precancerous lesions in the kidney

Precancerous Lesions in the KidneyHein Van Poppel, Sten Nilsson, Ferran Algaba, Ulf Bergerheim, Paola Dal Cin, Stewart Fleming,Sverker Hellsten, Ziya Kirkali, Laurence Klotz, Per Lindblad, Borje Ljungberg, Peter Mulders,Tania Roskams, Ronald K. Ross, Cheryl Walker and Peter Wersall

From the Department of Urology, University Hospital K.U. Leuven, Leuven, Belgium

Scand J Urol Nephrol Suppl 205: 136–165, 2000

Renal cell carcinoma (RCC), although occurring less frequently than prostate and bladder cancer, is actually the mostmalignant urologic disease, killing >35% of affected patients. Therefore, investigation of the nature of premalignant lesions ofthe kidney is a relevant issue. Following the most recent histological classification RCC can be subdivided into four categories:conventional RCC; papillary RCC; chromophobe RCC; and collecting duct carcinoma.

In contrast to many genitourinary malignancies, premalignant alterations in the kidney are scarcely described. Intratubularepithelial dysplasia has been recognized as the most common precursor of RCC. In analogy to prostatic intraepithelial neoplasia(PIN), the premalignant lesions of the kidney are described as high or low-grade renal intratubular neoplasia. In contrast,precancerous lesions have been described as part of the von Hippel–Lindau syndrome (VHL) where the evolution from a simplecyst to an atypical cyst with epithelial hyperplasia to cystic or solid conventional-type RCC is well documented. Finally, in thegenesis of papillary RCC an adenoma–carcinoma sequence has been recognized with specific genetic changes. There are nodata on the epidemiology of premalignant lesions of the kidney, but research into the etiology of RCC has been extendedsubstantially. Familial and genetic factors are well documented in VHL disease, in hereditary papillary RCC, in the tuberoussclerosis complex and in familial RCC. Cigarette smoking and obesity are established risk factors for RCC. Hypertension or itsmedication has also been associated with an increased risk. Among dietary factors an inverse relation between risk andconsumption of vegetables and fruit has been found. Occupational exposure to substances such as asbestos and solvents hasbeen linked to an increased risk of RCC.

Specific RCC variants have distinctive chromosome alterations and several genes have been implicated in the developmentof RCC. Loss of material from the 3p chromosome characterizes conventional RCC and the deletion of the VHL suppressorgene plays an important role in the genesis of this RCC variant. In contrast, numerical changes with trisomy of chromosomes 7and 17 and loss of the sex chromosome are typical changes in papillary tumors, whereas papillary RCC have additionaltrisomies. Chromophobe RCC is characterized by loss of chromosomes with a combination of monosomies. Less consistentgenetic alterations are associated with collecting duct carcinoma.

The traditional treatment of RCC is surgery by radical or partial nephrectomy. The latter approach carries a risk of tumorrecurrence as a result of unrecognized satellite lesions or premalignant lesions that might have been present at the time ofsurgery. However, the reported recurrence rates after partial nephrectomy are <1% and therefore the possible presence ofpremalignant disease does not alter the actual treatment strategy advocated. Although multifocality and bilateral occurrence ofRCC are much more likely in cases of papillary RCC, biopsy of the renal remnant or contralateral kidney is not justified even inpatients with this tumor type. Conversely, patients with RIN in a partial or radical nephrectomy specimen or in a renal biopsytaken for whatever reason should be subjected to closer follow-up with regularly repeated ultrasound.

When an effective chemopreventive regimen becomes available it might be useful for patients with an inherited risk of RCCas well as in those who are at risk of tumor recurrence after intervention. Mass screening with the purpose of detecting RCC atits earliest stage is not recommended at the present time, but screening focused on certain risk groups can be advocated. Furtherresearch is needed to identify avoidable risks, develop effective chemoprevention and recognize patients at risk.

Key words:kidney, precancerous lesions.

Hein Van Poppel, Department of Urology, University Hospital K.U. Leuven, B-3000 Leuven, Belgium. (E-mail:[email protected])

DESCRIPTION OF THE PROBLEM

In contrast to many genitourinary malignancies, pre-malignant alterations in the kidney are at present poorlydescribed. This is primarily due to the fact that tumors ofthe kidney tend to be asymptomatic until very late in thecourse of disease progression. In contrast, renal malig-nancies comprise a very heterogeneous class of tumorsarising from different cell types within the nephron.

Renal cell carcinoma (RCC) is an adult-onsetepithelial malignancy, accounting for 80–90% of allmalignant kidney tumors. During the latter part of the20th century, when the burden of cancer becameincreasingly well characterized, the incidence of, andmortality arising from, RCC increased in many parts ofthe world. The estimated values in 1990 were�30000in North America and 20000 in the European Union(1, 2). The incidence of RCC has increased by >30% in

2000 Taylor & Francis.ISSN 0036–5599 Scand J Urol Nephrol Suppl 205

the past two decades,a reflection of both earlierdiagnosisandthelink betweenRCCandenvironmentalrisk factors.

Thesetumorshavea poor prognosisbecauseof thepropensityof RCCto invadesmall vascularstructures(3). Historically, 30% of patients have exhibitedmetastaticdiseaseat the time of initial presentation.In patients with apparently localized disease,50%ultimately progress with distant metastasesafterremoval of the primary tumor (4). Early detectionandstagemigrationmay lower theseproportions.

RCC has an extremely variable clinical course.Prognosisis notwell correlatedwith specifichistologicsubtypesof RCC but can be predictedby the tumorstageandgrade.It is oneof the few malignanciesforwhich spontaneousregressionis well documented(5, 6) and the 5-year survival rate for patientswithsolitarymetastaseshasbeenshownin severalseriestorangefrom 5% to 50% (7–13).Diseaseprogressionisvariable even in metastaticdisease,and few clinicalprognostic factors are available to guide treatment,being essentiallylimited to high performancestatusandlung-onlymetastaticdisease(14–17).

Cytotoxicchemotherapeuticstrategieswhicharethestandardtreatmentfor other solid malignancieshavebeenonly minimally efficaciousfor RCC (18). As aresultof thelack of demonstrableefficacyof cytotoxicagents,thereis presentlynostandardchemotherapeuticapproachfor RCCandno recognizedsystemictherapyis effectiveat reducingtheprobability of relapse(15).Immunotherapeuticagentssuch as interferon-a andinterleukin-2 have been shown to have moderateefficacy,with objectiveresponseratesrangingbetween10%and30%,althoughtheresponsehasbeendurablein only a limited numberof patients(14).Thusthereisa greatneedfor the developmentof novel therapeuticagentsfor RCC. A better understandingof the mol-ecularetiologyof this diseaseis animportantfirst steptowardsthis goal.

The World HealthOrganization(WHO) nomencla-ture subdividesrenal tumorsinto adenomas,carcino-mas(with and without papillary growth pattern)andothers (19). Our limited knowledge regarding thefundamentalsof diseaseprogressionof RCC meansthatthedistinctionbetweenadenomaandcarcinomaisstill unresolved.It is clear that size alone (cut-off =3 cm in diameter) is not sufficiently prognostic,astumorsas small as 0.5cm havebeendocumentedtometastasize(20). At present there is no way toaccurately determine the prognosis of these smalllesionsthatarebeingdetectedmorefrequentlybecauseof the increasedsensitivity of modernimaging tech-niques(21).Datafrom autopsystudiessuggestthattheincidenceof smallrenaltumorsis muchmorefrequentthan that of overt malignantRCC (22). This makes

researchon the prevalenceand relevanceof pre-malignantlesionsthataretheprecursorsof thesesmallrenaltumorsevenmoredifficult.

DIAGNOSTIC CRITERIA

Introduction

The classificationof RCCthat previouslyappliedwasbasedon thecytologicappearanceandthecell typeoforigin (23). It distinguishedbetweenclear and chro-mophilic cell tumors (thought to arise from theproximal tubule), chromophobeand oncocytic celltumors(thoughtto arisefrom thecortical tubules)andBellini duct tumors(thoughtto arisefrom the corticalcollecting ducts).Variantsof thesebasictumor typesinclude eosinophilic variants, that result from anaccumulationof mitochondria,andspindlecell/pleio-morphic variants that arise as a consequenceof thepresenceof sarcomatoidtransformationin portionsofthe tumor.Furtherrefinementof this classificationhasbeenundertakenbasedonourcurrentunderstandingofthe geneticalterationswhich lead to the formationofrenal cancer(24). This classificationrecognizesRCCof conventional,papillary, chromophobeand collect-ing-ductcarcinomatypes(25).

ConventionalRCC, predominantlyof the clear-cellvariety, constitutes �80% of all adult epithelialneoplasmsof thekidneyandis by far thepredominantvariantseenin the clinic. Papillary tumors,which areusually chromophilic,makeup another10% of RCCvariants and chromophobeRCCs and oncocytomasaccountfor mostof theremaining10%of solid tumorsthat occur in adults (20). Diagnosisand subsequentprognosisof RCCareoftencomplicatedby therangeofphenotypicdiversitydisplayedby thesetumors(4).Forexample,in contrastto the poor prognosisassociatedwith othertypesof renaltumor,theoncocytomavariantis consideredbenignandhasagoodprognosis(26–28).

A decade’sworthof valuabledataon thegeneticsofRCC has given us a deeper understandingof themolecularpathogenesis.RCCof theconventionaltype(clear-cellvariant) is consistentlyassociatedwith theloss of the short arm of chromosome3 and, in themajority of cases,with mutationor inactivationof thevon Hippel–Lindau tumor suppressorgene (29,30).PapillaryRCC hasa characteristicmorphologyandisassociatedwith trisomiesof chromosomes7 and17andloss of the Y chromosomein males (31,32). Twodifferent subtypesof papillary RCC may be identified(33), althoughthis remainsto be confirmed.Chromo-phobeRCC is reportedto havea betterprognosisandexhibits multiple chromosomallossesand is hypodi-ploid by DNA flow cytometry(34,35).Collecting-ductcarcinomais the leastcommonbut mostaggressiveof

ScandJ Urol NephrolSuppl205

Precancerouslesionsin the kidney 137

renal epithelial neoplasms,being associatedwith a>90% mortality rate. It accountsfor 1–2% of adultrenal tumors but detailedgeneticanalysisis lacking(36,37).

TheRCCpathogenesiscontroversywasinitiatedbyPaul Grawitz in 1883 when he comparedrenal cellswith corticaladrenalcells.Following Cohnheim’scellrest theory,he consideredthat the RCC premalignantlesion was the result of a small ectopic adrenalrest(strumasuprarenalisaberrata).In 1893Sudeckidenti-fied atypical changesin tubular cells near to renalcancer,but the majority of authorsfavoredGrawitz’shypothesisandfor a long time the adrenalcells wereconsideredto be the initial lesion. Stoerk, in 1908,returned to Sudeck’s theory, supportedby Ewing(1919) and Wright (1922), and convincing evidencefor thetubularorigin of RCCwasobtainedby Oberling(38) [for historical review seeDelahuntandThornton(39)].

Despiterecentadvancesin theunderstandingof themolecular alterationswhich are seen in establishedneoplasms,very little is known about the molecularbiology of precancerouschanges,the sequenceofaccumulationof molecularalterationsandthepathwayof progressionfrom normalrenaltubularepitheliumtocarcinoma.Threeprecancerouslesionsof the kidneywill be consideredin this review,namelyintratubularepithelial dysplasia,preneoplasticlesionsin the vonHippel–Lindausyndromeandtheadenoma–carcinomasequenceof papillary RCC.

Intratubular epithelial dysplasia.The first reportsondysplasticchangesin the animal kidney dateback tothe 1950s.Tubular dysplasiain hamsterswas foundto be a precursor lesion for kidney cancer anddevelopmentof micro-invasivecarcinomafrom thesedysplastic changeswas described (40). RCC wasinduced experimentally in male Syrian hamstersexposedto high-doseestrogens(41). This hormonalinduction of renal tumorsmay be relevantto humandiseaseas the diffuse pattern of dysplasiais morecommon in women than in men. Most tumorsinduced by ethyl nitroso urea (papillary type) werealsoassociatedwith tubulardysplasia(42).

The existenceof a dysplastictubular epitheliuminhumankidneysremovedfor RCCis nowrecognized.Inthemid 1970sintratubulardysplasiawasdescribedforthe first time, occurring in threecasesof a seriesofpapillary RCCs. The authors consideredthat thesedysplasticlesionswerein collectingductsandthat thetumors were collecting-duct carcinomas(43). Sincethen epithelial dysplasiain collecting ductshasbeenconsistentlyrecognizedin a proportion of casesofcollecting-ductcarcinoma(36).

Epithelial hyperplasiais consideredto be theorigin

of neoplasiain patientswith polycystickidneydisease(44). There is also a high incidence of dysplasticchangesin the adjacenttubulesin papillary (45) andcollecting-duct carcinomas(46). Dysplastic changesmay be observedin the tubulesof normalkidneysbutthesechangesare not as severeas thoseidentified inassociationwith RCC(47).

Intratubular epithelial dysplasia(IED), associatedwith other variants of RCC, particularly the morecommon forms, was systematicallyinvestigatedbyMourad et al. (48). They identified morphologicalchangesin the tubules of the kidney, which theyconsideredto bedysplastic,in 30of 110casesof RCC.Review of 23 nephrectomiesperformed for non-malignantconditionsfailed to demonstratethis intra-tubulardysplasia.The lesionsweremainly seenin therenal cortex and were predominantlyperiglomerularand focal, although a diffuse change pattern wasidentifiedin six of the 30 cases.IED, andparticularlythe diffuse variant, was slightly more common inwomenthanin men.IED wasfound in 32% of clear-cell RCCs (equivalent to conventionalRCC types),14% of tubulopapillaryRCCs(equivalentto papillaryRCC),16% of oncocytic/granularcell-typeRCCs(nodirect equivalentin the current classification)and in75%of RCCsshowingsarcomatoidtransformation.

IED was also describedin associationwith threecasesof cystic RCC (probablyof conventionaltype)(49).In amoredetailedanalysisof 62casesof RCCforthepresenceof IED thesechangeswereseenin 23%ofcases,againmorefrequentlyin conventionalRCCbutalso in papillary RCC and in RCC showingsarcoma-toid transformation(47). Immunocytochemical inves-tigationssuggestthatIED lesionsshowsimilar profilesto RCC, with co-expressionof keratinsandvimentin,increasedexpressionof PCNA and expressionofepidermal growth factor receptor and cathepsinD(47).Thisobservationmaysupportthesuggestionof analteration in biologic behavior involving geneticchanges during the progression of carcinogenesisfrom dysplasiato carcinoma.Despite immunocyto-chemicalevidenceof p53expressionin kidneytubulesadjacentto RCC, but not in neoplastickidney (50),thereareno datato suggestthat geneticalterationsinp53occurin precursorlesionsof RCC.

Currently the lesionsaredefinedby morphologicalcriteria. IED is manifest by nuclear crowding andnuclear enlargement(Fig. 1). The dysplastic cellspossessnuclei two to three times the size of normalor reactive tubular epithelial cells (47,48). Theenlargednuclei arevesicularandrequirethe presenceof prominent,usually eosinophilic,nucleoli. Mitoticfiguresmay be present.Thesehistologicchangesmaybe either focal or diffuse. The cytoplasmfrequentlyshows a clear cell change. Carcinoma in situ is


138 H. Van Poppelet al.

recognizedby a multilayered and solid intratubularexpansionof cells with the samecytologic featuresasseenin IED. The tubularbasementmembraneshouldbe intact in orderto sustainthediagnosisof IED. This

mayrequiretheuseof eitherprostateantigenserumorsilver-stainedpreparations.

Premalignantlesionsof thekidneymaybegradedbythe severityof the lesion.They can be subdividedas

Fig. 1. Pathologicalfinding of premalignantlesionsin thekidney. (a) Thenormalrenalcortexconsistsof glomeruli andtubules.Thereis amixture of proximal anddistal tubules.Both typesare lined by cuboidalcells with eosinophiliccytoplasmandnuclei placedtowardsthebase.The cells of the proximal tubules arelargerthanthoseof the distal tubule.The cells arewell orientatedwith no nuclearcrowdingornuclearatypia.(b) Early dysplasticchangesareseenwithin tubules.Thesechangesarerecognizedby clearcell changein the cytoplasm,increasednuclearsizeandnuclearcrowding.In mild changessuchasthoseseenherethecells remainasa singlelayer lining thetubule.(c)Moreseverechangeswith markednucleiatypia,crowdingandearlymulti-layeringof theepithelium.Thenucleicontainprominentnucleoli.Someinterstitialfibrosisandinflammationaccompaniesthesechanges.(d) Thenormalcollectingductsarelined by cuboidalcellswith clearcytoplasmanda centrally placednucleus.Collecting-ductepitheliumis of larger cell size thanproximal or distal tubularepithelium.(e)Dysplasiaof thecollectingduct is evidencedby increasednuclearsize,chromatinstainingandpleomorphism.Nucleoli appearwithin thesenuclei. (f) Nuclearpleomorphismis moremarked. Thereis somemulti-layeringor early papilla formationwithin thecollectingducts.Thecollecting-ductbasementmembraneis often thickenedandthereis someinterstitial fibrosisandinflammation.



mild, moderateor severedysplasia,carcinomain situor renal intratubulardysplasiaI, II and III, or renalintratubularneoplasia(RIN; low or high grade).Usingthe latter system, severe dysplastic changeswithknown preneoplasticfeaturescan be separatedfrommilder lesionswith lesscertainpreneoplasticcapacityandbiologic behavior.In two recentarticlesconsider-ing the recommendationsfor the reporting of renaltumors, renal tubular dysplasia is not mentioned(51,52). Interpretationand reportingof theselesionsmayyield animprovedunderstandingof their biologicnature.

Precancerouslesionsin the von Hippel–Lindausyn-drome. The von Hippel–Lindausyndrome(VHL) isan autosomal-dominantinherited cancer predisposi-tion syndrome affecting 1/35000–1/40000 people

Fig. 2. (a) In VHL syndromenumerouscystsandmicrocystsarepresentthroughoutthekidney.Thelining of someof theseshowdysplasia.Thereis nuclearpleomorphismwith prominentnucleoli andsomedouble-or multi-layering of the lining of the tubules.Adjacentmicro-cystsshowsimilar changes.(b) In VHL syndromecarcinomain situdevelopswith multi-layeringandeventualsolidificationof thetubule.Inthis dysplastictubuleatypical cells occupythe whole tubulardiameter.

Fig. 3. (a)A smallpapillaryadenoma<5mm in diameter.Thereis a papillaryarchitectureandthepapillaearelined by cuboidalcells.Thesecells exhibit mild pleiomorphismandremainsingle-layered.(b) A largerpapillary adenomastill <5 mm in diameterbut showinga morecomplexpapillary architecturewith occasionalsolid areasandmorevariation in nuclei andcellular structure.

Fig. 4. Worldwide incidenceratesof kidney cancerin malesandfemales.



(53). Patientswith VHL havea predispositionto thedevelopment of a wide variety of tumors, mostfrequently hemangioblastomaof the central nervoussystem and retina, pheochromocytomaand RCC.They also develop cystic lesions of the kidney,pancreasand epididymis. These patients inherit amutantform of theVHL genebut tumor developmentis accompaniedby loss of the wild-type allele. Themost important prognosticfeature in the VHL syn-dromeis thedevelopmentof RCC.

The renal pathology of the VHL syndrome ischaracterizedby threeseparategroupsof lesions(54).A single layer of tubular epithelium without anycytologicalatypialinessimplecysts.Atypical cystsornon-cystictubulesdemonstratefocal or diffuseepithe-lial hyperplasiawith stratificationto threeor morecelllayers of tubular epithelium (Fig. 2). Theseatypicalcysts may show a rangeof cytological atypia, frommild nuclear hyperchromasiaand increasednuclearsizeto markedpleomorphismwith mitotic activity andatypicalmitoses.In theVHL syndrome,RCCsmaybeeither solid or cystic. These are almost always ofconventionaltype,with usuallya clear-ellmorphologyandasolidacinaror trabeculararchitecture.Within thisclassificationof renal lesionsin VHL diseasethereisconsideredto be progressionfrom simple cysts toatypical cyststo RCCs.Somegeneticdataappeartosupportthis idea, with a loss of the wild-type VHLallele in atypical cysts and in RCC. These renalpathologiesare numerousin the kidneys of patientswith VHL syndromeandhasbeensuggestedthat theremay be >1000 simple or atypical cysts and severalhundredcarcinomasin the typical VHL kidney (55).

Tuberoussclerosis is an autosomal-dominantin-herited disorderwith multi-organ abnormalities.Al-though the most frequently encounteredrenal lesionis theangiomyolipoma,patientswith tuberoussclerosishave an increasedrisk of the developmentof RCC.These RCCs are usually of the conventional typeand may show anaplasticor sarcomatoidchanges.Renalcystshavebeendescribedin tuberoussclerosisbut a detailed study comparableto that performedfor the VHL syndromeis not available.The presenceof precancerousstatesin tuberoussclerosisis thereforelesswell defined.

Adenoma–carcinomasequenceof papillary RCC.Several pieces of evidence support the hypothesisthat there is an adenoma–carcinoma sequenceduringthe genesisof papillary RCC. In kidneysremovedaspart of the surgical managementof papillary RCC,coexisting papillary tumors, frequently multiple, area common finding elsewherein the renal cortex.Detailedstudy of thesehassuggestedthat a meanofeight adenomasmay be identifiedin kidneysremoved

for papillary renalcarcinoma(56). Thesedatasuggestthat papillary adenomasoccur more frequently thanpapillary RCC but that a proportionof the adenomasmay evolve into carcinoma,with increasedgrowthand the acquisition of metastatic potential. Thesechangesare probably the result of the accumulationof further geneticchange.Adenomasmustbe <5mmin diameter,have a papillary architectureand be ofchromophilic cell type (Fig. 3). Cytogeneticdata ofpapillary adenomasand RCCs show similar geneticfindings, particularly the presenceof trisomies ofchromosomes7 and 17 and loss of the Y chromo-somein males.The trisomy of chromosome7 mayalsobe associatedwith activatingmutationsof the c-MET proto-oncogene(57). However,papillary RCCsappearto accumulateadditionaltrisomiesof chromo-somes8, 12, 16 and 20. Papillary renal tumors,bothadenomasand carcinomas,are a frequentfinding inpatientswith acquiredcystic disease,usuallyamongstthe hemodialysispopulation(56). Theselesionshavebeen shown not to contain alterationsin the VHLtumor suppressorgene, supporting the concept thatthey are a distinct biologic entity (58). Preliminaryinvestigationsof nuclearhistomorphometricalanaly-sis have suggestedthat in acquired cystic diseasethere is a gradualevolution of nuclearchangefromnormal tubulesthroughatrophictubules,cysts,papil-lary adenomaandcarcinoma.

The HeidelbergandMayo Clinic consensusclassi-fications distinguish papillary renal adenoma andpapillary RCCby a combinationof geneticalterations(24,59). Although the latter classification had acriterion based on size the maximum acceptablediameter for diagnosis of a papillary adenomais5 mm. The validity of sucha criterion will continueto be reviewedasgeneticevidencesuggeststhat sizealone is an unreliablecriterion to distinguishbenignandmalignantneoplasms.

A further aspect of the adenoma–carcinomase-quenceis suggestedby the interrelationshipof onco-cytoma and chromophobeRCC. There are certainlymicroscopic, ultrastructural, biochemical and histo-chemicalsimilarities betweenthesetwo tumor types(60–62).Thoeneset al. (23) concludedthat theyarosefrom the samepart of the nephron.Recentmoleculargenetic studies have shown that a proportion ofoncocytomashaveinvolvementof the 11q13chromo-somebutthatothersshowlossesof chromosomes1 andY. Theselatter lossesarealsoseenin a proportionofchromophobecarcinomas(60). Althoughtheevidenceis circumstantialatbest,it doesraisethepossibilitythatsomeoncocytomashave the potential to progress,ifuntreated,to chromophobeRCC(63).Someoncocyticadenomasmay representprecursorlesions for chro-mophobeRCC.



Conclusions

The precancerousabnormalities described in thekidneyarethreedistinctlesionswhichmaybeinvolvedat intermediatestagesin the sequenceof changesbywhich RCCdevelopsfrom normalkidney tubules.Allsubtypesof RCCmaybeprecededby IED progressingto carcinomain situ andeventuallyto invasiveRCC.This dysplastic change has more frequently beenrecognizedin associationwith collecting-ductcarcino-ma but in the last 5 years has increasingly beendescribed in associationwith the other subtypes.Patientswith the VHL syndromehave an increasedrisk of thedevelopmentof RCC,whichappearsto arisefollowing dysplasticchangeswithin the cysts asso-ciated with the VHL syndrome. Finally, there isepidemiologic,pathologicandgeneticevidenceof anadenoma–carcinomasequencein the developmentofpapillary RCC. Thereareareasof urologic pathologywhich merit further investigationin order to enhanceour understandingof the sequenceof eventsof renalcarcinogenesis.

EPIDEMIOLOGY

Introduction

It has been estimatedthat kidney cancer afflicted150000 individuals worldwide in 1990, thereby ac-countingfor 1.9%of all malignantdiseases(64). Thetrue incidenceis probablyhigher,becausemanycasesremain undetectedbefore death(65). Globally, inci-denceratesof RCC vary more than ten-fold (2). Theincidence is highest in North America, Northern,WesternandEasternEuropeandAustralia,intermedi-atein SouthernEuropeandJapanandlow elsewhereinAsia andin Africa andthe Pacific(Fig. 4) (2,66). Anincreasingtrendin incidencehasbeenobservedin menandwomenfrom almostall areasof theworld. In men,annual ratesof increasehave varied from a high of5.9%to a low of 1.5%(67). In theUSthereis agreaterannualincreaseamongthe black populationthan forthe white population(68). In someareasof Japantheincreasein the incidencerate is as great as that ofprostatecancer(69).

Most casesof RCC occur in the fourth to sixthdecadesof life, with thepeakincidencein thesixth toseventh decades.Male:female ratios are generallybetween1.5:1and2.5:1,thehigherratiosbeingfoundin suchcountriesasFrance,Italy, SpainandJapan(66).

Trends of mortality from kidney cancer haveaccuratelyparalleledtrendsin incidence,with risesinalmostall countries.Theestimatedworldwidemortal-ity rate for kidney cancer for 1990 was �78000,therebyaccountingfor 1.5%of all cancerdeaths(70).

There are no data on epidemiologic factors of

premalignantlesionsof the kidney. There is only apresumptionthatthesameissuesasthosethatrelatetothe epidemiology of RCC will also apply to suchlesionswhenandif theyareidentified.Althoughpartofthegeographicvariationsin incidencecouldbeduetodifferencesin diagnosticintensity and autopsyrates,the importanceof environmentalfactors in renal cellcarcinogenesisis likely to besubstantial.

During the last few years,researchinto theetiologyof RCC hasexpandedsubstantially.Thereareseveralwell-documentedrisk factors for RCC. Establishedenvironmentalrisk factors for RCC, mainly cigarettesmoking,obesityandhypertensionor its medications,canexplainonly a small partof theoccurrenceof thismalignancy.Other factorsthat may alsoinfluencetherisk of RCC aredietaryandoccupationalfactors(seebelow).

Familial and genetic factors. Although the vastmajority of RCCs occur in sporadic form, severalgeneticdiseasesare associatedwith RCC, includingVHL syndrome,hereditarypapillary renal carcinoma(HPRC)andtuberoussclerosis.The genesunderlyingeachof theseconditionshavebeenclonedandgerm-line mutationsin affectedpatientshave beenidenti-fied (Table I) (71). Strong correlations have beenfound betweensome of the genesinvolved in thepathogenesisof renal tumorsand the histopathologicand clinical behavioralfeatures(seethe sectionMol-ecularbiologyandgenetics).

In the kidneys, patientswith VHL often developmultiple bilateraltumorsandcysts;thecumulativeriskof RCC is >70% by the ageof 60. Becausemultipleorgansystemsareaffected,thissyndromeis associatedwith significantmorbidity, with RCC being the mostcommoncauseof deathin VHL patients(72). World-wide, �20 families without VHL syndrome aredescribedwith familial RCC, mostly of the papillarytype. The incidenceis much higher in men than inwomenandat youngerages(73).

HPRC is characterizedby multifocal, bilateralpapillary RCC. This diseasehasan autosomal-domi-nant inheritancepattern, distinct from that of otherhereditaryRCCs,someof whicharedueto agerm-line

TableI. Genespredisposingto inheritedRCC[from Gnarra (71)]

Inheriteddisease GeneandlocusConsequenceof mutation

VHL syndrome VHL tumor suppressorgene,3p25.5

Inactivating

Hereditarypapillaryrenalcarcinoma

c-MET proto-oncogene,7q31.1–34

Activating

Tuberoussclerosis1 TSC1,9q34.3 InactivatingTuberoussclerosis2 TSC2,16p13.3 Inactivating



mutation of the c-MET proto-oncogeneon chromo-some7q (73,74).

Tuberoussclerosiscomplex(TSC) is anautosomal-dominantdiseaseoccurring in 1/140000 individuals,and is more commonamongwomen.The diseaseischaracterizedby seizures, mental retardation andhamartomatouslesionsin many organs.The kidneysareaffectedmainlyby angiomyolipomasandcysts,andmorerarelyby RCC.Tuberoussclerosiscanbecausedby mutationsin either TSC1or TSC2,which map tochromosomes9q34 and 16p13, respectively.Otherthanthepropensityto developrenalcysts,therearenoobvious differencesbetweenthe clinical featuresofpatients with either mutation. In addition, largedeletions of TSC2 involving the PICD1 gene areassociatedwith severeand early onsetof polycystickidney disease.Both genesarepresumedto be tumorsuppressorgenes(75–78).

Familial casescarrying constitutionaltranslocationinvolving the 3p region and developing multiplebilateral clear-cellcarcinomas,i.e. t(3;8) (p14.2;q24)(79) and t (3;6) (p13:q25) (80), have been used todefineputative tumor suppressorgenes[for a reviewsee van den Berg and Buys (81)]. Recently, twofamilies with RCC and with balancedchromosomaltranslocationinvolving the long arm (q) of chromo-some3, i.e. t (2;3) (q35;q21)and t (3;6) (q12;q15),werereported(82).

A largeincreasein therisk of RCCdevelopmentwasindicatedin carriersof (familial) reciprocalchromo-some3 translocationsaffectingeithertheshortor longarm, in particularwhen the translocationbreakpointswere locatedpericentromerically. Of particular inter-est, Balzarini et al. (83) found 58 casesof sporadicRCCs,papillaryaswell asnon-papillary,characterizedby translocationsinvolving the3q chromosome,eitherat 3q11–12or 3q21. This notion may have conse-quencesfor theidentificationof personsat risk, for thedevelopmentof geneticcounselingstrategiesand fortheclinical managementof patients(82).

The molecular changesthat occur in clear-cellcarcinomas in the 3;8 translocation family haverecently been defined. The tumors that have beenstudied have shown a loss of the chromosome8carrying the distal part of chromosome3p. Someofthe tumors have also shown mutationsin the VHLgene. Theseresults suggestthat tumors develop inpatients with the 3;8 translocationvia a three-stepprocess:firstly, the inheritanceof the balancedchro-mosome3;8 translocation;secondly,the loss of thechromosome8 bearingthe distal part of chromosome3p;andlastly,mutationof theVHL geneonthenormal(non-translocated)chromosome3 (57,84).

Few analyticalepidemiologicstudieshavereportedonafamily historyof RCC.Onestudy,however,found

a 60%increasedrisk for RCCif a first-degreerelativewasaffectedwith the disease,whereasno associationwasfound in a smallerstudy(85,86).

Somaticinactivationof theVHL geneis commonlyfound in sporadic RCCs, suggestingthat this is acritical molecularevent in renal carcinogenesis.Onerecentpaperidentifieda “hotspot” for mutationsof theVHL geneassociatedwith exposureto trichloroethy-lene (87). Little information is available on therelationships between VHL gene inactivation andmajor establishedenvironmentalrisk factorsin RCC.Forexample,wedonotknowif VHL inactivationis animportantmoleculareventin themulti-stepprocessoftobacco-inducedRCC development.Moreover,we donot understandthe preciserole of the VHL geneinkidney cancerdevelopmentin VHL disease,andevenlessis understoodin sporadicRCC.

Autosomal-dominant polycystic kidney diseaseisthe most common renal cystic disease, with anincidence of 1/1000 people. Fifty percent of thesepatients will develop terminal renal failure. Thepredisposinggenes,PKD1 and PKD2, are locatedonchromosomes16 and 4, respectively.Although earlydata suggesteda link with RCC (44), large clinicalseriesdonotsupporta link betweenRCCandinheritedpolycystickidneydisease(88,89).

Cigarette smoking. The International Agency forResearchon Cancer(IARC) found, in 1986, “suffi-cient evidence” for a carcinogenicrole of tobaccosmokingin cancerof the renalpelvis. For RCC therewas insufficient evidencefor a causalrelation (90).However, subsequentepidemiologic, mostly case-control, studieshave convincingly demonstratedthatcigarettesmoking is a causeof RCC. In generaltherelative risks are moderate,ranging from 1.2–2.3to2.0–3.0 for heavy smokers. A majority of case-control studies demonstrateda dose–responserela-tionship in both men and women. Furthermore,thedecline in risk following cessation,demonstratedinseveralstudies,supportsa causalinterpretationof theassociationbetweencigarettesmokingandRCC.Thisinterpretationis also supportedby findings of higherrisks for current than for former smokers(91–94).The two largestcase-controlstudiesperformedpro-ducedcomparableresults for cigarettesmoking andrisk of RCC (Table II) (92,94). The proportion ofRCC that can be attributed to cigarettesmoking isbetween21% and30% for menandbetween9% and24% for women (92,94,95), dependingon the spe-cific populationstudied.

It is not clearwhich of the constituentsof cigarettesmokeare responsiblefor RCC, but N-nitroso com-pounds,especiallyN-nitrosodimethylamine, found intobaccosmoke,causesrenalcancerin severalanimal



species(90,96). Most of thesesubstancesaremetab-olized or excreted through the urinary tract. As acorollary, the urine of cigarette smokers showsincreasedmutagenicactivity (90,97,98).

This possibleetiologicrole of N-nitrosocompoundshasrecentlygainedsupportfrom anexperimentalstudyin which N-nitrosodimethylamine-inducedrat clear-cell renaltumorswereidentifiedwith VHL mutations.This study, the first to link VHL gene mutationsexperimentally with chemical exposure,provides apossiblemolecularpathwaylinking tobaccosmokingwith RCC(71,99).

Polymorphicgenesthat code for key enzymesincritical activationor detoxificationpathwayshavebeenshownto be modifiersof smoking-relatedcancerrisk.For example,NAT2 slow acetylatorsandGSTM1nullindividuals have beenfound to experienceincreasedrisk of smoking-relatedbladdercancer(100). Severalsimilar polymorphismshave beenfound to be asso-ciated with RCC (101) but their link with tobaccocarcinogenesisis not known. An associationbetweenGSTM1 and cancerrisk has beenobservedin case-controlstudiesof lung,bladderandcoloncancerswhileother studies have reported contrary findings. Theimportanceof thesepolymorphismsin mediatingtherisk of smoking-relatedcancers remains generallyunproven(57).

Although cigarettesmoking is an establishedriskfactor for RCC, the mechanismto explainrisk altera-tion hasnot beenwell elucidated.At present,little isknownabouttheextentto which tobacco-metabolizinggenesinfluencethe risk of renal cancer.Suchknowl-edgewill have the addedbenefit of pinpointing therelevantagent(s)in tobaccosmokethatareresponsiblefor renal cancer developmentamong smokers.Forexample,anobservedassociationbetweentheCYP2E1genotypeandRCC risk would supportnitrosamineasoneclassof tobaccoconstituentsthatplaysadirectrolein renal carcinogenesisamongsmokers(102). Thereare, however,also contradictoryfindings showing a

lack of associationbetweenCYP2E1 genetic poly-morphismandRCC(103).

Obesity.The mostconsistentfinding in epidemiologicstudiesof RCC hasbeenthe positiveassociationwithobesity or high relative weight (104). About 30studies have investigated the relationship amongwomen and/or men. Most case-controlstudies andalso severalcohort studieshave confirmedthe asso-ciation, which has been strongestand most consis-tently found amongwomen.An elevatedrisk of RCCfor obesemen is also supportedby the majority ofcase-control studies, including some nested case-control studies,anda dose–effecttrendwasobservedin severalof thesestudies.The resultsfrom prospec-tive cohortstudiesarein most instancesin agreementwith the case-controlstudies regarding weight orbody massindex as a risk factor for RCC. The twolargeststudiespublishedto date,both case-controlindesign,showedsignificantassociationsbetweenbodymassindex (BMI) and risk of RCC in both genders(105,106). Table III showsBMI findings for a fewselected case-control studies. The risk estimatesrangedfrom 1.3 to �4 in both womenand men forthe highest vs the lowest BMI categoriesreported.Somevariation with respectto the magnitudeof theassociationexistsbut it seemsthat body weight hasaslightly strongerimpact on RCC in women than inmen. Weight was evaluatedat different agesamongthe reportedstudiesandtherewerealsodifferencesinhow, and to what degree,control for confoundingfactors,especiallyhypertension,wasmanaged.

The mechanismsby which obesityinfluencesrenalcarcinogenesisare not clear. There are severalplau-sible biologic explanations.For example,sex steroidhormones may affect renal cell proliferation andgrowth by direct endocrinereceptor-mediatedeffects,by regulation of receptorconcentrationsor throughparacrinegrowthfactors,e.g.epidermalgrowthfactor.Furthermore, obesity is related to a number of

TableII. Main resultsfrom the two largestcase-controlstudiesinvestigatingtheassociationbetweencigarettesmokingandRCC

Area (Reference)Numberofcases/controls Risk estimates Observations Attributablerisk

Australia,Denmark,Germany,Sweden,USA (92)

1732/2309 Ever: 1.3 Trendwith intensity:� Men: 24%;women:9%

Current:1.4 Decreasedrisk followingcessation:25%> 15yearsquit

Former:1.2Heavysmokers:2.1

California,USA (94) 1204/1204 Ever: 1.35 Trendwith intensity:� Men: 21%;women:11%Current:1.53 Decreasedrisk following

cessation:30%> 10yearsquitFormer:1.24Heavysmokers:1.90



endocrineconsequencessuchasdecreasedlevelsof sexhormonebinding globulin andprogesterone,anovula-tion, insulin resistanceand increasedlevels of biolo-gically active insulin-like growth factor (IGF-I)(107,108). If obesity is related to RCC through anassociation with endocrine influences, then it ispossiblethat elevatedlevels of circulating androgensandgrowthfactors,aswell asof unopposedestrogens,may play a role in the etiology of RCC.Of particularinterestis a possibleassociationwith androgens,astheage-standardizedincidenceof renal cancer is �50–100% higher in men comparedwith womenin mostpopulations,a largely unexplainedphenomenon.Hy-pertensionmay be an intermediatestepin the causalpathwaybetweenobesityandRCC by inducing renaldamage through an associationwith metabolic orfunctional changeswithin the renal tubules whichincreasethesusceptibilityof thekidneyto carcinogensor promotingagents.

There are also other hypothesizedmechanismsbywhichobesityleadsto increasedcancerrisk.Onegrouprecently proposedthat lipid peroxidationmay be acommonetiologic pathwayfor obesityand hyperten-sion,aswell aspossiblyalsofor cigarettesmokingandfor the apparent protective effect of antioxidantvitamins (109). By-products of lipid peroxidationhave beenshown to react with renal DNA to formadducts(110). In animalmodels,lipid peroxidationisonemechanisticpathwayin chemically-inducedrenalcarcinogenesis(111). Furtherevaluationsof this uni-fying hypothesisshouldbeanotherresearchpriority.

Thecumulativeevidencefrom analyticalepidemio-logic studiesis most consistent,with obesitybeing arisk factor for RCC in womenand men.A US case-control studyestimatedtheattributablerisk of obesityfor both menandwomento be 21% (43% for womenalone)(112)andin anAustralianstudytheattributablerisk for bothmenandwomenwas13%(113).

Reproductivefactors. Few previousanalytical epide-miologic studiesof RCC have focusedon reproduc-tive factors or exogenoushormones. In general,

investigators have found little evidence that suchfactorsareimportant(95,114–116).

Severalpreviousstudiesfound eitherno, or a non-significant, association between parity and RCC(95,113,115,117–119),whereassomerecentstudiesshoweda positiveassociationfor parity (85,108,120).A few studiesshowedno associationbetweenhyster-ectomyor oophorectomyand risk of RCC (95,115),whereasothershavefound a positiveassociationwithhysterectomywith or without oophorectomy(120,121).

Weak positive associationshave been reportedoccasionallyfor oral contraceptives(118) andfor useof replacementestrogensin some(118,122),butnotall(95,123), studies.A multicenter case-controlstudy,however,found a significantly reducedrisk of RCCfollowing oral contraceptiveuse,with a suggestionofincreasedreductionwith durationof use(108); how-ever,anotherrecentstudyfoundno relationship(121).

The findings from the relatively few investigationsperformed regarding reproductive history and hor-monesremain enigmaticand inconsistent.However,there is sufficient evidence that certain hormone-related factors are associatedwith the risk of RCC.This areaof researchis worth further investigation.

Medical conditions and treatments.A number ofmedical conditionshave beenassociatedwith RCC,althoughthe evidenceis consistentfor only a few ofthem. A major concern arising from case-controlstudies exploring the relation between RCC andmedical conditions, especially urinary tract condi-tions, is recall bias.

Amongurologicdisorders,kidneystoneshavebeenfound to have positive associationsin some case-control studies(86,95,113,114,124,125), whereasacohort study found no increasedrisk of RCC (126).Amongotherdisorders,kidneyinfections(86,95,127)havebeenfoundto haveapositiveassociation,ashaveunspecifiedurinary tract infections in women (85),unspecifiedurologic disease(128) and lower urinarytract infection (113,114). Possibleassociationsbe-

TableIII. Resultsfrom selectedcase-controlstudiesinvestigatingthe associationbetweenobesityandRCC

Area (Reference)Numberofcases/controls Risk estimates Observations

Minnesota,USA (95) 495/697 BMI highestquartilecf. lowest:men1.5; women2.3 UsualweightTop 5%: men1.6; women5.9

Ontario,Canada(85) 518/1381 BMI highestquartilecf. lowest:men1.3; women2.5 5 yearsprior to diagnosisAustralia,Denmark,Germany,

Sweden,USA (105)1732/2309 BMI highestquartilecf. lowest:men1.6; women2.0 Usualweight

Top 5%: men1.6; women3.6California,USA (106) 1204/1204 BMI > 30 cf. <22: men4.6; women4.0 UsualweightWashington,USA (307) 238/616 BMI highestquartilecf. lowest:men2.3; women3.3 Maximum weight

Top 10%:men2.9; women5.2



tween RCC and thyroid disease(86,129), thyroidcancer(130) and lymphoid malignancies(131) havealsobeenreported.

In case-controlstudiesthat investigatedthe role ofdiabetesmellitus, some investigationsfound no in-creasedrisk (106,115,118,132),othersreportednon-significantlyelevatedrisks (113,124)andsomefoundan associationconfinedto women(85,95,133–135).An increasedrisk was seenin a recent large studyamong both women and men (86). A cohort studyreporteda significantly increasedincidenceof RCCamongfemale diabetic patients(136), while anotherstudy found a significantly increasedrisk of RCC inboth women and men (137) and others found noassociation(138–140).Thebiologic mechanismof therelationship betweendiabetesand RCC is unclear.However,elevatedgrowth factors and growth factorreceptorsmaybeinvolvedin thedevelopmentof RCCamongdiabeticpatients(137).

Acquired cystic kidney disease(ACKD), whichoccursin end-stagerenal disease(ESRD), is stronglyassociatedwith the developmentof RCC. ACKD hasbeenobservedin 7–22%of patientswith ESRDprior todialysis,but the proportionincreasesto 90% after 10yearsof dialysis.A notablegenderdifferencehasalsobeenobserved,with menhavingahigherincidenceandmoreseverecysticchangethanwomen(141–143).Theincidenceof RCCin ESRDpatientshasbeenreportedto beupto 40-foldhigherthanin thegeneralpopulation(141,142,144). An increasedrisk is also seenin thenative kidneys after renal transplantation(145–147).While proliferationof proximal tubularepithelialcellshasbeenidentifiedasthemajorpathogeneticmechan-ism of cyst formation,hormones(e.g. estrogens)andgrowth factorsand their receptorsmay stimulatecellproliferation andpromotecarcinogenesis.This mech-anismmay alsoexplain,in part, the onsetof multiplerenaladenomasandbilateralcarcinomasthat developin ACKD (148).

Analgesics. Phenacetin-containinganalgesics havebeenimplicated in the etiology of urothelial carcino-ma of the renal pelvis (149). Their role in theetiology of RCC is lessclear, althougha numberofstudieshave reportedmoderatelyelevatedrisks withregular or long-term use (85,95,118,128,150–152).Linked-register studies of patients with conditionslikely to requireuseof analgesicshavealso reportedincreasedrisks of RCC (153,154). Furthermore,arecent large case-controlstudy from the US showedelevatedrisk associatedwith all major typesof anal-gesics(155). A large internationalmulticenterstudy(156) found no relationshipbetweenRCC anduseofany type of analgesic(paracetamol,phenacetin,pyr-azolonesand salicylates,mainly aspirin). Moreover,

no consistent association was found between in-creasedrisk and consumptionlevel. There was norestrictionto exclusiveusersof onetype of analgesic.The lack of associationwith phenacetincould be dueto the drug beingunavailablefor manyyearsin mostof theparticipatingcountriesof themulticenterstudy.

Conjugationof reactivemetabolitesto glutathionerepresentsa major detoxificationpathwayfor analge-sic-inducedrenaltoxicity. Therefore,it is conceivablethat the polymorphicGST (gluthatione-S-transferase)geneswill haveobservableeffectson renalcancerriskamong usersof analgesics.At present,there is noinformation on suchpossibleanalgesic–geneinterac-tion effects on risk. Studies to date support thehypothesisthat the unwanted side effects of non-steroidal anti-inflammatory drugs (NSAIDs) in thekidney are due to their ability to inhibit COX-1(cyclooxygenase1) (157). COX-1-derivedrenalpros-taglandinsare important for renal homeostasisandinhibition of COX-1-relatedprostaglandinsis thoughtto be responsiblefor renal papillary necrosis(158).Renalpapillarynecrosisis anirreversiblerenaldamagemechanismwhich has been shown to increasecellproliferationin the tubularepithelium(159).Thusonepossible mechanismby which chronic NSAID usemight lead to RCC is through inhibitory effects ofNSAIDs on COX-1. As there are no data on thispossibleetiologic pathway, this mechanismremainshighly speculative.

Diuretics, antihypertensivedrugs and hypertension.An increasedrisk of RCC with diuretic usehasbeenreportedin a numberof epidemiologicstudiessincethe first report by Yu et al. (116). The magnitudeofthe risk has varied substantially among studies,mostly among women (116,127,160–163). Excessrisk, to a lesserdegree,hasalsobeenreportedamongmen (85,164). A recent mortality cohort studyshoweda moderatelyincreasedrisk of RCC with theuseof diureticsamongwomenbut not men;however,no adjustmentwas madefor the presenceof hyper-tensionor obesity (165). In contrast,a nestedcase-control study did find an increased risk, whichdisappearedafter adjustmentfor hypertension.This isin agreementwith a large multicenter case-controlstudy(166)andalsoin agreementwith anotherrecentlargecase-controlstudy,which reportedthat diureticsmight not be an independentrisk factor (106). Inaddition, dose–responsecharacteristicsof variousmedicationshavebeenlacking.

Theassociationbetweennon-diureticanti-hyperten-sive use and risk of RCC has only been the topicof a few studies.Oneof thesereportedan associationwith non-diureticanti-hypertensivedrugs,particularlyb-blockers (113). However, in a multicenter study,



there was an excessrisk that was not specificallyconfinedto anycategoryof non-diureticanti-hyperten-sive drug. This risk was reducedafter adjustmentforhypertension,but remainedsignificant(166).A reduc-tion in effectafteradjustmentfor hypertensionwasalsoreportedin arecentcase-controlstudy(167).In anotherlarge case-controlstudy regular use of non-diureticmedicationwas associatedwith an increasedrisk ofRCC. However,useby hypertensivesubjectsdid notfurther increasethe risk, and among normotensiveregularuserstherewasno association(106).Difficul-tieshavearisenin isolatingthepossibleeffectsof anti-hypertensivemedicationfrom the effectsof hyperten-sion, which itself has been linked to RCC in manystudies(106,113,118,161,165–171).

It is consistentthat hypertensionis an independentrisk factor for RCC but therearestill uncertaintiesinseparatingthe effect of hypertensionfrom that of itsmedicaltreatment.

Dietary factors.Resultsfrom analyticalstudiesof therelation betweendiet and RCC come mostly fromcase-controlstudies.The evidenceaccumulatedfromthesestudiessuggeststhat diet may havea role in thedevelopmentof RCC, althoughpresentlythere is noestablishedlink betweenany food item or nutrientandrisk of RCC(104).

Severalstudieshave shown a positive associationwith dietary intake of meat,milk, margarine,oils orbutter. A majority of case-controlstudiesand somecohort studies have found a protective effect ofvegetablesand/orfruit (104,172,173).Stronginverseassociationshave been found for dark green andcruciferousvegetables(172,173).

Associationsbetweenspecific nutrients and RCChavebeenanalyzedin a limited numberof case-controlstudies.Thesestudieshave shown elevatedrisks inrelationto proteinintakeanddietaryfat.Decreasedriskhasbeenfound to be associatedwith dietarycarotene(125,173) and vitamin C intake (173,174). In twostudies this associationwas stronger among non-smokers(172,174). An inverseassociationhas alsobeen documentedwith other dietary antioxidants(104,172–174).

The most consistentfindings regardingdiet havebeenthe inverserelationshipwith vegetablesandfruitandtheirnutrientcomponents.Althoughtherearelittledata from humanstudieson individual antioxidants,ascorbicacid, a-tocopherol,selenium,b-caroteneandisothiocynateshave all been implicated as humancancerprotectants(175,176).

Despiterecentadvances,theanticarcinogenicmech-anismsof antioxidantsasyet remainunclear.It is mostplausible that thesesubstancesact togetherand thatprotection is not conferredby a single nutrient in a

particular food. There are inconsistenciesand con-troversialfindingsconcerningtherole thatdietplaysinthe etiology of RCC. Nevertheless,there is evidencethat theremay be a protectiverole for vegetableandfruit consumption.

Occupationalexposures.RCC is generally not con-sideredan occupation-relatedcancer, in contrast tobladdercancer.However, a numberof studieshavereportedan associationwith asbestosexposure(177,178). Associations have also been reported withexposureto gasolineandotherpetroleumproductsinsome, but not all, studies (177,178). Studies havealso reportedassociationswith occupationssuch asdry-cleaningand laundry and those involving expo-sureto hydrocarbonsand cadmium(177,178). How-ever, except for asbestos,no occupationalexposureor occupationhas been consistentlyassociatedwithRCC(178).

TheIARC considersthatbothtrichloroethyleneandtetrachloroethylenearecarcinogenicin animals,andareprobablehuman carcinogens.The main usesof tri-chloroethyleneand tetrachloroethylene are in metaldegreasinganddry-cleaning,respectively(179).How-ever,in acritical reviewof bothcohortandcase-controlstudiesit wasconcludedthat it wasnot clearwhetherthesesolventsposea risk in humans.Recall bias islikely in case-controlstudiesandthefindingsfrom thefewcohortstudieswith adequatestatisticalpowerargueagainsta causalrelationship(180). After this reviewtwo mortality follow-up studieshave found a non-significant excess of kidney cancer for trichloro-ethylene,with no dose relation (181,182). Anotherrecent investigation,a case-controlstudy, showedasignificantlyincreasedrisk for trichloroethylene(183).However, this study had somemethodologicalflawsthatmakeinterpretationof theresultsunreliable(184).In a recentinvestigationof RCCs,trichloroethylene-exposedpatientsshowedfrequentsomaticVHL muta-tions. There was also an associationbetween thenumberof mutationsandtheseverityof trichloroethy-leneexposure(87). This might beanotherexample,inadditionto cigarettesmoking,whereanenvironmentalexposuremightberelatedtoaspecificgenealterationinRCC. However, the occurrenceof this carcinogen-inducedalterationin premalignantalterationsin TRI-exposedindividualshasnot beenexamined.

Overall, the resultsof theserecentstudiessuggestthatoccupationmaybemoreimportantin theetiologyof RCCthanwasindicatedby earlierstudies.

Conclusion

A limited numberof risk factorsfor thedevelopmentofRCChavebeenestablishedwhile manyothersremaincontroversial(Table IV). Onecould presumethat the



sameepidemiologicfactorsapply to the developmentof precursorlesionsof RCC.It remainsto beelucidatedwhatthepotentialimportanceof theserisk factorsis interms of the identification and early detection ofpremalignantlesions.Further researchis requiredtobetterunderstandtheetiologyandpathogenesisof thisdisease.Specific geneticalterationshave beenasso-ciated with specific environmental exposures.Thesearchfor environmentalexposurein the etiology ofRCC must considerthe recentlyobtainedknowledgethatRCCconsistsof different typesof carcinomawithspecific genetic molecular characteristics,and thatthere might exist a modulating effect of geneticpolymorphismsamong metabolic activation and de-toxification enzymes.

MOLECULAR BIOLOGY AND GENETICS

Introduction

Specific RCC variants have distinctive chromosomealterations(24). In fact, it hasbeensuggestedthat agenetic approachto tumor diagnosisbasedon thepatternof chromosomealterations,observedin parti-cular types of tumors, is a more appropriateandclinically relevant strategy than the currently usedhistologic and cytologic criteria (32,185,186). Forexample, papillary tumors exhibit combinationsofnumericalchanges,i.e. trisomiesof chromosomes7, 812, 16, 17 and 20, and loss of the Y chromosome,whereasconventional(clear-cell)RCCsarecharacter-ized by chromosome3p deletions(187), suggestingthat different molecularalterationsareinvolved in thedevelopmentof thesetwo RCCvariants.

Identifying thespecificmolecularalterationsunder-lying the pathogenesisof specific tumor phenotypeswouldbeanimportantsteptowardsthedevelopmentofclinically relevantbiomarkers.Recently,themoleculartools with which to begin such an approachhavebecome available (188). Several genes have beenimplicatedin the developmentof RCC, including theVHL and TSC-2 tumor suppressorgenesand the c-MET proto-oncogene.The VHL tumor suppressorgeneis locatedon chromosome3p andis involved inboth sporadicand hereditary RCC (189–191).This

geneis inactivatedin tumorsby severalmechanisms,includingmutationandsilencingby DNA methylation(29,192–194).EventhoughtheVHL tumorsuppressorgeneis ubiquitouslyexpressed(195), its involvementin tumorigenesisappearsto be restrictedto RCC andsome of the otherwiserare malignanciesthat occurwith a high frequencyin the VHL syndrome,suchashemangioblastoma(196,197). This suggeststhat itstumor suppressoractivity is tissue-specific.VHL is auseful illustration of the fact that specific molecularalterationscould be diagnosticfor specific types ofRCC, as this geneis involved exclusivelyin conven-tional RCC but not in other RCC variants(29,198).This is consistentwith the hypothesisthat differentRCCvariantsmayhavedifferentmolecularetiologies.

Hereditary papillary RCC was recently shown tooccurasa resultof a germ-linemutationin thec-METproto-oncogene(74). In epithelial cells, c-MET func-tions as a receptor tyrosine kinase for hepatocytegrowth factor (HGF) and germ-line mutationsin c-MET constitutivelyactivateits kinasefunction. Inter-estingly,the role of c-MET in sporadicpapillaryRCCis quite variable. In a recent report only 17 of 129sporadicpapillary RCCscontainedc-MET mutations(199). The differential frequencywith which c-METbecomesactivatedin hereditaryvs sporadicpapillaryRCCsuggeststhat,evenwithin thisoneclassof tumor,differencesexistin themolecularetiologyof hereditaryvs sporadicformsof this disease.

ThehumanTSC-2tumorsuppressorgenelocatedonchromosome16wasclonedby theEuropeanTuberousSclerosisConsortium(200) and its tumor suppressorfunction was first demonstratedfor RCC in the Ekerrat. Lossof heterozygosity(LOH) of TSC-2 hasalsobeenshownto occur in renal hamartomas(201–204)and other tumors that frequently occur in tuberoussclerosispatients(202)whoarealsoat increasedrisk ofthedevelopmentof renaltumors(205).

In addition to thesegeneswhich havebeenclearlyshownto be involved in the etiology of RCC due totheirability to predisposeto thesetumorsvia germ-linemutations,several other genes,such as PTEN andFHIT, havebeenshownto bealteredin RCC,althoughtheirassociationwith specificsubtypesof RCChasnotbeenelucidated(206–209).Two recentreportssuggest

TableIV. Risk factorsfor RCC

Established Controversial Needingfurther study

Cigarettesmoking Diuretics Phenacetin/acetaminophen/NSAIDSsObesity Otherantihypertensivemedication VegetablesandfruitHypertension Otherurologic diseases Proteinand/orfat dietAcquiredcystic kidney disease Amphetamines Otherdietaryfactors

ReproductivefactorsFamilial andgenetic(e.g.VHL) Occupationalexposures



that molecularetiology may be useful in determiningindividual risk of RCC and responseto therapeuticagents.An analysisof geneticpolymorphismsin genesthat metabolizechemicalcarcinogens,such as thosepresentin tobacco,indicatesan associationbetweenseveralgeneticpolymorphismsand the risk of devel-oping RCC (101). Geneticpolymorphismsare gener-ally associatedwith functionalalterationof enzymaticactivity, and may increasean individual’s risk ofdevelopingRCCasa resultof a carcinogenicexposureby increasingcarcinogenmetabolismor decreasingdetoxification.Thus,geneticpolymorphismshavethepotential to serveas useful screeningtools in earlydetectionor preventionstrategies.In terms of thera-peutic response,a recent study utilizing a seriesofRCC-derivedcell lines indicatesthat the statusof theVHL tumorsuppressorgenecanpredictresponseto thetherapeuticagentC225, an antibodydirectedagainstEGF-R(210).RCCcellslackingawild-typeVHL genefail to undergogrowtharrestin responseto C225duetoan inability to upregulatethe cell-cycle inhibitor p27.In contrast, RCC cells expressing normal VHLupregulatep27 andundergogrowth arrest,suggestingthat VHL gene status in RCC may predict patientresponseto C225 and potentially other therapeuticagentsdesignedto inducecell-cyclearrest.

Alterations associatedwith conventionalRCC. Lossof the chromosomalmaterial at 3p13 → pter is thecharacteristicchromosomalabnormality in conven-tional RCC. This partial monosomyof 3p can occurasa deletion(terminalor interstitial) of only theshortarm (Fig. 5A) or may also involve a proximal regionof the long arm of chromosome3 (Fig. 5D). It mayalso occur as an unbalancedtranslocationbetweenthe 3p 14–21 region and another chromosomealregion(Fig. 5B), with 5q22asthe preferentialpartnerregion (Fig. 5C). In addition to the finding of partialmonosomyof 3p, severalrecurrentcytogeneticaltera-tions associated with progression have beendescribed.Numerical changesthat have beenidenti-fied are monosomyof chromosomes8, 9, 13 and 14andtrisomy of chromosomes12 and20. Non-randomstructural alterationsinvolving 5q, 6q, 8p, 10q, and14q havealso beendescribedin thesetumors (211).In the caseof chromosomes8p, 9p and14q, LOH oftheseregionshasbeendescribedmainly in advancedstages(212).

Most of thebilateralsolitaryRCCshavebeenhisto-logically classifiedas non-papillary.Cytogeneticin-vestigationsin thosecasesreporta differentkaryotypein eachkidney, suggestingan independentpathoge-neticmechanism[for areviewseeDal Cin etal. (213)].

For the genesisof conventionalRCC, deletion oftheVHL geneis regardedashavinganimportantrole.

In thecaseof hereditaryconventionalRCCassociatedwith VHL disease, kidneys of affected patientsfrequently display numerousbilateral renal cysts asprecursorlesions.In onestudyof VHL patients,thesecysts,which arelined with a singlelayer of epithelialcells,wereexaminedfor alterationsin the VHL gene.Allelic lossof the VHL genewasobservedin 25 outof 26 of theselesions,suggestingthat lossof this genewas an early event in diseaseprogressionleading tocystogenesisandthatgeneticalterations,in additiontoloss of VHL function, were required to progresstomalignancy(214). The VHL geneis inactivatedviaseveralmechanisms,including mutationandsilencingby DNA methylation(192).Furthermore,lossof VHLgenefunction correlateswith increasedexpressionofthe angiogenesisfactor vascular endothelial growthfactor, which is overexpressedin vascularizedtumorssuchasRCC andmay be involved in the progressionof RCC (215–218).More recent studies have sug-gestedthat multiple interstitial deletionsin fact occurin the regionof 3p in which the VHL geneis located.Deletionsof chromosome3p segmentswereshowntooccurbetween3p11.2and3p14.1in 98% of conven-tional RCCs(219). The existenceof multiple loci inchromosome3 in RCChasbeensuggestedby vandenBerg and Buys (81), who showedthat allelic loss ofthe VHL and 3p12–p14region was found in adeno-mas,but thatLOH of the3p21regionwasadditionally

Fig. 5. Partial G-bandedkaryotypesshowing a few examplesofmonosomyof chromosome3p in conventionalRCC: (a) del(3)(p13); (b) der(3)t(2;3)(p12;p13);(c) der(3)t(3;5)(p14;q22); (d) der(20)t(3;20)(q12;p13).



observedin associationwith malignantdevelopmentof conventional RCC. A number of studies haveshown that additional genetic alterations, includingLOH at 14q24–2, 8p12–21.1 and 9p21, are alsoassociatedwith more malignant behavior of thetumors(212,220).

In contrastto VHL alterationswhichappearto occurvery early in the developmentof RCC, alterationsinthep53tumorsuppressorgeneappearto bea lateeventin diseaseprogressionassociatedwith malignancyand/or metastasis(221–227).Similarly, alterationsin p15/16 tumor suppressorgeneswhich inhibit cell-cycleprogressionarefoundin RCC-derivedcell linesbutnotprimarytumors,suggestingthatthesealterationsoccurin cells as a function of adaptationto autonomousgrowth in vitro (228).

Alterations associatedwith papillary RCC. In papil-lary tumors, a combination of numerical changes(sex chromosomeloss with trisomy or tetrasomyofchromosome7 and trisomy of chromosome17) isfrequently found as the only karyotypic alteration(Fig. 6). This combinationof cytogeneticalterationsoccursirrespectiveof tumor size and grade,withoutinvasiveand metastaticgrowth, and thereforeidenti-fies a class of papillary tumors that is consideredbenign and should be called papillary adenoma(229,230). It was arguedthat simultaneouschromo-some7 and 17 gain with or without Y chromosomeloss characterizesa distinct entity, the metanephricadenoma,suggestingthat thesetumors may be pre-cursorsof the papillary RCC (231).However,Stummet al. (232) recentlysuggestedthe role of a possibletumor suppressorgeneon the short arm of chromo-some2 in the developmentof metanephricadenoma.

When,however,additionaltrisomiesof chromosomes12, 16 and20 appearthey areassociatedwith a moreaggressivebehavior and with progression to thepapillary carcinoma stage (Fig. 7) (233). In thesecasestrisomy of chromosomes3/�3q is also fre-quentlyfound.

Anothersubgroupof papillary tumorshasalsobeenidentified basedon their cytogeneticprofile, i.e. Xautosometranslocations(234). The most frequentlyreportedtranslocationpartnerof theXp11.2bandis the1q21 band,followed by anotherchromosome1 band(1p34). In three casesa different chromosomere-arrangementinvolving theXp11 regionwasobserved,i.e. t(X;17)(p11.2;q25),at (X;10)(p11.2;q23)and adel(X)(p11).Thegenesinvolvedin theclassicalt(X;1)in these tumors have recently been identified. Thetranslocationresults in a fusion of the transcriptionfactor TFE3 on the X chromosometo a novel gene,designatedPRCC,on chromosome1 (235). Interest-ingly, the agedistribution of thesecasesis strikinglydifferent from that of kidney tumorsin general.Morethan 50% of the patientsare youngerthan 45 years,with a striking male predominance(ratio 4:1). Histo-logically, most of the caseshad a papillary growthpattern,with someclear cell-like featuresdue to thedeposition of fat and glycogen, which may haveresultedin a misdiagnosisof conventionalRCC byotherinvestigators.Thereareno dataon theprognosisof this subsetof papillary RCC. The finding in onestudy of metastases31 yearsafter nephrectomyin apatientwith this typeof papillaryRCC(234)couldbean isolatedoccurrence,andits significanceis unclear.Bilateralmultifocal RCCswerealwaysof thepapillarytypeandthekaryotypesshowedmoreor lessthesamenumericalchanges,with trisomiesin different combi-

Fig. 6. G-bandedkaryotypeof a papillary adenomashowingthe characteristiccombinationof tetrasomyofchromosome7, trisomy ofchromosome17 andlossofthe Y chromosome.



nationsin tumorswithin thesamekidneyaswell asintheoppositekidney (213).

Alterationsassociatedwith chromophobeRCC.Cyto-genetics,comparativegenomichybridization (CGH),microsatelliteand DNA cytometric analysissuggestthat a combinationof monosomiesof chromosomes1, 2, 6, 10, 13, 17 and 21 characterizesthis typeof RCC (34,35,236,237) (Fig. 8). Correct geneticcharacterizationis not trivial for chromophobeRCCas the two variants,the typical one and the eosino-philic one, may be misdiagnosedas clear-cell RCCand oncocytoma,respectively.However, these twoothertypesof renaltumorexhibit different chromoso-mal patterns: loss of 3p material in conventionalRCCs; and at least three different cytogeneticsub-groupsin oncocytomas,i.e. thosewith coincidentlossof chromosomesY and 1, thosewith involvementof11q13 (238) and those in which 12q13 is affected(239) (Fig. 9).

Alterations associatedwith collecting-duct carcino-ma. Dataconcerningthe cytogeneticabnormalitiesofthis tumor are limited to seven cases (240–242).Conflicting findings have been reported, with bothmonosomies(241) and trisomies (240,242) beingdescribed.Molecularstudiesof this particulartype ofkidney tumor havesuggestedfrequentlossof hetero-zygosityof 8p and13q,andno lossof materialof theshortarm of chromosome3 (243,244).

Chromosomalabnormalitiesin non-neoplasticrenaltissue. Several cytogenetic investigationshave de-monstratedthe presenceof chromosomeaberrationsin non-neoplasticrenal tissue surroundinga RCC.Trisomy of chromosome7, trisomy of chromosome

10 and lossof the Y chromosomewere the most fre-quentchangesobservedeitherby classicalor molecu-lar cytogenetics,i.e. fluorescencein situ hybridization.Cells carrying trisomy of chromosome7 as the solechromosomeaberrationhave been found in severaltypesof tumor, mostly of epithelial origin (245). Thesignificanceof this finding, aswell aswhich cell typegives rise to trisomy of chromosome7 in non-neoplasticparenchymaof RCC-bearingkidneys, isstill unknown.Epithelial cells (246) and cells of theimmunesystem(247)havebothbeenimplicated.

Experimentaldata on alterationsin preneoplasticle-sions.An animal model for RCC hasbeendescribedwhich recapitulatesmanyof the characteristicsof thehumandisease.The Eker rat (248) exhibits a heredi-tary predispositionto developRCC due to germ-lineinactivation of the TSC-2 tumor suppressorgene(249–251). The availability of this animal modelpresentsmany opportunitiesfor observingthe earlyevents associatedwith RCC, as both tumors andpreneoplasticlesions can be readily obtained fromtheseanimals.Overexpressionof transforminggrowthfactor a (TGF-a) and its cognatereceptor,the EGFreceptor,is a ubiquitouspropertyof RCCsof severalhistologic types. Using the Eker rat model, over-expressionof TGF-a wasfound to occurin adenomasand dysplasias,the earliesthistologically identifiableprecursorlesions for renal tumors in rodents(252).This indicates that overexpressionof this growthfactor is an early event in the developmentof thesetumors.

Conclusion

Duringthelastfew decadesnewinsightsinto cyto-and

Fig. 7. G-bandedkaryotypeof a papillary RCC showingthe presenceof trisomiesofchromosomes3 and12, inaddition to tetrasomyofchromosome7, trisomy ofchromosome17 andlossofthe Y chromosome.



molecular geneticshave recognizedcytogeneticandmolecularcorrelatesfor differentRCCtypesbasedoncertaingeneticalterations.Four different RCC tumortypes have been identified—conventional,papillary,chromophobe and collecting-duct carcinoma—inwhich translocationsor non-specificchangesoccurwith well-documentednumericalchanges(Table V).Concerningprecursorlesionsof RCC no significantreportsof molecularalterationsare available.As ourknowledgebaseregardingthe molecularetiology ofRCC increasescommensuratelywith increasedunder-standing of the mechanismof action of potentialtherapeuticstrategies,thereis reasonto be optimisticthat the future will seeimprovementsin our ability toprevent,diagnoseandtreatthesetumors.Similarly, ofthe known molecularand cytogeneticalterationsthat

have been identified for RCC, none has yet beendemonstratedto haveutility for early detection,as abiomarker of environmentalexposureor to predictresponseto therapeutic agents. Future studies tocorrelatethesealterationswith diseaseprogressionorprognosismaybeusefulto identify whichof themmayfind utility asbiomarkersfor RCC.

CLINICAL SIGNIFICANCE OF PREMALIGNANTLESIONSAND IMPACT ON CLINICALMANAGEMENT

Introduction

Malignant transformation is a multistage processinvolving initiation andpromotion.Theseeventsoccur

Fig. 8. G-bandedkaryotypeshowingthe low chromosomenumberin chromophobeRCC.

Fig. 9. PartialG-bandedkaryotypesshowingthe mostcharacteristicstructuralchromosomalchangesfoundin oncocytomas.



within the epithelium over a period of time. It isbelievedthatmostcancershavea latencyperiodof 10–20 years. However, our understandingof all theseaspectsof cancerbiology is still incomplete.Identifica-tion of early changesin cancermay lead to earlierdiagnosisandbetteroutcome.

The traditional treatmentfor RCC is radical neph-rectomy.RIN, which co-existsin thetumor-containingkidney, is not of clinical significancein such casesinsofar as it is removedwith the tumor. The majorclinical significanceof this lesionis in patientswhosetumors are managedby partial nephrectomy,wherepre-malignantdiseasemayremainin therenalremnant.In this review, the implications of RIN in patientsundergoingpartial nephrectomywill besummarized.

Nephron-sparingsurgery.Multiple reportsfrom manycentersthroughout the world confirm that nephron-sparingsurgery(NSS) hasbeenwidely implementedfor the managementof small renal tumors over thelast few decades(253–264).Thereareseveralreasonsfor this (265). The increasing use of abdominalultrasoundand other diagnostic techniqueshas re-sultedin a stagemigration phenomenon(266). ManyRCCs are now diagnosedincidentally, prior to theonset of symptoms. Frequently, these tumors areamenableto a nephron-sparingapproach(264). Pa-tientsoftenpreferthis approachwherepossible,partlybecausea proportionof renaltumorsarebenign(266–269). In the absenceof an accuratemeansto identifythese preoperatively, a nephron-sparingapproachavoids the risk of a nephrectomyfor benign disease(270). This approachalso dramatically improvesthetreatment options available for those patients whosubsequentlydevelopa contralateraltumor.

The most cogent rationale for the use of partialnephrectomyis the largefavorableclinical experiencereportedafter initial reportsin the early 1990s(258,259,262) on the outcome of this operation whenperformed electively, i.e. in the case of a normalcontralateralkidney. Theseseriesdemonstrateexcel-lent local control with a nephron-sparingapproach

(254,259,261,271–273).Interpretationof thesedatamust take into accountthe issuesof patientselectionandindicationsfor NSS.Historically, partial nephrec-tomy was associatedwith a 10% risk of local recur-rence.This reflectedthe experiencewith non-electiveindications(patientswho hadbilateraltumors,chronicrenal failure, a solitary kidney or VHL). In contrast,thereportedlocal recurrencefollowing electivepartialnephrectomyis dramaticallylower (274).

Table VI summarizesthe results of single-centerstudies of elective NSS in patients with a normalcontralateralkidney.Of 418patientsmanagedthusfar,three patients have had a local recurrence(0.7%).Remarkably, this is lower than the incidence ofcontralateral disease.The series with the longestfollow-up (10 years) reported a 6% incidence ofcontralateraldiseaseand a 1% incidence of localrecurrence(254). Theseexcellent resultsrepresentapowerfulargumentin favor of conservativesurgery,atleastwith respectto cancercontrol.

The multifocal natureof RCC is well documented(274).The overall risk of unexpectedmultifocality ofRCChasbeenreportedto beashigh as25%,althoughin a surgicalsettingit is only �6%. This multifocalitycan be responsiblefor ipsilateral tumor recurrenceinthe remnant kidney. Thus there is a fascinatingdiscrepancybetween the predictable likelihood ofleaving multifocal diseaseor dysplasiabehind in arenal remnant(6–25%),and the observedrecurrencerate in theseremnants(0.7–1%). There are severalpossiblereasonsfor this. Theunknownnaturalhistoryof small secondary lesions may account for thisdiscrepancy.Somesecondarylesionsthat are left inthe renal remnant after NSS do not evolve to aclinically detectabletumor.This may be analogoustothe well-known phenomenonof the spontaneousregressionof metastasesafter tumor nephrectomy,presumablyimmunologicallymediated.Cytotoxic im-munocytes could induce the regression of smallsatellitelesions(274).Anotherexplanationis that theinherentnaturalhistory of theselesionsis benign,i.e.that they behavelike renaladenomas.This is particu-

TableV. Molecular alterationsin RCC: correlation with stage

Alteration RCCvariant Stage Other

VHL Conventional Early Predisposingalterationin hereditaryRCC, lost in cystsassociatedwith VHL syndrome

TGF-a All types Early Overexpressedin dysplasiasandadenomasc-MET Papillary Early Predisposingalterationin hereditarypapillary RCCTrisomy/tetrasomyof chromosomes7 and17 Papillary Early Associatedwith papillary adenomasp53 Late Low-frequencyalterationassociatedwith advanceddiseasep15/16 Not known Late Alterationsfound in RCC-derivedcell lines but not primary

tumorsChromosome8p, 9p and14q Conventional Late Associatedwith advancedstagesof RCCTrisomy of chromosomes12, 16 and20 Papillary Late Associatedwith progressionto carcinoma



larly true for small, well-differentiatedtumors,whichare referred to by Thoeneset al. (23) as being “ofdoubtfulbenignity”. Finally, it is possiblethat the truelocal recurrencerateis higherthanhasbeenreported,but that the worseseriesreflectingtheseresultshavenot beenpublished(publicationbias).

Theseissuesarealso relevantto understandingthenatural history of premalignant lesions includingdysplasia.The natural history of IED must also beinterpretedin thelight of thelow rateof localrecurrencereportedin NSSseries.It is likely that,besidessatellitetumors,IED waspresentin therenalremnantof someofthe 418 patientsmanagedby partial nephrectomyintheseseries.The absenceof progressionsuggeststhatIED hasa benignphenotype.Themedianfollow-up ofthereportsto datevariesfrom 30 to 120months(range12–228months)andit might bethat theprogressionissufficientlyslow thatit hasnotmanifesteditself withinthe rangeof follow-up of thesestudies.Serieswithlongerfollow-up shouldclarify this question.

Is biopsyof the remaining/contralateralkidneyjustified?

Elective NSS with a normal contralateral kidney.Kidneys being subjectedto partial resectionmust becompletelymobilized and the entire kidney carefullyinspectedfor satellite lesions or multifocal tumors.Any suspiciousareasshould be biopsied. The pre-sence of a positive resection margin is usuallyconsideredan indication for nephrectomy in thissetting.The presenceof manysmall tumorsis alsoanindication for nephrectomy.A kidney that has de-monstrated the phenotype of diffuse multifocalityshould be removed completely if the contralateralkidney is normal, as the risk of a subsequentrecurrenceoutweighsthe benefitof nephronsparing.In the faceof limited multifocality, i.e. the finding ofa singlesmall tumor found elsewherein the kidney atthe time of surgery, local excision of the secondtumor is appropriate if possible. In between the

situation of a secondsmall tumor amenableto localexcision and diffuse multifocality requiring a com-pletenephrectomythereis a grayzonewheresurgicaljudgment is required.Clearly, the low rate of localrecurrencedoesnot justify a routinepolicy of biopsyof thepartof thekidney with a normalappearance.

Biopsy of the contralateral normal kidney. Biopsiesof a contralateralkidney with a normal appearanceare not indicated, even in the presenceof multi-focality in the other kidney. There are three reasonsfor this:

1. The rate of contralateraltumors is reportedto bebetween1 and6%. Thustheyield is low;

2. In theabsenceof anidentifiabletarget(i.e. a mass),thereis a substantialrisk of missingthe pathologywith needlebiopsiesevenin the presenceof RIN.Multiple systematicbiopsycoreswouldprobablyberequired. This creates a considerable risk ofhemorrhageandothercomplications;

3. Thenaturalhistoryof RIN is not known;hencetheoptimalmanagementof a patientwith this lesionisuncertain.However,onecanpredictthereactionofapatientwith ahistoryof RCCononesidewhowasfound to have RIN on the contralateralside by arandom biopsy. Even with conservativemanage-ment, the attendantanxiety engenderedby thisfindingwouldhaveasubstantialnegativeimpactonthe patient’s quality of life. In contrast, regularfollow-up imagingof the contralateralkidney withultrasoundor computerizedtomography(CT) willidentify a new lesionat a point in its developmentwhere the likelihood of cure is still high. Thusbiopsiesof a normalcontralateralkidney to detectRIN arenot justified.

Non-elective NSS. In this setting, with a solitarykidney or bilateral tumors,nephronsparingis impor-tant in order to minimize the risk of renal insuffi-ciency.As such,the finding of dysplasiaat biopsyof

TableVI. Resultsof partial nephrectomywith a normal contralateralkidney

Reference Numberof patients Meansizeof tumor (cm) Follow-up (months) Disease-freesurvival (%) Local recurrence

260 20 3.4 31 90 0259 19 2.6 35 100 0308 17 3.6 38 100 0309 15 3.5 40 100 0310 72 3.2 40 90 2 (2.8%)271 105 4.0 42 100 0256 20 3.1 46 100 0311 10 3.7 48 100 0312 19 3.34 70 96 0273 51 3.0 78 98 0254 70 3.0 120 97 1 (1.3%)Total 418 3 (0.7%)



a kidney with an otherwisenormalappearancewouldnot be an indication for further resection,as thismight further compromiserenal function. This ap-proachwould be alteredin the face of data demon-stratingan aggressivenaturalhistory associatedwithRIN. In theabsenceof suchdata,biopsiesof a kidneywith a normalappearancecannotbe justified.

Biopsy performed for other medical problemsshowingRIN. Patientsfrequentlyhavea renalbiopsyfor evaluationof suspectedrenalparenchymaldisease.The indications for biopsy are azotemia,proteinuriaor active castsin the urinary sediment.Occasionally,thesebiopsiesmay show an unsuspectedfinding ofRIN. This finding is usuallyincidental,insofarasRINis generally focal and unlikely to have clinicalmanifestations.This finding will often produceanxi-ety on the part of the patient and uncertaintyin themind of the physician.The natural history data thatexist suggestthat this lesion usually progressesto aninvasivephenotypeslowly, if at all. Thusconservativemanagementis generally called for, with periodicimaging of the affectedkidney at follow-up. Thera-peuticinterventionfor this lesionis not warranted.

Follow-up of the remnant/contralateralkidney.Localrecurrence,even in the face of suspectedor knownRIN, is uncommon.The mediantime to recurrenceisapproximately 4 years (254,261). Substantialdataindicatethat RCCs<2.5cm in diameterrarely metas-tasize.Thuspatientswho havehadNSSfor RCC,or adiagnosisof RIN of the kidney, must be followedclosely for a prolonged period of time with serialimagingstudies.For mostpatients,renalultrasoundissufficientto detecta newlesion1–2cm in diameter.

The reportedfollow-up scheduleafter NSS varies.Oneguidelineproposedno monitoring for T1, yearlychestX-ray and abdominalCT every 2 yearsfor T2,andCT every6 monthsfor 2 yearsfor T3 (275). It isappropriateto modify this scheduleaccordingto thelikelihood of recurrencebasedon pathologiccriteria.For example, a patient with known multifocality/bilaterality or high-grade tumor would warrant amore intensive follow-up regimen.A suggestedfol-low-up scheduleafter NSS consisting of physicalexamination,goodquality ultrasoundor CT, urinalysisandserumcreatininedeterminationis asfollows: every6 monthsfor years1–3 and annually thereafter.Thepresentdata do not allow one to identify a point atwhich patientsareno longerat risk.

Conclusion

It is possiblethatpatientswith RIN areat increasedriskof multifocal tumors and/or local or contralateralrecurrenceof NSS. However, given the extremelylow rateof local recurrencefollowing NSSandthelack

of dataon thenaturalhistoryof this entity, thefindingof RIN shouldnot substantiallyalter the managementof RCC.

PREVENTION,SCREENINGAND EARLYDETECTION

Introduction

It is well establishedthat most renal parenchymaltumorsarisefrom epithelialcells of the tubulesof thekidney. This is true both for RCC and the so-calledrenal adenomas(38,276). Because there are nomorphologicalcriteriawhichcanbeusedto distinguishrenal adenomasfrom RCC and becauseeven smallglandulartumorswith a benigncytologicalpicturecanprovetheir aggressivenessby infiltrating thesurround-ing tissueor by metastasizing,it hasbeenclaimedthatsmall renal cortical glandular neoplasmsshould beregardedassmallRCCs,eventhoughtheir potentialtometastasizeis minimal (277).It is alsowell knownthat,consideringall cortical glandular tumors, there is adirect relationshipbetweentumor size and frequencyof local aggressivenessin termsof capsularinvasion,vasculargrowthandmetastases(65).Previouslyit wasproposedto considercortical tumorswith a diameter<3cm asbenignadenomasasthey rarely metastasize(278).However,it is well knownthatevensmall renaltumors with a diameter <2cm may definitely bemalignantwith a potentialfor metastaticspread(258,279). Small, clinically unrecognized,renal epithelialtumors are common incidental findings in patientsoperatedon becauseof clinically diagnosedRCC(280,281). This has to be consideredin patientsundergoingNSS as, if overlooked,suchtumorsmayimply an increasedrisk of local recurrences(282).However, the frequency of such recurrencesasreported in most series is even lower than theoccurrenceof small,often multifocal, cortical tumors.This indicatesthat manyof thesetumorsmay remaindormant for a long time period (273). It should,however, be underlined that the natural history ofsuch small single or multifocal renal tumors isundeterminedat the presenttime (283). Furthermore,it is notyet resolvedto whatextentRCCarisesdenovofrom the renal tubular cell or by evolution ofadenomatoushyperplasiaandrenal cortical adenoma.It is alsonoteworthythat adenomatoushyperplasiaoradenomahave been found in as many as 14% ofpatientswith RCC(284)andthatsmallcortical tumorscanbe discoveredin up to 20% of autopsiesin adults(285,286).

Chemoprevention.Chemopreventionrefers to theprevention of cancer or the reduction of risk insusceptibleindividualsby the administrationof natur-



al or syntheticdrugswith minimal toxicity, in orderto suppress,delay or reversecarcinogenesis(287).Chemopreventionmight be particularly useful inpatientswith an inheritedrisk of RCC, suchas thosewith VHL syndromeor familial RCC. Patientswhohave undergoneradical nephrectomyfor RCC andwho are at risk of recurrencebecauseof locallyadvanceddiseaseor lymph node metastasismightalso be consideredgood candidatesfor preventiveapproaches(288,289).

Evidenceof the effectivenessof chemopreventionagainstRCC is supportedmainly by epidemiologicalstudies (vitamins A, C and E) (290–293). Recentanimal studieshavefurther suggestedthat vitamin Dhasantiproliferativeactivity in RCC(294,295).How-ever,thebenefitof thesefactorsneedsto beconfirmedin prospectiverandomizedtrials.

Screening.It hasbeenshownthat patientswith inci-dentalRCC havetumorswith smallersize and lowerstage and that their prognosis is more favorablecomparedwith that of symptomaticpatients(296).Sofar only a few studieshave beenperformedon theeffect of massscreeningwith the purposeto detectRCC (297–301).The largestof suchscreeningtrialsfocusedon RCC aloneand investigatedthe practica-bility and effectivenessof a systematicscreeningforRCC by ultrasound in two German centers. Thescreening procedure was divided into two phasesover the course of 1 year. In the first phase allcitizens>40 yearsof agewere invited to voluntarilyattend a cost-free ultrasound investigation. Thesecondphasewas a follow-up investigationfor theentire screeningpopulation. The screeningprogramrecruited 10000 volunteers; the mean age was 60yearsand13 RCCs(0.13%)weredetected.The inci-denceof other findings, such as renal or liver cystsand stonesin the urinary or biliary tract, was 15%,andnoneof thesefindingsrequiredfurther treatment.The authorsconcludedthat systematicscreeningiscost-effective and highly acceptable, despite theextremelylow detectionrate (297). Theseresultsarepreliminary and a thoroughcost-effectiveanalysisisstill awaitingpublication.

The introduction of screeningprogramsrequiresethical,scientificand,if possible,financialjustification.The validation for screeningproceduresimplies theexistenceof conclusiveevidencethat screeningcanalter the natural history of diseasein a “significantproportion” of those screened (302). Ideally thehypothesesshouldbe testedby randomizingeitherthescreening procedure or the therapy subsequenttoscreening.

Consideringthestudiesperformedsofar, thereis notenoughevidenceto recommendmassscreeningof a

largepopulationfor thedetectionof RCCat thepresenttime.Thebarriersto a randomizedtrial of screeningina healthy populationare substantial,and include therequirementfor a hugecohortsize,prolongedfollow-up and a tremendousfinancial burden. In contrast,screeningfocused on certain risk groups has beenproposed,suchason patientswith VHL syndromeandESRDandafter renaltransplantation(146,303–306).

Early detection.A numberof studieshaveshownthatby investigationof patientswith asymptomaticmicro-hematuriait is possibleto detect a few caseswithmalignant urologic tumors. However, the detectionrate of RCC is low and therefore the rationale forsuch investigations is questionable.An importantissuein early detectionof RCC is the effect of theincreasinguseof ultrasoundand CT in patientswitha history of abdominalsymptoms.This has resultedin an increasingrate of detectionof incidental renaltumors during the last 10–20 years. Such tumorscomprisenearly50%of newly detectedRCCs.

Conclusion

The preventionof RCC or reduction of the risk ofdevelopingthediseasein susceptibleindividualsby theadministrationof natural or synthetic drugs is sup-portedprimarily by animalandepidemiologicstudies.A greatdeal of work remainsto be donein order toconfirmthebenefitof theproposedfactorsin prospec-tive randomizedtrials.

Some studies have investigatedsystematicmassscreeningof the population >40 years of age withultrasound,in an attempt to detectRCC at an earlystage.At thepresenttime,theresultsof thesescreeningtrials do not fulfil the criteria for the introductionof amassscreeningprogram.In contrast,screeningof high-risk groups,suchaspatientswith VHL syndromeandESRD and following renal transplantation,is war-ranted.Theincreaseduseof abdominalultrasoundandCT hasresultedin stagemigration,with an increasingrateof earlier-detectedRCC, i.e. incidentaloma.Suchtumors are detected at a smaller mean size thansymptomaticRCC.

Furtherwork to increasetheknowledgeof thecausalfactors is needed to establish successful primaryprevention of RCC, for identification of effectivechemopreventivefactorsand for screeningof better-definedrisk groups.

REFERENCES

1. Black RJ, Bray F, Ferlay J, Parkin DM. Cancerincidenceandmortality in theEuropeanUnion: cancerregistry data and estimatesof national incidencefor1990.Eur J Cancer1997;33: 1075–107.



2. Parkin DM, Pisani P, Ferlay J. Estimates of theworldwide incidenceof 25 major cancersin 1990.IntJ Cancer1999;80: 827–41.

3. VanPoppelH, VandendriesscheH, BoelK, MertensV,Goethuys H, HaustermansK, et al. Microscopicvascularinvasion is the most relevantprognosticatorafter radical nephrectomyfor clinically nonmetastaticrenalcell carcinoma.J Urol 1997;158: 45–9.

4. Motzer RJ, Bander NH, Nanus DM. Renal-cellcarcinoma.N Engl J Med 1996;335: 865–75.

5. Vogelzang NJ, Priest ER, Borden L. Spontaneousregressionof histologically proved pulmonary meta-stasesfrom renal cell carcinoma:a casewith 5-yearfollowup. J Urol 1992;148: 1247–8.

6. PapacRJ. Spontaneousregressionof cancer.CancerTreatRev 1996;22: 395–423.

7. Dineen MK, Pastore RD, Emrich LJ, Huben RP.Resultsof surgical treatmentof renal cell carcinomawith solitary metastasis.J Urol 1988;140: 277–9.

8. Golimbu M, JoshiP, SperberA, TesslerA, Al-AskariS, Morales P. Renal cell carcinoma: survival andprognosticfactors.Urology 1986;27: 291–301.

9. Maldazys JD, deKernion JB. Prognostic factors inmetastaticrenalcarcinoma.J Urol 1986;136: 376–9.

10. Middleton RG. Surgery for metastatic renal cellcarcinoma.J Urol 1967;97: 973–7.

11. WagleDG, ScalDR. Renalcell carcinoma—areviewof 256 cases.J SurgOncol 1970;2: 23–32.

12. Cerfolio RJ, Allen MS, DeschampsC, Daly RC,Wallrichs SL, TrastekVF, et al. Pulmonaryresectionof metastaticrenal cell carcinoma.Ann Thorac Surg1994;57: 339–44.

13. Van PoppelH, BaertLV. Palliativesurgicaltheraphy.In: PetrovichZ, BaertL, BradyLW, eds.Carcinomaofthe kidney and testis,and rare urologic malignancies.Innovationsin management.Berlin: Springer-Verlag,1999:95–109.

14. Figlin RA. Renal cell carcinoma: managementofadvanceddisease.J Urol 1999;161: 381–6.

15. MotzerJR,VogelzangNJ.Chemotherapyfor renalcellcarcinoma.In: RaghavanD, SherHI, Leibel S, LangeP, eds. Principles and practice of genitourinaryoncology.NewYork: Lippincott-Raven,1997:885–96.

16. Bostwick DG, Eble JN, Murphy G. Conferencesummary: Diagnosis and prognosis of renal cellcarcinoma.Cancer1997;80: 975–6.

17. Van PoppelH, BeckersG, BaertL. Prognosticfactorsin cancer.UICC. Secondedition.M. Gospodarowiczetal. (Eds).Wiley, New York 2001(in press).

18. YagodaA, Abi-RachedB, PetrylakD. Chemotherapyfor advancedrenal-cellcarcinoma:1983–1993.SeminSurgOncol 1995;22: 42–60.

19. Murphy WM, Beckwith JB, Farrow GM Washington,DC: Armed ForcesInstituteof Pathology,1994.

20. True LD, Grignon D. Pathologyof renal cancers.In:Raghavan D, Sher HI, Leibel S, Lange P, eds.Principles and practice of genitourinary oncology.New York: Lippincott-Raven,1997:799–811.

21. BosniakMA, BirnbaumBA, Krinsky GA, WaismanJ.Small renal parenchymalneoplasms:further observa-tions on growth.Radiology1995;197: 589–97.

22. Ebley JN, Warfeld KA. Precursorlesionsand humanrenal cortical epithelial neoplasia.J CancerRes ClinOncol 1990;116(Suppl):1228.

23. ThoenesW, Storkel S, Rumpelt HJ. Histopathology

and classification of renal cell tumors (adenomas,oncocytomasand carcinomas).The basic cytologicalandhistopathologicalelementsand their usefor diag-nostics.PatholResPract1986;181: 125–43.

24. KovacsG, Akhtar M, Beckwith BJ, BugertP, CooperCS,DelahuntB, et al. TheHeidelbergclassificationofrenalcell tumours.J Pathol1997;183: 131–3.

25. Fleming S, O’Donnell M. Surgicalpathologyof renalepithelial neoplasms:recent advancesand currentstatus.Histopathology2000;36: 195–202.

26. KovacsG, Welter C, Wilkens L, Blin N, DeRieseW.Renaloncocytoma.A phenotypicandgenotypicentityof renalparenchymaltumors.Am J Pathol1989;134:967–71.

27. LehmannHD, BlessingMH. Renaloncocytoma(path-ology, preoperativediagnosis,therapy).ProgClin BiolRes1982;100: 589–96.

28. PearseHD, HoughtonDC. Renaloncocytoma.Urology1979;13: 74–7.

29. GnarraJR,Tory K, WengY, SchmidtL, Wei MH, LiH, et al. Mutationsof theVHL tumoursuppressorgenein renalcarcinoma.Nat Genet1994;7: 85–90.

30. Foster K, Prowse A, van den Berg A, Fleming S,HulsbeekMM, CrosseyPA, etal. Somaticmutationsofthevon Hippel-Lindaudiseasetumoursuppressorgenein non-familial clear cell renal carcinoma.Hum MolGenet1994;3: 2169–73.

31. KovacsG. Molecularcytogeneticsof renalcell tumors.Adv CancerRes1993;62: 89–124.

32. KovacsG. The valueof moleculargeneticanalysisinthe diagnosisand prognosisof renal cell tumours.World J Urol 1994;12: 64–8.

33. DelahuntB, Eble JN. Papillaryrenalcell carcinoma:aclinicopathologicand immunohistochemicalstudy of105 tumors.Mod Pathol1997;10: 537–44.

34. Bugert P, Gaul C, Weber K, HerbersJ, Akhtar M,LjungbergB, et al. Specificgeneticchangesof diag-nostic importancein chromophoberenal cell carcino-mas.Lab Invest1997;76: 203–8.

35. BugertP, Gaul C, WeberK. Specificgeneticchangesof diagnosticimportancein chromophobecarcinoma.Am J SurgPathol1997;21: 1185–95.

36. FlemingS, Lewi HJ. Collectingduct carcinomaof thekidney.Histopathology1986;10: 1131–41.

37. RumpeltHJ,StorkelS,Moll R, ScharfeT, ThoenesW.Bellini duct carcinoma:further evidencefor this rarevariant of renal cell carcinoma.Histopathology1991;18: 115–22.

38. OberlingC, Riviere M, HaguenauF. Ultrastructureoftheclearcellsin renalcarcinomaandits importanceforthe demonstrationsof their renal origin. Nature1960;186: 402.

39. Delahunt B, Thornton A. Renal cell carcinoma.Ahistoricalperspective.J Urol Pathol1996;4: 31–49.

40. Horning ES, Whittick JW. The histogenesis ofstilboestrol-inducedrenal tumors in the male goldenhamster.Br J Cancer1954;8: 451–7.

41. Goldfarb S, Pugh TD. Morphology and anatomiclocalization of renal microneoplasmsand proximaltubuledysplasiasinducedby four differentestrogensinthe hamster.CancerRes1990;50: 113–9.

42. Lombard LS, Rice JM, Vesselinovitch SD. Renaltumors in mice: light microscopic observationsofepithelial tumors inducedby ethylnitrosourea.J NatlCancerInst 1974;53: 1677–85.



43. Mancilla-JimenezR, StanleyRJ, Blath RA. Papillaryrenalcell carcinoma:a clinical, radiologic,andpatho-logic studyof 34 cases.Cancer1976;38: 2469–80.

44. Bernstein J, Evan AP, Gardner Jr KD. Epithelialhyperplasiain humanpolycystic kidney diseases.Itsrole in pathogenesisandrisk of neoplasia.Am J Pathol1987;129: 92–101.

45. Amin MB, Corless CL, RenshawAA, Tickoo SK,Kubus J, Schultz DS. Papillary (chromophil) renalcell carcinoma:histomorphologiccharacteristicsandevaluationof conventionalpathologicprognosticpar-ametersin 62 cases.Am J SurgPathol1997;21: 621–35.

46. KennedySM, Merino MJ, LinehanWM, RobertsJR,RobertsonCN, NeumannRD. Collecting duct carci-nomaof the kidney. Hum Pathol1990;21: 449–56.

47. Yorukoglu K, Aktas S, Mungan MU, Kirkali Z.Tubular dysplasiaand carcinomain situ: precursorsof renalcell carcinoma.Urology 1999;53: 684–9.

48. Mourad WA, NestokBR, SalehGY, SolezK, PowerRF, Jewell LD. Dysplastic tubular epithelium in“normal” kidney associatedwith renalcell carcinoma.Am J SurgPathol1994;18: 1117–24.

49. DoddLG, LayfieldLJ. Renalepithelialdysplasia.Am JSurgPathol1995;19: 1335–7.

50. Lai R, el Dabbagh L, Mourad WA. Mutant p53expressionin kidney tubules adjacent to renal cellcarcinoma:evidenceof a precursorlesion.Mod Pathol1996;9: 690–5.

51. Nochomovitz LE, JohanssonSL, Merino M, LeslieKO. Recommendationsfor the reporting of resectedneoplasmsof the kidney. Associationof DirectorsofAnatomical and Surgical Pathology.Virchows Arch1996;428: 71–3.

52. Eble JN. Recommendationsfor examiningandreport-ing tumor-bearing kidney specimens from adults.SeminDiagn Pathol1998;15: 77–82.

53. Choyke PL, Glenn GM, Walther MM, PatronasNJ,Linehan WM, Zbar B. von Hippel-Lindau disease:genetic, clinical, and imaging features. Radiology1995;194: 629–42.

54. Solomon D, Schwartz A. Renal pathology in vonHippel-Lindaudisease.Hum Pathol1988;19: 1072–9.

55. Walther MM, Lubensky IA, Venzon D, Zbar B,Linehan WM. Prevalenceof microscopic lesions ingrossly normal renal parenchymafrom patientswithvon Hippel-Lindau disease,sporadicrenal cell carci-nomaandno renaldisease:clinical implications.JUrol1995;154: 2010–4.

56. Ishikawa I, Kovacs G. High incidence of papillaryrenal cell tumours in patientson chronic haemodia-lysis. Histopathology1993;22: 135–9.

57. FryerAA, JonesPW. Interactionsbetweendetoxifyingenzymepolymorphismsand susceptibility to cancer.IARC Sci Publ 1999;148: 303–22.

58. Hughson MD, Schmidt L, Zbar B, Daugherty S,Meloni AM, Silva FG, et al. Renalcell carcinomaofend-stagerenaldisease:a histopathologicandmolecu-lar geneticstudy.J Am SocNephrol1996;7: 2461–8.

59. Storkel S, Eble JN, Adlakha K, Amin M, Blute ML,Bostwick DG, et al. Classificationof renal cell carci-noma:WorkgroupNo. 1. Union InternationaleContrele Cancer(UICC) and the AmericanJoint Committeeon Cancer(AJCC).Cancer1997;80: 987–9.

60. HennW, WelterC, Wullich B, ZangKD, Blin N, Seitz

C. Chromophoberenalcell carcinomaandrenalonco-cytoma.A cytogeneticand cytological comparison.JUrol Pathol1993;1: 145–55.

61. HesO, Michal M, SkalovaA. RenaloncocytomaswithHale’scolloidal iron positivecytoplasm.J Urol Pathol1997;7: 177–87.

62. Latham B, Dickersin GR, Oliva E. Subtypes ofchromophobecell renal carcinoma:an ultrastructuralandhistochemicalstudyof 13 cases.Am J SurgPathol1999;23: 530–5.

63. Storkel S. Chromophoberenal cell carcinoma(differ-ential diagnosiswith clear/glanularcell carcinomaandoncocytoma.Rev EspPathol1999;32: 381–2.

64. Parkin DM. The global burden of cancer. SeminCancerBiol 1998;8: 219–35.

65. HellstenS,BergeT, Wehlin L. Unrecognizedrenalcellcarcinoma.Clinical and pathologicalaspects.ScandJUrol Nephrol1981;15: 273–8.

66. Ferlay J, Parkin DM, Pisani P GLOBOCAN: cancerincidence and mortality worldwide. Lyon, France:IARC SciencePublishers,1998:CancerBase3.

67. McCredieM. Bladderandkidneycancers.CancerSurv1994;19: 343–68.

68. Chow WH, DevesaSS, Warren JL, FraumeniJr JF.Rising incidenceof renal cell cancer in the UnitedStates.JAMA 1999;281: 1628–31.

69. NakataS, OhtakeN, KubotaY, Imai K, YamanakaH,Ito Y, et al. Incidenceof urogenitalcancersin GunmaPrefecture,Japan:a 10-yearsummary.Int J Urol 1998;5: 364–9.

70. PisaniP,ParkinDM, Bray F, FerlayJ.Estimatesof theworldwide mortality from 25 cancersin 1990. Int JCancer1999;83: 18–29.

71. GnarraJR.vonHippel-Lindaugenemutationsin humanand rodent renal tumors—associationwith clear cellphenotype.J Natl CancerInst 1998;90: 1685–7.

72. Maher ER. Inherited renal cell carcinoma.Br J Urol1996;78: 542–5.

73. ZbarB, GlennG, LubenskyI, ChoykeP,WaltherMM,MagnussonG, et al. Hereditary papillary renal cellcarcinoma:clinical studiesin 10 families.J Urol 1995;153: 907–12.

74. Schmidt L, Duh FM, Chen F, Kishida T, Glenn G,ChoykeP,et al. Germlineandsomaticmutationsin thetyrosinekinasedomainof the MET proto-oncogeneinpapillaryrenalcarcinomas.NatGenet1997;16:68–73.

75. BjornssonJ, Short MP, Kwiatkowski DJ, HenskeEP.Tuberous sclerosis-associatedrenal cell carcinoma.Clinical, pathological, and genetic features. Am JPathol1996;149: 1201–8.

76. Brook-Carter PT, Peral B, Ward CJ, ThompsonP,HughesJ,MaheshwarMM, et al. Deletionof theTSC2and PKD1 genes associatedwith severe infantilepolycystic kidney disease—acontiguous gene syn-drome.Nat Genet1994;8: 328–32.

77. PeaM, Bonetti F, Martignoni G, HenskeEP, ManfrinE, Colato C, et al. Apparentrenal cell carcinomasintuberoussclerosisareheterogeneous:the identificationof malignantepithelioid angiomyolipoma.Am J SurgPathol1998;22: 180–7.

78. van SlegtenhorstM, de Hoogt R, HermansC, NellistM, JanssenB, Verhoef S, et al. Identification of thetuberoussclerosisgeneTSC1 on chromosome9q34.Science1997;277: 805–8.

79. CohenAJ, Li FP,BergS,MarchettoDJ,TsaiS,Jacobs



SC, et al. Hereditary renal-cell carcinomaassociatedwith a chromosomaltranslocation.N Engl JMed 1979;301: 592–5.

80. Kovacs G, Brusa P, De Riese W. Tissue-specificexpressionof a constitutional3;6 translocation:devel-opmentof multiple bilateral renal-cellcarcinomas.IntJ Cancer1989;43: 422–7.

81. van den Berg A, Buys CH. Involvementof multipleloci on chromosome3 in renal cell cancerdevelop-ment.GenesChromosomesCancer1997;19: 59–76.

82. van KesselAG, WijnhovenH, BodmerD, EleveldM,Kiemeney L, Mulders P, et al. Renal cell cancer:Chromosome3 translocationsas risk factors. J NatlCancerInst 1999;91: 1159–60.

83. Balzarini P, Dal Cin P, RoskamsT, Polito P, VanPoppelH, Van DammeB, et al. Histologymaydependon thepresenceof partialmonosomyor partial trisomy3 in renal cell carcinoma.Cancer Genet Cytogenet1998;105: 6–10.

84. van den Berg A, van der Veen AY, HulsbeekMM,KovacsG, Gemmill RM, Drabkin HA, et al. Definingthe position of the breakpoint of the constitutionalt(3;6) occurringin a family with renalcell carcinoma.GenesChromosomesCancer1995;12: 224–8.

85. Kreiger N, Marrett LD, Dodds L, Hilditch S,Darlington GA. Risk factorsfor renal cell carcinoma:results of a population-basedcase-control study.CancerCausesControl 1993;4: 101–10.

86. SchlehoferB, PommerW, MellemgaardA, StewartJH,McCredie M, Niwa S, et al. Internationalrenal-cell-cancer study. VI. the role of medical and familyhistory. Int J Cancer1996;11: 723–6.

87. BrauchH, Weirich G, HornauerMA, StorkelS, WohlT, Bruning T. Trichloroethyleneexposureandspecificsomaticmutationsin patientswith renal cell carcino-ma. J Natl CancerInst 1999;91: 854–61.

88. Gabow PA, Bennett WM. Renal manifestations:complicationmanagementand long-term outcomeofautosomaldominantpolycystic kidney disease.SeminNephrol1991;11: 643–52.

89. Zeier M, GeberthS, Ritz E, JaegerT, Waldherr R.Adult dominant polycystic kidney disease—clinicalproblems.Nephron1988;49: 177–83.

90. InternationalAgencyfor Researchon Cancer.Tobaccosmoking.In: IARC MonographEvaluationof Carcino-genic Risk of Chemicalsto Humans.Lyon, France:IARC, 1986:35–394.

91. Doll R. Cancersweakly relatedto smoking.Br MedBull 1996;52: 35–49.

92. McLaughlin JK, Lindblad P, MellemgaardA, McCre-die M, Mandel JS, SchlehoferB, et al. Internationalrenal-cell cancerstudy. I. Tobaccouse. Int J Cancer1995;60: 194–8.

93. MuscatJE,HoffmannD, WynderEL. The epidemiol-ogy of renal cell carcinoma.A secondlook. Cancer1995;75: 2552–7.

94. Yuan JM, CastelaoJE, Gago-DominguezM, Yu MC,Ross RK. Tobacco use in relation to renal cellcarcinoma.CancerEpidemiol BiomarkersPrev 1998;7: 429–33.

95. McLaughlin JK, Mandel JS, Blot WJ, SchumanLM,Mehl ES, FraumeniJr JF. A population-basedcase-control study of renal cell carcinoma.J Natl CancerInst 1984;72: 275–84.

96. Hard GC. Experimental models for the sequential

analysisof chemically-inducedrenal carcinogenesis.Toxicol Pathol1986;14: 112–22.

97. HechtSS,HoffmannD. Tobacco-specificnitrosamines,an important group of carcinogensin tobacco andtobaccosmoke.Carcinogenesis1988;9: 875–84.

98. YamasakiE, Ames BN. Concentrationof mutagensfrom urineby absorptionwith thenonpolarresinXAD-2: cigarettesmokershavemutagenicurine. Proc NatlAcad Sci USA 1977;74: 3555–9.

99. ShiaoYH, Rice JM, AndersonLM, Diwan BA, HardGC. von Hippel-Lindaugenemutationsin N-nitroso-dimethylamine-inducedrat renal epithelial tumors. JNatl CancerInst 1998;90: 1720–3.

100.BrockmollerJ, CascorbiI, Kerb R, RootsI. Combinedanalysisof inherited polymorphismsin arylamineN-acetyltransferase2, glutathioneS-transferasesM1 andT1, microsomal epoxide hydrolase,and cytochromeP450 enzymesas modulatorsof bladdercancerrisk.CancerRes1996;56: 3915–25.

101.LonguemauxS, DelomenieC, Gallou C, Mejean A,Vincent-Viry M, Bouvier R, et al. Candidategeneticmodifiers of individual susceptibility to renal cellcarcinoma:a studyof polymorphichumanxenobiotic-metabolizingenzymes.CancerRes1999;59: 2903–8.

102.MarchandLL, Wilkinson GR, Wilkens LR. Geneticanddietarypredictorsof CYP2E1activity: a phenotyp-ing study in Hawaii Japaneseusing chlorzoxazone.CancerEpidemiolBiomarkersPrev1999;8: 495–500.

103.Farker K, LehmannMH, OelschlagelB, Haerting J,Hoffmann A, Janitzky V, et al. Impact of CYP2E1genotypein renal cell and urothelial cancerpatients.Exp Toxicol Pathol1998;50: 425–31.

104.Wolk A, Lindblad P, Adami HO. Nutrition and renalcell cancer.CancerCausesControl 1996;7: 5–18.

105.MellemgaardA, Lindblad P, SchlehoferB, BergstromR, MandelJS,McCredieM, et al. Internationalrenal-cell cancerstudy.III. Role of weight, height,physicalactivity, anduseof amphetamines.Int J Cancer1995;60: 350–4.

106.YuanJM, CastelaoJE,Gago-DominguezM, RossRK,Yu MC. Hypertension,obesityandtheir medicationsinrelationto renalcell carcinoma.Br J Cancer1998;77:1508–13.

107.Lindblad P, Wolk A, BergstromR, PerssonI, AdamiHO. The role of obesityandweight fluctuationsin theetiology of renalcell cancer:a population-basedcase-control study. Cancer Epidemiol Biomarkers Prev1994;3: 631–9.

108.LindbladP,MellemgaardA, SchlehoferB, Adami HO,McCredie M, McLaughlin JK, et al. Internationalrenal-cellcancerstudy.V. Reproductivefactors,gyne-cologic operationsand exogenoushormones.Int JCancer1995;61: 192–8.

109.Gago-DominguezM, CastelaoJE, Yuan JM, Yu MC,RossRK (Submitted).

110.Toyokuni S, Mori T, Dizdaroglu M. DNA basemodificationsin renalchromatinof Wistar ratstreatedwith a renal carcinogen,ferric nitrilotriacetate.Int JCancer1994;57: 123–8.

111.Li JL, OkadaS, HamazakiS, EbinaY, MidorikawaO.Subacutenephrotoxicity and induction of renal cellcarcinomain mice treatedwith ferric nitrilotriacetate.CancerRes1987;47: 1867–9.

112.BenichouJ, Chow WH, McLaughlin JK, Mandel JS,FraumeniJr JF. Populationattributablerisk of renal



cell cancerin Minnesota.Am J Epidemiol1998;148:424–30.

113.McCredie M, Stewart JH. Risk factors for kidneycancer in New South Wales, Australia. II. Urologicdisease,hypertension,obesity, and hormonal factors.CancerCausesControl 1992;3: 323–31.

114.Talamini R, BaronAE, BarraS, Bidoli E, La VecchiaC, Negri E, et al. A case-controlstudyof risk factorsfor renal cell cancerin northernItaly. CancerCausesControl 1990;1: 125–31.

115.WynderEL, MabuchiK, WhitmoreJr WF. Epidemiol-ogy of adenocarcinomaof the kidney. J Natl CancerInst 1974;53: 1619–34.

116.Yu MC, Mack TM, HanischR, Cicioni C, HendersonBE. Cigarette smoking, obesity, diuretic use, andcoffee consumption as risk factors for renal cellcarcinoma.J Natl CancerInst 1986;77: 351–6.

117.Cantor KP, Lynch CF, Johnson D. Reproductivefactorsandrisk of brain,colon,andothermalignanciesin Iowa (United States).CancerCausesControl 1993;4: 505–11.

118.McLaughlin JK, Gao YT, Gao RN, ZhengW, Ji BT,Blot WJ, et al. Risk factors for renal-cell cancerinShanghai,China.Int J Cancer1992;52: 562–5.

119.MellemgaardA, Engholm G, McLaughlin JK, OlsenJH. Risk factorsfor renal-cellcarcinomain Denmark.III. Role of weight, physicalactivity andreproductivefactors.Int J Cancer1994;56: 66–71.

120.ChowWH, McLaughlinJK, MandelJS,Blot WJ,NiwaS, FraumeniJr JF.Reproductivefactorsandtherisk ofrenalcell canceramongwomen.Int J Cancer1995;60:321–4.

121.Gago-DominguezM, CastelaoJE,YuanJM, RossRK,Yu MC. Increasedrisk of renal cell carcinomasubse-quentto hysterectomy.CancerEpidemiol BiomarkersPrev1999;8: 999–1003.

122.Asal NR, RisserDR, KadamaniS, GeyerJR,Lee ET,Cherng N. Risk factors in renal cell carcinoma: I.Methodology, demographics,tobacco,beverageuse,andobesity.CancerDetectPrev1988;11: 359–77.

123.Adami HO, PerssonI, HooverR, SchairerC, BergkvistL. Risk of cancer in women receiving hormonereplacementtherapy.Int J Cancer1989;44: 833–9.

124.Asal NR, GeyerJR,RisserDR, Lee ET, KadamaniS,Cherng N. Risk factors in renal cell carcinoma.II.Medicalhistory,occupation,multivariateanalysis,andconclusions.CancerDetectPrev1988;13: 263–79.

125.MaclureM, Willett W. A case-controlstudyof dietandrisk of renal adenocarcinoma.Epidemiology1990; 1:430–40.

126.Chow WH, Lindblad P, Gridley G, Nyren O,McLaughlin JK, Linet MS, et al. Risk of urinary tractcancers following kidney or ureter stones. J NatlCancerInst 1997;89: 1453–7.

127.Hiatt RA, Tolan K, QuesenberryJr CP. Renal cellcarcinomaand thiazide use:a historical, case-controlstudy(California,USA). CancerCausesControl 1994;5: 319–25.

128.McCredie M, Ford JM, StewartJH. Risk factors forcancerof therenalparenchyma.Int J Cancer1988;42:13–6.

129.RosenbergAG, DexeusF, SwansonDA, von Eschen-bachAC. Relationshipof thyroid diseaseto renalcellcarcinoma.An epidemiologicstudy.Urology 1990;35:492–8.

130.TuckerMA, BoiceJr JD,HoffmanDA. Secondcancerfollowing cutaneousmelanomaand cancersof thebrain, thyroid, connective tissue, bone, and eye inConnecticut,1935–82.Natl CancerInst Monogr1985;68: 161–89.

131.BaristaI. An associationbetweenrenalcell carcinomaand lymphoid malignancies:a case series of eightpatients.Cancer1997;80: 1004–5.

132.La Vecchia C, Negri E, FranceschiS, D’Avanzo B,Boyle P. A case-controlstudyof diabetesmellitus andcancerrisk. Br J Cancer1994;70: 950–3.

133.GoodmanMT, MorgensternH, Wynder EL. A case-control study of factorsaffecting the developmentofrenalcell cancer.Am J Epidemiol1986;124: 926–41.

134.Mellemgaard A, Niwa S, Mehl ES, Engholm G,McLaughlin JK, OlsenJH. Risk factorsfor renal cellcarcinoma in Denmark: Role of medication andmedicalhistory. Int J Epidemiol1994;23: 923–30.

135.O’Mara BA, ByersT, SchoenfeldE. Diabetesmellitusand cancer risk: a multisite case-control study. JChronicDis 1985;38: 435–41.

136.Adami HO, McLaughlin J, Ekbom A, Berne C,SilvermanD, HackerD, et al. Cancerrisk in patientswith diabetesmellitus.CancerCausesControl1991;2:307–14.

137.LindbladP,ChowWH, ChanJ,BergstromA, Wolk A,Gridley G, et al. The role of diabetesmellitus in theaetiologyof renal cell cancer.Diabetologia1999; 42:107–12.

138.Green A, JensenOM. Frequencyof cancer amonginsulin-treateddiabeticpatientsin Denmark.Diabeto-logia 1985;28: 128–30.

139.Kessler II. Cancermortality amongdiabetics.J NatlCancerInst 1970;44: 673–86.

140.Smith GD, EggerM, Shipley MJ, Marmot MG. Post-challenge glucose concentration, impaired glucosetolerance,diabetes,and cancermortality in men. AmJ Epidemiol1992;136: 1110–4.

141. Ishikawa I. Uremic acquired renal cystic disease.Naturalhistory andcomplications.Nephron1991;58:257–67.

142.MarpleJT,MacDougallM, ChonkoAM. Renalcancercomplicatingacquiredcystickidneydisease.J Am SocNephrol1994;4: 1951–6.

143.Fick GM, GabowPA. Hereditaryand acquiredcysticdiseaseof the kidney. Kidney Int 1994;46: 951–64.

144.MaisonneuveP, Agodoa L, Gellert R, Stewart J,BucciantiG, LowenfelsA, et al. Cancerin patientsondialysis for end-stagerenal disease:an internationalcollaborativestudy.Lancet1999;354: 93–9.

145.DoubletJD,PeraldiMN, GattegnoB, ThibaultP,SraerJD. Renalcell carcinomaof native kidneys:prospec-tive studyof 129renaltransplantpatients.J Urol 1997;158: 42–4.

146.Kliem V, Kolditz M, Behrend M, Ehlerding G,Pichlmayr R, Koch KM, et al. Risk of renal cellcarcinomaafterkidneytransplantation.Clin Transplant1997;11: 255–8.

147.Levine LA, GburekBM. Acquired cystic diseaseandrenal adenocarcinomafollowing renal transplantation.J Urol 1994;151: 129–32.

148.Concolino G, Lubrano C, Ombres M, SantonatiA,Flammia GP, Di Silverio F. Acquired cystic kidneydisease:the hormonalhypothesis.Urology 1993; 41:170–5.



149. InternationalAgencyfor Researchon Cancer.Analge-sic mixtures containingphenacetin.In: IARC Mono-graphEvaluationof CarcinogenicRiskof ChemicalstoHumans.Updatingof IARC MonographsVols. 1–42.Lyon, France:IARC, 1987:Suppl7, 310–2.

150.Maclure M, MacMahon B. Phenactinand cancerofurinary tract. N Engl J Med 1985;313: 1479.

151.McCredieM, StewartJH, Day NE. Different rolesforphenacetinandparacetamolin cancerof thekidneyandrenalpelvis. Int J Cancer1993;53: 245–9.

152.McLaughlin JK, Blot WJ, Mehl ES, FraumeniJr JF.Relationof analgesicuseto renal cancer:Population-basedfindings. Natl Cancer Inst Monogr 1985; 69:217–22.

153.Gridley G, McLaughlin JK, Ekbom A, KlareskogL,Adami HO, Hacker DG, et al. Incidence of canceramongpatientswith rheumatoidarthritis.JNatl CancerInst 1993;85: 307–11.

154.MellemgaardA, Moller H, JensenOM, Halberg P,OlsenJH. Risk of kidney cancerin analgesicsusers.JClin Epidemiol1992;45: 1021–4.

155.Gago-DominguezM, YuanJM, CastelaoJE,RossRK,Yu MC. Regularuseof analgesicsis a risk factor forrenalcell carcinoma.Br J Cancer1999;81: 542–8.

156.McCredieM, PommerW, McLaughlinJK, StewartJH,Lindblad P, Mandel JS, et al. Internationalrenal-cellcancerstudy. II. Analgesics.Int J Cancer1995; 60:345–9.

157.Vane JR, Botting RM. Mechanismof action of anti-inflammatorydrugs.Int J TissueReact1998;20: 3–15.

158.SabatiniS.Pathophysiologicmechanismsin analgesic-induced papillary necrosis.Am J Kidney Dis 1996;28(Suppl1): S34–8.

159.Gobe GC, Axelsen RA. The role of apoptosisin thedevelopmentof renal cortical tubular atrophy asso-ciatedwith healedexperimentalrenalpapillary necro-sis.Pathology1991;23: 213–23.

160.Finkle WD, McLaughlin JK, RasgonSA, Yeoh HH,Low JE. Increasedrisk of renal cell canceramongwomen using diuretics in the United States.CancerCausesControl 1993;4: 555–8.

161.FraserGE, Phillips RL, BeesonWL. Hypertension,antihypertensivemedicationandrisk of renalcarcino-ma in California Seventh-Day Adventists. Int JEpidemiol1990;19: 832–8.

162.McLaughlin JK, Blot WJ, FraumeniJr JF. Diureticsandrenalcell cancer.J Natl CancerInst 1988;80: 378.

163.PrineasRJ,FolsomAR, ZhangZM, SellersTA, PotterJ. Nutrition and other risk factors for renal cellcarcinomain postmenopausalwomen. Epidemiology1997;8: 31–6.

164.Weinmann S, Glass AG, Weiss NS, Psaty BM,Siscovick DS, White E. Use of diuretics and otherantihypertensivemedicationsin relation to the risk ofrenalcell cancer.Am JEpidemiol1994;140:792–804.

165.Heath Jr CW, Lally CA, Calle EE, McLaughlin JK,ThunMJ. Hypertension,diuretics,andantihypertensivemedications as possible risk factors for renal cellcancer.Am J Epidemiol1997;145: 607–13.

166.McLaughlin JK, Chow WH, MandelJS,MellemgaardA, McCredieM, Lindblad P, et al. Internationalrenal-cell cancerstudy. VIII. Role of diuretics,other anti-hypertensive medications and hypertension. Int JCancer1995;9: 216–21.

167.ShapiroJA, Williams MA, Weiss NS, StergachisA,

LaCroix AZ, Barlow WE. Hypertension,antihyperten-sive medicationuse,and risk of renal cell carcinoma.Am J Epidemiol1999;149: 521–30.

168.Buck C, DonnerA. Cancerincidencein hypertensives.Cancer1987;59: 1386–90.

169.Coughlin SS, Neaton JD, Randall B, SenguptaA.Predictorsof mortality from kidney cancerin 332,547menscreenedfor theMultiple Risk FactorInterventionTrial. Cancer1997;79: 2171–7.

170.GambleJF,PearlmanED, Nicolich MJ. A nestedcase-control study of kidney canceramongrefinery/petro-chemicalworkers.EnvironHealthPerspect1996;104:642–50.

171.Raynor Jr WJ, ShekelleRB, RossofAH, Maliza C,Paul O. High blood pressureand 17-year cancermortality in the WesternElectric Health Study.Am JEpidemiol1981;113: 371–7.

172.Wolk A, Gridley G, Niwa S,LindbladP,McCredieM,MellemgaardA, et al. Internationalrenal cell cancerstudy.VII. Role of diet. Int J Cancer1996;65: 67–73.

173.Yuan JM, Gago-DominguezM, CastelaoJE, HankinJH, Ross RK, Yu MC. Cruciferous vegetablesinrelationto renalcell carcinoma.Int J Cancer1998;77:211–6.

174.Lindblad P, Wolk A, BergstromR, Adami HO. Dietandrisk of renalcell cancer:a population-basedcase-control study. Cancer Epidemiol Biomarkers Prev1997;6: 215–23.

175.DragstedLO, StrubeM, LarsenJC. Cancer-protectivefactors in fruits and vegetables:biochemical andbiological background.PharmacolToxicol 1993; 72(Suppl1): 116–35.

176.Hecht SS. Chemopreventionof cancer by isothio-cyanates,modifiersof carcinogenmetabolism.J Nutr1999;129: 768S–74S.

177.Mandel JS, McLaughlin JK, SchlehoferB, Mellem-gaardA, Helmert U, Lindblad P, et al. Internationalrenal-cell cancerstudy. IV. Occupation.Int J Cancer1995;61: 601–5.

178.McLaughlin JK, Blot WJ, DevesaSS,FraumeniJr JF.Renalcancer.In: SchottenfeldD, FraumeniJFJr.,eds.Cancerepidemiologyand prevention,2nd edn. NewYork: Oxford University Press,1996:1142–55.

179.McGregorDB, HeseltineE, Moller H. Dry cleaning,somesolventsusedin dry cleaningandotherindustrialchemicals.ScandJ Work Environ Health 1995; 21:310–2.

180.McLaughlinJ,Blot W. A critical reviewof epidemiol-ogy studiesof trichloroethyleneandperchloroethyleneandrisk of renal-cellcancer.Int Arch OccupEnvironHealth1997;70: 222–31.

181.Blair A, HartgeP, StewartPA, McAdamsM, Lubin J.Mortality andcancerincidenceof aircraftmaintenanceworkersexposedto trichloroethyleneandotherorganicsolventsand chemicals:extendedfollow up. OccupEnviron Med 1998;55: 161–71.

182.MorganR, KelshM, ZhaoK, HeringerS. Mortality ofaerospaceworkers exposedto trichloroethylene.Epi-demiology1998;9: 424–31.

183.VamvakasS, Bruning T, ThomassonB, LammertM,Baumuller A, Bolt HM, et al. Renal cell cancercorrelated with occupational exposure to trichloro-ethene.J CancerResClin Oncol 1998;124: 374–82.

184.Green L, Lash T. Re: “Renal cell cancercorrelated



with occupationalexposureto trichloroethylene”. JCancerResClin Oncol 1999;125: 430–2.

185.KovacsG. Application of molecularcytogenetictech-niquesto theevaluationof renalparenchymaltumors.JCancerResClin Oncol 1990;116: 318–23.

186. van denBerg E, Dijkhuizen T, OosterhuisJW, Geurtsvan Kessel A, de Jong B, Storkel S. Cytogeneticclassification of renal cell cancer. Cancer GenetCytogenet1997;95: 103–7.

187.Kovacs G. Molecular differential pathology of renalcell tumours.Histopathology1993;22: 1–8.

188.Erlandsson R. Molecular genetics of renal cellcarcinoma.CancerGenetCytogenet1998;104: 1–18.

189.GnarraJR,LermanMI, ZbarB, LinehanWM. Geneticsof renal-cellcarcinomaandevidencefor a critical rolefor von Hippel-Lindau in renal tumorigenesis.SeminOncol 1995;22: 3–8.

190.Decker HJ, Weidt EJ, Brieger J. The von Hippel-Lindau tumor suppressorgene.A rare and intriguingdiseaseopeningnew insight into basicmechanismsofcarcinogenesis.CancerGenetCytogenet1997;93: 74–83.

191.Latif F, Tory K, GnarraJ,YaoM, DuhFM, OrcuttML,et al. Identification of the von Hippel-Lindaudiseasetumor suppressorgene.Science1993;260: 1317–20.

192.HermanJG,Latif F, WengY, LermanMI, ZbarB, LiuS, et al. Silencingof the VHL tumor-suppressorgeneby DNA methylation in renal carcinoma.Proc NatlAcad Sci USA 1994;91: 9700–4.

193.Shuin T, Kondo K, Torigoe S, Kishida T, Kubota Y,HosakaM, et al. Frequentsomaticmutationsand lossof heterozygosityof the von Hippel-Lindau tumorsuppressorgenein primary humanrenal cell carcino-mas.CancerRes1994;54: 2852–5.

194.Whaley JM, Naglich J, Gelbert L, Hsia YE, LamiellJM, GreenJS, et al. Germ-line mutationsin the vonHippel-Lindau tumor-suppressorgene are similar tosomatic von Hippel-Lindau aberrationsin sporadicrenal cell carcinoma.Am J Hum Genet 1994; 55:1092–102.

195.Los M, JansenGH, Kaelin WG, Lips CJ,Blijham GH,VoestEE.Expressionpatternof thevonHippel-Lindauprotein in humantissues.Lab Invest1996;75: 231–8.

196.Kanno H, Kondo K, Ito S, Yamamoto I, Fujii S,TorigoeS, et al. Somaticmutationsof thevon Hippel-Lindau tumor suppressorgene in sporadic centralnervoussystemhemangioblastomas.CancerRes1994;54: 4845–7.

197.OberstrassJ,ReifenbergerG, ReifenbergerJ,WechslerW, Collins VP. Mutation of the Von Hippel-Lindautumoursuppressorgenein capillary haemangioblasto-masof thecentralnervoussystem.J Pathol1996;179:151–6.

198.KenckC, Wilhelm M, BugertP,StaehlerG, KovacsG.Mutation of the VHL gene is associatedexclusivelywith the development of non-papillary renal cellcarcinomas.J Pathol1996;179: 157–61.

199.Schmidt L, Junker K, Nakaigawa N, Kinjerski T,Weirich G, Miller M, et al. Novel mutationsof theMET proto-oncogenein papillary renal carcinomas.Oncogene1999;18: 2343–50.

200.ConsortiumECTS. Identificationand characterizationof the tuberoussclerosisgeneon chromosome16. TheEuropeanChromosome16 TuberousSclerosisCon-sortium.Cell 1993;75: 1305–15.

201.Carbonara C, Longa L, Grosso E, Mazzucco G,Borrone C, Garre ML, et al. Apparent preferentiallossof heterozygosityat TSC2over TSC1chromoso-mal region in tuberoussclerosishamartomas.GenesChromosomesCancer1996;15: 18–25.

202.GreenAJ, Smith M, YatesJR.Lossof heterozygosityon chromosome16p13.3in hamartomasfrom tuberoussclerosispatients.Nat Genet1994;6: 193–6.

203.HenskeEP, NeumannHP, ScheithauerBW, HerbstEW, Short MP, Kwiatkowski DJ. Loss of hetero-zygosity in the tuberoussclerosis(TSC2) region ofchromosomeband16p13occursin sporadicaswell asTSC-associatedrenal angiomyolipomas.GenesChro-mosomesCancer1995;13: 295–8.

204.SeppT, YatesJR,GreenAJ. Lossof heterozygosityintuberoussclerosishamartomas.J Med Genet1996;33:962–4.

205.WasheckaR, Hanna M. Malignant renal tumors intuberoussclerosis.Urology 1991;37: 340–3.

206.HadaczekP, Siprashvili Z, Markiewski M, DomagalaW, Druck T, McCuePA, et al. Absenceor reductionofFhit expressionin most clear cell renal carcinomas.CancerRes1998;58: 2946–51.

207.SteckPA, PershouseMA, JasserSA, YungWK, Lin H,Ligon AH, et al. Identificationof a candidatetumoursuppressorgene, MMAC1, at chromosome10q23.3that is mutated in multiple advancedcancers.NatGenet1997;15: 356–62.

208.TengDH, Hu R, Lin H, DavisT, Iliev D, FryeC, et al.MMAC1/PTEN mutationsin primarytumorspecimensandtumor cell lines. CancerRes1997;57: 5221–5.

209.Velickovic M, DelahuntB, GrebeSK. Lossof hetero-zygosityat 3p14.2in clearcell renalcell carcinomaisanearlyeventandis highly localizedto theFHIT genelocus.CancerRes1999;59: 1323–6.

210.PereraAD, KleymenovaEV, WalkerCL. Requirementfor the von Hippel Lindau tumor suppressorgeneforfunctional EGF receptor blockade by monoclonalantibodyC225in renalcell carcinoma.Clinical CancerRes2000: in press.

211.Kalble T, KovacsG. Molecular geneticsin the diag-nosisandprognosisof renalcancer.Klin Lab 1994;40:1209–34.

212.Schullerus D, Herbers J, Chudek J, Kanamaru H,KovacsG. Lossof heterozygosityat chromosomes8p,9p, and14q is associatedwith stageandgradeof non-papillary renal cell carcinomas.J Pathol 1997; 183:151–5.

213.Dal Cin P, vanPoppelH, vanDammeB, BaertL, Vanden BergheH. Cytogeneticinvestigationof synchro-nous bilateral renal tumors.CancerGenetCytogenet1996;89: 57–60.

214.Lubensky IA, GnarraJR, BertheauP, Walther MM,LinehanWM, ZhuangZ. Allelic deletionsof the VHLgenedetectedin multiple microscopicclear cell renallesions in von Hippel-Lindau diseasepatients.Am JPathol1996;149: 2089–94.

215.NakagawaM, Emoto A, HanadaT, NasuN, NomuraY. Tubulogenesisby microvascularendothelialcells ismediated by vascular endothelial growth factor(VEGF) in renal cell carcinoma.Br J Urol 1997; 79:681–7.

216.Nicol D, Hii SI, Walsh M, Teh B, Thompson L,Kennett C, et al. Vascularendothelialgrowth factor



expressionis increasedin renalcell carcinoma.J Urol1997;157: 1482–6.

217.Siemeister G, Weindel K, Mohrs K, Barleon B,Martiny-BaronG, MarmeD. Reversionof deregulatedexpressionof vascular endothelial growth factor inhuman renal carcinomacells by von Hippel-Lindautumorsuppressorprotein.CancerRes1996;56: 2299–301.

218.Tomisawa M, TokunagaT, Oshika Y, Tsuchida T,FukushimaY, Sato H, et al. Expressionpattern ofvascularendothelialgrowth factor isoform is closelycorrelatedwith tumour stage and vascularisationinrenalcell carcinoma.Eur J Cancer1999;35: 133–7.

219.BugertP, KenckC, Wilhelm M, KovacsG. Refiningaproximal breakpointclusterat chromosome3p11.2innon-papillaryrenalcell carcinomas.Int J Cancer1996;68: 723–6.

220.BeroudC, FournetJC, JeanpierreC, Droz D, BouvierR, FrogerD, et al. Correlationsof allelic imbalanceofchromosome14 with adverseprognosticparametersin148 renal cell carcinomas. Genes ChromosomesCancer1996;17: 215–24.

221.HofmockelG, WittmannA, DammrichJ,BassukasID.Expressionof p53 and bcl-2 in primary locally con-finedrenalcell carcinomas:no evidencefor prognosticsignificance.AnticancerRes1996;16: 3807–11.

222. Imai Y, StrohmeyerTG, FleischhackerM, SlamonDJ,Koeffler HP. p53 mutationsandMDM-2 amplificationin renalcell cancers.Mod Pathol1994;7: 766–70.

223. Jennings SB, Gnarra JR, Walther MM, Zbar B,LinehanWM. Renalcell carcinoma.Moleculargenet-ics and clinical implications.Surg Oncol Clin N Am1995;4: 219–29.

224.NishiyamaY, SuwaH, OkamotoK, FukumotoM, HiaiH, Toyokuni S. Low incidenceof point mutationsinH-, K- and N-ras oncogenesand p53 tumor sup-pressorgene in renal cell carcinomaand peritonealmesotheliomaof Wistar rats inducedby ferric nitrilo-triacetate.JpnJ CancerRes1995;86: 1150–8.

225.ReiterRE, AnglardP, Liu S, GnarraJR,LinehanWM.Chromosome17pdeletionsandp53mutationsin renalcell carcinoma.CancerRes1993;53: 3092–7.

226.Uchida T, WadaC, ShitaraT, EgawaS, MashimoS,KoshibaK. Infrequentinvolvementof p53 mutationsandlossof heterozygosityof 17p in the tumorigenesisof renalcell carcinoma.J Urol 1993;150: 1298–301.

227.Uchida T, Wada C, Wang C, Egawa S, Ohtani H,Koshiba K. Genomic instability of microsatelliterepeatsand mutationsof H-, K-, and N-ras, and p53genesin renal cell carcinoma.CancerRes 1994; 54:3682–5.

228.CairnsP, Tokino K, Eby Y, SidranskyD. Localizationof tumor suppressorloci on chromosome9 in primaryhuman renal cell carcinomas.CancerRes 1995; 55:224–7.

229.Dal Cin P, GaetaJ, Huben R, Li FP, Prout Jr GR,Sandberg AA. Renal cortical tumors. Cytogeneticcharacterization.Am J Clin Pathol1989;92: 408–14.

230.KovacsG, FuzesiL, EmanualA, Kung HF. Cytoge-netics of papillary renal cell tumors.GenesChromo-somesCancer1991;3: 249–55.

231.Brown JA, Anderl KL, Borell TJ, Qian J, BostwickDG, JenkinsRB. Simultaneouschromosome7 and17gain and sex chromosomeloss provide evidencethat

renalmetanephricadenomais relatedto papillary renalcell carcinoma.J Urol 1997;158: 370–4.

232.StummM, Koch A, WieackerPF,Phillip C, SteinbachF, Allhoff EP,et al. Partialmonosomy2p asthesinglechromosomalanomalyin a caseof renal metanephricadenoma.CancerGenetCytogenet1999;115: 82–5.

233.KattarMM, GrignonDJ,Wallis T, HaasGP,SakrWA,Pontes JE, et al. Clinicopathologic and interphasecytogeneticanalysisof papillary (chromophilic) renalcell carcinoma.Mod Pathol1997;10: 1143–50.

234.Dal Cin P, StasM, Sciot R, De WeverI, Van DammeB, Van den Berghe H. Translocation(X;1) revealsmetastasis31 yearsafter renalcell carcinoma.CancerGenetCytogenet1998;101: 58–61.

235.WetermanMA, Wilbrink M, Geurts van Kessel A.Fusionof thetranscriptionfactorTFE3geneto a novelgene, PRCC, in t(X;1)(p11;q21)-positive papillaryrenalcell carcinomas.ProcNatl Acad Sci USA 1996;93: 15294–8.

236.Shuin T, Kondo K, Sakai N, Kaneko S, Yao M,NagashimaY, et al. A caseof chromophoberenalcellcarcinomaassociatedwith low chromosomenumberandmicrosatelliteinstability. CancerGenetCytogenet1996;86: 69–71.

237.SpeicherMR, SchoellB, duManoirS,SchrockE, RiedT, CremerT, et al. Specificlossof chromosomes1, 2,6, 10, 13, 17, and 21 in chromophoberenal cellcarcinomasrevealedby comparativegenomichybridi-zation.Am J Pathol1994;145: 356–64.

238.NeuhausC, Dijkhuizen T, van denBerg E, StorkelS,Stockle M, Mensch B, et al. Involvement of thechromosomalregion11q13in renaloncocytoma:casereport and literature review. CancerGenetCytogenet1997;94: 95–8.

239.Dal Cin P,VandenBergheH, VanPoppelH, RoskamsT. Involvementof 12q12-13is a nonrandomchromo-some changein renal oncocytoma.GenesChromo-somesCancer1999;24: 94.

240.CavazzanaAO. Bellini duct carcinoma.A clinical andin vitro study.Eur Urol 1996;30: 340–4.

241.Fuzesi L, Cober M, Mittermayer C. Collecting ductcarcinoma:cytogeneticcharacterization.Histopathol-ogy 1992;21: 155–60.

242.Gregori-RomeroMA, Morell-QuadrenyL, Llombart-BoschA. Cytogeneticanalysisof threeprimaryBelliniduct carcinomas.GenesChromosomesCancer1996;15: 170–2.

243.PolascikTJ, Cairns P, EpsteinJI, Fuzesi L, Ro JY,MarshallFF,et al. Distal nephronrenaltumors:micro-satelliteallelotype.CancerRes1996;56: 1892–5.

244.SchoenbergM, Cairns P, Brooks JD, Marshall FF,EpsteinJI, IsaacsWB, et al. Frequentlossof chromo-somearms 8p and 13q in collecting duct carcinoma(CDC) of the kidney. GenesChromosomesCancer1995;12: 76–80.

245. JohanssonB, Heim S, MandahlN, MertensF, Mitel-man F. Trisomy 7 in nonneoplasticcells. GenesChromosomesCancer1993;6: 199–205.

246.Elfving P,AmanP,MandahlN, LundgrenR, MitelmanF. Trisomy 7 in nonneoplasticepithelial kidney cells.CytogenetCell Genet1995;69: 90–6.

247.Dal Cin P,Aly MS, DelabieJ,CeuppensJL, Van GoolS, Van DammeB, et al. Trisomy 7 and trisomy 10characterizesubpopulationsof tumor-infiltrating lym-phocytes in kidney tumors and in the surrounding



kidney tissue. Proc Natl Acad Sci USA 1992; 89:9744–8.

248.Eker R, Mossige J. A dominant gene for renaladenomasin the rat. Nature1961;189: 858–9.

249.Kobayashi T, Hirayama Y, Kobayashi E, Kubo Y,Hino O. A germlineinsertionin the tuberoussclerosis(Tsc2) gene gives rise to the Eker rat model ofdominantlyinheritedcancer.Nat Genet1995;9: 70–4.

250.Walker C, Goldsworthy TL, Wolf DC, Everitt J.Predispositionto renalcell carcinomadueto alterationof a cancer susceptibility gene. Science1992; 255:1693–5.

251.Yeung RS, Xiao GH, Jin F, Lee WC, Testa JR,KnudsonAG. Predispositionto renalcarcinomain theEker rat is determinedby germ-line mutation of thetuberoussclerosis2 (TSC2)gene.ProcNatl Acad SciUSA 1994;91: 11413–6.

252.Everitt JI, Walker CL, GoldsworthyTW, Wolf DC.Altered expressionof transforming growth factor-alpha:an early eventin renalcell carcinomadevelop-ment.Mol Carcinog1997;19: 213–9.

253.BazeedMA, ScharfeT, BechtE, JurincicC, Alken P,Thuroff JW. Conservative surgery of renal cellcarcinoma.Eur Urol 1986;12: 238–43.

254.Herr HW. Partial nephrectomyfor unilateral renalcarcinomaand a normal contralateralkidney: 10-yearfollowup. J Urol 1999;161: 33–4.

255.Lerner SE, Tsai H, Flanigan RC, Trump DL,FleischmannJ. Renal cell carcinoma:considerationsfor nephron-sparingsurgery.Urology 1995;45: 574–7.

256.MorganWR, Zincke H. Progressionandsurvival afterrenal-conservingsurgery for renal cell carcinoma:experiencein 104 patientsand extendedfollowup. JUrol 1990;144: 852–7.

257.Novick AC. Partial nephrectomy for renal cellcarcinoma.Urology 1995;46: 149–52.

258.PetritschPH, RauchenwaldM, ZechnerO, Ludvik W,PummerK, UrlesbergerH, et al. Resultsafter organ-preserving surgery for renal cell carcinoma. AnAustrianmulticenterstudy.Eur Urol 1990;18: 84–7.

259.ProvetJ, TesslerA, Brown J, GolimbuM, BosniakM,MoralesP. Partialnephrectomyfor renalcell carcino-ma: indications,resultsandimplications.J Urol 1991;145: 472–6.

260.Selli C, Lapini A, Carini M. Conservativesurgeryofkidney tumors.ProgClin Biol Res1991;370: 9–17.

261.Steinbach F, Stockle M, Muller SC, Thuroff JW,Melchior SW, Stein R, et al. Conservativesurgeryofrenal cell tumors in 140 patients:21 yearsof experi-ence.J Urol 1992;148: 24–9.

262.Van PoppelH, ClaesH, WillemenP,OyenR, BaertL.Is therea place for conservativesurgeryin the treat-mentof renalcarcinoma?Br J Urol 1991;67: 129–33.

263. van PoppelHP, BamelisF, Baert LV. Non-metastaticrenalcell carcinoam.Curr Opin Urol 1996;6: 241–4.

264.WunderlichH, ReicheltO, SchumannS, SchlichterA,KosmehlH, WernerW, et al. Nephronsparingsurgeryfor renal cell carcinoma4 cm. or less in diameter:indicatedor undertreated?J Urol 1998;159: 1465–9.

265.Van PoppelH, BaertLV. Nephron-sparingsurgery.In:PetrovichZ, BaertL, BradyLW, eds.Carcinomaof thekidney and testis, and rare urologic malignancies.Innovationsin management.Berlin: Springer-Verlag,1999:79–93.

266.Curry NS, SchabelSI, Betsill Jr WL. Small renal

neoplasms:diagnostic imaging, pathologic features,andclinical course.Radiology1986;158: 113–7.

267.FosterWL, HalvorsenRA, Dunnick NR. The clandes-tine renal cell carcinoma:atypical appearancesandpresentations.Radiographics1985;5: 175–92.

268.Kavoussi LR, Torrence RJ, Catalona WJ. Renaloncocytomawith synchronouscontralateralrenal cellcarcinoma.J Urol 1985;134: 1193–6.

269.LondonNJ, MessiosN, Kinder RB, SmartJG,OsbornDE, Watkin EM, et al. A prospectivestudyof thevalueof conventionalct, dynamic ct, ultrasonographyandarteriographyfor stagingrenal carcinoma.Br J Urol1989;64: 209–17.

270.Van PoppelH, BamelisB, BaertL. Electivenephron-sparing surgery for renal cell carcinoma.EuropeanUrological UpdateSeries(EUUS) 1997;6: 8–12.

271.Moll V, BechtE, ZieglerM. Kidneypreservingsurgeryin renalcell tumors:indications,techniquesandresultsin 152 patients.J Urol 1993;150: 319–23.

272.Carini M, Selli C, Barbanti G, Lapini A, Turini D,CostantiniA. Conservativesurgicaltreatmentof renalcell carcinoma:clinical experienceand reappraisalofindications.J Urol 1988;140: 725–31.

273.Van PoppelH, Bamelis B, Oyen R, Baert L. Partialnephrectomyfor renal cell carcinoma can achievelong-termtumor control. J Urol 1998;160: 674–8.

274.Van Poppel H, Baert LV. Organ-sparingsurgery:magicbullet or tragicinsanity?In: Kurth KH, MickischGH, Schroder FH, eds. Renal, bladder and prostaticcancer,an update.New York: Parthenon-PublishingGroup,1999:23–9.

275.HafezKS, Novick AC, CampbellSC.Patternsof tumorrecurrenceand guidelinesfor followup after nephronsparing surgery for sporadicrenal cell carcinoma.JUrol 1997;157: 2067–70.

276.Holthofer H. Immunohistologyof renal carcinomas.Eur Urol 1990;18(Suppl2):15–6.

277.BenningtonJL. Proceedings:Cancerof the kidney—etiology, epidemiology,and pathology.Cancer1973;32: 1017–29.

278.Bell ET Renaldiseases,2ndedn.Philadelphia,PA: LeaandFebiger,1950.

279.Hajdu SI, Thomas AG. Renal cell carcinoma atautopsy.J Urol 1967;97: 978–82.

280.MukamelE, KonichezkyM, EngelsteinD, ServadioC.Incidentalsmall renal tumorsaccompanyingclinicallyovert renalcell carcinoma.J Urol 1988;140: 22–4.

281.EschwegeP, SaussineC, Steichen G, Delepaul B,Drelon L, JacqminD. Radical nephrectomyfor renalcell carcinoma 30mm. or less: long-term followresults.J Urol 1996;155: 1196–9.

282.ChengWS, Farrow GM, Zincke H. The incidenceofmulticentricity in renal cell carcinoma.J Urol 1991;146: 1221–3.

283.Butler BP,Novick AC, Miller DP,CampbellSA, LichtMR. Management of small unilateral renal cellcarcinomas:radical versus nephron-sparingsurgery.Urology 1995;45: 34–41.

284.Holland JM. Proceedings:Cancer of the kidney—naturalhistoryandstaging.Cancer1973;32: 1030–42.

285.Xipell JM. The incidenceof benign renal nodules(aclinicopathologicstudy).J Urol 1971;106: 503–6.

286.Olsen S Tumors of the kidney and urinary tract.Copenhagen:Munsgaard,1984.

287.Sporn MB, Dunlop NM, Newton DL, Smith JM.



Preventionof chemical carcinogenesisby vitamin Aand its syntheticanalogs(retinoids). Fed Proc 1976;35: 1332–8.

288.HanashKA, Aquilina JW, Barrett DM, Blute ML,CockettAT, CoricaFA, et al. Clinical researchpriori-tiesin renalcell carcinoma:renalcell carcinomachemo-prevention strategies including target populations,proposedagents,andclinical trial designs—WorkgroupNo. 6. Union InternationaleContrele Cancer(UICC)andtheAme. Cancer1997;80: 999–1001.

289.KamatAM, LammDL. Chemopreventionof urologicalcancer.J Urol 1999;161: 1748–60.

290.Malvy DJ, Burtschy B, Arnaud J, Sommelet D,LevergerG, DostalovaL, et al. Serumbeta-caroteneandantioxidantmicronutrientsin childrenwith cancer.The ‘Cancer in Children and Antioxidant Micronu-trients’ FrenchStudyGroup.Int J Epidemiol1993;22:761–71.

291.Ryuto M, Jimi S, Ono M, Naito S, NakayamaY,Yamada Y, et al. All-trans-retinoic acid-dependentinhibition of E-cadherin-basedcell adhesion withconcomitantdephosphorylationof beta-cateninin meta-static humanrenal carcinomacells. Jpn J CancerRes1997;88: 982–91.

292.Liehr JG.Vitamin C reducestheincidenceandseverityof renal tumorsinducedby estradiolor diethylstilbes-trol. Am J Clin Nutr 1991;54: 1256S–60S.

293.Nikiforova NV, Chumakov AM, Kirpatovskii VI,KomarovaVA. Fat-solublevitamins E and A in thetumortissueandin thebloodof patientswith renal-cellcancer.Urol Nefrol (Mosk) 1996;6: 23–7.

294.Okamoto T, Fujioka T, Horiuchi S. A study of themetabolismof vitamin D in patientswith renal cellcarcinoma—withspecial referenceto serumconcen-tration of 1 alpha, 25-(OH)2D and its clinical signi-ficance.Nippon Hinyokika Gakkai Zasshi 1991; 82:890–9.

295.Fujioka T, HasegawaM, Ishikura K, MatsushitaY,Sato M, Tanji S. Inhibition of tumor growth andangiogenesisby vitamin D3 agentsin murinerenalcellcarcinoma.J Urol 1998;160: 247–51.

296.BretheauD, LechevallierE, EghazarianC, Grisoni V,CoulangeC. Prognosticsignificanceof incidentalrenalcell carcinoma.Eur Urol 1995;27: 319–23.

297.FilipasD, SpixC, Schultz-LampelD, BlackP,KaterB,Roth S, et al. Pilotstudiezur sonographischenFruher-kennung des Nierenzellkarzinoms.Radiologe 1999;39: 350–3.

298.MiharaS,NaganoK, KurodaK, YoshiokaR, SawatariM, Koba H, et al. Efficacy of ultrasonicmasssurveyfor abdominalcancer.J Med Syst1998;22: 55–62.

299.SekidoN, ShimazuiT, NoguchiR, AkazaH, Koiso K.

Transabdominal ultrasonography for detection ofurologic diseasesin the generalhealthcheck.NipponHinyokika GakkaiZasshi1994;85: 1089–96.

300.Horii Y, TsumataniK, NatsumeO, UemuraH, SasakiK, Komada S, et al. Usefulnessof computedtomo-graphyand ultrasonographyfor the early detectionofrenal cell carcinoma.Hinyokika Kiyo 1987; 33: 998–1004.

301.RinshoK, IshikawaS, UchidaK, Koiso K. The valueof ultrasonographyin early detection of renal cellcarcinoma.JpnJ Clin Oncol 1984;14: 329–34.

302.CochraneAL, Holland WW. Validation of screeningprocedures.Br Med Bull 1971;27: 3–8.

303.Hes FJ, Feldberg MA. Von Hippel-Lindau disease:strategiesin early detection(renal-,adrenal-,pancrea-tic masses).Eur Radiol 1999;9: 598–610.

304.SarasinFP,WongJB,LeveyAS, MeyerKB. Screeningfor acquiredcystic kidney disease:a decisionanalyticperspective.Kidney Int 1995;48: 207–19.

305.Curley SA, Lott ST, Luca JW, Frazier ML, KillaryAM. Surgicaldecision-makingaffectedby clinical andgeneticscreeningof a novel kindredwith von Hippel-Lindau diseaseand pancreaticislet cell tumors.AnnSurg1998;227: 229–35.

306.BretanJr PN. Acquiredrenalcystic disease:is thereaneedfor screeningfor renalcell carcinomain patientswith renal failure?SeminUrol 1994;12: 89–92.

307.ShapiroJA, Williams MA, WeissNS.Bodymassindexand risk of renal cell carcinoma.Epidemiology1999;10: 188–91.

308.Licht MR, Novick AC, Goormastic M. Nephronsparing surgery in incidental versussuspectedrenalcell carcinoma.J Urol 1994;152: 39–42.

309.Brisset JM, LugagnePM, Veillon B, Vallancien G,Charton M, Andre-BougaranJ. Parenchymal-sparingsurgery for small renal cell cancer: are there anyreasonablearguments?ProgClin Biol Res1989;303:153–9.

310.Steinbach F, Stockle M, Thuroff JW, Storkel S,Melchior SW, Kiewel R, et al. Long-termexperiencewith conservativesurgery of renal tumors. Surgicaltechnique, complications, results, DNA-cytometry.ScandJ Urol NephrolSuppl1991;138: 207–13.

311.TaariK, SaloJO,RannikkoS,KarkkainenP,NordlingS, Lehtonen T. Parenchyma-conservingsurgery forrenalcell carcinoma.Ann Chir GynaecolSuppl1993;206: 54–8.

312.D’Armiento M, Damiano R, FeleppaB, PerdonaS,Oriani G, De Sio M. Electiveconservativesurgeryforrenal carcinomaversus radical nephrectomy:a pro-spectivestudy.Br J Urol 1997;79: 15–9.



From the WHO ConsensusConference

PremalignantLesionsin the Genito-urinary Tract

Stockholm,Sweden,June8–9,2000

Copyright of Scandinavian Journal of Urology & Nephrology is the property of Taylor & Francis Ltd and its

content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's

express written permission. However, users may print, download, or email articles for individual use.

precancerous lesions in the kidney

Documents