Translating genomic information into clinical medicine: Lung cancer as a paradigm.

2012·1.1 MB·English

by  Levy, Mia A.

Perspective
Translating genomic information into clinical
medicine: Lung cancer as a paradigm
Mia A. Levy,1,2,3 Christine M. Lovly,2,3 and William Pao2,3,4
1DepartmentofBiomedicalInformatics,2Vanderbilt-IngramCancerCenter,3DepartmentofMedicine,
VanderbiltUniversityMedicalSchool,Nashville,Tennessee37212,USA
Wearecurrentlyinaneraofrapidlyexpandingknowledgeaboutthegeneticlandscapeandarchitecturalblueprintsof
various cancers. These discoveries have led to a new taxonomy of malignant diseases based upon clinically relevant
molecular alterations in addition to histology or tissue of origin. The new molecularly based classification holds the
promiseofrationalratherthanempiricapproachesforthetreatmentofcancerpatients.However,theacceleratedpaceof
discovery and the expanding number of targeted anti-cancer therapies present a significant challenge for healthcare
practitionerstoremaininformedandup-to-dateonhowtoapplycutting-edgediscoveriesintodailyclinicalpractice.In
thisPerspective,weuselungcancerasaparadigmtodiscusschallengesrelatedtotranslatinggenomicinformationintothe
clinic,andwepresentoneapproachwetookatVanderbilt-IngramCancerCentertoaddressthesechallenges.
Knowledge about genetic alterations that drive and sustain the Treatment of lung cancer with chemotherapy
growth of various cancers is exploding. Using lung cancer as
amodel,wedescribeevolvingtrendsandchallengesinleveraging Althoughlungcancerisaheterogeneousdisease,theparadigmfor
geneticinformationtoinformcancercare.Webrieflyreviewthe thetreatmentofmetastaticdiseasefromthelate1970’stothemid-
history of chemotherapy and targeted therapies in lung cancer, 2000’swaslargelyempiric,baseduponobservationsmadeinclin-
withafocusonthecurrenteraoftherapiestargetingtheprotein icaltrialsinvolvingcytotoxicchemotherapies.Themajorclassesof
products of driver mutations occurringin single oncogenes.We chemotherapies include alkylating agents (e.g., cisplatin), anti-
discusschallengesrelatedtohowtumorswillbeprofiled,including metabolites (gemcitabine), microtubule inhibitors (e.g., pacli-
issuesrelatedtowhoshouldperformtestingandwithwhatplat- taxel),topoisomeraseinhibitors(e.g.,etoposide),andcytotoxic
form,whoshouldpayfortestingandhowgenepatentsmayaffect antibiotics(e.g.,doxorubicin).Theseagentsgenerallykillcellsthat
costs, which tumors to profile and when, and the concept of dividerapidly,oneofthekeypropertiesofmostcancercells.The
companiondiagnostics.Wealsodiscussissuesrelatedtoclinical dogma by the late 1990’s was that all patients with metastatic
interpretationsofgenomicinformation:howwilltheybereported NSCLC should receive ‘‘modern platinum-based chemotherapy
toclinicians,inwhatformat,andusingwhatknowledgeresources doublets’’ (i.e., carboplatin plus an agent such as paclitaxel or
(Fig. 1). Finally, we present one approach that we took at the gemcitabine), regardless of histology (Schiller et al. 2002). With
Vanderbilt-IngramCancerCentertoaddresstheseissuesfor our this‘‘one-size-fitsall’’approach,thetherapeuticefficacyoftreat-
personalizedcancermedicineinitiative. mentreachedaplateau.Alandmarktrialinwhich;1200patients
with advanced NSCLC were randomly assigned to one of four
platinum-based chemotherapy regimens showed no significant
Lung cancer
advantageofanyoneregimenovertheothersintermsofoverall
Lung cancer is the leading cause of cancer-related death in the survival(;8mo)(Schilleretal.2002).Theadditiontoplatinum-
UnitedStates(Siegeletal.2011).Theoverall5-yrsurvivalforall based doublets of bevacizumab, a monoclonal antibody against
stagesis;16%.Sinceatleast1968,whentheinternationaltumor- vascular endothelial growth factor (a major regulator of angio-
node-metastasis (TNM) staging system was officially adopted as genesisinnormalandmalignanttissue),ledtoonlya2-mosur-
the‘‘internationallanguage’’fordiagnosisandtreatmentoflung vival benefit (12 mo versus 10 mo) in one study (Sandler et al.
cancer,patients’tumorshavebeenclassifiedprimarilyaccording 2006)butnotanother(Recketal.2009)andwasassociatedwithan
to tumor histology (International Union Against Cancer 1968). increased risk of treatment-related deaths (Sandler et al. 2006).
Clinically,thetwomainsubtypesoflungcancerincludesmallcell Despite multiple efforts to identify molecular predictors of re-
lungcancer(SCLC;;10%ofcases)andnon-smallcelllungcancer sponse to cytotoxic chemotherapy, none are used routinely in
(NSCLC;;90%ofcases).ThisPerspectivewillfocusprimarilyon clinicalpractice(Andrewsetal.2011).
NSCLC, the major histologic subtype in which advances in the
Treatment of lung cancer with targeted therapies
genomicsoflungcancerhavebeenmade.NSCLCisfurthersub-
dividedintoadenocarcinoma,squamouscellcarcinoma,andlarge in the beginning of the genomic era: The new
cell carcinoma. Unfortunately, more than half of patients with taxonomy of single gene driver mutations
NSCLCarediagnosedatadvancedstages,wheretreatmentisnot
curative. The first draft of the human genome was published in 2001
(Lander et al. 2001; Venter et al. 2001). Although a handful of
genes(i.e.,KRAS,NRAS,PTEN)wereshowntobemutatedinlung
4Correspondingauthor cancerspriorto2001,mosttumor-specificgenealterationshave
[email protected] beenidentifiedsincethattime(Fig.2).Withtheadventofwhole
Article published online before print. Article and publication date are at
exome,wholegenome,andwholetranscriptomesequencing,the
http://www.genome.org/cgi/doi/10.1101/gr.131128.111.Freelyavailableonline
throughtheGenomeResearchOpenAccessoption. paceofdiscoveryisaccelerating.
22:2101–2108 (cid:2)2012,PublishedbyColdSpringHarborLaboratoryPress;ISSN1088-9051/12;www.genome.org GenomeResearch 2101
www.genome.org
Levy et al.
(<1%) (Table 1).For details on thetypes of recurrentgenetical-
terations that affect these genes in lung cancer, see Pao and
Chmielecki(2010),Weissetal.(2010),Paoetal.(2011),Bergethon
et al. (2012), Lovly and Pao (2012), and Pao and Hutchinson
(2012).Importantly,atumorwithamutationinoneofthesegenes
usually does not have a mutation in another gene (except for
PIK3CA). Thisfindingimpliesthatthese mutationsdefineclini-
callydistinctentities.
A major reason why this classification scheme has been
adopted clinically is that specific ‘‘driver’’ mutations are linked
withsensitivityand/orresistancetoexistingoremergingtargeted
therapies. Such therapies are different from conventional cyto-
toxicchemotherapiesinthatthey‘‘target’’amutantproteinspe-
cifictotumors,notfoundinnormalcells.Thus,thetherapeutic
windowofdrugefficacycanpotentiallybegreaterwithfewerof
thesideeffectsseenwithcytotoxicchemotherapy.
Thefirstsuccessfulexampleofatargetedtherapybasedupon
driver mutations in lung cancer involves the epidermal growth
factor receptor (EGFR). EGFR is a receptor tyrosine kinase that
normally regulates intracellular signaling pathways to promote
Figure 1. Complexities of genetically informed cancer medicine. As
shownhere,multiplefactorsneedtobeaddressedinordertoeffectively cellgrowthandproliferation.Lungtumorsharboringspecificac-
translateknowledgeoftumorgenemutationsintoroutineclinicalpractice. tivating mutations in the EGFR kinase domain are exquisitely
sensitivetothe EGFRtyrosinekinaseinhibitors(TKIs), erlotinib
(Tarceva),andgefitinib(Iressa).Morethan70%ofpatientswith
Genomic profiling using sequencing and other high- EGFRmutanttumorswilldisplayaradiographicresponsetoEGFR
throughput technologies has led to an explosion of genomic TKIs,whileresponserates totraditional platinum-basedchemo-
classifiers for lung cancers (Chen et al. 2007; Weir et al. 2007; therapyregimensare;30%(Mitsudomietal.2009;Maemondo
Sheddenetal.2008;Yuetal.2008).Notallofthesehavedirect etal.2010).Only1%ofEGFRwild-type(i.e.,non-EGFRmutant)
clinicalrelevance.Oneofthemostwidelyadoptedclassification tumorswillrespondtoEGFRTKItherapy(Moketal.2009).The
schemaintheclinicinvolvestheuseof‘‘driver’’mutations.Basic latter‘‘wild-type’’categoryincludestumorswithothergenetical-
and clinical research has revealed that some genetic lesions are terations,suchasKRASmutations,ALKfusions,ROS1fusions,etc.
notonlynecessaryfortheinitialdevelopmentorprogressionof Unfortunately,eventhoughpatientstreatedwithEGFRTKIsare
aspecifictumorbutarealsorequiredforthemaintenanceofthat stillnotcuredoflungcancer,progression-freesurvivalandquality
tumor’s survival; a concept referred to as ‘‘oncogene addiction’’ oflifeareimprovedwiththistargetedtherapyversustraditional
(Weinstein 2002). Such ‘‘driver’’ mutations sustain tumors, but chemotherapy(Moketal.2009).
simultaneouslycanserveascancer-specifictargetsofvulnerability Thelistofsuchexamplesoftargetableoncogenicdrivermu-
to be exploited therapeutically. This notion is best exemplified tationscontinuestogrow.Patientswithlungtumorsthatarere-
bythesuccessofimatinib(Gleevec),anABLkinaseinhibitor,in fractory tostandardchemotherapy but that harborspecificALK
treatingpatientswithchronicmyelogenousleukemiadrivenand genefusionsdisplayresponseratesof;60%totheALKtyrosine
sustained by the BCR-ABL oncogene
(Drukeretal.2001).
Currently, mostofthekey‘‘driver’’
mutationsinNSCLChavethusfarbeen
found in genes encoding signaling pro-
teins, such as protein kinases. By trans-
ferring phosphate groups from ATP to
specific target proteins, these signaling
proteinsarecriticalregulatorsofcellular
proliferation and survival. ‘‘Driver’’ mu-
tationsarealsofoundinGTPases,which
are intracellular enzymes that typically
functiondownstreamfromproteinkinases
topropagatecellgrowth,proliferation,and
survivalsignals.
Inlungcancer,themaingeneswith
knownorpotential‘‘driver’’mutationsta-
tus currently include EGFR (10%–30%),
KRAS (15%–30%), FGFR1 (20%), ERBB2
(alsoknownasHER2)(2%–5%),PIK3CA
(2%–5%),ALK(3%–5%),BRAF(1%–3%),
ROS1(1%),MAP2K1(alsoknownasMEK1) Figure2. Timelineforthediscoveryofsignificantmolecularalterationsinlungcancer.(Adenoca)
(1%), RET (1%), NRAS (1%), and AKT1 adenocarcinoma;(CNVs)copynumbervariants;(SCLC)smallcelllungcancer.
2102 GenomeResearch
www.genome.org
Genetically informed lung cancer medicine
Table1. MolecularsubsetsofNSCLCdefinedby‘‘driver’’ Howwilltumorsbeprofiled?
mutations
DNA sequencingtechnology isundergoing a revolution, andad-
Gene Frequency References vancesinmassivelyparalleltechnologieshavedramaticallyreduced
thecostofsequencing(MacconaillandGarraway2010).Thesede-
AKT1 #1% Bleekeretal.2008 velopmentsimplythatcomprehensivetumorgenomeprofilingwill
Doetal.2008
notonlyoccuraspartofpilotprograms(Roychowdhuryetal.2011)
Malangaetal.2008
ALK 3%–7% Sodaetal.2007 butwillalsobecomewidelyadoptedinthefuture.Tobeclinically
Koivunenetal.2008 applicable,suchprofilingshouldbeperformedintheappropriate
Kwaketal.2010 settings,i.e.,inlaboratoriesthatarecompliantwiththeClinical
BRAF 1%–3% Pratilasetal.2008
Laboratory Improvement Amendment (CLIA) of 1988 adminis-
Leeetal.2010
teredbytheCenterforMedicareServices(CMS)andinaccordwith
Paiketal.2011
EGFR 10%–30% Lynchetal.2004 guidelinessetforthbytheCollegeofAmericanPathologists(CAP)
Paezetal.2004 (fordetails,seeFebboetal.2011).Presumably,mostassayswillbe
Paoetal.2004 performed using formalin-fixed paraffin-embedded tissues and
FGFR1 20% Weissetal.2010
notfreshfrozentissues,sinceinthevastmajorityofpatientswith
Duttetal.2011
HER2 2%–5% Sonobeetal.2006 metastaticdisease,onlytheformerisavailable.
Leeetal.2010
KRAS 15%–30% Tsaoetal.2006
Rielyetal.2008 Whowillperformthetesting,localhospitals
MEK1 1% Marksetal.2008 orreferencelaboratories?
NRAS 1% Broseetal.2002
Dingetal.2008 Some local and many academic hospitals are able to perform a
PIK3CA 1%–3% Kawanoetal.2006 limited number of molecular and cytogenetic tests on a select
Leeetal.2010 numberofcancergenesusingtechnologiesthathavebeenwidely
RET 1% Kohnoetal.2012
availablein CLIAlabsformanyyears.However,thetechnology
Takeuchietal.2012
Lipsonetal.2012 platformsfor highthroughputtumorgenemutationtestingare
Juetal.2012 changingeveryyear.Howwillmoleculardiagnosticslaboratories
ROS1 1% Bergethonetal.2012 keepupwiththecostofreplacingtheirtechnologyandvalidating
thenewassayseveryfewyears?Caneachlocalhospitalaffordto
develop/performsuchassaysonitsown,orwillsuchassaysneedto
kinaseinhibitor,crizotinib(Kwaketal.2010).Incontrast,typical
be outsourced because of their inherent complexity? Until the
responseratesinthesettingofpretreatedlungcanceraretypically
‘‘landscape’’issettled,referencelaboratoriesandonlyahandfulof
around10%(Hannaetal.2004).PatientswithALKfusion-positive
well-endowed institutions are likely to be able to offer compre-
lungcancerdonotrespondtotreatmentwithEGFRTKIs(Shaw
hensivetumor profilingfor patients. Perhapseven more impor-
etal.2009).Similarly,ROS1fusion-positivetumorsarealsohighly
tantly,therearealimitednumberoftrainedmolecularpatholo-
sensitivetocrizotinib(toutedasanALKTKI,butwithactivityalso
gists worldwide who can run such labs. The training of future
againstROS1andMET)(Bergethonetal.2012).Inaddition,ERBB2
pathologistswillneedtochangetoaccommodatetheriseinde-
(HER2)mutanttumorsareresponsivetoHER2TKIs(DeGreveetal.
mandformoleculartesting.
2012),andBRAFmutanttumorsmayberesponsivetothemutant-
specificBRAFinhibitor,vemurafenib(Gautschietal.2012).
As a result of these compelling data, multiplex mutational Whowillpayforthemoleculartesting?
profilingoflungtumorsisnowpartofacceptedstandardclinical The cost of current molecular diagnostic assays can vary from
andpathologypracticeatlocationsthroughouttheUS(Paoetal. hundredsofdollars(forsinglegenesequencing)tothousandsof
2009;Dias-Santagataetal.2010;Sequistetal.2011;Suetal.2011). dollars (for multigene profiling or fluorescent in situ hybridiza-
Such‘‘first-generation’’profilingofabout10genesimplicatedin tion).Thecostsofsuchtestsarereimbursablebyinsurers,butat
lung-cancerpathogenesiswilldetectmutationsinroughlyhalfof somepoint,asthe‘‘$1000genome’’becomesareality(Macconaill
unselectedNSCLCsandnearly90%oflungadenocarcinomasfrom andGarraway2010),thecostofcomprehensivetumorprofiling
EastAsianneversmokers(Sunetal.2010;Lietal.2011). willactuallybecheaperandmorepracticalthanindividualgene
tests.Whenthattimecomes,itwillbemorecosteffectiveona
Maximizing the potential of genetically societallevelforpatientsdiagnosedwithcancertoundergoroutine
informed cancer medicine comprehensive tumor profiling at various stages of treatment,
providedthattheuseofexpensivetherapiesisrestrictedtoonly
Theexplosionofgenomicinformationoffersgreatpromisetoward thosemostlikelytobenefit.
an era of ‘‘precision’’ or genetically informed cancer medicine,
where patients are treated rationally according to the genetic
Howwillgenepatentlicensingbeaddressed?
makeupoftheirindividualtumorsratherthanempiricallybased
uponhistology.Wearejustatthebeginningofthisparadigmshift, In2005,itwasestimatedthatnearly20%ofhumangeneswere
especially because translation of basic scientific findings to the explicitlyclaimedasU.S.intellectualproperty(JensenandMurray
clinicoftentakestimeandeducationforallinvolved.Inaddition, 2005).Manyimportantcancergenesarecoveredbypatents,sonot
theknowledgeexplosionpresentsmanychallenges(Fig.1).Below everyone can legally analyze them for commercial purposes
wediscussmanyoftheissuesthatremaintobeaddressedandour withoutproperlicenses(Kean2011).Infact,investigatingallofthe
preliminaryexperienceattheVanderbilt-IngramCancerCenterin relevantpatentclaims(issuedandpending)forpossibleinfringe-
solvingthem. mentonthetestingofabout100cancergeneswasestimatedin
GenomeResearch 2103
www.genome.org
Levy et al.
2011tocostatleast$35million(Kean2011).Creativesolutions siteofmetastasisharboringagivendrugresistancemechanismand
among all stakeholders, including lawyers, businesspeople, and another site harboring a different mechanism (Engelman et al.
policymakers,willneedtobemadetopavethewayforward.For- 2007).Theseobservationsraisethequestionastowhetherpatients
tunately, recent arguments have been made that perhaps these shouldhavemorethanonetumorexaminedatanygiventime.
claimshavebeenoverstatedandthereappearstobelittleevidence Feasibilityinmostinstanceswilllimitwhetherornotthispractice
thatwouldsupportanassumptionthatgenepatentsposeasub- occurs(e.g.,canmultiplesitesbebiopsiedsafely?).Anotherques-
stantialimpedimenttotechnologieslikewhole-genomesequenc- tion is whether patients should undergo repeat tumor profiling
ing(Holman2012). after each treatment course to assist in prioritizing additional/
futuretherapies.Cost-benefitratioswilllikelyguidethispractice.
Whatsequencingplatformshouldbeused?
Howwillcliniciansbeassistedininterpretingtheresults
Howmuchaveragegenedepthandcoverageareappropriatefor
inordertomakeactionabledecisionsforpatientcare?
profiling?Isvariablecoverageacrosskeygenesacceptable?Does
thewholeexomeneedtobedone,thewholegenome,orthewhole Informationregardingthetherapeuticimplicationsoftumorgene
transcriptome?Orshouldjust‘‘cancergenes’’beexamined?Im- mutation testing results is clinically complex. Mutations in the
portantly, canthetestand theinterpretationbeperformedin a samegenemayconferprimarysensitivity,primaryresistance,or
clinicallyrelevanttimeframe,i.e.,within1to2wkofthefirstclinic secondaryresistancetothesamedrug(PaoandChmielecki2010).
visit,sothatpatientswithmetastaticdiseasecanstarttreatmentas Forinstance,inlungadenocarcinoma,mutationsinEGFRexon19
soonaspossible?Onecurrent‘‘second-generation’’approachin- deletions and exon 21 point mutations (L858R) confer primary
volvestargetedcaptureandresequencingofabout200genesim- sensitivitytoEGFRTKI’s(Lynchetal.2004;Paezetal.2004;Pao
plicated in cancer with high coverage (Lipson et al. 2012). This et al. 2004), whereas EGFR exon 20 mutations (T790M) confer
particular platform is attractive because it detects all known ge- primaryorsecondaryresistancetothesamedrugs(Kobayashietal.
nomicalterationsthatcouldhaverelevance,i.e.,pointmutations, 2005;Paoetal.2005a).TumorsharboringmutationsinKRAS(Pao
insertions, deletions, rearrangements, and copy-number alter- etal.2005b;Rielyetal.2009)orALK(Shawetal.2009)areusually
ations.Moreover,theassayisperformedinaCLIA-certifiedlabo- unresponsivetoEGFRTKIs.
ratorywithaquickturnaroundtime. TumorsmayharboralterationsthataffectnotonlyDNAbut
alsoRNAorprotein,eachofwhichmayhavedifferenttherapeutic
implications.Forexample,inlungcancer,EGFRDNAmutations
Whataboutcompaniondiagnostics?
predict for high sensitivity to EGFR tyrosine kinase inhibitors,
AccordingtotheUSFoodandDrugAdministration,acompanion whileEGFRDNAcopy-numberlevels,mRNAexpressionlevels,or
diagnosticdeviceisaninvitrodiagnostic(IVD)devicethatpro- proteinlevelshavelittleornopredictivevalue.Providersinabusy
videsinformationthatisessentialforthesafeandeffectiveuseof medical practice will be challenged to keep up with all of this
a corresponding therapeutic product. Examples include an ALK multifactorialandclinicallysignificantdetail.
fluorescenceinsituhybridization(FISH)testtodetectALKgene
rearrangements for the use of crizotinib in lung cancer and an
Whowillinterpretandreportgenomicresultstophysicians?
allele-specificPCR-basedtesttodetectBRAFV600Emutationsforthe
use of vemurafanib (a BRAF inhibitor) in melanoma. Currently, Traditionally,moleculardiagnosticlaboratoriesgenerateaninter-
when an appropriate scientific rationale supports such an ap- pretivereportregardingbasicinformationonthetestperformance
proach, the FDA encourages the development of therapeutic andclinicalsignificanceoftheresults.Evenwithonlyafewgene
productsthatdependontheuseofapprovedorclearedIVDcom- mutationstestedintheclinicallabtoday,busylaboratorydirectors
paniondiagnosticdevices.However,howistheuseofcompanion maynothavesufficienttimetostaycurrentwiththelatestknowl-
diagnostics compatible with multiplex mutational profiling? edge. This challenge will continue to grow as molecular testing
If 100 different mutations are found that have implications for techniquesscaletoincludefullsequencingofhundredsorthou-
therapy,willtheFDAmandateuseof100differentdiagnosticde- sandsofgenesforclinicaluse.Generally,thereisalackofpublically
vices,eachofwhichcouldcostthousandsofdollars?Howisthat accessibleknowledgeresourcesavailabletoassistbothmolecular
feasiblefromapracticalstandpoint,bothfromatissueavailability diagnosticlabsandpracticingcliniciansininterpretingtheclinical
perspective and from a financial perspective? The FDA and/or significanceoftheseresults.
otherregulatoryagencieswillneedtoreconsiderthisapproachas
reliablegenome-widesequencingplatformsbecomeavailable.
Whatwillbereported?
Itiscurrentlyacceptedthatknowledgeoftumordrivermutation
Whosetumorsshouldbeprofiledandwhen?
status is clinically helpful, allowing for the prioritization of tar-
Ataminimum,tumorswithahighprevalenceofpotentiallyac- getedtherapytreatment.However,shouldcliniciansandpatients
tionablemutations,suchasNSCLCs,shouldbetestedtoday.Ide- alsoknowaboutthestatusofothermutations,suchaspassenger
ally,tobeclinicallyuseful,patients’tumorsshouldbeprofiledat mutations that are currentlynot thoughtto be biologicallyand
thetimeofdiagnosis,allowingfortreatmentprioritization.How- clinically relevant? What about mutations with prognostic sig-
ever,recentdatademonstratesthattumorsundergoevolutionover nificance?Ormutationsofunknownsignificance?
time, acquiring resistance with new mutations and genomic al- Amoleculardiagnostictestcanhaveanalyticvalidity,clinical
terations (Lovly and Pao 2012). While it is common practice to validity, and clinical utility (Febbo et al. 2011). Briefly, analytic
reassess the molecular status of liquid tumors, invasive biopsy validityatteststothereproducibilityandqualityofthetest.Clin-
techniques make this less practical in solid tumors. Moreover, icalvalidityimpliesthatthemarkeridentifiestwogroupsthatcan
metastaticlesionswithinpatientscanbeheterogeneous,withone bedistinguishedbiologicallyandhavedifferentoutcomes,butthis
2104 GenomeResearch
www.genome.org
Genetically informed lung cancer medicine
observationmay not indicatethat it should be usedfor routine adifficultchallengetosolvegiventhecurrentlevelofclinicalev-
clinicalcare.Clinicalutilityimpliesthathigh-levelevidenceshows idence;however,itwillbecomemoreandmoreimportantasnew
thatuseofthemarkerimprovespatientoutcomesufficientlyto targetsandnewtherapiesemerge.
justifyitsincorporationintoroutineclinicalcare.Asmutationsare Therearealsoseverallimitationswithrespecttotheformatof
uncovered,wewilleventuallyknowalloftheonesassociatedwith thereportsthemselves.Free-textreportsaretypicallyscannedinto
clinicalutility,i.e.,thatcanhelpprioritizetherapy.Untilthattime, the electronic health record (EHR) or transmitted via HL7 mes-
however,wewillneedtobeflexibleandinclusive,butcognizant sagingfromthelaboratoryinformationsystemastextdocuments.
thatmostofthemutationsidentifiedwillbeofunknownsignifi- Free-textreportsandimagefilesarenotcomputablewithrespectto
cance.Oneapproachthatmanyplacesareadoptinginvolvesthe enablingdownstreamclinicaldecisionsupportfor treating phy-
reportingofclinicallyactionable mutations,withstatementsre- sicians.However,therearealimitednumberoflaboratoryinfor-
garding (1) whether a mutation has been linked to an FDA- mation systems that have adequate structuredrepresentation of
approveddrugwithinthespecificdiseasebeingtested,(2)whether genetictestingresults.OrganizationssuchasHL7areworkingin
a mutation has been linked to an FDA-approved drug within conjunction with molecular diagnostic and testing societies to
adifferentdiseasefromthatbeingtested(sinceeffectivetargeted develop an information model to adequately represent genetic
treatmentinonediseasedoesnotalwaystranslatetoanother),(3) testing results. However, the challenges with respect to next-
whetheramutationhasbeenlinkedtoanon-FDAapproveddrug generationsequencingtechnologiesaregreatinthisregard.
inpreclinicalmodelsorearlyphaseclinicaltrials,and(4)whether Finally,theinterpretivereportsarestaticdocumentsthatare
amutationhasnotbeenlinkedtoanydrug. onlyrelevantatamomentintime,arequirementformedical-legal
purposes.Whiletheresultsofthemutationtestingforagivendate
willnotchange(e.g.,EGFRT790Mmutationdetectedonspecimen
Howwilltumormutationresultsbereported?
fromJanuary15,2011),theinterpretationoftheclinicalsignifi-
Currentmethodsforreportingtumorgenemutationtestingresults canceoftheresultswillchangeovertime.Thus,anopportunity
for clinical use have several limitations with respect to report exists to separate the static from the dynamic components of
contentandformat.Today,mostmoleculardiagnosticslaborato- moleculardiagnosticreports.Thecurrentapproachofonereport
riesperformtumorgenemutationtestingononeortwogenesat foronegenedoesnotscalewhencomprehensivelyreportingthe
atime,andonlytestforthefewmostcommonmutationswithin results for multiple genes and mutations. Thus, multiparameter
thatgene.Thereportcontentconsistsofafull-pagetextdescription datavisualizationandpresentationrepresentanothersignificant
of each gene tested. The report contains static information re- challenge.
gardingthetestingtechniqueandtheclinicalsignificanceofthe
genemutationdetectedwithrespecttodrugtherapiesforapar-
Howwilltumormutationresultsbeusedinclinical
ticulardisease(i.e.,‘‘tumorswithEGFRexon19deletionaresen-
decisionsupportsystems(CDSSs)?
sitivetoEGFRTKI’sinnon-smallcelllungcancer’’).
While reports typically contain one or two references to Beyond the traditional approach of providing decision support
supportthesetypesofhigh-levelstatementsregardingdrugsen- throughinterpretivereports,opportunitiesexisttoutilizetumor
sitivity,theymaynothavethemostup-to-dateinformation.The genemutationtestingresultsdirectlyinclinicaldecisionsupport
referencesareessentialtocliniciansinthattheyprovideimportant systemsforcancertreatmentplanselection.CDSSshaverecently
informationregardingtheappropriateclinicalcontextforuseof been developed to support pathway and guideline compliance
thesedrugs(e.g.,inthemetastaticoradjuvantsetting).Thislevel (Neubaueretal.2010).Thesetoolsmaybestandaloneapplications
ofdetailassiststhetreatingphysicianinmakingactionabletreat- orintegrateddirectlyintoelectronichealthrecords.Theyprovide
mentdecisionstakingintoaccountthetumorgenemutationre- recommendations for treatment selection based on multiple tu-
sults.Evenwhentheinformationismostcurrent,itisonlyrelevant morandpatientfeatures.ClinicalpracticeguidelinesforASCOand
forthatmomentintimewhenthereportiscreated.Sixmonths the National Cancer Comprehensive Network (NCCN) already
later, when the patient’s tumor has progressed and a new treat- recommendEGFRtestinginlungcancerasstandardofcare,but
mentisconsidered,theinterpretationintheoriginalreportmay theyprovidelittlesupportforvariantlevelinterpretationofresults
nolongerbeup-to-dategiventhepaceatwhichnewknowledgein toguidetreatment.ItisanticipatedthatCDSSswillutilizethese
this area is emerging. How will providers and patients remain newtumorfeaturesintheirrecommendationsfortreatmentpri-
updatedregardingchangesintheinterpretationofresultsasnew oritization,butexactlyhowremainstobeseen.
evidenceemerges?
Inaddition,thesereportslackinformationregardingtumor/
Howwillknowledgebasesthatcatalogtheassociationsbetween
genemutation-directedclinicaltrials,oneofthemainactionable
mutationsandclinicaloutcomesbemaintainedandfunded?
treatment decisions that can result from this type of molecular
testing today. However, clinical trial availability changes even From the above discussion, it is clear that knowledge bases are
morerapidlythantheclinicalevidence.Evenwhenopenclinical neededtosupportclinicians,molecularpathologists,andpatients
trialsareincludedintheinterpretivereportatthetimeitisgen- inunderstandingtheevolvingclinicalsignificanceoftumorgene
erated,howwillprovidersandpatientsbecomenotifiedwhennew mutations. Who will fund these knowledge bases and who will
trialsareavailablethatmayberelevanttothepatient’scare? maintainthem?Withsomemutationsoccurringatafrequencyof
Furthermore, the current approach of reporting the signifi- <1%,clinicaltrialscannotbeexpectedtoanswereveryquestion
canceofonegeneatatimedoesnotaddresstheneedtoaccount regarding the significance of rare mutations. Similarly, observa-
forhowmultiplegenesaffectthesensitivityofaparticulardrugor tionalcohortstudiesandcasereportsintheliteraturecanprovide
drugregimen.Thisisinpartduetothegene-orientedviewofmost astartingpointforknowledgeextraction,butalsowillnotprovide
reports as opposed to a drug-oriented view where the results of sufficient power to account for rare mutations and multiple si-
multiplegenescanbetakenintoaccountsimultaneously.Thisis multaneousmutations.Largedatabasesofpatientlevelcasesfrom
GenomeResearch 2105
www.genome.org
Levy et al.
multipleinstitutionswillbeneededtoaggregateknowledgesuf- aboveissues,wemadethekeydecisiontodecouplereportingof
ficientfordiscoveryandinterpretationofclinicalsignificance.We thetumorgenemutationtestingresultsfromtheinterpretationof
havestartedonesuchdatabaseforEGFRmutationsinlungcancer theirclinicalsignificance.Tumorgenemutationresultsarenow
(Hornetal.2011). reportedasdiscretedataelementsinthelaboratoryinformation
system.Eachgenemutationinthepaneliscodedandreportedas
Implementation of genome directed cancer ‘‘Detected’’ or ‘‘Not Detected’’ which makes this information
computable. The results are then transmitted to the electronic
treatment at Vanderbilt
healthrecordandvisualizedinmultiplewaysthataggregateand
The Vanderbilt-Ingram Cancer Center was faced with the chal- abstract the results to enable rapid review of a large amount of
lengesdiscussedabovewhenweimplemented‘‘first-generation’’ information (Fig. 3). Finally, individual results are linked to a
(SNaPshot-based)tumorgenemutationtestingforNSCLCinour knowledgebasecalled‘‘MyCancerGenome’’thatprovidesinfor-
CLIA-certified molecular diagnostics lab in July 2010 (Su et al. mationontheclinicalsignificanceofspecificgenemutationsfora
2011).Initially,weneededtodeterminewhichtumorstotestand particulardiseaseincludinginformationongene-associatedclini-
when.WechoseNSCLCandmelanomabecausetheyhadaction- caltrialsopenatVanderbiltandworldwide.TheinformationinMy
ablemutationsaspartofstandardofcareandactiveclinicaltrials. CancerGenomeiscontinuouslyupdated,creatingadynamicin-
Tumorsaretestedatdiagnosisandattimeofrebiopsyafterdisease terpretation for clinicians who may review the results several
recurrence or progression. We took a practical and economic ap- monthslaterasthepatient’sstatechanges.Thiscombinationof
proachtoselectionofthemoleculardiagnostictestingplatformin reportingstatictestingresultsasdiscretedata,abstractdatavisu-
choosingtouseanABISNaPshotplatformthatwasalreadyinusein alization,andlinkstodynamicknowledgeserviceshasprovided
ourCLIAcertifieddiagnosticlab.Giventhatalimitednumberof aninitialscalablesolutionatourinstitution.Inadditiontobeing
mutationscouldbetestedsimultaneouslyusingthisplatform(;40), integrated into Vanderbilt’s EHR, My Cancer Genome is freely
wechosetocreatedisease-specifictestingpanelsforlungcancerand availableonline(www.mycancergenome.org)forpublicuse.
melanoma.Weselectedvariantstotestthatwereprevalentinatleast
1%oftumorsineachrespectivedisease.Thishelpedmaximizethe Conclusion and future directions
likelihoodofamutationbeingdetectedusingthisassay.
Faced with reporting over 40 different mutations in nine Antimicrobialtestingtodayistakenforgrantedasaroutinepartof
genesforlungcancer,weneededanalternatesolutiontothetra- medicine,buthistorysuggeststhatitactuallytookalongtimeto
ditionalone-geneone-reportapproachtoreporting.Toaddressthe become standardized. In 1928, Alexander Fleming discovered
Figure3. Screenshotvisualizingtumorgenemutationresultsintheelectronichealthrecordlinkedtodecisionsupport.(A)Intheproviderdashboard,
patientsarelistedinseparaterowswiththeirrespectivegenemutationresultsshownascoloredindicatorsineachcolumn(yellow:mutationdetected;
gray:mutationnotdetected).‘‘H-GFR’’and‘‘H-SLP’’refertothespecificdiagnostictestsperformed,asnamedbythemoleculardiagnosticlab,whilethe
remainingninecolumnsrefertothespecificgenesassessedformutations.Theuserclicksonaspecificgeneindicatortoviewthespecificgenemutation
resultsshowninB.GenemutationsarereportedusingtheREFSEQnomenclaturewithresultsreportedasDetectedorNotDetected.Theuserclicksonthe
specificgenemutationandistakentoC.Summaryoftheclinicalsignificanceforthespecificgenemutationinthepatient’sspecificcancerdiagnosis.
2106 GenomeResearch
www.genome.org
Genetically informed lung cancer medicine
penicillin, markingthe start of modern antibiotics.In the same BroseMS,VolpeP,FeldmanM,KumarM,RishiI,GerreroR,EinhornE,
decade,hemadeacontributiontomodernantibioticsusceptibility HerlynM,MinnaJ,NicholsonA,etal.2002.BRAFandRASmutationsin
humanlungcancerandmelanoma.CancerRes62:6997–7000.
testing by developing a forerunner of contemporary minimum ChenHY,YuSL,ChenCH,ChangGC,ChenCY,YuanA,ChengCL,Wang
inhibitoryconcentrationmethodology.Diffusionmethodsofan- CH,TerngHJ,KaoSF,etal.2007.Afive-genesignatureandclinical
tibioticsusceptibilitytestingweredevelopedfurtherintheensuing outcomeinnon-small-celllungcancer.NEnglJMed356:11–20.
DeGreveJ,TeugelsE,GeersC,DecosterL,GaldermansD,DeMeyJ,Everaert
decades,butitwasnotuntil1975,whenaspecificmethod(the
H,UmeloI,VeldPI,SchallierD.2012.Clinicalactivityofafatinib(BIBW
Kirby-Bauerdiscdiffusiontechnique)becamethebasisforarou- 2992)inpatientswithlungadenocarcinomawithmutationsinthe
tine standardized national method accredited by the National kinasedomainofHER2/neu.LungCancer76:123–127.
Dias-SantagataD,AkhavanfardS,DavidSS,VernovskyK,KuhlmannG,
CommitteeforLaboratoryStandards(Wheat2001).Suchantibi-
BoisvertSL,StubbsH,McDermottU,SettlemanJ,KwakEL,etal.2010.
oticsusceptibilitytestingreportsarethestapleofmedicinetoday, Rapidtargetedmutationalanalysisofhumantumours:Aclinical
allowing physicians to make rational and effective choices re- platformtoguidepersonalizedcancermedicine.EMBOMolMed2:
146–158.
gardingtheuseofantibioticstotreatinfectionsarisingfromdi-
DingL,GetzG,WheelerDA,MardisER,McLellanMD,CibulskisK,Sougnez
verseorigins(e.g.,gram-positivecocci,gram-negativerods,etc.) C,GreulichH,MuznyDM,MorganMB,etal.2008.Somaticmutations
Analogously, decades of research designed to elucidate the affectkeypathwaysinlungadenocarcinoma.Nature455:1069–1075.
geneticbasisofcancershavenowrevealedmultipleclinicallyrel- DoH,SolomonB,MitchellPL,FoxSB,DobrovicA.2008.Detectionofthe
transformingAKT1mutationE17Kinnon-smallcelllungcancerby
evantmolecularsubsetsofdisease.Thesediscoveriesareleadingto highresolutionmelting.BMCResNotes1:14.doi:10.1186/1756-0500-
anewtaxonomyofcancer,whichholdsthepromiseofrational 1-14.
rather than empiric approaches for treatment. Lung cancer cur- DrukerBJ,TalpazM,RestaDJ,PengB,BuchdungerE,FordJM,LydonNB,
KantarjianH,CapdevilleR,Ohno-JonesS,etal.2001.Efficacyandsafety
rently represents one of the best paradigms for this new shift.
ofaspecificinhibitoroftheBCR-ABLtyrosinekinaseinchronicmyeloid
Tumorscannowalreadybetestedformultipledifferent‘‘driver’’ leukemia.NEnglJMed344:1031–1037.
mutationsatatime,helpingtoprioritizetheuseoftargetedther- DuttA,RamosAH,HammermanPS,MermelC,ChoJ,SharifniaT,Chande
A,TanakaKE,StranskyN,GreulichH,etal.2011.Inhibitor-sensitive
apiesmostlikelytobenefitindividualpatients.However,weare
FGFR1amplificationinhumannon-smallcelllungcancer.PloSOne6:
justatthebeginningofthisnewparadigm.Greateffortsstillneed e20351.doi:10.1371/journal.pone.0020351.
tobemadetoovercomethemanychallengescurrentlyfacingthe EngelmanJA,ZejnullahuK,MitsudomiT,SongY,HylandC,ParkJO,
LindemanN,GaleCM,ZhaoX,ChristensenJ,etal.2007.MET
field. When comprehensive mutational profiling becomes effi-
amplificationleadstogefitinibresistanceinlungcancerbyactivating
cient, cost-effective, and easily interpretable by practicing clini- ERBB3signaling.Science316:1039–1043.
cians,weareconfidentthatjustlikeantibioticsusceptibilitytest- FebboPG,LadanyiM,AldapeKD,DeMarzoAM,HammondME,HayesDF,
ing,itwillbecomeapartofroutinecancercare. IafrateAJ,KelleyRK,MarcucciG,OginoSetal.2011.NCCNTaskForce
report:Evaluatingtheclinicalutilityoftumormarkersinoncology.J
NatlComprCancNetw9:S1–S32.
GautschiO,PauliC,StrobelK,HirschmannA,PrintzenG,AebiS,DieboldJ.
Competing interest statement
2012.ApatientwithBRAFV600Elungadenocarcinomarespondingto
vemurafenib.JThoracOncoldoi:10.1097/JTO.0b0.13e318262903.
William Pao reports consulting for MolecularMD, AstraZeneca, HannaN,ShepherdFA,FossellaFV,PereiraJR,DeMarinisF,vonPawelJ,
Bristol-MyersSquibb,SymphonyEvolution,andClovisOncology. GatzemeierU,TsaoTC,PlessM,MullerT,etal.2004.Randomizedphase
IIItrialofpemetrexedversusdocetaxelinpatientswithnon-small-cell
His lab has received research funding from Enzon, Xcovery,
lungcancerpreviouslytreatedwithchemotherapy.JClinOncol22:
AstraZeneca,andSymphogen.RightstoEGFRT790Mtestingwere 1589–1597.
licensedonhisbehalfandotherstoMolecularMDbyMSKCC.The HolmanCM.2012.Debunkingthemyththatwhole-genomesequencing
infringesthousandsofgenepatents.NatBiotechnol30:240–244.
otherauthorshavenoconflictstodisclose.
HornL,ChenH,LovlyCM,AndrewsJ,YehP,LevyMA,PaoW.2011.
DIRECT:DNA-mutationInventorytoRefineandEnhanceCancer
Treatment—Acatalogueofclinicallyrelevantsomaticmutationsinlung
Acknowledgments cancer.InJClinOncol29,Supplement,abstractno.7575.2011ASCO
AnnualMeeting,Chicago,Illinois.
WeacknowledgesupportfromtheVICCSpecializedProgramof InternationalUnionAgainstCancer(UICC).1968.TNMclassificationof
ResearchExcellenceinLungCancergrant(CA90949),theVanderbilt- malignanttumours.Geneva,Switzerland.
IngramCancerCenterCoregrant(CA68485),theVFoundation, JensenK,MurrayF.2005.Intellectualpropertylandscapeofthehuman
genome.Science310:239–240.
theTJMartellFoundation,theRobertJ.KlebergJr.andHelenC.
JuYS,LeeWC,ShinJY,LeeS,BleazardT,WonJK,KimYT,KimJI,KangJH,
Kleberg Foundation, and an anonymous donor. We thank John SeoJS.2012.AtransformingKIF5BandRETgenefusioninlung
Iafrate and Dora Dias-Santagata at MGH for assistance with es- adenocarcinomarevealedfromwhole-genomeandtranscriptome
tablishing the lung and melanoma SNaPshot assays. This work sequencing.GenomeRes22:436–445.
KawanoO,SasakiH,EndoK,SuzukiE,HanedaH,YukiueH,KobayashiY,
would not have been possible without contributions from the
YanoM,FujiiY.2006.PIK3CAmutationstatusinJapaneselungcancer
entire Personalized Cancer Medicine Initiative and My Cancer patients.LungCancer54:209–215.
GenometeamsatVanderbilt. KeanS.2011.Humangenome10thanniversary.TheHumanGenome
Author contributions: M.A.L., C.M.L., and W.P. conceived, (patent)Project.Science331:530–531.
KobayashiS,BoggonTJ,DayaramT,JannePA,KocherO,MeyersonM,
wrote,andapprovedthemanuscript.
JohnsonBE,EckMJ,TenenDG,HalmosB.2005.EGFRmutationand
resistanceofnon-small-celllungcancertogefitinib.NEnglJMed352:
References 786–792.
KohnoT,IchikawaH,TotokiY,YasudaK,HiramotoM,NammoT,Sakamoto
H,TsutaK,FurutaK,ShimadaY,etal.2012.KIF5B-RETfusionsinlung
AndrewsJ,YehP,PaoW,HornL.2011.Molecularpredictorsofresponseto adenocarcinoma.NatMed18:375–377.
chemotherapyinnon-smallcelllungcancer.CancerJ17:104–113. KoivunenJP,MermelC,ZejnullahuK,MurphyC,LifshitsE,HolmesAJ,
BergethonK,ShawAT,IgnatiusOuSH,KatayamaR,LovlyCM,McDonald ChoiHG,KimJ,ChiangD,ThomasR,etal.2008.EML4-ALKfusiongene
NT,MassionPP,Siwak-TappC,GonzalezA,FangR,etal.2012.ROS1 andefficacyofanALKkinaseinhibitorinlungcancer.ClinCancerRes
rearrangementsdefineauniquemolecularclassoflungcancers.JClin 14:4275–4283.
Oncol30:863–870. KwakEL,BangYJ,CamidgeDR,ShawAT,SolomonB,MakiRG,OuSH,
BleekerFE,FelicioniL,ButtittaF,LambaS,CardoneL,RodolfoM,ScarpaA, DezubeBJ,JannePA,CostaDB,etal.2010.Anaplasticlymphoma
LeenstraS,FrattiniM,BarbareschiM,etal.2008.AKT1E17Kinhuman kinaseinhibitioninnon-small-celllungcancer.NEnglJMed363:
solidtumours.Oncogene27:5648–5650. 1693–1703.
GenomeResearch 2107
www.genome.org
Levy et al.
LanderES,LintonLM,BirrenB,NusbaumC,ZodyMC,BaldwinJ,DevonK, GeneticpredictorsofMEKdependenceinnon-smallcelllungcancer.
DewarK,DoyleM,FitzHughW,etal.2001.Initialsequencingand CancerRes68:9375–9383.
analysisofthehumangenome.Nature409:860–921. ReckM,vonPawelJ,ZatloukalP,RamlauR,GorbounovaV,HirshV,Leighl
LeeW,JiangZ,LiuJ,HavertyPM,GuanY,StinsonJ,YueP,ZhangY,PantKP, N,MezgerJ,ArcherV,MooreN,etal.2009.PhaseIIItrialofcisplatinplus
BhattD,etal.2010.Themutationspectrumrevealedbypairedgenome gemcitabinewitheitherplaceboorbevacizumabasfirst-linetherapy
sequencesfromalungcancerpatient.Nature465:473–477. fornonsquamousnon-small-celllungcancer:AVAil.JClinOncol27:
LiC,FangR,SunY,HanX,LiF,GaoB,IafrateAJ,LiuXY,PaoW,ChenH, 1227–1234.
etal.2011.Spectrumofoncogenicdrivermutationsinlung RielyGJ,KrisMG,RosenbaumD,MarksJ,LiA,ChitaleDA,NafaK,Riedel
adenocarcinomasfromEastAsianneversmokers.PLoSONE6:e28204. ER,HsuM,PaoW,etal.2008.Frequencyanddistinctivespectrumof
doi:10.1371/journal.pone.0028204. KRASmutationsinneversmokerswithlungadenocarcinoma.Clin
LipsonD,CapellettiM,YelenskyR,OttoG,ParkerA,JaroszM,CurranJA, CancerRes14:5731–5734.
BalasubramanianS,BloomT,BrennanKW,etal.2012.Identificationof RielyGJ,MarksJ,PaoW.2009.KRASmutationsinnon-smallcelllung
newALKandRETgenefusionsfromcolorectalandlungcancerbiopsies. cancer.ProcAmThoracSoc6:201–205.
NatMed18:382–384. RoychowdhuryS,IyerMK,RobinsonDR,LonigroRJ,WuYM,CaoX,
LovlyCM,PaoW.2012.EscapingALKinhibition:Mechanismsofand Kalyana-SundaramS,SamL,BalbinOA,QuistMJ,etal.2011.
strategiestoovercomeresistance.SciTranslMed4:120ps2.doi:10.1126/ Personalizedoncologythroughintegrativehigh-throughputsequencing:
scitranslmed.3003728. apilotstudy.SciTranslMed3:doi:10.1126/scitranslmed.3003261.
LynchTJ,BellDW,SordellaR,GurubhagavatulaS,OkimotoRA,Brannigan SandlerA,GrayR,PerryMC,BrahmerJ,SchillerJH,DowlatiA,LilenbaumR,
BW,HarrisPL,HaserlatSM,SupkoJG,HaluskaFG,etal.2004.Activating JohnsonDH.2006.Paclitaxel-carboplatinaloneorwithbevacizumab
mutationsintheepidermalgrowthfactorreceptorunderlying fornon-small-celllungcancer.NEnglJMed355:2542–2550.
responsivenessofnon-small-celllungcancertogefitinib.NEnglJMed SchillerJ,HarringtonD,BelaniC,LangerC,SandlerA,KrookJ,ZhuJ,
350:2129–2139. JohnsonD,GroupTECO.2002.Comparisonoffourchemotherapy
MacconaillLE,GarrawayLA.2010.Clinicalimplicationsofthecancer regimensforadvancednon-small-celllungcancer.NEnglJMed346:
genome.JClinOncol28:5219–5228. 92–98.
MaemondoM,InoueA,KobayashiK,SugawaraS,OizumiS,IsobeH, SequistLV,HeistRS,ShawAT,FidiasP,RosovskyR,TemelJS,LennesIT,
GemmaA,HaradaM,YoshizawaH,KinoshitaI,etal.2010.Gefitinibor DigumarthyS,WaltmanBA,BastE,etal.2011.Implementing
chemotherapyfornon-small-celllungcancerwithmutatedEGFR.N multiplexedgenotypingofnon-small-celllungcancersintoroutine
EnglJMed362:2380–2388. clinicalpractice.AnnOncol22:2616–2624.
MalangaD,ScrimaM,DeMarcoC,FabianiF,DeRosaN,DeGisiS,MalaraN, ShawAT,YeapBY,Mino-KenudsonM,DigumarthySR,CostaDB,HeistRS,
SavinoR,RoccoG,ChiappettaG,etal.2008.ActivatingE17Kmutation SolomonB,StubbsH,AdmaneS,McDermottU,etal.2009.Clinical
inthegeneencodingtheproteinkinaseAKT1inasubsetofsquamous featuresandoutcomeofpatientswithnon-small-celllungcancerwho
cellcarcinomaofthelung.CellCycle7:665–669. harborEML4-ALK.JClinOncol27:4247–4253.
MarksJL,GongY,ChitaleD,GolasB,McLellanMD,KasaiY,DingL,Mardis SheddenK,TaylorJM,EnkemannSA,TsaoMS,YeatmanTJ,GeraldWL,
ER,WilsonRK,SolitD,etal.2008.NovelMEK1mutationidentifiedby EschrichS,JurisicaI,GiordanoTJ,MisekDE,etal.2008.Gene
mutationalanalysisofepidermalgrowthfactorreceptorsignaling expression-basedsurvivalpredictioninlungadenocarcinoma:amulti-
pathwaygenesinlungadenocarcinoma.CancerRes68:5524–5528. site,blindedvalidationstudy.NatMed14:822–827.
MitsudomiT,MoritaS,YatabeY,NegoroS,OkamotoI,TsurutaniJ,SetoT, SiegelR,WardE,BrawleyO,JemalA.2011.Cancerstatistics,2011:The
SatouchiM,TadaH,HirashimaT,etal.2009.Gefitinibversuscisplatin impactofeliminatingsocioeconomicandracialdisparitieson
plusdocetaxelinpatientswithnon-small-celllungcancerharbouring prematurecancerdeaths.CACancerJClin61:212–236.
mutationsoftheepidermalgrowthfactorreceptor(WJTOG3405):An SodaM,ChoiYL,EnomotoM,TakadaS,YamashitaY,IshikawaS,FujiwaraS,
openlabel,randomisedphase3trial.LancetOncol11:121–128. WatanabeH,KurashinaK,HatanakaH,etal.2007.Identificationofthe
MokTS,WuYL,ThongprasertS,YangCH,ChuDT,SaijoN,Sunpaweravong transformingEML4-ALKfusiongeneinnon-small-celllungcancer.
P,HanB,MargonoB,IchinoseY,etal.2009.Gefitiniborcarboplatin- Nature448:561–566.
paclitaxelinpulmonaryadenocarcinoma.NEnglJMed361:947–957. SonobeM,ManabeT,WadaH,TanakaF.2006.Lungadenocarcinoma
NeubauerMA,HovermanJR,KolodziejM,ReismanL,GruschkusSK,HoangS, harboringmutationsintheERBB2kinasedomain.JMolDiagn8:
AlvaAA,McArthurM,ForsythM,RothermelT,etal.2010.Cost 351–356.
effectivenessofevidence-basedtreatmentguidelinesforthetreatmentof SuZ,Dias-SantagataD,DukeM,HutchinsonK,LinYL,BorgerDR,Chung
non-small-celllungcancerinthecommunitysetting.JOncolPract6:12–18. CH,MassionPP,Vnencak-JonesCL,IafrateAJ,etal.2011.Aplatformfor
PaezJG,JannePA,LeeJC,TracyS,GreulichH,GabrielS,HermanP,KayeFJ, rapiddetectionofmultipleoncogenicmutationswithrelevanceto
LindemanN,BoggonTJ,etal.2004.EGFRmutationsinlungcancer: targetedtherapyinnon-small-celllungcancer.JMolDiagn13:74–84.
Correlationwithclinicalresponsetogefitinibtherapy.Science304: SunY,RenY,FangZ,LiC,FangR,GaoB,HanX,TianW,PaoW,ChenH,
1497–1500. etal.2010.LungadenocarcinomafromEastAsiannever-smokersis
PaikPK,ArcilaME,FaraM,SimaCS,MillerVA,KrisMG,LadanyiM,RielyGJ. adiseaselargelydefinedbytargetableoncogenicmutantkinases.JClin
2011.Clinicalcharacteristicsofpatientswithlungadenocarcinomas Oncol28:4616–4620.
harboringBRAFmutations.JClinOncol29:2046–2051. TakeuchiK,SodaM,TogashiY,SuzukiR,SakataS,HatanoS,AsakaR,
PaoW,ChmieleckiJ.2010.Rational,biologicallybasedtreatmentofEGFR- HamanakaW,NinomiyaH,UeharaH,etal.2012.RET,ROS1andALK
mutantnon-small-celllungcancer.NatRevCancer10:760–774. fusionsinlungcancer.NatureMed18:378–381.
PaoW,HutchinsonKE.2012.Chippingawayatthelungcancergenome. TsaoAS,TangXM,SabloffB,XiaoL,ShigematsuH,RothJ,SpitzM,Hong
NatMed18:349–351. WK,GazdarA,WistubaI.2006.Clinicopathologiccharacteristicsofthe
PaoW,MillerV,ZakowskiM,DohertyJ,PolitiK,SarkariaI,SinghB,Heelan EGFRgenemutationinnon-smallcelllungcancer.JThoracOncol1:
R,RuschV,FultonL,etal.2004.EGFreceptorgenemutationsare 231–239.
commoninlungcancersfrom‘‘neversmokers’’andareassociatedwith VenterJC,AdamsMD,MyersEW,LiPW,MuralRJ,SuttonGG,SmithHO,
sensitivityoftumorstogefitinibanderlotinib.ProcNatlAcadSci101: YandellM,EvansCA,HoltRA,etal.2001.Thesequenceofthehuman
13306–13311. genome.Science291:1304–1351.
PaoW,MillerVA,PolitiKA,RielyGJ,SomwarR,ZakowskiMF,KrisMG, WeinsteinI.2002.Cancer.Addictiontooncogenes–theAchilleshealof
VarmusH.2005a.Acquiredresistanceoflungadenocarcinomasto cancer.Science297:63–64.
gefitiniborerlotinibisassociatedwithasecondmutationintheEGFR WeirBA,WooMS,GetzG,PernerS,DingL,BeroukhimR,LinWM,Province
kinasedomain.PLoSMed2:e73.doi:10.1371/journal.pmed.0020073. MA,KrajaA,JohnsonLA,etal.2007.Characterizingthecancergenome
PaoW,WangTY,RielyGJ,MillerVA,PanQ,LadanyiM,ZakowskiMF, inlungadenocarcinoma.Nature450:893–898.
HeelanRT,KrisMG,VarmusHE.2005b.KRASmutationsandprimary WeissJ,SosML,SeidelD,PeiferM,ZanderT,HeuckmannJM,UllrichRT,
resistanceoflungadenocarcinomastogefitiniborerlotinib.PLoSMed2: MenonR,MaierS,SoltermannA,etal.2010.FrequentandfocalFGFR1
e17.doi:10.1371/journal.pmed.0020017. amplificationassociateswiththerapeuticallytractableFGFR1
PaoW,KrisMG,IafrateAJ,LadanyiM,JannePA,WistubaII,Miake-LyeR, dependencyinsquamouscelllungcancer.SciTranslMed2:62ra93.doi:
HerbstRS,CarboneDP,JohnsonBE,etal.2009.Integrationofmolecular 10.1126/scitranslmed.3001451.
profilingintothelungcancerclinic.ClinCancerRes15:5317–5322. WheatPF.2001.Historyanddevelopmentofantimicrobialsusceptibility
PaoW,IafrateAJ,SuZ.2011.Geneticallyinformedlungcancermedicine. testingmethodology.JAntimicrobChemother48(Suppl1):1–4.
JPathol223:230–240. YuSL,ChenHY,ChangGC,ChenCY,ChenHW,SinghS,ChengCL,YuCJ,
PratilasCA,HanrahanAJ,HalilovicE,PersaudY,SohJ,ChitaleD, LeeYC,ChenHS,etal.2008.MicroRNAsignaturepredictssurvivaland
ShigematsuH,YamamotoH,SawaiA,JanakiramanM,etal.2008. relapseinlungcancer.CancerCell13:48–57.
2108 GenomeResearch
www.genome.org