nasopharyngeal carcinoma

15
Seminar Introduction Nasopharyngeal carcinoma is a non-lymphomatous, squamous-cell carcinoma that occurs in the epithelial lining of the nasopharynx. This neoplasm shows varying degrees of differentiation and is frequently seen at the pharyngeal recess (Rosenmüller’s fossa) posteromedial to the medial crura of the eustachian tube opening in the nasopharynx. 1 The first report on a group of 14 patients who had this type of tumour was published in 1901. 2 A further clinical study of 79 patients was published in 1922. 3 The first comprehensive study of nasopharyngeal carcinoma was done in 1941, and described clinicopathological features in 114 patients. 4 This neoplasm is an uncommon disease in most countries, and its age-adjusted incidence for both sexes is less than one per 100 000 population. 5 However, the disease occurs with much greater frequency in southern China, northern Africa, and Alaska. The Inuits of Alaska 6 and ethnic Chinese people living in the province of Guangdong are especially prone to the disease. The reported incidence of nasopharyngeal carcinoma among men and women in Hong Kong (geographically adjacent to Guangdong province) is 20–30 per 100 000 and 15–20 per 100 000, respectively. 5 That the incidence of nasopharyngeal carcinoma remains high among Chinese people who have immigrated to southeast Asia or North America, but is lower among Chinese people born in North America than in those born in southern China, is noteworthy. 7,8 This finding suggests that genetic, ethnic, and environmental factors could have a role in the cause of the disease. Pathology The malignant epithelial cells of the nasopharynx are large polygonal cells with a syncytial composition. Their nuclei are round or oval with scanty chromatin and distinct nucleoli. The cells show no parakeratosis or cornification and are frequently intermingled with lymphoid cells in the nasopharynx, giving rise to the introduction of the term lymphoepithelioma. 9 Electronmicroscopy studies have established that these tumour cells are of squamous origin and that the undifferentiated carcinoma is a form of squamous-cell carcinoma. 10,11 Epstein-Barr virus (EBV) is consistently detected in patients with nasopharyngeal carcinoma from regions of high and low incidence. EBV-encoded RNA signal has been shown, by in-situ hybridisation, to be present in nearly all tumour cells, whereas EBV-encoded RNA is absent from the adjacent normal tissue, except perhaps for a few scattered lymphoid cells. Premalignant lesions of the nasopharyngeal epithelium have also been shown to harbour EBV, which suggests that the infection occurs in the early phases of carcinogenesis. 12 Detection of a single form of viral DNA suggests that the tumours are clonal proliferations of a single cell that was initially infected with EBV. Specific EBV latent genes are consistently expressed in nasopharyngeal carcinomas and in early, dysplastic lesions. The corresponding latent viral proteins (latent membrane protein 1 and 2) have substantial effects on cellular gene expression and cellular growth, resulting in the highly invasive, malignant growth of the carcinoma. 13,14 The histological classification of nasopharyngeal carcinoma proposed by WHO in 1978, categorised tumours into three groups: type I included typical keratinising squamous-cell carcinomas, similar to those Lancet 2005; 365: 2041–54 Department of Surgery (Prof W I Wei FRCS) and Department of Clinical Oncology (Prof J S T Sham FRCR), University of Hong Kong Medical Centre, Queen Mary Hospital, Hong Kong SAR, China Correspondence to: Professor William I Wei [email protected] www.thelancet.com Vol 365 June 11, 2005 2041 Nasopharyngeal carcinoma William I Wei, Jonathan S T Sham Incidence of nasopharyngeal carcinoma has remained high in endemic regions. Diagnosing the disease in the early stages requires a high index of clinical acumen and, although most cross-sectional imaging investigations show the tumour with precision, confirmation is dependent on histology. Epstein-Barr virus (EBV)-encoded RNA signal is present in all nasopharyngeal carcinoma cells, and early diagnosis of the disease is possible through the detection of raised antibodies against EBV. The quantity of EBV DNA detected in blood indicates the stage and prognosis of the disease. Radiotherapy with concomitant chemotherapy has increased survival, and improved techniques (such as intensity-modulated radiotherapy), early detection of recurrence, and application of appropriate surgical salvage procedures have contributed to improved therapeutic results. Screening of high-risk individuals in endemic regions together with developments in gene therapy and immunotherapy might further improve outcome. Search strategy and selection criteria We did an extensive search of published work about nasopharyngeal carcinoma through the Pubmed/MEDLINE database from 2004 back to the mid-60’s, and this search included the OLDMEDLINE. We also searched The Cochrane Library for review articles published between 1990 and 2003. Search terms included: “nasopharyngeal carcinoma”, “nasopharynx cancer”, “radiotherapy”, “chemotherapy”, “salvage therapy”, “nasopharyngectomy”, “combined treatment modality”, “clinical trials”, “randomized controlled trials”, and “meta-analysis”. The searches included reports of studies on human beings and were those published in English. Priority of selection was large contemporary clinical trials or studies. Abstracts of recent pertinent medical conferences were also included for information about latest developments. Most of the publications used were from work published within the past 15 years, although we have also included a few highly regarded old articles.

Upload: chainurridha-sii-dedek

Post on 18-Aug-2015

212 views

Category:

Documents


0 download

DESCRIPTION

NPC

TRANSCRIPT

SeminarIntroductionNasopharyngealcarcinomaisanon-lymphomatous,squamous-cellcarcinomathatoccursintheepitheliallining of the nasopharynx. This neoplasm shows varyingdegreesofdifferentiationandisfrequentlyseenatthepharyngealrecess(Rosenmllersfossa)posteromedialto the medial crura of the eustachian tube opening in thenasopharynx.1The rst report on a group of 14 patients who had thistype of tumour was published in 1901.2A further clinicalstudyof79patientswaspublishedin1922.3Therstcomprehensivestudyofnasopharyngealcarcinomawasdone in 1941, and described clinicopathological featuresin 114 patients.4This neoplasm is an uncommon diseasein most countries, and its age-adjusted incidence for bothsexes is less than one per 100 000 population.5However,thediseaseoccurswithmuchgreaterfrequencyinsouthern China, northern Africa, and Alaska. The InuitsofAlaska6andethnicChinesepeoplelivingintheprovinceofGuangdongareespeciallypronetothedisease.ThereportedincidenceofnasopharyngealcarcinomaamongmenandwomeninHongKong(geographicallyadjacenttoGuangdongprovince)is2030per100 000and1520per100 000,respectively.5That the incidence of nasopharyngeal carcinoma remainshighamongChinesepeoplewhohaveimmigratedtosoutheastAsiaorNorthAmerica,butisloweramongChinesepeopleborninNorthAmericathaninthoseborninsouthernChina,isnoteworthy.7,8Thisndingsuggests that genetic, ethnic, and environmental factorscould have a role in the cause of the disease. Pathology The malignant epithelial cells of the nasopharynx are largepolygonal cells with a syncytial composition. Their nucleiareroundorovalwithscantychromatinanddistinctnucleoli. The cells show no parakeratosis or cornicationand are frequently intermingled with lymphoid cells in thenasopharynx,givingrisetotheintroductionofthetermlymphoepithelioma.9Electronmicroscopystudieshaveestablished that these tumour cells are of squamous originandthattheundifferentiatedcarcinomaisaformofsquamous-cell carcinoma.10,11Epstein-Barrvirus(EBV)isconsistentlydetectedinpatients with nasopharyngeal carcinoma from regions ofhighandlowincidence.EBV-encodedRNAsignalhasbeenshown,byin-situ hybridisation,tobepresentinnearlyalltumourcells,whereasEBV-encodedRNAisabsent from the adjacent normal tissue, except perhaps fora few scattered lymphoid cells. Premalignant lesions of thenasopharyngealepitheliumhavealsobeenshowntoharbour EBV, which suggests that the infection occurs inthe early phases of carcinogenesis.12 Detection of a singleformofviralDNAsuggeststhatthetumoursareclonalproliferations of a single cell that was initially infected withEBV. Specic EBV latent genes are consistently expressedinnasopharyngealcarcinomasandinearly,dysplasticlesions.Thecorrespondinglatentviralproteins(latentmembraneprotein1and2)havesubstantialeffectsoncellular gene expression and cellular growth, resulting inthe highly invasive, malignant growth of the carcinoma.13,14ThehistologicalclassicationofnasopharyngealcarcinomaproposedbyWHOin1978,categorisedtumoursintothreegroups:typeIincludedtypicalkeratinising squamous-cell carcinomas, similar to thoseLancet 2005; 365: 204154Department of Surgery(Prof WI Wei FRCS) andDepartment of ClinicalOncology(Prof J S T ShamFRCR),University of Hong KongMedical Centre, Queen MaryHospital, Hong Kong SAR,ChinaCorrespondence to: Professor William I [email protected] Vol 365 June 11, 20052041Nasopharyngeal carcinomaWilliam I Wei, Jonathan S T ShamIncidence of nasopharyngeal carcinoma has remained high in endemic regions. Diagnosing the disease in the earlystages requires a high index of clinical acumen and, although most cross-sectional imaging investigations show thetumourwithprecision,conrmationisdependentonhistology.Epstein-Barrvirus(EBV)-encodedRNAsignalispresent in all nasopharyngeal carcinoma cells, and early diagnosis of the disease is possible through the detection ofraised antibodies against EBV. The quantity of EBV DNA detected in blood indicates the stage and prognosis of thedisease.Radiotherapywithconcomitantchemotherapyhasincreasedsurvival,andimprovedtechniques(suchasintensity-modulatedradiotherapy),earlydetectionofrecurrence,andapplicationofappropriatesurgicalsalvageprocedures have contributed to improved therapeutic results. Screening of high-risk individuals in endemic regionstogether with developments in gene therapy and immunotherapy might further improve outcome.Search strategy and selection criteriaWe did an extensive search of published work aboutnasopharyngeal carcinoma through the Pubmed/MEDLINEdatabase from 2004 back to the mid-60s, and this searchincluded the OLDMEDLINE. We also searched The CochraneLibrary for review articles published between 1990 and 2003.Search terms included: nasopharyngeal carcinoma,nasopharynx cancer, radiotherapy, chemotherapy,salvage therapy, nasopharyngectomy, combinedtreatment modality, clinical trials, randomized controlledtrials, and meta-analysis. The searches included reports ofstudies on human beings and were those published inEnglish. Priority of selection was large contemporary clinicaltrials or studies. Abstracts of recent pertinent medicalconferences were also included for information about latestdevelopments. Most of the publications used were from workpublished within the past 15 years, although we have alsoincluded a few highly regarded old articles. Seminarfound in the rest of the upper aerodigestive tract; type IIincludednon-keratinisingsquamouscarcinomas;andtypeIIIincludedundifferentiatedcarcinomas(panel).15Analternativeclassicationhasdividedtumoursintotwo histological types, namely squamous-cell carcinomasandundifferentiatedcarcinomasofthenasopharyngealtype.16ThesecondclassicationiscorrelatedwithEBVserology: patients with squamous-cell carcinomas have areducedEBVtitre,whereasthosewithundifferentiatedcarcinomas of the nasopharyngeal type have raised titres.In North America, around 25% of tumour patients havetype I histology, 12% have type II, and 63% have type III.ThehistologicaldistributioninsouthernChinesepatients is 2%, 3%, and 95%, respectively.17Biopsiesobtainedfromnasopharyngealcarcinomassometimesshowamixedhistologicalpattern,andthispatternvariesamongdifferentpartsofthetumour.ThemostrecentWHOclassicationhastakenthismixedpattern into account as well as the association of EBV withtypeIIandtypeIIItumours.Thehistologicaltypesofnasopharyngealcarcinomaarenowdenedeitherassquamous-cellcarcinomasornon-keratinisingcarcinomas,andthesecondgroupissubdividedintodifferentiatedandundifferentiatedcarcinomas.18Thisclassicationismoreapplicableforepidemiologicalresearchandhasalsobeenshowntohaveaprognosticbearing. Undifferentiated carcinomas have a higher localtumour control rate with treatment and a higher incidenceof distant metastasis than do differentiated carcinomas.19,20Symptoms and serological diagnosisPatientswithnasopharyngealcarcinomacanpresentwith symptoms from one or more of four categories. Thecategories consist of (1) presence of tumour mass in thenasopharynx(epistaxis,nasalobstruction,anddischarge);(2)dysfunctionoftheeustachiantube,associatedwiththelateroposteriorextensionofthetumourtotheparanasopharyngealspace(tinnitusanddeafness);(3)skull-baseerosionandpalsyofthefthandsixthcranialnerves,associatedwiththesuperiorextension of the tumour (headache, diplopia, facial painand numbness); and (4) neck masses, usually appearingrst in the upper neck. Symptoms such as anorexia andweightlossareuncommoninpatientswithnasopharyngealcarcinomasanddistantspreadshouldbesuspectedwhensuchsymptomsarepresent.Unfortunately, because of the non-specic nature of thenasal and aural symptoms and the difculty of making aclinicalexaminationofthenasopharynx,mostpatientswiththediseasearediagnosedonlywhenthetumourhas reached an advanced stage (stages III and IV).Aretrospectiveanalysisof4768patientsidentiedsymptoms of nasopharyngeal carcinoma at presentationasneckmass(76%),nasaldysfunction(73%),auraldysfunction(62%),headache(35%),diplopia(11%),facialnumbness(8%),weightloss(7%),andtrismus(3%).Thephysicalsignspresentatdiagnosiswereenlarged neck node (75%) and cranial nerve palsy (20%).Thecranialnervesmostcommonlyaffectedwerethethird,fth,sixth,and12thnerves.21,22Thepresentingsymptomsinyoungpatientswereingeneralsimilartothose reported in adults.23Patientswhopresentwithsymptomsofnasopharyngealcarcinomashouldbeclinicallyassessedfor physical signs of the disease. A positive EBV serologytestwillgivefurthergroundsforsuspicionandwouldjustify an endoscopic examination and a biopsy from thenasopharynx. If the clinical suspicion for nasopharyngealcarcinomaishigh,evenifthesuspectedtumourisnotvisualisedwithendoscopicexamination,cross-sectionalimaging by CT or MRI should be undertaken. A denitivediagnosisofnasopharyngealcarcinomaneedsapositivebiopsytakenfromthetumourinthenasopharynx,supportedeitherbyitsvisualisationinthenasopharynxor (in the case of predominantly submucosal tumours) itsvisualisation with cross-sectional imaging. Population screeningIn southern China, where nasopharyngeal carcinoma isendemic,EBVserologyhasbeenusedforpopulationscreening.InastudyundertakeninWuzhou(Guangxiprovince,China)24intheearly1980s,1136individualsidentied as positive for immunoglobulin A against viralcapsid antigen received regular clinical examinations ofthenasopharynxandneckfor4 years.Duringthisfollow-up period, 35 cases of nasopharyngeal carcinomaweredetected,mostofwhich(92%)werediagnosedearlyateitherstageIorstageII.Theannualdetectionrateofnasopharyngealcarcinomaforthisgroupwas317 times higher than for the population as a whole. 2042 www.thelancet.com Vol 365 June 11, 2005 Panel: WHO histological classication of nasopharyngealcarcinoma15Keratinising squamous-cell carcinoma (WHO type I)This type of nasopharyngeal carcinoma shows squamousdifferentiation with the presence of intercellular bridgesand/or keratinisation over most of its extent.Non-keratinising carcinomaThis group comprises a differentiated type of non-keratinisingcarcinoma and an undifferentiated type. These tumours aregenerally more radiosensitive than squamous cell carcinomaand have stronger relationships with the Epstein-Barr virus.1. Differentiated non-keratinising carcinoma (WHO type II)The tumour cells show differentiation with a maturationsequence that results in cells in which squamousdifferentiation is not evident on light microscopy.2. Undifferentiated carcinoma (WHO type III)The tumour cells have oval or round vesicular nuclei andprominent nucleoli. The cell margins are indistinct, and thetumour exhibits a syncytial rather than pavementedappearance.SeminarSimilar results were reported from another study doneinZhongshan(Guangdongprovince,China).25Thesensitivity and predictive value of serology in populationscreeningwasproposedtobeimprovedbytestingagainst a panel of EBV antibodies.26The predictive valueof EBV serology for nasopharyngeal carcinoma was lentsupportbyamorerecentreportfromTaiwan.27Inthisstudy, the initial EBV serology of 9699 study participantswas cross-checked against the cancer registry and deathregistryintheensuing15-yearperiod.Thedurationoffollow-upwascorrelatedwiththedifferenceinthecumulativeincidenceofnasopharyngealcarcinomabetweenseropositiveandseronegativepatients.Prospectivestudiesarenowneededtoassesstheeffectofsuchpopulation-basedscreening,intermsofthereductioninmortalityrelatedtonasopharyngealcarcinomainthescreenedpopulation;therisk-benetratio (risks from endoscopic examination and biopsies);and cost-effectiveness. Imaging studiesBefore the introduction of cross-sectional imaging, littlewasknownaboutthenaturalbehaviourandroutesofextensionofnasopharyngealcarcinomasintheearlystagesofdevelopment.Surgerywasnotaprimarytreatment,andpost-mortemexaminationsofpatientswho died from nasopharyngeal carcinoma were of littleimportancesincethetumourswereusuallyveryadvancedbythetimeofdeathandhadundergonesignicantsecondarychangesasaresultoftreatment.The best that could be done was to use plain radiographsto assess bone destruction and soft tissue mass abuttingontheupperairway,butthesetechniqueswereoflowsensitivityandspecicityandaddedlittletoourknowledge of invasion and extension of the disease.Clinicalexamination(includingendoscopicexamination)canprovidevaluableinformationaboutmucosalinvolvementandtumourextensionintothenasalfossaeandoropharynx,butcannotascertaindeepextension,skull-baseerosion,orintracranialspread,exceptwheretherearetell-talesymptomsandsignsofgrossextensionalongtheseroutes.Cross-sectionalimaginghasrevolutionisedandimprovedtheeffectiveness of treatment for nasopharyngeal carcinoma.Intermsofcontributiontostaging,CThasidentiedparanasopharyngealextensionasoneofthemostcommonmodesofextensionofnasopharyngealcarcinoma28andhasshownperineuralspreadthroughthe foramen ovale to be an important route of intracranialextension. Perineural spread through the foramen ovalealsoaccountsfortheCTevidenceofcavernoussinusinvolvement without skull-base erosion.29MRIisbetterthanCTfordisplayingbothsupercialanddeepnasopharyngealsofttissueandfordifferentiatingtumourfromsofttissue.MRIisalsomoresensitiveforassessmentofretropharyngealanddeepcervicalnodalmetastases.30However,thetechniqueisoflimitedeffectivenessforassessingbonedetails and CT should be undertaken when the status ofthe base of the skull cannot be satisfactorily establishedwithMRI.31Intermsofstaging,MRIisabletodetectmarrowinltrationbytumours,whereasCTcannotdetect this kind of inltration unless there is associatedbone erosion. This kind of marrow inltration has beensuggestedtobeassociatedwithanincreasedriskofdistant metastases.32Detectionofdistantmetastasesatdiagnosiswithconventionalradiographs,CT,andMRIisnotusuallysuccessful.Severalreportshaveconcludedthatbonescans,33liver scintigraphy,34 abdominal ultrasonography,35and marrow biopsy36are of little value in routine stagingandhaverecommendedthattheyneednotbeused.Astudyconcludedthattherewasnoevidencetolendsupporttodistantimagingforlow-risk(N0orstageI)disease,butrecommendedthathigh-risk(N3)diseaseshould be fully staged with chest radiography, bone scan,andliverultrasonography.37Theroleofpositronemissiontomography(PET)inthedetectionofdistantmetastases in other malignancies has been established,38butitsapplicationinthestagingofnasopharyngealcarcinoma has not been ascertained. Cross-sectionalimagingdisplaystheextentoftheprimarytumourwithunprecedentedprecision.Thisaccuracy enables radiotherapy treatment to be designedand administered more accurately, effectively improvingtheoutcomesforthisformoftreatment.39Evenbetterresultshavebecomepossiblewithintensitymodulatedradiotherapy,whichmakesuseofcompositeCT-MRItargets40enabling radiotherapy to be targeted even moreaccurately onto tumours, sparing adjacent tissues.Whenusedtomonitorapatientsconditionaftertreatment,bothCTandMRIhavelowsensitivityandmoderatespecicityinthedetectionoftumourrecurrence;41although, in general, MRI is better than CTinshowingtumourrecurrenceandpostradiationcomplications.42Recurrentnasopharyngealcarcinomascanexhibitavarietyofsignalintensitiesandcontours,and these can be difcult to interpret.43However, CT canshow bone regeneration after treatment, which could bean indication of the complete eradication of the tumourintheaffectedarea44suggestingafavourableprognosiswith related clinical ndings.45PET has been reported tobe more sensitive than CT and MRI at detecting residualand recurrent tumours in the nasopharynx.46Staging systemTherearevariouswaysofclassifyingnasopharyngealcarcinomas.AtpresenttheAmericanJointCommitteeonCancerStagingandEndResultReporting/InternationalUnionAgainstCancer(AJC/UICC)systemispreferredinEuropeandAmerica,47whereasHossystemisfrequentlyusedinAsia.48,49ThenodalclassicationinHossystemhasincorporatedprognosticsignicance,butthestraticationofthewww.thelancet.com Vol 365 June 11, 2005 2043SeminarT stagesintovesectorsdiffersfrommoststagingsystems.Thedevelopmentofarevisedstagingsysteminthepastdecadewasmotivatedbythedesiretoincorporateexperiencesgainedfromvariouscentresaroundthe world,takingintoaccountmanyprognosticfactors, includingskull-baseerosion,involvementof cranial nerves,50primarytumourextensiontoparanasopharyngeal space,51and the level and size of thecervicalnodes.52ArevisedAJC/UICCstagingsystemwas published in 1997;53in this new staging system theT1 stage included tumours classied as both T1 and T2undertheoldsystem.ThenewT2stagecoveredtumoursthathadextendedtothenasalfossa,oropharynx,orparanasopharyngealspace.ThenewT3stagecoveredtumoursthathadextendedtotheskullbaseorotherparanasalsinuses.ThenewT4stagecoveredtumoursthathadextendedintotheinfratemporalfossa,orbit,hypopharynx,andcranium,ortothecranialnerves.Forcervicalnodalstaging,N1underthenewsystemreferredtounilateralnodalinvolvement;N2tobilateralnodaldiseasethathadnotreachedN3designation,irrespectiveofthesize,number, and anatomical location of the nodes; and N3 tolymphnodeslargerthan6cm(N3a),ornodesthathadextendedtothesupraclavicularfossa(N3b).54Thenewstagingsystemhasenabledpatientstobestagedmoresensitivelyandisabetterpredictorofsurvivalthantheold system (table 1).55,56PrognosisAswithmostothertumours,theextentofanasopharyngealcarcinomaasembodiedintheTMNstaging system (table 1) is the most important prognosticfactor. Indeed, most other known prognostic factors aredirectlyorindirectlyrelatedtotheextentorbulkofthetumour.Thechangesinprognosticfactorsidentiedandreportedatdifferenttimesinthepastprobablyrepresented adoption of these known adverse factors inthenewstagingsystems,ortheuseoftreatmentstrategiestoaddresstheseknownadverseprognosticfactors and to nullify their adverse effects.Areportin199057showedthat,besidestheTandN stages,otherprognosticfactorsincludedsizeanddegree of xation of neck nodes, sex, age, the presence ofcranialnervepalsy,andearsymptomsatpresentation.ThefactorsofsizeoflymphnodeandearsymptomsprobablysuggestthelackofrecognitionofnodalsizeandparanasopharyngealextensionintheTandN staging system used at that time. A study reported in199258showedthatthetumourshistologicaltypeandthe radiotherapy dose and coverage were also signicantindependent prognostic factors. The adverse prognosticfactor of histological type is shown in this report, whichincluded mainly the white population with WHO type Ihistology.Paranasopharyngealextensionwasanindependentprognosticfactorcorrelatedwithadverselocal tumour control and increased distant spread.59 Evenafter paranasopharyngeal extension of tumour had beenincorporatedintothe1997AJC/UICCstageclassications,theadverseprognosticeffectremainsvalid despite the use of concurrent chemoradiotherapy.60AlargevariationoftumourvolumeispresentinT stagesofdifferentstagingsystems,andprimarytumourvolumerepresentsanindependentprognosticfactoroflocalcontrolandismorepredictivewiththeAJC/UICCstagingsystemthanwithHosTstageclassication.61ValidityoftumourvolumehasbeenconrmedinpatientswithT3andT4tumours,62andthere is an estimated 1% increase in risk of local failurefor every 1 cm3increase in primary tumour volume.63Inadditiontodirectmeasurementoftumourvolume,quantitativeanalysisofcirculatingEBVDNAinnasopharyngealcarcinomahasshownapositivecorrelation with disease stage and a strong relation withclinicalevents,aswellasexhibitingprognosticimportance.64Basedonthedifferenceinfailurepatterns,differentprognosticcategoriescanbedenedacrossstages.Theseare(1)T12N01(relativelygoodtreatmentoutcome);(2)T34N01(mainlylocalfailure);(3) T12N23 (mainly regional and distant failure); and(4) T34N23 (local, regional and distant failure). Theseprognosticgroupingswillhaveimportantimplicationsfor the selection of appropriate treatment strategies andthedesignoffutureclinicaltrialstoaddressdifferentfailurepatterns.65Thereisearlyevidencethatforadvanceddiseases,addingchemotherapytoradiotherapywillimprovetreatmentoutcome,bothinterms of locoregional control and distant metastases.66,67Treatment RadiotherapyRadiotherapyisthestandardtreatmentfornasopharyngealcarcinoma.Unfortunately,itcanproduceundesirablecomplicationsaftertreatmentbecause of the location of the tumour at the base of skull,closely surrounded by and in close proximity to radiationdose-limitingorgans,includingthebrainstem,spinalcord,pituitary-hypothalamicaxis,temporallobes,eyes,middleandinnerears,andparotidglands.Sincenasopharyngeal carcinomas tend to inltrate and spreadtowards these dose-limiting organs, they are even moredifcult to protect.One of the most common radiotherapy approaches fornasopharyngealcarcinomaistostartphaseItreatmentwith large lateral opposing faciocervical elds that covertheprimarytumourandtheuppernecklymphaticsinone volume, with matching lower anterior cervical eldforlowernecklymphatics.Whenthespinalcorddosereaches4045Gy,therearetwooptionsforphaseIItreatment.Treatmentcaneitherbechangedtolateralopposingfacialeldswithanteriorfacialeldfortheprimarytumour,withmatchinganteriorcervicaleld2044 www.thelancet.com Vol 365 June 11, 2005 Seminarforthenecklymphatics. Alternatively,treatmentcanbecontinuedwiththelateralopposingfaciocervicaleldsbut with shrinkage of elds to avoid the spinal cord, andbytreatingthesuperior-posteriorlymphaticwithelectronelds.68,69Themajorobjectiontotreatingtheprimary tumour and the neck lymphatic in two separatevolumes (both of these phase II treatment techniques) isthatthereisadangerofunderdosingtheparanasopharyngealextensionofthetumourandtheuppernecknodesatthejunctionbetweentheprimarytumour and neck lymphatic target volumes.In radiotherapy a dose of 6575 Gy is normally given tothe primary tumour and 6570 Gy to the involved necknodes, whereas the dose for prophylactic treatment for anode-negativeneckis5060Gy.Thistreatmenthassuccessfully controlled T1 and T2 tumours in 7590% ofcasesandT3andT4tumoursin5075%ofcases.65,6871Nodal control is achieved in 90% of N0 and N1 cases, butthe control rate drops to 70% for N2 and N3 cases.65Asinterrupted or prolonged treatment reduces the benetsof radiotherapy, every effort should be made to maintainthe treatment schedule.72Because of the high incidencewww.thelancet.com Vol 365 June 11, 2005 2045The American Joint Committee on Cancer Staging53Ho Staging49Tumour in nasopharynx (T) Primary tumour (T)T1 Tumour conned to the nasopharynx T1 Tumour conned to nasopharynx (space behind choanal orices and nasal septum and aboveposterior margin of soft palate in resting position)T2 Tumour extends to soft tissues of oropharynx and/or nasal fossa T2 Tumour extended to nasal fossa, oropharynx, or adjacent muscles or nerves below base of skullT2a without parapharyngeal extensionT2b with parapharyngeal extensionT3 Tumour invades bony structures and/or paranasal sinuses T3 Tumour extended beyond T2 limits and subclassied as follows:T3a Bone involvement below base of skull (oor of sphenoid sinus is included in this category)T3b Involvement of base of skullT3c Involvement of cranial nerve(s)T3d Involvement of orbits, laryngopharynx (hypopharynx), or infratemporal fossaT4 Tumour with intracranial extension and/or involvement of cranial nerves, infratemporal fossa, hypopharynx, or orbitRegional lymph nodes (N) Regional lymph nodes (N)The distribution and the prognostic effect of regional lymph node spread from nasopharynx cancer, especially of the undifferentiated type, is different from that of other head and neck mucosal cancers and justies use of a different N classication scheme.NX Regional lymph nodes cannot be assessedN0 No regional lymph node metastasis N0 Node palpable or thought to be benignN1 Unilateral metastasis in lymph node(s), 6 cm or less in greatest dimension, above theN1 Node(s) wholly in upper cervical level, bounded below by the skin crease extending laterally and supraclavicular fossa backward from or just below thyroid notch (laryngeal eminence) N2 Bilateral metastasis in lymph node(s), 6 cm or less in greatest dimension, above theN2 Node(s) palpable between crease and supraclavicular fossa, the upper limit being a line joining the supraclavicular fossa upper margin of the sternal end of the clavicle and the angle formed by the lateral surface of the neck and the superior margin of the trapeziusN3 Metastasis in a lymph node(s) N3 Node(s) palpable in the supraclavicular fossa and/or skin involvement in the form of carcinoma en N3a greater than 6 cm in dimension cuirasse or satellite nodules above the claviclesN3b extension to the supraclavicular fossaDistant metastasis (M) Metastases (M)MX Distant metastasis cannot be assessedM0 No distant metastasis M0 No haematogenous metastasesM1 Distant metastasis M1 Haematogenous metastases present, and/or lymph nodal metastases below the clavicleStage grouping Stage groupingStage 0 T1s N0 M0Stage I T1 N0 M0 Stage I T1,N0Stage IIA T2a N0 M0 Stage II T2 and/or N1Stage IIB T1 N1 M0T2 N1 M0T2a N1 M0T2b N0 M0T2b N1 M0Stage IIIT1 N2 M0 Stage III T3 and/or N2T2a N2 M0T2b N2 M0T3 N0 M0 T3 N1 M0T3 N2 M0Stage IVAT4 N0 M0Stage IV N3 (any T)T4 N1 M0T4 N2 M0Stage IVBAny TN3 M0 Stage IVCAny TAny NM1Stage V M1Table 1: Staging systems for nasopharyngeal carcinomaSeminarofoccultnecknodeinvolvement,prophylacticneckradiationisusuallyrecommended.73Goodlocoregionalcontrol should be the prime objective of treatment sincelocoregionalrelapsesrepresentasignicantriskfactorfor the development of distant metastases.74For T1 andT2tumours,aboosterdosebyuseofintracavitarybrachytherapyimprovedtumourcontrolby16%.75Althoughstereotacticradiosurgeryhasalsobeenusedfortheboosterdose,76itisprobablybetterreservedforthetreatmentofpersistentandrecurrentnasopharyngealcarcinomasbecauseoftheundesirableside-effects associated with hypofractionated treatment.77Themajorlimitationsof2Dplanningfornasopharyngealcarcinomacannowbeovercomewith3Dconformalradiotherapyandintensity-modulatedradiotherapy.78,79Whenappliedalone,3Dconformalboost is not effective,80thus conformal radiotherapy andintensity-modulatedradiotherapyshouldbeadoptedthroughoutthetreatment.Inthecaseofextensivetumours, and when the tumour extension is close to thedose-limitingorgans,intensity-modulatedradiotherapyisdistinctlypreferableto3Dconformalplanningbecauseitfurtherimprovesthedosedifferentialbetweenthetumourandthedose-limitingorgans.81,82Intensity-modulatedradiotherapyalsoresolvestheproblem of dose uncertainty at the junction between theprimary tumour and neck lymphatic target volumes as itenablestheprimarytumourandtheuppernecknodestobetreatedinonevolumethroughout.Althoughthistechnique theoretically allows very good dose differentialbetweenthetumourandthesensitiveadjacentnormaltissuestructures,theoptimumsafetymarginneededbetween gross tumour and adjacent tissues has still notbeenestablished.Untilthisdosedifferentialinformation is available, the clinical target volume of theprimarytumourshouldbedenedcautiouslyinplanning.Randomisedprospectivetrialsshouldenablethe clinical target volume to be more accurately dened. Intensity-modulatedradiotherapyhasachievedexcellentlocoregionalcontrolofnasopharyngealcarcinomas.83A study that prospectively assessed salivaryfunctionsconrmedthegradualrecoveryofparotidfunctionwithin2yearsaftercompletionofintensity-modulatedradiotherapy.84Satisfactorydosimetricresultswerealsoachievedwiththistreatmentforrecurrentnasopharyngeal carcinomas, and the degree of short-termcontrol was encouraging.85 Other attempts to enhance thebiologicaleffectsofradiotherapyhavebeenreported.Theseattemptsincludeacceleratedfractionation,86acceleratedhyperfractionation,87andacombinationofoneorotherofthesetreatmentswithchemotherapy.88,89However,hyperfractionationradiotherapyfornasopharyngealcarcinomashouldbeusedwithcare,becauseastudyofacceleratedhyperfractionationby2DradiotherapyplanninghasreportedanincreaseinradiationdamagetotheCNSwithoutimprovementintumour control.90ChemotherapySeveral studies in the past two decades have reported theresults of the use of chemotherapy in combination withradiotherapyforthemanagementoflocoregionaladvancedcasesofnasopharyngealcarcinoma.Twelverandomisedcontrolledtrialshavereportedonneoadjuvant,concurrent,andadjuvanttherapy,oronacombination of these approaches. Nine of these studieswerereportedbefore2004andincludedfourneo-adjuvantchemotherapystudies,9194 threeconcurrentchemotherapy studies,66,67,95and two adjuvant studies.96,97Oneoftheconcurrentstudies95hasrecentlybeenupdated,98andtwooftheneoadjuvantstudies92,93havebeenupdatedandpooledformeta-analysis.99ThreemoreconcurrentchemotherapystudieshavebeenreportedfromHongKongandSingapore.100102 Resultsdifferedbetweenstudiesthatusedneoadjuvant,thosethatusedconcurrent,andthosethatusedadjuvantchemotherapyincombinationwithradiotherapy.Inadditiontothedifferenceinchemotherapyschedules,the effect of staging classication and of stage migrationforstudiesreportedatdifferenttimescouldexplainthese reported differences in results (table 2).The Intergroup 1997 study66was the rst study to showthatuseofchemotherapyalongsideradiotherapyimprovedoverallsurvivalcomparedwithradiotherapyalone.Becausethisstudyincludedcasesofwell-differentiated carcinomas, there were initial doubts as towhethertheresultswereapplicabletonasopharyngealcarcinomainendemicareas.However,asubsequentreport from Taiwan67lent support to the benets of thisapproach.Infact,thesestudiesweretheonlytwotoshow an improvement in both relapse-free survival andoverall survival. As far as the other concurrent studies were concerned,one study102reported an improvement in overall survival,and another study100reported a borderline improvementinoverallsurvival.However,neitherstudyshowedanyevidenceofimprovedrelapse-freesurvival,andthedisparitybetweenoverallsurvivalandrelapse-freesurvivalwasapparentlyexplainedbyimprovementincontrol of distant metastases in the absence of improvedlocoregionalcontrol.Theupdatedconcurrentstudy98also reported an improvement in overall survival but noimprovementinrelapse-freesurvival.Afurtherstudy101reportedanimprovementinactuarialloco-regionalcontrolat3 years,butnoimprovementinrelapse-freesurvival or overall survival, and a signicant increase inototoxicity in the treatment group. The long-term follow-up reports from these more recent studies are expectedto provide more denitive results than at present.Twoofthefourneoadjuvantstudies91,93reportedimprovementinrelapse-freesurvivalbutnoimprovement in overall survival. The others92,94reportednoimprovementgenerally.Ameta-analysis99notedimprovementsinrelapse-freesurvivalanddisease-specicsurvival.However,overallsurvivalwasnot2046 www.thelancet.com Vol 365 June 11, 2005 Seminarimproved because of the increase in intercurrent deathsin the treatment group. Promising results have also beenreportedfromphaseIIstudiesofadvancednasopharyngealcarcinomatreatedwithalternatingweeklychemotherapywithcisplatinand5-uorouracil/folinicacid.103Thetwoadjuvantchemotherapystudies96,97reportednoimprovementeither in relapse-free survival or overall survival.Studiestryingtoimproveontheuseofconcurrentchemotherapyplusadjuvantchemotherapyhavebeenreportedwithobjectivesvaryingfromimprovementoftolerance and side-effects to improvement of efcacy forthemoreadvancedcases.Thepoorcompliancewithadjuvantchemotherapyafterconcurrentchemo-radiotherapy can be overcome by the use of neoadjuvantchemotherapy.Astudyonneoadjuvantchemotherapyfollowedbyconcomitantchemoradiotherapyhasreportedexcellentoverallsurvivalandacceptabletoxicity.104Replacementofcisplatinwithotherchemotherapeutic agents in part of the treatment couldovercomeorreducetheototoxicityassociatedwithsixcourses of cisplatin. A study using cisplatin concurrentlywithradiotherapy,followedbyadjuvantifosfamide, 5-uorouracil,andleucovorin,forpatientswithstageIVbnasopharyngealcarcinomahasbeenreported.105Althoughthepatientsconcernedhaddiseaseatamoreadvancedstage,theoutcomesofthisgroupwerecomparablewiththosereportedinotherseriesofpatients with less advanced disease for whom platinum-basedadjuvantchemotherapywasused.Thechemotherapy regime also has an acceptable compliancerate.AlthoughthestageIandIIcasesweregenerallyconsidered to have resulted in relatively good treatmentoutcomes,analysishasshownthattheAmericanJointCommitteeonCancerScreening1997systemhasallowedmorepatientswithapoorprognosistobegrouped under stage II.106With such stage migration ofthe more advanced cases to stage II, and early evidencethatdisease-freesurvivalismuchthesameforstageIIpatientswitharaisedtumourburdenaftertreatmentwithconcurrentchemoradiotherapyandforstageIpatientstreatedwithradiotherapyalone,theroleofconcurrentchemoradiotherapyshouldbeexploredforstageIIpatients.107Aninternationaleffortisnowunderwaytoundertakeameta-analysisofupdateddatafrommanyofthesereportedrandomisedcontrolledtrials, with more than 1700 patients. The results of thismeta-analysis are eagerly awaited.ThereisnowgeneralagreementthatthepositiveresultsreportedintheIntergroup1997study66areapplicabletonasopharyngealcarcinomainendemicareas,buttheconictingevidenceofchemotherapyonlocalcontrolanddistantmetastaseshavegenerateddiscussion. The conclusion seems to be that of the threebasicapproachestestedinthesestudies(neoadjuvant,concurrent,andadjuvantchemotherapy),concurrentchemoradiotherapy is the most efcacious. Nevertheless,theclassicprinciplesofchemoradiotherapytiming(namely,thatconcurrentchemoradiotherapyprovidesmoreeffectivelocalcontrol,whereassequentialuseofchemotherapyandradiotherapyismoreeffectivewithdistant metastases) have not been borne out by the studyresults.108Despitetheuseofconcurrentchemoradiotherapy,distantmetastasesremainthemajorcauseoftreatmentfailure,32andtheoutlookforstage IV patients remains poor.109Follow-upClinicalDocumentationofcompleteremissioninthenasopharynxandnecklymphatics,withtheapplicationof clinical examination, endoscopic examination with orwithoutbiopsy,andimagingstudies,isimportant.Forassessmentofcompleteremissioninthenasopharynx,the decision about where to draw the line between a slowregressingtumourandaresidualtumourremainswww.thelancet.com Vol 365 June 11, 2005 2047Number ofTiming of chemotherapy % disease-free% overall patients survival (years) survival (years)RT RT/CT RT RT/CTInstitute Nationale229 Adjuvant: vincristine, doxorubicin,56 (4) 58 (4) 67 (4) 59 (4)Tumori, Italy94cyclophosphamide 6 cyclesInternational NPC339 Neoadjuvant: bleomycin,30 (5)* 39 (5)* 46 (5) 40 (5)Study Group89epirubicin, cisplatin 3 cyclesIntergroup 009964147 Concurrent: cisplatin 3 cycles, then 24 (3)* 69 (3)* 47 (3)* 78 (3)*adjuvant: cisplatin, 5-FU for 3 cyclesAsian Oceanian Clinical334 Neoadjuvant: epirubicin,42 (3) 48 (3) 71 (3) 78 (3)Oncology Association90cisplatin 23 cyclesSun Yat-sen University,456 Neoadjuvant: bleomycin,49 (5)* 59 (5)* 56 (5) 63 (5)Guangzhou91cisplatin, 5-FU 23 cyclesSun Yat-sen University,784 Neoadjuvant: bleomycin, cisplatin,43 (5)* 51 (5)* 58 (5)* 64 (5)*Guangzhou + Asian5-FU 23 cyclesOceanian ClinicalOROncology AssociationNeoadjuvant: epirubicin, (pooled updated)97cisplatin 23 cyclesPrince of Wales Hospital,350 Concurrent: cisplatin 69 (2) 76 (2)Hong Kong 938 cycles weeklyPrince of Wales Hospital,350 Concurrent: cisplatin52 (5) 62 (5) 59 (5)* 72 (5)*Hong Kong (updated)968 cycles weeklySapporo Medical80 Neoadjuvant: cisplatin,43 (5) 55 (5) 48 (5) 60 (5)University, Japan925-FU 2 cyclesNational Yang-Ming157 Adjuvant: cisplatin,50 (5) 54 (5) 61 (5) 55 (5)University, Taiwan955-FU 9 cycles weeklyTaichung Veterans284 Concurrent: cisplatin,53 (5)* 72 (5)* 54 (5)* 72 (5)*General Hospital, Taiwan1015-FU for 2 cyclesQueen Mary Hospital,219 Concurrent: uracil, then adjuvant:58 (3) 69 (3) 77 (3) 87 (3)Hong Kong98cisplatin, 5-FU alternating with vincristine, bleomycin, methotrexate 6 cycles overallNational Cancer Centre,220 Concurrent: cisplatin 3 cycles,62 (2) 76 (2) 77 (2)*85 (2)*Singapore100then adjuvant cisplatin, 5-FU for 3 cyclesHong Kong NPC348 Concurrent: cisplatin 3 cycles,61 (3) 69 (3) 79 (3) 78 (3)Study Group99then adjuvant: cisplatin, 5-FU for 3 cycles* p005; disease-specic survival; p=006; RT=radiotherapy alone; RT/CT=combined radiotherapy and chemotherapy arm; 5FU=5-uorouracil.Table 2: Results of randomised prospective studies on application of chemotherapy and radiotherapy inthe management of nasopharyngeal carcinoma Seminarproblematic, but in most cases salvage treatment shouldnotbedelayedforlongerthanabout10weeks.110Residualtumoursinthenasopharynxcanbetreatedwitheitherconedownelds111orbrachytherapy112withgoodresults,andresidualnecknodediseaseisamenable to radical neck dissection. Clinicalandimagingfollow-upofpatientsisrecommended because locoregional relapses, if detectedearly,areamenabletoradicalsalvagetreatment.113Therecommendedfollow-upproceduresincludeclinicalexaminationofthenasopharynx(includinganendoscopic examination) and neck, and regular imagingevery46monthsduringtheinitial35yearsaftertreatment.114,115Endoscopicexaminationshouldbeusedto detect supercial tumours, and cross-section imagingshouldbeusedtodetectdeepinltratingtumoursnotassociatedwithmucosallesion.31AnimagingstudycomparingPETwithMRIfordetectionofresidualandrecurrenttumourhasreportedPETasthesuperiormodality.46Forthedetectionofdistantmetastases,theuseofserumEBVDNAhasbeenshowntobemoresensitive and reliable than other options.116EBV geneCirculating free EBV DNA has been reported in patientswithnasopharyngealcarcinoma,117andtheincreasednumberofcopiesofEBVDNAinthebloodduringtheinitialphaseofradiotherapysuggeststhattheviralDNAwasreleasedintothecirculationaftercelldeath.118ThequantityoffreeplasmaEBVDNA,asmeasuredbyreal-timequantitativePCR,isrelatedtothestageofthedisease.ThequantitiesofEBVDNAcopiesbeforeandaftertreatmentaresignicantlyrelatedtotheratesofoverallanddisease-freesurvival.119Astudyhasreportedthat the levels of post-treatment EBV DNA compared withpretreatmentEBVDNAareagoodpredictorofprogression-freesurvival.120WhenEBVDNAwasusedtogether with immunoglobulin A against the viral capsidantigenofEBV,thesensitivityofearlydiagnosisofnasopharyngealcarcinomaincreased.121RaisedlevelsofEBVDNAwereonlydetectedin67%ofpatientswithlocoregional recurrence,116,122 although in those with distantmetastasislevelsofEBVDNAcopieswereheightenedbefore the appearance of clinical abnormality.116Sequelae of therapySurvivorsofnasopharyngealcarcinomahaveimpairedhealth-related quality of life.123,124Patients who survive thediseasecanhaveseverallatecomplications,manyofwhichresultfromtheeffectsofradiationonthedose-limitingorgansadjacenttothenasopharynxandnecknodes. The use of chemotherapy in more advanced casesaddstotheside-effects,whichincludeototoxicityassociatedwithcisplatin.100Asmallproportionofthelong-termsequelaerepresenttheeffectsofunhealedresidual damage by the tumour, such as residual cranialnervepalsiesandserousotitismediaresultingfrompersistentdisturbanceoftheeustachian-tubefunction.Thesesequelaeincludeneuro-endocrine125andauditory126complications,drymouth,poororalanddentalhygiene,127,128radiation-inducedsofttissuebrosis,129andcarotidarterystenosis.130Themostdebilitatingsequelaeareneurologicalcomplications.Thesecanincludeseriousdisorderssuchastemporallobe necrosis,131cranial nerve palsies132and dysphagia,133andalsolessobviouseffectssuchasmemoryloss,134cognitivedysfunction,135andneuropsychologicaldysfunction136(table 3).Aseriesofcasesinwhichhypofractionatedradiotherapy was used in combination with 2D planningproduced a 60% actuarial risk of complication and a 28%riskofneurologicalcomplications.137Cuttingdownlatecomplicationsoftreatmentshouldbeoneofthemainobjectivesoffutureclinicaltrials.Shieldingofthepituitary-hypothalamicaxisin2Dplanningandtreatmenthasbeenshowntosignicantlyreduceneuroendocrinecomplications.138Useofintensity-modulatedradiotherapyhasbeenshowntoimprovesalivaryfunction,84butotherbenetsneedalongerfollow-up period to conrm. 2048 www.thelancet.com Vol 365 June 11, 2005 Time of assessment Risk and details of sequelaeSide-effects associated with hypofractionated10 years 31% of patients developed one or more late irradiation sequelae. Most were mild soft-tissue damages. Neurological damage thatradiotherapy135occurred in 10% of patients constituted the major morbidity and accounted for most of the treatment mortalities (1%)Hypothalamic-pituitary function1235 years With life table analysis, the cumulative probability of endocrine dysfunction was estimated to be 62% after 5 years. These includedisturbances of growth hormone, gonadotropins, corticotropin, and/or thyrotropin. The progressive impairment in hypothalamic pituitary function leading to endocrine dysfunction that requires treatment occurs in 50% of patients at 5 years after cranial irradiationLong-term sensorineural hearing decit1242 years Within 3 months after radiotherapy, deterioration of bone conduction threshold at 4 kHz was noted in 31% of ears. In 40% of these ears, recovery was evident at 2 yearsDental and oral hygiene12714 years Xerostomia (92%), trismus (29%), higher prevalence of clinical candidosis (24%)Carotid artery stenosis1285 years15 times increase in risk of developing signicant carotid stenosis Temporal lobe necrosis1295 years 5-year actuarial incidence ranged from 0% to 14% (dependent on fractional and total dose)Cranial nerve palsy130120 years Hypoglossal palsy, vagus palsy, recurrent laryngeal nerve palsy, and accessory palsies. Often associated with marked neck brosisDysphagia131Not stated Swallowing function continues to deteriorate over time, even many years after radiation therapyMemory loss1322 years Visual memory performance deteriorated with time, while verbal memory remained more stableCognitive function associated with temporal1 year For patients who developed temporal lobe necrosis after radiotherapy, memory, language, motor ability, and executive functionslobe necrosis133were signicantly impaired, although their general intelligence remained relatively intactTable 3: Late complications of radiotherapy for the treatment of nasopharyngeal carcinomaSeminarManagement of residual or recurrent diseaseDespite the effectiveness of radiation and chemotherapyin the management of nasopharyngeal carcinoma, localfailureorregionalfailurepresentingaspersistentorrecurrenttumourstilloccurs.Toattainahighsalvagerate,earlydetectionandtreatmentisessential.18FDG-PETisbetterthanCTindetectingresidualorrecurrentdiseaseinthenasopharynx,139anditsresultscanusuallybeconrmedwithbiopsythroughendoscopicexamination.Residualorrecurrenttumourintheneckafterradiotherapyisnotoriouslydifculttoconrmbecauseinsomelymphnodesonlyclustersoftumourcellsarepresent.140Aggressivetreatmentforlocallyrecurrentnasopharyngealcarcinomaiswarranted,especiallyincaseswherethediseaseisconned to the nasopharynx. Survival after retreatmentformoreextensivediseaseremainspoor,butisstillhigherthaninpatientsreceivingsupportivetreatmentonly.113Even for patients with synchronous locoregionalfailures,aggressivetreatmentshouldbeconsideredforselected patients141(table 4).Disease in the neck Aftercombinedchemoradiationfornasopharyngealcarcinoma, isolated failure in the neck is less than 5%.146If cancer persists or recurs in the cervical lymph nodes,as evidenced by imaging studies or clinical progressionofthelymphnodes,salvagetherapyisneeded.Whenmanagedwithanothercourseofexternalradiotherapy,the overall 5-year survival rate is around 20%.147Radicalneckdissectionasaformofsurgicalsalvagehasachieved a 5-year tumour control rate of 66% in the neckand a 5-year actuarial survival of 38%.142When tumour inthe neck node extends beyond the connes of the lymphnode,brachytherapyshouldbeappliedtothetumourbedinadditiontoradicalneckdissection.Withthisadjuvant therapy, a similar tumour control rate has beenachieved as for radical neck dissection for less extensiveneck disease.148Disease in the nasopharynxResidual or recurrent disease in the nasopharynx can bemanaged with a second course of external radiotherapy.Thedosageshouldbegreaterthantheinitialradiationdose. Although a salvage rate of 32% has been achieved,thecumulativeincidenceoflatesequelaeafterre-irradiationis24%withtreatmentmortalityof18%.149Toavoidthehighincidenceofcomplicationsresultingfromre-irradiation,stereotacticradiotherapy,brachytherapy, and surgical resection have been used forpatientswithsmalllocalisedtumoursinthenasopharynx.Stereotacticradiotherapy,whenusedforthemanagementofresidualorrecurrenttumour,isassociatedwitha2-yearlocaltumourcontrolrateof72%.77However,onlyafewpatientshavebeentreatedwith this method, and long-term follow-up informationis not available.150BrachytherapyWith brachytherapy, the radiation dose decreases rapidlyfromtheradiationsource,enablingahighdoseofirradiationtobedeliveredtotheresidualorrecurrenttumour in the nasopharynx but a much smaller dose tothesurroundingtissue.Brachytherapyalsodeliversradiation at a continuous low dose rate, which gives it afurtherradiobiologicaladvantageoverfractionatedexternal radiation. Intracavitary brachytherapy has beenusedfornasopharyngealcarcinomas.151Theradiationsourcewasplacedeitherinatubeoramouldbeforeinsertion into the nasopharynx. In view of the irregularcontourofthenasopharynx,accurateapplicationoftheradiationsourcetoprovideatumoricidaldoseisdifcult.Tocircumventthisproblem,radioactiveinterstitialimplantshavebeenusedtotreatsmalllocalisedresidualorrecurrenttumoursinthenasopharynx.152Radioactive gold grains (198Au) are the most frequentlyusedradiationsourceforthispurpose.Goldgrainscanbeimplantedeithertransnasallyorwiththesplit-palateapproach.143The split-palate approach gives the surgeon adirect view of the tumour site and enables him or her toimplantthegoldgrainspermanentlyintothetumourwithgreatprecision.Fortumourslocalisedinthenasopharynx,withoutboneinvasion,thismethodhasprovidedeffectivesalvagewithminimummorbidity.153Where gold grain implants were used to treat persistentand recurrent tumours after radiotherapy, the 5-year localtumourcontrolrateswere87%and63%,respectively,andthecorresponding5-yeardisease-freesurvivalrateswere68%and60%.154Otherstudiesusingintracavitarybrachytherapy have also reported success.155,156NasopharyngectomyIf the residual or recurrent tumour in the nasopharynx istooextensiveforbrachytherapyorhasextendedtotheparanasopharyngealspace,nasopharyngectomycanachievesalvageinselectedpatientswithlocalisedwww.thelancet.com Vol 365 June 11, 2005 2049Number ofStage ofLocal control/survival rate Complications Treatment- patients recurrence related mortality Reirradiation706 T1 to T3 5-year survival rate 14%;Late sequelae18%(conventional)14210-year survival rate 9% 24%Stereotactic18 T1 and T2 2-year local control rate 72% 56% 0radiosurgery75Re-irradiation (IMRT)7749 T1 to T4 9-month local control 100% 0 0Brachytherapy 106 T1 5-year local control, residual19% 0(Gold grain)14387%, recurrent 62%; 5-year disease-free survival, residual 68%, recurrent 60%Nasopharyngectomy 109 T1 5-year local control rate 68%,25% 0(Maxillary swing)1445-year disease-free survival 54%Nasopharyngectomy 37 T1 to T3 5-year local control rate 67%,54% 3%(transpalatal)1455-year disease-free survival 52%IMRT= intensity-modulated radiation therapy.Table 4: Results of different types of salvage therapy for residual or recurrent tumour in thenasopharynx after radical external radiotherapySeminardisease.Becauseoftheawkwardpositionofthenasopharynxinthemiddleofthehead,exposureforoncologicalextirpationofthetumourhaspresentedadifculttechnicalchallenge.Variousapproacheshavebeenreported,includinganinfratemporalapproachfromthelateralaspect;157transpalatal,transmaxillary,andtranscervicalapproachesfromtheinferioraspect;145,158 andanantereolateralapproach.144Themortalitiesassociatedwiththesesalvagesurgicalprocedureshavebeenlow,andsinceallthepatientsconcernedhadpreviouslyundergoneradicalradiotherapy,theassociatedmorbiditiesinsomepatients,suchastrismusandpalatalstula,wereacceptable. As long as the residual or recurrent tumourcanberemovedadequately,thelong-termresultshavebeensatisfactory.The5-yearactuarialcontroloftumours in the nasopharynx is about 65% and the 5-yeardisease-free survival rate is around 54%.159,160External radiotherapyFormoreadvancedorinltrativetumours,asecondcourseofexternalradiotherapyisneeded.161Asecondcourse of external radiotherapy given concurrently withchemotherapy has been tried; this approach was built ontheexperiencegainedfromtheuseofconcurrentchemoradiotherapyinprimarytreatment.Thistreatmenthasbeenreportedtogivea5-yearactuarialoverallsurvivalrateof26%,althoughtheriskofmajorlatetoxicitieswassignicant.162Theuseofprecisionradiotherapysuchasintensity-modulatedradiotherapycouldimprovethetherapeuticratioforlocalcontrol;promisinginitialresultshavebeenreported,85butdistant metastases will remain a major issue for patientswith local relapse.Distant metastasisCisplatin-basedcombinationchemotherapyisthemosteffectivetreatmentformetastaticnasopharyngealcarcinoma.Cisplatinandinfusional5-uorouracilhasbecome the standard treatment with a 6676% responserate.163Several phase II studies of the newer agents havebeenreported.164167Moreintensivecombinationsgiveahigher response rate, but are also usually associated withincreased toxicities.168170None of these combinations hasyet been compared with the combination of cisplatin andinfusional 5-uorouracil.Treatmentofmetastaticnasopharyngealcarcinomas,mainlywithchemotherapy,isessentiallypalliative,althoughlong-termdisease-freesurvivorshavebeenreported.171Forselectedpatientswithfewmetastases,additionallocoregionaltreatmentcangiveextendeddisease control. Resection of lung metastases can resultin longer control for patients in whom the spread of thecarcinoma to the lung has been limited.172In cases wherethere has been little spread to the mediastinal nodes, theadditionofradiotherapytochemotherapycouldalsoresult in protracted tumour control.173Recent developmentsInadditiontothenoveltreatmentapproachesthataregenerallyapplicabletocancersatothersites,thecloseassociation between EBV and nasopharyngeal carcinomagivesfurtheropportunitiesfornoveltreatment.StrategiestargetedatEBVincludegenetherapyandimmunetherapy,andtheproof-of-principlesstudieshavebeendoneinlaboratories.Genetherapywithanovelreplication-decientadenovirusvectorinwhichtransgeneexpressionisunderthetranscriptionalregulation of oriP of EBV has been reported.174Immunetherapyapproacheshaveincludedtherapeuticaugmentation of cytotoxic T-lymphocyte responses175andadoptivetransferofautologousEBV-speciccytotoxicT-cells.176FuturestudiesontherolesofthevirusintransformationandfunctionsofEBVlatentproteinscould help to identify other novel treatment targets.177Conict of interest statementWe declare that we have no conict of interest.References1 Sham JS, Choy D, Wei WI, et al. Detection of subclinicalnasopharyngeal carcinoma by breoptic endoscopy and multiplebiopsy. Lancet 1990; 335: 37174.2 Jackson C. Primary carcinoma of the nasopharynx: a table of cases.JAMA 1901; 37: 37177.3 New GB. Syndrome of malignant tumors of the nasopharynx,a report of seventy-nine cases. JAMA 1992; 79: 1014.4 Digby KH, Fook WL, Che YT. Nasopharyngeal carcinoma.Br J Surg 1941; 28: 51737.5 Parkin DM, Whelan SL, Ferlay J, Raymond L, Young J, eds. Cancerincidence in ve continents, vol 7. IARC 1997; 143: 81415.6 Nielsen NH, Mikkelsen F, Hansen JP. Nasopharyngeal cancer inGreenland: the incidence in an Arctic Eskimo population.Acta Pathol Microbiol Scand 1977; 85: 85058.7 Dickson RI, Flores AD. Nasopharyngeal carcinoma: an evaluationof 134 patients treated between 19711980. Laryngoscope 1985; 95:27683.8 Buell P. The effect of migration on the risk of nasopharyngealcancer among Chinese. Cancer Res 1974; 34: 118991.9 Godtfredsen E. On the histopathology of malignantnasopharyngeal tumors. Acta Pathol Microbiol Scand 1944;55 (suppl): 38319.10 Svoboda D, Kirchner F, Shanmugaratnam K. Ultrastructure ofnasopharyngeal carcinoma in American and Chinese patients: anapplication of electron microscopy to geographic pathology.Exp Mol Pathol 1965; 4: 189204.11 Prasad U. Cells of origin of nasopharyngeal carcinoma: an electronmicroscopical study. J Laryngol Otol 1974; 88: 1087.12 Gulley ML. Molecular diagnosis of Epstein-Barr virus-relateddiseases. J Mol Diagn 2001; 3: 110.13 Raab-Traub N. Epstein-Barr virus in the pathogenesis of NPC.Semin Cancer Biol 2002; 12: 43141. 14 Young LS, Murray PG. Epstein-Barr virus and oncogenesis: fromlatent genes to tumours. Oncogene 2003; 22: 510821.15 Shanmugaratnam K, Sobin LH. Histological typing of tumours ofthe upper respiratory tract and ear. In: Shanmugaratnam K,Sobin LH, eds. International histological classication of tumours:no 19. Geneva: WHO, 1991: 3233.16 Michaeu C, Rilke F, Pilotti S. Proposal for a new histopathologicalclassication of the carcinomas of the nasopharynx. Tumori 1978;64: 51318.17 Nicholls JM. Nasopharyngeal carcinoma: classication andhistological appearances. Adv Anat Path 1997; 4: 7184.18 Shanmugaratnam K, Sobin LH. Histological typing of tumors ofupper respiratory tract and ear. In: Shanmugaratnam K , Sobin LH,eds. International histological classication of tumours, 2nd edn.Geneva: WHO, 1991: 3233.2050 www.thelancet.com Vol 365 June 11, 2005 Seminar19 Reddy SP, Raslan WF, Gooneratne S, Kathuria S, Marks JE.Prognostic signicance of keratinization in nasopharygealcarcinoma. Am J Otolaryngol 1995; 16: 10308.20 Marks JE, Philips JL, Menck HR. The National Cancer Data Basereport on the relationship of race and national origin to thehistology of nasopharyngeal carcinoma. Cancer 1998: 83: 58288.21 Lee AWM, Foo W, Law SCK, et al. Nasopharyngeal carcinomapresenting symptoms and duration before diagnosis. HK Med J1997; 3: 35561.22 Ozyar E, Atahan IL, Akyol FH, Gurkaynak M, Zorlu AF. Cranialnerve involvement in nasopharyngeal carcinoma: its prognosticrole and response to radiotherapy. Radiat Med 1994; 12: 6568. 23 Sham JS, Poon YF, Wei WI, Choy D. Nasopharyngeal carcinoma inyoung patients. Cancer 1990; 65: 260610. 24 Zeng Y, Zhang LG, Wu YC, et al. Prospective studies onnasopharyngeal carcinoma in Epstein-Barr virus IgA/VCA antibody-positive persons in Wuzhou City, China. Int J Cancer 1985; 36: 54547. 25 Zong YS, Sham JS, Ng MH, et al. Immunoglobulin A against viralcapsid antigen of Epstein-Barr virus and indirect mirrorexamination of the nasopharynx in the detection of asymptomaticnasopharyngeal carcinoma. Cancer 1992; 69: 37. 26 Cheng WM, Chan KH, Chen HL, et al. Assessing the risk ofnasopharyngeal carcinoma on the basis of EBV antibody spectrum.Int J Cancer 2002; 97: 48992. 27 Chien YC, Chen JY, Liu MY, et al. Serologic markers of Epstein-Barr virus infection and nasopharyngeal carcinoma in Taiwanesemen. N Engl J Med 2001; 345: 187782. 28 Sham JS, Cheung YK, Choy D, Chan FL, Leong L. Nasopharyngealcarcinoma: CT evaluation of patterns of tumor spread.Am J Neuroradiol 1991; 12: 26570. 29 Chong VF, Fan YF, Khoo JB. Nasopharyngeal carcinoma withintracranial spread: CT and MR characteristics.J Comput Assist Tomogr 1996; 20: 56369. 30 Dillon WP, Mills CM, Kjos B, DeGroot J, Brant-Zawadzki M.Magnetic resonance imaging of the nasopharynx. Radiology 1984;152: 73138. 31 Olmi P, Fallai C, Colagrande S, Giannardi G. Staging and follow-up of nasopharyngeal carcinoma: magnetic resonance imagingversus computerized tomography. Int J Radiat Oncol Biol Phys1995; 32: 795800. 32 Cheng SH, Jian JJ, Tsai SY, et al. Prognostic features and treatmentoutcome in locoregionally advanced nasopharyngeal carcinomafollowing concurrent chemotherapy and radiotherapy.Int J Radiat Oncol Biol Phys 1998; 41: 75562. 33 Sham JS, Tong CM, Choy D, Yeung DW. Role of bone scanning indetection of subclinical bone metastasis in nasopharyngealcarcinoma. Clin Nucl Med 1991; 16: 2729. 34 Kraiphibul P, Atichartakarn V, Clongsusuek P, Kulapaditharom B,Ratanatharathorn V, Chokewattanaskul P. Nasopharyngealcarcinoma: value of bone and liver scintigraphy in the pre-treatment and follow-up period. J Med Assoc Thai 1991; 74: 27679. 35 Leung SF, Metreweli C, Tsao SY, Van Hasselt CA. Stagingabdominal ultrasonography in nasopharyngeal carcinoma.Australas Radiol 1991; 35: 3132. 36 Sham JS, Chan LC, Loke SL, Choy D. Nasopharyngeal carcinoma:role of marrow biopsy at diagnosis. Oncology 1991; 48: 48082. 37 Kumar MB, Lu JJ, Loh KS, et al. Tailoring distant metastaticimaging for patients with clinically localized undifferentiatednasopharyngeal carcinoma. Int J Radiat Oncol Biol Phys 2004; 58:68893. 38 Nakamoto Y, Osman M, Wahl RL. Prevalence and patterns of bonemetastases detected with positron emission tomography usingF-18 FDG. Clin Nucl Med 2003; 28: 30207.39 Cellai E, Olmi P, Chiavacci A, et al. Computed tomography innasopharyngeal carcinomapart II: impact on survival.Int J Radiat Oncol Biol Phys 1990; 19: 117782. 40 Emami B, Sethi A, Petruzzelli GJ. Inuence of MRI on targetvolume delineation and IMRT planning in nasopharyngealcarcinoma. Int J Radiat Oncol Biol Phys 2003; 57: 48188. 41 Chong VF, Fan YF. Detection of recurrent nasopharyngealcarcinoma: MR imaging versus CT. Radiology 1997; 202: 46370. 42 Chong VF, Mukherji SK, Ng SH, et al. Nasopharyngeal carcinoma:review of how imaging affects staging. J Comput Assist Tomogr1999; 23: 98493. 43 Ng SH, Chang JT, Ko SF, Wan YL, Tang LM, Chen WC. MRI inrecurrent nasopharyngeal carcinoma. Neuroradiology 1999; 41:85562. 44 Fang FM, Leung SW, Wang CJ, et al. Computed tomographyndings of bony regeneration after radiotherapy for nasopharyngealcarcinoma with skull base destruction: implications for localcontrol. Int J Radiat Oncol Biol Phys 1999; 44: 30509. 45 Lu TX, Mai WY, Teh BS, et al. Important prognostic factors inpatients with skull base erosion from nasopharyngeal carcinomaafter radiotherapy. Int J Radiat Oncol Biol Phys 2001; 51: 58998. 46 Yen RF, Hung RL, Pan MH, et al. 18-uoro-2-deoxyglucosepositron emission tomography in detecting residual/recurrentnasopharyngeal carcinomas and comparison with magneticresonance imaging. Cancer 2003; 98: 28387. 47 Sobin LH, Wittekind, eds. TNM classication of malignanttumours, 5th edn. New York: Wiley-Liss, 1997: 2530.48 Ho JHC. An epidemiologic and clinical study of nasopharyngealcarcinoma. Int J Radiat Oncol Biol Phys 1978; 4: 18298.49 Fleming ID, Cooper JS, Henson DE, et al, eds. AJCC CancerStaging Manual, 5th edn. Philadelphia: Lippincott-Raven, 1997:3335.50 Sham JST, Cheung YK, Choy D, Chan FL, Leong L. Cranial nerveinvolvement and base of the skull erosion in nasopharyngealcarcinoma. Cancer 1991; 68: 42226.51 Chua DTT, Sham JST, Kwong DLW, Choy D, Au GKH, Wu PM.Prognostic value of paranasopharyngeal extension ofnasopharyngeal carcinoma. Cancer 1996; 78: 20210.52 Teo P, Yu P, Lee WY, et al. Signicant prognosticator after primaryradiotherapy in 903 nondisseminated nasopharyngeal carcinomaevaluated by computer tomography. Int J Radiat Oncol Biol Phys1996; 36: 291304. 53 Ho JH. Stage classication of nasopharyngeal carcinoma: a review.International Agency for Research on Cancer, publication no. 20,1978; 99113.54 Lee AW, Foo W, Law SC, et al. Staging of nasopharyngealcarcinoma: from Hos to the new UICC system. Int J Cancer 1999;84: 17987.55 Cooper JS, Cohen R, Stevens RE. A comparision of staging systemsfor nasopharyngeal carcinoma. Cancer 1998; 83: 21319.56 zyar E, Yildiz F, Akyol FH, Atahan II. Comparison of AJCC 1988and 1997 classications for nasopharyngeal carcinoma. Int J RadiatOncol Biol Phys 1999; 44: 107987. 57 Sham JS, Choy D. Prognostic factors of nasopharyngeal carcinoma:a review of 759 patients. Br J Radiol 1990; 63: 5158. 58 Perez CA, Devineni VR, Marcial-Vega V, Marks JE, Simpson JR,Kucik N. Carcinoma of the nasopharynx: factors affectingprognosis. Int J Radiat Oncol Biol Phys 1992; 23: 27180. 59 Sham JS, Choy D. Prognostic value of paranasopharyngealextension of nasopharyngeal carcinoma on local control and short-term survival. Head Neck 1991; 13: 298310. 60 Cheng SH, Yen KL, Jian JJ, et al. Examining prognostic factors andpatterns of failure in nasopharyngeal carcinoma followingconcomitant radiotherapy and chemotherapy: impact on futureclinical trials. Int J Radiat Oncol Biol Phys 2001; 50: 71726. 61 Chua DT, Sham JS, Kwong DL, et al. Volumetric analysis of tumorextent in nasopharyngeal carcinoma and correlation with treatmentoutcome. Int J Radiat Oncol Biol Phys 1997; 39: 71119. 62 Chang CC, Chen MK, Liu MT, Wu HK. The effect of primarytumor volumes in advanced T-staged nasopharyngeal tumors.Head Neck 2002; 24: 94046. 63 Sze WM, Lee AW, Yau TK, et al. Primary tumor volume ofnasopharyngeal carcinoma: prognostic signicance for localcontrol. Int J Radiat Oncol Biol Phys 2004; 59: 2127. 64 Lo YM. Quantitative analysis of Epstein-Barr virus DNA in plasmaand serum: applications to tumor detection and monitoring.Ann N Y Acad Sci 2001; 945: 6872.65 Chua DT, Sham JS, Wei WI, Ho WK, Au GK. The predictive valueof the 1997 American Joint Committee on Cancer stageclassication in determining failure patterns in nasopharyngealcarcinoma. Cancer 2001; 92: 284555. 66 Al-Sarraf M, Leblanc M, Giri S, et al. Chemoradiotherapy versusradiotherapy in patients with advanced nasopharyngeal cancer:phase III randomized Intergroup Study 0099. J Clin Oncol 1998;16: 131017.www.thelancet.com Vol 365 June 11, 2005 2051Seminar67 Lin JC, Jan JS, Hsu CY, et al. Phase III study of concurrentchemoradiotherapy versus radiotherapy alone for advancednasopharyngeal carcinoma: positive effect on overall andprogression-free survival. J Clin Oncol 2003; 21: 63137.68 Mesic JB, Fletcher GH, Goepfert H. Megavoltage irradiation ofepithelial tumors of the nasopharynx. Int J Radiat Oncol Biol Phys1981; 7: 44753. 69 Hoppe RT, Gofnet DR, Bagshaw MA. Carcinoma of thenasopharynxeighteen years experience with megavoltageradiation therapy. Cancer 1976; 37: 260512. 70 Lee AW, Poon YF, Foo W, et al. Retrospective analysis of 5037patients with nasopharyngeal carcinoma treated during19761985: overall survival and patterns of failure. Int J RadiatOncol Biol Phys 1992; 23: 26170. 71 Wang CC. Improved local control of nasopharyngeal carcinomaafter intracavitary brachytherapy boost. Am J Clin Oncol 1991; 14:58. 72 Kwong DL, Sham JS, Chua DT, Choy DT, Au GK, Wu PM. Theeffect of interruptions and prolonged treatment time inradiotherapy for nasopharyngeal carcinoma. Int J RadiatOncol Biol Phys 1997; 39: 70310. 73 Lee AW, Sham JS, Poon YF, Ho JH. Treatment of stage Inasopharyngeal carcinoma: analysis of the patterns of relapse andthe results of withholding elective neck irradiation. Int J RadiatOncol Biol Phys 1989; 17: 118390. 74 Kwong D, Sham J, Choy D. The effect of loco-regional control ondistant metastatic dissemination in carcinoma of the nasopharynx:an analysis of 1301 patients. Int J Radiat Oncol Biol Phys 1994; 30:102936. 75 Levendag PC, Lagerwaard FJ, de Pan C, et al. High-dose, high-precision treatment options for boosting cancer of thenasopharynx. Radiother Oncol 2002; 63: 6774. 76 Le QT, Tate D, Koong A, et al. Improved local control withstereotactic radiosurgical boost in patients with nasopharyngealcarcinoma. Int J Radiat Oncol Biol Phys 2003; 56: 104654. 77 Chua DT, Sham JS, Kwong PW, Hung KN, Leung LH. Linearaccelerator-based stereotactic radiosurgery for limited, locallypersistent, and recurrent nasopharyngeal carcinoma: efcacy andcomplications. Int J Radiat Oncol Biol Phys 2003; 56: 17783. 78 Waldron J, Tin MM, Keller A, et al. Limitation of conventional twodimensional radiation therapy planning in nasopharyngealcarcinoma. Radiother Oncol 2003; 68: 15361.79 Cheng JC, Chao KS, Low D. Comparison of intensity modulatedradiation therapy (IMRT) treatment techniques fornasopharyngeal carcinoma. Int J Cancer 2001; 96: 12631. 80 Wolden SL, Zelefsky MJ, Hunt MA, et al. Failure of a 3Dconformal boost to improve radiotherapy for nasopharyngealcarcinoma. Int J Radiat Oncol Biol Phys 2001; 49: 122934. 81 Wu VW, Kwong DL, Sham JS. Target dose conformity in3-dimensional conformal radiotherapy and intensity modulatedradiotherapy. Radiother Oncol 2004; 71: 20106. 82 Hsiung CY, Yorke ED, Chui CS, et al. Intensity-modulatedradiotherapy versus conventional three-dimensional conformalradiotherapy for boost or salvage treatment of nasopharyngealcarcinoma. Int J Radiat Oncol Biol Phys 2002; 53: 63847.83 Lee N, Xia P, Quivey JM, et al. Intensity-modulated radiotherapy in the treatment of nasopharyngeal carcinoma: an update of the UCSF experience. Int J Radiat Oncol Biol Phys 2002; 53:1222. 84 Kwong DL, Pow EH, Sham JS, et al. Intensity-modulatedradiotherapy for early-stage nasopharyngeal carcinoma: aprospective study on disease control and preservation of salivaryfunction. Cancer. 2004; 101: 158493. 85 Lu TX, Mai WY, Teh BS, et al. Initial experience using intensity-modulated radiotherapy for recurrent nasopharyngeal carcinoma.Int J Radiat Oncol Biol Phys 2004; 58: 68287. 86 Lee AW, Sze WM, Yau TK, Yeung RM, Chappell R, Fowler JF.Retrospective analysis on treating nasopharyngeal carcinoma with accelerated fractionation (6 fractions per week) in comparisonwith conventional fractionation (5 fractions per week): report on 3-year tumor control and normal tissue toxicity. Radiother Oncol2001; 58: 12130. 87 Franchin G, Vaccher E, Talamini R, et al. Nasopharyngeal cancerWHO type II-III: monoinstitutional retrospective analysis withstandard and accelerated hyperfractionated radiation therapy.Oral Oncol 2002; 38: 13744. 88 Wolden SL, Zelefsky MJ, Kraus DH, et al. Accelerated concomitantboost radiotherapy and chemotherapy for advanced nasopharyngealcarcinoma. J Clin Oncol 2001; 19: 110510.89 Jian JJ, Cheng SH, Tsai SY, et al. Improvement of local control ofT3 and T4 nasopharyngeal carcinoma by hyperfractionatedradiotherapy and concomitant chemotherapy. Int J Radiat OncolBiol Phys 2002; 53: 34452.90 Teo PM, Leung SF, Chan AT, et al. Final report of a randomizedtrial on altered-fractionated radiotherapy in nasopharyngealcarcinoma prematurely terminated by signicant increase inneurologic complications. Int J Radiat Oncol Biol Phys 2000; 48:131122. 91 International Nasopharynx Cancer Study Group VUMCA I trial.Preliminary results of a randomized trial comparing neoadjuvantchemotherapy (cisplatin, epirubicin, bleomycin) plus radiotherapyvs. radiotherapy alone in stage IV(> or = N2, M0) undifferentiatednasopharyngeal carcinoma: a positive effect on progression-freesurvival. Int J Radiat Oncol Biol Phys 1996; 35: 46369.92 Chua DT, Sham JST, Choy D, et al. Preliminary report of theAsian-Oceanian Clinical Oncology Association randomized trialcomparing cisplatin and epirubicin followed by radiotherapyversus radiotherapy alone in the treatment of patients withlocoregionally advanced nasopharyngeal carcinoma. Cancer 1998;83: 227083.93 Ma J, Mai HQ, Hong MH, et al. Results of a prospectiverandomized trial comparing neoadjuvant chemotherapy plusradiotherapy with radiotherapy alone in patients withlocoregionally advanced nasopharyngeal carcinoma. J Clin Oncol2001; 19: 135057.94 Hareyama M, Sakata K, Shirato H, et al. A prospective, randomizedtrial comparing neoadjuvant chemotherapy with radiotherapy alonein patients with advanced nasopharyngeal carcinoma. Cancer 2002;94: 221723.95 Chan ATC, Teo PML, Ngan RK, et al. Concurrent chemotherapy-radiotherapy compared with radiotherapy alone in loco-regionallyadvanced nasopharyngeal carcinoma: progression-free survivalanalysis of a Phase III randomized trial. J Clin Oncol 2002; 20:203844.96 Rossi A, Molinari R, Boracchi P, et al. Adjuvant chemotherapy withvincristine, cyclophosphamide, and doxorubicin after radiotherapyin local-regional nasopharyngeal cancer: Results of a 4-yearmulticenter randomized study. J Clin Oncol 1988; 6: 140110.97 Chi KH, Chang YC, Guo WY, et al. A phase III study of adjuvantchemotherapy in advanced nasopharyngeal carcinoma patients.Int J Radiat Oncol Biol Phys 2002; 52: 123844.98 Chan AT, Ngan R, Teo P, et al. Final results of a phase IIIrandomized study of concurrent weekly cisplatin-RT versus RTalone in locoregionally advanced nasopharyngeal carcinoma (NPC).Proc Am Soc Clin Oncol 2004 (abstract 5523): 492.99 Chua DTT, Ma J, Sham JST. Long-term survival after cisplatin-based induction chemotherapy and radiotherapy fornasopharyngeal carcinoma: a pooled data analysis of two phase IIItrials. Proc Am Soc Clin Oncol 2004 (abstract 5524) 492.100 Kwong DL, Sham JS, Au GK, et al. Concurrent and adjuvantchemotherapy for nasopharyngeal carcinoma: a factorial study.J Clin Oncol 2004; 22: 264353. 101 Lee AWM, Lau WH, Tung SY, et al. Prospective randomized studyon therapeutic gain achieved by addition of chemotherapy forT14N23M0 Nasopharyngeal Carcinoma (NPC). Proc Am Soc Clin Oncol 2004 (abstract 5506): 488.102 Wee J, Tan EH, Tai BC, et al. Phase III randomized trial ofradiotherapy versus concurrent chemo-radiotherapy followed byadjuvant chemotherapy in patients with AJCC/UICC (1997) stage 3and 4 nasopharyngeal cancer of the endemic variety. Proc Am SocClin Oncol 2004 (abstract 5500): 487.103 Lin JC, Jan JS, Hsu CY, Jiang RS, Wang WY. Outpatient weeklyneoadjuvant chemotherapy followed by radiotherapy for advancednasopharyngeal carcinoma: high complete response and lowtoxicity rates. Br J Cancer 2003; 88: 18794.104 Oh JL, Vokes EE, Kies MS, et al. Induction chemotherapy followedby concomitant chemoradiotherapy in the treatment oflocoregionally advanced nasopharyngeal cancer. Ann Oncol 2003;14: 56469. 2052 www.thelancet.com Vol 365 June 11, 2005 Seminar105 Chua DT, Sham JS, Au GK. A concurrent chemoirradiation withcisplatin followed by adjuvant chemotherapy with ifosfamide,5-uorouracil, and leucovorin for stage IV nasopharyngealcarcinoma. Head Neck 2004; 26: 11826. 106 Chua DT, Sham JS, Kwong DL, Au GK. Treatment outcome afterradiotherapy alone for patients with Stage I-II nasopharyngealcarcinoma. Cancer 2003; 98: 7480. 107 Cheng SH, Tsai SY, Yen KL, et al. Concomitant radiotherapy andchemotherapy for early-stage nasopharyngeal carcinoma.J Clin Oncol 2000; 18: 204045. 108 Sanguineti G, Bossi P, Pou A, Licitra L. Timing ofchemoradiotherapy and patient selection for locally advancednasopharyngeal carcinoma. Clin Oncol (R Coll Radiol) 2003; 15:45160. 109 Cheng SH, Jian JJ, Tsai SY, et al. Long-term survival ofnasopharyngeal carcinoma following concomitant radiotherapy andchemotherapy. Int J Radiat Oncol Biol Phys 2000; 48: 132330. 110 Kwong DL, Nicholls J, Wei WI, et al. The time course of histologicremission after treatment of patients with nasopharyngealcarcinoma. Cancer 1999; 85: 144653. 111 Yan JH, Xu GZ, Hu YH, et al. Management of local residualprimary lesion of nasopharyngeal carcinoma: II. Results ofprospective randomized trial on booster dose. Int J Radiat OncolBiol Phys 1990; 18: 29598. 112 Leung TW, Tung SY, Sze WK, Sze WM, Wong VY, O SK. Salvagebrachytherapy for patients with locally persistent nasopharyngealcarcinoma. Int J Radiat Oncol Biol Phys 2000; 47: 40512. 113 Chua DT, Sham JS, Kwong DL, Wei WI, Au GK, Choy D. Locallyrecurrent nasopharyngeal carcinoma: treatment results for patientswith computed tomography assessment. Int J Radiat OncolBiol Phys 1998; 41: 37986. 114 Sham JS, Choy D, Wei WI, Yau CC. Value of clinical follow-up forlocal nasopharyngeal carcinoma relapse. Head Neck 1992; 14:20817. 115 Chiesa F, De Paoli F. Distant metastases from nasopharyngealcancer. ORL J Otorhinolaryngol Relat Spec 2001; 63: 21416. 116 Hong RL, Lin CY, Ting LL, Ko JY, Hsu MM. Comparison ofclinical and molecular surveillance in patients with advancednasopharyngeal carcinoma after primary therapy: the potential roleof quantitative analysis of circulating Epstein-Barr virus DNA.Cancer 2004; 100: 142937. 117 Mutiranura A, Pornthanakasem W, Theamboonlers A, et al.Epstein-Barr viral DNA in serum of patients with nasopharyngealcarcinoma. Clin Cancer Res 1998; 4: 66569.118 Lo DYM, Leung SF, Chan LYS, et al. Kinetics of plasma Epstein-Barr virus DNA during radiation therapy for nasopharyngealcarcinoma. Cancer Res 2000; 60: 235155.119 Lin JC, Wang WY, Chen KY, et al. Quantication of plasmaEpstein-Barr virus DNA in patients with advanced nasopharyngealcarcinoma. N Engl J Med 2004; 350: 246170.120 Chan AT, Lo YM, Zee B, et al. Plasma Epstein-Barr virus DNA andresidual disease after radiotherapy for undifferentiatednasopharyngeal carcinoma. Natl Cancer Inst 2002; 94: 161419.121 Leung SF, Tam JS, Chan AT, et al. Improved accuracy of detectionof nasopharyngeal carcinoma by combined application ofcirculating Epstein-Barr virus DNA and anti-Epstein-Barr viralcapsid antigen IgA antibody. Clin Cancer Res 2003; 15: 343134.122 Wei WI, Yuen AP, Ng RW, Ho WK, Kwong DL, Sham JS.Quantitative analysis of plasma cell-free Epstein-Barr virus DNA innasopharyngeal carcinoma after salvage nasopharyngectomy. aprospective study. Head Neck 2004; 26: 87883.123 Fang FM, Chiu HC, Kuo WR, et al. Health-related quality of life fornasopharyngeal carcinoma patients with cancer-free survival aftertreatment. Int J Radiat Oncol Biol Phys 2002; 53: 95968. 124 Millan AS, Pow EH, Leung WK, Wong MC, Kwong DL. Oralhealth-related quality of life in southern Chinese followingradiotherapy for nasopharyngeal carcinoma. J Oral Rehab 2004; 31:60008.125 Lam KS, Tse VK, Wang C, Yeung RT, Ho JH. Effects of cranialirradiation on hypothalamic-pituitary functiona 5-yearlongitudinal study in patients with nasopharyngeal carcinoma.Q J Med 1991; 78: 16576. 126 Ho WK, Wei WI, Kwong DL, et al. Long-term sensorineuralhearing decit following radiotherapy in patients suffering fromnasopharyngeal carcinoma: a prospective study. Head Neck 1999;21: 54753. 127 Pow EH, McMillan AS, Leung WK, Wong MC, Kwong DL. Salivarygland function and xerostomia in southern Chinese followingradiotherapy for nasopharyngeal carcinoma. Clin Oral Investig2003; 7: 23034. 128 Pow EH, McMillan AS, Leung WK, Kwong DL, Wong MC. Oralhealth condition in southern Chinese after radiotherapy fornasopharyngeal carcinoma: extent and nature of the problem.Oral Dis 2003; 9: 196202. 129 Leung SF, Zheng Y, Choi CY, et al. Quantitative measurement ofpost-irradiation neck brosis based on the young modulus:description of a new method and clinical results. Cancer 2002; 95:65662. 130 Cheng SW, Ting AC, Lam LK, Wei WI. Carotid stenosis afterradiotherapy for nasopharyngeal carcinoma. Arch OtolaryngolHead Neck Surg 2000; 126: 51721. 131 Lee AW, Kwong DL, Leung SF, et al. Factors affecting risk ofsymptomatic temporal lobe necrosis: signicance of fractionaldose and treatment time. Int J Radiat Oncol Biol Phys 2002; 53:7585. 132 Lin YS, Jen YM, Lin JC. Radiation-related cranial nerve palsy inpatients with nasopharyngeal carcinoma. Cancer 2002; 95: 40409. 133 Chang YC, Chen SY, Lui LT, et al. Dysphagia in patients withnasopharyngeal cancer after radiation therapy: a videouoroscopicswallowing study. Dysphagia 2003; 18: 13543. 134 Lam LC, Leung SF, Chan YL. Progress of memory function afterradiation therapy in patients with nasopharyngeal carcinoma.J Neuropsychiatry Clin Neurosci 2003; 15: 9097. 135 Cheung M, Chan AS, Law SC, Chan JH, Tse VK. Cognitivefunction of patients with nasopharyngeal carcinoma with andwithout temporal lobe radionecrosis. Arch Neurol 2000; 57:134752. 136 Lee PW, Hung BK, Woo EK, Tai PT, Choi DT. Effects of radiationtherapy on neuropsychological functioning in patients withnasopharyngeal carcinoma. J Neurol Neurosurg Psychiatry 1989; 52:48892. 137 Lee AW, Law SC, Ng SH, et al. Retrospective analysis ofnasopharyngeal carcinoma treated during 19761985: latecomplications following megavoltage irradiation. Br J Radiol 1992;65: 91828. 138 Sham J, Choy D, Kwong PW, et al. Radiotherapy fornasopharyngeal carcinoma: shielding the pituitary may improvetherapeutic ratio. Int J Radiat Oncol Biol Phys 1994; 29: 699704. 139 Kao CH, Tsai SC, Wang JJ, Ho YJ, Yen RF, Ho ST. Comparing18-uoro-2-deoxyglucose positron emission tomography with acombination of technetium 99m tetrofosmin single photonemission computed tomography and computed tomography todetect recurrent or persistent nasopharyngeal carcinomas afterradiotherapy. Cancer 2001; 92: 43439.140 Wei WI, Ho CM, Wong MP, Ng WF, Lau SK, Lam KH.Pathological basis of surgery in the management ofpostradiotherapy cervical metastasis in nasopharyngeal carcinoma.Arch Otolaryngol Head Neck Surg 1992; 118: 92329.141 Chua DT, Wei WI, Sham JS, Cheng AC, Au G. Treatment outcomefor synchronous locoregional failures of nasopharyngealcarcinoma. Head Neck 2003; 25: 58594.142 Wei WI, Lam KH, Ho CM, Sham JS, Lau SK. Efcacy of radicalneck dissection for the control of cervical metastasis afterradiotherapy for nasopharyngeal carcinoma. Am J Surg 1990; 160:43942.143 Wei WI, Sham JS, Choy D, Ho CM, Lam KH. Split-palate approachfor gold grain implantation in nasopharyngeal carcinoma.Arch Otolaryngol Head Neck Surg 1990; 116: 57882.144 Wei WI, Lam KH, Sham JS. New approach to the nasopharynx: themaxillary swing approach. Head Neck 1991; 13: 20007.145 Morton RP, Liavaag PG, McLean M, Freeman JL. Transcervico-mandibulo-palatal approach for surgical salvage of recurrentnasopharyngeal cancer. Head Neck 1996; 18: 35258.146 Huang SC, Lui LT, Lynn TC. Nasopharyngeal cancer: study III.A review of 1206 patients treated with combined modalities.Int J Radiat Oncol Biol Phys 1985; 11: 178993.147 Sham JS, Choy D. Nasopharyngeal carcinoma: treatment of necknode recurrence by radiotherapy. Australas Radiol 1991; 35: 37073.www.thelancet.com Vol 365 June 11, 2005 2053Seminar148 Wei WI, Ho WK, Cheng AC, et al. Management of extensivecervical nodal metastasis in nasopharyngeal carcinoma afterradiotherapy: a clinicopathological study. Arch Otolaryngol HeadNeck Surg 2001; 127: 145762. 149 Lee AW, Law SC, Foo W, et al. Retrospective analysis of patientswith nasopharyngeal carcinoma treated during 19761985: survivalafter local recurrence. Int J Radiat Oncol Biol Phys 1993; 26: 77382.150 Xiao J, Xu G, Miao Y. Fractionated stereotactic radiosurgery for50 patients with recurrent or residual nasopharyngeal carcinoma.Int J Radiat Oncol Biol Phys 2001; 51: 16470.151 Wang CC, Busse J, Gitterman M. A simple afterloading applicatorfor intracavitary irradiation of carcinoma of the nasopharynx.Radiology 1975; 115: 73738. 152 Harrison LB, Weissberg JB. A technique for interstitialnasopharyngeal brachytherapy. Int J Radiat Oncol Biol Phys 1987;13: 45153.153 Choy D, Sham JS, Wei WI, Ho CM, Wu PM. Transpalatal insertionof radioactive gold grain for the treatment of persistent andrecurrent nasopharyngeal carcinoma.Int J Radiat Oncol Biol Phys1993; 25: 50512. 154 Kwong DL, Wei WI, Cheng AC, et al. Long term results ofradioactive gold grain implantation for the treatment of persistentand recurrent nasopharyngeal carcinoma. Cancer 2001; 91:110513. 155 Leung TW, Tung SY, Wong VY, et al. High dose rate intracavitarybrachytherapy in the treatment of nasopharyngeal carcinoma.Acta Oncol 1996; 35: 4347. 156 Law SC, Lam WK, Ng MF, Au SK, Mak WT, Lau WH.Reirradiation of nasopharyngeal carcinoma with intracavitary moldbrachytherapy: an effective means of local salvage. Int J RadiatOncol Biol Phys 2002; 54: 1095113. 157 Fisch U. The infratemporal fossa approach for nasopharyngealtumors. Laryngoscope 1983; 93: 3644.158 Fee WE Jr, Roberson JB Jr, Gofnet DR. Long-term survival aftersurgical resection for recurrent nasopharyngeal cancer afterradiotherapy failure. Arch Otolaryngol Head Neck Surg 1991; 117:123336.159 Fee WE Jr, Moir MS, Choi EC, Gofnet D. Nasopharyngectomy for recurrent nasopharyngeal cancer: a 2- to 17-year follow-up. Arch Otolaryngol Head Neck Surg 2002; 128: 28084.160 Wei WI. Nasopharyngeal cancer: current status of management.Arch Otolaryngol Head Neck Surg 2001; 127: 76669.161 Leung TW, Tung SY, Sze WK, et al. Salvage radiation therapy forlocally recurrent nasopharyngeal carcinoma. Int J Radiat OncolBiol Phys 2000; 48: 133138. 162 Poon D, Yap SP, Wong ZW, et al. Concurrent chemoradiotherapyin locoregionally recurrent nasopharyngeal carcinoma. Int J RadiatOncol Biol Phys 2004; 59: 131218. 163 Wang TL, Tan YO. Cisplatin and 5-uorouracil continuousinfusion for metastatic nasopharyngeal carcinoma. Ann Acad MedSingapore 1991; 20: 60103. 164 Chua DT, Sham JS, Au GK. A phase II study of capecitabine inpatients with recurrent and metastatic nasopharyngeal carcinomapretreated with platinum-based chemotherapy. Oral Oncol 2003;39: 36166.165 Ngan RK, Yiu HH, Lau WH, et al. Combination gemcitabine andcisplatin chemotherapy for metastatic or recurrent nasopharyngealcarcinoma: report of a phase II study. Ann Oncol 2002; 13: 125258.166 Tan EH, Khoo KS, Wee J, et al. Phase II trial of a paclitaxel andcarboplatin combination in Asian patients with metastaticnasopharyngeal carcinoma. Ann Oncol 1999; 10: 23537. 167 Chua DT, Kwong DL, Sham JS, Au GK, Choy D. A phase II studyof ifosfamide, 5-uorouracil and leucovorin in patients withrecurrent nasopharyngeal carcinoma previously treated withplatinum chemotherapy. Eur J Cancer 2000; 36: 73641.168 Taamma A, Fandi A, Azli N, et al. Phase II trial of chemotherapywith 5-uorouracil, bleomycin, epirubicin, and cisplatin forpatients with locally advanced, metastatic, or recurrentundifferentiated carcinoma of the nasopharyngeal type. Cancer1999; 86: 110108.169 Siu LL, Czaykowski PM, Tannock IF. Phase I/II study of theCAPABLE regimen for patients with poorly differentiatedcarcinoma of the nasopharynx. J Clin Oncol 1998; 16: 251421.170 Boussen H, Cvitkovic E, Wendling JL, et al. Chemotherapy ofmetastatic and/or recurrent undifferentiated nasopharyngealcarcinoma with cisplatin, bleomycin, and uorouracil. J Clin Oncol1991; 9: 167581. 171 Fandi A, Bachouchi M, Azli N, et al. Long-term disease-freesurvivors in metastatic undifferentiated carcinoma ofnasopharyngeal type. J Clin Oncol 2000; 18: 132430. 172 Cheng LC, Sham JS, Chiu CS, Fu KH, Lee JW, Mok CK. Surgicalresection of pulmonary metastases from nasopharyngealcarcinoma. Aust N Z J Surg 1996; 66: 7173.173 Kwan WH, Teo PM, Chow LT, Choi PH, Johnson PJ.Nasopharyngeal carcinoma with metastatic disease to mediastinaland hilar lymph nodes: an indication for more aggressivetreatment. Clin Oncol (R Coll Radiol) 1996; 8: 5558.174 Li JH, Chia M, Shi W, et al. Tumor-targeted gene therapy fornasopharyngeal carcinoma. Cancer Res 2002; 62: 17178. 175 Duraiswamy J, Sherritt M, Thomson S, et al. Therapeutic LMP1polyepitope vaccine for EBV-associated Hodgkin disease andnasopharyngeal carcinoma. Blood 2003; 101: 315056.176 Chua D, Huang J, Zheng B, et al. Adoptive transfer of autologousEpstein-Barr virus-specic cytotoxic T cells for nasopharyngealcarcinoma. Int J Cancer 2001; 94: 7380. 177 Lopes V, Young LS, Murray PG. Epstein-Barr virus-associatedcancers: aetiology and treatment. Herpes 2003; 10: 7882. 2054 www.thelancet.com Vol 365 June 11, 2005