Review/Praca poglądowa
Biology and management of myeloma-related bone disease
Evangelos Terpos
1,*, Nikolaos Kanellias
1, Krzysztof Giannopoulos
2,3,41DepartmentofClinicalTherapeutics,NationalandKapodistrianUniversityofAthens,SchoolofMedicine, Athens,Greece
2ExperimentalHematooncologyDepartment,MedicalUniversityofLublin,Poland
3DepartmentofHematooncologyandBoneMarrowTrasplantation,MedicalUniversityofLublin,Poland
4PolishMyelomaConsortium,Poznan,Poland
Introduction
Multiplemyeloma(MM)isaplasmacell malignancywhichis characterizedby thepresence of bone destructiondue to an
elevated function of osteoclasts that is not balanced by a comparable elevation of osteoblast function. This bone destruction develops lytic lesions that lead to bone pain, hypercalcemia and skeletal-related events (SREs) such as pathologicalfractures,requirementforsurgeryand/orradiation article info
Articlehistory:
Received:24.03.2014 Accepted:31.03.2014 Availableonline:13.04.2014
Keywords:
Multiplemyeloma
RANKL
Bisphosphonates
Zoledronicacid
Denosumab
abstract
Bonedisease isoneofthemostcommoncomplicationsofmultiplemyeloma.Itisthe result ofincreased osteoclastactivity whichis not compensatedby osteoblastactivity andleadstoosteolyticlesionscharacterizedbybonepainandincreasedriskforpatholo- gicalfracture,spinalcordcompressionandneedforradiotherapyorsurgerytothebone.
Recent studies have revealednovel pathways and moleculesthat are involved in the biologyofmyelomabonediseaseincludingthereceptoractivatorofnuclearfactor-kappa Bligand/osteoprotegerinpathway,theWntsignalinginhibitorsdickkopf-1andsclerostin, macrophage inflammatory proteins, activin A, and others. A thorough studyof these pathwayshaveprovidednovelagentsthatmayplayacriticalroleinthemanagementof myeloma related bone disease in the near future, such as denosumab (anti-RANKL), sotatercept(activinA antagonist),romosozumab(anti-sclerostin)or BHQ-880(anti-dick- kopf 1).Currently, bisphosphonates are the cornerstonein the treatment ofmyeloma relatedbonedisease.Zoledronicacidandpamidronateareusedinthissettingwithvery goodresultsinreducingskeletal-relatedevents,buttheycannotbeusedinpatientswith severerenalimpairment. Furthermore,theyhave somerarebutseriousadverse events includingosteonecrosisofthejawandacuterenalinsufficiency.Thisreviewpaperfocu- sesonthelatestadvances inthepathophysiologyofmyelomabonedisease andinthe currentandfuturetreatmentoptionsforitsmanagement.
©2014PolskieTowarzystwoHematologówiTransfuzjologów,InstytutHematologiii Transfuzjologii.PublishedbyElsevierUrban&PartnerSp.zo.o.Allrightsreserved.
*Correspondingauthorat:DepartmentofClinicalTherapeutics,NationalandKapodistrianUniversityofAthensSchoolofMedicine, AlexandraGeneralHospital,80Vas.SofiasAvenue,11528Athens,Greece.Tel.:+302132162846;fax:+302132162511.
E-mailaddress:eterpos@med.uoa.gr(E.Terpos).
ContentslistsavailableatScienceDirect
Acta Haematologica Polonica
journalhomepage:www.elsevier.com/locate/achaem
http://dx.doi.org/10.1016/j.achaem.2014.04.010
0001-5814/©2014PolskieTowarzystwoHematologówiTransfuzjologów,InstytutHematologiiiTransfuzjologii.PublishedbyElsevier Urban&PartnerSp.zo.o.Allrightsreserved.
totheboneandspinalcordcompression(SCC)[1].Atdiagnosis 70%ofthepatientspresentwithbonepain,whileduringthe courseofthedisease50%ofthepatientsdevelopatleastone SRE if they do not receive abone-targeted agent [2]. Intwo studies, Melton et al. has shown that MM patients have generalizedbonelossandosteoporosisthatmakethemvulner- abletoosteoporoticfractures[3].Theyalsoshowedthateven patientswithmonoclonal gammopathyofundeterminedsig- nificance(MGUS)havea>2-foldincreaseinfracturerateofthe axial skeleton[4]. Bone disease has aserious impact on the qualityoflifeandsurvivalofMMpatients[5]andaffectsboth clinicalandeconomicaspectsoftheirlife[6].Theriskofdeath inMMpatientswhodevelopapathologicfractureincreasesby 20% in comparison with MM patients without pathologic fractures[7].Thus,itis importantto diagnoseearlyandtreat properly bone disease and its complications. This paper reviews the latest available details of pathophysiology and treatmentofmyelomarelatedbonedisease.
Biology of multiple myeloma bone disease
Intheadultskeleton,skeletalintegrityiscoordinatedbythe synchronizedactivityof threecell types.Osteoblasts create new bone matrix; osteoclasts are responsible for bone resorptionand osteocytesregulateboneturnover.Inmulti- ple myeloma patients, bone disease is the result of an uncouplinginboneremodeling.Itconsistsofanincreasein theosteoclast-mediatedboneresorption,whichiscombined with suppression in the osteoblast, mediated bone miner- alizationanddefectsinosteocytefunctions [8].Untiltoday, several direct and indirect interactions between myeloma and stromal cells in the bone marrow microenvironment havebeenrecognized.Thefactthatosteolyticlesionsoccur close to MM cells suggests that factors secreted by tumor cellsleadtodirectstimulationof osteoclastmediatedbone resorption and inhibition of osteoblast mediated bonefor- mation [9]. In addition, the increased bone resorptive progressleadstothereleaseofgrowthfactorsthatincrease thegrowthofMMcells, leadingtoaviciouscycle oftumor expansion and bone destruction. Apart from this, interac- tionsviaadhesionbetweenMMcellsandbonemarrowcells resultin the production of factors that promoteangiogen- esisandmakethemyelomacellsresistanttochemotherapy [10, 11]. One example is that of T-regulatory and T-helper cells. In MM patients the stimulated T-regulatory cells by myeloma cells up-regulate pro-osteoclastic molecules and havebeen implicatedwithdiseaseprogression,whereas T- helpercellssecrete IL-17whichpromotesosteoclastforma- tion[12–14].Ontheotherhand,Yaccobyetal.showed that osteoblasts inhibitMM cell growthin mostof the patients [15].
Increasedosteoclastactivity
Themainregulatoroftheosteoclaststimulationandactiva- tionisthesystemofthereceptoractivatorofnuclearfactor- kappaB (RANK),its ligand (RANKL) and itsdecoy receptor, osteoprotegerin (OPG). Animportant stepin the osteoclast stimulation is the binding of myeloma cells to the bone
marrowstromalcells(BMSCs).Thisadhesionismediatedby interactions between a4b1 on myeloma cells and vascular cell adhesionmolecule1(VCAM-1) onBMSCs,and leadsto the up-regulation of a variety of pro-osteoclasticcytokines and chemokines which directly or indirectly stimulate osteoclast formation differentiation and activity. These factorsincludeinterleukin-6(IL-6),IL-1a,IL-1b,IL-11,macro- phage-colony stimulating factor (M-CSF), tumor necrosis factor alpha and beta (TNF-a and TNF-b), macrophage inflammatory proteins-1 alpha and beta (MIP-1a and b), parathyroid hormone-related peptide (PTHrP), vascular endothelial growthfactor (VEGF) and others [16–18]. These factorsare excretedbyMMcellsdirectly, orindirectlyafter stimulationofbonemarrowcellsbytheMMcells.
TNFSuperfamilymembers–theRANK/RANKLsignaling pathway
RANKisatransmembranesignalingreceptor.Itislocatedon the surface of osteoclast precursors [19, 20]. RANKL is expressedbyarangeofcelltypes,includingmarrowstromal cells and osteoclasts. Its expression is stimulated by cyto- kinesthatstimulateboneresorption[21]suchasparathyroid hormone (PTH), 1,25-dihydroxy vitamin D3 and prostaglan- dins [22, 23]. RANKL binds to its receptor on osteoclast precursors and stimulates osteoclast differentiation forma- tionandsurvival.Thesefunctionsaremediatedthroughthe nuclearfactorkappa-B(NFkB)andp38MAP-kinasepathways.
Apart fromthis, RANKLhas direct enhancementeffects on mature osteoclasts thatinhibit theirapoptosis. The impor- tance of the role of RANKLin osteoclastogenesis hasbeen shown in RANKL or RANK gene knockout mice. These animals lackosteoclastsandasaresulttheydeveloposteo- petrosis[24–27].IntheabsenceofRANKLalmostnochemo- kinewithosteoclastactivitycanact.
OPG,anothermemberoftheTNFreceptorsuperfamily,is a soluble decoy receptor for RANKL[28]. It is producedby several cells, including osteoblasts, and interacts with RANKL, causing inhibition of its action, thereby reducing osteoclastogenesis. The important role of OPG has been shown in studieswith knock-outmice. OPG deficient mice develop severe osteopenia and osteoporosis [29–31]. An abnormal RANKL/OPG ratio is found in the majority of malignantbonedisorders[32].
MyelomacellsturnthebalanceoftheRANKL/OPGratioin favorofRANKL.Inthebonemarrowmicroenvironment,MM cellsplayadoublerole:theyinducetheexpressionofRANKL from stromal cells, while they directly express RANKL, although in low amounts [33–37]. Apart from this they decreasetheOPGavailabilitywithinthebonemarrowmicro- environment. Thisismaintainedintwodifferentways.The MMcellsreduceOPGsecretionfromosteoblastsandstromal cells.Inaddition,theyremovetheremainingOPGbylysoso- mal degradation [38, 39]. The up-regulation of RANKL, in combination with down-regulation of OPG, leads to the formationandactivationofosteoclasts.LevelsofRANKLand OPG have been shown to correlate with clinical activity of MM, severity of bone disease and poor prognosis[40]. In individuals with MGUS, the RANKL/OPG is also increased when compared to that in control subjects but remains significantlylower thanthatinpatientswith myeloma [41],
partly explaining the higher incidence of osteoporosis in thesepatients.
Macrophageinflammatoryproteins-1alphaandbeta (MIP-1a,-b)
Thesetwocytokinesplayanimportantroleinthebiologyof myeloma bone disease. Both are produced and secreted by MM cells. MIP-1a, also known as chemokine (C-C motif) ligand3(CCL-3),isalowmolecularweightchemokine,which belongstotheRANTES(regulatedonactivation,normalTcell expressed and secreted) family of chemokines, primarily associated with cell adhesion and migration. MIP-1a is chemotacticformonocytesandmonocyte-likecells,including osteoclastprecursors. MIP-1a induces latestage differentia- tioninhumanosteoclastprogenitorsandpromotesosteoclast formationinadose-dependentwayinbonemarrowcultures [17,42–44].MIP-1aenhancestheeffectsofRANKLandIL-6on osteoclastformation[45].However,ithasalsobeenshownto enhanceosteoclastformationindependentofRANKL.MIP-1a andMIP-1b enhance theRANKLexpression instromalcells [27].MIP-1bisahighlyhomologouschemokineofMIP-1athat similarly to MIP-1a induces the development of osteolytic bonelesions[46].
Both MIP-1a and MIP-1b are produced and secreted by myelomacells.MIP-1am-RNAhasbeendetectedinMMcells, whileMIP-1aproteinwasfoundelevatedinthemicroenvir- onmentofMMpatientsinwhomitcorrelatedwithstageand disease activity. MIP-1a was also elevated in the blood of myeloma patients with severe bone disorders, but not in MGUSpatientswithincreasedboneresorption[41,47].Gene expressionprofilingshowedthatMIP-1aisoneofthegenes thatishighlycorrelatedwithbonedestructioninMM[48].
Furthermore,MIP-1ahasdirect actiononmyelomacells, since they express the receptor CCR5, promoting growth, survival and migration of myeloma cells [49]. MIP-1a up- regulatestheexpressionofb1integrinonMMcells,increas- ing adhesive interactions between MM cells and marrow stromalcells.ThisresultsinincreasedproductionofRANKL, IL-6, VEGF and TNF-a by marrow stromal cells, which further enhances MM cell growth, angiogenesis and bone destruction[27].
Interleukin-3
IL-3 mRNA levels were foundto be increased in myeloma cells andIL-3 protein levels werefoundto beincreased in bonemarrowplasmafromMMpatients.IL-3incombination withMIP-1aorRANKLsignificantlyenhanceshumanosteo- clastformation andboneresorption comparedwithMIP-1a orRANKLalone.IL-3alsostimulatesthegrowthofmyeloma cells independently of the presence of IL-6. These data suggestthatincreasedIL-3levelsarepresentinthemarrow microenvironment of myeloma patients, increasing bone destructionandtumorcellgrowth[50,51].
Interleukin-6
IL-6 is a growth factor for both osteoclasts and myeloma cells,promotingtheirsurvivalandpreventingtheirapopto- sis. IL-6 causes an increase in the osteoclast precursors, which leads to the increase in the number of mature osteoclasts. The levels of circulating IL-6 and its receptor
(IL-6R) are increased in MM and correlate with stage, advanced myeloma featuresand disease-free survival [52].
Levels of IL-6 are elevated in MM patients with osteolytic bone disease when compared with MM patients without bonedisease,aswellasinpatientswithMGUS[53].
Interleukin-1b
IL-1bhaspotentosteoclastogenesisactivity:itenhancesthe expressionofadhesionmoleculesandinducesparacrineIL- 6 production, resulting inosteolytic disease. IL-b has been foundtobeincreasedinmyelomacellcultures[50].Elevated IL-1b m-RNA levels were also detected in MM patients, while anti-IL-1b antibodies failed completely to abolish osteoclastogenesisactivityofmyelomabonemarrow[54].
Tumornecrosisfactoralpha(TNF-a)
High plasma levels of TNF-a have been found in patients with MM [55]. TNF-a causes proteolytic breakdown of I-kappaB(theinhibitorofNF-kB),leadingtoNF-kBactivation and enhancement of genetranscription, including IL-6 and adhesion molecules,which are involvedinpromoting bone resorption[56].
Hepatocytegrowthfactor(HGF)
Myeloma cells can transform HGF to its active form. HGF plays animportant rolein osteoclastactivationand angio- genesis. HGF can up-regulate the osteoclast-like cell- mediatedIL-11expression[57].
Vascularendothelialgrowthfactor(VEGF)
VEGF plays a major role in tumor neovascularization and hasbeenrecentlyimplicatedinosteoclastogenesisinMM.It isexcretedbymyelomacellsandbindstoVEGFR-1receptor that ismainly expressedbyosteoclasts.Ithasadirect role in enhancing osteoclast function and survival [11]. VEGF stimulates the IL-6 production by stromal cells, while IL-6 enhances VEGFsecretion by myelomacells, suggesting the existence ofparacrine interactionsamong stromal andMM cells[58].
Osteopontin
Osteopontin is a non-collagenous matrix protein which is produced by differentcells including osteoblasts and mye- loma cells. It is involved in tumor metastasis, adhesion, apoptosis and angiogenesis. Marrow cells from myeloma patientswithadvanceddiseaseproducedincreasedlevelsof osteopontincomparedwiththatfromasymptomaticMMor MGUS patients. Furthermore, plasma osteopontin levels of MM patientsweresignificantlyhigher than those of MGUS and controls, and correlatedwith bothdisease progression and bone destruction. These observations suggest that myeloma cells actively produce osteopontin, which contri- butestoosteoclasticboneresorption[59].
Stromal-derivedfactor-1a(SDF-1a)
SDF-1a is another chemokine which is expressed by both stromal and myeloma cells. MM patients have elevated plasmalevels of SDF-1a whencompared withnormal,age- matched subjects. The SDF-1a levels have been correlated withmultipleradiologicalosteolyticlesionsinMMpatients.
SDF-1a binds to its receptor CXCR4, which is widely expressed on leukocytes, mature dendritic cells, osteoclast precursors,andmyelomacells,andup-regulatestheexpres- sionofthematrixdegradingenzyme,matrixmetalloprotei- nase 9 (MMP-9),promoting the recruitment, migration and activationoftheosteoclasts[27].
Parathyroidhormone-relatedprotein(PTHrP)
PTHrPisproducedbyanumberoftumorsthatgrowinbone andmediatesthedevelopmentofbonemetastases,particu- larly of breast and lung cancer [60–62]. It is possible that PTHrPstimulatesboneresorptionandmediatesitseffectby up-regulating RANKL in osteoblasts via the PTH-R1 [62].
PTHrP has been shown to beexpressed by myeloma cells andPTHrPsignaling,viathePTH-RI,increasesexpressionof RANKLinmyelomacells[63,64].
Osteoblastsuppression
Theinhibition of osteoblasts isanother crucial stepinthe pathogenesisofmyelomabonedisease.Asmyelomaburden increases, osteoblast-driven bone formation is suppressed whichfurtherresultsinthedevelopmentofosteolyticbone lesions. Osteoblast suppression is maintained even in patients in long-term remission. Osteoblast inhibition is maintainedthroughthesecretionofcytokines,whichisthe resultof interactions between MMcells and osteoblasts or osteocytes[55].
Wntsignalingpathway
The osteoblastfunctionismaintained byseveralpathways, including the canonical Wingless-type(Wnt) pathway.Wnt proteinsbindtotheWntreceptoranditsco-receptorsLRP5/
LRP6andlead toastabilizationofb-catenin. Thisresultsin the increase of cytoplasmic levels of b-catenin, leading to translocation into the nucleus. This event stimulates the expressionof osteoblastic target genes [65]. When theWnt signalisabsent,b-cateninisphosphorylatedanddegradedby theproteasome.WntantagonistspreventthebindingofWnt glycoproteins to their receptors and include the following molecules [66]. Members of the dickkopf (DKK) family and sclerostinbindto theLRP5/LRP6component, whilesecreted frizzled-relatedproteins(sFRP),forexamplesFRP-2andsFRP- 3,bindtoWntproteins.BothresultinasuppressionofWnt signalingandareducedosteoblastfunction.
DKK-1issecretedbymyelomacellsandhasbeenshown toinhibitdifferentiationofosteoblastprecursorcellsinvitro.
In MM patients with lytic lesions, immunohistochemical analysisofbonemarrowbiopsiesshowedthatmyelomacells overexpressDKK-1.Infact,bonemarrowplasmafromnewly diagnosedMMpatientscontainsnearly3times moreDKK-1 protein compared to that from control subjects: marrow plasmafrom patientswithMMthatcontained>12ng/mlof DKK-1inhibitedosteoblastdifferentiation.Furthermore,gene expression levels of DKK-1 correlated with extensive bone disease[67].DKK-1isincreasedintheserumofMMpatients [68]andcorrelateswiththeextentofbonedisease[69].Serum DKK-1 decreases in myeloma patients who respond to therapy,butnotinthosewhodidnotrespond[70].DKK-1is secretedinvivomainlybymyelomacells.Furthermore,since
WntsignalinginosteoblastsincreasestheexpressionofOPG and downregulatestheexpressionof RANKL[71,72],inhibi- tion of Wnt signaling promotes osteoclastogenesis. Taken together,DKK-1seemstobeakeyregulatorofbonemetabo- lisminmyeloma.
Soluble FRP-2 is secreted from MM cells and inhibits mineralizednoduleformationand osteoblastdifferentiation induced by bone morphogenetic protein 2 (BMP-2) [73]. It inhibits osteoblastic differentiation at multiple steps, not only early osteoblastic differentiation to express alkaline phosphatase (ALP), but also terminal differentiation to acquire mineralizingproperties.Itisthoughttobeadecoy receptor that interferes with Wnt binding to its receptor, Frizzled. MM patients with advanced bone disease had elevatedexpressionofsFRP-2intheirmyelomacells[73].
Finally, sclerostin is a cysteine-knot-containing protein, which is produced by osteocytes, inhibits canonical Wnt pathwayandthusinhibitsosteoblastfunction[74].Circulat- ing sclerostin reflect bonemarrow plasmasclerostin levels [75]. Inpatients withmultiplemyeloma,sclerostin isover- produced in the marrow microenvironment either by the myeloma cells [76] or the osteocytes, and its circulating levels correlatewithadvanced bone disease and abnormal boneremodeling[77].
Activin-A
Activinisamemberofthetransforminggrowthfactor-beta (TGF-b) superfamily with complex effects on the bones.
Activin-A hasbeenshowntoinhibitboneformationandin some studies promote osteoclastic bone resorption, although thismayprovetobecontextspecific.Activin-Ais increased in the bone marrow of patients with myeloma and serum level is increased in patientswith newly diag- nosed myeloma andassociated withelevated boneresorp- tion [78,79].Activin-A signalingoccursthroughtheActivin A type IIA receptor toinhibit osteoblastic boneformation.
Blockingactivin-A signalingusingasolubleActRIIAmurine Fc fusionprotein (ActRIImuRc)has been shown toprevent activinAmediatedosteoblastsuppression,buthasnoeffect onosteoclastformationinvitro[80].
Diagnosis and monitoring of myeloma bone disease
Thediagnosticproceduresthatareusedwidelytodayinthe diagnosis and monitoringof myelomarelatedbonedisease include conventional radiography, computed tomography (CT),magnetic resonanceimaging(MRI) andpositronemis- siontomography/CT(PET/CT).
Conventionalradiography
It isthe standard diagnostic procedure for the detection of skeletal involvement, although it lacks sensitivity as it requires a 30–50% of the trabecular bone loss to reveal a detectable lytic lesion [81]. The skeletal survey should includeanteroposteriorandlateralviewsoftheskull,poster- oanteriorviewofthechest,anteroposteriorandlateralviews ofthethoracic,lumbarandcervicalspine(includinganopen
mouthview), humeri andfemora, and anteroposteriorview ofthepelvis.Inaddition,symptomaticareasshouldalsobe specifically visualized. Approximately 75% of patients with multiplemyelomahave abnormalskeletal radiographs.The mostcommonsitesinvolvedincludethecentralskeleton,the skull,andthefemur,whereasinvolvementofdistalbonesis notveryfrequent[27].Theosteolyticlesionsofmyelomaare well circumscribed, and sclerosis of surrounding bone is usuallyabsent.Inapproximately15%ofpatients,generalized osteopeniaistheonlybonemanifestationofmyeloma[82].In cases of disease progression, skeletal survey should be performedagain[83].
Computedtomography
Itisasensitivetoolforthedetectionofthebone-destructive effects in MM as it can detect small osteolytic lesions, unseenwithplainradiographs.CT isnotnecessaryforthe initialstaging ofpatientswithMM,but isuseful fordirect- ingneedlebiopsyfor histologicaldiagnosis.UrgentCTmay beusedwhenSCCissuspectedand MRIiscontraindicated (intraorbital metallic foreign bodies, cardiac pacemakers) dueto patientintolerance or unavailability[27, 84]. A new CT technique for the whole body with low dose radiation (LDWBCT) has now been recognized as a simple and very sensitive method for the depiction of lytic lesions in myelomapatients,butitsvalueisstillunderinvestigation.
Magneticresonanceimaging
Itcansamplealargevolumeofbonemarrowanddepictbone marrow abnormalities in MM with greater sensitivity than conventionalradiographyand CT.MRI shouldbeperformed inall MM patients with negative skeletal survey[83]. Focal lesions are identified in more than one half of patients lackingosteolysisinplainradiography.Theconverse,detec- tionoffocallesionsonplainradiographywithoutcorrespond- ingMRIabnormalities,wasseenin20%ofpatients[27].MRI isausefultoolinthedetectionandstagingofnonsecretory and macrofocal myeloma or relapse. Whole spine MRI is astagingtoolinpatientswithsolitaryplasmacytomaofbone, irrespectiveofthesiteoftheindexlesion.MRIplaysarolein determining the infiltration of the bone marrow and the adjacentsofttissuestructures.This canlead todetectionof bonemarrowalterations,beforebonedestructionisdetected inconventionalradiographyorinCTscans[84–86].Myeloma- tous lesions of bone marrow can be classified into three patterns: focal, diffuseand variegated. MRI pattern of bone marrowinvolvementcorrelateswithprognosisinMM.Dimo- poulos et al. found that patients with diffuse pattern had a median survival of 24 months; patients with variegated pattern had 52 months, patients with focal pattern had 51 months while those with normal pattern had 56 months (p=0.001) [87]. MRI images accurately reflect response to treatmentbyshowingadecreaseorresolutionoffocallesions seenoninitialstudies,whereaslyticlesionsareseenonCT even if a patient has complete remission (CR) on MRI.
Resolution of diffuse disease can also be identified [87].
Complete response to therapy, as accessed by MRI, favors prolongedsurvival, especiallyamongpatientswith ahigher
number of focal lesions [87]. These data justify the wider application of MRI in MM, as the appropriate imaging tool that permitsearly detection of eventuallydevastating focal lesions and asan independentstaging toolwith prognostic implications[83].
Positronemissiontomography/computedtomography (PET/CT)
PET/CT is a technique that combines both anatomical and functional characteristics. It consists of the injection of labeledradiopharmaceuticalssuchasFDG,followedbytomo- graphicimaging.Itcombinesahighresolutioncontrastofpet alongwith ahighresolutionofCT.Focallesions showhigh glucose utilization, due to their high metabolic rate [88].
Lammeren-Veneva et al. showed that, in comparison with conventionalradiography,PET-CTrevealedmorelyticlesions withtheexceptionofthoseintheskull[89].Inanotherstudy, PET-CTwasfoundtohave92%specificityand85%sensitivity in the detection of myelomatous involvement. This study demonstrated the superiority of PET-CT inthe detection of extramedullary diseaseincomparison withMRI and radio- graphicbonesurvey.PET-CTcanalsoplayanimportantrole in the assessmentof responseto treatment and mainlyin the mostaccuratedefinitionof stringentcompleteresponse inMM[90].
Bonemarkers
With the exception of MRI, imaging modalities do not provide information about the rate of the bone turnover.
Bone remodeling in MM patients has been tried to be monitoredthroughbiochemicalmarkers.Theassessmentof boneresorptionhasbeenmademainlythroughtheurinary andserumproductsofbonecollagendegradation.Theseare the C-and N-terminalcross-linkingtelopeptide ofcollagen type-I (NTX, and CTX orICTP, respectively)and the serum levels of tartrate-resistant acid phosphatase type-5b (TRACP-5b), an enzyme produced by activated osteoclasts.
For the evaluationof boneformation wecan evaluatetwo enzymes that are produced by the osteoblasts: the bone- specific ALP (bALP) and the osteocalcin(OC) [84]. Coleman et al.have shown that high levels of NTX correlatedwith increasedriskofdiseaseprogressionandskeletalcomplica- tions in comparison with low NTX levels (p<0.001) [85].
The levels of bALP correlated withriskof negativeclinical outcomes. Urinary NTX and serum ICTP are sensitive markersfortheidentificationofpatientswithincreasedrisk ofearlybonediseaseprogression[86].
Treatment of myeloma bone disease
RadiationtherapyRadiotherapyismainlyusedforthemanagementofsolitary plasmacytoma,andlessfrequentlywhenthereisevidenceof symptomatic SCC, extensive and symptomatic lyticlesions, andforthepreventionofpathologicfractures.Approximately 20% of patients with MM required radiation therapy inthe
past,butsincecurrentnovelagentsworkrapidlytheneedfor palliativeradiotherapyhasbeendecreased[91].Itshouldbe clearthatradiotherapycanleadtodelaysinothertreatments including drugs that may be effective anti-MM agents but quiteradiosensitizingsuchasanthracyclinesandproteasome inhibitors.TheInternationalMyeloma Workinggroupguide- linessuggestthatalow-doseradiation therapy(upto30Gy) is useful as palliative treatment for uncontrolled pain, for impending pathologic fracture, or impending SCC [92].
Upfrontexternalbeamradiationtherapyisusefulforpatients withplasmacytoma,extramedullarymassesandSCC.How- ever, radiotherapy for palliation and local disease control should be used with caution taking into consideration the priorhistoryoftreatment,responseandtheneedforurgent response.Itshouldbelimited,inordertosparethepatient's marrowfunction.Novelagentshavedecreasedtheneedfor palliativeradiotherapy.
Kyphoplastyandvertebroplasty
Vertebroplasty consists of percutaneous injection of poly- methylmethacrylate (PMMA) into the vertebral body under fluoroscopy guidance, and is used in the treatment of painful vertebral compression fractures (VCFs). Approxi- mately80% of patients withpain unresponsive tomedical treatmentexperience pain relief[93]. The role of vertebro- plasty has not been studied in MM patients. Kyphoplasty representsamodificationofvertebroplasty,whereaballoon is inflated prior to PMMA injection. This can stabilizethe fractured vertebral body, reduce kyphotic deformity and restore vertebral height [94, 95]. According to the latest IMWG guidelines, balloon kyphoplasty (BKP) should be considered for symptomatic VCFs and is the procedure of choice to improve QoL in patients with painful VCFs [92].
Howeverriskforsubsequent fracturesignificantlyincreases inpatients undergoingvertebroplasty or kyphoplasty com- pared with that in patients with previous VCFs who were nottreatedwitheitherprocedures[96].
Surgery
ThereshouldbeaclosecooperationandcontinuousOrtho- pedicconsultationregardinglong-bonefractures,bonycom- pressionof the spinalcord, or vertebral columninstability (gradeD). Considerationand indicationsfor surgeryshould bedoneinconsultation withthetreatingoncologist/hema- tologistand theorthopedicandneurosurgeontodetermine whenMMtreatmentcanbesafelyrestarted[92].
Bisphosphonates
Bisphosphonatesareartificialanalogsofpyrophosphates.In comparison withnatural pyrophosphates, bisphosphonates areresistanttophosphataseinducedhydrolysis[97].Bispho- sphonates cause osteoclast suppression. They bind to cal- cium containing molecules such as hydroxyapatite [98].
Osteoclast-induced bone resorption causes exposure of hydroxyapatite.Bisphosphonatesbindtotheexposedmole- culesofhydroxyapatite.Thisfactleadstoincreasedconcen- trationofbisphosphonateswithinthelyticlesions[98–100].
There are two main groups of bisphosphonates,each with adifferentlyproposedmechanismofaction[98].Non-nitro- gencontainingbisphosphonatesinduceosteoclastapoptosis viatheircytotoxicATPanalogs.Ontheotherhand,nitrogen containing bisphosphonates downregulate osteoclast activ- ity by inhibiting the HMG-CoA reductase pathway. Etidro- nate and clodronate are non-nitrogen containing bisphosphonates.Zoledronicacid,ibandronate,pamidronate and risedronate are nitrogen-containing bisphosphonates.
All bisphosphonates have similar physicochemical proper- ties;however,theiranti-resorbingactivityisdifferent.Their activity is drastically increased when an amino group is enteredinto thealiphatic carbonchain.Thus,pamidronate is 100- and 700-fold more potent than etidronate, while zoledronicacidandibandronatehave10000-to100000-fold higher potency than etidronate, both in vitro and in vivo [101].Bisphosphonatesalsoappeartoaffectthemicroenvir- onment in which tumor cells grow and may have direct anti-tumor activity [102–107]. Possible mechanisms include thereductionofIL-6secretionbyBMSCsortheexpansionof gamma/delta T-cells with possible anti-MM activity. The aim of bisphosphonates use is the reduction of SREs in patientswithmyelomabonedisease[27].
Etidronate
Etidronate was found to be ineffective in two placebo- controlledstudiesinmyelomapatients[108,109].
Clodronate
Twomajor,placebo-controlled,randomizedtrialshavebeen performed to date in MM. Lahtinen et al. reported the reductionofnewosteolyticlesionsbyapproximately50%in myeloma patients who received oral clodronate for two years. Thebenefits of clodronate wereindependent of the presence of lytic lesions at baseline [110, 111]. McCloskey etal.showedasurvivaladvantageinpatientswhoreceived clodronate and who did not have vertebral fractures at diagnosis (59 vs. 37 months), even though there was no differenceregardingoverallsurvivalinthetwogroups.After one year of follow-up, both vertebral and non-vertebral fractures aswell as thetime tofirst non-vertebralfracture and severe hypercalcemia were reduced in the clodronate group. At two years, the patientswho received clodronate had better performance status and less myeloma-related painthanpatientstreatedwithplacebo[112,113].
Pamidronate
Pamidronate is an aminobisphosphonate which has been administered either orally or intravenously. In one trial, patients withatleast one lyticlesionandadvanceddisease were randomized to placebo or intravenous pamidronate [114,115].Inthepamidronategroup,therewasareductionin time to the firstskeletal event and in the total number of SREs per year. At nine months, the incidenceof SREs was nearly 50%lower inMMpatients treated with pamidronate comparedwithplacebo(24%vs.41%,respectively;p<0.001), and at 21months the difference remainedsignificant.Pain scoresandqualityoflifewerealsosignificantlyimproved in thepamidronategroup.Inanothertrialpatientswererando- mized to receive either placebo or oral pamidronate, in
additiontoconventionaltherapy.Therewerenoreductionin SREsalthoughtherewasreductioninseverepain.Theoverall negative result of the study was possibly due to the low absorptionoforallyadministeredbisphosphonates[116].
Zoledronicacid
Ina randomized trialBerensonet al.compared the effects of zoledronic acid and pamidronate. Both pamidronate at a dose of 90mgand zoledronic acid at doses of 2mgand 4mgincomparisonwithzoledronicacid0.4mgsignificantly reducedSREs [117]. Thistrial didnotshow anysuperiority of zoledronic acid in comparison with pamidronate, in terms of SREs in myeloma population. In a large rando- mizedphaseIII,double-blind,studytheeffectsofzoledronic acidandpamidronatewerecompared[118].Regarding time to first SRE there were no differences between the study groups.Patientstreatedwithzoledronicacid(4mg)showed slightly lower skeletal morbidity rate. However,the use of radiationtobonewassignificantlylowerinpatientstreated with 4mg zoledronic acid compared with pamidronate. In patients treated with zoledronic acid in comparison with those treatedwithpamidronate,the levels of NTXshowed better normalization. A subsequent analysis of data from a long-term (25-month) extension phase of this study confirmed the equivocal findings that zoledronic acid and pamidronate had similar efficacy in reducing the risk of SREsinMMpatients[119].Thereisarelativelyrecentstudy exploring the role of zoledronic acid in patients with asymptomaticmyeloma.Nodifferencewasobservedregard- ingthe timetoprogression tosymptomatic diseaserequir- ing chemotherapy between patients receiving zoledronic acid and the patients who were only observed. However, SREswerereducedinthezoledronicacid groupat progres- sion(55.5%)vs.theobservationgroup(78.3%;p=0.041)[120].
InMRC-IXstudy,therewasacomparisonbetweenintrave- nous zoledronic acid (4mg every 3–4 weeks or at doses accordingtocreatinine clearancerates)andoralclodronate (1600mg orally daily) in newly diagnosed patients with symptomaticMM(n=1960evaluableforefficacy).Zoledronic acid reduced the incidence of SREs in both myeloma patients with and without bonelesions as assessed using conventional radiography, compared to clodronate [121, 123]. Themediannumber ofSREs after amedianperiodof 3.7 years was 35% for patientsreceiving clodronate versus 27% of patients receiving zoledronic acid (p=0.004). More importantly,zoledronicacidreducedmortalityandextended median survival. Further subset analysis showed that this treatmentextended survivalby10 monthsoverclodronate for patients with osteolytic disease at diagnosis, whereas myeloma patients without bone disease at diagnosis as assessed using conventional radiography had no survival advantagewithzoledronicacid [122].Theseresultsconfirm preclinicalstudies suggesting indirect and direct anti-mye- lomaeffectsofzoledronicacid[123].
Bisphosphonatesadverseevents
Even though bisphosphonate therapy is well tolerated in patientswithMM, clinicians shouldbealert for symptoms andsignssuggestingadverseevents(AEs)andpatientsand
healthcare professionals should be instructed on how to prevent and recognize AEs. Potential AEs associated with bisphosphonate administration include hypocalcemia and hypophosphatemia,gastrointestinaleventsafteroraladmin- istration, inflammatory reactions at the injection site, and acute-phase reactions after IV administration of aminobi- sphosphonates. Renal impairmentand ONJ represent infre- quentbutpotentiallyseriousAEswithbisphosphonateuse.
Hypocalcemia
Hypocalcemia is usually relatively mild and asymptomatic withbisphosphonateuseinmostMMpatients.Theincidence ofsymptomatichypocalcemiaismuchlowerinMMpatients compared to that in patients with solid tumors. Although severe hypocalcemia has been observed in some patients [124]theseeventsareusuallypreventableviatheadministra- tionoforalcalciumandvitaminD3.Patientsshouldroutinely receive calcium (600mg/day) and vitamin D3 (400IU/day) supplementation since 60% of MM patients have vitamin Ddeficiencyorinsufficiency[125,126].InvitaminDdeficient patients there isan increase inbone remodeling. This fact shows that MM patients should be calcium and vitamin D sufficient[127]. Calciumsupplementationshouldbe used withcautioninpatientswithrenalinsufficiency.
Renalimpairment
Bisphosphonateinfusionsareassociatedwithbothdose-and infusionrate-dependenteffectsonrenalfunction.Thepoten- tial for renaldamageis dependenton the concentration of bisphosphonate in thebloodstream, and the highestrisk is observedafteradministrationof highdosagesorrapidinfu- sion. Both zoledronic acid and pamidronate have produced acute renal damage or increases in serum creatinine [115, 119].Patientsshouldbeclosely monitoredfor compromised renal function by measuring CrCl before administration of each IVbisphosphonate infusion. Current guideline recom- mendations[92]statethatthedosagesofzoledronicacidand clodronate, when administered intravenously, should be reducedforpatientswhohavepreexistingrenalimpairment (CrCl30–60mL/min)buttherearenoclinicalstudiesdemon- stratingtheefficacyof thisapproach.ForpatientswithCrCl between 30 and 60mL/min,zoledronic acid doseshould be adjusted.The effectofzoledronicacidhasnotbeenstudied in patients presented with severe renal impairment (CrCl
<30mL/min),andit isnot recommendedfor thesepatients.
We suggest that pamidronate may be given at a dose of 90mginfusedover4–6hformyelomapatientswithosteoly- ticdiseaseandrenalinsufficiency.Furthermore,serumcrea- tinineand CrClshouldbemeasured beforeeachinfusionof pamidronate or zoledronic acid, while BPs should not be administered inshort infusiontimes(<2h for pamidronate and less than 15min for zoledronic acid). Bisphosphonate therapycanberesumedafterwithholdingzoledronicacidor pamidronate for patients who develop renal deterioration duringtherapy,whenserumcreatininereturnstowithin10%
ofbaseline[92].
OsteonecrosisoftheJaw
Itisanuncommoncomplication ofintravenousbisphospho- nates.Itispotentiallyseriousanditsmaincharacteristicisthe
development ofexposed bone inthemouth. Incidencemay varyfrom2to10%[128–130].Riskfactorsincludetheinvasive dental procedures, poor oral hygiene, older age, increased durationandnumberofbisphosphonateinfusions,andzole- dronic acid use [129, 130]. In approximately one half of patients,ONJlesionswill heal[131],but inthe other50%of patientswhorestartbisphosphonate afterhaving stoppedit, recurrence of ONJ will develop. According to recent IMWG guidelines preventivestrategies should be adopted toavoid ONJ[85].Adentalexaminationisnecessarybeforebeginning the bisphosphonate course. Patients should also be alerted regardingdentalhygiene.Allexistingdentalconditionshould be treatedbeforeinitiation ofbisphosphonate therapy. After bisphosphonate treatment initiation, unnecessary invasive dentalproceduresshouldbeavoided,anddentalhealthstatus should be monitored on an annual basis. Patients' dental health status should be monitored by a physician and adentist.Dentalproblemsshouldbemanagedconservatively ifpossible. Ifinvasive dentalproceduresarenecessary there shouldbetemporarysuspensionofbisphosphonatetreatment.
The panel consensus suggests the interruption of bispho- sphonatesbeforeandafterdentalproceduresforatotalof180 days (90days before and 90 days after procedures such as tooth extraction, dental implants and surgery to the jaw).
Bisphosphonatesdonot needtobe discontinued forroutine dental procedures including root canal. Initial treatment of ONJshouldincludediscontinuationofbisphosphonatesuntil healingoccurs.Thephysicianshouldconsidertheadvantages anddisadvantagesof continuedtreatmentwith bisphospho- nates, especially in the relapsed/refractory MM setting. Pre- ventivemeasuresduringbisphosphonatetreatmenthavethe potential to reduce the incidence of ONJ about 75% [132].
ProphylacticantibiotictreatmentmaypreventONJoccurrence after dental procedures [133]. Management of patients dependsonONJstage.StageI(asymptomaticexposedbone;
nosofttissueinfection) canbemanagedconservativelywith oral antimicrobial rinses. Stage II (exposed bone and asso- ciated pain/swelling and/or soft tissue infection) requires culture-directed long-term and maintenance antimicrobial therapy,analgesicmanagementandoccasionally,minorbony debridement. Stage III disease (pathological fracture and exposed bone or soft tissue infection not manageable with antibiotics)requiressurgical resectioninordertoreducethe volumeofnecroticboneinadditiontothemeasuresdescribed instageII[134].WhenONJoccursinitialtherapyshouldinclude discontinuationofbisphosphonatesuntilhealingoccurs[98].
The administration of medical ozone (O3) as an oil suspensiondirectlytotheONJlesionsthatarebelow2.5cm may be another possible therapeutic strategy for those patients who fail to respond to conservative treatment. In such patients, thereare reports suggesting thatONJlesions resolvedwithcompletereconstitutionoforalandjawtissue, with3–10applications[135,136].Inaddition,treatmentwith hyperbaricoxygenhasbeenreportedtobehelpful.
Futuretreatmentoptions
RANKLantagonists
PreclinicalmodelsofMMdemonstratedthatRANKLinhibition canpreventbonedestructionfromMM.RANKLinhibitionwith
recombinant RANK-Fcproteinnot onlyreducedMM-induced osteolysis,butalsocausedamarkeddeclineintumorburden [35,137].SimilarresultswereobtainedusingrecombinantOPG for the treatment of MM-bearing animals [138]. These data gavetherationale forusingRANKL inhibitionintheclinical setting.
Denosumab, a fully human monoclonal antibody, has showed high affinityand specificityinbinding RANKL and inhibits RANKL-RANK interaction, mimicking the endogen- ouseffectsofOPG.Inknock-inmicewithchimeric(murine/
human) RANKL expression, denosumab showed inhibition ofboneresorption[139].
InaphaseItrial,54patientswithbreastcancer(n=29)or MM(n=25)withosteolytic lesions receivedasingle doseof denosumab or pamidronate. Denosumab decreased bone resorptionwithin24hoursofadministration, asreflectedby levels of urinary and serum NTX. This was similar in magnitudebutmoresustainedthanwithintravenouspami- dronate[140].Theseresultswereconfirmedinanotherphase Itrial,inwhichdenosumabwasgivenatmultipledoses[141].
InaphaseIItrial,theabilityofdenosumab(120mggiven monthlyasasubcutaneousinjection) toaffectbone resorp- tionmarkersandmonoclonalproteinlevelsinMMpatients, whorelapsedafterresponsetopriortherapy,andinpatients who responded to most recent therapy and had stable disease for 3 or more months was evaluated. No patient experiencedcompleteorpartialresponse(50%reductionin M-protein) but seven patients had maximum reduction of
25% in serum M-protein. Bone resorption markers were reducedbymorethan50%withdenosumab[142].
Inanother phaseIItrial,Fizazietal.evaluatedtheeffect ofdenosumabinpatientswithbonemetastasesandelevated urinary NTXlevelsdespiteongoingintravenousbisphospho- natetherapy.Patientswerestratifiedbytumortype(total111 patients; nine patients with multiple myeloma, 50 patients with prostatecancer,46 patientswithbreastcancer andsix patientswithanothersolidtumor)andscreeningNTXlevels andrandomly assignedtoreceivesubcutaneousdenosumab 180mgeveryfourorevery12weeksorcontinueintravenous bisphosphonates every four weeks. Denosumab normalized urinaryNTXlevelsmorefrequentlythanthecontinuationof intravenous bisphosphonate (64% vs. 37% respectively;
p=0.01)andreducedon-studySREscomparedtointravenous bisphosphonate (8% vs. 17%)[143]. This study showedthat denosumabinhibitsboneresorptionandpreventsSREseven inpatientswhoarerefractorytobisphosphonatetherapy.
A meta-analysis of major phase 3 studies comparing denosumab versus zoledronic acid that included mainly patientswithbonemetastasesduetosolidtumors showed thatdenosumabwassuperiorintermsofdelayingthetime tofirst on-studySREby 8monthsand reducingthe riskof the first SREby17%. Nodifference betweenthe twodrugs was reported regarding disease progression and overall survival. Hypocalcaemia wasmore commonin denosumab arm,whileONJwassimilarwiththetwodrugs[144].
Denosumabappearstohavelittletoxicity,mainlyasthe- nia, andmultiplephase IIItrialsof denosumab inpatients with bonemetastasis are ongoing. However it is crucialto mention that RANKL is involved in dendritic cell survival andthattheanti-RANKLstrategymayhaveaneffectonthe
immune system and a possibleincrease in infection rate, especiallyin cancerpatients who have already had severe immunodeficiency.ForMMpatients,while denosumabwas comparable to zoledronic acid with respect to the occur- rence of SREs, inferior survival occurred in denosumab compared to zoledronic acid-treatedpatients, but this was asubset analysis from alarge phaseIII trial thatinvolved mostly solid tumor patients with metastatic bone disease [145]. Interpretationislimitedbasedonthe smallnumbers of MM patients who wereenrolled in the trial and imbal- anceinbaselinediseasecharacteristics.
Activin-Ainhibitors
Sotatercept (ACE-011) is a novel fusion protein of the extracellular domain of the activin receptor IIA (ActRIIA) andhumanimmunoglobulinG(IgG)Fcdomainwithpotent inhibitoryeffectonactivin,enhancingthedepositionofnew bone tissue and preventing bone loss. In the preclinical setting, RAP-011, a murine counterpart of sotatercept, pre- ventedthe formation of osteolyticlesions inamurine MM modelbystimulating boneformation butwithnoeffecton osteoclastactivity[146].
Inaphase1study,asingle-doseofsotaterceptdecreased bone resorption and increased bone formation in healthy postmenopausal volunteers with no severe AEs [147]. In a multicenter phase IIa study weevaluated the safetyand tolerabilityofsotaterceptanditseffectsonbonemetabolism and hematopoiesis in newly diagnosed and relapsed MM patients. Patients were randomized to receive four 28-day cyclesofsotatercept(0.1,0.3,or0.5mg/kg)orplacebo.Patients also received six cycles of combination oral melphalan, prednisolone, and thalidomide (MPT). Thirty patients were enrolled;sixreceivedplaceboand24receivedsotatercept.In patientswithoutbisphosphonateuse,anabolicimprovements in bone mineral density and in bone formation relative to placebo occurred, whereas bone resorption was minimally affected.Increasesinhemoglobinlevels,versusbaseline,and thedurationof theincreaseswerehigherinthesotatercept- treatedpatients,withatrendsuggestingadose-relatedeffect [148].Furtherresearchisneededtosupportthesefindings.
Moreover,increasedactivin-Asecretionwasenhancedby lenalidomideandwasinhibitedbytheadditionofanactivin A-neutralizing antibody. This effectively restored osteoblast functionandsubsequentlyinhibitedmyeloma-relatedosteo- lysiswithoutabrogatingthecytotoxiceffectsoflenalidomide onmalignantcells[149]andthussupportingthecombination oflenalidomidewithananti-activin-Amolecule.
DKK-1antagonists
DKK-1plays animportant role inthe dysfunctionof osteo- blastsobserved inMM. The production of thissoluble Wnt inhibitor by MM cells inhibits osteoblast activity, and its serumlevelreflectstheextensionoffocalbonelesionsinMM [67,150].SerumDKK-1isincreasednotonlyinsymptomatic MMpatientsatdiagnosisbutalsoinrelapsedMM,correlating with advanced disease features and the presence of lytic lesions,while serumDKK-1levelsof asymptomatic patients at diagnosis and plateau do not differ from control values [68,151].
BHQ880,aphage-derivedIgGantibody,thefirst-in-class, fully human anti-DKK-1 neutralizing antibody, seems to promoteboneformationinhibitingtumor-inducedosteoly- tic disease in preclinical studies [152]. Inhibiting DKK-1 with BHQ880 in the 5T2MM murine model of myeloma reduced the development of osteolytic bone lesions and invivogrowthofMMcells[153].Finallyinarecentstudyin humansBHQ880managedtoincreasebonestrengthinthe majorityofmyelomapatientswithrelapsedand/or refrac- torydisease[154].
Sclerostinantagonists
Circulating sclerostiniselevated inpatientswithmyeloma and extended bone disease [77]. Romosozumab is ahumanizedmonoclonalantibodythattargetssclerostin.In aphaseIIstudyinwomenwithpostmenopausalosteoporo- sis, romosozumab demonstrated increases in the bone mineral density of the lumbar spine after 12 months of therapy[155].StudiesinMMaregoingtostartsoon.
Effects of antimyeloma agents on bone metabolism
BortezomibBortezomib is a first-in-class proteasome inhibitor with knownactivityagainstmyeloma.Bortezomibplaysanimpor- tantroleinosteoclastfunctionanddifferentiation.Itaffects bothlateandearlystagesofosteoclastdifferentiationcausing reduction of subsequent bone resorption [156–158]. Clinical trials with bortezomib indicated that it may also increase osteoblastactivityand inducenewbone formation.Inmice bortezomib inducesmesenchymalstemcellstodifferentiate into osteoblasts [159]. Bortezomibupregulatesthetranscrip- tionfactorRunx2/Cbfa1activityinhumanosteoblastprogeni- tors and osteoblasts [160]. Bortezomib administration in relapsed/refractorypatientsresultedinasignificantreduction of DKK-1, enhancement of bone formation and increase in bone mineral density[161–163]. Furthermore,bortezomib in combinationwith thalidomideanddexamethasone(VTD)as consolidation therapy post autologous transplantation pro- duces no SREs in patients with no progressive disease, indicating that patients who respond toconsolidation may notneedconcomitantbisphosphonateadministration[164].
Immunomodulatoryagents
Immunomodulatory drugs (IMiDs), such as thalidomide, lenalidomideandpomalidomide,arehighlyactiveagentsin thetreatmentofbothnewlydiagnosedandrelapsed/refrac- toryMM.Theseagentsalsoalterinteractionsbetweenbone marrowmicroenvironmentandmalignantplasmacells,and modifyabnormalbonemetabolisminMM[27].
Thalidomide almost completely blocks RANKL-induced osteoclastformationinvitro.Inrelapsed/refractoryMMpati- ents, intermediatedoseof thalidomide(200mg/d)incombi- nationwithdexamethasoneproducedasignificantreduction of serum markers of bone resorption [C-telopeptide of collagentype-I(CTX)andTRACP-5b]andalsoofsRANKL/OPG ratio[165].
Lenalidomide also inhibited osteoclast formation, by targetingPU.1,acriticaltranscriptionfactorforthedevelop- ment of osteoclasts, and downregulating cathepsin K. The downregulation of PU.1 in hematopoietic progenitor cells resultedinacompleteshiftoflineagedevelopmenttoward granulocytes.Lenalidomidealsoreducedtheserumlevelsof sRANKL/OPGratioinMM patients[166]. However,lenalido- mideseemstohavemodestornoeffectonboneformation ofmyelomapatients[167].
Pomalidomide, like thalidomide, blocks RANKL-induced osteoclastogenesis in vitro, even at concentrations of one mM, which is similar or even lower than that achieved in vivo after the therapeutic administration of this agent.
Pomalidomide downregulates transcription factor PU.1, affecting the lineage commitment of osteoclast precursors towardgranulocytesinsteadofmatureosteoclasts[168].
IMWG recommendations for treatment of myeloma-related bone disease
The International Myeloma Working group recently pro- duced recommendations regarding the management of myeloma-related bone disease. The IMWG experts recom- mendedthat allpatientswithMM, whoare receiving anti- myelomatherapywithorwithoutosteolyticbonelesions,as wellaspatientswithosteopeniaorosteoporosisduetoMM should receive bisphosphonates. Intravenous pamidronate andzoledronicacidarerecommendedforthepreventionof SREsinpatientswithactivediseaseduetotheirefficacyin SREs reduction. Intravenous zoledronic acid has shown better efficacy in SREs prevention compared with oral clodronate. Zoledronic acid is recommended for MM patients with bone disease at diagnosis rather than CLO.
Thisisbecause ZOLhas shown bettersurvival benefitand has potential antimyeloma activity. MM patients who are ineligiblefortransplantationmaybenefitfromthecombina- tion of antimyeloma treatment with zoledronic acid. In patients with asymptomatic MM of low and intermediate risk,bisphosphonates are recommendedwhen dual-energy X-ray absorptiometry (DXA) scan reveals osteoporosis. For patientswithhigh-riskasymptomaticMM,orifitisunclear whetherbonelossisMMoragerelated,scheduleanddose of bisphosphonates should follow those of symptomatic MM, especially in patients with abnormal MRI pattern.
Intravenous bisphosphonates should beadministered at 3- to4-weekintervalstoallpatientswithactiveMM.Zoledro- nicacid improves OSand reduces SREs overclodronate in patients who have been treatedfor more than two years;
thusitshouldbegivenuntildiseaseprogressioninpatients, not in CR or a very good partial remission (VGPR) and further continued at relapse. There is no similar evidence for pamidronate. Zoledronic acid and pamidronate should bediscontinuedafter1or2yearsinpatientsinCRorVGPR becauseofthehigherratesofONJ[92].
Authors' contributions/Wkład autorów
Accordingtoorder.
Conflict of interest/Konflikt interesu
Evangelos Terpos has received honoraria by Janssen-Cilag, Celgene, Novartis and Amgen. Nikolaos Kanellias has no conflictstodeclare.KrzysztofGiannopouloshasnoconflicts todeclare.
Financial support/Finansowanie
Nonedeclared.
Ethics/Etyka
The work described in this article has been carried out in accordance with The Code of Ethics of the World Medical Association (Declaration of Helsinki) for experiments invol- ving humans; EU Directive 2010/63/EU for animal experi- ments;andUniformRequirementsformanuscriptssubmitted toBiomedicaljournals.
references/pi smiennictwo
[1] KyleRA,GertzMA,WitzigTE,etal.Reviewof1027patients withnewlydiagnosedmultiplemyeloma.MayoClinProc 2003;78:21–33.
[2] Melton3rdLJ,KyleRA,AchenbachSJ,ObergAL,Rajkumar SV.Fractureriskwithmultiplemyeloma:apopulation- basedstudy.JBoneMinerRes2005;20:487–493.
[3] MeltonLJ,KyleRA,AchenbackSJ,ObergAL,RajkumarSV.
Fractureriskwithmultiplemyeloma:apopulation-based study.JBoneMinerRes2005;20:487–493.
[4] MeltonLJ,RajkumarSV,KhoslaS,AchenbackSJ,ObergAL, KyleRA.Fractureriskinmonoclonalgammopathyof undeterminedsignificance.JBoneMinerRes2004;19:
25–30.
[5] CocksK,CohenD,WisløffF,etal.Aninternationalfield studyofthereliabilityandvalidityofadisease-specific questionnairemodule(theQLQ-MY20)inassessingthe qualityoflifeofpatientswithmultiplemyeloma.EurJ Cancer2007;43:1670–1678.
[6] BruceNJ,McCloskeyEV,KanisJA,GuestJF.Economic impactofusingclodronateinthemanagementofpatients withmultiplemyeloma.BrJHaematol1999;104:358–364.
[7] SaadF,LiptonA,CookR,ChenYM,SmithM,ColemanR.
Pathologicfracturescorrelatewithreducedsurvivalin patientswithmalignantbonedisease.Cancer 2007;110:1860–1867.
[8] BatailleR,ChappardD,MarcelliC,etal.Recruitmentof newosteoblastsandosteoclastsistheearliestcritical eventinthepathogenesisofhumanmultiplemyeloma.J ClinInvest1991;88:62–66.
[9] TerposE,DimopoulosMA.Myelomabonedisease:
pathophysiologyandmanagement.AnnOncol 2005;16:1223–1231.
[10] AbeM,HiuraK,WildeJ,etal.Osteoclastsenhance myelomacellgrowthandsurvivalviacell-cellcontact:a viciouscyclebetweenbonedestructionandmyeloma expansion.Blood2004;104:2484–2491.
[11] TanakaY,AbeM,HiasaM,etal.Myelomacell-osteoclast interactionenhancesangiogenesistogetherwithbone
resorption:aroleforvascularendothelialcellgrowth factorandosteopontin.ClinCancerRes2007;13:816–823.
[12] RajaKR,HajekR.ContributionofregulatoryTcellsto immunosuppressionanddiseaseprogressioninmultiple myelomapatients.Oncoimmunology2013;2:e25619.
[13] GiulianiN,CollaS,SalaR,etal.Humanmyelomacells stimulatethereceptoractivatorofnuclearfactor-kappaB ligand(RANKL)inTlymphocytes:apotentialrolein multiplemyelomabonedisease.Blood2002;100:4615–4621.
[14] NoonanK,MarchionniL,AndersonJ,PardollD,Roodman GD,BorrelloI.AnovelroleofIL-17-producinglymphocytes inmediatinglyticbonediseaseinmultiplemyeloma.
Blood2010;4116:3554–3563.
[15] YaccobyS,WezemanMJ,ZangariM,etal.Inhibitory effectsofosteoblastsandincreasedboneformationon myelomainnovelculturesystemsandamyelomatous mousemodel.Haematologica2006;91:192–199.
[16] SatiHI,GreavesM,ApperleyJF,RussellRG,CroucherPI.
Expressionofinterleukin-1betaandtumournecrosis factor-alphainplasmacellsfrompatientswithmultiple myeloma.BrJHaematol1999;104:350–357.
[17] OyajobiBO,FranchinG,WilliamsPJ,etal.Dualeffectsof macrophageinflammatoryprotein-1alphaonosteolysis andtumorburdeninthemurine5TGM1modelof myelomabonedisease.Blood2003;102:311–319.
[18] MitsiadesCS,MitsiadesNS,MunshiNC,RichardsonPG, AndersonKC.Theroleofthebonemicroenvironmentin thepathophysiologyandtherapeuticmanagementof multiplemyeloma:interplayofgrowthfactors,their receptorsandstromalinteractions.EurJCancer 2006;42:1564–1573.
[19] HsuH,LaceyDL,DunstanCR,etal.Tumornecrosisfactor receptorfamilymemberRANKmediatesosteoclast differentiationandactivationinducedbyosteoprotegerin ligand.ProcNatlAcadSciUSA1999;96:3540–3545.
[20] NakagawaN,KinosakiM,YamaguchiK,etal.RANKisthe essentialsignalingreceptorforosteoclastdifferentiation factorinosteoclastogenesis.BiochemBiophysRes Commun1998;253:395–400.
[21] BoyleWJ,SimonetWS,LaceyDL.Osteoclastdifferentiation andactivation.Nature2003;423:337–342.
[22] YasudaH,ShimaN,NakagawaN,etal.Osteoclast differentiationfactorisaligandforosteoprotegerin/
osteoclastogenesis-inhibitoryfactorandisidenticalto TRANCE/RANKL.ProcNatlAcadSciUSA1998;95:3597–3602.
[23] HofbauerLC,HeufelderAE.Osteoprotegerinandits cognateligand:anewparadigmofosteoclastogenesis.Eur JEndocrinol1998;139:152–154.
[24] DougallWC,GlaccumM,CharrierK,etal.RANKis essentialforosteoclastandlymphnodedevelopment.
GenesDev1999;13:2412–2424.
[25] TsukiiK,ShimaN,MochizukiS,etal.Osteoclast
differentiationfactormediatesanessentialsignalforbone resorptioninducedby1alpha,25-dihydroxyvitaminD3, prostaglandinE2,orparathyroidhormoneinthe microenvironmentofbone.BiochemBiophysRes Commun1998;246:337–341.
[26] KongYY,YoshidaH,SarosiI,etal.OPGLisakeyregulator ofosteoclastogenesis,lymphocytedevelopmentand lymph-nodeorganogenesis.Nature1999;397:315–323.
[27] ChristoulasD,TerposE,DimopoulosMA.Pathogenesis andmanagementofmyelomabonedisease.ExpertRev Hematol2009;2:385–398.
[28] LaceyDL,TimmsE,TanHL,etal.Osteoprotegerinligandis acytokinethatregulatesosteoclastdifferentiationand activation.Cell1998;93:165–176.
[29] BucayN,SarosiI,DunstanCR,etal.Osteoprotegerin- deficientmicedevelopearlyonsetosteoporosisand arterialcalcification.GenesDev1998;12:1260–1268.
[30] MizunoA,AmizukaN,IrieK,etal.Severeosteoporosisin micelackingosteoclastogenesisinhibitoryfactor/
osteoprotegerin.BiochemBiophysResCommun 1998;247:610–615.
[31] SimonetWS,LaceyDL,DunstanCR,etal.Osteoprotegerin:
anovelsecretedproteininvolvedintheregulationofbone density.Cell1997;89:309–319.
[32] HofbauerLC,KhoslaS,DunstanCR,LaceyDL,BoyleWJ, RiggsBL.Therolesofosteoprotegerinandosteoprotegerin ligandintheparacrineregulationofboneresorption.J BoneMinerRes2000;15:2–12.
[33] PearseRN,SordilloEM,YaccobyS,etal.Multiplemyeloma disruptstheTRANCE/osteoprotegerincytokineaxisto triggerbonedestructionandpromotetumorprogression.
ProcNatlAcadSciUSA2001;98:11581–11586.
[34] GiulianiN,BatailleR,ManciniC,LazzarettiM,BarilléS.
Myelomacellsinduceimbalanceintheosteoprotegerin/
osteoprotegerinligandsysteminthehumanbonemarrow environment.Blood2001;98:3527–3533.
[35] CroucherPI,ShipmanCM,LippittJ,etal.Osteoprotegerin inhibitsthedevelopmentofosteolyticbonediseasein multiplemyeloma.Blood2001;98:3534–3540.
[36] HeiderU,LangelotzC,JakobC,etal.Expressionofreceptor activatorofnuclearfactorkappaBligandonbonemarrow plasmacellscorrelateswithosteolyticbonediseasein patientswithmultiplemyeloma.ClinCancerRes 2003;9:1436–1440.
[37] FarrugiaAN,AtkinsGJ,ToLB,etal.Receptoractivatorof nuclearfactor-kappaBligandexpressionbyhuman myelomacellsmediatesosteoclastformationinvitroand correlateswithbonedestructioninvivo.CancerRes 2003;63:5438–5445.
[38] QiangYW,ChenY,StephensO,etal.Myeloma-derived Dickkopf-1disruptsWnt-regulatedosteoprotegerinand RANKLproductionbyosteoblasts:apotentialmechanism underlyingosteolyticbonelesionsinmultiplemyeloma.
Blood2008;112:196–207.
[39] StandalT,SeidelC,HjertnerO,etal.Osteoprotegerinis bound,internalized,anddegradedbymultiplemyeloma cells.Blood2002;100:3002–3007.
[40] TerposE,SzydloR,ApperleyJF,etal.Solublereceptor activatorofnuclearfactorkappaBligandosteoprotegerin ratiopredictssurvivalinmultiplemyeloma:proposalfora novelprognosticindex.Blood2003;102:1064–1069.
[41] PolitouM,TerposE,AnagnostopoulosA,etal.Roleof receptoractivatorofnuclearfactor-kappaBligand (RANKL),osteoprotegerinandmacrophageprotein1-alpha (MIP-1a)inmonoclonalgammopathyofundetermined significance(MGUS).BrJHaematol2004;126:686–689.
[42] ChoiSJ,CruzJC,CraigF,etal.Macrophageinflammatory protein1-alphaisapotentialosteoclaststimulatoryfactor inmultiplemyeloma.Blood2000;96:671–675.
[43] AbeM,HiuraK,WildeJ,etal.Roleformacrophage inflammatoryprotein(MIP)-1alphaandMIP-1betainthe developmentofosteolyticlesionsinmultiplemyeloma.
Blood2002;100:2195–2202.
[44] ChoiSJ,ObaY,GazittY,etal.Antisenseinhibitionof macrophageinflammatoryprotein1-alphablocksbone destructioninamodelofmyelomabonedisease.JClin Invest2001;108:1833–1841.
[45] HanJH,ChoiSJ,KuriharaN,KoideM,ObaY,RoodmanGD.
Macrophageinflammatoryprotein-1alphaisan
osteoclastogenicfactorinmyelomathatisindependentof receptoractivatorofnuclearfactorkappaBligand.Blood 2001;97:3349–3353.
[46] RoodmanGD.Mechanismsofbonelesionsinmultiple myelomaandlymphoma.Cancer1997;80:1557–1563.
[47] TerposE,PolitouM,SzydloR,GoldmanJM,ApperleyJF, RahemtullaA.Serumlevelsofmacrophageinflammatory
protein-1alpha(MIP-1a)correlatewiththeextentofbone diseaseandsurvivalinpatientswithMultipleMyeloma.Br JHaematol2003;123:106–109.
[48] MagrangeasF,NasserV,Avet-LoiseauH,etal.Gene expressionprofilingofmultiplemyelomareveals molecularportraitsinrelationtothepathogenesisofthe disease.Blood2003;101:4998–5006.
[49] LentzschS,GriesM,JanzM,BargouR,DörkenB,Mapara MY.Macrophageinflammatoryprotein1-alpha(MIP-1 alpha)triggersmigrationandsignalingcascades
mediatingsurvivalandproliferationinmultiplemyeloma (MM)cells.Blood2003;101:3568–3573.
[50] LeeJW,ChungHY,EhrlichLA,etal.IL-3expression bymyelomacellsincreasesbothosteoclastformation andgrowthofmyelomacells.Blood2004;103:2308–
2315.
[51] EhrlichLA,ChungHY,GhobrialI,etal.IL-3isapotential inhibitorofosteoblastdifferentiationinmultiple myeloma.Blood2005;106:1407–1414.
[52] KyrstsonisMC,DedoussisG,BaxevanisC,StamatelouM, ManiatisA.Seruminterleukin-6(IL-6)andinterleukin-4 (IL-4)inpatientswithmultiplemyeloma(MM).BrJ Haematol1996;92:420–422.
[53] SatiHI,ApperleyJF,GreavesM,etal.Interleukin-6is expressedbyplasmacellsfrompatientswithmultiple myelomaandmonoclonalgammopathyofundetermined significance.BrJHaematol1998;101:287–295.
[54] DonovanKA,LacyMQ,GertzMA,LustJA.IL-1beta expressioninIgMmonoclonalgammopathyandits relationshiptomultiplemyeloma.Leukemia2002;16:
382–385.
[55] SilvestrisF,CafforioP,CalvaniN,DammaccoF.Impaired osteoblastogenesisinmyelomabonedisease:roleof upregulatedapoptosisbycytokinesandmalignantplasma cells.BrJHaematol2004;126:475–486.
[56] LamJ,TakeshitaS,BarkerJE,KanagawaO,RossFP, TeitelbaumSL.TNF-alphainducesosteoclastogenesisby directstimulationofmacrophagesexposedtopermissive levelsofRANKligand.JClinInvest2000;106:1481–1488.
[57] TjinEP,DerksenPW,KataokaH,SpaargarenM,PalsST.
Multiplemyelomacellscatalyzehepatocytegrowthfactor (HGF)activationbysecretingtheserineproteaseHGF- activator.Blood2004;104:2172–2175.
[58] DankbarB,PadroT,LeoR,etal.Vascularendothelial growthfactorandinterleukin-6inparacrinetumor- stromalcellinteractionsinmultiplemyeloma.Blood 2000;95:2630–2636.
[59] StandalT,Hjorth-HansenH,RasmussenT,etal.
Osteopontinisanadhesivefactorformyelomacellsandis foundinincreasedlevelsinplasmafrompatientswith multiplemyeloma.Haematologica2004;89:174–182.
[60] LiaoJ,McCauleyLK.Skeletalmetastasis:Establishedand emergingrolesofparathyroidhormonerelatedprotein (PTHrP).CancerMetastRev2006;25:559–571.
[61] GuiseTA.Parathyroidhormone-relatedproteinandbone metastases.Cancer1997;80:1572–1580.
[62] LeeSK,LorenzoJA.Parathyroidhormonestimulates TRANCEandinhibitsosteoprotegerinmessenger ribonucleicacidexpressioninmurinebonemarrow cultures:correlationwithosteoclast-likecellformation.
Endocrinology1999;140:3552–3561.
[63] OtsukiT,YamadaO,KurebayashiJ,etal.Expressionand invitromodificationofparathyroidhormone-related protein(PTHrP)andPTH/PTHrP-receptorinhuman myelomacells.LeukLymph2001;41:397–409.
[64] CafforioP,SavonarolaA,StucciS,etal.PTHrPproducedby myelomaplasmacellsregulatestheirsurvivalandpro- osteoclastactivityforbonediseaseprogression.JBone MinerRes2014;29:55–66.
[65] WestendorfJJ,KahlerRA,SchroederTM.Wntsignalingin osteoblastsandbonediseases.Gene2004;341:19–39.
[66] KawanoY,KyptaR.SecretedantagonistsoftheWnt signallingpathway.JCellSci2003;116:2627–2634.
[67] TianE,ZhanF,WalkerR,etal.TheroleoftheWnt- signalingantagonistDKK1inthedevelopmentof osteolyticlesionsinmultiplemyeloma.NEnglJMed 2003;349:2483–2494.
[68] PolitouMC,HeathDJ,RahemtullaA,etal.Serum concentrationsofDickkopf-1proteinareincreasedin patientswithmultiplemyelomaandreducedafter autologousstemcelltransplantation.IntJCancer 2006;119:1728–1731.
[69] KaiserM,MiethM,LiebischP,etal.Serumconcentrations ofDKK-1correlatewiththeextentofbonediseasein patientswithmultiplemyeloma.EurJHaematol 2008;80:490–494.
[70] HeiderU,KaiserM,MiethM,etal.Serumconcentrationsof DKK-1decreaseinpatientswithmultiplemyeloma respondingtoanti-myelomatreatment.EurJHaematol 2009;82:31–38.
[71] GlassIIDA,BialekP,AhnJD,etal.CanonicalWntsignaling indifferentiatedosteoblastscontrolsosteoclast
differentiation.DevCell2005;8:751–764.
[72] SpencerGJ,UttingJC,EtheridgeSL,ArnettTR,GeneverPG.
Wntsignallinginosteoblastsregulatesexpressionofthe receptoractivatorofNFkappaBligandandinhibits osteoclastogenesisinvitro.JCellSci2006;119:1283–1296.
[73] OshimaT,AbeM,AsanoJ,etal.Myelomacellssuppress boneformationbysecretingasolubleWntinhibitor,sFRP- 2.Blood2005;106:3160–3165.
[74] GkotzamanidouM,DimopoulosMA,KastritisE, ChristoulasD,MoulopoulosLA,TerposE.Sclerostin:a possibletargetforthemanagementofcancer-induced bonedisease.ExpertOpinTherTargets2012;16:761–769.
[75] ClarkeBL,DrakeMT.Clinicalutilityofserumsclerostin measurements.BonekeyRep2013;2:361.
[76] BrunettiG,OrangerA,MoriG,etal.Sclerostinis overexpressedbyplasmacellsofmultiplemyeloma patients.AnnNYAcadSci2011;1237:19–23.
[77] TerposE,ChristoulasD,KatodritouE,etal.Elevated circulatingsclerostincorrelateswithadvanceddisease featuresandabnormalboneremodelinginsymptomatic myeloma:reductionpost-bortezomibmonotherapy.IntJ Cancer2012;131:1466–1471.
[78] ValletS,MukherjeeS,VaghelaN,etal.ActivinApromotes multiplemyeloma-inducedosteolysisandisapromising targetformyelomabonedisease.ProcNatlAcadSciUSA 2010;107:5124–5129.
[79] TerposE,KastritisE,ChristoulasD,etal.Circulating activin-Aiselevatedinpatientswithadvancedmultiple myelomaandcorrelateswithextensiveboneinvolvement andinferiorsurvival;noalterationspost-lenalidomideand dexamethasonetherapy.AnnOncol2012;23:2681–2686.
[80] ChantryAD,HeathD,MulivorAW,etal.Inhibitingactivin- Asignalingstimulatesboneformationandprevents cancer-inducedbonedestructioninvivo.JBoneMinerRes 2010;25:2633–2646.
[81] AngtuacoEJ,FassasAB,WalkerR,SethiR,BarlogieB.
Multiplemyeloma:clinicalreviewanddiagnosticimaging.
Radiology2004;231:11–23.
[82] DimopoulosMA,MoulopoulosLA,DatserisI,etal.Imaging ofmyelomabonedisease-implicationsforstaging, prognosisandfollow-up.ActaOncol2000;39:823–827.
[83] DimopoulosM,TerposE,ComenzoRL,etal.International myelomaworkinggroupconsensusstatementand guidelinesregardingthecurrentroleofimaging techniquesinthediagnosisandmonitoringofmultiple Myeloma.Leukemia2009;23:1545–1556.
[84] TerposE.Biochemicalmarkersofbonemetabolismin multiplemyeloma.CancerTreatRev2006;32:15–19.
[85] ColemanRE,MajorP,LiptonA,etal.Predictivevalueof boneresorptionandformationmarkersincancerpatients withbonemetastasesreceivingthebisphosphonate zoledronicacid.JClinOncol2005;23:4925–4935.
[86] AbildgaardN,BrixenK,KristensenJE,EriksenEF,Nielsen JL,HeickendorffL.Comparisonoffivebiochemical markersofboneresorptioninmultiplemyeloma:elevated pre-treatmentlevelsofS-ICTPandU-Ntxarepredictivefor earlyprogressionofthebonediseaseduringstandard chemotherapy.BrJHaematol2003;120:235–242.
[87] DimopoulosMA,MoulopoulosLA,ManiatisA,Alexanian R.Solitaryplasmacytomaofboneandasymptomatic multiplemyeloma.Blood2000;96:2037–2044.
[88] KoppulaB,KaptuchJ,HanramanCJ.Imagingofmultiple myeloma:usefulnessofMRIandPET/CT.Semin UltrasoundCTMRI2013;34:566–577.
[89] VanLammeren-VenemaD,RegelinkJC,RiphagenII,etal.
18F-fluoro-deoxyglucosepositronemissiontomographyin assessmentofmyeloma-relatedbonedisease:A systematicreview.Cancer2012;118:1971–1981.
[90] BredellaMA,SteinbachL,CaputoG,SegallG,HawkinsR.
ValueofFDGPETintheassessmentofpatientswith multiplemyeloma.AmJRoentgenol2005;184:1199–1204.
[91] BalducciM,ChiesaS,ManfridaS,etal.Impactof radiotherapyonpainreliefandrecalcificationinplasma cellneoplasms:Long-termexperience.StrahlentherOnkol 2011;187:114–119.
[92] TerposE,MorganG,DimopoulosMA,etal.International MyelomaWorkingGrouprecommendationsforthe treatmentofmultiplemyeloma-relatedbonedisease.J ClinOncol2013;31:2347–2357.
[93] PehWC,GilulaLA.Percutaneousvertebroplasty:
indications,contraindications,andtechnique.BrJRadiol 2003;76:69–75.
[94] DudeneyS,LiebermanIH,ReinhardtMK,HusseinM.
Kyphoplastyinthetreatmentofosteolyticvertebral compressionfracturesasaresultofmultiplemyeloma.J ClinOncol2002;20:2382–2387.
[95] FribourgD,TangC,SraP,DelamarterR,BaeH.Incidenceof subsequentvertebralfractureafterkyphoplasty.Spine 2004;29:2270–2276.
[96] MudanoAS,BianJ,CopeJU,etal.Vertebroplastyand kyphoplastyareassociatedwithanincreasedriskof secondaryvertebralcompressionfractures:apopulation- basedcohortstudy.OsteoporosInt2008;5:819–826.
[97] RogersMJ,GordonS,BenfordHL,etal.Cellularand molecularmechanismsofactionofbisphosphonates.
Cancer2000;88:2961.
[98] TerposE,SezerO,CroucherPI,etal.Theuseof
bisphosphonatesinmultiplemyelomarecommendations ofanexpertpanelonbehalfoftheEuropeanMyeloma Network.AnnOncol2009;20:1303.
[99] BoonekampPM,vanderWee-PalsLJ,vanWijk-vanLennep MM,etal.Twomodesofactionofbisphosphonateson osteoclasticresorptionofmineralizedmatrix.BoneMiner 1986;1:27.
[100]RoweDJ,EtreLA,LovdahlMJ,PietrzykDJ.Relationship betweenbisphosphonateconcentrationandosteoclast activityandviability.InVitroCellDevBiolAnim 1999;35:383.
[101]TerposE,BerensonJ,RajeN,RoodmanGD.Management ofbonediseaseinmultiplemyeloma.ExpertRevHematol 2014;7:113–125.
[102]MundyGR,YonedaT.Bisphosphonatesasanticancer drugs.NEnglJMed1998;339–398.
[103]YinJJ,SelanderK,ChirgwinJM,etal.TGF-betasignaling blockadeinhibitsPTHrPsecretionbybreastcancercells
andbonemetastasesdevelopment.JClinInvest 1999;103:197.
[104]DielIJ,SolomayerEF,CostaSD,etal.Reductioninnew metastasesinbreastcancerwithadjuvantclodronate treatment.NEnglJMed1998;339:357.
[105]AparicioA,GardnerA,TuY,etal.Invitrocytoreductive effectsonmultiplemyelomacellsinducedby
bisphosphonates.Leukemia1998;12:220.
[106]ShipmanCM,RogersMJ,ApperleyJF,etal.
Bisphosphonatesinduceapoptosisinhumanmyeloma celllines:anovelanti-tumouractivity.BrJHaematol 1997;98:665.
[107]DhodapkarMV,SinghJ,MehtaJ,etal.Anti-myeloma activityofpamidronateinvivo.BrJHaematol 1998;103:530.
[108]BelchAR,BergsagelDE,WilsonK,etal.Effectofdaily etidronateontheosteolysisofmultiplemyeloma.JClin Oncol1991;9:1397–1402.
[109]DaragonA,HumezC,MichotC,etal.Treatmentof multiplemyelomawithetidronate:resultsofa
multicentredouble-blindstudy.EurJMed1993;2:449–452.
[110]LahtinenR,LaaksoM,PalvaI,VirkkunenP,ElomaaI.
Randomised,placebo-controlledmulticentretrialof clodronateinmultiplemyeloma.Lancet1992;340:1049–1052.
[111]LaaksoM,LahtinenR,VirkkunenP,ElomaaI.Subgroup andcost-benefitanalysisoftheFinnishmulticentretrial ofclodronateinmultiplemyeloma.BrJHaematol 1994;87:725–729.
[112]McCloskeyEV,MacLennanIC,DraysonMT,etal.A randomizedtrialoftheeffectofclodronateonskeletal morbidityinmultiplemyeloma.MRCWorkingPartyon LeukaemiainAdults.BrJHaematol1998;100:317–325.
[113]McCloskeyEV,DunnJA,KanisJA,MacLennanIC,Drayson MT.Long-termfollow-upofaprospective,double-blind, placebo-controlledrandomizedtrialofclodronatein multiplemyeloma.BrJHaematol2001;113:1035–1043.
[114]BerensonJR,LichtensteinA,PorterL,etal.Efficacyof pamidronateinreducingskeletaleventsinpatientswith advancedmultiplemyeloma.MyelomaArediaStudy Group.NEnglJMed1996;334:488–493.
[115]BerensonJR,LichtensteinA,PorterL,etal.Long-term pamidronatetreatmentofadvancedmultiplemyeloma patientsreducesskeletalevents.MyelomaArediaStudy Group.JClinOncol1998;16:593–602.
[116]BrinckerH,WestinJ,AbildgaardN,etal.Failureoforal pamidronatetoreduceskeletalmorbidityinmultiple myeloma:adouble-blindplacebo-controlledtrial.Danish- Swedishco-operativestudygroup.BrJHaematol 1998;101:280–286.
[117]BerensonJR,RosenLS,HowellA,etal.Zoledronicacid reducesskeletal-relatedeventsinpatientswithosteolytic metastases.Cancer2001;91:1191–1200.
[118]RosenLS,GordonD,KaminskiM,etal.Zoledronicacid versuspamidronateinthetreatmentofskeletal metastasesinpatientswithbreastcancerorosteolytic lesionsofmultiplemyeloma:aphaseIII,double-blind, comparativetrial.CancerJ2001;7:377–387.
[119]RosenLS,GordonD,KaminskiM,etal.Long-termefficacy andsafetyofzoledronicacidcomparedwithpamidronate disodiuminthetreatmentofskeletalcomplicationsin patientswithadvancedmultiplemyelomaorbreast carcinoma:arandomized,double-blind,multicenter, comparativetrial.Cancer2003;98:1735–1744.
[120]MustoP,PetrucciMT,BringhenS,etal.Amulticenter, randomizedclinicaltrialcomparingzoledronicacidversus observationinpatientswithasymptomaticmyeloma.
Cancer2008;113:1588–1595.
[121]MorganGJ,DaviesFE,GregoryWM,etal.First-line treatmentwithzoledronicacidascomparedwith
