Pathophysiology of BRONJ: Drug-related osteoclastic disease of the jaw Oral Science International

ContentslistsavailableatSciVerseScienceDirect

Oral Science International

j ou rna l h o m e p a g e :w w w . e l s e v i e r . c o m / l o c a t e / o s i

Review

Pathophysiology of BRONJ: Drug-related osteoclastic disease of the jaw

Tetsuro Ikebe

DepartmentofOralandMaxillofacialSurgery,FukuokaDentalCollege,2-15-1Tamura,Sawara-ku,Fukuoka814-0193,Japan

a r t i c l e i n f o

Articlehistory:

Received27February2012

Receivedinrevisedform21March2012 Accepted22March2012

Keywords:

Osteonecrosis Bisphosphonate Osteoclast Mevalonate

a b s t r a c t

Sincethefirstarticleaboutbisphosphonate-relatedosteonecrosisofthejaw(BRONJ)waspublishedin 2003,clinicalandbasicresearchforBRONJhascontinuedworldwidetounderstandthisnoveldisease.

Severalorganizationshaveproposedthedefinition,diagnosticcriteria,riskfactors,andtreatmentstrat- egyforBRONJ.Recently,somenewdrugsusedforcancerpatientssuchasbevacizumabandsunitinib havealsobeenreportedtobeinvolvedinosteonecrosisofthejaw(ONJ).BecauseONJappearstobeini- tiallyderivedfromosteoclastinhibition,anewcategoryofdiseasesnamedas“drug-relatedosteoclastic diseaseofthejaw”maybeassumed.ConsideringtheaccumulatedknowledgerelatedtoBRONJ,including osteoclastbiology,bisphosphonatepharmacology,animalexperiments,andclinicopathologicalfindings, aperspectiveofBRONJfromthepathophysiologicalviewpointisproposedinthisreview.

© 2012 Japanese Stomatological Society. Published by Elsevier Ltd. All rights reserved.

Contents

1. Introduction... 2

2. IncidenceofandriskfactorsforBRONJ. . . ... 2

2.1. Incidence... 2

2.2. Toothextraction. . . ... 2

2.3. Glucocorticoids... 2

3. DiagnosticcriteriaandstageclassificationofBRONJ. . . ... 2

3.1. Criteriaandstage... 2

3.2. Stage0. . . ... 3

4. Relationshipbetweenstage0andsuppurativeosteomyelitisofthejaw... 3

5. Mechanismofnitrogen-containingBPactions. . . ... 4

6. OthertargetcellsofBP... 4

6.1. Osteoblasts... 4

6.2. Osteocytes... 4

6.3. Vascularendothelialcells... 4

6.4. Keratinocytes... 5

6.5. Macrophages/monocytes... 5

7. WhydoesBRONJoccurinthejaws?... 5

7.1. Boneturnoverrate... 5

7.2. Vulnerabilityofthejawtobacterialinfection... 5

8. ConfusedeffectofnBPonosteomyelitisofthejaw... 5

9. PathophysiologicalmechanismofBRONJ... 5

10. TreatmentofBRONJ... 6

10.1. Drugholiday... 6

10.2. Hyperbaricoxygen... 6

10.3. Teriparatide(recombinanthumanparathyroidhormone1–34)... 6

10.4. Prospectivetherapy... 6

11. Drug-relatedosteoclasticdiseaseofthejaw... 7

References... 7

∗ Correspondingauthor.Tel.:+81928010411;fax:+81928011288.

E-mailaddress:ikb@college.fdcnet.ac.jp

1348-8643/$–seefrontmatter © 2012 Japanese Stomatological Society. Published by Elsevier Ltd. All rights reserved.

http://dx.doi.org/10.1016/S1348-8643(12)00045-6

1. Introduction

The number of the patients with bisphosphonate-related osteonecrosisofthejaw(BRONJ)hasincreasedyearbyyearsince Marxfirstreportedcasesofosteonecrosisofthejaw(ONJ)occur- ringinpatientswithmultiplemyelomatreatedwithintravenously administeredbisphosphonatein2003[1].Thediseaseconceptof BRONJseemstobeestablishednowsincemanyarticlesandreviews aboutBRONJhavebeenpublished.

Bisphosphonate(BP)wasoriginallydevelopedasadrugtotreat boneresorbingdiseasessuchasmultiplemyelomaandbonemetas- tasis,whosetypicaloriginsarebreastcancerandprostatecancer,as wellastumor-relatedhypercalcemia,andthenusedforthepatients withosteoporosistopreventtheirpathologicalfracture.InJapan, approximately3millionpatientshavereceivedBPtherapy.Aspe- cifictarget cellofBPisosteoclast whichplaysa centralrole in physiologicalandpathologicalboneresorption.

Since BP possesses a pyrophosphate-like structure, whose hydroxylgroupconfersastrongaffinitytothehydroxyapatite,80%

ofBPadministeredintravenouslyororallydepositstothebones.

ThehalflifeofBPinhumans,whichispoorlymetabolizedbybio- logicalenzymes,isthoughttobeabout10years,andthusBPmay beenrichedinthebones,towhereosteoclasts,whichdifferentiate frombonemarrowstemcells,migrateandadheretofunction.That iswhyBPwasexpectedtoworkselectivelyonosteoclastswith- outadverseeffects.Nitrogen-containingBP(nBP)hasmorepotent abilitytoinhibitmineraldissolutionandhigherpotencytoinduce BRONJthanearlyBPwithoutnitrogen-containingside-chain[2].In thisreview,“BP”occasionallyindicatesbothnBPandearlyBP.

LikeBP,denosumab,whichisanti-humanRANKligand(RANKL) monoclonalantibodyanddevelopedtoinhibitosteoclastsforthe treatment of bone-resorbing diseases, is also suspected to be involvedintheoccurrenceofONJthatissimilartoBRONJ[3,4].The symptomsseeninchronicosteomyelitisofthejawoccurringinthe patientswithosteopetrosis,inwhichosteoclastfunctionsareinhib- itedbyageneticmutation,alsoseemtobesimilartothoseofBRONJ.

Sincetheselesionsappeartobederivedfromosteoclastinhibition, wecouldclassifythemintoanewcategoryofdiseasespresumably namedas“drug-relatedosteoclasticdiseaseofthejaw”.Thisideais basedontheconceptthatthe“osteoclastisaself-defenseregulator ofthejaw.”Thedefensemechanismofthejawshouldbeimpaired inthelackofosteoclasts,whichislikentothatwebecomesus- ceptibletoinfectionwhenourimmunecellsareinhibited.Inthis review,hence,theclinicalaswellaspathophysiologicalaspectsof BRONJwillbediscussedaccordingtothisconcept.

2. IncidenceofandriskfactorsforBRONJ 2.1. Incidence

TheincidenceofBRONJintheearlyUSAreportwas0.8–12%

withintravenousBPand 0.7/100,000(0.0007%)person-yearsof exposurewithoralBP[5].InEurope,theincidencewas95/100,000 (0.095%)person-yearsofexposurewithintravenousadministra- tionand1/100,000(0.001%)person-yearsof exposurewithoral administration [6]. BRONJ occurred in 1.15% of intravenously administered patients and in 0.04% of the orally administered patientsinAustralia[7].AnotherreportfromtheUSAindicatedthat theincidenceofBRONJwas0.1%withoralBPadministration[8].

Takentogether,thesereportssuggestthattheincidenceofBRONJ withoralBPis0.01–0.1%.

MostofthepatientstakingoralBPhaveosteoporosis.Approx- imately20%ofthe10millionosteoporosispatientsinJapanare thoughttobetreatedwithmedicationincludingBPs.Astheelderly populationincreases,thenumberofosteoporosispatientstaking

oralBPwillincreaseandtheperiodforBPadministrationwillpro- long,whichwillresultinanincreasedincidenceofBRONJ.Actually, anationwideretrospectivecohortstudywithaquestionnaireon BRONJinJapan,whichwasundertakenbytheJapaneseSocietyof OralandMaxillofacialSurgeons,demonstratedthat263casesof BRONJoccurredintheentryhospitalsfrom2006to2008,and39.5%

ofthemwerepatientstakingoralBPs[9].

2.2. Toothextraction

Among various risk factors, including glucocorticoids (GCs), anti-cancer drugs, alcohol, smoking,and malnutrition,to make patientssusceptible toBRONJ,toothextractionand othersurgi- calprocedurestothejawsaremostproblematictodentistsand oralandmaxillofacialsurgeons.Toothextractionraisedtheinci- dence of BRONJ about 8-fold, from 0.04% to 0.34% in the oral BP-administeredpatientsandfrom1.15%to9.1%intheintravenous BP-administeredpatientsinAustralia[7].BRONJ occurredin6%

in66patientsorallytakingaledronateaftertoothextraction[10].

Toothextractionwasrelatedto77%ofBRONJcasesoccurringinthe patientsintowhomBPwereadministeredintravenously[11].Sur- gicaldamagetothejaw,especiallyalveolarbone,islikelytobethe mostpotenttriggertoBRONJwhiletheimmunosuppressiveagents suchasGCsandanti-cancerdrugs,couldlowerthethresholdforits occurrence.

Without any risk factors, BRONJ is known to occur spon- taneously. The lingual side of the alveolus in mandible is a predisposedsiteforspontaneousBRONJ,whichisthoughttobe triggeredbyaninjuryonthemucosa.Likewise,bonyprominence seeninpalataltorusandmandibulartorusisalsosusceptibleto BRONJ.

2.3. Glucocorticoids

The use of GCs with BP is known to raise the risk for BRONJbecauseGCssuppresstheactivitiesofinflammatorycells and immune cells to make thepatients immunocompromized.

SinceGCsnot onlyactivate theosteoclasts,butalsoinhibitthe osteoblasts,thelong-termuseofGCsinduceosteoporosis.Thatis whypatientsundergoingGCtherapyarealsoadministeredBPto preventosteoporosis.Rheumatoidarthritis(RA)isatypicaldisease treatedbothwithBPandGCs.SomeofthepatientswithRAarealso treatedwithmethotrexate(MTX)togetherwithBPandGC.MTXisa folicacidantagonistandusedasananti-cancerdrugaswellasanti- rheumaticdrugbyinhibitingtheinflammatorycellsanddecreasing variouscytokineproductions,whichislikelytoincreasetheriskof BRONJ.RApatientsshouldbepaidspecialattentionforpreventing BRONJ.

3. DiagnosticcriteriaandstageclassificationofBRONJ 3.1. Criteriaandstage

In2007,criteriatodiagnoseBRONJwereprovidedinaposition paperbytheAmericanAssociationofOralandMaxillofacialSur- geonsasfollows:(1)currentorprevioustreatmentwithaBP;(2) exposedboneinthemaxillofacialregionthathaspersistedformore than8weeks;and(3)nohistoryofradiationtherapytothejaws[5].

Althoughanexposureofnecroticbone,sequester,inthejawisthe mostprominentcharacteristicofBRONJ,itisoccasionallyobserved inchronicsuppurativeosteomyelitisofthejaw.Thatseemstobea reasonwhythecriteriaincludethesecondpoint.Thecriterion(3) istoexcludethepossibilityofosteoradionecrosisofthejaw.

Thepositionpaperin2007alsoproposedastagingclassifica- tionofBRONJ,stage1,stage2,andstage3.Stage1showsonlyan exposureofthebonewithoutinfectioussigns.Stage2includesnot

Table1

Comparisonofthelevelbetweenjawinflammationandbisphosphonate-related osteonecrosisofthejaw(BRONJ).

Category Levelofthelesion

Alveolus Jawbody

Ostitis Alveolarostitis Periostitis

Periodontitis

Abscess Gingivalabscess Subperiostealabscess

Alveolarabscess

BRONJ Stage2 Stage3

onlyaboneexposure,butalsoinflammatorysymptomsbybacte- rialinfectionsuchaspainandswelling.Theextentofthelesion instage2islimitedinthealveolarbone.Whenthelesionextends tothejawitselfbeyondthealveolarlevel,then,itwillbeclassi- fiedtostage3,wherethebonenecrosisorosteolysismayextend totheinferiorborderofmandiblewithorwithoutextraoralfis- tulaandpathologicalfracture.Thedifferencebetweenstage2and stage3willbesimilartothatbetweenperiodontitisandperiostitis, dependingontheaffectedarea(Table1).

3.2. Stage0

AllthestagesofBRONJrequireanexposedboneasadefinition.

However,itisknownthattherehavebeenpatientstakingoralBP whohavenoboneexposureinthejaw,butsufferedfromsomenon- specificandrefractorysymptomssuchasdiffusealveolarswelling, gingivalredness,orinternaldentalfistulawhoseoriginsarenot identified.Thesesymptomshaveoccasionallysubsidedafterthe discontinuationofBPtherapy.TheAmericanAssociationofOral andMaxillofacialSurgeonspositionpaperin2009addedastage 0intotheBRONJstagingforthecaseswithnoclinicalevidence ofnecroticbone,butnonspecificclinicalfindingsandsymptoms [12].Fig.1showsaplausibleexplanationaboutthedevelopment ofstage0.Instage0,anopenalveolarsocketmaypersistaftertooth extraction,wherethesurfaceofthesocketisconcealedwithgran- ulationtissues,butthemostpartofthebaresocketisleftbehind withabonysurface,presumablybecausethethickenedlaminadura [13],whichcannotberesorbedbyBP,maypreventtheinflamma- torycellsfrominfiltrating,impairthewoundhealingprocesses,and allowtheinfectiontopersist(Fig.1).

4. Relationshipbetweenstage0andsuppurative osteomyelitisofthejaw

Dentists or oral and maxillofacial surgeons may diagnose the patients with stage 0 simply as suppurative periostitis or osteomyelitisofthejaw,beingunawareoftheirBPadministration.

Ifstage0lesionisaprecursorofBRONJ,bacterialinfectionmayplay acriticalroleintheinitiationofBRONJ.

Kosetal.[14]reportedthatwhenpatientswithosteomyelitis of the jaws were divided into two groups, the ones undergo- ingBPtreatmentandtheotherstaking noBP, Actinomyceswas significantlydetectedontheexposed necroticbonesin theBP- administeredpatients,suggestingtheinvolvementofActinomyces in BRONJ. Ganguli et al.[15] suggest theaffinity of bacteria to BP by showing increased adhesion of Staphylococcus aureus to hydroxyapatiteinjointprosthesescoatedwithpamidronate.On theotherhand,Hansenetal.[16]alsodetectedActinomycesinthe necroticbonesofBRONJaswellasosteoradionecrosisinanequally highfrequency.BothnecroticbonesfromBRONJandsuppurative osteomyelitisofthejawswerefoundtobecoveredwiththebac- terialbiofilms,but,Actinomycespredominatedinthebiofilmfrom suppurativeosteomyelitiswhilethebiofilmfromBRONJincluded morediversebacterialorganismsinadditiontofungalorganisms notobservedinsuppurativeosteomyelitis[17].Althoughitremains unknownwhetherbacterialinfectionsincludingActinomycesoccur primarilyorsecondarilyintheBRONJprocess,infectionislikelyto worsenthesymptomsofBRONJ.Thus, BRONJstage0shouldbe initiallytreatedwithantibacterialmouthrinseandantibiotics.

As mentioned below, BP inhibits the growth of vascular endothelialcellsandimpairstheangiogenesisinboneandperios- teum.Thedecreaseinbloodsupplymayleadtoavascularnecrosis ofthejawinBRONJ,whichislikeosteoradionecrosisofthejaws.

But,itisunlikelythatavascularandasepticnecrosisofthebonesisa primarycauseofBRONJsincesomeinfectiouslesionssuchasgingi- valabscessoftenprecedethedevelopmentofBRONJinmanycases [18],andstage0ofBRONJusuallyshowssomeinfectionsigns.Saia etal.[19]reportedthatamong60casesoftoothextractionunder BPtreatment,BRONJoccurredin3caseswithin3months,suggest- ingthatsuppurativeosteomyelitismightbethemostinfluential riskfactortobringoutBRONJ.

RegardlessofwhetherbacterialinfectioncausesBRONJornot, itisimportanttofindoutthecasesofstage0asearlyaspossible andpreventthemfromprogressingtothelaterstages.Hutchinson etal.foundtheradiographicfeatureswithpanoramicradiography andcomputedtomography(CT),includingdiffuseosteosclerosis,

Fig.1.WhathappensinthejawofStage0?Thickenedlaminadurawillpreventwhitebloodcellsfrominfiltratingandgranulationtissuefromforminginthesocket,resulting indelayedwoundhealingandpersistentinfection.

Fig.2. Bisphosphonate (BP) targetingthe mevalonate pathway. FPP,farnesyl diphosphate;PP,pyrophosphate.

thickenedlaminadura,prominenceoftheinferioralveolarcanal, andcorticaldisruption,in10casesamong30patientswithstage0 [20].Inmyopinion,hence,CTorconebeamCTseemstobeuseful todetecttheseinitialchangesinthebonesfortheearlydiagnosis ofstage0.

5. Mechanismofnitrogen-containingBPactions

ThemoleculartargetofnBPisafarnesyldiphosphatesynthase (FPPsynthase)inosteoclasts[2,21,22].Atfirst,nBPwassupposed toinhibitthemineraldissolutionbyitsbindingtomineralcom- ponentsoutsideosteoclasts,butnowitisknownthatnBPenters intotheosteoclaststoinhibittheactivityofFPPsynthasewhichis anenzymeinvolvedinmevalonatepathwayofosteoclasts(Fig.2).

ThiseffectofnBPwasfoundintheresearchforanagenttolower thelevelofserumcholesterolwhichissynthesizedthroughmeval- onatepathway[2].FPPsynthaseisanenzymetoconvertgeranyl diphosphatetofarnesyldiphosphate,whichisinturnconvertedto geranylgeranyldiphosphate.Geranylgeranyldiphosphateactivates thesmallGTPases,suchasRab,Rac,Ras,andRho,throughtheiriso- prenylationandgeranylgeranylation.SmallGTPasesthenregulate thecytoskeletalarrangement,vesiculartrafficking,andmembrane rufflingwhichareinvolvedintheprocessesofboneresorptionby osteoclasts,includingtheirmigrationtobones,adhesiononbone surfaceandtransportationofbone-resorbingenzymestotheruf- fledborder.Therefore,theinhibitionofFPPsynthasewithnBPcan preventosteoclastsfromdestructingthebones(Fig.2).

TheremaybeevidencetoshowthatnBPdisturbsthecytoskele- talregulation. Weinstein et al. observed theosteoclasts in the bone-biopsyspecimensofiliacboneobtainedfromhealthypost- menopausalwomen receiving3-year oral alendronate therapy, comparedwiththe placebo-controlledonesin a doubled-blind, randomizedtrial[23].Intheory,itwasexpectedthattherewould befewosteoclastsinthebonesfromthepatientstakingnBP.Sur- prisingly,however,thenumberoftheabnormalosteoclasts,which weregiant,hypernucleated(with20–40 nuclei),detached from bone surface, increased as the cumulative dose of alendronate increased.Ofthesegiantosteoclasts,20–37%wereapoptotic.They suggestthatnBPprotractstheapoptosisofosteoclasts,wherecell fusioncontinuestogeneratethedistinctivemorphologysuchas giantismandhypernucleation.Thesefeaturesmaybeattributedto nBP-mediatedlossofsmallGTPase.

Suchanincreaseinthenumberofgiant,hypernucleatedosteo- clastsisalsoobservedinthebonesfrompatientswithSchönberg diseaseorautosomaldominantosteopetrosistypeII,inwhichthe abilityofosteoclaststodecayboneisimpairedbecauseofthemuta- tionsinCLCN7gene encodinga chloridechannel[24].nBPmay inhibitthefunctionofCLCN7[25].Hansenetal.[16]alsofound manyosteoclastsinthenecroticbonesfromtheBRONJpatients

administeredwithpamidronateorzoledronate.Thesedatasug- gestthatnBPbringsaboutthefunctionaldisorder,ratherthanthe depletion,ofosteoclasts.

6. OthertargetcellsofBP

Althoughosteoclastshavebeenthoughttobeaspecifictarget ofnBP,itisdifficulttoattributetheoccurrenceofexposedbone necrosisonlytoosteoclastdisorder.Someothercellsarethought totakepartinestablishingBRONJ.

6.1. Osteoblasts

Osteoclasts are known to indirectly assist bone formation, whereosteoclasticboneresorptioncanreleasesomegrowthfactors buriedinthebone,whichstimulateosteoblaststomakebones[26].

Thecooperationofosteoclastswithosteoblasts,which iscalled

“coupling”,isthoughttobeessentialforboneturnover.Thus,the lossofosteoclastsis likelytoimpairtheactivityofosteoblasts, resulting in the bone turnover arrest to osteonecrosis. On the otherhand,thedirecteffectsofnBPonosteoblastshavealsobeen investigated.Zoledronateinhibitedthegrowthandviabilityofcul- turedosteoblasts[27,28],whilepamidronateandzoledronatewere reportedtoinhibitthecellgrowth,butincreasetheboneformation ofimmortalizedfetalosteoblasts[29].Thedamagetoosteoblasts maybeassociatedwiththeoccurrenceofBRONJ,whichhypoth- esismaybesupportedbythereportthat parathyroidhormone mightimprovetheBRONJbystimulatingosteoblasts,asmentioned below.

6.2. Osteocytes

AllenandBurr[30]examinedthebonematrixinthemandible frombeagledogs treatedwithclinically relevantdoses ofdaily oral alendronate for 3 years and found significant amounts of necrotic bone matrix in approximately25% ofnBP-treated ani- mals,whichwerepredominantlypresentinthealveolarportionof themandible.Thenon-viablebonematrixshowedemptylacunae withoutosteocytes.Ithasbeenreportedthatfluorescentlylabeled risedronateinjectedintravenouslyintomicewasobservedaround thelacunae[31].Becauseosteocytesarethemostabundantbone cellsandformanintricatecommunicationnetworkthroughoutthe bonematrix,whereosteocytesarethoughttoprovideasignalto osteoclaststocleanupthenecroticbones,thelossofosteocytes mayplayaroleinthepathophysiologyofBRONJ[32].Itislikelythat themicronecrosisofthebonemayaccumulateinlargeamounts becauseofthelackofosteocyte–osteoclastcommunication[33,34].

6.3. Vascularendothelialcells

ItisassumedthatBPinhibitsboneangiogenesisbysuppress- ingthegrowthofvascularendothelialcellstoresultinavascular necrosisofthebone.ThereareanumberofreportstoshowthatnBP directlyinhibitsangiogenesisinvitroorinvivo[35,36,28]although thestudiesontheeffectsof nBPonangiogenesisinbone mar- rowandperiosteumofthejawremaintobedone[37].nBPmay alsoexertindirecteffectsonthesuppressionofboneangiogene- sissinceosteoclastsmayberequiredtomakevesselspassthrough bonematrix[38].ThisimportanceofnBP-mediatedinhibitionof angiogenesisinthejawcouldbesupportedbythereportsthatthe administrationofzoledronatereducedtheserumlevelsofvascular endothelialgrowthfactor(VEGF)inpatientswithbonemetasta- sis[39]andthatbevacizumab,ananti-VEGFantibody,mayinduce osteonecrosisofthejaw[40].

6.4. Keratinocytes

nBPisreportedtoinhibitthegrowthofculturedkeratinocytes [41].nBPdecreasedthenumberofp63-positivekeratinocytepro- genitorcellsandpreventedthegingivalfibroblastsfromproducing keratinocyte growth factor (KGF) [42,43]. It is presumed that inflammationinthejaws,i.e.toothextraction,releasesthenBP buriedinthebones,whichinturninhibitsthekeratinocytegrowth toworsen theexposureofbones,especiallyinpatientswithan ulceroraninjuryformedonthemucosabyunsuitabledentures.

ThemucosaldamagewithnBPisdemonstratedinacasethatthe contactofnBPonmucosainducedstomatitisinpatientswhohad thehabittoholdthealendronatetabletsinthemouthforawhile beforeswallowingthem[44,45].

6.5. Macrophages/monocytes

Sinceosteoclastsdifferentiatefrommacrophage/monocytelin- eage,itisplausiblethatnBPinhibitstheactivityofmacrophages.

Actually,nBPinhibitedtheproductionofcytokinesinmacrophages and monocytes [31]. The differentiation of dendritic cells, a macrophagelineage, wasinterferedwithby nBP[46,47].These macrophages/monocytesplayanimportantroleintheself-defense systemofimmunityasantigen-presentingcells. Thus,nBPmay suppresstheimmunologicalreactionstomakebonessusceptible tobacterialinfection.Inthisrespect,thereasonwhyGCisariskfac- torforBRONJisthoughttobeattributedtoitsimmunosuppressive action[48,49].Inaddition,itislikelythatnBPinhibitsthediffer- entiationfrommacrophage/monocytetoosteoclast.Kimachietal.

[50]reportedthatzoledronateinhibitedthetumornecrosisfactor-

␣-induceddifferentiationofamurinemacrophage/monocytecell line,Raw264.7tomatureosteoclasts.

7. WhydoesBRONJoccurinthejaws?

It isthoughtthat jawis themostpredisposedtoBP-related osteonecrosisbecause(1)theturnoverrateofjaws,especiallyalve- olarbones,issorapidand(2)jawshaveteethandgumthatmay becomeaneasyentranceforbacterialinfection[51].

7.1. Boneturnoverrate

Theremodelingratesofthecorticalboneinthejaware10–20 timeshigherthaninthecortexofiliaccrestinhumans[52,53].The levelofintracorticalboneremodelinginthemandiblewasfound tobemorethan10timeshigherthaninthetibialcortexindogs, especially,thealveolarportionofthemandiblehavingmorethan 8timeshigherrateofturnoverthanintheotherportionsofthe mandible[37].Incontrast,somearticlesshowthattheturnover ofthejawisnotthehighestamongthebones.HujaandBeck[54]

reportedthattheboneformationrateofthefemurishigherthan thatofthemandibleandmaxillaindogs.Inathymicrats,theuptake offluorescentlylabeledpamidronatetothebonesandtherelease ofitperunitcalciumwerelowerintheoralbonesthanintheaxial bones[55].Apartfromthecontroversy,bonedamagesuchassur- gicalinterventionandinfectionmayacceleratetheturnoverrateof thejaw.

Whenfluorescentpamidronatewasinjectedintravenouslyinto mice,90%ofthefluorescentsignallocalizedtobonewithin2–6hof injection,wheremoresignalsweredetectedandretainedlongerin themandibleratherthaninthefemur[56].Moreover,fluorescent pamidronatewasfoundtobedepositedinthealveolarboneand bonesurroundingtheperiodontalligamentandmolarroots[56].

Takentogether,alveolarboneofthemandibleismostsusceptible

tothenBPeffectsbecauseofrapidturnoverrateandhighaffinity tonBP.

7.2. Vulnerabilityofthejawtobacterialinfection

Amongallthebones,jawseemstobethemostliabletobacte- rialinfectionsincemucosacoveringthealveolarboneisverythin andvulnerable,and teetheasilybecomeapathwayforbacteria fromtheoutsideintothebone.Althoughitiscertainthatbacterial infectionmakesBRONJworse,itremainstobedeterminedwhether bacterialinfectionprecedesosteonecrosisofthejaw.Aghalooetal.

[57]foundthatthenecrosisofthealveolarbonesdevelopedinthe rats,whichunderwent theplacementofa wireligaturearound thecrownofmaxillarymolarinaperiodontaldiseasemodel,after intraperitonealinjectionwithzoledronate.Thenecroticboneswere notexposedintheseratswhereasthecontrolanimalsshowedthe resorptionofalveolarbone,butnotanyosteonecrosis.Theresults imply thatperiodontitis, whichis presumablyinfection-related, cantriggertheosteonecrosis.

Whenperiodontitisoccurs,inflammatorycellsarerecruitedto thesitestoeliminatethecausativepathogens.However,theblock- ade of boneresorption withBPmaymake it difficultfor these cellstoaccesstothepathogens,allowingtheinfectiontopersist.

Theresultingaccumulationofbacterialtoxinsandinflammation- generated superoxides will promote the bone necrosis. Thus, bacterialinfectionislikelytobeinvolvedintheformationofBRONJ.

From theseassumptions, it seemsto be reasonable totreat BRONJwiththoroughlocalrinseandantibioticadministrationasa firststeptocontroltheinfection.

8. ConfusedeffectofnBPonosteomyelitisofthejaw

Ontheotherhand,pamidronateisusedtotreatpatientswith chronicnon-bacterialosteomyelitis(CNO),includingchronicrecur- rent multifocal osteomyelitis (CRMO), SAPHO (synovitis, acne, pustulosis,hyperostosis, and osteitis),and juvenilediffuse scle- rosingosteomyelitis(DSO),whichoccasionallyoccurinthejaws [58–61]. Although the causative mechanism of CNO remains unknown,theeffectivenessofBPonCNOmayrathersupportthe hypothesisthatbacterialinfectionisinvolvedintheinitiationof BRONJ.BPcancurenon-bacterialosteomyelitiswhileBPworsens suppurativeosteomyelitistoBRONJ.

Whendentalsurgicalprocedureswereappliedto22patients with osteogenesis imperfecta who underwent BP therapy, no BRONJcaseswereobserved[62].Itremainsunknownwhytheinci- denceofBRONJissolowwhenBPtherapyisappliedtoosteogenesis imperfectaandCNO.

9. PathophysiologicalmechanismofBRONJ

Based onourknowledge assummarized above,theputative mechanism of BRONJ formation is summarized in Fig. 3. After administration,nBPsdepositandaccumulateinthealveolarbone, whichhaveahigherturnoverrate.Duringaphysiologicalremod- eling of the bone, osteocytes are exposed to the buried nBPs and damaged, leading to the micronecrosis of the bones. The amountsofthenecroticbonesgraduallyincreasebecauseosteo- clastscannotdegradethenecroticbones.Whentoothextraction is performed, then, the inflammatory cytokines are produced in theperiodontal tissues,which exert theserial inflammatory reactions,wherethetissue-degradingproteasesmaybeinvolved in the increased release of nBPs from the alveolar bones. The releasednBPsmayblockangiogenesisanddelaythemigrationof neutrophils,macrophages,andosteoclastprogenitorstothesite, inhibitingtheformationofgranulationtissues,whichareneeded

Fig.3. Hypothesisofpathophysiologyofbisphosphonate(BP)-relatedosteonecrosisofthejawdevelopment.

forwoundhealingandboneremodeling.Thepoorvascularityand sequestrationinperiodontaltissuesmakesthebacterialinfection persistbecausetherecruitmentofneutrophilsandlymphocytes areblocked.Thefistulaformationtodischargepusinsideboneis alsopreventedbecauseofthelackofosteoclasts.Thepersisting infectionpromotestheincreaseinosteonecrosiswhiletheuncou- plingbetweenosteoblastsandosteoclastscausesthesuppression ofboneturnover.ThereleasednBPsfinallyinhibittheproliferation ofmucosalkeratinocytes,leadingtotheexposureofnecroticbones.

10. TreatmentofBRONJ

ThestrategiesforthetreatmentofBRONJhavebeenproposedin thepositionpapersdependingonthestages[5,9,12].BRONJstage 3maybetreatedwithasurgicalremovalofthejawwhichcontains thenecrotic boneswhilea conservativetherapyisindicated for stage1orstage2.Oneofthebasicconservativetherapiesisalocal irrigationwithsalineortheuseoforalantimicrobialrinse.

10.1. Drugholiday

InthepatientstreatedwithoralBPtherapy,adrugholiday,dis- continuationofBPtherapy,seemstobeeffectivetorestorebone turnoverofthejawandsupportthetreatmentofBRONJ.Inthe positionpapers, the discontinuation ofBP is recommended for morethan3monthsbeforetoothextractionifthepatientshave receivedoralBPtherapyformorethan3years.Itisexpectedthat boneturnoverofthejawoncedisturbedwithBPwillberecovered duringa3-monthdrugholiday.Thisconceptissupportedbythe reportofMarxetal.[63],wheretheserumconcentrationsofCTX, whichisacross-linkingpeptideoftypeIcollagen,releasedfrom thebonesduringboneresorptionandutilizedasamarkerofbone turnover,weremeasuredinthepatientswithBRONJ.Theyfound thattheserumconcentrationsofCTXincreasedfrom72.4pg/ml to150pg/ml,whichisinthelevelforlowriskofBRONJ,withthe discontinuationofBPfor3–4months.Hence,a3-monthholiday fromBPmay restorebone turnoverof thejaw.Infact, Treister andWoo[64]reportedthatthemucosacompletelycoveredthe exposedboneinstage2ofBRONJ4monthsafterthediscontinu- ationofBPandlocalrinse.DiscontinuationoforalBPtherapyis thoughttobethefirststeptocareforpatientswithBRONJ.

ItremainscontroversialwhetherserumCTXconcentrationcan beusedasapredictivemarkerforBRONJ,becausethevaluesofCTX concentrationseemtovaryfromcasetocase[65,66].

10.2. Hyperbaricoxygen

Sincehyperbaricoxygen(HBO)isknowntobeeffectiveadjunc- tive therapy for the treatment of chronic osteomyelitis and osteoradionecrosisofthejaw,then,ithasalsobeenappliedtothe treatmentofBRONJ.TheefficacyofHBOinthetreatmentofBRONJ hasnotyetbeenelucidated,butmaybepromisingbecauseitis expectedtoimprovethehypoxiaconditioninthejawandgener- atereactiveoxygenspecies(ROS)tostimulatethedifferentiation andactivityofosteoclasts[67].

10.3. Teriparatide(recombinanthumanparathyroidhormone 1–34)

Recently, it hasbeen reported that BRONJ wasdramatically improvedwithteriparatidetherapy[68–70].Teriparatideconsists of1–34aminoacidsofrecombinanthumanparathyroidhormone (PTH)(afullPTHincludes84aminoacids),andhasbeenusedfor thetreatmentofGC-inducedosteoporosis.PTHplaysanessential roleintheregulationofcalciummetabolism.WhenPTHlevelsare continuouslyelevated,bonesareseverelydegradedtoincreasethe serumcalciumconcentrationbecauseofosteoclastactivation.In contrast,intermittentpulsatileadministrationofPTHstimulates thedifferentiationandfunctionofosteoblasts,ratherthanosteo- clasts,toleadtoanaboliceffectsonbone[71].Hence,intermittent subcutaneousinjectionofteriparatideonceadayisexpectedto promoteboneformation.InthetreatmentofBRONJ,teriparatide maycauseactivationofosteoblaststorestoretheboneturnover onceinhibitedwithnBP,andpromotetheproductionofreceptor activationofnuclearfactor-kBligand(RANKL)fromosteoblaststo reactivateosteoclasts.

10.4. Prospectivetherapy

Asmentionedabove,nBPsimpairosteoclaststhroughinhibiting FPPsynthaseinthemevalonatepathway(Fig.2).Ifadownstream moleculeoffarnesyldiphosphatewhosegenerationismediated

Fig.4.Aroleofosteoclastsinself-defenseofthejaw.Osteoclastsplayanimportantroleinexcludingbacteriafromtheinsideofjaw.BP,bisphosphonate.

Table2

Listofthedrugsrelatedtoosteonecrosisofthejaw.

Drugsrelatedtoosteonecrosisofthejaw Bisphosphonate

Glucocorticoid Bevacizumab Denosumab Sunitinib

withFPPsynthaseinthemevalonatepathwayissuppliedintothe nBP-treatedosteoclasts,theinhibitoryeffectsofnBPonosteoclasts wouldbecompensated.Kimachietal.[50]foundthattheaddition ofgeranylgeranyldiphosphate,whichisadownstream metabo- liteofFPPsynthase,dramaticallyrestoredthemotilityandRANK expressionoftheculturedosteoclastprecursorswhichwereonce inhibitedwithzoledronate,implyingthatgeranylgeranyldiphos- phateneutralizestheeffectofnBP.Suchanintermediatemetabolite inthemevalonatepathwayisexpectedtobecomeadrugtorevive theosteoclastsinthepatientswithBRONJ[72].

11. Drug-relatedosteoclasticdiseaseofthejaw

Recently,ithasbeensuspectedthatONJisinducednotonlyby nBP,butalsobythemolecule-targetingdrugs,whicharenewly developed,suchasdenosumab,anti-RANKLantibody,andbeva- cizumab, an anti-VEGF antibody [2,40,73]. The number of case reportsaboutONJoccurringincancerpatientstreatedwiththese new anti-cancer drugs appears to be increasing (Table 2).The incidenceof ONJwas0.3–0.4%inthe patientswithbreast can- cerundergoingbevacizumabtherapywhereastheadministration ofbevacizumab togetherwithnBP increasedtheONJincidence to0.9–2.4%[74].Thecombination ofnBP withsunitinib, which inhibitstyrosinekinaseinvolvedinvariousgrowthfactorreceptors totreatrenalcellcarcinoma,isreportedtoraisetheriskofBRONJ, wheresunitinib-inducedoralmucositisisthoughttotriggerthe occurrenceofBRONJ[75].

Accordingtorecentadvancesinmedicalscience,anumberof newdrugs have beendeveloped and appliedtopatients, espe- ciallythosewithcancer.Oursocietywillspeedupthissituation inthefuture.Ontheotherhand,themorenewdrugsaredevel- oped,themoreadverseeffectswillemerge.Newdrugsmayinduce newdiseasessuchasBRONJ.Itisplausiblethatthejawisthemost vulnerablesitetotheadverseeffectsofnewdrugsbecauseofits specialanatomicalcharacteristics.Furthermore,ajawiscomposed ofavarietyofcells,allofwhichshouldtakepartinthelifeofthejaw.

EvendamageonlytoosteoclastswithBPwilldisturbthenetwork ofeachcellthroughthejaw.Fromthisviewpoint,ourexperiences

aboutBRONJgivebirthtoaconceptthatosteoclastsplayacriti- calroleintheself-defenseofthejaw(Fig.4).Hence,itsuggests thatBRONJorotherrelatedjawdiseaseisnewlyclassifiedasdrug- relatedosteoclasticdiseaseofthejaw.Weshouldprepareforthe challengesfromsuchnewdiseases.

References

[1]MarxRE. Pamidronate (Aredia) and zoledronate(Zometa) inducedavas- cular necrosis of the jaws: a growing epidemic. J Oral Maxillofac Surg 2003;61:1115–7.

[2]KimmeDB.Mechanismofaction,pharmacokineticandpharmacodynamicpro- file,andclinicalapplicationsofnitrogen-containingbisphosphonates.JDentRes 2007;86:1022–33.

[3]TaylorKH,MiddlefellLS,MizenKD.Osteonecrosisofthejawsinducebyanti- RANKligandtherapy.BrJOralMaxillofacSurg2010;48:221–313.

[4]AghalooTL,FelsenfeldAL,TetradisS.Osteonecrosisofthejawinapatienton denosumab.JOralMaxillofacSurg2010;68:959–63.

[5]AdvisoryTaskForceonBisphosphonate-RelatedOsteonecrosisoftheJaw.

American Association of Oral and Maxillofacial Surgeons position paper onbisphosphonate-relatedosteonecrosisofthejaws.JOralMaxillofacSurg 2007;65:369–76.

[6]The impact of osteonecrosis of the jaw on osteonecrosis manage- ment: executive summary of a European society on clinical and eco- nomic aspects of osteoporosis and foundation for research on osteo- porosis and other bone diseases working group meeting. Available at http://www.ecceo8.org/mailing/esceo/esceo0530.html.

[7]MavrokokkiT,ChengA,SteinB,etal.Natureandfrequencyofbisphosphonate- associated osteonecrosis of the jaws in Australia. J Oral Maxillofac Surg 2007;65:415–23.

[8] LoJC,O’RyanFS,GordonNP,etal.Predictingriskofosteonecrosisofthe jawwithoralbisphosphonateexposure(PROBE)investigators.Prevalenceof osteonecrosisofthejawinpatientswithoralbisphosphonateexposure.JOral MaxillofacSurg2010;68:243–53.

[9]UradeM,TanakaN,FurusawaK,etal.Nationwidesurveyforbisphosphonate- related osteonecrosis of the jaws in Japan. J Oral Maxillofac Surg 2011;69:e364–71.

[10]SedghizadehPP,StanleyK,CaligiuriM,etal.Oralbisphosphonateuseandthe prevalenceofosteonecrosisofthejaw:aninstitutionalinquiry.JAmDentAssoc 2009;140:61–6.

[11]Boonyapakorn T, SchirmerI, Reichart PA, et al. Bisphosphonate-induced osteonecrosisofthejaws:prospectivestudy of80 patientswithmultiple myelomaandothermalignancies.OralOncol2008;44:857–69.

[12]RuggieroSL,DodsonTB,AssaelLA,etal.AmericanAssociationofOralandMax- illofacialSurgeonsAmericanAssociationofOralandMaxillofacialSurgeons positionpaperonbisphosphonate-relatedosteonecrosisofthejaws–2009 update.JOralMaxillofacSurg2009;67(Suppl.1):2–12.

[13] ArceK,AssaelLA,WeissmanJL,etal.Imagingfindingsinbisphosphonate- relatedosteonecrosisofjaws.JOralMaxillofacSurg2009;67(Suppl.1):75–84.

[14]Kos M, Brusco D, Kuebler J, et al. Clinical comparison of patients with osteonecrosis of the jaws, with and without a history of bisphosphonatesadministration.IntJOralMaxillofacSurg2010;39:1097–120.

[15]GanguliA,StewardC,ButlerSL,etal.Bacterialadhesiontobisphosphonate coatedhydroxyapatite.JMaterSciMaterMed2005;16:283–7.

[16] HansenT,KunkelM,WeberA,etal.Osteonecrosisofthejawsinpatients treatedwithbisphosphonates– histomorphologicanalysisincomparisonwith infectedosteoradionecrosis.JOralPatholMed2006;35:155–60.