The cancer registry as an ally in monitoring treatment effectiveness

www.journalpulmonology.org

ORIGINAL

ARTICLE

The

cancer

registry

as

an

ally

in

monitoring

treatment

effectiveness

F.A.

Costa

_,

_C.

_Ramos

_,

_R.

_Murteira

_,

_T.

_Almodovar

_,

_J.L.

_{Passos-Coelho}

_,

M.I.

Carvalho

_,

_L.

_Costa

_,

_M.J.

_Brito

_,

_S.

_Ramos

_,

_M.

_Ferreira

_,

_A.C.

_Miranda

b_{InstitutoPortuguêsdeOncologiadeLisboaFranciscoGentil,Portugal} c_{HospitaldaLuz,Portugal}

d_{CentroHospitalardeSãoJoão,Portugal} e_{CentroHospitalardeLisboaNorte,Portugal} f_{HospitalGarciadeOrta,Portugal}

g_{CentroHospitalardeLisboaOcidental,Portugal} h_{HospitalProf.DoutorFernandodaFonseca,Portugal}

Received5January2018;accepted18May2018 Availableonline22October2018

KEYWORDS Immunotherapy; Registries; Lungneoplasms; Treatment effectiveness; Survivalanalysis; Drug-relatedside effectsandadverse reactions

Abstract

Objective: Toevaluateifthecancerregistrydatabasecanbeusedtomonitortreatment effec-tivenessusingnivolumabtreatmentofnon-smallcelllungcancer(NSCLC)asanexample. Method: Anobservationalinceptioncohortwasused,whereallregisteredcasesofNSCLCwith authorisationtoinitiatetreatmentwithnivolumabweremonitoredretrospectivelytoevaluate diseasecharacteristicsandresponsetopriortreatments.Currentexposuretonivolumabwas prospectivelycharacterisedandtreatmentoutcomesclassifiedbasedontheclinicalinformation registeredinthepatientmedicalrecord.Themainoutcomemeasureusedtoassesstreatment effectivenesswasoverallsurvival(OS).Secondaryoutcomesconsideredwereprogressionfree survival(PFS)asameasureofeffectivenessandoccurrenceofAdverseDrugReaction(ADRs)as ameasureofsafety.DatawereanalysedusingSPSS,version24.

Results:Atotalof115patientsreceivedtreatmentwithnivolumabforNSCLC,between Novem-ber 1st 2015and July31st 2016, andwere registered inthe database. The majoritywere non-squamoustype(n=107).ThemedianOSwas11.4months_{CI95%:11.1---11.7},witha1-year survivalof44%,inlinewithclinicaltrialdata.MedianPFSwas5.4months{CI95%:2.8---7.9}. Treatment wasdiscontinued in82cases,mostfrequently duetodisease progression.There were38casesofADRsdocumentedinthepatientmedicalchart,21ofwhichledtotreatment discontinuation.

Abbreviations: OS,overallsurvival;PFS,progression-freesurvival;ADRs,adversedrugreactions.

∗_{Correspondingauthor.}

E-mailaddress:facosta@ipolisboa.min-saude.pt(F.A.Costa). https://doi.org/10.1016/j.pulmoe.2018.05.007

2531-0437/©2018SociedadePortuguesadePneumologia.PublishedbyElsevierEspa˜na,S.L.U.ThisisanopenaccessarticleundertheCC BY-NC-NDlicense(http://creativecommons.org/licenses/by-nc-nd/4.0/).

Conclusion:Theanalyseddatasuggestthatthecancerregistryisapowerfultooltomonitor treatmenteffectiveness,althoughconsiderableinvestmentisneededtoimprovethemedical cultureofrecordingtreatmentexposure,particularlydocumentationofADRs.

©2018SociedadePortuguesadePneumologia.PublishedbyElsevierEspa˜na,S.L.U.Thisisan openaccessarticleundertheCCBY-NC-NDlicense( http://creativecommons.org/licenses/by-nc-nd/4.0/).

Introduction

Lung cancer is one of the leading causes of mortality in Portugalandworldwide.1 _{Thevisibleimprovementsin}

sur-vival achieved during the last decade are likely to be a resultoftheprogressiveincreaseinearlydetectionofthe disease. The effectiveness of the smokingban legislation implementedmayhavehadarole,althoughtheimpacthas notyetbeenclearlyidentified.2

Apart from surgery, treatments are mainly developed toincrease length of survival but very few lead to cure. Nonetheless,the search for effective treatments remains a hope for all and is of particular interest given the ris-ing costs of marketed cancer treatments.3 _{The highest}

proportionofthecostsforthePortugueseHealthCare Sys-temconstantlycorrespondtothe therapeuticgroups used incancertreatment, includingbiological, immunomodula-tory and cytotoxic ones.4 _{The aforementioned treatment}

related costs are constantly rising and immunotherapy is not only an important factor in this trend but also an area where patient access is sometimes delayed by the standardproceduresimplemented.Ensuringpatient’searly accesstomedicineshaslongbeenasubjectofconcernfor the World Health Organisation (WHO) and for all govern-ments.

Several mechanisms exist to favour early access,such astheuseofspecialutilisationauthorisations(SUA).These arenormallyusedwhenthemedicinesarestillunder eco-nomic evaluation and before negotiations are finalised. Adaptive pathways are a common process through which patientsin Portugalcanget accesstomedicinesahead of full market authorisation,5 _{namely the early access}

pro-gramthatwas recentlyusedfor Nivolumab. Nivolumab is an immunotherapy drug that modulates the programmed death 1 (PD-1) pathway. The binding of the ligands PD-L1 (prevalentin NSCLC) and PD-L2, expressed by tumour cells,totheirreceptorPD-1,expressedbyactivatedTcells, leadstoadownregulationintheimmuneresponse, favour-ingtumourcellsurvival.Nivolumab isan antibodyagainst thePD-1immune-checkpoint-inhibitor,whichbydisrupting thePD-1mediated signalling, promotes a normalimmune response against tumour cells. Nivolumab was approved onthe19thJune 2015 bythe EuropeanMedicines Agency (EMA) following documentation of improvement in OS in clinicaltrialsinadvancedNSCLCpatients.6,7 _Theinclusion

criteria for the early access program were: stage IIIB or IV NSCLC who had progressed on previous chemotherapy treatment; poor expected outcome with available treat-ment options; and expected life expectancy of at least 3months.

These adaptivemechanisms were setup tocreate the environmentforthedifferentstakeholdersinvolvedtomake informeddecisions oninnovative medicines. However,for theseconversationstobemeaningful,realworlddatamust beavailableandobjectivelymonitored.Themostcommon sources of data for effectiveness studies are population based registries. In Portugal, until July 2017, there were four regional cancer registries South, North, Centre and AzoresIslandsandrecentlythecreationofasinglenational registry has been approved.8 _{Currently, the ROR-Sul, the}

cancer registryforSouthernPortugaland MadeiraIslands, hasathoroughcollectionofdataincludingpatient identi-fication,cancerdiagnosis,tumourcharacteristics,detailed lines of treatment and patient vitalstatus. In most insti-tutions, all clinical and medical data is entered into the ROR-Suldatabase.

Given the political and science-based context high-lighted,wesuggestedthatthecancerregistrytraditionally usedtomonitor diseaseincidenceandsurvival,couldalso beusedtomonitorresponsetotreatment.

Methods

Anobservationalinceptioncohortwasused,whereall reg-istered casesof lung cancer withauthorisationtoinitiate treatment with nivolumabin the early accessprogramme thattookplacefromNovember1st2015toJuly31st2016 weremonitoredretrospectively,retrievingdisease charac-teristicsandtypeandresponsetopriortreatment;andalso prospectively to characterise exposure to nivolumab and treatmentoutcomes.

Datasources: The Portuguese Drug Regulatory Agency, INFARMED(AutoridadeNacionaldoMedicamentoeProdutos deSaúde,I.P.)hasadatabaseofSUAgrantedfornivolumab in thecontext ofthe early accessprogramme,whichwas usedtoascertainthetheoreticalstudypopulation.The ROR-Sul database wasthen usedto extractthese cases which comprisedallgrantedSUAthat weretreatedaccordingto theregulator’sinstructions.

Eligibilitycriteriawerethosematchingthe characteris-tics defined by the Regulatory Agencyfor early accessto medication, i.e. patients with a histologically confirmed diagnosis of non-small cell lung cancer, locally advanced ormetastatic. Caseswereselectedbasedonthefollowing topographiccodes, usingthe10thInternational Classifica-tionofDiseaseforOncology9_{:C34.0Mainbronchus;C34.1}

Upperlobe,lung;C34.2Middlelobe,lung;C34.3Lowerlobe, lung;C34.8Overlappinglesionoflung;C34.9Lung,NOS.No additional criteria were defined as we intended to study real world exposure to nivolumab. As such, the detailed

characteristicsofthesampleextractedarepresentedinthe resultssection.

Therapeutic response was classified according to the RECISTcriteria,10_{basedontheinformationavailableonthe}

medicalrecordsandclinicalfiles.

Themainoutcomemeasureusedtoevaluatetreatment effectivenesswasOS.ToestimateOS,theperiodconsidered wasfrom the day of treatment initiation withNivolumab untildeathfromanycause.

PFS was considered a secondary outcome to evaluate treatmenteffectiveness.To estimatePFS,theperiod con-sideredwasalsofromthedayoftreatmentinitiationwith Nivolumab until disease progression. Clinical and imagio-logical progressionswerebothconsidered. Otheroutcome measuresofinterestconsideredwereresponsetotreatment andADRsdocumentedinthemedicalrecord.

Statisticalanalysis:DatawerestoredintheROR-Sul can-cerregistrydatabase,whichwasthenextractedtoExceland subsequently analysed usingSPSSversion24.0 (IBM statis-tics).Analysisfocusedontheunivariatecharacterisationof studysubjects,diseaseandtreatmentcharacteristics. Sub-analysis focused on different subgroups, such asstage of diseaseattreatmentinitiationandhistology(squamous ver-susnon-squamous).Survivalanalysiswasusedtoascertain thetimegrantedbysubmittingthepatienttotherapy, mea-suring the time from the day of starting treatment with Nivolumab until the event (death or disease progression, respectivelyforOSandPFS).One-yearsurvivalwith95%CI wasusedforcomparisonwithpublishedclinicaltrialdata.6,7

Theprimaryreferenceconsideredfocusesonnon-squamous carcinomas,6 _{and as such the analysis was subsequently}

restrictedtosuchcasesforthepurposesofcomparison.

Results

Samplecharacteristics

Atotalof225SUAweregrantedbyINFARMED,froma max-imalof250initiallyestablished.These225SUEweregiven to 217 patients in 32 different institutions, both private andpublic,throughoutthecountry.Theaveragetimefrom submission of request by the institutions’ Pharmacy and TherapeuticsCommitteeuntildecisionbytheregulatorwas 6.5±5.1days.

The majorityofthesecaseswereregisteredinROR-Sul database(n=115;53.0%),althoughmostSUAs(84.3%)which enabledanalysisoftreatmenteffectivenessoriginatedfrom theSouthregionofPortugal.

Most cases were male patients (n=73; 63.5%), with a medianageatdiagnosisof62years,withsignificant differ-encebetweengenders[male64.0,female58.5(p=0.017)]. MostcaseswerestageIVNSCLCatdiagnosis(n=63;54.8%). Themajoritywereadenocarcinomas(n=104;90.4%),and49 (42.6%)wereconfinedtotheupperlobeofthelung.

Priortotreatmentwithnivolumab,51.3%ofpatientshad beentreatedwithradiotherapy(n=59),31.3%withsurgery (n=36)and100%withchemotherapy(n=115).Mostpatients weretreatedwithtwo(n=41;35.6%),threeormore(n=39; 33.9%)priorlinesofchemotherapy andtheremaininghad receivedonly one line of treatment (n=34; 29.6%). In 98 patients (85.2%) nivolumab was givento treat metastatic

cancer. As nivolumab is only indicated for metastatic or locallyadvanceddisease,themediantimeuntiltreatment initiationwasdetermineddifferently,accordingtothestage atdiagnosis.ForstagesIIIBandIV,thedateofdiagnosiswas usedinthiscalculation;forstagesI,IIandIIIA,thedateused wasthatofrecurrenceand/orprogressiontostagesIIIBorIV. Hence,themediantimefromdiagnosisofadvanceddisease toinitiationoftreatmentwithnivolumabwas1.2years±1.5 {CI95%:27days---7.6years}.

Median duration of treatment with nivolumab was 5.6±4.5months,and33patients(28.7%)werestill under-goingtreatmentatthetimeofdatabaseclosure(May31st 2017).Thedoseadministeredattreatmentinitiationranged from 100 to 351mg (M=198.8±44mg) and was adminis-tered every 15 days, as approved, in 93% of cases. The completenessoftheregistryforthesetwotreatment varia-bles(notmandatoryintheregistry)wererespectively94.8% and93.0%.

Secondaryoutcomes

Ofthe 82 patients whostopped treatment, in 76 (92.7%) thereasonfortreatmentwithdrawalwasdocumentedinthe medicalrecord.Mostpatientsdiscontinuedtreatmentdue todisease progression, although a high proportion didso asaresultofadversedrugreactions.Atotalof38adverse drugreactions were documented, amongwhich 20 led to treatmentdiscontinuation(Table1).

Responsetotreatmentwithnivolumabwasevaluatedin allpatients:44(38.3%)haddiseaseprogression,40(34.8%) stabledisease (34.8%), 18 partial response and one com-pleteresponse.Itwasnotpossibletodetermineresponseto treatmentfor12ofthepatients,tenofwhomdiscontinued treatment.

Disease progression was documented for 70 patients (60.9%). Median PFS was 165 days {CI95%: 83.3---246.7}, equivalentto5.4 months.The proportionofpatients who wereprogressionfreeatone-yearwas36%.

Primaryoutcomes

From the 115 patients that started treatment with nivolumab,56havedied(48.7%).MedianOSconsideringthis entiresample was11.4monthswithaone-yearOSof44%. TheconfidenceintervalforthemedianOSwasnotpossible toestimate.

Restricting the analysis topatients with non-squamous carcinomas(n=107), for purposesof comparisonwiththe clinicaltrialsleadingtodrugapproval,therewere53deaths (49.5%). Median OS in this subgroup was the same, 11.4 months,alsowiththesameone-yearOS,44%(Fig.1).

MedianPFSintheoverallsample(n=115)was5.4months {CI95%:2.7---8.1},slightlyreducedwhenconsideringthe non-squamoussubgroup(n=107)to5.3months{CI95%:2.8---7.9}. One-yearPFS for thenon-squamous subgroup was36%. In Table 2, the comparison of treatment response is made undercomparableconditionstotheclinicaltrial,i.e.,only considering non-squamous carcinomas(n=107) andin the case of therapeutic response evaluation, only considering thosetreatedfor9weeksorover(n=76),anotealso indi-catedinthetablefooter(Table2).

Table1 Causesoftreatmentinterruptionanddistribution ofadversedrugreactions.

Adverseeffectsleadingto treatmentinterruption

Diseaseprogression 40

Adversedrugreactions 20

Death 12

Patientrefusal 2

Detectionofprogressionofanothertumour 1

Hospitalisationforotherreasons 1

Unknown 6

Total 82

Adversedrugreactions Leadingto treatment interruption

Overall

Pneumonitis 4 7

Hypothyroidism 0 6

Shortnessofbreathand respiratoryproblems

3 4

Gastrointestinaltoxicity 2 3

Hematologictoxicity 2 2

Weaknessandasthenia 2 2

Metabolicacidosis 2 2 Cutaneoustoxicity 1 2 Diarrhoea 1 1 Macularrash 0 1 Hepatictoxicity 1 1 Neurologictoxicity 1 1 Hyperthyroidism 1 1 Diabetes 0 1 Pneumothorax 0 1

Deepveinthrombosis 0 1

Lowerlimboedema 0 1

Dermatophytosisand bacterialinfection 0 1 Total 20 38 Survival function non-squamous NSCLC 1,0 0,8 0,6 0,4 0,2 0,0 0 90 180 270 360 450 22 54 70 89 107 At risk Overall survival Survival function Censored Cum ulativ Cum ulativ e sur viv al

Figure 1 Overall survival observed in the cohort of non-squamouslungcancerpatients(n=107)cumulativesurvival.

Table2 Comparisonofclinicaltrialdatawithreal-world effectivenessdata.

Realworlddata (n=107)

Clinicaltrialdata6

Patientdemographicandclinicalcharacteristics

Agemedian {min---max} (years) 65_{37---83} 61_{37---84} ≥75years 13(12.1%) 20(7.0%) Previouschemotherapy Oneregimen 34(29.6%) 256(88.0%) Tworegimens 41(35.6%) 35(12.0%) Threeormore regimensa 39(33.9%) 1(<1.0%) Secondaryoutcomes Adversedrug reactions 38(35.5%) 199(69.0%) Therapeuticresponse(n=76)b Complete response 1(1.3%) 4(1.0%) Partialresponse 14(18.4%) 52(18.0%) Stabledisease 37(48.7%) 74(25.0%) Disease progression 18(23.7%) 129(44.0%) Notevaluated 6(7.9%) 33(11.0%)

Progressionfreesurvival

MedianPFS (95%CI) 5.3months {2.8---7.9} 2.3months {2.2---3.3} 1-yearPFS 36.0% 19.0%_{{14.0---23.0}} Primaryoutcomes Overallsurvival MedianOS (95%CI) 11.4months {11.1---11.7} 12.2{9.7---15.0} 1-yearOS 44.0% 51.0%{45.0---56.0} a _Missing=1.

b _{Therapeuticresponsewascomparedconsideringonlythose} patientstreatedfor9weeksorlonger,asreportedelsewhere.6

Discussion

Theclinicaloutcomesthatwereobservedinthiscohortof patients with NSCLCtreated with nivolumab weresimilar tothosereportedintheclinical trial whichhadledtoits approvalinnon-squamousNSCLCpatients.6_ThemedianOS

observedinrealworlddatawas11.4months,slightly infe-rior to the 12.2 months reported in the clinical trial but stillsuperiortothe9.4monthsreportedfortreatmentwith docetaxel.Worthmentioning,wepresumethatthemedian OSwithdocetaxelinrealworlddatawouldmost likelybe shorter.Furthermore,thecharacteristicsofpatients receiv-ingnivolumab indicatetheyhadworseprognosticfactors, withahigherproportionofpatientsbotholderthan75years andhavingundergonethreeormorelinesofchemotherapy priortotreatmentwithnivolumab.

It is worth noting that although the observed OS was slightlyinferiortothatreportedintheclinicaltrial,thePFS wassuperior(5.3versus2.3months).6_{Oneplausiblereason}

isthatinclinicaltrials,patientsaremorefrequentlysubject toCTs,henceprogressionsaredetectedearlier.Inreallife, CTsareoftendelayed bywaitingtimesassociatedwithan overloadedhealthcaresystem.PFSatone-yearwashigher than in the clinical trial (36.0% versus 19.0%).6 _This

find-ingseemssurprisingandwe believeitcouldbearesultof somestudylimitations,namelythefactthatconsidering dis-easeprogressioncouldbearesultofobservedcomputerised tomographies(CTs),butalsoofclinicaljudgement;also,in reallifevariousoncologistsjudgediseaseprogression,which couldleadtomisclassificationbias.

WehavedemonstratedthatthecancerRegistrydatabase maybesuccessfullyusedtomonitortreatment outcomes, althoughagreateffortneedstobeplacedonensuringthe exhaustivenessofregistereddata.

The main limitation of thisstudy is thescant informa-tiononpatientstreatedin centralandnorth regions.The newlegislation for theNational OncologyRegistry8 _which

anticipatesmandatoryregistrationofcases,asopposedto the voluntary registration observed at the regional level observedsince1988,11_{willcertainlybefundamentalin}

over-comingthis problem.As aconsequence ofthis limitation, the sample is smaller than ideal,limiting our capacityto undertakepotentiallyinterestingsub-analysis.Forexample, itcouldhavebeenusefultocompareOSofpatients under-goingnivolumabassecondlinetreatmentcomparedtothose exposedasthird,fourthorevenfifthlineoftreatment.

The other limitations of this study derived from sometimes insufficient information in medical records. Oncologistsinvolvedinthestudyhadadeterminantrolein thisaspect.Agreateffortwasmadetomotivatecolleagues todocumentin detailall aspectsoftreatment. Tradition-ally,physiciansfocusondeliveringclinicalcareand,dueto timeconstraintsinherenttothehealthsystemstructureand its demands, minimise documentation in patient medical records.Asaresult,somevariablesaremoreoftendetailed than others.Tumour characterisation is oneexample of a setofvariables whereaccuracyis extremelyhigh.In con-trast, outsideclinical trials, treatment characterisationis lessdetailed.Thereneedstobeachangeinmentalityand workorganisationinthisneweraofreallifedatacollection andanalysis.Treatmentmustbeseenasanessential compo-nentoftheregistry.Tofullycharacterisedrugexposure,itis ofutmostimportancetoknowthetreatmentduration,dose delivered,adversedrugreactionsobservedandresponseto treatment.

The proportionofADRs reportedwaslow.This is prob-ably a result of under registration rather than extreme safety of nivolumab. Therefore,results need to be inter-pretedwithcaution.Among registeredADRs, grades were veryseldomdocumented,supportingtheprevious assump-tion. We are currently working in the cancer registry on upgrading the fields for monitoring ADRs to enable the quantification of grades and also to document all observed ADRs and not merely the ones leading to treatment discontinuation. This is particularly important because most of these drugs are subject to additional monitoring according to the pharmacovigilance European legislation.

To the bestof ourknowledge, this isthe firstnational studytoshowthatapopulation-basedcancer registrycan be usedto monitor the effectiveness of new treatments,

providingvaluableevidenceforpolicymakerstobase reim-bursement decisions.In the future, data may be used to determinefullorpartialreimbursementaccordingtoresults fromdifferent effectiveness profiles. As an example, our data suggests that the benefit obtained using nivolumab in non-squamous cell lung cancers is substantially higher thanthebenefitobtainedinsquamouscellcancer(datanot shown).Althoughthisresultshouldbeinterpretedwith cau-tionduetothelimitedsamplesizeofthesquamousgroup, thefindingissupportedbythepublishedclinicaltrialdata.

Conclusion

Thecancerregistryisapowerfulsourceofdatatomonitor thetreatmentofcancer,providededucation,trainingand, perhaps,incentives areusedtochange thementalityand thework organisationsothat complete treatmentdata is collectedroutinely.

Funding

Thisworkreceivednospecificgrantfromfundingagencies inthepublic,commercialornot-for-profitsectors.

Conflict

of

interest

Theauthorshavenoconflictofinteresttodeclare.

Acknowledgments

Toallcancercoordinators,cancerregistrars,hospital phar-macists and other allied health care professionals who ensuredthe exhaustiveness of data registry, from all the hospitalunitsthatcollaboratedinthisstudy.

References

1.MirandaN,PortugalC,etal.Portugal---Doenc¸asOncológicas emNúmeros2015.Direc¸ãoGeraldeSaúde,ISSN:2183-0681. 2.López-Campos JL, Ruiz-Ramos M, Fernandez E,

Sori-ano JB. Recent lung cancer mortality trends in Europe: effect of national smoke-free legislation strength-ening. Eur J Cancer Prev. 2017, http://dx.doi.org/ 10.1097/CEJ.0000000000000354.

3.Dusetzina SB. Drug pricing trends for orally adminis-tered anticancer medications reimbursed by commercial health plans, 2000---2014. JAMA Oncol. 2016;2:960---1, http://dx.doi.org/10.1001/jamaoncol.2016.0648.

4.INFARMED.Monitorizac¸ãomensaldoconsumodemedicamentos emMeioHospitalar.Janeiro-Outubro2016.

5.EichlerHG,OyeK,BairdLG,AbadieE,BrownJ,DrumCL,etal. Adaptivelicensing:takingthenextstepintheevolutionofdrug approval.ClinPharmacolTher.2012;91:426---37.

6.BorghaeiH,Paz-AresL,HornL,SpigelDR,SteinsM,ReadyNE, et al.Nivolumab versusdocetaxelin advancednonsquamous non-small-celllungcancer.NEnglJMed.2015;373:1627---39. 7.BrahmerJ,ReckampKL, BaasP,CrinòL,EberhardtWE,

Pod-dubskaya E, et al. Nivolumab versus docetaxelin advanced squamous-cell non-small-cell lung cancer. N Engl J Med. 2015;373:123---35.

8.Lein◦53/2017,de14deJulho,PublishedinDiáriodaRepública, n◦135,SérieI.CreatestheNationalOncologyRegistry.

9.FritzA,PercyC,etal.ICD-OInternationalclassificationof dis-easesforoncology.thirded.WorldHealthOrganization;2013. 10.Eisenhauer E, Therasse P,Bogaerts J, Schwartz LH, Sargent D, Ford R, et al. New response evaluation criteria in solid

tumours:revisedRECISTguideline(version1.1).EurJCancer. 2009;45:228---47.

11.Portarian◦ 35/88,de 18 de Janeiro, Published inDiário da Repúblican.◦13,SérieI.CreatestheRegionalRegistries.