cancer world 44

� Julio Celis: bridging science and politics � ESO’s Masterclass: building an army ofbrilliant and caring young oncologists � ESTRO at 30: still focused on its vision fora cure � These big survival gaps are not acceptable: it’s time for governments to act

Julio Celis

Education & knowledge through people & facts

Number 44, September-October 2011

CancerWorld

44SEPTEM

BER-OCTO

BER2011

CANCER WORLD � SEPTEMBER/OCTOBER 2011 � 1

Contents

3 EditorialSecuring quality cancer care: Governments must step up to the plate

4 Cover StoryJulio Celis: bridging science and politics

13 e-Grand RoundState-of-the-art treatment for advanced melanoma

24 Best Cancer ReporterThe future of cancer as told through the story of Renee

34 MasterclassThe ESO Masterclass: where those most eager to learn meet thosemost willing to teach

42 Impact FactorDenosumab – a new option for solid tumours metastatic to bonePerioperative therapy improves gastro-oesophageal cancer survival

50 NewsroundSelected news reports

56 Spotlight on...Thirty years on, ESTRO remains focused on its vision for a cure

60 Systems & ServicesA minimum acceptable standard of care for every patient

EditorKathy [email protected]

Assistant EditorAnna Wagstaff

Editorial AssistantCorinne Hall

Editorial AdvisorsJacques BernierFatima CardosoFranco CavalliAlberto CostaVincent T. DeVita

Contributing WritersMarc Beishon, David CunninghamJanet Fricker, Mark HendersonPeter McIntyre, Philip SaylorTom Waddell, Anna Wagstaff

Publishing AdvisorsGillian Griffith, Fedele Gubitosi

Website LiaisonCorinne Hall

Art EditorJason Harris

ProductionHarrisDPIwww.harrisdpi.co.uk

Printed byGrafiche Porpora

Cover photographJens Nargaard Larsen

Published byEuropean School of Oncology

Direttore responsabileAlberto Costa

Registrazione Tribunale di RomaDecreto n. 436 del 8.11.2004

All enquiries about Cancer Worldshould be made to:ESO Editorial OfficeVia del Bollo 420123 Milan, Italye-mail: [email protected]: +39 02 8546 4522Fax: +39 02 8546 4545All correspondence should be sentto the Editor at [email protected]

Copyright ©2011 European School of Oncology.All rights reserved

Cancer World is published six times per year by the European School of Oncology.It is distributed at major conferences, mailed to subscribers and to Europeanopinion leaders, and is available online at www.cancerworld.org


Editorial

Yourchances of being alive five yearsafterbeingdiagnosedwithcolorectalcancer are around38%higher if you

live inEngland than in theSlovakRepublic, andaround 63% higher if you live in Germany.Similar variations in survival are found formostcancers. In an attempt to explain why this ishappening, theOrganisation forEconomicCo-operation andDevelopment (OECD) recentlymade a cross-country analysis of how cancercare systemsperform. It found that variations inresources, and in access to care, the effective-nessofcare, and theway thecancercare systemis organised, are all important predictors ofsurvival. The countries that perform best arethose that invest in cancer care and healthinfrastructure ingeneral, haveanational cancerplan, set cancer-specific targets, develop net-works for servicedelivery andusequality assur-ancemechanisms toensurepatientsgainaccessto high-quality cancer care.

The message is that underperformingcountries need not only to invest moreresources intocancercare, but also to improvetheprocess quality and governanceof cancercontrol. Sadly, many European governmentsare failing to address these issues adequately,while some are doing nothing at all.

There is broad agreement that cancer careshould be patient-centred, evidence-based,safe, effective and integrated. But definingwhat thismeans in practice and how it can bemeasured can be difficult.A number of Euro-pean organisations have taken on the chal-lenge of defining standards and qualityindicators for specific components of cancer

� Kathy Redmond � EDITOR

care and introducingmechanisms for improv-ingquality.TheOrganisationofEuropeanCan-cer Institutes recently awardedComprehensiveCancerCentre accreditation to fiveEuropeancentres thatmeet quality criteria and are com-mitted to continuous quality improvement.EUSOMAhas introduced voluntary certifica-tion for specialist breast units, with 22 unitsacross Europe achieving certification to date,while the EBMT has accredited more than100European transplantationcentres throughits long-standing JACIE programme. Othergroupsare just beginning to setupprogrammesto improve the quality of cancer care. SIOPEhas identifiedEuropeanStandards ofCare forChildren with Cancer and is working to raiseawareness of the need for all children to haveaccess to high-quality paediatric cancer care.ECCO is creating a European colorectalcancer audit structure – EURECCA– to helpdrive improvements incolorectal cancer surgery.

Theseareall voluntary efforts that arebeingimplemented with little or no funding fromgovernments but lots of enthusiasm frompro-fessionalswho recognise thepotential todobet-ter by embracing the concept of continuousquality improvement. If theOECD is right inasserting that process quality and governanceare keypredictors of cancer survival, then whyare we still so dependent on ad hoc, voluntaryefforts todefineandmeasure thequalityof can-cer care? Governments should now shouldertheir responsibilities to implement, in partner-ship with relevant stakeholders, a systematicapproach to quality assurance at all levels ofcancer control.What are theywaiting for?

Securing quality cancer care:

governments muststep up to the plate

CoverStory

4 � CANCER WORLD � SEPTEMBER/OCTOBER 2011

Julio Celis:bridging science and politics

� Marc Beishon

JulioCelis is a scientist to his core, and can lay claim to being one of the fathers of proteomics. But

it is inhelping shape the structures andvisionofEurope’s cancer researcheffort thathehas arguably

had the greatest impact.Better coordination andnetworkinghave shownwhat’s possible, saysCelis,

the challengenow is tomove frompiecemeal short-term initiatives to somethingmore sustainable.

Thereare twomainaspects to complexityin cancer. The first is the hugely chal-lenging biological nature of cancer, asresearchers dig ever deeper into themolecular structures and pathways of

tumour cells, uncovering layer upon layer of com-plexityas theygo.Thesecond is theworldof ‘oncopol-itics’ and the way in which healthcare and researchorganisations are set up to tackle cancer at nationaland regional levels – and there is arguably no morecomplexworld inoncology than theEuropeanUnionand its relationship with an expanding number ofmember states.

Understanding how best to bring these twodemanding areas together is vital if resources are tobe deployed to best effect, and Julio Celis has beenbridging both science and politics for some timenow to this end. A research biochemist of long-standing, based at the Danish Cancer Society inCopenhagen, and a veteran of many international

committees, he is now taking his experience to theheart of Europe on behalf of ECCO to help drive aresearchpolicy that is formulatedbyscientists andcli-nicians in the field – andnot only by administrators.

“The problem has been that the research com-munity has not looked into the future and preparedtheground to influencedecisionmakers aboutwhereweshouldbegoing,” saysCelis. “Wedon’t tend togetinvolved until we are affected in a big way, such aswith the clinical trials directive, and we have neverreally had aEurope-wide vision for cancer research.We have left it to policymakers to decide for us.”

Now, if Celis and colleagues on ECCO’spolicy committee have their way, Europe’s can-cer community will be one of the leaders in thedebate on how pan-European research fundscan be best be allocated to pursue critical fieldssuch as translational research, where attempts tocreate collaborative networks among the majorcomprehensive cancer centres and basic

CoverStory


research centres have so far been piecemeal.But the obstacles are formidable, as he says.

Health systems are primarily the provenance ofthe member states, not the EU, and there is a bigdisconnect between the existing EU directoratefor consumers and public health, DG SANCO,and the research and innovation directorate.Meanwhile, the principal research structure – theframework programmes – have been limited inscope and sustainability.

Further, about 95%of all cancer research spendis at individual country level, and the instrumentsthrough which member states might channelresearch funds for long-term translational researchplatforms donot exist. Large sums are at stake here,saysCelis– theUK’sNationalCancerResearchInsti-tute alone spent more than €550 million on cancerresearch last year, nearly double the amount it spentaround ten years ago.

Despite the barriers, Celis insists he knows no

moreexciting time in researchpolitics inEurope.Thisis partly because, after many years of committeework in variouspan-Europeanagencies, the researchandclinical sidesof thecancer community are finallyuniting with a much stronger voice. ECCO hasestablished a policy committee to represent theviews of the 60,000 strong European cancer multi-disciplinarycommunity, and thedreamofaEuropeancancer institute or centre is also still verymuchalive.Celis andcolleaguesatECCOcalled recently for thecreationof such abody as a logical next step tounitethe continent’s researchers, at least in virtual form.

Meanwhile, at the political level there has beenprogress in recognising the need to establish aresearch strategy forEuropean science, and for can-cer inparticular. “Keymilestoneswere thecreationoftheEuropeanResearchArea [ERA] in2000, cham-pioned by commissioner Phillipe Busquin, the cre-ation of the European Research Council [ERC],and the initiativebyBusquin in2004tosetupawork-

JEN

SN

ARG

AARD

LARSE

N

ing group to look at the fragmentation of cancerresearch in Europe and to identify barriers,” saysCelis. “I was a member of that group, which latermade an application to the sixth framework pro-gramme [FP6] to set up the Eurocan+Plus projectthat identified where lack of coordination was par-ticularly detrimental to the progress of scientificknowledge and the quality of care.

“One of the outcomes of the project, whichwasledbyPeterBoyle from the InternationalAgency forResearch on Cancer, was the recommendation toestablish aworld-class infrastructure– aplatformofEuropeancancer centres for translational research.”

There then followed the Stockholm Declara-

tion, led byUlrik Ringborg at the Karolinska, whichis a manifesto drawn up by 18 cancer centres toachieve such a platform. “It was the first time headsof cancer centres had really sat down to look at howtheycouldstructure translational research inEurope,”says Celis. “Following this, and recommendationsfrom the ECCO oncopolicy committee and theOrganisationofEuropeanCancer Institutes (OECI),the European Commission funded under FP7 anetwork of excellence, led byRingborg, to structuretranslational cancer research between cancerresearch centres in Europe, namely the Eurocanplatform” (seewww.eurocanplatform.eu).

The European Partnership for Action Against

“It was the first time cancer centres had looked at how

they could structure translational research in Europe”

CoverStory


JEN

SN

ARG

AARD

LARSE

N

experiments onanimals. “Chile thendevelopedclin-ical biochemistry as a career option, and Iwas able tostudy subjects such as histology, physiology andpharmacology aswell as thecore science,whichwasvery valuable tome later,” saysCelis.

He left for the US to do a PhD, and on hisreturn toChilewas able to get a posting to theMed-ical ResearchCouncil laboratory ofmolecular biol-ogy in Cambridge in the UK, financed by theWellcome Trust. “I can’t emphasise how valuablenetworking was to me while I was in Chile undermymentor JorgeAllende – and these are skills I’vealso carried forward.Without networking, it is dif-ficult to progress far in a career.”

Certainly, that progressionwas stratospheric, forCelis foundhimselfworkingalongsideFrancisCrick,one of the famousCambridgeDNAduo, and otherluminaries such asMaxPerutz, Fred Sanger and, inparticular, SydneyBrenner, aNobelwinner forworkonnematodes (ringworms).Brennerextendedhis stayin England as a member of the staff when the mili-tary coup in Chile made Celis reluctant to returnhomewith his family.

Therewas hardly a better place forCelis to startworking in thebooming fieldofmolecularbiology, andit was Brenner who set him to work on proteins,which has been the core of his research ever since.“But I left Cambridge after five years to take a per-manentposition inDenmarkat thenewDepartmentof Biostructural Chemistry, headed by Brian Clark,whowasalso fromCambridge– itwasacriticalmovethathelped securemycareer.There is one importantaspectofwork inEnglandIhave takenwithme,how-ever, and that is that everyonewasworking in the lab,not sitting inofficesplayingwithmodels –except forCrick, that is.”

Beingahands-on scientistmeansbetterproblemsolving, especially in research that requires extensivevalidation suchas cancer, and thishasbeena featureof seniorEnglish biologists, he adds. “This is not thecase in every country – I see several places where itis notunusual for evenyounger scientists to leave thelab and direct research from their offices.”

Cancer (EPAAC), with its various work packages,is another major programme that has research aspart of its remit, and aims to ensure that one-thirdof all cancer research funding is coordinated acrossEurope by 2013 – a sum of some €1.5 billion.“This is not achievable of course,” saysCelis, but itis an indication of what the research communitybelieves to be necessary.

Celis also expects the health component of thenext framework programme to have a focus on per-sonalised/stratifiedmedicine.Thiswould tie inwiththeEUnowhavingan ‘innovationunion’strategy thatstretchesahead to2020,which focusesonmajor soci-etal challenges, and may foster better cooperationamongmember states.

“The biggest words in Europe today are co-ordination and sustainability,” saysCelis. “We havecreated many networks so far, but little in the wayof ensuring things continue. The visionwe have atECCO is to have a sustainable network of com-prehensive cancer centres and institutes like myown to build a critical mass of expertise andresources across all the many niches we have inresearch to solve problems in different combina-tions – a sort of ‘variable geometry’.”

Celis, more than most, is also enmeshed in oneof themost complex scientific geometries in cancer– probing the world of proteomics. Until the end of2011, when hewill make Brussels and oncopoliticshis main priority, his job is scientific director of theInstitute of Cancer Biology at the Danish CancerSociety inCopenhagen.Thesociety is the largest can-cer organisation in Denmark, with some 450,000individual members. Unlike other charities such asCancer Research UK, it directly employs its ownresearch teamsacross sixdepartments.ForCelis, thisposition has allowed him to focus exclusively oncancer in the latter stages of his research career, andto get fully involved in oncopolitics. His pedigree asa scientist on this stage could hardly be better.

Born in Chile, he was an early convert at schoolto the then relatively new subject of biochemistry,enthused by a teacher who was carrying out some

CoverStory


“Everyone was working in the lab, not sitting in

offices playing with models – except for Crick, that is”

Celis had landed in theUniversity ofAarhus inDen-mark, with his late wifeAriana, amedical technolo-gist, where he was to spend many years on cellbiology research, continuingwith aproductivenicheinproteins,beforemoving to theDanishCancerSoci-ety. Here he was able to devote himself full time totranslational research from the basic science stand-point, especially on breast cancer.

“At first I took forwardwork I had been doing inCambridge,with JohnSmith, on suppressor transferRNAs,but found therewerealreadymanyexperts onthis topic in Denmark. So we turned to cell biology,inparticular to thecell cycle andcell transformation,drawingonwork Ihadbeendoingon separatingpro-teins using gels to study theseprocesses using abio-chemical approach. By 1981wehad discovered thePCNA protein [proliferating cell nuclear antigen]simultaneously with a group at the SCRIPPSResearch Institute, amolecule that is central to celllife and death, and in 1982 we published the firstextensivework onprotein expressionprofiles of nor-mal and transformed cells.”

PCNAhas been described as the ‘ringmaster ofthe genome’and the implications of uncovering thisand other proteins involved in normal and trans-formed cells has great importance for cancer, butas Celis says, this was just the beginning of whathas turned out to be one of the most complexbranches of molecular biology. There are nowknown to be as many as two million proteins,thanks to mechanisms such as ‘splice variants’from the body’s 25,000 or so genes that encode pro-teins, and the field of proteomics – for whichCelisis rightly seen as one of the founding fathers – iseven more complex than genomics.

“At the start I thoughtwecould lookatmanypro-teins at the same time as a way of profiling changesin cells, and indeedwe started to build protein data-bases of what we found, and later as identificationtechniquesbecameavailablewewereable to tellwhatthe proteins were,” he says.

Celishimselfhas longuseda laboratory techniquecalled 2D gel electrophoresis, pioneered by Patrick

O’Farrell,which separatesproteins according to theirmolecularweight andcharge andallows researchersto compare proteomic profiles in different samples.It is the kind of hands-on laboratorywork he is keento promote, and he points out that, with the adventof high-throughput technologies suchasmass spec-trometry and protein arrays, researchers can losesight of biological questions they should be seekingto answer and instead get wrapped up in a constantdiscoverymode forworkonbiomarkers, for example.“In cancer you need a validationmode too,” he says.

In 1995, when he turned to look at clinical can-cer questions using tissue samples rather than celllines, he sayshehad to “convert” himself to apathol-ogist as well. “I realised I couldn’t do anything with-out understanding the histology and pathology oftissue samples, whichmeant doing part of theworkmyself so I could interpret the data.”

Key questions that Celis and the internationalproteomics community are addressing includethe search for biomarkers for early detection, find-ing new drug targets and predicting how a tumourwill respond to a therapy (see also the thematic‘oncoproteomics’ issue of Molecular Oncology(2010; vol. 4, pp 459–566), where new approachessuch as nanotechnology and imaging spectrome-try are discussed).

AtAarhus, Celis worked on bladder cancer bio-markers,usingantibodies to locate theproteins indis-eased and normal tissue, as he recognised that anexpression level on its own may not be meaningfuldue to the daunting heterogeneity of the tissues.Thiswas laboriousworkmadepossible only throughclose cooperation with Hans Wolf, a cancer sur-geon, and Torben Ørntoft, a clinical biochemist,whowere committed to obtaining tumour and con-trol tissue, and points to a much wider problem hesees in translational research.

“This type ofwork not only requires fresh tissue,which is very challenging to organise, but also set-ting up tumour, blood and urine biobanks so whenwe come to validate a biomarker, for example, con-sistent samples are at hand. But samples are often

“They can lose sight of biological questions they should

be seeking to answer, and get stuck in discovery mode”

CoverStory


line andanimal samples that areused so extensively.During the1990s,Celis began to representDen-

mark on European organisations in the molecularbiology arena (inparticular theEuropeanMolecularBiologyLaboratory andEuropeanMolecularBiologyConference) andsteppeduphis activity in theEuro-pean Molecular Biology Organization (EMBO). In1999hebecame secretary general of theFederationof EuropeanBiochemical Societies (FEBS).

It was a this point that the idea of proactivelyworking for a coherentEuropean strategy and struc-ture for scientific research began to take shape.

Working with others in the biology sector, Celisandcolleagues suchasFotisKafatos andFrankGan-non created the European Life Science Forum(ELSF) to put pressure on theEUover the need forbasic research and in particular towork towards thecreation of theEuropeanResearchCouncil (ERC).“Wesoon realised that to achieve this goalwehad toalso involve all theotherbranchesof science, andwecreated the Initiative for Science in Europe (ISE),which I chaired after José Mariano Gago, who wasscienceminister inPortugal, andwhich led to the for-mation of the ERC.”

Thiswasabig achievement, asmembergroupsofISE included existing science organisations such asEuroscience and the European Science Founda-tion, among others. “It tookus four years, butwhat Ihave learnt is howweneed to turn adiffuse idea intoan object of desire – so that politicians then see thevalue of it. The European Commission was some-what sceptical at the beginning, but it got to such alevel that they couldn’t stop it and so took it over.”

Then in2008,LexEggermont, thenpresident ofECCO, invited Celis to chair ECCO’s new policycommittee,whichaimed topull levers at thehighestlevels to try to progress the pan-European cancerresearchvision,with initiatives suchas theOncopol-icyForumandtheEuropeanAcademyofCancerSci-ences, an independent entity that can providescience-based advice to policymakers.

So there are many projects, meetings and com-

stored in different systems and at different times –we may have had a focus on translational researchin the currentEuropean frameworkprogramme, butwehave failed to create consistent infrastructure tosupport it – and tumour banks are one of the mainresources. It’s this kind of omission that means weare often paying the price of short-term solutions tolong-term problems.”

WhenCelismoved to theDanishCancer Soci-ety he switched to translational breast cancerresearchwith the support of pathologist Fritz Rank,taking a year tounderstand thehistory of thedisease,andhis institute is now a key niche player in certainaspects of thismajor tumour group, such as lookingat biomarkers for endocrine resistance and earlydetection.Naturally he also runs a proteomics sub-group alongside those on breast cancer, cell cycle,apoptosis, metastasis and tumour microenviron-ment and genomics.

Ontwosidesofhis largeoffice inCopenhagenhehas pinned up many pictures of protein profiles inbreast cancer tissues,which showwhere theproteinis beingmade, in some cases from only certain cellsin a tumour. “This gives you some idea of the incred-ible heterogeneitywe are finding in tumours – thereare no two sites on a tumour that look exactly alike,”he says.There are images tooofpremalignant tissue,fromwhich his proteomics lab is working on identi-fyingprogenitor cell types, andwhich could eventu-ally result inbiomarkers forearlydetectionofcells thataremore likely to progress to tumours.

The validation problem with biomarkers is wellillustrated by the fact that, of the thousands of arti-cles published on biomarkers, so far fewer than 100of themarebeingused.Althoughhe is philosophicalabout theenormouscomplexity thatnature isputtingin the way of the cancer community, Celis says theonly way to find more answers is to build the infra-structure thatwill allow researchers to take the timethey need to work on the most clinically relevantquestions involvinghuman tissue and fluids, aswellas themucheasier, butmuchmorehomogenous, cell

CoverStory


“This gives you some idea of the incredible

heterogeneity we are finding in tumours”

what type of trialswecoulddo, andwhat novel proj-ectswecan take frombeginning to end. Ifwedo thatit will be first time it is done on this scale anywhere.Wealsoneed todemonstrate success stories to inter-est policymakers.

“Then the next step is sustainability, where weneednew instruments for organisingmember statesand theEuropeanCommission.AtECCOwehaveproposed for example that comprehensive cancercentreswould getmatched funding from theEUona competitive basis, but overall weneednew instru-ments for members states to cooperate on meetingtheir societal challenges so thatwecanapproach thatgoal of increasing thecoordinationof research fund-ing.Oneenablerwouldbe tohavenationalhealthandresearchministersmeet in the samewayas their col-leagues do for economics.”

Celis says that in addition to setting up a virtualEuropean cancer centre, ECCOwould like to see alargerEuropean Institute ofHealth aswell as a newmechanism to support top researchers through‘dream teams’, allowing industry to contribute cashand new drugs for study. He would also like to seemore support for research intocancerpreventionandearly detection.

The ECCO policy committee has set outmuch of its current thinking in a response to theEuropean Commission’s green paper, ‘Future EUresearch and innovation funding programmes:common strategic framework’. This paper setsout the aims for research and innovation fundingfor the next framework programme at both EUand national levels from 2014 to 2020, and byfocusing on instruments ‘with proven Europeanadded value’ there should also be specific fundingproposals by the end of this year. In a bid to connectwith the public imagination, the framework 8 pro-gramme will do its bit to further European inno-vation and research under the nameHorizon 2020.

Celis says it is imperative that basic scienceorganises itself as well as other sectors to advisepolicy makers and influence funding. “Scientists

“I have learnt how we need to turn a diffuse idea into

an object of desire, so that politicians see the value of it”

CoverStory


mittees now in European cancer – and to cap it all,Celis is also the current president of the EuropeanAssociation of Cancer Research (EACR), whichgiveshim fullECCOboardmembership.What out-puts does he now see coming from all this activity?

“At the science level we need to firmly establisha structure for translational research–howwe stan-dardise technology, share data, exchange patients,

JEN

SN

ARG

AARD

LARSE

N

Celis manages to integrate pathologists into theproteomic research he does on tumour progres-sion and tumour subtypes, “in a way that theclinical questions steer the research strategies.”

Having worked alongside Celis in some bigEuropeanprojects, including theEurocanplatform,Ringborg also singles out his “unusual capability toorganise and convince people about both infra-structures and research strategies,” and also hisgrasp of the need to find new types of collaborationand financial support to improve the innovativepotential in cancer research.

Celis’s first wife and fellow researcher Arianasadlydied13years agoandhehas since foundacom-panion, againa scientist, biochemistTeresaCabezòn,whoalsoworks at theDanishCancerSociety.Hehasthree children and six grandchildren, so family lifetends to take all his free time. After he steps downfrom his research post in 2012 he will be spendingmost of his time in Brussels to concentrate ononcopolitical duties.

With his scientific achievements now largelybehind him,Celis is looking for a success in gettingmoreopportunities for researchers in thenext frame-workprogramme, andmoreprojectsunder theEuro-canumbrella. “And ifwecangetmember states andthe commission to agree to look for new sustainableopportunitieswewill be in good shape,” he says. “Weneed to be sure we will be consulted in time whenpolicymakersworkonanythingcancer related. I’d likealso to seemorepeople fromoutsideEuropecomingto do their research here.”

Of course, there are some who would now saythat thewholeEuropean ‘project’ is inperil followingthe economic collapse in countries such asGreece,Spain and Ireland, and there is huge strain now evi-dent inhigh-level political relations.Meanwhile theEuropeanCommission is looking for a five per centincrease in the EU budget to meet 2020 goals.Whicheverway this goes, forCelis therewill beno letup in pushing for sustainable structures that tapintomember state funding to securea solid future forcancer research in Europe.

are not as visible as doctors and patient groups –we need to set out our priorities for the long termthat will fit alongside the more short-term projectsthat tend to dominate decisions,” he says.

From next year, no doubt Celis could spendvirtually his entire working time in meetings andconferences as the scenario for 2020 unfolds,and with other commitments such as president ofthe EACR, which was a founder member ofECCO, and has 10,000 researchers in its ranksthrough national society membership. The EACRis naturally lending its support to the coordinationeffort and the expected focus on personalisedmedicine in the next framework programme. Itwill hold its biannual Congress at Barcelona nextyear, which Celis will be chairing.

He is concerned that the US is still attractingmany of the world’s top researchers, and thecancer community there enjoys a close relation-ship with industry. “One of the aims of the Euro-can platform is to help make Europe a moreattractive place for researchers from say Asia tocome and work.”

Developing the next generation of Europeanresearchers is also important, and Celis is one ofthe organisers of a FEBS lecture course on trans-lational cancer research, taking place in Portugalthis September. “This is not just about work – it’sin a nice summer school setting and is a greatchance to build a personal network,” he says. Asif this isn’t enough, he is also editor in chief of thejournal, Molecular Oncology, which majors onissues such as the next steps in proteomics. Asalways, Celis will be urging that the technologyrace in techniques such as protein profiling is tem-pered with realistic biochemistry.

Ringborg, director of Stockholm’s KarolinskaCancer Centre, describes Celis as “a remarkableman”, in both his scientific work and his capacityfor networking. “I have never met a basicresearcher and cancer biologist with such a ded-icated interest in the problems of the patients,” hesays, and comments particularly on how well

CoverStory


“National health and research ministers could meet

in the same way as their colleagues do for economics”

e-GrandRound


State-of-the-art treatmentfor advanced melanoma

Progress in treating melanoma has lagged behind many other cancers for decades. But a

number of novel approaches, based on better understanding of the role of the immune system,

better selection of patients and the identification of targetable mutations, is now offering a

glimmer of hope.

Before 2010, advanced mela-noma was considered a diseasewith very poor prognosis. Mor-

tality was increasing compared to othercancers, and median survival remainedat only six to nine months in most stud-ies. There were few, if any, effectivetherapies. Interferon (IFN) had limitedeffect, and dacarbazine (DTIC) andhigh-dose interleukin-2 (IL-2) had noconfirmed effect on overall survival.High-dose IL-2, which was a standardin the US, had significant toxicity.There were no positive phase III trialsfor overall survival.

For cytotoxic chemotherapy, there iscurrently no evidence that singleagents, combination chemotherapy orthe addition of tamoxifen or IFN toDTIC is superior to DTIC alone.Although some data suggest that thecombination of carboplatin and pacli-taxel might be superior to DTIC alone,these treatment approaches have notbeen formally compared.

High-dose IL-2 therapy becamethe standard in the US in 1998,based on data showing a responserate of about 16%. Some responses

The European School of Oncology pres-ents weekly e-grandrounds which offerparticipants the opportunity to discussa range of cutting-edge issues, fromcontroversial areas and the latest sci-entific developments to challengingclinical cases, with leading Europeanexperts in the field. One of these isselected for publication in each issue ofCancer World.In this issue, Michael B. Atkins, from theBeth Israel Deaconess Medical Center,Dana Farber/Harvard Cancer Center andHarvard Medical School, Boston, USA,provides an update on the latest devel-opments in treating patients withadvanced melanoma, and looks at whatthe future may bring.Daniel Helbling, Onkozentrum Zurich,

Switzerland, poses questions sent in byparticipants during the e-grandround livepresentation, which is summarised hereby Susan Mayor.

The recorded version of this and other e-grandrounds is available at www.e-eso.net

were very durable, with the medianduration of response being 8.9months and the median not beingreached for complete responders.There was limited impact on overallsurvival, but among responders, anypatient who was still responding at30 months has remained in responseformore than 10 years (JCO 17: 2105–16,Cancer J Sci Am 6 [suppl 1]:S11–S14).

High-dose IL-2 appears to be use-ful and we still use it in the US, but itis toxic and requires inpatient treat-ment, making it expensive andimpractical. Therefore, its use is lim-ited to selected patients treated atexperienced centres. Efforts to betterselect patients who might benefitfrom IL-2 therapy are warranted andwe are currently actively investig-ating this.

THE THERAPEUTICLANDSCAPE IN 2010Looking at the therapeutic landscapein 2010, results may be improving alittle bit for non-specific reasons: ear-lier treatment, better patient selec-tion, improved treatment of brainmetastases and better systemic ther-apy. But new approaches are clearlynecessary. Unless new approachesare unquestionably active, it is likelythat they will need to be studied inphase III trials.

Question: Is there a predictive tool inmetastatic melanoma to select on thebasis of genetic profile?Answer: There are mutations that helpselect therapies, particularly for therapiesthat target protein products of C-KIT orB-RAF mutations. But these mutationsmay also help us select for who mightrespond to immunotherapy. We mayalso need to come up with therapiesfor patients whose tumours don’t havethose mutations.

PROMISING THERAPEUTICAPPROACHESImmunotherapyThe benefit from immunotherapy maybe limited by the ability of T cells toinfiltrate a tumour. Melanomas mayvary in their degree of immune infil-tration, with about 24% of tumourshaving a high degree of immune infil-tration and 40% having very littleT cell infiltration. The higher thedegree of immune infiltration, the bet-ter the outcome. This is particularlytrue if the infiltrating cells are CD8+T cells.

Using an immune signature basedon gene expression profiling, we havefound that patients whose tumourshave this signature are more likely torespond to high-dose IL-2 and have asignificantly longer median progres-sion-free survival, of 19.4 months forthe class 2 immune versus 2.5 monthsfor class 1 antigenic gene expressionsignatures, respectively (JCO 27:15S,abstract 9003).

Another factor potentially associ-ated with response to high-dose IL-2,which may be worth investigating withnovel immunotherapies, is mutationalstatus. A significantly larger propor-tion of patients with B-RAF and, par-ticularlyN-RAS, mutations are likely torespond to high-dose IL-2 compared tothose with wildtype tumours (JCO28:15S, abstract 8597). More data areneeded on this.

Elevated lactate dehydrogenase(LDH) status is a negative predictor forresponse to high-dose IL-2. If thepatient has a high level of LDHreleased from the tumour, which is apoor prognostic finding, he/she will beless likely to respond to high-doseIL-2 therapy, with a response rate of6% and no complete responses.

It is becoming clear that LDH reg-ulation is associated with hypoxia.Other hypoxia-related genes such as


e-GrandRound

VEGF (JCO 27:2645–52) are alsoinversely associated with likelihood ofbenefiting from high-dose IL-2 therapy.

In summary,with regard tohigh-doseIL-2 tissue-basedpredictive biomarkers,a novel immune-based gene expressionprofile appears to predict for better pro-gression-free survival and, possibly,response tohigh-dose IL-2.The associa-tion of clinical benefit with immuneresponse signature suggests a possiblemechanism for high-dose IL-2 anti-tumour effect based on immune cellsthat arealreadypresent in the tumourandblocked for a particular reason.

This is part of the reason we were soexcited about some of the novelimmunotherapies. The fact that bothB-RAF and N-RAS mutational statusmay predict for favourable response tohigh-dose IL-2 suggests opportunities forcombinationstudies.AnelevatedLDHorVEGFmaypredict for lackof response tohigh-dose IL-2, so different treatmentsmay be necessary for those patients.

Question:Which patients do you treatwith high-dose IL-2? How do you selectthese patients?Answer: In the US, we treat patientswho have good performance status (0–1),good heart and lung function, no CNSmetastases and, based on the data I havejust described, normal LDH.We are notyet selecting patients based on muta-tional status of their tumours, but we arecarrying out a prospective trial to seewhether factors including mutationalstatus and immune signature will predictwhich patients benefit from IL-2 basedtherapy.

NOVEL IMMUNOTHERAPYIpilimumab is a novel type ofimmunotherapy. When the immunesystem recognises an antigen on anantigen-presenting cell (APC), bindingoccurs between a Tcell receptor (TCR)and a co-stimulatory molecule, CD28.

and can show tumour flare beforeresponse.

The CT scan in the figure belowshows extensive disease on the patient’sabdominal wall at baseline, which getsworse three weeks into therapy, buthas completely disappeared by fourmonths. This pattern is seen in about10-20% of patients who respond toipilimumab. It calls into question thestandard RECIST criteria for diseaseprogression, and may have confoundedthe interpretation of some of the earlystudies with ipilimumab.

When this binding takes place, CTLA4is upregulated on the surface of theT cell. This out-competes CD28 forbinding to be severed on the antigen-presenting cell, which leads to ashutting off of T cell function. CTLA4antibodies such as ipilimumab blockbinding between CTLA4 and B7 andallow for that brake on the immunesystem to be removed, and for theimmune system to continue to expand(see figure above).

The two antibodies blockingCTLA4that have been studied extensively –ipilimumab and tremilimumab – bothproduce responses in 8–15%of patientswith refractory melanoma. Theseresponses are associated with autoim-munity, because taking the brakes offthe immune system happens not justwithin the tumour but also withinorgans that are protected from autoim-munity by CTLA4. Responses aredurable in many patients, with themajority lasting longer than two years:20–30%of patients have durable diseasecontrol longer than three years.

In contrast to IL-2, activity is seenin the central nervous system and is notprevented by steroid co-administra-tion. Responses are delayed at onset

Pivotal trials with ipilimumabA trial presented at ASCO last year(2010) by Steve Hodi (NEJM363:711–723) randomised patientswho were originally HLA (humanleukocyte antigen)A2+ to a low dose ofipilimumab (3 mg/kg), either alone orin combination with gp100 peptidevaccine, or to gp100 vaccine alone.Results showed an overall survivaladvantage for the ipilimumab-con-taining arms but no advantage for thevaccine, either alone or when added toipilimumab.

A pivotal phase III trial with DTIC+/- ipilimumab (10 mg/kg) completedaccrual a couple of years ago and therequired events have just beenachieved. Positive results showing animprovement in overall survival for theDTIC + ipilimumab arm relative toDTIC alone were recently reported(NEJM 364:2517–26). The CytokineWorking Group has looked at ipili-mumab in patients with CNS metas-tases and we have seen similar activityto that seen in systemic disease.

Adjuvant studies are currentlyongoing in patients with stage III dis-ease: one in Europe comparing ipili-mumab to observation, which isnearing completion, and a second that

e-GrandRound


CTLA4 BLOCKERS: IT CAN GET WORSE BEFORE IT GETS BETTER

This typicalpattern of a flareup followed byresponse mustbe borne inmind whenevaluatingresponse totreatmentSource: Courtesy

of Michael B

Atkins

IPILIMUMAB TAKES THE BRAKES OFF THE IMMUNE SYSTEM

By blocking thebinding betweenCTLA4 and B7, thisnew type of drugallows the immunesystem to do its jobSource: Adapted

from Lebbe et al,

oral presentation

769O at

ESMO 2008

has just opened in the US, comparingipilimumab to interferon.

The second-line phase III trial pre-sented at ASCO in 2010 randomisedpatients with pretreated metastaticmelanoma to ipilimumab plus gp100(60%), ipilimumab plus placebo (20%)or gp100 plus placebo (20%). Both theipilimumab arms showed superior over-all survival to gp100 alone, with 44–46% of patients alive at one year,compared with 25% with placebo alone(see figure above). Twice as manypatients treated with ipilimumab werealive at two years compared to thosegiven placebo alone.

There was a fair amount of toxicityrelated to ipilimumab. This was pri-marily autoimmune toxicity, includingdermatologic side-effects, gastroin-testinal side-effects such as colitis, anda small percentage of patients withendocrine or hepatic side-effects.These are themajor side-effects relatedto ipilimumab, which need to be keptin mind and treated aggressively withsteroids if they occur.

Summing up, it appears thatCTLA4 antibody enablesimmune responses and anti-tumour responses in someindividuals. Tumour expres-sion of PD-1 ligandmay pre-vent immune response evenwith CTLA4 blockade byinducingTcell death, servingas ‘barbed wire’ to thoseimmune cells that may be inthe tumour. Given thatmelanoma cells have beenshown to express PD-1 lig-and,CTLA4antibody effectsmight be augmented byantibodies that inhibit thePD-1 pathway. Controllingimmune regulationmay pro-vide a way forward forimmunotherapy of patientswith melanoma.

Question: Auto-immunity is a majorconcern. My first impression was thatipilimumab is not very targeted, becauseit releases the immune system to fightagainst anything it finds. Do you think itis really a targeted approach formelanoma patients?Are you not partic-ularly concerned about auto-immunity?Answer: I think peopleshould have a healthy respectfor auto-immunity. Patientscan get into very serious trou-ble if the auto-immune side-effects are not treated quicklywith steroids and, if neces-sary, infliximab. It is inter-esting that treating theauto-immune side-effects cancontrol them without pre-venting the anti-tumourresponse, so you do not haveto be worried about givingimmunosuppressants the wayyou might be with otherimmunotherapies.

If you delay treating auto-immunity, theeffects can be severe. But if you areattuned to the problem and reactquickly, you can provide this therapysafely in an outpatient setting andpatients can achieve the benefit.Nonetheless, it is not as targeted to thetumour as we would like.Question:Can ipilimumab be used as aneoadjuvant therapy in patients who arepotentially resectable?Answer: I think that is a potentiallyuseful research tool and ongoing studiesare using ipilimumab as neoadjuvanttherapy. However, fewer than one-thirdof patients show responses. I would notadvise this outside a research study, as thetoxicity might complicate surgery.Question:Are auto-immunity and side-effects related to response – can they besurrogate markers?Answer: Yes, it appears that auto-immunity is related to response. Theresponse rate in patients who get auto-immune side-effects is 40–50%, whileit is closer to 5% in those who do not.This tells us that the mechanism of theresponse is probably unleashing latentauto-immunity against the tumour cell.At the same time, those patients who


e-GrandRound

IPILIMUMAB: SURVIVAL ANALYSIS

Kaplan-Meier analysis shows a clear survival advantagefor both ipilimumab arms in this trial for use as asecond-line therapySource: FS Hodi et al. NEJM (2010) 19:363:711–723. Printed

by permission from Massachusetts Medical Society

PD-1 ANTIBODY: CHANGE IN TUMOUR BURDEN

Tumour shrinkage occurred in more than 50% ofpatients treated with various doses of a PD-1 antibodyin this phase I trialSource:M Sznol et al ASCO 2010 presentation, abstract 2506

body are now underway, and otherstudies are being considered. Overall,it is possible that the PD-1 antibodymay offer even more tumour-specifictargeted immunotherapy.

TARGETING MOLECULARALTERATIONS IN MELANOMAAround 50–60% of melanomas havemutations in a very specific area,V600E or K mutations in B-RAF. Acomplementary 15% of tumours havemutations inN-RAS and about 30% ofpatients with mucosal or acral/lentigi-nousmelanomas have amplifications ormutations inC-KIT (see figure below).

These mutations are not distrib-uted randomly across melanomas.Tumours in moles or in non-chronicsun-damaged skin are more likely tohaveB-RAFmutations, but rarely haveC-KITmutations. Tumours in chronicsun-damaged skin, acral/lentiginousor mucosal areas are more likely tohave C-KIT amplification or muta-tions. Tumours that haveC-KITmuta-tions are very sensitive to inhibitors ofC-KIT such as imatinib. The figureoverleaf shows a PET scan before andafter four weeks of treatment, withdramatic reduction in the PET uptakeof multiple tumour metastases (JCO26:2046–51).

develop immune reactions against theirtumours probably also have some sortof defect in their immune regulationthat allows them to also develop reac-tions against their colon, their skin ortheir liver.

PD-1 antibodies and inhibitorsPD-1 antibodies, or PD-1 inhibitors,may provide more tumour-specificimmune suppression. When T cellsare exposed to the tumour, they upregulate PD-1 on their surface. PD-1 isa member of the CD28 family involvedin T cell regulation. When it binds totumour or antigen that is expressingPD-L1, this causes apoptosis orexhaustion of the immune cells. Thishappens primarily in the tumour or in

chronic inflammatory situa-tions. If you could block theinteraction you might beable to restore the activity ofthe immune system selec-tively within tumours.

A phase I study of anantibody to PD-1 presentedat ASCO in 2010 (JCO28:15S, abstract 2506)treated 46 patients withmelanoma with three dif-ferent doses. Almost 33%(15/46) exhibited tumourresponses. All of theseresponses were ongoing(with the longest >18months) when the resultswere reported. More than50% of patients had tumourshrinkage on a waterfall plotat different dose levels (seefigure opposite page, below).

The figure (this page, topleft) gives an example ofresponse in a patient withvery extensive liver metas-tases as well as some lungmetastases that reduced dra-matically within two cycles

of treatment with PD-1 antibody, evenat the 1mg/kg dose.

Early trials show tumour responsein more than 30% of heavilypretreated patients withadvanced melanoma (JCO28:15S, abstract 2506),which is very exciting.Responses have also beenseen in other cancers,including lung, colon andkidney cancer, which aredurable to date. The toxicityseen so far has been rela-tively mild, without thedegree of auto-immunityseen with ipilimumab.Combination studies withipilimumab and PD-1 anti-

e-GrandRound


PD-1 ANTIBODY: OBJECTIVE RESPONSE

Response to PD-1 antibody (1 mg/kg) of liver (upper) andlung (lower) metastases in a 66-year-old male patient withmelanoma, who had progressed on high-dose IL-2. A shows thebaseline and B shows after one cycle of treatment. The patientmet the partial response criteria after three cycles of treatment,and the response was still continuing at 12+ monthsSource: Courtesy of M Sznol, Yale University

MOLECULAR ALTERATIONS IN MELANOMA

inhibitor of mutant B-RAF (NEJM363:809–819); andGSK 2118436, alsoa selective inhibitor of B-RAF (JCO28:15S abstract 8503). Both of thesehave shown promising data.

Studies show that PLX4720inhibits tumour growth in B-RAFmutant tumours but has no activity inwildtype tumours. Results from a phaseI trial of PLX4032 in patients withmutant B-RAF tumours (NEJM363:809–819) showed tumourresponse in 70% of patients, includingone complete response, even thoughmany had M1c disease (metastases

involving visceral sitesbeyond lung and/or anelevated LDH) (see bar-chart opposite). A dra-matic response was seenon PET scan within 15days, showing very signif-icant reduction in glucoseuptake in disease in thelungs (see scans opposite).

Although this was nota randomised study, sur-vival curves from thisphase I study show sig-nificant prolongation inprogression-free survival,with a median progres-sion-free survival ofaround eight months inpatients with V600Emutations treated at theoptimal dose (see graphopposite, lower).

This treatment isassociated with some tox-icity, including arthralgia,photosensitivity, rash,fatigue, pruritus and pal-mar-plantar dysaesthesia,although mostly not seri-ous (grade I and II)(NEJM 10:363:809–819). The most troublingside-effect is cutaneous


e-GrandRound

Experience with C-KIT inhibitorsC-KIT inhibitors have dramatic effectsin patients with melanomas containinga variety of C-KIT mutations. How-ever, these mutations occur in less than1% of all melanomas. The role of C-KIT inhibition in C-KIT amplifiedtumours is yet to be fully establishedand it is possible that those with ampli-fications inC-KITmay be less respon-sive than those with mutations inC-KIT or not responsive at all.

Despite being rare tumours inEurope and the US, mucosalmelanomas make up the majority ofmelanomas in Asia, par-ticularly in India andChina. C-KIT mutatedtumours are, therefore,much more prevalent.Multiple studies are cur-rently underway with ima-tinib, sunitinib, dasatiniband nilotinib, including aninternational phase IIItrial comparing nilotinibversus DTIC in patientswith C-KIT mutatedmucosal melanomas. Thisis all very exciting, andprovides a proof of princi-ple, but it is not theanswer for the majority ofpatients with melanoma.

B-RAF inhibitors inmelanomaInitial studies to deter-mine whether B-RAFinhibitors had activity inmelanoma did not selectby mutational status andused sorafenib, which is avery poorB-RAF inhibitor.Studies showed limited orno activity with single-agent sorafenib.A phase IItrial of sorafenib com-bined with DTIC showed

some improvement in progression-freesurvival (JCO 26:2178–85). But therewas no additional benefit when com-bined with carboplatin/paclitaxel(E2603, PRISM), either in a first-lineco-operative group trial or a second-lineindustry-sponsored trial.

We did not know whether the poorresponse was because sorafenib wasa poor B-RAF inhibitor or becauseB-RAF was not an important target inmelanoma. We needed better tools toanswer these questions, and two ofthese have now come along: PLX 4032/RG7204, which is a more selective

C-KIT INHIBITOR IN PATIENT WITH KIT-MUTANT MELANOMA

Source: S Hodi (2008) JCO 26:2046–51, reprinted with permission © American

Society of Clinical Oncology. All rights reserved

Pretreatment PET scan PET scan after 4 weeks

malignancies, which looklike squamous cell cancers.They occur in about one-quarter of patients andrequire follow-up with adermatologist.

Conclusions from thephase I study are thatPLX4032 is tolerable andhighly effective, even inpatients with extensive priortreatment. Results provideproof of concept thatB-RAFmutations are critical onco-genic drivers in B-RAFmutant melanoma.

There is a lot of excite-ment around this therapy,which is justified, but itstrue efficacy will depend onthe durability of responseand the ability to impact onoverall survival.

Larger phase II and ran-domised phase III studies are neces-sary to confirm this benefit. Thesetrials have been completed and the

results were reported at ASCO 2010.The phase II trial confirmed the effi-cacy of PLX4032 (vemurafenib) in

patients with B-RAFmutant melanoma (JCO29[15S)]:abstract 8509),and the phase III trialshowed a significantimprovement in overall sur-vival for treatment-naïvepatients receiving vemu-rafenib compared to DTIC(NEJM 364:2507–16).

Findings with B-RAFinhibitors have implicationsfor howwe select patients forvarious therapies. In thefuture, I think thatmelanomastudies will divide patientsinto three classifications:V600mutant tumours; V600wildtype tumours and V600mutant tumours that haveprogressed after selectiveB-RAF inhibitor therapy. It isvery important to carry outstudies designed to enhance

theefficacyof selectiveB-RAF inhibitors,such as studies combining them withMEK inhibitors, with other agents

e-GrandRound


PHASE 1 TRIAL OF PLX4032 IN B-RAF MUTANT MELANOMA PATIENTS

Above: Interim results for the maximum tumour shrinkage for patientsinvolved in the study show a 70% response rateRight: These PET scans show a dramatic response to the B-RAF inhibitor

Source: Bar chart – K Flaherty et al. (2010) NEJM 363:809-819; Scans courtesy of Jeff Sosman, Vanderbilt University

Interim best overall response Reduced glucose uptake by day 15

INTERIM PFS FOR PLX4032

This plot of interim phase 1 data show that patients with the V600 mutationon B-RAF show a much better progression-free survival in response toPLX4032 than those with wildtype (WT) B-RAFSource: Courtesy of Keith Flaherty, Massachusetts General Hospital

that might inhibit the development ofresistance, and in combination withimmunotherapy.

Question: If you have a patient withmetastatic melanoma, do you alwaystest for B-RAF and C-KIT now? Is thisstandard?Answer: We test B-RAF status inpatients who develop metastatic diseaseor in those with very high-risk stage IIIdisease with intransit metastases. Wethen incorporate the result into decisionmaking on which the treatments to offerthem. We do not test patients who onlyhave sentinel node involvement or whoonly have primary tumours as there is, asyet, no role established for B-RAFinhibitors in the adjuvant setting.

WHAT CAN BE DONE FORPATIENTS WHO HAVEELEVATED LDH?We are reluctant to give immunother-apy to patients with elevated LDH orwith B-RAF wildtype tumours. Onething that is becoming clear is thatpatients who have elevated LDH maybe the same as those with elevatedVEGF, and elevated VEGF levelswithin tumours correlates with pooroutcome.

The phase II BEAM study ran-domised 200 patients treated with car-boplatin/paclitaxel in a 2:1 ratio tobevacizumab-containing therapy versusplacebo (Advanced Melanoma: Eur JCancer Suppls 7[3]:13).

Survival curves at one year showeda significant, or nearly significant,improvement in overall survival (52%for the bevacizumab-containing arm,compared to 39% for the placebo arm).This was particularly true for patientswith M1c disease or those with M1cdisease and elevated LDH, with sig-nificant benefit for those patientsreceiving bevacizumab.

This has led to a trial proposed in

the US Intergroup by Ryan Sullivantaking patients who have primarilyB-RAF wildtype tumours or poten-tially those with tumours containingB-RAFmutations who are resistant toB-RAF inhibitors, and randomisingthem to either carboplatin/paclitaxel+ bevacizumab or carboplatin/pacli-taxel + placebo.

WHERE ARE WE IN 2011?We are beginning to see a glimmer ofhope on the horizon, with novelimmunotherapies, some specific andhighly active tumour-targeted thera-pies, antiangiogenic therapies, and thepotential ability to select patients forparticular tumour types based onmolecular profiling.

Question: There has been very littlechange in adjuvant therapy over the pastyears. Which patients do you treat withadjuvant therapy? Do you think thosenew promises will also translate intobenefits in the adjuvant setting?Answer:We still use the standard high-dose interferon regimen for patients withstage IIB and stage III melanomawho arenot eligible for research protocols, butwho are physiologically 70 years or less,and able to tolerate interferon.We do notyet knowwhether the new agents that areshowing activity in the metastatic settingwill be active and tolerable in the adju-vant setting.The study carried out in Europe

comparing ipilimumab to control inpatients with stage III disease has hadsome difficulty with toxicity. So itremains to be seen whether any benefitseen, or the number cured, is sufficientto justify the toxicity that patients mayhave to undergo. A trial in the US com-paring ipilimumab to interferon willbe a truer test as patients in the US maybe reluctant to take a placebo.Whether the B-RAF inhibitors will

have a role in the adjuvant setting is

unclear because, at the moment, incontrast to immunotherapy, they areprimarily palliative. They cause dra-matic tumour shrinkage, but do notappear to be causing durable completeresponses. Whether this type of activityin the metastatic setting will translateinto eliminating tumour cells, which iswhat we would like to see in the adju-vant setting, requires investigation.Question: For practical purposes,how do you use interferon? Do youthink it can be replaced by Pegintron[peginterferon alpha-2b], because thatis better tolerated, or do you give onemonth of high-dose interferon followedby low-dose for the rest of the 11months?Answer: In the US, we believe that thehigh-dose, four-week induction periodis the most important component ofinterferon treatment, so we are reluc-tant to adjust that in any way. We havea lower threshold for reducing or stop-ping therapy, because we think thatmost of the benefit of interferon hap-pens within the first 4–12 weeks. If weare going to modify the treatment, it ismore likely in the last nine months oftherapy, either using a lower dose oreven omitting it, rather than modifyingthe four-week induction period. There-fore, we have not moved towards Pegin-tron, at least in our patient population.Question: AreB-RAF inhibitors activein patients who have B-RAFmutationsin their melanoma and who also haveCNS metastasis?Answer: PLX4032 has not been for-mally studied in patients with brainmetastases as these patients have beenexcluded from the trials. But theGSKB-RAF inhibitor has been studied in somepatients with CNS disease. A studyreported at ECCO in 2010 (AdvancedMelanoma: Eur J Cancer Suppls, 8[3])showed activity in the central nervoussystem.We will see more studies includ-ing these patients in the future.


e-GrandRound

The future of cancer,as told through the story of Renee

Merrimack, New Hampshire,usually gets a white Christ-mas.AndwhenReneeWeaver

felt apain inherbackover theholiday sea-son, she assumedshemusthavepulled amuscle while shovelling snow.

When her sight became blurred afew weeks later, there also seemed tobe a logical cause. “I’m about to be40,” she thought. “I must need glasses.”But her optician revealed that neitherailment had so benign an explanation.A mass was pressing on Renee’s righteye but, doctors told her, its cellshadn’t started out there. Though just39, and a non-smoker since her twen-ties, Renee had advanced lung cancer.It had already spread to her brain,bones and liver.

For the mother of Emily, 13, andJacob, 10, the diagnosis could hardlyhave been more devastating. Medianlife expectancy for a patient with somany metastases is just eight months,and the five-year survival rate is lower

than 10%. “The doctor, when she toldus, I really think she thought I washeaded for death,” she said. “Like anymother, I just thought of my children.My God, I won’t see them grown.” Aserendipitous coincidence, however,has thrust Renee to the heart of a gath-ering medical revolution that is startingto give patients like her genuine hope

of beating cancers that would oncehave carried the bleakest prognosis.

With the help of a surgeon whowas treating her husband, Tom, for aheart complaint, she came under thecare of one of the world’s leading can-cer centres, at Massachusetts GeneralHospital (MGH), 50 miles away inBoston. There, she has become oneof the first patients to benefit from anew approach to cancer, based on sci-ence’s growing understanding of thehuman genetic code.

Under this strategy, cancer is nolonger considered as a single disease, oreven as the 200 or more forms thatafflict different organs in distinct ways.Genomic insights are instead definingtumours according to the DNA muta-tions that drive them.

This newparadigm is providing doc-tors with the intelligence they need toattack cancer with smart weapons,


BestReporter

Explaining in everyday language how the science that brought us the Human Genome Project

is giving hope to cancer patients of today and tomorrow is no easy task. Mark Henderson,

science editor for The Times in London, won aBestCancer ReporterAward for this piece, which

was originally published under the title, ‘Making cancer history: killing tumours’.

Mark Henderson

ceptin) and imatinib (Glivec) neutralisethe rogue proteins made by defectivegenes, killing tumours or weakeningthem so the body can finish the job.They have proved to be capable assas-sins, often prolonging life by years, butthey cannot be deployed indiscrimi-nately. For the most part, they workonly against cancers with the geneticsignatures they are designed to target.Doctors must know their enemy whenplanning attacks.

What this means, according toDaniel Haber, director of the MGHCancer Centre, is that “you can nolonger do cutting-edge oncology with-out genetic tests”. It is not enough todiagnose cancer in a patient accordingto how a tumour looks under themicro-scope, and where it is in the body. You

BestReporter


calibrated for individual patients, inplace of the blunt instruments of tradi-tional oncology. It has already started tochange the landscape of medicine, totransform its capacity to contain – andeven sometimes cure – this dreadeddisease.And it is coming to Britain.

This year, Cancer Research UK willbegin establishing a network of centresthat will use similar genetic techniquesto guide treatment decisions for Britishpatients. It will be the precursor totreating every NHS [National HealthService] cancer patient this way, per-haps in as little as five years. Renee’sexperience, though groundbreaking fornow, could soon be expected tobecome routine. You might call it thefuture of cancer.

There is a tendency to think of can-

cer as an environmental disease, trig-gered by exposures such as smoking orultraviolet light. But at root, it is a dis-ease of the genes. It is the result ofDNA defects that cause cells to growunchecked; carcinogens are danger-ous because they inflict this damage.The nucleus of a tumour cell is a placeof genetic chaos, with many thousandsof mutations. It is these that are the lifeforce of cancer cells, feeding theirappetite for proliferation and destruc-tion. But they are also weaknesses thatcan be attacked.

As scientists have started to identifythese mutations – abetted by the workof the Human Genome Project – theyhave begun to develop drugs that canknock them out. Agents such aserlotinib (Tarceva), trastuzumab (Her-

Agents such as erlotinib neutralise the rogue proteins

made by defective genes, killing or weakening tumours

A mission to inform. Thestory of non-smoker Renee,whose advanced lung cancerresponded dramatically totreatment with a targetedtherapy, brings readersrealistic hope temperedwith caution

BestReporter


have to read the molecular instruc-tions that are driving it, the genes thatmake it tick.

Some tests that do this have beenavailable for a few years. Indeed, theNHS requires that they be conductedbefore certain drugs are approved.Trastuzumab is given only to breastcancer patients with a defective HER2gene, while cetuximab (Erbitux) is pre-scribed only for colon cancers withoutmutation in a gene called KRAS.

Provision of testing, however, is stillpatchy, and those patients who do get itare investigated only for the defect thatmost commonly afflicts their particularcancer.A bowel cancer patientmight geta KRAS test, but nothing else.

The MGH programme is takingsuch testing to a whole new level. AtRenee’s first appointment, her doctor,Lecia Sequist, ordered a battery ofgenetic investigations. A sample of hertumourwas screened for about 120 dif-ferent mutations in 13 genes that areknown to affect drug response or prog-nosis. The hope was that she might besuitable for a targeted therapy – or thatif nothing appropriate was available,she might be able to join a clinical trial.

“For someone like Renee, a young,female non-smoker, we had a high indexof suspicion that she might have a suit-able mutation,” Dr Sequist said. “Wepushed the lab to go as fast as possibleand got the results back in eight days.”

They provided a filament of hope.“Dr Sequist had told us that if the testwas negative, we shouldn’t give up,”Renee said. “But when she said, ‘We’vegot your results back,’ therewas a tone inher voice.We could tell shewas happy.”

Dr Sequist was happy because the

test had revealed that her patient’stumour had a mutation in a gene calledEGFR. This is present in about 10–20% of lung cancers, and it is morecommon still in patients with Renee’sage, sex and smoking history. It meantthat she could be treated with erlotinib,a drug that inhibits EGFR.

The test greatly improved her prog-nosis: about 60%of patientswithEGFRmutations respond well to erlotinib. Ithas also spared her the gruelling effectsof chemotherapy: instead of intravenouscourses of highly toxic drugs in hospital,she can take pills at home.

Renee does not look like a typicalcancer patient. The only visible clues toher treatment are a headscarf, to hidehair lost during radiotherapy thatshrinks her brain lesions, and a little“teenage acne”. She has had some diar-rhoea and nausea, and she has lostsome weight. But as her husband says:“We’d rather you were alive with a bit ofa rash. I just think you’re a trooper.”

The beauty of MGH’s wide-rangingtest is that, even had Renee’s tumourlacked an EGFR mutation, it couldhave revealed other genetic guides. Amutation called ALK, for example,present in about 5% of lung cancers, iscommon in patients with her profile.

“What we’re doing is to capture allwe can of what’s happening geneticallyin the tumour, to look at everythingthat might possibly inform a treatmentdecision,” said Leif Ellisen, co-directorof MGH’s translational research labo-ratory, which developed the panel oftests, known as Snapshot. “If you reallywant to have personalised medicine,you have to test a broad spectrumof mutations in every tumour.”

“We’re testing for all the majormutations we feel can affect therapy,either now or in the near term,”Dr Sequist said. “We’re looking at themutations for which there are licensedtargeted drugs, as well as for sometreatments that are currently in devel-opment. That way, when the drugs areready, we’ll be ready too.”

Since March, this approach hasbecome the standard atMGH for everypatient with advanced lung, breast,gastrointestinal or brain tumours, aswell as for malignant melanoma, themost aggressive form of skin cancer.About 900 people with these cancers –the types for which molecular diagno-sis is currently most useful – will havethe Snapshot test this year.

Manymore can expect to benefit infuture. New genes are being added allthe time – there are now 16 on thepanel, three more than when Reneewas tested in February. The cancers forwhich it is indicated will widen furtheras the International Cancer GenomeConsortium, a £600million [€665mil-lion] project to find all the mutationsthat drive 50 common tumours, beginsto deliver results. The initiative hasalready borne fruit: in December, theentire genomes of a lung tumour and amelanoma were published, identifyingmany newdefects that could be drivers.

The next challenge is to roll outsuch programmes, so that they reachpatients in small hospitals, and applythem earlier to treat patients whosecancers have yet to spread.

Trastuzumab has already made thissecond step, slashing recurrence ratesafter surgery for HER2-positive breasttumours.

A sample of her tumour was screened for about 120

mutations in 13 genes... the results came back in eight days


At a cost of about $1000 [€700] andfalling, the Snapshot test is well withinthe affordable range of health providerssuch as the NHS. And MGH’s oncol-ogists are convinced that their methodswill soon become commonplace. “Forthis to become standard in five years isnot unrealistic,” Dr Sequist said.

It is a position with which ProfessorPeter Johnson, the chief clinician ofCancer ResearchUK, concurs. “There’sno doubt inmymind that this is thewaycancer medicine is going,” he said.

So convinced is the charity of thepotential of broad-spectrum geneticdiagnosis that it is establishing a pro-gramme to provide similar services inBritain. In the pilot phase, it plans toset up genetic testing centres at up tosix hospitals, with the capacity to scanabout 6000 patients a year for a rangeof mutations. It hopes to extend thescheme to every NHS hospital.

“We think it’s clear that this train isalready moving,” Professor Johnsonsaid. “It’s time to take the initiative toapply these insights. Molecular typingof cancer will be in routine practice formany people before long. It’s a matterof making sure we’re ready.”

For Renee, the roll-out cannotcome too quickly. “If you’re just some-one who lives in a small town anddoesn’t have access to MGH, is yourlife not as important as someone wholives near Boston?” she said. “I’m luckyto be an hour away. This should bestandard everywhere.”

The pace of advance has also awak-ened business to the possibilities of amajor new market. For example, Foun-dation Medicine, a Boston-based com-pany that launched in April, aims todevelop a one-stop-shop for genomiccancer diagnosis that is accessible toany hospital. Its advisers include Eric

Lander, a pioneer of the HumanGenome Project.

This enthusiasm for widespreadgenomic diagnosis of cancer hasemerged because of the astonishingprogress made by targeted therapiesin the past decade. Their potential firstbecame clear in 2001 with the adventof imatinib (Glivec), a drug designed toshut down a mutant gene that causeschronic myeloid leukaemia (CML). Ittransformed treatment of the disease.Patients who would once have beenexpected to die within months are stillalive today, often living ordinary lives.The team who developed it, BrianDruker, Nicholas Lydon and CharlesSawyers, were awarded the Lasker-DeBakey Clinical Medical ResearchAward last year for “converting a fatalcancer into a manageable chronic con-dition”. Similar targeted therapies havefollowed for some solid tumours,

BestReporter

At a cost of about $1000 and falling, the Snapshot test

is well within the affordable range of health providers

MUTATIONS EXPLAINED

The DNA (deoxyribonucleic acid) in a cell’s nucleus makes up genes(bits of code). Genes dictate what a cell does, what proteins it pro-duces (and thus how we are made) and when it reproduces. This repli-cation enables the body to grow and repair itself, but it is also whenmutations, i.e., mistakes, occur. These can be inherited or result from

environmental factors such as UV light or smoking (called mutagens).Sometimes our cells can fix these mutations, but if they can’t theyare passed on to future copies of the cells. These cells normally can-not survive but in some cases they keep on dividing until a lump, ortumour, is formed.

especially those of the lung, bowel andbreast. They have even begun to crackone of the deadliest and least tractablecancers of all: melanoma.

This skin cancer is readily treat-able if caught early, but once it hasstarted to spread it is a reliable killer.Chemotherapy rarely works, and fewerthan one in ten patients live for a year.A diagnosis of metastatic melanoma isthe proverbial death sentence.

On the wall of Professor MikeStratton’s office, though, hangs aremarkable picture. On the left is apositron emission tomography (PET)scan of a melanoma patient, whosebody is riddled with cancer. It is sopockmarked that it resembles a Dal-matian. Next to it is a PET scan of thesame body, taken 15 days later. It isalmost completely clear: the cancerhas melted away.

It is an image of which ProfessorStratton, who heads the CancerGenomeProject at theWellcomeTrustSanger Institute in Cambridge, is justi-fiably proud. For the drug that causedthis extraordinary transformation wasdeveloped as a result of a genetic dis-coverymade by his team just eight yearsago. “It is an incredible kick for me andmy colleagues to look at,” he said. “Onbad days, I look at that picture and Ithink things are all right. It gives ushuge satisfaction to know that our workcan make that sort of difference.”

That difference began with the dis-covery in 2002 that a mutated genecalled BRAF is present in about 70% ofmelanoma tumours. This insightallowed Plexxicon, a biotechnologycompany, to develop a BRAF inhibitor

called PLX4032. In initial trials, 80% ofpatients responded so well thattumours often vanished from theirscans; the PET images are from one ofthe trial’s participants.Almost two yearsafter starting treatment, about one infive patients who responded remainsclear. “The response was far more spec-tacular than we expected,” said KeithFlaherty, of MGH, a leader of the trials.The drug is expected to be licensednext year.

The BRAF experience also high-lights another fascinating aspect ofcancer genomics that is changing the

way the disease is diagnosed andtreated. This is because melanoma,caused by exposure to ultraviolet radi-ation from tanning salons or the Sun, isnot the only cancer in which this genemalfunctions. BRAF can also bemutated in colon, lung and thyroidcancer – all organs on which the Sundoesn’t shine. The same is true else-where: EGFR mutations are found ingastrointestinal and brain cancers, andthe ALK mutation that drives somelung tumours was originally identifiedin lymphoma.

What this suggests is that cancers


BestReporter

They have even begun to crack one of the deadliest

and least tractable cancers of all: melanoma

A copy of the PET scan images of a melanoma patient that hang on Mike Stratton’s office wall.On the left, it shows the patient before treatment with PLX4032, and on the right, 15 days afterstarting the therapy

Prior to treatment After 14 days of treatment

that oncologists would never beforehave grouped together might wellbenefit from a similar therapeuticapproach. The tissue in which atumour is found may even be less of aguide to the best treatment than itsmolecular subtype.

While this is still a hypothesis,research is now putting it to the test.Trials are already under way to examineif BRAF and EGFR inhibitors work oncancers no matter where they began.

Dr Sequist said: “We used to thinkthere was one cookbook for colon can-cer and another for lung cancer. Butthere may be general treatment algo-rithms for colon cancer with BRAFthat are the same for lung cancer withBRAF. The paradigm that certain drugsonly work for cancers in a certain tissueis just old-fashioned.”

For all the improvements that tar-geted therapy has delivered, however,none tell of unqualified success. Inthe first place, treatments still existonly for a subset of cancers: there is noagent, for instance, suitable for boweltumours with mutant KRAS. In thesecond, though most patients whosetumours match a tailored drug respondfor a time, their cancers often return.

PLX4032, the BRAF inhibitor, is aprime example. In trials, its effectsgenerally delivered about tenmonths ofremission, after which tumours startedto progress: some patients who initiallyhad results as spectacular as ProfessorStratton’s scans have since died. “That’sstill real progress,” Dr Flaherty said.“For a metastatic melanoma patient,even ten months is a reprieve you rarelyget with chemotherapy. Then you fac-tor in the quality of life. But of course

it’s not as much as we’d like. We’re notcalling it quits.”

The problem is that cancers, likeviruses and bacteria, can evolve resist-ance to drugs. This can happen in twoways. A tumour can acquire a newmutation that drives it forward evenwhen BRAF or EGFR is knocked out.Or a few cells in a primary tumourmay be resistant from the outset.Whenthe larger number of susceptible cellshave been killed, the resistant onescan take their place.

But if resistance remains a signifi-cant hurdle, there are grounds for opti-mism that it will not always be aninsuperable one. “We’re still losingmost of our patients,” Dr Sequist said.“But I do think that will improve.”

Several trials are under way inwhich targeted therapies are beinggiven alongside other drugs that, it ishoped, will prevent or delay resistance.Renee is participating in one of them:as well as erlotinib, she is receiving adrug called hydroxychloroquine, origi-nally developed as an antimalarial,which may have an anti-cancer action.The hope is that this cocktail will pro-long the effectiveness of her treatment.

The MGH doctors all feel that thismultidrug approach to cancer willbecome increasingly important. Severalmade the analogy with HIV, which canbe controlled with combination ther-apy, in which different antivirals guardagainst the development of resistance.

The toxicity of cancer drugs maysometimes prevent them being givenall at once, as with HIV agents, but itmay be possible to deliver whatDr Sequist calls “pulses” of treatment– a few weeks of erlotinib, followed

by a few weeks of something else –to hold a cancer in check. Shouldtumours regrow, it will also be neces-sary to retest them for mutations, topick up any new ones that require freshtherapeutic tactics.

There is another reason why cancerresearchers and clinicians speak oftenabout HIV. This is that while there isno HIV vaccine, and no cure, the viruscan be suppressed for long periods oftime. HIV-positive people, such asLord Smith of Finsbury, the formerCulture Secretary, have taken combi-nation therapies for two decades,remaining well enough to hold downhigh-powered jobs.A once-fatal infec-tion has become a chronic one thatcan be managed so effectively it isquestionable whether one should thinkof its carriers as ill.

The advent of genetically targetedtherapies, and the molecular diagnos-tics that underpin them, has brought asimilar goal into view for cancer. AsEunice Kwak, another MGH doctor,put it: “If you could cure cancer thatwould be phenomenal, but if you couldmake it similar to HIV, so that insteadof being a lethal disease it becomessomething you can live with and man-age, that would be a huge advance-ment. It is maybe not so far away.”

Dr Haber goes farther, noting thatfor patients whose cancer is caughtearly, before it has spread, targetedtherapy could really change the mean-ing of the ‘C-word’ to cure. “We do OKwith metastatic cancers, but it’s a bigchallenge to fight a cancer that has gotthat big,” he said. “The key is to get tar-geted therapies in there early.”

Professor Stratton agrees. “I think

If resistance remains a significant hurdle, there are

grounds for optimism that it will not be insuperable

BestReporter


we will increasinglymake cancers chronic,”he said. “Butwe shouldbe aiming for cures –wecan aspire to a highergoal.” Genomic treat-ment, he says, will dofor many tumours whatadvances in chemo-therapy did for testicular cancerpatients such as LanceArmstrong. “Wedon’t think much about testicular can-cer now: 40years ago it was a 100%killer, now 90% of young men arecured,” he said. “These therapies willadd to that group of patientswho can betreated and hopefully cured. Genomicstrategies are proving so extraordinarythat I would personally voice optimismthat ways will be found.”

As genomics brings cancer moresharply into focus, these incrementalsteps forward are starting to becomestrides. “As I see it, 20, 30, 40 years ofgenetics are now coming to an applica-tionwedidn’t have before, to new thera-pies that are smartly designed,”DrHaber

said. “I really think it is a revolution.”It is a revolution that reached the

Weaver household on May 5, when,just sevenweeks after starting erlotinib,Renee returned to MGH for the resultsof her second scan. She was braced forthe worst. Dr Sequist, though, wassmiling as she opened the door.

“She told me she had some goodnews, and showed me the scans,”Renee said. “It shocked us all: the pri-mary tumour had shrunk by about 80%,and you couldn’t see any other spots atall. I’d hoped she might say there hadbeen a little shrinkage, but I hadn’tdared to hope for anything like this.”

She is not out of the woods justyet: when erlotinib works, the response

typically lasts a year. But, as Dr Sequistsaid, hardly any patients like Reneewho have chemotherapy do so well.There is a chance, too, that she couldbecome one of the growing group ofpatients in whom erlotinib controlslung cancer for years.

“I’m taking it a day at a time,” Reneesaid. “Whether the results will be thisgreat next time, who knows? But I’mfeeling good. I’m trying to go back to anormal life. Thank God for theresearchers and doctors who havemade this happen. I don’t want to thinkabout where I’d be without them.”

This article was first published by The Times in theJune 2010 edition of its monthly science magazineEureka, and is republished here with permission


BestReporter

“I think we will increasingly make cancers chronic ...

but we should be aiming for cures”

The Weaver family,including daughter Emily

and son Jacob, posetogether at home. Reneesaid her worst fear whenshe was diagnosed wasthe thought of not being

able to watch her childrengrow up. Targeted therapy

has, at the very least,bought her more time to

spend with them

The ESO Masterclass:where those most eager to learn meet

those most willing to teach

� Peter McIntyre

Once a year, in a quiet location somewhere in Europe, 60 of the continent’s brightest andmost

motivated young oncologists gather for a learning experience that can define their careers. This

week-longMasterclass lies at the very heart of ESO’smission, building an army of brilliant and

caring young oncologists, many of whomwill be the leaders of tomorrow.


Masterclass

would not let them go for a week. So the courseswere cut back to one or two days.

“Now we have 60 students for a week. This isexactly what people of this age need. They are nowat the agewhen they have to choosewhat specialtythey will follow but options are still open to them.You cannot comehere through paying a registrationfee. You have to write a motivation about why youwant to come and you have to prepare a case studyand then youmay be accepted.”

LEARNING FROM DIFFICULT CASESWhat is now theESO–ESMOMasterclass inClin-icalOncology celebrated its 10th anniversary inApril2011 at theWolfsbergCentre, by LakeConstance,inSwitzerland.The venue (it belongs toUBS invest-ment bankers) exudes a sense of calm order. TheMasterclass is a ferment of bubbling activity.

The daily routine is rigorous: breakfast by8.00am, first presentation at 8.30am, and (withbreaks) presentations through till 4.30pm.Then theday begins again, as the participants split intogroups and present and discuss case studies fromtheir own hospitals. By the time they close, it hasbeen 11 hours of intense concentration.

Thepresentations read like aWho’s WhoofEuro-peanoncology:AronGoldhirschonbreast cancer, JanVermorkenongynaecological cancers, JacquesBernieron head and neck cancers, Eric van Cutsem oncolorectal cancer, Rolf Stahel on lung cancer.

For the case study discussions, nobody brings astraightforward case. The history is revealed step bystep, usually with recurrence or complications fol-lowing treatment and some tough choices at theend.With 15 to 20 young oncologists in each groupthere are echoes of the hit TV series House, inwhichbright youngdoctors outdo eachother indiag-noses and suggested treatment.

The faculty member for one group is NicholasPavlidis, professor at the Department of MedicalOncology in Ioannina,Greece, and joint chair of theMasterclass.He’s far too nice to be ‘DrHouse’, butthere is somethingmagical about his ability to popupwith a slide containing themost relevant researchresults during discussion.

And (unIike the malignant Dr House) Pavlidisalways brings the students back to the human per-spective– these arenot casesbutpeople. Soonecasestudy involvesamotherof twochildrendiagnosedwith

The European School of Oncology wasfamously founded on a misdiagnosis andits unique selling point is to prevent tragedies

based on ignorance and lack of education.The legacy on which the School was launched

in 1982was bequeathed by an Italian businessmanwho was treated for a year for arthritis before dis-covering that his pain came from bone metastasesoriginating from undiagnosed prostate cancer.

Thewholepremise of theSchool is thatmistakeslike this would not happen if doctors were bettereducated about cancer and if specialists were up todate with and applied the latest research and bestclinical practice. The Milan group who createdESO was evangelical about improving doctors’knowledge and understanding about latest treat-ments and diagnostic tools.

ESOeducation is about incorporating researchthat updates state-of-the-art best practice, whileretaining tried and tested treatments thatwork. It isthe application of knowledge to the treatment ofpatients that is at the heart of ESO initiatives,exemplified in its motto ‘Learning to care’.

It does this by enlisting the best experts fromaroundEurope to present the evidence for first-linetherapies and subsequent treatments, asking andanswering “what if” questions on the way. Theemphasis is on multidisciplinary ways of workingand the importance of becoming a true specialist sothat all patients are in the hands of doctors whoknow the evidence and how to apply it.

In the variety of ESO courses – Inside Track,Insight conferences, a course formedical students– the Masterclass in Clinical Oncology for youngoncologists has come tobe seen as a key event bring-ing together those most eager to learn and thosemost willing to teach.

The annual Masterclass is a week of intensivelearning and interaction for 60 of the brightest andbest: 30- to 40-year-olds on the cusp of decidingtheir ultimate oncology specialism. The course isdedicated to medical and clinical oncologists andthe focus is on the big killers: lung, breast, prostate,colorectal and gynaecological cancers.

AlbertoCosta, director of ESO, says that theseMasterclasses take the School back to its roots.“Whenwe startedESO, courseswere like this – oneweek residential courses. But then people had nomoney and peoplewere too busy, and the hospitals

Masterclass


ENNIO

LEANZA

colorectal cancer in 2008. She has been treated foraggressive disease and liver metastases with cyto-reductive surgery and hyperthermic intraperitonealchemotherapy (HIPEC).ByMarch2011, aftermanycyclesof treatment, there isnoobvious signofdisease.Should they be adopting invasive diagnostic tests?Pavlidis askswhat theoutcomewill be if thecancerhasreturned and all agree that they will have run out ofoptions. Gently, he suggests that the additional testswill bring distress to the patient without benefit.

Another case concerns a youngwomanwhopre-sented with adenocarcinoma, and was initiallytreated apparently successfully. But, after recurrenceand further treatment, hepatic metastases werefound. Despite hepatic metastasectomy, it is clearthat the cancer has not been cured. Pavlidis pointsout that she is still alive six years after her initial diag-nosis, and formuchof that timehas been able to live

a normal life. Is this a treatment that has failed, asthe young doctormay feel, or one that has given thisyoung woman years of good-quality life, criticalyears with her young children?

Pavlidis has been described as the ‘father’ of theMasterclass. Almost 30 years ago, in 1982, as ayoung Fellow at the RoyalMarsdenHospital in theUK, he attended the first ESO course inMilan. “Ifyou look at who were participants then, more than50%of themarenowprofessors inuniversities insideoroutsideEurope,”hesays.He is sure thispatternwillbe repeatedwith the currentMasterclass series.

“Weoffer real educationbydistinguished expertsfromEurope.Whatmakes the difference is that notonly do you teach the big killers but you also havethese people sitting in these kind of groups pre-senting their ownclinical cases.Weanalyse andcrit-icise and give directions.”


Masterclass

What if…? Students bring difficult case studies for discussionwith international experts such as lung cancer specialist RolfStahel, pictured here

There is something magical about his ability to pop

up with a slide containing the most relevant research

ENNIO

LEANZA

UP FOR THE CHALLENGEThe participants are clearly up for the challenge.JochanBauerqualifiedas adoctor sevenyears agoandnowworks in internalmedicine at theEsslingenHos-pital in Germany, specialising in cancer and haema-tology. “My consultant took part in this Masterclassseveral yearsagoandrecommended itwarmly,”hesaid.“He is certainly right about this. It updates youon thefacts and on standard diagnosis and treatment. It ishardwork, but it is definitely worth it.”

Andreia Costa, a medical oncology fellow atthe Hospital de São João, Portugal, takes her finalmedical oncology exams in 2012. “This is a fantas-tic revision for me and it is very important for mytraining. It is very intensive, but we do get somebreaks between sessions. I particularly value thedis-cussion of the cases in small groups andover coffee.”

Davit Zohrabyan, a medical oncologist at theYerevanStateMedicalUniversityHospital inArme-nia, sees the course as a lifeline to knowledge thatis not easily available in his home country.He savedup the questions he could not answer at home sohecould ask the experts here. “Armenia is a smallcountry and it is not rich and it does not have bigresearch programmes. It is very important for meand for my country that I get acquainted with thenew methods and practice. I want to be informedwith all the important new scientific knowledge. Iwant to work together with science.”

Inevitably, after 10 years, the pupils start tobecome teachers. Elżbieta Senkus-Konefka wasone of 53 students on the very firstMasterclass heldinMontecantini, Italy, in 2002.Now a specialist inradiotherapy and medical oncology in Gdansk,Poland, she has returned for the past two years tolecture on treatment for metastatic breast cancer.

“My career is an example of someonewho grad-uates from these courses and then becomes anexpert.When I applied Iwas just starting onmedicaloncology. I wrote that this was at a very importantpoint formebecause I had just switched to a secondspecialty. Ten years ago hardly anyone from eastern

Europe went to international conferences or wasconsidered an expert. Iwas one of the few to do so.”

Razvan Popescu, medical oncologist at theHirslandenClinic inAarau, Switzerland, is one oftwo scientific coordinators. “One of the reasonswhy I became involved is because of the experi-ence of both good and bad quality teaching that Ihave had myself.”

Popescu is active with the European Society forMedical Oncology (ESMO), and in his work on theESMOawardscommitteehesawmanyCVsof youngdoctors from eastern Europe who needed support.“Certaincountries, not exclusively incentral andeast-ernEurope, have teaching that ismandatory for themto attend, but it does not take them forward – theydon’t learn. These young people are the future of

“More than 50% of the class of 1982 are now

professors in universities inside or outside Europe”

Masterclass


AN EXCELLENT REPORT CARD

Students rate the Masterclass highly for useful and relevant educationA review of doctors attending the first nine Masterclasses shows thatalmost 60% were aged 30–34, and 37% were aged 35–45. Partic-ipants were evenly divided, males and females.The vast majority of participants (72%) were medical oncologists, with6% radiation oncologists and 6% clinical oncologists. Some had notyet finally decided on their areas of specialisation. Two thirds (67%)came from European Union countries, 25% from non-EU countriesof Europe and 8% from outside Europe – mainly the Middle East andLatin America.Feedback by students is very positive. Over the first nine years theoverall median score for the quality of education was of 3.54 outof 4. Information was rated as useful and relevant (median 3.64) andwell balanced with good evidence (3.59), with adequate time for dis-cussion and questions (3.51).Masterclasses have been held in Italy, Spain, Cyprus, Malta, Bulgariaand Portugal, but the Wolfsberg Centre, at Ermatingen, Switzerland,received almost perfect scores for facilities, management andorganisation, and it is clear that ESO hopes it will be able to return.Source: N Pavlidis et al. (2010) The Masterclass of the European School of

Oncology: The ‘key educational event’ of the school. Eur J Cancer 46:2159–65

oncology in Europe.What ESOhas been doing, andI am delighted it is now doing it in partnership withESMO, has been to set up a very, very good course.One of themain drivers forme to continue to partic-ipate is the live interactionwhichdevelopsbetweenthefaculty and students.Theyget thecrèmeofEuropeanoncology coming and lecturing, but also interactingwith them.”

Should the course ease up a little? Can anyoneconcentrate for 11 hours a day? The other coordi-nator,WolfgangGatzemeier, believes that the ben-efit isworth thepain. “At thebeginning, I say you arehere for a marathon. I did it for 10 years and I sur-vived and you will survive as well. I know it is noteasy to followpresentations for such a long time, butthey also have time to reflect and talk to eachother.”He points out that the students do not haveto remember everything as they get a book of pre-sentations and access to the slides online.

THE LATEST DATA - IN CONTEXTGatzemeier, a breast surgeon, believes that theMasterclasses have added value over the large sci-entific conferences. “When you go to themajor con-ferences and congresses you can get lost in all thatthey have to offer. Here, in thismore relaxed ambi-ence, people are taught by top faculty in a veryintensive way. The most important information ispresented and everybodyhas an opportunity to gainasmuch as possible fromwhat is offered.

“We ask people at the end if it is too much andyes, some complain that it is too compressed, butthey also always ask us to do something more aswell! Fromour follow-up,we see that formany par-ticipants this was a crucial point in their career, sothat theymay have the opportunity to becomeheadof a department.”

ESO is not resting on its laurels. InMay this yearit held aBalkanMasterclass inClinicalOncology inDubrovnik,Croatia, co-chaired byPopescu,who isoriginally fromRomania, and by Semir Bešlija fromthe Institute ofOncology in Sarajevo, Bosnia. Likethe full Masterclass, this was a residential course


Masterclass

“From our follow-up, we see that for many

participants this was a crucial point in their career”

It’s for nurses too. Twenty-two cancer nurses attendeda nursing Masterclass held in parallel; this group isdiscussing issues in patient communication

paid for by sponsors, although slightly shorter, atthree days. TheBalkans coursewas also open to sur-geons,who arenot coveredby themainMasterclass.

Popescu explained that the aimwas to devise acourse that was particularly relevant to the clinicalenvironment in the region. “Some people comeand hear things that they won’t ever be able to putinto practice, like monoclonals and targeted smallagents that cost a fortune.

“There are a fewpeople from the regionwho areexceptionally bright and sometimesmanage to cometo western institutions and do superbly well. It ispainful that if they want to return to their ownregion they find they are not only struggling withfinancial issues, they are also strugglingwith systemicissues and the culture.”

A MASTERCLASS IN CANCER NURSINGTheMasterclass in Switzerlandwas not only aboutdoctors: 22nurses also attended from15countries.It was the tenth Masterclass for doctors, and thefourth for nurses. Funding is in place for thenursing

PETERMCINTYRE

course to takeplace alongside themedical course forthe next two years. Some presentations are sharedwithmedics, but thenurses also break out to look atthe advanced practice of nursing cancer patients.

SaraFaithfull, President of theEuropeanOncol-ogy Nursing Society (EONS), chaired the nursingcoursewithmany of the sameobjectives. “The ideais that we are training future leaders, so this is verymuchabout clinical experience andabout looking atkey skills and developing practice.We have lookedat putting in place the advanced assessment skillsthat inmany countries aren’t systematic or taught. Itis not just about the latest breast cancer treatment,it is about how you would, as a nurse, manage tar-geted therapies andmanage the side-effects and lookmore specifically at the quality of life issues andenhancing care.”

The pedagogical methods used by the nurses areverydifferent,basedaroundgroupworkandgroupdis-cussion. In a communication segment they discusshow to respondwhen apatient inevitably asks: “Howlong have I got?” or when a relative says: “Don’t tellthemwhat they have got.”

Nurse training and practice differ across Europe,andEONShas beenworking to improve online train-ingopportunities fornurses,manyofwhomget less thanthree days a year away from the ward for training. Butaswith thedoctors, there is little thatcanbeat the face-to-face interaction.

Sandrine Decosterd, from theGenevaUniversityHospitals, feels thatthispersonal sharingof experienceandmotivation is most valuable. “Thiscoursealsogives youa lotofknowledgein a short time–both scientific knowl-edge and nursing knowledge.”

Elana Laska teaches palliativecare nursing inAlbania, where thereare few oncology nurses.As an advo-cate to improve the role and skills ofnurses, she was keen to share herknowledge when she got home.“Every time I go for training I prepare

a report. I hold meetings with the team and givethem my experiences. I write notes every nightaboutwhat I have seen and all the things I am think-ing about howwe can do something differently.”

EONSisplanning togetacademicaccreditation forthe next nurseMasterclass using case studies onlineafterwards to see how it has changednurses’practice.

AlbertoCosta feels that theMasterclass conceptis much stronger when it includes doctors andnurses and he praises the sponsors for puttingmoney into education,withoutwanting to put theirnames all over everything.

Cutting the cake to mark the 10th anniversaryof the medical Masterclass, he pointed out thatthere are now more than 500 people who havebeen through the course, “a little army of brilliantyoung oncologists all over Europe”.

The average age of those attending the courseseems tobedropping and themajority areunder35. Itwillnotbemanyyearsbefore thestudentsattending theMasterclass will not even have been bornwhenESObegan.Fornow, it canbe said that theyhave grownuptogetherandthat together theyundoubtedlymakeadif-ference to the practice of oncology.

The Masterclass is funded by an unrestricted grant from a pool ofsponsors through ESO’s Sharing Progress in Cancer Care programme

Masterclass


Happy 10th anniversary. ChatrinaMelcher, one of the organisers, withESO director Alberto Costa (right)and co-chair Nicolas Pavlidis (left)celebrate with cake and presents E

NNIO

LEANZA

ImpactFactor

Denosumab – a new option forsolid tumours metastatic to bone� Philip Saylor

Zoledronic acid is a potent bisphosphonate used as the standard therapy for the prevention of

skeletal-related events in patientswith solid tumoursmetastatic to bone. Three phase III studies

have reported head-to-head comparisons of zoledronic acid with denosumab, an inhibitor of the

RANK signaling pathway.

Skeletalmorbidity is a substantial bur-den inmanypatientswith advancedsolid tumours. Pathologic fractures,

pain, spinal cord compression and hyper-calcaemia are among thepotential compli-cations of bone metastases. Osteoclast-mediatedbone resorptionhasan importantrole in thepathophysiology of bonemetas-tases as it weakens the bone and liberatesgrowth factors that can stimulatebothcan-cer growth and further bone turnover.1

Therefore, inhibitingosteoclasts is a rationaltherapeutic strategy and bisphosphonatesand denosumab are two available types ofosteoclast inhibitors.

Bisphosphonates are analogues ofpyrophosphate – a normal component ofbonematrix.Oncedepositedwithin bone,they inhibit osteoclasts locally. Zoledronicacid is the most potent bisphosphonateavailable. Since its approval in 2002, ithas been the standardbone-targeted treat-ment for the prevention of skeletal-relatedevents (SREs) in patients with solid

tumours that have metastasised to bone.Denosumabnowrepresentsanewclass

of osteoclast-targeted therapy as it inhibitsthe receptor activator of nuclear factorκB (RANK) signalling pathway. RANK ispresent on osteoclasts throughoutmost oftheir life cycle and its signalling promotesosteoclast differentiation, activation, andsurvival.2 Osteoprotegerin (OPG) is anendogenous decoy receptor to RANK-ligand (RANKL) that negatively regulatesthis pathway.Denosumab is amonoclonalantibody that functions as an exogenousOPG, suppressing markers of osteoclastactivity for severalmonthsaftera singledosein some settings.3

Three randomised phase III studieshave recently compared denosumab withzoledronic acid. Fizazi et al.4 reported thatdenosumab is superior to zoledronicacid formenwithcastration-resistantprostate can-cer (CRPC) thathasmetastasised tobone.This phase III study enrolled 1904 menwithCRPCwhowere randomlyassigned to

treatment with denosumab (120 mgadministered subcutaneously every fourweeks) or zoledronic acid (4 mg adminis-tered intravenously every fourweeks). Theprimaryendpointwas time to first on-studySRE(pathologic fracture, radiation therapyto bone, surgery to bone, or spinal cordcompression) and the primary objectivewas todemonstratenoninferiority of deno-sumab to zoledronic acid. One of the sec-ondary objectives was to demonstrate itssuperiority. Median time to first on-studySRE was 20.7 months with denosumaband 17.1 months with zoledronic acid(HR=0.82; 95%CI 0.71–0.95; P=0.0002for noninferiority; P=0.008 for superior-ity). Time to disease progression and over-all survival rates were similar between thetwo groups.

In another phase III study, Henry andhis co-authors reported that denosumab isnoninferior to zoledronic acid in patientswith metastases due to non-prostate andnon-breast cancer.5 This study enrolled


This article was first published in Nature Reviews Clinical Oncology 2011 vol.8 no.6, and is published withpermission. © 2011 Nature Publishing Group. doi:10.1038/nrclinonc.2011.70, www.nature.com/nrclinonc

ImpactFactor

1776 patients with advanced cancer ormultiplemyeloma thathadmetastasised tobone. Median time to first SRE was 20.6monthswithdenosumaband16.3monthswith zoledronic acid. Noninferiority, butnot superiority,wasdemonstrated fordeno-sumab. The disease progression and over-all survival rates were similar between thetwo groups.5

A third similarly designed study ofwomen with breast cancer metastatic tobonehadpreviously reported superiority ofdenosumab(HR=0.82;95%CI0.71–0.95;P=0.01) comparedwith zoledronic acid inreducingSREs.6 Takentogether, thesestud-iesestablishdenosumabassuperior to zole-dronic acid in breast cancer and prostatecancerandnoninferior to zoledronicacid inother solid tumours. On the strength ofthese three studies, the FDA approveddenosumab inNovember2010 for thepre-vention of SREs among patientswith solidtumours thathavemetastasised to thebone.

Denosumab represents an effectivenewoption for themanagement of skeletalmorbidity,buthowis theclinician tochoosebetweendenosumab and zoledronic acid?It is worth noting that none of the threetrials demonstrated a difference in overalldiseaseprogressionor overall survival. Thechoice, therefore, must be driven by con-sideration of drug toxicities and efficacy inpreventing SREs.

Inprostatecancer, denosumab is supe-rior to zoledronicacid. Is it superior enoughfor clinicians to reach for it exclusively?Zoledronicacidbecamestandardofcare forbone-metastaticCRPCwhen itwas shownto prolong median time to first SRE from10.7months to16.3months–a5.6-monthimprovement compared with placebo.7 Inthe trial reportedbyFizazietal.4, timeto firstSRE improved by 3.6 months with deno-sumabcomparedwithzoledronicacid.Sim-ply put, the magnitude of the additionalbenefit of denosumab over zoledronic acid

is about two-thirds of the benefit of zole-dronic acid over placebo. This significantadditional benefit should leadus to choosedenosumab for men with prostate cancer.Thehazard ratio for first on-studySREwasidentical in the breast cancer trial, indicat-ing thatdenosumabshouldbe thestandardof care in this population aswell.

In non-prostate, non-breast tumours,the difference between denosumab andzoledronicacid is simply tooclose tocall. Inthe trial byHenry andcollaborators, enrol-ment featured approximately 40% ofpatients with non-small-cell lung cancer(NSCLC) and 50%with a variety of othernon-breast, non-prostate solid tumours. Inthe absence of any specific contraindica-tions, either drug is a reasonable choice.Denosumab seemed to be inferior to zole-dronic acid in the subset of patients withmultiplemyelomaandhenceshouldnotbeused for this indication.

In patients with renal dysfunction,denosumab offers a potential advantage.Zoledronic acid is recommended atreduced doses for stable mild renal insuf-ficiency (glomerular filtration rate (GFR)of30–60 ml/min) and is contraindicated inpatients with evolving renal dysfunctionor aGFRof <30ml/min.Denosumab hasa long half-life (28 days)8 and its clearanceis not dependent on kidney function.Althoughdenosumab isa rationalchoice forpatients with renal dysfunction, availabledataare limitedbecause suchpatientswereexcluded from the phase III trials thatincluded zoledronic acid arms.

Do the toxicityprofiles of the twodrugshelp in deciding which of the two is pref-erential? Zoledronic acid often causes aflu-like acute-phase reaction, but this isgenerallymildandresolveswithoutmedicalintervention. InhibitingRANKsignalling incells from the immune system raises thepossibilityof riskof infectionafter treatmentwith denosumab,9 but the incidence of

infectious adverse events did not signifi-cantly differ during the2–3 years of followup in any of the phase III trials.4–6

Hypocalcaemia and osteonecrosis ofthe jaw (ONJ) are important potentialadverse effects of both drugs. Hypocal-caemiacanoccurwithanypotentosteoclastinhibitor, although it seems to be morecommonwith denosumab thanwith zole-dronic acid (13% vs 6% in the study byFizazi et al. and 10.% vs 5.8% in the studyby Henry et al.). Because vitamin D defi-ciency raises the risk for hypocalcaemia, itis important to verify a normal serum vita-minD level before therapy and to encour-age calcium and vitamin D supplemen-tation during therapy. ONJ is a rare butimportant potential toxic effect of bothdrugs, occurring in1–2%of theparticipantsin these trials. Appropriate dental carebefore initiation of therapy is likely themost important preventativemeasure.10

These threephase III trials arepractice-changing for patients with solid tumoursthat are metastatic to bone. Denosumabshould be our first choice in men withCRPCandwomenwithbreast cancer.Forother solid tumours, zoledronic acid anddenosumabare equally supportedbyhigh-level evidence.

Details of the references cited in this article can be

accessed at www.cancerworld.org


Practice pointThree phase III trials establish deno-sumab as an effective new option toreduce skeletal morbidity in patientswith solid tumours that have metasta-sised tobone.Denosumab is superior tozoledronicacid forpatientswithprostateor breast cancers and is noninferior forpatients with other solid tumours.

Acknowledgement: Philip Saylor is supported by Young Investigator Awards from the Prostate Cancer Foundation and the ASCO Cancer Foundation. Author affiliation: Division ofHematology–Oncology, Massachusetts General Hospital Cancer Center, Massachusetts General Hospital, Boston, Massachusetts

ImpactFactor

Perioperative therapy improvesgastro-oesophageal cancer survival

� Tom Waddell and David Cunningham

A randomised phase III study has reported significant improvements in R0 resection rate and

overall survival associated with perioperative cisplatin and 5-fluorouracil treatment compared

with surgery alone in patients with gastro-oesophageal adenocarcinoma. These data support the

results of the randomised phase III MAGIC study that reported a 13% five-year survival

benefit from perioperative chemotherapy.

In 2008, the estimated worldwideincidence of gastro-oesophagealcancerwas 1.47million, accounting

for 11.6% of all cancer diagnoses. Per-haps more importantly, the estimatednumber of deaths in that year attribut-able to gastro-oesophageal cancer wasmore than 1.1 million, making this thesecond most common cause of cancerdeath after lung cancer.1 Due to theaggressive nature of these cancers,mostof the patients have inoperable diseaseat presentation. However, even in thecontext of locally-confined disease, the

five-year survival rates with surgeryalone are less than 25%.2,3 A studyrecently published by Ychou et al.2

supports the use of perioperativechemotherapy as a combined modalitytherapy in this disease setting with a14% improvement in five-year overallsurvival compared with surgery alone.

This approachhas beenwidely prac-ticed throughout Europe since the pub-lication of the results of the MedicalResearch Council Adjuvant GastricInfusional Chemotherapy (MAGIC)trial in 2006.3 The MAGIC study

reported a clinically and statistically sig-nificant improvement in progression-free survival (PFS) and overall survivalwith the addition of perioperative ECF(epirubicin, cisplatin and 5-fluorouracil(5-FU)) to surgery. As a result of thesedata, perioperative chemotherapybecame the standard of care in mostEuropean and Australasian countries.By contrast, patients in North Americaare routinely treatedwith primary resec-tion followed by post-operative 5-FU-basedchemoradiotherapy.As reportedbyMacDonald et al.,4 this approach also


This articlewas first published inNature Reviews Clinical Oncology as an advanceonlinepublicationon7 June2011andis publishedwithpermission.©2011NaturePublishingGroup. doi:10.1038/nrclinonc.2011.87,www.nature.com/nrclinonc

ImpactFactor

improves overall survival comparedwithsurgery alone, and a recent update withlong-term follow-up confirmed ongoingbenefit for survival (HR=1.32,P=0.04).4

Nevertheless, the two approaches havenever been compared in a randomisedtrial, and cross-trial comparison cannotestablish superiority of either strategydue to important differences in the pop-ulations. In particular, whereas theMAGIC trial3 required patients to havepotentially resectable tumours, the Inter-group 0116 trial4 required a completedcurative resection for eligibility.

Another geographical variation in thetreatment of gastro-oesophageal adeno-carcinoma exists in Japan, where theprevalence and natural history of thesetumours differ greatly from those inWestern populations. S-1, an oral fluoro-pyrimidine, has become the standard ofcare in the adjuvant setting in Japan fol-lowing the results of a Japanese studythat demonstrated a significant improve-ment in three-year overall survival.5

However, this approach has not beenadopted as a standard approach outsideof Japan as S-1 is not licensed in many

countries and globally capecitabine is themost established oral fluoropyrimidine.A recent meta-analysis of studies ongastric cancer, with individual data from3838 patients treatedwithinEuropean,American andAsian trials, supports theuse of adjuvant chemotherapy with a5.7% overall improvement in five-yearsurvival.6 Notably, this overall survivalbenefit with adjuvant chemotherapyseems to be less than the 13–14%benefit reported with the use of a peri-operative approach.2,3

Both theMAGIC trial3 and the studyby Ychou et al.2 included patients withadenocarcinoma of the lower oesopha-gus, gastro-oesophageal junction andstomach, although the recruited patientpopulations in the two studies differgreatly in terms of the distribution oftumour sites. Selected demographicsand the results of these two studies areshown in the table.Of particular note, inthe MAGIC trial, 42% of the patientscompleted six cycles of perioperativechemotherapy whereas only 23% of thepatients completed chemotherapy inthe trial byYchou et al.Although preop-

erative chemotherapy was completedsuccessfully in most of the patients,postoperative treatment delivery waslimited by factors that included diseaseprogression, postoperative complica-tions, and treatment toxicity. As onco-logical and surgical expertise continue toimprovewith theperioperative approach,successful completion of therapy shouldbecome possible in most of the cases.

The 36% five-year survival ratereportedwithECF in theMAGICstudyis similar to the 38% rate reported in thestudy byYchou et al. that usedCF (cis-platin and 5-FU). This inevitably raisesthe question as towhether epirubicin isnecessary in this setting. To address thisquestion, differences between the pop-ulations in the two studies must beexamined. The study byYchou and col-laborators did not formally stage patientsat trial entry and included a predomi-nance of tumours of the gastro-oesophageal junction,2 whichmay haveimproved outcomes, as a subgroupanalysis of theMAGIC trial data demon-strated that junctional tumours seemedto benefit most from perioperative


CROSS-TRIAL COMPARISON BETWEEN MAGIC AND FFCD/FNCLCC TRIALS2,3

Trial (n) Site of Tumour Chemotherapy Tolerance to Curative resection OStumour stage to regimen chemotherapy (% patients) benefit(% patients) be eligible (% patients Peri-op Surgery (HR)

completing) chemo alone

MAGIC3 LO 14.5 > Stage II Epirubicin=50 mg/m2 Preoperative: 67.6 65.6 13%(503) GEJ 11.5 Resectable Cisplatin=60 mg/m2 86% (3 cycles) (0.74)

Stomach 5-FU=200 mg/m2/day Perioperative:74.0 for 21 days 42% completed

6 cycles

FFCD2 LO 11.2 Not staged Cisplatin=100 mg/m2 Preoperative: 84.1 73.0 14%(224) GEJ 64.3 Suitable 5-FU=800 mg/m2/day 87% (2 cycles) (0.69)

Stomach for curative for 5 days Perioperative:24.5 resection 23% completed

6 cycles

GEJ – gastro-oesophageal junction, LO – lower oesophagus, OS – overall survival

ImpactFactor

chemotherapy.3 Furthermore, a meta-analysis from the Cochrane Collabora-tion carried out in 2010 explored thebenefits associated with variouschemotherapy regimens in metastaticgastric cancer, based upon data frommore than 5000 patients in 35 trials. Insubgroup analyses, the researchersdemonstrated significant overall survivalbenefits associated with regimens thatincluded 5-FU, an anthracycline andcisplatin compared with regimens thatlacked either the anthracycline or cis-platin.7 These data support the use ofregimens such as ECF in the advanceddisease setting, which can be reasonablyexpected to confer benefit to patientswith localised disease.

To improve tolerability and patientconvenience, capecitabine is increas-ingly substituting infused 5-FU in regi-mens such asECX (epirubicin, cisplatinand capecitabine),whichwas confirmedtobenon-inferior toECF in themetasta-tic setting.8 The ECX regimen is cur-rently under evaluation as preoperativetherapy in the OEO5 study comparedwith the standard CF regimen for theneoadjuvant therapy of oesophageal can-cer (UKCRN trial no. 854).

As therapeutic options for the treat-ment of gastro-oesophageal cancersexpand, clinical trials of multimodalitytherapy for early-stage disease mustincorporate targeted therapies to evalu-ate the outcome of their addition toperioperative, adjuvant chemotherapyor chemoradiation. In the advanced-disease setting the international ran-domised phase III ToGA study (n=594)recently demonstrated a significantimprovement in response rate, PFS andoverall survival when the anti-HER2monoclonal antibody, trastuzumab,wasadded to a cisplatin–5-FU doublet inpatients with HER2 positive adenocar-

cinomas of the stomach or gastro-oesophageal junction.9 Trastuzumab iscurrently being evaluated in combinationwith capecitabine and oxaliplatinchemotherapy in a Spanish phase IIperioperative gastric cancer study (Clin-icalTrials.gov identifier:NCT01130337),and following neoadjuvant chemoradia-tion and surgery for oesophageal andoesophagogastric junction cancers intheRTOG-1010 trial (ClinicalTrials.govidentifier NCT01196390). The small-molecule inhibitor ofHER2andEGFR,lapatinib, and twomonoclonal antibod-ies targetingEGFR, cetuximab andpan-itumumab, are currently undergoingevaluation in the first-line advanced-disease setting and, if successful, willlikely be evaluated in patientswith early-stage disease.

Even though the results of these fur-ther studies are awaited, the study byYchou and collaborators strengthens thecurrent evidence for perioperativechemotherapy with improvements inR0 resection rate, disease-free survivalandoverall survival in gastro-oesophagealcancer. This study confirms the benefitsof this approach and further validatesperioperative chemotherapy as a stan-dard treatment option in this diseasesetting. We expect that the addition ofmolecularly targeted therapies mayfurther improve patient outcomes inthe future.

References

1. J Ferlay et al. (2010) Estimates of worldwide

burden of cancer in 2008: GLOBOCAN 2008.

Int J Cancer 127:2893–2917

2. M Ychou et al. (2011) Perioperative

chemotherapy compared with surgery alone for

resectable gastroesophageal adenocarcinoma: a

FNCLCC and FFCD multicenter phase III trial.

JCO 29:1715–21

3. D Cunningham et al. (2006) Perioperative

chemotherapy versus surgery alone for resectable

gastroesophageal cancer. NEJM 355:11–20

4. J MacDonald et al. (2009) Chemoradiation of

resected gastric cancer: a 10-year follow-up of the

phase III trial INT0116 (SWOG 9008). ASCO

Annual Meeting; 2009 [abstract]. JCO 27:4515

5. S Sakuramoto et al. (2007) Adjuvant

chemotherapy for gastric cancer with S-1, an oral

fluoropyrimidine. NEJM 357:1810–20

6. X Paoletti et al. (2010) Benefit of adjuvant

chemotherapy for resectable gastric cancer: a

meta-analysis. JAMA 303:1729–37

7. AD Wagner et al. (2010) Chemotherapy for

advanced gastric cancer. Cochrane database syst.

rev. issue 3. art. no.: CD004064. doi:

10.1002/14651858.CD004064.pub3

8. D Cunningham et al. (2008) Capecitabine and

oxaliplatin for advanced esophagogastric cancer.

NEJM 358:36–46

9. YJ Bang et al. (2010) Trastuzumab in

combination with chemotherapy versus

chemotherapy alone for treatment of HER2-

positive advanced gastric or gastro-oesophageal

junction cancer (ToGA): a phase 3, open-label,

randomised controlled trial. Lancet 376:687–697


Practice points� Two large randomised phase III

trials have now confirmed a 13–14% absolute improvement inoverall survival associated withperioperative chemotherapy com-paredwith surgery alone in gastro-oesophageal adenocarcinoma

� Significant geographical variationin practice will continue in theabsence of a head-to-head trialto confirm superiority of oneapproach over another

� Addition of novel targeted agentsto perioperative chemotherapy islikely to further improve patientoutcomes in the future

Author affiliations: Department of Medicine, Royal Marsden Hospital, Sutton UK (Tom Waddelland David Cunningham).Competing interests statement: David Cunningham declares associations with the following companies: Amgen, Merck Serono, Roche. Tom Waddell declares no competing interests.

NewsRound


N E W S R O U N DSelec ted repo r t s ed i t ed by Jane t F r i cke r

Cancer familyhistories foundto be inaccurate� JNCI

General population reports on their familyhistory for lung, colorectal, breast and

prostate cancer were not found to be highlyaccurate, the 2001 Connecticut Family Healthstudy has reported. Accuracy was greater forreports on first-degree relatives (FDR) thansecond-degree relatives (SDR).

It is well recognised that knowledge of apatient’s family cancer history is essential forestimating their individual cancer risk andmak-ing clinical recommendations regarding screen-ing and referral to cancer genetics clinics. It hasnot been clear, however,whether reported fam-ily cancer history is sufficiently accurate forthis purpose.

In the current study, Phuong Mai and col-leagues, from the National Cancer Institute(Bethesda, Maryland), undertook a random-digit dial survey involving 1019 participants. Inthe telephone interviews respondents wereasked to list all biological FDRs (parents, siblingsand children) and SDRs (grandparents, uncles,aunts, nieces, and nephews) who had sufferedfrom cancer. Altogether the participants

tory of lung, colorectal, breast, and prostatecancers in this population-based survey werelow tomoderate, especially amongSDR, but thespecificity and NPV (negative predictive value)were high,”write the authors, adding that thereis a need to promote family history awarenessand to find better tools to capture it accuratelyto ensure that appropriate risk assessment andclinical care recommendations can be made.“Improved knowledge about cancer mightencourage people to be more willing to com-municate about it with others, either whensharing information about their owndiagnosesorwhen asking for information about their rel-atives’ diagnoses,” they write.

In an accompanying commentary, RachelFreedman and JudyGarber, from theDana Far-ber Cancer Institute (Boston, Massachusetts),said that although for the foreseeable futuredetailed family cancer historieswill continue toprovide the basis for identification of suscepti-bility genes, ultimately genomic analyses willbecome a routine part of cancer with predis-positions to cancer identified at a young age.

� PL Mai, AO Garceau, BI Graubard, et al.

Confirmation of family cancer history reported in

a population based survey. JNCI 18 May 2011,

103:788–797

� R Feedman, J Garber. Family cancer history:

healthy scepticism required. ibid pp 776–777

reported 20,578 FDRs and SDRs, ofwhich 2605were sampled for confirmationof cancer reportson breast, colorectal, prostate and lung cancerusing state cancer registries, Medicare data-bases, the National Death Index, death certifi-cates andheathcare facility records. The state ofConnecticutwas chosen for the study becauseit has the oldest population-based cancer reg-istry in the US, with records dating back to1935, thereby facilitating the process of con-firming cancer reports.

Results showed that, for lung cancer, thesensitivity value was 60.2% and positive pre-dictive valuewas 40%; that for colorectal can-cer the sensitivity value was 27.3% and thepositive predictive value was 53.5%; for breastcancer the sensitivity valuewas 61.1%and thepositive predictive value was 61.3% and forprostate cancer the sensitivity valuewas 32.0%and the positive predictive value was 53.4%.Overall, cancer history reports on FDR weremore accurate than reports onSDR. For prostatecancer, FDRhad 58.9% sensitivity versus 21.5%for SDR (P=0.002); for lung cancer FDR had78.1% sensitivity versus 31.7% for SDR(P<0.001); for colorectal cancer FDRhad85.8%sensitivity versus 43.5% for SDR (P=0.004); andfor breast cancer FDR had 79.9% sensitivityversus 53.6% for SDR (P=0.02).

“In summary, the sensitivity and PPV (pos-itive predictive value) of a reported family his-

NewsRound


FOLFIRINOX improvessurvival in metastaticpancreatic cancer� NEJM

Incomparisonwith single-agent gemcitabine,FOLFIRINOX was associated with a survival

advantage as a first-line treatment in patientswithmetastatic pancreatic cancer, a phase II-IIItrial has found. However, the FOLFIRINOX reg-imen (oxaliplatin, irinotecan, fluorouracil andleucovorin) showed increased toxicity.

In 2010pancreatic adenocarcinomawas thefourth leading causeof death fromcancer in theUS,with five-year survival rates of 6% in Europeand the US. Since a randomised trial showedsignificant improvements in median overallsurvival for gemcitabine comparedwith fluoro-uracil, gemcitabinehas been the reference treat-ment regimen. However the combination ofgemcitabine with a variety of cytotoxic andtargeted agents has generally shownno signif-icant survival advantage as compared withgemcitabine alone. Data are lacking on the effi-cacy and safety of the combination chemo-therapy regimen FOLFIRINOX compared withgemcitabine as a first-line therapy inmetasta-tic pancreatic cancer.

Between December 2005 and October2009, Thierry Conroy from Nancy University(France) and colleagues from48 French centresrandomised 342 patients in a ratio of 1:1 toreceive FOLFIRINOX (n=171) or gemcitabine(n=171). Inclusion criteria included an EasternCooperativeOncologyGroup performance sta-tus score of 0–1 (on a scale of 0–5, with higherscores indicating greater severity of illness). Sixmonths of chemotherapywas recommended inboth groups for patients who had a response.

Results show that the overall survival at amedian duration of follow-up of 26.6 monthswas 11.1months in the FOLFIRINOXgroup com-pared with 6.8 months in the gemcitabinegroup (HR=0.57, 95%CI 0.45–0.73; P<0.001).The median progression-free survival was6.4months in the FOLFIRINOXgroup comparedwith 3.3 months in the gemcitabine group

(HR=0.47, 95%CI 0.37–0.59; P<0.001). Thesafety profile of FOLFIRINOXwas less favourablethan that for gemcitabine, with FOLFIRINOXassociatedwith a higher incidence of grade 3or4 neutropenia, febrile neutropenia, thrombo-cytopenia, diarrhoea, and sensory neuropathy,as well as grade 2 alopecia. Despite this higherincidence of adverse events, however, a signif-icant increase in the time to definitive deterio-ration in the quality of life was observed in theFOLFIRINOX group. At six months, 31% of thepatients in the FOLFIRINOX group had a defin-itive degradation of the quality of life versus66% in the gemcitabine group (HR=0.47,95%CI, 0.30–0.70; P<0.001).

“Our findings suggest that FOLFIRINOX is afirst-line option for patients with metastaticpancreatic cancer who are younger than 76years andwhohave a goodperformance status(ECOG0 or 1), no cardiac ischemia, and normalor nearly normal bilirubin levels,” write theauthors. The success of the trial over previousstudies, they suggest, may be due to the factthat the selection criteria were more rigorous.

� T Conroy, F Desseigne, M Ychou, et al.

FOLFIRINOX versus gemcitabine for metastatic

pancreatic cancer. NEJM 12 May 2011,

364:1817–25

15 years follow-up:radical prostatectomybeats watchful waiting� NEJM

After 15 years of follow-up, radical prosta-tectomy continues to be associated with

a reduction in the rate of death from prostatecancer as compared to watchful waiting, thelatest results of the Scandinavian ProstateCancer Group 4 (SPCG-4) study have shown.Subgroup analyses also showed that the riskof death from prostate cancer after radicalprostatectomy was increased by a factor ofseven for men who had tumours with extra-capsular growth, and that the benefits of

prostatectomy were confined to men under65 years old.

In 2008 the SPCG-4 study group reportedthat radical prostatectomy, as compared withwatchful waiting, reduced the rate of deathfrom prostate cancer. The current study, whichpresents an additional three years of follow-up, represents theonly randomised investigationthus far todemonstrate that surgery reduces therisk of mortality from prostate cancer.

Between October 1989 and December1999, Anna Bill Axelson and colleagues, fromUppsala University Hospital (Uppsala, Sweden),randomly assigned 695 men with newly diag-nosed localised prostate cancer to radicalprostatectomy (n=347) or watchful waiting(n=348).

Results at a median follow-up of 12.8years show that 166 of the 347 men in theradical prostatectomy group and 201 of the348 in the watchful waiting group died(P=0.007). Death was due to prostate cancerin the case of 55 of the men assigned to sur-gery and 81 assigned to watchful waiting. At15 years the cumulative incidence of deathfrom prostate cancer was 14.6% for the sur-gical groups versus 20.7% for the watchfulwaiting group (P=0.01).

Subgroup analysis showed that the sur-vival benefit was limited to men under 65years of age, and that the risk of death fromprostate cancer for men with extra-capsulartumour growthwhounderwent radical prosta-tectomywas seven times that formenwithoutextra-capsular tumour growth undergoingthe same procedure.

Gleason scores were also highly predictiveof the risk of death fromprostate cancer: among129menwhohad tumourswithGleason scoresof 2 to 6, only 5 died from prostate cancer. Thecurrent analysis showed that the numberneeded to treatwith surgery to avert one deathwas 15overall and seven formen younger than65 years of age. This compared with 19 in theearlier analysis.

“Although extra capsular growth is not aperfect predictor of lethal disease, our findingsindicate that these men could be a group forwhich adjuvant local or systemic therapy

NewsRound


would be beneficial,” write the authors, addingthat continued follow-up data from theSPCG-4 study might allow them to identifyprognostic markers inmen assigned towatch-ful waiting that can serve as trigger points foractive treatment.

In an accompanying commentary,Matthew Smith from Massachusetts GeneralHospital Cancer Center (Boston, Massachu-setts) writes, “The survival benefit with prosta-tectomy in men with low-risk disease is themost important new finding of the SPCG-4.”He adds, however, that the findings may notbe relevant for men today who have early-stage low-risk prostate cancers identified byprostate-specific antigen screening.

� A Bill-Axelson, L Holmberg, M Ruutu et al.

Radical prostatectomy versus watchful waiting in

early prostate cancer. NEJM 5 May 2011,

364:1708–17

� MR Smith. Effective treatment for early-stage

prostate cancer – possible, necessary or both? ibid

pp 1770–72

Sunitinib improvesPFS in pancreaticneuroendocrine tumours� NEJM

Continuous daily administration of suni-tinib doubled progression-free survival

(PFS) in patients with advanced pancreaticneuroendocrine tumours (NETs) in comparisonto patients receiving placebo, a phase III trialhas concluded. The trial also showed objectiveresponse rates and overall survival data thatconsistently favoured the sunitinib arm. Thetrial was terminated early by the data moni-toring committee due to the risk of seriousadverse events, disease progression and deathamong patients receiving placebo.

The incidence of pancreatic NETs, whicharise from endocrine cells in the pancreas, isincreasing, but five-year survival rates arestill below 43%. Surgery has been the main-

VEGFRs and PDGFR,” write the authors, addingthat the improvements were achievedwithoutadversely affecting quality of life.

Although early termination of clinicaltrials may result in overestimation of treat-ment effects, the magnitude of the observedtreatment effect, the consistency of the haz-ard ratio for disease progression or death andthe favourable survival data provide strongevidence of a clinically meaningful benefitfor sunitinib, say the authors.

� E Raymond, L Dahan and JL Raoul. Sunitinib

malate for the treatment of pancreatic neuro-

endocrine tumours. NEJM 10 February 2011,

364:501–513

Opportunities found forimproving colorectal cancerpatients’ quality of life� British Journal of Cancer

Most factors that adversely affect qualityof life in patients with colorectal cancer

(CRC) can be modified, a UK study hassuggested.

Advances in treatment for CRC, whichrepresents the third most common cancer inwestern countries, is resulting in more peoplebeing cured and also surviving longer with thedisease. The adverse effects of both the diseaseand treatment can be longterm and includelack of energy, bowel problems, poor bodyimage and emotional problems, as well asdifficulties with sleep, fear of recurrence, anx-iety, depression, sensory neuropathy, gas-trointestinal problems, urinary incontinenceand sexual dysfunction. As many people withCRC are more elderly, they often have addi-tional functional limitations and comorbiditiessuch as geriatric syndromes, heart disease,chronic obstructive pulmonary disease orother cancers.

To tackle the challenge of improving qual-ity of life in patients with CRC, Nicola Gray andcolleagues from the Centre of Academic

stay of treatment, with somatostatin ana-logues used to relieve symptoms resultingfrom hormone hypersecretion. The onlyapproved chemotherapeutic agents remainstreptozocin alone or in combination withdoxorubicin. In both preclinical models andphase I and II trials the multitargeted tyrosinekinase inhibitor sunitinib, whichwas rationallydesigned to inhibit VEGFR and PDGFR, showedactivity against pancreatic NET tumours.

In the current study, conducted betweenJune 2007 and April 2009, Eric Raymond fromHôpital Beaujon, (Clichy, France) and col-leagues from 42 centres in 11 different coun-tries randomly assigned 171 patients in a 1:1ratio, to receive best supportive care witheither sunitinib at a dose of 37.5 mg per day(n=86) or placebo (n=85). Eligible patientshad pathologically confirmed, well-differen-tiated pancreatic endocrine tumours thatwere advanced, metastatic, or both, and werenot candidates for surgery.

In February 2009 the data and safetymonitoring committee recommended dis-continuation of the trial because of thegreater number of deaths and serious adverseevents in the placebo group, and differencesin PFS. Results show that median PFS was 11.4months in the sunitinib group versus 5.5months in the placebo group (HR for pro-gression or death = 0.42, 95%CI 0.26–0.66;P<0.001). The objective response rate was9.3% in the sunitinib group versus 0% in theplacebo group. At the data cut-off point,nine deaths (10%) were reported in the suni-tinib group versus 21 deaths (25%) in theplacebo group (HR=0.41, 95%CI 0.19–0.89;P=0.02). The most frequent adverse events inthe sunitinib group were diarrhoea, nausea,vomiting, asthenia, and fatigue. Most suni-tinib-related adverse events, report theauthors, were manageable through doseinterruption or modification.

“The improvement in progression-freesurvival observed among patients whoreceived sunitinib provides support for previ-ous preclinical and clinical data suggestingthat neuroendocrine tumors may be particu-larly sensitive to combined inhibition of

NewsRound


Primary Care at the University of Aberdeen(Scotland), set out to identify the potentiallymodifiable and fixed factors most associatedwith better or worse quality of life.

In the study 496 people diagnosed withCRC completed the EORTC-QLQ-C30 quality oflife questionnaire, which comprises five func-tional scales (physical, role, cognitive, emo-tional and social) and three symptom scales(fatigue, pain, nausea and vomiting). Addi-tional symptoms commonly reported by peo-ple with cancer (dyspnoea, loss of appetite,insomnia, constipation and diarrhoea) andthe perceived financial impact of the diseasewere also assessed. The mean age of partici-pants in the study was 66 years, with 70%being over the age of 60.

Results show that, of theunmodifiable fac-tors, female sex (P<0.001), more self-reportedcomorbidities (P=0.006) andhavingmetastasesat diagnosis (P=0.036) significantly predictedpoorer quality of life, but explained little of thevariability of the model, giving a correlationcoefficient of R2=0.064 (1 = perfect correlationand 0 = no correlation). However, when themodifiable risk factors poorer role (P<0.001)and poorer social functioning (P=0.003),together with fatigue (P=0.001), dyspnoea(P=0.001), anorexia (P<0.001), depression(P<0.001) and worse perceived consequences(P=0.013), were introduced, the model fitimproved considerably (R2=0.574).

“We found that physical, psychologicaland social factors were all significantly andindependently associated with overall QoL.Most predictors were modifiable, with symp-toms, depression and limitations to usualactivities being most important,” write theauthors, adding that the influence of unmod-ifiable factors was small, with the remainingindependent predictors offering the potentialfor intervention.

Fatigue, for example, has been shown torespond in a variety of diseases to pro-grammes of graded activity, say the authors,while depression and anxiety have improvedwith nurse led interventions, exercise andantidepressants, and difficulties with travellingmay be helped by interventions to reduce the

need for it, such as providing more locallybased follow-up.

“If we wish to improve QoL in peoplewith colorectal cancer, then we need first toidentify those most at risk, and second tointervene to address factors which are mod-ifiable,” conclude the authors, adding thatany future interventions will require rigorousevaluation.

� NM Gray, SJ Hall and S Browne. Modifiable

and fixed factors predicting quality of life in people

with colorectal cancer. Br J Cancer 24 May 2011,

104:1697–1703

Short-term radiotherapyadded to surgerydelivers long-termgains in rectal cancer� Lancet Oncology

For patients with resectable rectal cancer,preoperative short-term radiotherapy

reduced local recurrence by more than 50%compared to surgery alone, reports the DutchColorectal Cancer Group. In the long termfollow-up of the total mesorectal excision(TME) trial, investigators found that a reduc-tion in local recurrence was maintained, andthat overall survival was improved in a subsetof patients.

The TME trial, undertaken by Cornelis vande Velde and colleagues at Leiden UniversityMedical Centre (Leiden, Netherlands), was thefirst study to suggest that, in combinationwith TME, short-term preoperative radiother-apy delivered added value. Results at two yearsshowed a decreased risk of local recurrence forirradiated patients (2% vs 8%, P<0.001); whileresults at six years again showed a decreasedrisk of local recurrence for irradiated patients(6% vs 11 %, P<0.0001). In both cases no dif-ferencewas found in overall survival. However,the possibility that radiotherapy might notprevent, butmerely postpone, local recurrencecould not be excluded. The current publication

reports on the long-term results after amedianof 11.6 years follow-up.

In the TME trial between January 1996and December 1999, 1861 patients withresectable rectal cancer without evidence ofdistant disease were randomly assigned, in a1:1 ratio, to TME preceded by 5x5 Gy radio-therapy (n=897) or TME alone (n=908). Thepatients were recruited from 118 Europeancentres and one Canadian centre.

Results show that the 10-year cumulativeincidence of local recurrence was 5% in thegroup assigned to the short-course preoper-ative radiotherapy plus TME versus 11% for theTME alone group (P<0.0001). Again overallsurvival did not differ between the two groups.However, in the subset of patients with TMEstage III and negative circumferential margins,survival was 40% in patients receiving justTME versus 50% in patients receiving TMEplus radiotherapy (P=0.031).

In an accompanying editorial, RobGlynne-Jones, from Mount Vernon CancerCentre (Northwood, London, UK), commentsthat the fact that the results do not show adifference in survival implies that some sub-groups may be being disadvantaged by radio-therapy in terms of survival. Indeed, he adds,the results showed that death from a secondmalignancy was more frequent in the radio-therapy group than in the TME-alone group(14% vs 9%).

“Preoperative short-term radiotherapysignificantly improved 10-year survival inpatients with a negative circumferential mar-gin and TNM stage III,” conclude the authors,adding that future staging techniques shouldoffer possibilities to select patient groups forwhich the balance between benefits and side-effects will result in sufficiently large gains.

� W van Gijn, C Marignen, I Nagtegaal et al.

Preoperative radiotherapy combined with total

mesorectal excision for resectable rectal cancer:

12-year follow-up of the multicentre, randomised

controlled TME trial. Lancet Oncol June 2011,

12:575–582

� R Glynne-Jones. …and a two-edged sword in

their hands. ibid pp 519–520

Thirty years on, ESTRO remains focusedon its vision for a cure

� Marc Beishon

ESTROwas set up in 1981 to ensure that radiation oncology could play its full part alongside

other disciplines in the developing field of cancer care. This May, thousands of members

celebrated its achievements at an anniversary conference, and then turned their focus onto

the many challenges ahead.

The European Society forRadiotherapy andOncology(ESTRO) celebrated its30th anniversary inLondonthis year at a special con-

ference that brought together its mainregular events as well as introducing anew one, the ESTRO InternationalOncology Forum, which included 'theopportunity to review 30 years of radia-tion oncology in 30 hours.’“You don’t usually have an anniver-

sary clinical track at conferences – itwas a unique chance to review theevidence-basedmedicine generated overthe last 30 years and how it prepares usfor the future,” says Jean Bourhis,ESTRO’s current president, andhead ofthe radiation oncology department atthe Institute Gustave Roussy in Paris.Although the history of radiation

oncology goes back much further, the

society can certainly be proud of itsachievements, which are chronicled in abook prepared for the occasion, ‘Threedecades of ESTRO: a vision for cancercure’. It details the desire of ESTRO’sfounders, who hadmostly trained in theUS, to establish a more united front inEurope after they had witnessed divi-sions in specialties in America, andindeed therewas an idea for a ‘EuropeanSociety ofOncology’, integrating all dis-ciplines. But the feeling prevailed thatradiotherapy needed its own distinctvoice, asmedical oncologists began theirrise, but they would also include allieddisciplines such as radiobiology andmedical physics.There was the challenge too of dis-

tinguishing radiation oncology from themuch larger community of radiology at atime when some saw radiotherapy as adying specialty, as the era of chemother-

apy started to take hold.After holding itssecond meeting at the European Asso-ciation of Radiology’s conference in1983, ESTRO went its own way in thefollowing years.Special mention is made of

Emmanuel van der Schueren, widelyrecognised as a brilliant clinical scientistand the driving force behind ESTRO,first as executive secretary and thenpresident, and also editor of the society’sjournal, Radiotherapy and Oncology. Aprofessor of radiotherapy atLeuven,Bel-gium, hedied far too soon fromcancer in1998, but has left a lasting impact onEuropean oncology.Van der Schueren’s fellow ESTRO

founders must be name-checked too,as they are also pivotal figures – KlaasBreur, JerzyEinhorn,Michael Peckhamand Maurice Tubiana – who togetherset in train an organisation that has built


Spotlighton...

up amembership of close to 5000.Asbefits a field that covers awide ter-

ritory, scientifically and clinically, thereare now several regularmeetings in addi-tion to the main ESTRO conference,which is annual but takes place atECCOevery other year. There are bian-nual meetings for physics and for theGECgroup (GroupeEuropéendeCurie-thérapie – a clinical trials meeting thatmerged with ESTRO in 1990).More recently, ESTRO has started

running PREVENT, on the side-effectsof radiotherapy, and an event on noveltargeted drugs and radiotherapy. Andthere is CERRO, an experimental radi-ation oncology meeting, as well newevents on head and neck oncology andmolecular imaging.As Bourhis notes, meetings and the

journal contribute to communication,one of three pillars thatESTRO is organ-ising its work around. “We are growingour own meetings and also organisingmorewithmultidisciplinary partners,” hesays, noting thatworkingwith other spe-cialties is in his opinion still the mostpressing issue at clinical level inEurope.

PULLING UP STANDARDSThe second pillar, education, is partic-ularly important because of the varia-tions in quality in using increasinglycomplex equipment and regimens.Bourhis says ESTRO has been carefulnot to expand its courses too quickly soas not to compromise quality, but itnow runs more than 30 courses bothinside and outside Europe – twice asmany as in 2005. Significant numbersof attendees come from easternEurope, Africa and the Far East, hesays, with about 40% from westernEurope. It all adds up to some 3000attendees and 200 teachers a year.ESTRO’s core curriculum has alsorecently been revised, and it has nowbeen integrated into the national teach-ing programmes of countries such asthe Netherlands and Spain.E-learning is playing an increasing

part in ESTRO’s education plans, addsBourhis. The society has a platformcalled Eagle (ESTRO's application forglobal learning), for online tutoredcourses, which has so far been used forrectal cancer, with courses for breastand head and neck to come. Anotheroffering, called Falcon, is a combinedworkshop/online platform for anatomicdelineation and contouring, which aimsto address one of the biggest and mostunpredictable sources of error in radia-tion oncology. “At ESTRO 30 we alsolaunched the ESTRO Fellow, a posi-tion that people can attain after gaininga certain number of credits and passingan exam,” says Bourhis.The third pillar of ESTRO’s work is

policy, which Bourhis says has probablybeen themost challenging for all cancersocieties in Europe, but which hasreceived a big boost now that ECCO ismore effectively presenting a joint front.ESTROpast-presidentMichaelBau-

mann,whose contribution to the field ofradiationoncologywasprofiled inCancer

Spotlighton...


An inspired leader. Emmanuel van der Schuerenwas a moving force behind the foundingof ESTRO 30 years ago

Spreading best practice. Delegates had theopportunity to visit University College Hospital inLondon to see the latest in precision beamdelivery from a high-energy linac. The custom-made thermoplastic face mask helps ensure thepatient is precisely positioned for every session

FEWER'SVIEWPHOTOGRAPHY©ESTRO

ESTRO

World Jan–Feb 2006, isthe current president ofECCO. He and otherESTRO leaders havebeenuncompromisingin fighting ESTRO’scorner in Europeanoncopolitics, but ithas been tough togain the level ofrepresentation theyfeel the society needswithin ECCO, and also politicallyon the European stage.The level of investment needed by

healthcare systems in radiotherapy ismuch higher than most other medicaltreatments, and ESTRO is acting onceagain to raise awareness of the economicbenefits of its specialty, recognising thatawareness of radiotherapy remains pooramong not only the public and politi-cians, but also healthcare purchasers.Anewproject calledHERO (HealthEco-nomics inRadiationOncology), run by ataskforce fromEurope andCanada, hasstarted to look at the need for radiother-apy, its provision and accessibility, wait-ing lists, staffing, cost accounting andeconomic evaluation.Taskforce member Yolande Lievens,

fromLeuven,explains that theproject is anextendedupdateofa2005

study that wasfunded by theEuropeanUnioncalled QUARTS(Quantificationof RadiotherapyInfrastructure andStaffing). The radi-ationoncology com-munity has still“barely started” evalu-ating the economicaspects of its treat-ments, she says, whileother disciplines suchas medical oncologyhave been doing betterfor some time.

Aims of the HERO project includethe development of economicmodels tocalculate the value for money of treat-ments, and using data for lobbying andpolicy making. Preliminary data from aquestionnairewere presented atESTRO30, with final results expected in 2013.

A VERY WIDE AGENDAESTROhas one of thewidest agendas ofany cancer specialty, covering as it does,for example, core European curriculafor clinicians, physicists and radiotherapytechnicians. This, togetherwith the var-ious meetings and educational activi-ties, makes it quite a challenge just toorganise its own affairs, let alone tackle

True to its founding vision. The anniversary bookchronicles ESTRO’s achievements so far. Theanniversary conference focused on carrying thatvision forward into the next decade

the wider multidisciplinary integrationthat Bourhis says is vital.A long list of actions are in train to

address issues such as increasingmem-bership and attracting younger mem-bers – the societywould like to increaseits number to 8000 by 2015, whichwould still be only about half of its target‘market’. Like other European oncologyorganisations, however, it faces the trickyproblem of whether and how to extendterms tonational radiation oncology soci-eties. Representing more of the varioussub-specialties at board level is anotherissue, as is forging ethical collaborationwith industry – the society has recentlyexpanded its corporatemembership pro-gramme, and now offers a ‘gold level’whereby companies canparticipate in anadvisory corporate council for promotingresearch and education.ESTRO has certainly made a con-

certedeffort to increase itsprofileandpro-fessional presentation, with the 30thconference being a flagship, and hasengaged top-levelpublic relations support,produced the anniversary book, createdthenewfellowshipandstartedadditionalactivities. With a solid platform in con-ferences and education, the emphasis isnow on pushing forward its standardsand guidelines in the multidisciplinarycontext,whichwill needboth consistentengagement of activists at local level andhigher-level European networking.To help finance this ambitious pro-

gramme, theESTROCancerFoundationwas launched at the anniversary confer-ence,which shouldhelp ensureESTROcancontinue toplay adynamic and lead-ing role inpushing forward theboundariesof cancer care in decades to come.


Spotlighton...

ESTRO is acting once again to raise awareness

of the economic benefits of its specialty

A minimum acceptable standardof care for every patient

� Anna Wagstaff

A great deal is already known about why cancer patients do better in some countries or some

parts of a country than in others. But how can that knowledge be transmitted to the people

who have the power to act, in a way they can quickly and easily understand?

Maximising survival andminimising the side-effects, the long-termeffects and the impactof the disease and its

treatment on patients’ quality of life arewhat good cancer care is all about. Butwhen funds are limited, anddiagnostic ormanagement tools are expensive, distinc-tions must be made between what isessential and what is desirable, with aview to ensuring that every citizen diag-nosedwith cancer has access to the basicessentials of care. This is the philosophybehind one of the latest EUROCHIPprojects, which has been piloting a newapproach to closing the survival gapsbetween thebest and theworst inEurope.The idea is simple. First, select a dis-

ease setting. Priority should be given tothose with curative potential and thoseaffecting largepopulations– low-riskchild-hoodALL(acute lymphoblastic leukaemia)

andearly andadvancedbreast cancerwereselected for the pilot studies. Next, ask agroupof experts to agree on the gold-stan-dard evidence-basedprotocol for diagnos-ing and managing the disease. Then,separate out the ‘minimum requirementsfor acceptable treatment’ from the ‘addi-tional [desirable] tools’– things thatmight,for instance, offer an extra little bit of cer-tainty, ormake thepatient feel lessunwell.This list ofminimumrequirementswill

represent the basic standard of treatmentthat every cancerpatient inEurope shouldhave the right to expect. It can be scannedto identify which elements are affordableeven in the poorest areas in Europe, andwhichare sufficiently costly toposeaprob-lemwherehealthbudgets are very tight.Bynarrowing the focus onto diagnostic andmanagement tools that are both essentialand potentially unaffordable, it becomespossible to concentrate efforts onproblemareas and look for alternative options that

aremoreaffordablebut equally effective (ifperhaps less desirable), or explore cheaperways of getting access to essential tools –greater sharingof expensivediagnostic test-ing facilities, for example.This pilot project, one of many initia-

tivesof theEuropeanCancerHealth Indi-cator Project (EUROCHIP) programmerun fromthe IstitutoNazionaledeiTumoriinMilan, focuses on reducing inequalitiesincancer incidenceandcancercareacrossEurope. (EUROCHIP’swork on improv-ing cervical cancer screening in six coun-tries was profiled in the May–June 2011issue ofCancer World.) The pilot tacklesan aspect of cancer control that almost allofEurope’s richer countries arenowstrug-glingwith, and that is evenmore essentialfor the poorer ones: cost–effectiveness,how to do the best for cancer patientswith themoney available.Where this approachdiffers from that

of existing bodies set up to performhealth


Systems&Services

technology assessment (HTA) and value-for-money analyses, such as the UK’sNICE (National Institute for Health andClinicalExcellence) andSweden’sDentaland Pharmaceutical Benefits Agency, isthat rather than takingcosts andbenefits asits starting point, it tries to identify wherecost constraints couldbe a significant fac-tor explaining why patients do so muchworse in somecountries or regions than inothers.And it tries to suggest solutions.

GOAL ORIENTEDFor Andrea Micheli, the EUROCHIPleader, thisproject is all about resultson theground, and thatmeans it isheavily gearedtoward Europe’s political leaders. “Theproblem is that individual specialists know


what has to be done, but this is not infor-mation known to politicians. WhatEUROCHIPdoes is to extract fromthesepeople some key proposals and to passthem on to the politicians. We need tosend simple messages to the EuropeanCommission: in this way we can quicklyimprove the situation inpoorer countries.”ChildhoodALLwasanobviouschoice

for a pilot study. Thanks to decades ofcooperative clinical studies by paediatriconcologists in theUSandEurope (notablyGermany), childhood ALL is now cur-

able inaround80%ofcases.Yetmanychil-drencontinue todieunnecessarily in somecountries and regions ofEurope.Further-more, thehigh level of collaboration in thisarea means that, while many questionsremain to be answered, there is a strongconsensus over the current gold-standardprotocol formanaging the disease.The rationale behind choosing breast

cancer, both early and advanced, for theother pilot studies was that it is the mostcommoncanceramongwomen inEurope,and the number of new cases is growing.

Systems&Services

COMITATOMARIALETIZIAVERGA/ATTILIOROSSETTI

We can do this. Not all ALL patients can receive the same level of care as this young girl at the SanGerardo Hospital in Monza, Italy, but Momcilo Jankovic, pictured here, hopes that simple messagesdesigned to address the main factors behind variations in survival will ensure greater access tominimum acceptable standards of care across Europe

“In this case, newdrugs areproposed, anddifferences insurvivalmaybe related to theavailability of these drugs or to otherthings,” saysMicheli.

A CHANCE TO HELPMomcilo Jankovic, a paediatric oncologistat San Gerardo Hospital in Monza, Italy,was delighted to be asked to participate intheALL pilot, alongside Kathy Pritchard-Jones, professor of paediatric oncology atUniversity College London and NickGoulden, consultant haematologist atGreatOrmondStreetChildren’sHospital,London. “When they askedme to cooper-ate todefinewhat the childneeds in orderto be treated according to local resourcesand to reduce the cost of this treatment, Ithought, this is not widely reported in theliterature, so it seemed to be a very goodopportunity to helpwithmy experience.”Jankovic’s experience in this field is

considerable. Not only does he have along track record treating young ALLpatients– includingcollaborating in inter-national clinical studies led by the ItalianAssociation of Paediatric Haematologyand Oncology (AIEOP) and the Berlin-Frankfurt-Munich (BFM) ALL group –buthealsohasexperiencehelping improveresults in countries where costs pose areal problem, including a collaborationwith Nicaraguan paediatric oncologists,which raised survival rates for childhoodALL from10% to an impressive 50%.Closer to home, Jankovic and his col-

leagues also built up a long-term collabo-rationwithdoctors inSerbia after years ofisolation during the Balkan war left themtrailing behind much of Europe. Hementions, as one important outcome, theinterest and support these paediatric

oncologists received from the politiciansonce they saw what was being achieved.“The government looked at the resultsthey obtained and wanted to promote anationalnetwork.Theynowpay for equip-ment, and they pay for doctors to attendmeetings or to visit outside the country.They are much more positive aboutresponding to the request of doctors.”Jankovic hopes that theEUROCHIP

projectwill help achieve similar improve-ments across Europe.

A BRIDGE TO POLITICIANSMicheli describes theproject as essentiallyan intellectual exercise, “using amethod-ologyderived fromour experienceover 10years” to facilitate discussion betweenexperts coming from different fields, toextract key messages, check whetherthese are widely accepted by others, andthenpass themon to the peoplewho candeliver change. “We are trying to build abridge and to find a common languagewith the politicians.”In the ALL pilot, as well as Jankovic

and his fellow physicians, the group ofexperts included researchers, epidemiol-ogists, healtheconomists andhealth tech-nology analysts.The final report has yet tobe written and validated among a widergroup of experts, but findings so far indi-cate that the cost of providing therapies isless of an issue than the cost of tests thatcan guide physicians in tailoring treat-ments. The only exception to thismay bePEG-asparaginase, a less toxic, but moreexpensive, variation of L-asparaginase.However, as affordable methods existfor managing the side-effects of theunpegylated version, PEG-asparaginasewas put under the ‘desirable’ rather than

the ‘minimum requirement’heading.More important, perhaps, are the

methods used to stratify patients intorisk levels as a guide to treatment, keyamongwhich aremeasurements ofmin-imal residual disease, showing how wellthe patient has responded to the initialinduction treatment (day 33) and thesecond induction treatment (day 72).The gold standard here is using quanti-tative PCR (polymerase chain reaction).“This is a very expensive methodologyand is not possible to adopt in every coun-try,” says Jankovic. However, PCR waslisted under the heading ‘desirable’,because an alternativemethod formeas-uringminimal residual disease does exist,in the form of cytofluorometry. Thoughless accurate than PCR, the team con-sidered it to be an ‘acceptable’alternative.While it is cheaper than PCR, cytofluo-rometry equipment nonetheless requiresa hefty investment, and it was thereforeflagged as a ‘minimum requirementwhere cost constraints could limit access’.

PART EVIDENCE PART EXPERIENCEAs Micheli readily admits, the approachtaken in thisEUROCHIPprojectdrawsasmuch on the experience of the experts asit does on hard peer-reviewed evidence.This is partly amatter of necessity, as evi-denceon thecost–effectivenessof specificprocedures or therapies in thecontext of aparticular indication is oftenhard to comeby.A thorough search of the literature oncost–effectiveness/cost–utility/cost–benefit/costminimisation analyses of theALLdiagnostic/management tools flaggedup as potentially unaffordable showedhow little there is out there–at least in theacademic literature. In the case of child-


Systems&Services

“Individual specialists know what has to be done,

but this is not informatoion that is known to politicians”’

hood ALL, the treatment protocol is sowidely accepted that there may be fewcalls to carry out such analyses.“Basically we still don’t have the evi-

dence in termsof the literature, economicstatistics and data, so we must go on thebasis of experience,” saysAnnalisaTrama,oneof the epidemiologists involved in theALL pilot study. But that experience, sheargues, offers some crucial insights thatwill probably form the basis of the mainrecommendations of the pilot study.She cites, in particular, reports by par-

ticipating oncologists of visits to hospitalswherechildrenbeing treated forALLwerenot losing their hair. “This is almost impos-sible if childrenget theappropriatedoseofchemotherapy.And it raised thequestionoftowhat extent childrenwere really receiv-ing the dose recommended by the proto-col.” In some cases, children were beingtreated in general haematology depart-ments, and sharing wards with adults.As a result of this discussion, says

Trama, theexperts groupstarted lookingatwhether, and to what extent, issues oforganisation – “how these well-knowndrugs andother interventions are actuallyprovided” – might be responsible for theobserved variations in survival. “In thecaseofALL,wearenot talkingaboutaveryexpensive treatment.Therearea fewtech-niques that are important and are expen-sive, but these alone cannot really explainthe difference we see in survival. So wesaid,probably this is an issueofquality andaccessibility of these treatments.” Thatsaid, she adds, there are important costimplications in improving organisationanddelivery of care throughgreater useofreferrals to specialist centres, linked in tonational andEuropeannetworks, and thisneeds to be explicit in any recommenda-tions coming out of the pilot.Micheli is conscious of the need for

any set of recommendations to carry thebacking of the leading voices in the field,andhewill becirculatingadraft of the final

report to bodies such as ECCO, ESMO(Europe’s medical oncologists), SIOPE(Europe’s paediatric oncologists) and theEuropean Leukaemia Network for theircomments and endorsement.Whatever the specific recommenda-

tions may be on ALL and on early andadvancedbreast cancer,Micheli hopes tosend three simple messages to the Com-missionabout closingcancer survival gapsacross Europe.� Studies exploring the relationshipbetween costs and outcomes for spe-cific cancer indications canhelp iden-tify and address the key issues that liebehind variations in survival betweencountries, to help ensure that allEurope’s cancer patients have accessto the minimum requirements foracceptable treatment.

� To facilitate such studies, memberstates shouldbe encouraged to gatherand share information relating to thecosts andbenefits of technologiesusedin specific cancer indications.

� EUfunding shouldbemadeavailablespecifically for studies that explore therelationbetweencancercosts andcan-cer outcomes in the next call for pro-posals for public health or medicalcancer science research.

Jankovic, who has seen the way the Ser-bian government responded to the evi-dence of improved survival when theiryoung ALL patients were treated effec-tively in accordance with minimumrequirements, endorses Micheli’s mes-sages. “If we give the Commission thecorrect information, they can help dif-ferent countries achieve theseminimumrequirements. So I believe in this type ofstudy, which is based on evidence forsome aspects and experience for others.In this waywe can offer the authorities away to ensure patients can be adequatelytreated at the lowest cost.”

Details of this and other EUROCHIP projects can be

found at www.tumori.net/eurochip/

Systems&Services


SIOPE/NURDOR

Survivors. Miloš, Ajla and Hena are among the many young ALL patients to have benefited from aninitiative to ensure all children have access to minimum acceptable treatment, which was spearheadedby Serbian paediatric oncologists in collaboration with a group of Italian specialists

cancer world 44

Documents

cancer worldshould

cancerspecific targets

rights reservedcancer

best arethose

vision fora

similar variations

variations inresources

politics esos masterclass