients with recurrent or unresectableGCTB were treated with subcutaneous denosumab 120 mgevery 28 days, with additional doses on days 8 and 15. Theprimary efficacy endpoint was the proportion of patientswith a tumor response, defined histopathologically as atleast 90% elimination of giant cells relative to baseline; or,where giant cells represented less than 5% of tumor cells atbaseline, complete elimination of giant cells. If histopath-ologic data were not available, tumor response was definedas a lack of radiologic progression of the target lesion byweek 25. Prespecified exploratory analyses included clinicalbenefit as reported by investigators and subjective evalua-tion of baseline and on-study tumor tissue specimens forRANKL and RANK expression. Histopathologic evaluationsof baseline and on-study tissue specimens from clinicalstudy patients were conducted by a central pathologist whowas blinded to the sequence of specimen collection (base-line or on-study).

Patient specimensFor the subset of 20 patients in the phase II study from

whom histologic specimens were available, we comparedbaseline and on-study tissue specimens to subjectivelyevaluate the effects of denosumab treatment on overall

tumor morphology. Evaluation included the extent of thetumor section composed of densely cellular ovoid or spin-dle-shaped mononuclear tumor stromal cells, increased denovo bone matrix, woven bone, or new bone, and thepresence of tumor giant cells. Changes in the degree anddistribution of expression of RANK, RANKL, Ki67, andactivated caspase 3 in the tissue sections and in tumor cellswere also noted. Baseline biopsy was carried out if clinicallyappropriate; otherwise, themost recent biopsy before deno-sumab treatment was used. On-study biopsies or resectionswere obtained from all patients between doses 5 and 9 ofdenosumab (months 3 and 7), unless the procedures werejudged to be clinically unsafe. Up to 10 unstained slides oftissue from each biopsy/resection were formalin-fixed andparaffin-embedded (FFPE) at clinical sites and sent to theAmgen Tissue Bank.

Tumor morphologyFFPE tissue sections (4–5 mm) were deparaffinized,

hydrated, and stained with hematoxylin and eosin (H&E).Baseline and on-study tissue samples from each patientwere evaluated for changes in overall tissue compositionand architecture by light microscopy. Significant and con-sistent differences in the content of dense fibro-osseousconnective tissue or new bone between baseline and on-study samples were qualitatively evaluated and described asincreased, decreased, or unchanged relative to baseline.Representative photomicrographs were obtained.

Expression of RANK and RANKLBaseline and on-study tissue specimens were evaluated

for RANK and RANKL expression by immunohistochemis-try (IHC). Dried tissue sections were incubated at 60�C for45 minutes or overnight, deparaffinized, hydrated, andprepared by heat retrieval in Diva Decloaker (Biocare Med-ical) or citrate buffer pH 6 (Citra, BioGenex LaboratoriesInc.). Immunohistochemical analysis was carried out usingautomated staining methods (Dako Autostainer: ThermoScientific; Dako North America). Tissue sections wereblocked against endogenous proteins and staining reagents.Tissue sections were stained with a mouse anti-humanRANK antibody, Amgen N-2B10.1 (5 mg/mL; Amgen Inc.)and mouse anti-human RANKL antibody, Amgen M366(0.75mg/mL; Amgen Inc.) with goat anti-mouse secondaryantibody. The specificity of the N-2B10.1 antibody forHuRANK and of the M366 antibody for HuRANKL hasbeen shown in transfected cell lines and in cell lines endog-enously expressing the appropriate antigen (25). In thesespecimens, the pattern of labeling by Western blotting andflowcytometrywas concordantwith expressionobservedbyIHC in tumor xenograft samples derived from these celllines. In addition, no labeling by flow cytometry or IHCwasobserved in untransfected parental cells or negative controlxenografts. An isotype control mouse IgG1 slide was ana-lyzed for each specimen and each antibody. Antibodyreactivity was visualized by 3,30-diaminobenzidine (DAB;Dako, # K3466; Dako North America) enhanced by tyra-mide amplification (Perkin Elmer). The relative degree and

Translational RelevanceGiant-cell tumor of bone (GCTB) is a rare, aggressive

osteolytic tumor that causes localized pain and impairedmobility and function. Surgery, the definitive therapy forGCT, is often associated with significant recurrence andmorbidity, including amputation. GCTB is characterizedby osteoclast-like giant cells and their precursors thatexpress RANK and mononuclear stromal cells thatexpress RANK ligand (RANKL), a key mediator of oste-oclast activation. Denosumab binds to RANKL, inhibit-ing bone destruction and eliminating giant cells. In anopen-label phase II proof-of-concept study of patientswithGCTB, 86%of patients had a tumor response.Usinghistologic analyses from this study, this article exploresthe mechanism of action of denosumab in GCTB, sup-porting continued evaluation of denosumab for treat-ment of GCTB.

Branstetter et al.

Clin Cancer Res; 18(16) August 15, 2012 Clinical Cancer Research4416




distribution of RANK and RANKL-positive cells in thetumor specimen were assessed subjectively for all samples.

Tumor proliferation and apoptosisTo assess evidence of cell proliferation and apoptosis, the

baseline and on-study tissue specimens were analyzed forKi67 and caspase 3 expression. Dried tissue sections wereincubated at 60�C for 45 minutes or overnight, deparaffi-nized, hydrated, and prepared by heat retrieval in DivaDecloaker Solution in Decloaking Chamber (Biocare Med-ical). IHC was carried out using automated staining meth-ods (Labvision Autostainer: Thermo Scientific; Dako Auto-stainer: DakoNorth America). Tissue sections were blockedagainst endogenous proteins and staining reagents andstained with a rabbit polyclonal Ki-67 antibody, Abcam#ab66155-200 (0.5 mg/mL; Abcam) and caspase 3 antibody(1:50; Cell Signaling, #9661L), followed by anti-rabbitsecondary antibody, EnVision, Dako #K4003. Antibodyreactivity was visualized by DAB.

Archival specimensTo establish that baseline patient samples from the phase

II study were morphologically representative of the overallpopulationofGCTB, 40archival FFPE specimensof primaryand recurrent GCTB, unrelated to the patient specimensdescribed above, were stained with H&E. We subjectivelyevaluated the extent of the tumor section composed ofdensely cellular ovoid or spindle-shaped mononucleartumor stromal cells, increased de novo bone matrix, wovenbone, or new bone, and the presence of tumor giant cells.

Microscopy methodsHistologic images were photographed using a Nikon

Eclipse E600 microscope with a Nikon DXM1200 digitalcamera. The resulting images were white-balanced usingPhotoshopCS software; no further imagemanipulationwasdone.

ResultsHistopathology resultsAmong the 37 patients treated with denosumab, 20

patients had baseline and on-study biopsies or histologicspecimens from resection. All 20 patients had specimensevaluable for tumor morphology, 19 for the presence ofRANKL, and 10 for the presence of RANK. The demographicand disease characteristics in the histologic subset weresimilar to those in the overall study population (20).Approximately half (55%) of patients were women, andapproximately two-thirds were Caucasian;mean age was 33years (range, 20–59 years; Table 1). Most (80%) had unre-sectable disease, either primary or recurrent; 70% hadprevious surgery (Table 1).Baseline tumor samples were typically composed of

densely cellular RANKL-positive tumor stromal cells andrandomly distributed abundant RANK-positive tumor giantcells. Numerous RANK-positive ovoid mononuclear cellswere also present. Very sparse focal areas composed of bonematrix, fibro-osseous tissue, and woven bone were present

in the tumor in 16 of 31 (52%) evaluable baseline samples;paired on-study samples were not available for all baselinesamples (Fig. 1A and B; Fig. 3A and B; Fig. 4A, C, and E; Fig.5A, C, and E).

The 40 archival GCTB samples unrelated to the phase IIpatient samples were similar in morphology to the phase II

Table 1. Demographic and diseasecharacteristics: histology subset

CharacteristicDenosumab-treatedpatients (N ¼ 20)

Female sex, n (%) 11 (55)Ethnic or racial group, n (%)White or Caucasian 13 (65)Black or African American 2 (10)Hispanic or Latino 3 (15)Asian 2 (10)

Age, y, mean (SD) 33 (10)GCTB disease type, n (%)Primary unresectablea 8 (40)Recurrent unresectablea 8 (40)Recurrent resectable 4 (20)

ECOG performance status0 7 (35)1 10 (50)2 1 (5)Missing 2 (10)

Anatomic site ofbiopsy/resection, n (%)

Baselineb On-studyb

Tibia 4/18 (22) 2/11 (18)Sacrum 3/18 (17) 2/11 (18)Femur distal 2/18 (11) 1/11 (9)Fibula 2/18 (11) 1/11 (9)Other 1/18 (6) 2/11 (18)Pubis 2/18 (11) 1/11 (9)Ulna 2/18 (11) 1/11 (9)Thoracic vertebrae 1/18 (6) 1/11 (9)Radius 1/18 (6) 0/11 (0)

Previous surgery, n (%)None 6 (30)1 4 (20)>1 10 (50)

Other previous therapies receivedChemotherapy 2 (10)Radiation 4 (20)Oral bisphosphonates 0 (0)Intravenous bisphosphonates 1 (5)Calcitonin 0 (0)IFN 0 (0)

aUnresectable disease includes disease for which resectioncould not be done without nerve damage or substantialimpairment of joint function.bIndicates the number of patients for whom the biopsy/resection site were identified.

Denosumab Induces Tumor Reduction and Bone Formation

www.aacrjournals.org Clin Cancer Res; 18(16) August 15, 2012 4417




baseline samples in the extent of densely cellular ovoid orspindle-shaped mononuclear tumor stromal cells and thepresence of osteolytic tumor giant cells.

Eighteen of 40 (45%) of the archival samples showedsmall foci of de novo bonematrix, woven bone, or new bonein a proportion similar to that seen in the phase II baselinesamples.

As previously reported (20), all 20 evaluable patientsshowed a decrease in tumor giant cells of 90% or moreupon histologic analysis of on-study tumor samples. We

also observed a marked reduction in the tumor contentof densely cellular RANKL-positive tumor stromal cellsand increased fibro-osseous tissue and/or new bone thatreplaced the original RANK-positive tumor cells (Fig.2A–D; Fig. 3C–F; Fig. 4B, D, and F; Fig. 5B, D, and F).In 20 pairs of patient samples evaluated, 65% of on-study samples showed a marked increase from baselinein the proportion of dense fibro-osseous tissue and/ornew bone and a consequent reduction in tumorcellularity.

Figure 1. Baseline H&E-stainedspecimen (subject 1). A, denselycellular tumor with numeroustumor giant cells surrounded byabundant spindle-shaped tumorstromal cells. B, areas of increasedbone matrix and woven bone(arrows).

Figure 2. On-study H&E-stainedspecimen (subject 1). A and B, lowmagnification fields showingtransition from residual denselycellular (DC) tumor cell areas tofibro-osseous intercellular areasand new bone (NB). C, highermagnification of residual denselycellular tumor. D, highermagnification showingreplacement of tumor withtrabeculae of woven bone andmore normal new bone.

Branstetter et al.





After treatment, small islands of residual tumor tissueremained, composed of tumor stromal cells expressingsignificant levels of RANKL and mononuclear cells expres-sing RANK, whereas tumor giant cells were typicallyabsent (Fig. 2A–C; Fig. 3 C and D; Fig.4D and F; Fig.5D and F). In baseline samples, Ki67-positive tumorstromal cells were observed, but tumor giant cells werenot Ki67 positive. In on-study samples, the proportion ofKi67-positive cells was reduced; however, the residualareas of densely cellular stromal cells contained a similar

proportion of Ki67-positive cells to that found in baselinesamples (Supplementary Fig. S1A and B). The proportionof activated caspase 3 was very low in both baseline andon-study samples and was not changed by treatment atthe time of on-study sample collection 3 to 7 months afterthe start of treatment.

Clinical benefitNineteen of the 20 patients in the histologic subset had

baseline and on-study investigator assessments of clinical

Figure 3. RANKL and RANK stains(subject 1). A, baseline section:numerous RANKL-positive tumorstromal cells surrounding RANKL-negative tumor giant cells. B,baseline section: RANK-positivetumor giant cells surrounded byRANK-negative tumor stromal cells.C, on-study section: residual area ofdense RANKL-positive tumorstromal cells. D, on-study section:residual densely cellular tumor cellswith numerous RANK-positivemononuclear cells but no RANK-positive tumor giant cells. E, on-study section: area where denselycellular tumor is replaced bytrabeculae of woven bone and moremature newbone that contain a smallnumber of RANKL-positive stromalcells. F, on-study section: areawheredensely cellular tumor cells arereplaced by trabeculae of wovenbone and more mature new bonewith no RANK-positive cellsremaining.






benefit. Of these patients, 17 (90%) were judged byinvestigators at various stages of treatment to havederived clinical benefit from denosumab (e.g., improvedfunctional status or reduced pain). Of the 13 patients inthis analysis for whom histologic evidence of new boneformation was observed, investigators reported clinicalbenefit for 11 patients and no clinical benefit for 2. Sevenpatients with no histologic evidence of new bone forma-tion also experienced clinical benefit. These resultsshould be interpreted with caution; the study populationwas small and the study was not statistically powered totest these correlations. The condition of patients at base-line varied considerably, and the timing of histologic

samples was independent of the reporting of clinicalbenefit.

DiscussionThis article summarizes the results of histologic analysis

of tissue specimens from patients with GCTB, providingevidence that the fully humanmAb denosumab significant-ly reduces or eliminates RANK-positive tumor giant cells,reduces the relative proportion of proliferative, denselycellular tumor stroma, and promotes the formation ofdifferentiated bone tissue.

Some aspects of the histogenesis of GCTB have beenpreviously characterized. GCTB contains 3 histologically

Figure 4. H&E, RANKL, and RANKstains (subject 2). A, baselinesection: densely cellular tumor withnumerous tumor giant cellssurrounded by abundant spindle-shaped tumor stromal cells. B, on-study section: transition fromresidual densely cellular tumor cellareas to fibro-osseous intercellularareas and new bone. C, baselinesection: numerous RANKL-positive tumor stromal cellssurrounding RANKL-negativetumor giant cells. D, on-studysection: residual area of RANKL-positive dense tumor stromal cellsseparated by wide areas of poorlymineralized fibro-osseous matrix.E, baseline section: RANK-positivetumor giant cells surrounded byRANK-negative tumor stromalcells. F, on-study section: residualdensely cellular tumor cells withnumerous RANK-positivemononuclear cells, but no RANK-positive tumor giant cellsseparated by wide areas of poorlymineralized fibro-osseous matrix.

Branstetter et al.





different cell types: round to oval polygonal or elongatedmononuclear stromal tumor cells, a minor component ofreactive rounded mononuclear cells of monocyte lineagerepresenting precursors of the giant cells, and evenly scat-tered giant cells. In many publications, the mononuclearstromal cells are reported to overexpress RANKL and arethought to be the true neoplastic components of GCTB(2, 7, 10, 13). These transformedmononuclear stromal cellshave been shown to express many characteristics of mes-enchymal stem cells that have differentiated along theosteoblast lineage but not to a completely differentiatedosteoblast phenotype, with minimal expression of such

factors as osteocalcin and alkaline phosphatase (14–17, 26, 27). They also show cytogenic markers indicatingtransformation, including telomere associations, but fail toshow consistent chromosomal changes and are reported tobe polyclonal (28–31). Interestingly, GCTB tumors areoccasionally reported to contain scant focal areas of osteoidmatrix and woven bone that is present within the tumorand separate from the tumor margins, in an incidenceranging from 22% to 52% (9, 18, 19, 32). Our own analysisof 31 baseline samples from the phase II study and anarchival set of 40 GCTB samples revealed a similar patternin 45% to 52% of the tumors evaluated, although

Figure 5. H&E, RANKL, and RANKstains (subject 3): A, baselinesection: densely cellular tumor withnumerous tumor giant cellssurrounded by abundant spindle-shaped tumor stromal cells. B, on-study section: transition fromabundant residual densely cellulartumor cell areas to scant regions offibro-osseous intercellular areas. C,baseline section: numerous RANKL-positive tumor stromal cellssurrounding RANKL-negative tumorgiant cells. D, on-study section:abundant residual area of denseRANKL-positive tumor stromal cellsseparated by scant areas of poorlymineralized fibro-osseous matrix. E,baseline section: RANK-positivetumor giant cells surrounded byRANK-negative tumor stromal cells.F, on-study section: abundantresidual densely cellular tumor cellswith numerous RANK-positivemononuclear cells but no RANK-positive tumor giant cells separatedby wide areas of poorly mineralizedfibro-osseous matrix.






localization of the biopsy to the tumor margin was notestablished.

Within the tumor environment, numerous soluble fac-tors that are chemotactic for myelomonocytic cells such asSDF-1, MIP-1a, and csf-1 have been identified (2, 11, 33–35). It is thought that the expression of these factors resultsin the accumulation of RANK-positive monocytic cells ofbonemarrow origin expressing a preosteoclastic phenotype(2–4, 7, 8, 12, 13). Signaling by RANKL on tumor stromalcells activates RANK on the preosteoclasts, inducing giantcell formation and resulting in the destructive osteolysis ofGCTB. Coculture of GCTB stromal cells with circulatingmononuclear cells has been shown to induce osteolyticgiant cells in vitro (32, 36–38).

The importance of the RANK/RANKL signaling pathwayin GCTB is thought to represent a neoplastic analog tonormal bone remodeling in which the RANKL-positiveosteoblast and RANK-positive osteoclasts are coupled toresult in regulated bone deposition and removal. In normalbone, this coupling is regulated as the bone architecture ismodified during growth and changing stresses (39). Inhi-bition of normal bone remodeling by interfering with thispathway has been clearly established (40, 41). The destruc-tive osteolysis that is characteristic of GCTB, in whichdysregulated stimulation of osteoclastic bone remodelingis observed, is also dependent on this pathway. In culturesofGCTB, it has been shown that the activationofmonocytesto form osteoclasts can be inhibited by osteoprotegerin, asoluble decoy receptor for RANKL (42). The results of theGCTB phase II study presented here also clearly show thatthe anti-RANKL antibody denosumab is clinically active inaltering the osteolytic effect of GCTB (20).

Surprisingly, histologic comparison of baseline and on-study samples of GCTB from this study also showed that theclassic architecture of GCTB, consisting of proliferativedensely cellular stromal cells, RANK-positive mononuclearcells, and tumor giant cells, was substantially changed bytreatment. In the on-study samples, a marked increase innonproliferativeosteoidmatrix,wovenbone, andnewbonewas present with only focal areas of proliferative RANKL-positive tumor stromal cells and RANK-positive mononu-clear cells, andwith greatly reduced numbers of tumor giantcells. This indicates that the tumor cell component wasinduced to differentiate toward a nonproliferative osteo-blastic phenotype not typical of untreated tumors, with amarked reduction in osteolysis at the tumor margins. It ispossible that these observations reflect a changing balanceof bone formation/resorption in favor of bone formation,and that the histologic findings are a manifestation of thischange in bonemetabolism. The reports that GCTB stromalcells express many markers of the osteoblastic lineage butminimally express the late-stage markers of osteocalcin andalkaline phosphatase (14, 25) suggest that these neoplasticcells are either incapable of further differentiation, lackcritical differentiation factors, or are being inhibited fromdifferentiation by factors in the tumor environment. How-ever, results from in vitro cultures of GCTB stromal cellsindicate that they are capable of differentiating into a

mature osteoblast phenotype when separated from theosteoclastic component and following treatment with fac-tors such as bonemorphogenic protein (BMP). In addition,GCTB stromal cells can formmature bone when implantedinto immunologically deficient mice (43). We report here,as others have, that limited de novo bonematrix formation isobserved in primary GCTB and also in GCTB that hasmetastasized to the lung (9, 18, 19, 32, 44).

In normal osteoblast–osteoclast coupling, the activatedosteoclast typically releases factors that stimulate bonedeposition (39–44). Because denosumab treatment signif-icantly reduces or eliminates osteoclasts, we would notexpect the increased bone matrix and new bone to beobserved. These observations indicate that the microenvi-ronment within GCTB decouples the interaction of thetumor stromal cells and tumor giant cells, resulting in verylimited bone formation and marked stimulation ofosteolysis.

It is thought that the RANK/RANKL pathway is the finalcommon signal for osteoclast formation from monocytes/macrophages. In contrast, numerous factors have beendescribed that can either stimulate or inhibit osteoblastmaturation such as BMPs and the Ephrin/Eph bidirectionalsignaling between osteoblast and osteoclast (44–49). Theresults of this study indicate that inhibiting RANKL signal-ing inGCTB induces bone formation by tumor stromal cellseither indirectly by eliminating factors associated withtumor giant cells or by direct action on the tumor stromalcell to formbone. Becauseneither published reports nor ouranalyses provide evidence ofRANKexpression on the tumorstromal cell, we would conclude that the increased boneformation induced by denosumab is an indirectmechanism.

Treatment of GCTB by local curettage or resection,accompanied in some cases by local adjuvant treatments(1, 5, 6, 9–11) and by systemic treatment with IFN-a2a ornitrogen-containing bisphosphonates, has been previouslyreviewed (5, 10). Bisphosphonates are reported to causeincreased apoptosis of GCTB stromal cells and tumor giantcells. In one study, no change in the morphologic structureof the tumor was observed; however, several patientsshowed an increase in radiologic density and bone mineralcontent (50).

The striking differentiation of the densely cellular tumortissue into more normal woven bone and connective tissuecomponents posed a limitation in this study; this beneficialbut unexpected changemade it difficult to accurately definethe region of interest that represented what had been theextent of pretreatment tumor when examining the on-studysamples. For this reason, the magnitude of decrease inRANK/RANKL or Ki67 expression and the reduction in thetumor content of tumor stromal cells or the increase infibro-osseous tissue and/or new bone were reported semi-quantitatively. However, examination of the typical histo-logic changes shown in Figs. 1 through 5 and Supplemen-tary Fig. S1 shows striking differences between the baselinesamples and the on-study samples. These observed histo-logic changes are consistent with the radiologic changes

Branstetter et al.





reported previously (20), showing a clear effect of treat-ment. We have also recently observed similar changes inpulmonary metastases of GCTB after treatment with deno-sumab (data not shown).In the phase II study reported here, denosumab treat-

ment caused a marked reduction in tumor giant cells andsignificant histologic evidence of treatment-induced dif-ferentiation of highly cellular proliferative tumor stromalcells to nonproliferative osteoid bone matrix, wovenbone, and mature bone. These findings are consistentwith the tendency of baseline tumors to show small fociof bone matrix before treatment and the published dataindicating that the neoplastic stroma of GCTB can beinduced to form bone when the tumor microenviron-ment is altered. These results confirm and explain theradiologic tumor response previously reported in thisstudy with denosumab (20). Treatment of GCTB withdenosumab offers a novel and effective treatment optionfor this aggressive tumor.

Disclosure of Potential Conflict of InterestD.G. Branstetter has held the role of Pathologist Executive Director at

Amgen. J.-Y. Blay is a consultant and an advisory board member of AmgenInc. S. Chawla is a consultant and advisory board member of Amgen Inc.D.M. Thomas is a recipient of honoraria from Speakers Bureau of Amgen Inc.S. Jun has been Executive Medical Director at Amgen Inc. and has ownershipinterest (including patents). I. Jacobs is an employee and stockholder ofAmgen Inc. No potential conflicts of interest were disclosed by the otherauthors.

Authors' ContributionsConception and design: D.G. Branstetter, D.M. Thomas, S. Jun, I. JacobsDevelopment ofmethodology:D.G. Branstetter, J.-Y. Blay, S. Jun, I. JacobsAcquisitionofdata (provided animals, acquired andmanagedpatients,provided facilities, etc.): D.G. Branstetter, J.C. Manivel, J.-Y. Blay, S.Chawla, D.M. Thomas, S. Jun, I. JacobsAnalysis and interpretation of data (e.g., statistical analysis, biosta-tistics, computational analysis): D.G. Branstetter, S.D. Nelson, J.C. Man-ivel, J.-Y. Blay, D.M. Thomas, S. Jun, I. JacobsWriting, review, and/or revision of themanuscript:D.G. Branstetter, J.C.Manivel, J.-Y. Blay, D.M. Thomas, S. Jun, I. JacobsAdministrative, technical, or material support (i.e., reporting or orga-nizing data, constructing databases): D.G. Branstetter, I. JacobsStudy supervision: S. Chawla, S. Jun, I. Jacobs

AcknowledgmentsThe authors thank Anne-Valerie Decouvelaere of the Centre Centre L�eon

B�erard, Lyon, France, for contributions to the analysis of histologic speci-mens; Jenn Hawkins, Cristina Costales, Ildiko Sarosi (deceased), and CarolJohnson, Amgen Tissue Bank; Li-Ya Huang, Rosalia Soriano, Kathy Rohr-bach, Larry Woody, and Jim Pretorius, Amgen Pathology; Jing Huang,Biostatistics; and the medical writing support of Karen Poole, ELS, of AmgenInc. and Sue Hudson, whose services were funded by Amgen Inc.

Grant SupportThe work was financially supported by Amgen Inc.The costs of publication of this article were defrayed in part by the

payment of page charges. This article must therefore be hereby markedadvertisement in accordance with 18 U.S.C. Section 1734 solely to indicatethis fact.

Received February 17, 2012; revised May 7, 2012; accepted June 6, 2012;published OnlineFirst June 18, 2012.

References1. Mendenhall WM, Zlotecki RA, Scarborough MT, Gibbs CP, Menden-

hall NP. Giant cell tumor of bone. Am J Clin Oncol 2006;29:96–9.2. Atkins GJ, Haynes DR, Graves SE, Evdokiou A, Hay S, Bouralexis S,

et al. Expression of osteoclast differentiation signals by stromal ele-ments of giant cell tumors. J Bone Miner Res 2000;15:640–9.

3. Atkins GJ, Kostakis P, Vincent C, Farrugia AN, Houchins JP, FindlayDM, et al. RANK expression as a cell surface marker of humanosteoclast precursors in peripheral blood, bone marrow, and giantcell tumors of bone. J Bone Miner Res 2006;21:1339–49.

4. Forsyth RG, De Boeck G, Baelde JJ, Taminiau AH, Uyttendaele D,Roels H, et al. CD33þ CD14- phenotype is characteristic of multinu-clear osteoclast-like cells in giant cell tumor of bone. J BoneMiner Res2009;24:70–7.

5. Garcia R, Inwards C, Unni K. Benign bone tumors—recent develop-ments. Semin Diagn Pathol 2011;28:73–85.

6. Goldring SR, Roelke MS, Petrison KK, Bhan AK. Human giant celltumors of bone identification and characterization of cell types. J ClinInvest 1987;79:483–91.

7. Huang L, Xu J, Wood DJ, Zheng MH. Gene expression of osteopro-tegerin ligand, osteoprotegerin, and receptor activator of NF-kappaBin giant cell tumor of bone: possible involvement in tumor cell-inducedosteoclast-like cell formation. Am J Pathol 2000;156:761–7.

8. Roux S, Amazit L, Meduri G, Guiochon-Mantel A, Milgrom E, MarietteX. RANK (receptor activator of nuclear factor kappa B) and RANKligand are expressed in giant cell tumors of bone. Am J Clin Pathol2002;117:210–6.

9. Schwartz H. Giant cell tumor of bone. Compr Ther 1993;19:63–8.10. Thomas DM, Skubitz KM. Giant cell tumor of bone. Curr Opin Oncol

2009;21:338–44.11. Werner M. Giant cell tumour of bone: morphological, biological, and

histogenetical aspects. Int Orthop 2006;30:484–9.12. Zheng MH, Robbins P, Xu J, Huang L, Wood DJ, Papadimitriou JM.

The histogenesis of giant cell tumour of bone: a model of interaction

between neoplastic cells and osteoclasts. Histol Histopathol2001;16:297–307.

13. Skubitz KM, Cheng EY, Clohisy DR, Thompson RC, Skubitz AP. Geneexpression in giant-cell tumors. J Lab Clin Med 2004;144:193–200.

14. Huang L, Teng XY, Cheng YY, Lee KM, Kumta SM. Expression ofpreosteoblast markers and Cbfa-1 and Osterix gene transcripts instromal tumour cells of giant cell tumour of bone. Bone 2004;34:393–401.

15. Nishimura M, Yuasa K, Mori K, Miyamoto N, Ito M, Tsurudome M,et al. Cytological properties of stromal cells derived from giant celltumor of bone (GCTSC) which can induce osteoclast formation ofhuman blood monocytes without cell to cell contact. J Orthop Res2005;23:979–87.

16. Robinson D, Segal M, Nevo Z. Giant cell tumor of bone. The role offibroblast growth factor 3 positive mesenchymal stem cells in itspathogenesis. Pathobiology 2002;70:333–42.

17. Wulling M, Delling G, Kaiser E. The origin of the neoplastic stromal cellin giant cell tumor of bone. Hum Pathol 2003;34:983–93.

18. Emura I, Inoue Y,Ohnishi Y,Morita T, Saito H, Tajima T. Histochemical,immunohistochemical and ultrastructural investigations of giant celltumors of bone. Acta Pathol Jpn 1986;36:691–702.

19. Fornasier VL, Protzner K, Zhang I, Mason L. The prognostic signifi-cance of histomorphometry and immunohistochemistry in giant celltumors of bone. Hum Pathol 1996;27:754–60.

20. ThomasD, HenshawR, Skubitz K, Chawla S, Staddon A, Blay JY, et al.Denosumab in patients with giant-cell tumour of bone: an open-label,phase 2 study. Lancet Oncol 2010;11:275–80.

21. Lewiecki EM. Clinical use of denosumab for the treatment for post-menopausal osteoporosis. Curr Med Res Opin 2010;26:2807–12.

22. Fizazi K, Carducci M, Smith M, Dami~ao R, Brown J, Karsh L, et al. Arandomised, double-blind study of denosumab versus zoledronic acidin the treatment of bone metastases in men with castration-resistantprostate cancer. Lancet 2011;377:813–22.






23. Henry D, Costa L, Goldwasser F, Hirsch V, Hungria V, Prausova J, et al.A randomized, double-blind study of denosumab versus zoledronicacid in the treatment of bone metastases in patients with advancedcancer (excluding breast and prostate cancer) or multiple myeloma. JClin Oncol 2011;29:1125–32.

24. Stopeck A, Lipton A, Body JJ, Steger GG, Tonkin KS, de Boer R, et al.Denosumab compared with zoledronic acid for the treatment of bonemetastases in patients with advanced breast cancer: a randomized,double-blind study. J Clin Oncol 2010;28:5132–9.

25. Gonzalez-Suarez E, Jacob AP, Jones J, Miller R, Roudier-Meyer MP,Erwert R, et al. RANK ligand mediates progestin-induced mammaryepithelial proliferation and carcinogenesis. Nature 2010;468:103–7.

26. Murata A, Fujita T, Kawahara N, Tsuchiya H, Tomita K. Osteoblastlineage properties in giant cell tumors of bone. J Orthop Sci 2005;10:581–8.

27. RobinsonD,Einhorn TA.Giant cell tumor of bone: a uniqueparadigmofstromal-hematopoietic cellular interactions. J Cell Biochem 1994;55:300–3.

28. Butler MG, Dahir GA, Schwartz HS. Molecular analysis of transforminggrowth factor beta in giant cell tumor of bone. CancerGenetCytogenet1993;66:108–12.

29. Horvai AE, Kramer MJ, Garcia JJ, O'Donnell RJ. Distribution andprognostic significance of human telomerase reverse transcriptase(hTERT) expression in giant-cell tumor of bone. Mod Pathol 2008;21:423–30.

30. Schwartz HS, Butler MG, Jenkins RB, Miller DA, Moses HL. Telomericassociations and consistent growth factor overexpression detected ingiant cell tumor of bone. Cancer Genet Cytogenet 1991;56:263–76.

31. Schwartz HS, Eskew JD, Butler MG. Clonality studies in giant celltumor of bone. J Orthop Res 2002;20:387–90.

32. Lau YS, Sabokbar A, Gibbons CL, Giele H, Athanasou N. Phenotypicand molecular studies of giant-cell tumors of bone and soft tissue.Hum Pathol 2005;36:945–54.

33. Baud'huin M, Renault R, Charrier C, Riet A, Moreau A, Brion R, et al.Interleukin-34 is expressed by giant cell tumours of bone and plays akey role in RANKL-induced osteoclastogenesis. J Pathol 2010;221:77–86.

34. Liao TS, Yurgelun MB, Chang SS, Zhang HZ, Murakami K, Blaine TA,etal.Recruitmentofosteoclastprecursorsbystromalcell derivedfactor-1 (SDF-1) in giant cell tumor of bone. J Orthop Res 2005;23:203–9.

35. Zheng MH, Fan Y, Smith A, Wysocki S, Papadimitriou JM, Wood DJ.Gene expression of monocyte chemoattractant protein-1 in giant celltumors of bone osteoclastoma: possible involvement in CD68þ mac-rophage-like cell migration. J Cell Biochem 1998;70:121–9.

36. Joyner CJ, Quinn JM, Triffitt JT, OwenME, Athanasou NA. Phenotypiccharacterisation of mononuclear and multinucleated cells of giant celltumour of bone. Bone Miner 1992;16:37–48.

37. Miyamoto N, Higuchi Y, TajimaM, ItoM, TsurudomeM, NishioM, et al.Spindle-shaped cells derived from giant-cell tumor of bone supportdifferentiation of blood monocytes to osteoclast-like cells. J OrthopRes 2000;18:647–54.

38. Roux S, Quinn J, Pichaud F, Orcel P, Chastre E, Jullienne A, et al.Human cord blood monocytes undergo terminal osteoclast differen-tiation in vitro in the presence of culture medium conditioned by giantcell tumor of bone. J Cell Physiol 1996;168:489–98.

39. Hofbauer LC, Khosla S, Dunstan CR, Lacey DL, Boyle WJ, Riggs BL.The roles of osteoprotegerin and osteoprotegerin ligand in theparacrine regulation of bone resorption. J Bone Miner Res 2000;15:2–12.

40. Dougall WC, GlaccumM, Charrier K, Rohrbach K, Brasel K, De SmedtT, et al. RANK is essential for osteoclast and lymph node development.Genes Dev 1999;13:2412–24.

41. Kong YY, Yoshida H, Sarosi I, Tan HL, Timms E, Capparelli C, et al.OPGL is a key regulator of osteoclastogenesis, lymphocyte develop-ment and lymph-node organogenesis. Nature 1999;397:315–23.

42. Atkins GJ, Bouralexis S, Haynes DR, Graves SE, Geary SM, EvdokiouA, et al. Osteoprotegerin inhibits osteoclast formation and boneresorbing activity in giant cell tumors of bone. Bone 2001;28:370–7.

43. James IE, Dodds RA, Olivera DL, Nuttall ME, Gowen M. Humanosteoclastoma-derived stromal cells: correlation of the ability to formmineralizednodules in vitrowith formationof bone in vivo. JBoneMinerRes 1996;11:1453–60.

44. Kudo N, Ogose A, Ariizumi T, Kawashima H, Hotta T, Hatano H, et al.Expressionof bonemorphogenetic proteins in giant cell tumor of bone.Anticancer Res 2009;29:2219–25.

45. Mundy GR, Elefteriou F. Boning up on ephrin signaling. Cell 2006;126:441–3.

46. Pierroz DD, Rufo A, Bianchi EN, Glatt V, Capulli M, Rucci N, et al.Beta-Arrestin2 regulates RANKL and ephrins gene expression inresponse to bone remodeling in mice. J Bone Miner Res 2009;24:775–84.

47. Sims NA, Gooi JH. Bone remodeling: multiple cellular interactionsrequired for coupling of bone formation and resorption. SeminCell DevBiol 2008;19:444–51.

48. Xing W, Kim J, Wergedal J, Chen ST, Mohan S. Ephrin B1 regulatesbone marrow stromal cell differentiation and bone formation by influ-encing TAZ transactivation via complex formation with NHERF1. MolCell Biol 2010;30:711–21.

49. Zhao C, Irie N, Takada Y, Shimoda K, Miyamoto T, Nishiwaki T, et al.Bidirectional ephrinB2-EphB4 signaling controls bone homeostasis.Cell Metab 2006;4:111–21.

50. Tse LF, Wong KC, Kumta SM, Huang L, Chow TC, Griffith JF. Bispho-sphonates reduce local recurrence in extremity giant cell tumor ofbone: a case-control study. Bone 2008;42:68–73.

Branstetter et al.





2012;18:4415-4424. Published OnlineFirst June 18, 2012.Clin Cancer Res Daniel G. Branstetter, Scott D. Nelson, J. Carlos Manivel, et al. Patients with Giant-Cell Tumor of BoneDenosumab Induces Tumor Reduction and Bone Formation in

Updated version

10.1158/1078-0432.CCR-12-0578doi:

Access the most recent version of this article at:

Material

Supplementary

http://clincancerres.aacrjournals.org/content/suppl/2012/06/18/1078-0432.CCR-12-0578.DC1

Access the most recent supplemental material at:

Cited articles

http://clincancerres.aacrjournals.org/content/18/16/4415.full#ref-list-1

This article cites 50 articles, 5 of which you can access for free at:

Citing articles

http://clincancerres.aacrjournals.org/content/18/16/4415.full#related-urls

This article has been cited by 21 HighWire-hosted articles. Access the articles at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected]

To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://clincancerres.aacrjournals.org/content/18/16/4415To request permission to re-use all or part of this article, use this link



http://clincancerres.aacrjournals.org/lookup/doi/10.1158/1078-0432.CCR-12-0578

http://clincancerres.aacrjournals.org/content/suppl/2012/06/18/1078-0432.CCR-12-0578.DC1

http://clincancerres.aacrjournals.org/content/18/16/4415.full#ref-list-1

http://clincancerres.aacrjournals.org/content/18/16/4415.full#related-urls

http://clincancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]

http://clincancerres.aacrjournals.org/content/18/16/4415


denosumab induces tumor reduction and bone formation in ... · denosumab in reducing...

Documents