Brief
Correspondence
Large-scale
Analysis
Demonstrates
Familial
Testicular
Cancer
to
have
Polygenic
Aetiology
Chey
Loveday
a,
Philip
Law
a,
Kevin
Litchfield
a,z,
Max
Levy
a,
Amy
Holroyd
a,
Peter
Broderick
a,
Zsofia
Kote-Jarai
a,
Alison
M.
Dunning
b,
Kenneth
Muir
c,d,
Julian
Peto
e,
Rosalind
Eeles
a,f,
Douglas
F.
Easton
b,g,
Darshna
Dudakia
a,
Nick
Orr
h,
Nora
Pashayan
i,
UK
Testicular
Cancer
Collaboration,
The
PRACTICAL
Consortium
z,
Alison
Reid
j,
Robert
A.
Huddart
k,
Richard
S.
Houlston
a,
Clare
Turnbull
a,l,m,n,*
aDivision of Genetics and Epidemiology, The Institute of Cancer Research, London, UK;bCentre for Cancer Genetic Epidemiology, Department of Oncology,
University of Cambridge, Cambridge, UK;cDivision of Health Sciences, Warwick Medical School, Warwick University, Warwick, UK;dInstitute of Population Health, University of Manchester, Manchester, UK;eDepartment of Non-communicable Disease Epidemiology, London School of Hygiene and Tropical Medicine, London, UK;fRoyal Marsden NHS Foundation Trust, London, UK;gCentre for Cancer Genetic Epidemiology, Department of Public Health and Primary Care, University of Cambridge, Cambridge, UK;hThe Breast Cancer Now Toby Robins Research Centre, The Institute of Cancer Research, London, UK; iDepartment of Applied Health Research, University College London, London, UK;jAcademic Uro-oncology Unit, The Royal Marsden NHS Foundation Trust,
Sutton, Surrey, UK;kAcademic Radiotherapy Unit, Institute of Cancer Research, Sutton, Surrey, UK;lWilliam Harvey Research Institute, Queen Mary University, London, UK;mDepartment of Clinical Genetics, Guys and St Thomas NHS Foundation Trust, London, UK;nPublic Health England, National Cancer Registration and Analysis Service, London, UK
a v ai l a b l e a t w w w . s c i e n c e d i r e c t . c o m
j o u r n al h o m e p a g e : w w w . e u r o p e an u r o l o g y . c o m
Articleinfo
Articlehistory: AcceptedMay31,2018
AssociateEditor: JamesCatto
Keywords: Familial Genetics GWAS
Polygenicriskscore Testiculargermcelltumour
Abstract
Testiculargermcelltumour(TGCT)isthemostcommoncancerinyoungmen.MultiplexTGCTfamilies havebeenwellreportedandanalysesofpopulationcancerregistrieshavedemonstratedafour-to eightfoldrisktomalerelativesofTGCTpatients.Earlylinkageanalysisandrecentlarge-scalegermline exomeanalysisinTGCTcasesdemonstrateabsenceofmajorhigh-penetranceTGCTsusceptibilitygene (s).Serialgenome-wideassociationstudyanalysesinsporadicTGCThaveintotalreported49 inde-pendent risk loci. Todate, it has notbeen demonstrated whetherfamilial TGCT arises dueto enrichmentofthesamecommonvariantsunderpinningsusceptibilitytosporadicTGCTorisdue tosharedenvironmental/lifestylefactorsordisparateraregeneticTGCTsusceptibilityfactors.Herewe presentpolygenicriskscoreanalysisof37TGCTsusceptibilitysingle-nucleotidepolymorphismsin 236familialand3931sporadicTGCTcases,and12368controls,whichdemonstratesclearenrichment forTGCTsusceptibilityallelesinfamilialcomparedtosporadiccases(p=0.0001),withthemajorityof familialcases(84–100%)beingattributabletopolygenicenrichment.TheseanalysesrevealTGCTasthe firstraremalignancyofearlyadulthoodinwhichfamilialclusteringisdrivenbytheaggregateeffectsof polygenicvariationintheabsenceofamajorhigh-penetrancesusceptibilitygene.
Patientsummary: Todate,ithasbeenunclearwhetherfamilialclustersoftesticulargermcell tumour(TGCT)ariseduetogeneticsorsharedenvironmentalorlifestylefactors.Wepresent large-scale genetic analyses comparing236 familial TGCTcases, 3931 isolated TGCT cases, and 12 368controls.WeshowthatfamilialTGCTiscaused,atleastinpart,bypresenceofahigherdose ofthesamecommongeneticvariantsthatcausesusceptibilitytoTGCTingeneral.
©2018TheAuthor(s).PublishedbyElsevierB.V.onbehalfofEuropeanAssociationofUrology. ThisisanopenaccessarticleundertheCCBYlicense(http://creativecommons.org/licenses/by/4.0/).
z MembersfromtheProstateCancerAssociationGrouptoInvestigateCancerAssociatedAlterations
intheGenome(PRACTICAL)consortiumareprovidedintheSupplementalmaterials.Informationof theconsortiumcanbefoundathttp://practical.icr.ac.uk/.
z Presentaddress:TranslationalCancerTherapeuticsLaboratory,TheFrancisCrickInstitute,London,UK.
*Correspondingauthor.DivisionofGeneticsandEpidemiology,TheInstituteofCancerResearch, London,SM25NG,UK.Tel.+442087224485;Fax:+442078825619.
E-mailaddress:clare.turnbull@icr.ac.uk (C.Turnbull).
https://doi.org/10.1016/j.eururo.2018.05.036
\
Testiculargerm cell tumour(TGCT)is the mostcommon cancerinyoungmen,withover18000newcasesofTGCT diagnosed annually in Europe [1]. Over the past 40 yr, severalauthorshavereportedfamilieswithmultiplecases of TGCT. Such observations, coupled with the higher concordance of TGCT in monozygotic twins than in dizygotic twinshave suggestedaheritablebasis to TGCT
[2]. In the 2000s, systematic familystudies, including in population-based registries, confirmed that first-degree relatives of patients with TGCT, have four- to eight-fold higher risk for TGCT. Based on data from the Swedish nationwideregistry,around2%ofTGCTcaseshavea first-degreerelativewithTGCT[3].WhilsttheclusteringofTGCT infamilieshasraisedthepossibilityofMendelian suscepti-bility,linkageanalysesandlarge-scaleexomesequencingof familial TGCT have not provided evidence for high-penetrancesusceptibilitygenes[4–6].
Meanwhile, recent genome-wide association studies (GWAS) haveidentified single-nucleotidepolymorphisms (SNPs) at49 independent lociassociated with TGCT risk
[7,8].Whiletheidentificationofsuchriskallelesprovesthe existence of inherited susceptibility, the genetic basis of familial TGCT is unclear. The identified risk SNPs are common,havemodesteffects,andhavebeendiscoveredby comparing unselected cases with controls. Although
statistical predictionssuggestthatcommonsusceptibility mayaccountforaround37%ofthefamilialrisk[7,8],thusfar nodirectevidenceforsuchapolygenicaetiologyhasbeen reported.WhilsttherapiddoublingofTGCTincidenceover the past40yrhasbeentakenas evidencefor significant environmental influences on TGCT aetiology, no specific environmentalriskfactorshavebeenrobustlyestablished
[1,9].Therefore,itisanopenquestionastowhetherfamilial TGCTisaconsequenceoftheco-existenceofunusuallyhigh numbersofcommonriskallelesorarisesduetootherrarer geneticfactors,sharedenvironmentalexposure,orcommon lifestylefactors.
To explorethe role of polygenic susceptibility in the aetiology of familial TGCT, we studied 236 familial and 3931sporadicTGCTcases,and12368healthypopulation controlsderivedfromtwopreviouslypublishedGWAS(see Supplementarymaterials).Briefly,casesandcontrolswere genotypedusingilluminaarrayswithrecoveryofuntyped genotypesbyimputation.Bothcasesandcontrolswereof Europeanancestry.WeextractedthegenotypesoftagSNPs for 37riskloci(SupplementaryTable2)whichhavebeen robustlyassociatedwithTGCTandforwhichhigh-quality directorimputedgenotypeswereavailableforbothGWAS datasets. We quantified risk allele burden using two approaches. First,we calculated thetotal number ofrisk
Fig.1–Testiculargermcelltumour(TGCT)riskallelesinTGCTcasesandcontrols.Polygenicriskscoresareplottedtoshow(A)probabilitydensity function(PDF)and(B)cumulativedistributionfunction(CDF)forfamilialTGCT(green,n=236),sporadicTGCT(blue,n=3931),andcontrols(red, n=12,368).PDFandCDFarealsoshownforriskallelecounts,unweightedbyeffectsize,in(C)and(D),respectively.
alleles and second, we calculated a polygenic risk score (PRS) combining the number of risk alleles weighted by theireffectsize.Toavoidstatisticalinflationresultingfrom theanalysisofmultiplerelatedfamilymembers, weonly included a single affected proband from each of the 236multiplexTGCTfamilies(SupplementaryTable1).
Comparingthedifferentgroups,weobserveda signifi-cantenrichmentofriskallelesinfamilialcasescompared with sporadic cases (average number of risk alleles,
p=0.0001;averagePRS,p=0.0001;Fig.1;Supplementary Table3).Notably,thesedifferenceswereseenineachofthe contributingGWASdatasets(SupplementaryFigs.1and2). Themostoverrepresentedriskalleleinfamilialcomparedto sporadiccaseswasSNPrs3751673(chromosome16q24.2; riskallelefrequency, 0.75 vs0.65, respectively; Cochran-Armitage trend test, p=0.006; Bonferroni-corrected,
p=0.22;SupplementaryTable4).
We next examined the relationship between PRS and familial TGCT in more detail. Although not statistically significant, likely on account of the limited number of
“large”familiesavailableforanalysis,PRSwasfoundtobe greaterinprobandsfromfamilieswiththreeormorecases
of TGCTcomparedwith two-case TGCTfamilies (Supple-mentaryFig.3).FamilialrelativerisksforTGCThavebeen reported in epidemiological studies to be higher for brothers than in father–son TGCT families [3]; however, wefoundnosignificantdifferenceinPRSbetweenfamilial casesanalyzedbypedigreestructure(SupplementaryFig. 3). To better quantify the extent of disease within each familyandtakeaccountofbilateraldisease,wegenerateda family history score that incorporates the number of affectedindividuals,degreeofrelatedness, andthe occur-rence of bilateral disease (see Supplementary materials). PRSwaspositivelycorrelatedwiththisfamilyhistoryscore, albeit again not reaching statistical significance(p=0.09,
Fig.2).Collectively,theseresultsprovide strongevidence for enrichment for known common TGCT risk alleles underpinningfamilialclusteringofTGCT.
WenextestimatedtheproportionoffamilialTGCTthatis attributabletoenrichmentofcommonriskallelesadopting thestrategyofHalvarssonetal[9],whichisbasedonthe premisethat,theoretically,aPRScanbedrawnfromoneof two distributions depending on aetiology. Cases caused byenrichmentofcommonTGCTriskalleleswill followa Fig.2–Relationshipbetweenrankedpolygenicriskscoreandfamilyhistoryscore.Spearmanrankcorrelationshowingtherelationshipbetween rankedpolygenicriskscore(y-axis)andrankedfamilyhistoryscore(x-axis)forthe228familialtesticulargermcelltumourcaseswithaknown pedigreestructure.Theunderlyingdistributionsofthosedataareshown,respectively,ontheoppositeaxes.
right-shiftedriskscoredistribution,whereascasescaused byotherfactors(notreflectedinPRS)willfollowthesame distribution as the population. To estimate the relative proportionofthetwounderlyingdistributions,wefitteda two-component Gaussianmixturemodel totheobserved PRSfor familialTGCTcases,restrictingonecomponent of the model to the distribution parameters defined by controls. The proportion of familial TGCT following a right-shifteddistribution (ie,enriched forTGCT suscepti-bilityalleles)inourdatasetwas100%,withthelowerbound of thisestimate at84% (Supplementary Table 5).Similar resultswereobtainedwhenusingthenumberofriskalleles, unweightedbyeffectsize(98%,95%confidenceinterval[CI]: 54–100%;SupplementaryTable5).Together,theseresults indicateattributiontoenrichmentfor commonTGCTrisk allelesforthemajorityofTGCTfamiliesanalyzed.
Survival from TGCT is high; however, the success of treatment is accompanied by long-term consequences associatedwith survivorship. TGCT isa model ofdisease in which prediction and early intervention could be impactful,astheprecursorcarcinoma-in-situ(CIS)lesion is reliably present from adolescence, with CIS cells exfoliated in the semen [10]. Early intervention could reducetheoccurrenceofinvasivecancerarisinginyoung men,reducingtheburdenofchemotherapy-related survi-vorshipissuesandreducingmortalityintheminoritywith treatment-refractory disease state. Therefore, whilst un-derstanding the inherited basis of TGCT is useful for counselingofindividualsconcernedaboutafamilyhistory ofthedisease,suchinformationmayalsobeimportantfor TGCT-screening programs targeting those at elevated a prioririsk.
In conclusion, we present the first evidence of clear enrichmentinfamilialTGCTforcommonTGCT susceptibili-tyalleles,demonstratingthatfamilialclusteringofTGCTis at least in part due to enrichment for the same genetic factors that confer susceptibility to sporadic TGCT. We demonstrateattribution topolygenic susceptibilityin the majorityofTGCTfamilies.TheenrichmentofcommonTGCT riskalleles amongfamilialcases issufficientlymodest in magnitudethat existenceof additionalgenetic and envi-ronmentaldriversoffamilialTGCTremainspossible.Future studieswillprovidefurtherelucidationofthegeneticbasis offamilialTGCTandempowerclinicalapplication.
Declarations
Ethicsapprovalandconsenttoparticipate:Collectionofbloodsamples andclinicalinformationfrompatientswasundertakenwithinformed writtenconsentandrelevantethicalreviewboardapprovalatrespective institutions.
Consentforpublication:Noidentifiablepersons’datawasusedinthis study.
Availabilityofdataandmaterials:CaseOncoArrrayGWASdataandcase/ controlUKPhaseIGWASdatahavebeendepositedintheEuropean Genome–phenome Archive (EGA) under accession codes EGAS00001001836andEGAS00001001302,respectively.
Authorcontributions:ClareAnnTurnbullhadfullaccesstoallthedatain thestudyandtakesresponsibilityfortheintegrityofthedataandthe accuracyofthedataanalysis.
Studyconceptanddesign:Turnbull,Houlston,Law,Loveday.
Acquisition of data: Levy,Dudakia,Holroyd,Broderick,Jarai,Dunning, Muir,Peto,Eeles,Easton,Orr,Reid,Huddart.
Analysis and interpretation of data: Turnbull, Houlston, Litchfield, Loveday.
Draftingofthemanuscript:Turnbull,Loveday.
Critical revision of the manuscript for important intellectual content: Turnbull,Houlston,Loveday.
Statistical analysis: Loveday,Law,Litchfield,Levy. Obtaining funding:Turnbull.
Administrative, technical, or material support:Levy,Broderick,Dudakia. Supervision:None.
Other(specify):None.
Financialdisclosures:ClareAnnTurnbullcertifiesthatallconflictsof interest, including specific financial interests and relationships and affiliationsrelevanttothesubjectmatterormaterialsdiscussedinthe manuscript(eg,employment/affiliation,grantsorfunding, consultan-cies,honoraria,stockownershiporoptions,experttestimony,royalties, orpatentsfiled,received,orpending),arethefollowing:None.
Funding/Supportandroleofthesponsor:Thisstudywassupportedby the Movember foundation and the Instituteof Cancer Research. K. LitchfieldwassupportedbyaPhDStudentshipfromCancerResearchUK. Richard S. Houlston and Peter Broderick are supported by Cancer ResearchUK.
GenotypingoftheOncoArraywasfundedbytheUSNationalInstitutesof Health(NIH)(U19CA148537forELucidatingLociInvolvedinProstate cancer SuscEptibility project and X01HG007492 to the Center for Inherited Disease Research under contract number HHSN268201200008I). Additional analytic support wasprovided by NIHNCIU01CA188392(PI:Schumacher).
ThePRACTICALconsortiumwassupportedbyCancerResearchUKGrants C5047/A7357, C1287/A10118, C1287/A16563, C5047/A3354, C5047/ A10692, C16913/A6135, European Commission's Seventh Framework Programmegrantagreementn223175(HEALTH-F2-2009-223175),and The National Institute of Health (NIH) Cancer Post-Cancer GWAS initiativegrant:No.1U19CA148537-01(theGAME-ONinitiative).
Acknowledgments:WethankthepatientswithTGCTandtheclinicians involved intheir carefor participation inthisstudy. Wethank the patients andall cliniciansforming partofthe UK Testicular Cancer Collaborationfortheirparticipationinthisstudy.Thisstudywouldnot have been possible without the contributions of M.K. Bolla (Breast CancerAssociationConsortium[BCAC]),Q.Wang(BCAC),K.Michailido (BCAC), J. Dennis (BCAC), P. Hall (Collaborative Oncological Gene-environmentStudy[COGS]),D.F.Easton(BCAC),A.Berchuck(Ovarian Cancer Association Consortium), R. Eeles(PRACTICAL),G. Chenevix-Trench (TheConsortiumofInvestigatorsofModifiersofBRCA1/2),J. Dennis,P.Pharoah,A.Dunning,K.Muir,J.Peto,A.Lee,andE.Dicks.We alsothankthefollowingfortheircontributionstothisproject:J.Simard, P.Kraft,C.Luccarini,andthestaffoftheCentreforGeneticEpidemiology Laboratory;andK.F.DohenyandthestaffoftheCenterforInherited DiseaseResearch(CIDR)genotypingfacility.Theresultspublishedhere arebasedinpartondatageneratedbytheTCGAResearchNetwork.This studymakesuseofdatageneratedbytheWellcomeTrustCaseControl Consortium2.
WearegratefulforsupportofNIHRfundingtotheNIHRBiomedical ResearchCentre atThe InstituteofCancer Research andThe Royal MarsdenNHSFoundationTrust.
Wethankalltheindividualswhotookpartinthesestudiesandallthe researchers,clinicians,technicians,andadministrativestaffwhohave enabledthisworktobecarriedout.
AppendixA. Supplementarydata
Supplementary material related to this article can be found, in the online version, at https://doi.org/10.1016/j. eururo.2018.05.036.
References
[1] LeCornetC,Lortet-TieulentJ,Forman D,etal.Testicularcancer incidencetoriseby25%by2025inEurope?Model-based predic-tionsin40countriesusingpopulation-basedregistrydata.EurJ Cancer2014;50:831–9.
[2] SwerdlowAJ,DeStavolaBL,SwanwickMA,MaconochieNE.Risksof breastandtesticularcancersinyoungadulttwinsinEnglandand Wales: evidence on prenatal and genetic aetiology. Lancet 1997;350:1723–8.
[3]HemminkiK,ChenB.Familialrisksintesticularcanceras aetiolo-gicalclues.IntJAndrol2006;1:205–10.
[4]LitchfieldK,LevyM,DudakiaD,etal.Raredisruptivemutationsin ciliaryfunctiongenescontributetotesticularcancersusceptibility. NatCommun2016;7:13840.
[5]LitchfieldK,LovedayC,LevyM,etal.Large-scalesequencingof testicular germ cell tumour(TGCT) cases excludes major TGCT predispositiongene.EurUrol2018;73:828–31.
[6]Stone L. No major predisposition gene in TGCT. Nat Rev Urol 2018;15:202.
[7]LitchfieldK,LevyM,OrlandoG,etal.Identificationof19newrisk lociandpotentialregulatorymechanismsinfluencingsusceptibility totesticulargermcelltumor.NatGenet2017;49:1133–40. [8]WangZ,McGlynnKA,Rajpert-DeMeytsE,etal.Meta-analysisof
fivegenome-wideassociationstudiesidentifiesmultiplenewloci associated with testicular germ cell tumor. Nat Genet 2017;49:1141–7.
[9]HalvarssonBM,Wihlborg AK,AliM,etal.Directevidencefora polygenic etiology in familial multiple myeloma. Blood Adv 2017;1:619–23.