Approved CAR T cell therapies

DRUDIS-2196;NoofPages8

Pleasecitethisarticleinpressas:Zheng,P.P.etal.ApprovedCARTcelltherapies:icebucketchallengesonglaringsafetyrisksandlong-termimpacts,DrugDiscovToday(2018),https:// doi.org/10.1016/j.drudis.2018.02.012

feature

Approved

CAR

T

cell

therapies:

ice

bucket

challenges

on

glaring

safety

risks

and

long-term

impacts

Ping-PinZheng1,2_{,p.zheng.1@erasmusmc.nl,JohanM.Kros}2_andJinLi3

Two

autologous

chimeric

antigen

receptor

(CAR)

T

cell

therapies

(Kymriah

TM

and

Yescarta

TM

)

were

recently

approved

by

the

FDA.

Kymriah

TM

is

for

the

treatment

of

pediatric

patients

and

young

adults

with

refractory

or

relapse

(R/R)

B

cell

precursor

acute

lymphoblastic

leukemia

and

Yescarta

TM

is

for

the

treatment

of

adult

patients

with

R/R

large

B

cell

lymphoma.

In

common,

both

are

CD19-specific

CAR

T

cell

therapies

lysing

CD19-positive

targets.

Their

dramatic

efficacy

in

the

short

term

has

been

highlighted

by

many

media

reports.

By

contrast,

their

glaring

safety

gaps

behind

the

miracles

remain

much

less

addressed.

Here,

we

focus

on

addressing

the

crucial

challenges

in

relation

to

the

gaps.

Introduction

Twochimericantigenreceptor(CAR)Tcell therapies(KymriahTMandYescartaTM)were recentlyapprovedbytheFDA[1,2].KymriahTM (tisagenlecleucel)isforthetreatmentof pedi-atricpatientsandyoungadultswithrefractory orrelapse(R/R)Bcellprecursoracute lym-phoblasticleukemia(ALL),whereasYescartaTM (axicabtageneciloleucel)isforthetreatmentof adultpatientswithR/RlargeBcelllymphoma. Theyarebothgeneticallymodifiedautologous TcellsexpressingaCD19-specificCAR,lysing CD19-positivetargets(normalandmalignantB lineagecells).Anoteddifferenceisshownin thevectorsusedforKymriahTM_(lentiviral vec-tor)andYescartaTM₍

F

_{-retroviralvector)[3].} Theoverallresponserate(ORR)intheshort termwasveryhigh(83%),solelybasedona singleinfusionofKymriahTM[1],where leuke-miacouldnotbecuredbyanyothermeans,

andpatientswentintoremissionwithin3 monthsofbeingtreatedwithKymriahTM.The recipientsofYescartaTMhad72%ORR[2]. Ob-viously,thereisnodoubtaboutthelifesaving potentialofthetreatmentsinthesehopeless cases.Numerousmedia reportshave dramati-callyhighlightedthelifesavingpotentialof KymriahTMandYescartaTM,andtheyhavebeen coinedas‘livingdrugs’.

Indeed,thisisahistory-rewritingprogressin cancermedicineandaquintessentiallymodern paradigmofclinicaloncology,whichnotonly giveshopebutalsodirectlydrivesinnovative cancersciencetopatientcareandleadstoa paradigmshiftfromprotocol-basedtreatment toreal-timepersonalizedtherapy unprece-dentedly.However,intherealworld,even thoughadrughasagreaterpotencyora medicaltechnologyprovidesdramaticbenefits, distinctandevenseriousadversehealthrisks

canbeassociatedeitherpredictablyor unpre-dictably[4].Ithasbeenevidentthatmanytypes ofanticancerdrugsormodalitiesincluding thosemodernoneswith‘breakthrough desig-nation’haveinducedlife-threatening compli-cations(e.g.cardiotoxicity)[5].KymriahTMand YescartaTMremainthereforenotonlywith seri-ouspatientsafetyeventsalreadynotedinthe shorttermbutalsowiththeirlong-termimpacts (efficacyandsafety)lacking.Asallthe stake-holdersstrivetounderstandthegreatsuccesses, inthemeantime,weshouldkeepinmindthe real-timechallengesandrealizegapsinthe dramaticefficacyversusglaringsafetyconcerns. Here,weanalyzethecrucialchallenges re-gardingthegapsimpactingquality-of-life(QOL) withthetherapies,andprovokeintensive debatesespeciallyregardingthesepotentially long-simmeringproblemsthathavenotyet beenfullyexplored.

Features

PERSPECTIVE

DrugDiscoveryTodayVolume00,Number00December2018 PERSPECTIVE

1359-6446/ã2018TheAuthors.PublishedbyElsevierLtd.ThisisanopenaccessarticleundertheCCBYlicense(http://creativecommons.org/licenses/by/4.0/).

EfficacyversusresistanceofKymriahTM andYescartaTM

Overall,theefficacyversustoxicityandsafetyof atreatmentmanifestsasshort-andlong-term effects.Despitetheexcellentclinicalresponses oftheR/RBALLpatientstoKymriahTM_[1]andR/ RlargeBcelllymphomapatientstoYescartaTM

[2],asignificantnumberofpatientstreatedby KymriahTMhaverelapsedmonthslater[6,7],and nearly30%ofpatientshadapartialresponse treatedbyYescartaTMandthetherapeutic effectstendedtowanebythe6-monthmarkin many[8].Thus,itremainsunknownastohow longthebenefitsofKymriahTMandYescartaTM mightlast(i.e.thereareconcernsabout long-termefficacy).Clinicalrelapsesuggeststhat cancercellsdevelopresistancetothe destruc-tionunleashedbythecytotoxicTlymphocytes

[9].Manybiologicalandbiochemicalfactors couldpotentiallyimpacttheefficacyandsafety ofKymriahTM_andYescartaTM_{(Table1).However,} thedefinitecausesunderlyingtheimmune re-sistanceorpartialresponsearenotfully un-derstood.Someimportantfactorspossibly accountingfortheefficacy,resistanceor ineffi-cacyareformulatedhere.

Challenges

in

synthetic

immunobiology

ExpansionandpersistenceoftheCAR-modifiedT cellsinthebodyarelinkedtomanyfactors(Table 1).Anyofthesefactorscouldcollectivelyor in-dividuallyinfluencetheresponseinthepatients treatedbyKymriahTMandYescartaTM[7,10–20].

Formulation

of

T

cell

subsets

EachTcellsubsethasauniquecytokineprofile, functionalpropertiesandpresumedrolesin pathogenesis[21]andholdsaspecificrolein protectiveimmunity[22].Functionally,Tcells canbeidentifiedaseitherbeneficial tumor-specificTcellsordeleteriouscounterparts[22]. Thus,controllingtheTcellsubsetswith favor-ablefunctioncompositionsofaCARTcell productisoneofthemostimportantaspectsfor manufacturingmore-effectiveclinicalTcell products[10,22].Thestrategyholdsthe po-tentialtoreduceproductvariability,improves theconsistencyofinvivoproliferationand providesreproduciblepotency[11,15,19,22,23]. Moreover,Tcellmaturationstatusisimportant aswell,anditwasfoundthatlessdifferentiated, stem-cell-likeTcellspossessgreatertherapeutic efficacy[24,25].

Immunosuppressive

tumor

microenvironment

Theimmunesystemhasadouble-edgedrole, beinginvolvedinsuppressingtumorgrowthby

destroyingcancercellsandshapingthe im-munogenicphenotypesoftumorstopromote tumorprogressionbyescaping immunosur-veillance[9,26].Theseinhibitoryand immuno-suppressivestimulicanimpedethefunctionof CARTcells[27]and‘armoredCARs’could improveTcellfunction[28].

CD19 variants

(antigen-loss

relapses)

CD19 ALLvariantsarebeingrecognizedwith increasingfrequency,renderingtheCARTcells ineffectiveagainstBcelltumorsandthus repre-sentingabarriertoprogressinCD19-directed immunotherapy[29,30].Severalnovel mecha-nismsassociatedwithCD19 ALLvariantshave beendiscovered[6,31–33](e.g.alternativemRNA splicing,CD19genedeletionormutation, CD19-negativeclonalevolution,inductionofamyeloid switch).Allogeneicstemcelltransplantation (allo-SCT)andco-targetingofmultiplemarkerson leukemiccellscouldbethepossiblesolutions[6]. Buttumor-specificantigensarerare,andthus multiple targeting potentially increasesoff-tumor, on-targettoxicities[5]includingneoreactivities (allo-HLAandautoreactiveactivity)inducedby mixedTcellreceptor(TCR)dimers[34].

CAR

protein

and

RNA

downregulation

CARexpressionisdecreaseduponrepeated stimulations[24,35,36]orwhenthereis accel-erateddifferentiationandexhaustionoftheT cells[24,36].Theseproblemsposeadditional challengesofCARinCARTcelltherapy.A possiblesolutionfortheproblemistodirecta CD19-specificCARtotheTCR

a

constant(TRAC) locusbyCRISPR/Cas9genomeediting[35], whichpotentiallyyieldssomebenefits(e.g. decreasedTcelldifferentiationandexhaustion

[22,37,38],minimizingtherisksofinsertional oncogenesisandTCR-inducedautoimmunity andalloreactivity[35]).

High

dose

of

corticosteroids

Itisunclearwhethertocilizumabhasany bene-ficialeffectsonneurotoxicities[39],becauseits sizemakesefficientblood–brainbarrier(BBB) penetrationunlikely[33,40].Thus,thefirst-line agenttotreatsevereneurotoxicitiesisoftenwith systemiccorticosteroidsratherthantocilizumab

[33,39].However,prolongeduseofhigh-dose corticosteroidsresultsinablationoftheCARTcell population[20,41].Moreover,inappropriateuse ofglucocorticoidsisassociatedwithriskforearly relapseofprimarydisease[41].

Extramedullary

disease

Thecentralnervoussystem(CNS)isa well-rec-ognizedreservoirwhereinleukemiacanescape

systemiccytotoxictherapy[42].TheCNS com-partmentisaffectedinroughlyone-thirdofALL relapses[43,44],whereasCNSinvolvementat relapseoccursmainlyinpatientswhowere CNS-negativeatinitialdiagnosis[44,45].Intriguingly, CD19CARTcellshavebeenidentifiedinthe cerebrospinalfluid(CSF)ofpatientsafter infu-sion[46–48],eventhoughmanyofthepatients (80%)didnothaveahistoryofCNSleukemia

[49],suggestingtheabilityofthesecellstocross theBBB[47,50].Thus,thetherapymightbe consideredtoreplacemultipledosesofeither prophylacticortherapeutic,intrathecal che-motherapyandradiationinleukemiapatients. Theoretically,thereplacementcouldreduce cognitiveimpairmentanddevelopmentaldelay resultingfromchemotherapyandradiationin thepatientpopulation,becauseALLismost commonlydiagnosedinchildrenunder8years ofage,acrucialtimeinbraindevelopment[51]. However,acontradictoryeventinparallel con-siderationisneurotoxicity–oneofthemajor complicationsofKymriahTM_andYescartaTM_.Asa result,cautionshouldalwaysbetakenwhen consideringthereplacement.Furthermore, de-tectionofCD19expressioninthebrain paren-chymaremainscontroversial[25],andthusthe capacityforclearanceofExtramedullarydisease (EMD)bythetherapyremainsuncertain[22]and furtherresearchinthisareaiswarranted.

CommontoxicitiesofKymriahTM, YescartaTMandbeyond

GiventheextremepotencyoftheCAR-modified Tcellsandsimilarmodeofaction,theuseof KymriahTM_andYescartaTM_{harborscommon} fataltoxicpotentialsthatcanbeasbadasor worsethantheoriginalconditionandeven lethal[1,2,10].Somehigherratesofserious adverseeventsmanifestedinacuteorsubacute formshavebeendemonstratedasimmediately life-threatening[1,2,10](Table2).Becausethe cellularimmunesystemhasbeenartificially boostedforanenhancedactivation,KymriahTM andYescartaTMactlike‘immuno-bombs’, remi-niscentoftheatomicbombinginHiroshimaand Nagasakiin1945,andtheimmuno-bombsdrop intothecirculationsystemofthehumanbody tononspecificallydestroycancercellsandtheir innocentcounterparts.Effectivepreventionof theseacuteandsubacutetoxicities(e.g.CRS: Cytokine-releasesyndromeandNT: neurotox-icity)remainsunfeasible,becauseeitherthe mechanismsofthesetoxicitiesremainpoorly understood(e.g.NT)[22]orCARTproductshave endogenouslyinheritedfeatures(e.g.CRS).To date,palliativesupportivecare(PSC)and im-munosuppressionremaintheonlyapproaches

PERSPECTIVE DrugDiscoveryTodayVolume00,Number00December2018

Features



Drug Discove ry Toda y  V olume 00, Number 00  December 2018 PERSPECTIVE DR UDIS-21 96; No of Pages 8 Please cite this article in press as: Zheng, P. P. et al. Approv ed CAR T cell therapies: ice bucket chall enges on glaring safety risks and long-term impacts, Dru g Discov T oday (2018), https:// doi.org/10.1016/j.drudi s.2018.02.012 TABLE 1

Potential biological and biochemical factors impacting the efficacy and safety of KymriahTMand YescartaTM

Potential factors Potential mechanisms or causes Major possible effects Possible solutions and remarks Refs

Number of the transduced T cells Transduction efficiency Impact the reproducible potency Control vector copy and CAR expression [7,20]

Cell lineage and differentiation state

Component variability of the product

Impact the reproducible potency Improve production method

[7,11,15,20,28,38]

Cell viability Nonviable cells Impact the efficacy and safety profile Improve production method [7,10]

Cellular impurities Non T cells (B lineage cells, blasts

and others)

Impact the efficacy and safety profile Improve production method [7,10]

Excipients (DMSO, dextran 40) Anaphylaxis Impact the safety profile Improve production method [1,7,10]

Manufacturing failures Poor starting autologous

leukapheresis cells

Jeopardize disease control and survival Use universal CAR19 T cells [7,10,18]

Specificity of the scFv domain Determinant for the CAR T cell

safety profile

Off-target activity and B cell aplasia Use therapeutic immunoglobulin,

anti-Fc

m

R CAR T, RNA CARs

[7,14,20]

Affinity of scFv binding CAR T cell activation and effector

functions

Impact the safety and activity Not restricted by MHC (applicable to any

MHC haplotype)

[7]

Functional T cell subsets and ratios

Component variability of the product

Impact the reproducible potency Improve production method

[7,11,20,22]

IFN-

g

production A prerequisite for CAR T cell activity An indicator of T cell activation In vitro data may not correlate to in vivo

efficacy, technical advances

[7]

CAR signaling domains and spacer variants

Off-target T cell activation Impact the safety and activity

(CD19-independent toxicities)

Biological optimization

[7,12,13,16,17,20]

Decreased CAR expression Repeated stimulations, accelerated

diff/ex

Reduced efficacy TRAC-CAR to decrease differentiation of T

cells, other approaches

[24,35,36]

T cell dose versus tumor burden An inverse correlation Impact the expansion and persistence Bridging therapy to reduce tumor burden

before KymriahTM

[11,28]

Immunosuppressive environment

Inhibitory and immunosuppressive stimuli

Impede the function of CAR T cells ‘Armored’ CAR T cells to enhance IS, risk of

cumulative toxicities

[22,30]

CD19-negative variants CD19 del/mut, CD19 clonal

evolution, lineage switch

Inefficacy Target multiple antigens, allo-HSCT, risk of

cumulative toxicities

[23,30– 34,55]

Anti-mCAR19 antibodies Immunogenicity Immunity anaphylaxis, impact the

efficacy and safety profile

Use human anti-CD19 CAR (HuCAR-19) [1,7]

Extramedullary disease (EMD) Sanctuary site relapse (e.g. CNS) Uncertain capacity for clearance of EMD Further studies for confirmation [22,25]

Lymphodepletion chemotherapy Conditioning regimen to reduce

tumor burden

Augment the antitumor effects Risk of cumulative toxicities [7,11,28]

High dose of corticosteroids Impede CAR T cell function Diminished efficacy owing to

immunosuppression

Tocilizumab, uncertain effects for neurotoxicities

[20] Abbreviations: IS, immune system; DMSO, dimethyl sulfoxide; MHC, major histocompatibility complex; IFN-g, interferon gamma; TRAC, T cell receptoraconstant locus; diff, differentiation; ex, exhaustion; del, deletion; mut, mutation; scFv, single-chain variable fragment; allo-HSCT, allogeneic stem cell transplantation; CNS, central nervous system.

www .drugdisco verytoday .com 3 FeaturesPERSPECTIVE

PERSPECTIVE Drug Discover y Today  V olume 00, Number 00  December 2018 DR UDIS-21 96; No of Pages 8 Please cite this article in press as: Zheng, P. P. et al. Approv ed CAR T cell therapies: ice bucket chall enges on glaring safety risks and long-term impacts, Dru g Discov T oday (2018), https:// doi.org/10.1016/j.drudi s.2018.02.012 TABLE 2

Common acute and subacute toxicities (incidences>50%, 20%, 10% and 2%) and long-term risks of KymriahTMand YescartaTM

Toxicities (T) and risks (R) Category 50% 20% 10% 2% Clinical form Potential mechanisms/

causes

Management strategies and comments

Refs

CRS T + Short term Activated T cells produce

high levels of cytokines

Familiar with FDA labels, REMS and ETASU

[1,2,7]

Neurotoxicities T + Short term Unknown Familiar with FDA labels,

REMS and ETASU

[1,2,7]

Serious infection T + Short term Acquired

hypogammaglobulinemia Familiar with FDA

labels, REMS and ETASU

[1,2,7]

Prolonged cytopenias T + Short term Miscellaneous causes (e.

g. CRS)

Familiar with FDA labels, REMS and ETASU

[1,2,7]

Acquired hypogammaglobulinemia T + Short term On-target off-tumor

toxicities (B cell aplasia)

Familiar with FDA labels, REMS and ETASU

[1,2,7]

Humoral immunogenicity T + Short term Anti-mCAR19 antibodies Familiar with FDA labels,

REMS and ETASU

[1,2,7]

Tachycardia T + Short term Miscellaneous cause (e.g.

CRS)

Familiar with FDA labels, REMS and ETASU

[1,2,7]

Gastrointestinal disorders T + Short term Miscellaneous causes (e.

g. CRS)

Familiar with FDA labels, REMS and ETASU

[1,2,7]

Acute kidney injury T + Short term Miscellaneous causes (e.

g. CRS)

Familiar with FDA labels, REMS and ETASU

[1,2,7]

Acute respiratory distress T + Short term Miscellaneous causes (e.

g. CRS)

Familiar with FDA labels, REMS and ETASU

[1,2,7]

Musculoskeletal disorders T + Short term Miscellaneous causes (e.

g. CRS)

Familiar with FDA labels, REMS and ETASU

[1,2,7]

Hypotension T + Short term Miscellaneous causes (e.

g. CRS)

Familiar with FDA labels, REMS and ETASU

[1,2,7]

Hypertension T + Short term Miscellaneous causes (e.

g. CRS)

Familiar with FDA labels, REMS and ETASU

[1,2,7]

Cardiac failure or arrest T + Short term Miscellaneous causes (e.

g. CRS)

Familiar with FDA labels, REMS and ETASU

[1,2,7]

TLS T + Short term Large amounts of tumor

cells lysed

Familiar with FDA labels, REMS and ETASU

[1,2,7]

DIC T + Short term Miscellaneous causes (e.

g. CRS)

Familiar with FDA labels, REMS and ETASU

[1,2,7]

MAS T + Short term Uncontrolled activation

of macrophages and T cells

Familiar with FDA labels, REMS and ETASU

[1,2,7]

Capillary leak syndrome (bleeding) T + Short term Miscellaneous causes (e.

g. CRS)

Familiar with FDA labels, REMS and ETASU

[1,2,7] 4 www .drugdisco verytoday .com 

Drug Discove ry Toda y  V olume 00, Number 00  December 2018 PERSPECTIVE DR UDIS-21 96; No of Pages 8 Please cite this article in press as: Zheng, P. P. et al. Approv ed CAR T cell therapies: ice bucket chall enges on glaring safety risks and long-term impacts, Dru g Discov T oday (2018), https:// doi.org/10.1016/j.drudi s.2018.02.012 TABLE 2 (Continued )

Toxicities (T) and risks (R) Category 50% 20% 10% 2% Clinical form Potential mechanisms/

causes

Management strategies and comments

Refs

Coagulopathy T + Short term Miscellaneous causes (e.

g. CRS)

Familiar with FDA labels, REMS and ETASU

[1,2,7]

Hypofibrinogenemia T + Short term Miscellaneous causes (e.

g. CRS)

Familiar with FDA labels, REMS and ETASU

[1,2,7]

GVHD R 1% Undefined Residual donor

lymphocytes from prior HSCT

Warning and intensive monitoring

[7]

Anaphylaxis R Undefined Excipients (e.g. DMSO,

dextran)

Warning and intensive monitoring

[1,2,7]

Secondary malignancies R Long term Insertional oncogenesis

and genotoxicity

Warning and lifelong monitoring

[1,2,7]

Developmental and reproductive toxicity R Long term Miscellaneous causes (e.

g. DAMPs, prolonged CRS)

Warning and lifelong monitoring

[1,2,7]

New incidence of neurologic disorders R Long term Miscellaneous causes (e.

g. DAMPs, prolonged CRS)

Warning and lifelong monitoring

[1,2,7]

Exacerbation of pre-existing neurologic disorders R Long term Miscellaneous causes (e.

g. DAMPs, prolonged CRS)

Warning and lifelong monitoring

[1,2,7]

New incidence of autoimmune disorders R Long term Miscellaneous causes (e.

g. DAMPs, prolonged CRS)

Warning and lifelong monitoring

[1,2,7]

Exacerbation of prior autoimmune disorders R Long term Miscellaneous causes (e.

g. DAMPs, prolonged CRS)

Warning and lifelong monitoring

[1,2,7]

Incidence and outcome of any pregnancy R Undefined Miscellaneous causes (e.

g. DAMPs, prolonged CRS)

Warnings and monitoring during the pregnancy

[1,2,7]

Abbreviations: CRS, cytokine release syndrome; TLS, tumor lysis syndrome; DIC, disseminated intravascular coagulation; MAS, macrophage activation syndrome; GVHD, graft-versus-host disease; DAMPS, damage-associated molecular patterns; HSCT, hematopoietic stem cell transplantation; REMS, Risk Evaluation and Mitigation Strategy; ETASU, Elements to Assure Safe Use.

www .drugdisco verytoday .com 5 FeaturesPERSPECTIVE

fortreatingthesecommoncomplications[1,2], evenconsideringthelatestnewguidelines[52]. Recently,ahumanstudyexploredthe mecha-nismofNTandsuggestedthatanincreasedBBB permeabilitymightexplainNT[53].Thestudy couldleadtofurtherstudiesfordevelopmentof noveltreatmentonthebasisofmechanisms.B cellaplasia(acquiredhypogammaglobulinemia) isanon-targetoff-tumortoxicityfor CD19-tar-getedCAR[1,2](i.e.aspecifictoxicityofCD19 CART)becauseCD19isacell-surface compo-nentofBcelllineage[3].Thereareseveral possiblesolutionstopotentiallyovercomingor minimizingBcellaplasia:(i)useofanti-Fc

m

R CART[14];(ii)useofRNACARs[20];(iii)infusion ofpooledimmunoglobulins[1,2].Beyondthis, additiveside-effects(secondaryortertiary toxicities)derivedfromcombiningorbridging agentsshouldnotbeoverlooked(e.g. tocilizu-mabwithanFDAwarningandprecautionlabels

[54],ibrutinibtopreventCRSafterusing anti-CD19CART[55]withknowncardiacconcerns[5]

andotherseriouscomplications[56,57]). Fur-thermore,theuseofhostlymphodepletion chemotherapywithimmunosuppressiveagents (e.g.cyclophosphamide)beforeaCART ap-proachisarequiredsteptoaugmentthe anti-tumoreffectsofthistreatment[1,2,5,7]. However,suchconcomitanttherapiescanlead toclinicalcardiotoxicity[5].Consequently,these combiningorbridgingagentsmightincrease somecumulativeorsynergistictoxicitiesforthe patients.

Uncertainlong-termoutcomesof KymriahTMandYescartaTM

Datawerefast-emergingontheearlyresponses toKymriahTM_andYescartaTM_{,thusmostofthe} patientsparticipatinginthetrialshaveonly beenfollowedforarelativelyshortperiodof time[1,2],limitingtheabilitytoassesstherisk oflong-termadverseeventsandrulethemout. Asaresult,long-termsequelaeandlatetoxic effectsofKymriahTMandYescartaTMremain unknownalthoughsomearetheoretically predictable(Table2).Theoretically,the after-mathoftheimmuno-bombinginthehuman bodycanbejustasdeadlyandfar-reaching, becausethesecellularandmolecularfallouts fromthesedamagedleukemiacellsandtheir normalcounterpartsinthebloodcirculation reachasfarasanysystemicorgans.Such damagetonormalcellsandtissuesmightbe long-termandprobablypermanentlytoxic

[7,58].Thisisin-linewiththerationalethatthe immunesystemnotonlyrespondstoforeign substances(i.e.pathogens)butalsoresponds toendogenouslyderivedmoleculesthatare

expressedasaresultoftissuedamageor stressedcells,knownasdamage-associated molecularpatterns(DAMPs)[59],whichcan causevariousdiseases(e.g.autoimmune dis-eases)[60,61].Further,lateonsetofNTis an-otherconcernforcognitivedysfunction.Little isknownabouttimingofthesecondaryand/or tertiarytoxicitiesresultingfromDAMPs. Re-ferringtothepathogenesisandlong-term courseofmanyautoimmunediseasesand neurocognitivedisorders,achronic, progres-sivediseaseprocessshouldbeanticipated. Giventheextremeimportancetotheyoung patientpopulationuniquelytargetedby Kym-riahTM,itisworthknowingthatclassical gen-otoxicityassaysandcarcinogenicityassessment invivo(rodentmodels)werenotperformedfor KymriahTM[7,10].Developmentaland repro-ductivetoxicitystudieswerenotconductedin thenonclinicalstudiesforKymriahTM_either

[7,10].Thus,detectionoflong-termproblems assuchwillnotonlybedependentona long-termfollow-upbutalsoenhancedclinical awarenessandsensitivedetectionalgorithms arerequiredforagoal-orientedevaluation. Takentogether,thesafetyprofilesandthetoxic potentialofKymriahTMandYescartaTMcannot beassessedinisolationforshort-term moni-toringandmanagementbutneedtobe con-sideredtogetherwithalong-termfollow-up.

LifesavingversusQOL-preservingof KymriahTMandYescartaTM

Immune-cell-basedtherapiesopenanew fron-tierforcancertreatments.Butthechanging landscapeofmedicalbenefitsandriskscreates newchallengesforallthestakeholdersin healthcareowingtopotentiallylethalside effectsofthetherapiesanduncertainlong-term impactsonQOL.Currently,becausethedata aboutthelong-termimpactsofKymriahTMand YescartaTMarenotavailableyet,thereis insuf-ficientvoicetoclaimmuchmorebenefitsthan medicallyacknowledged,insteadofbeing in-creasinglyawareoftheshort-andlong-term risks[58].Mediareportsoftenstate dispropor-tionatelyonriskbyoverstatingbenefitswhile understatingtheharms[4,58].Nevertheless,the FDAplaysacentralpartasanauthoritativevoice incommunicatingthebenefitsandrisksofa drug[4].Itisimportantforallthestakeholdersto becomefamiliarwiththeFDAlabelscontaining aRiskEvaluationandMitigationStrategy(REMS) andanElementstoAssureSafeUse(ETASU)

[1,2,7,10].Lifesavingcareandpreservingpatient QOLarethetasksofmodernmedicine,being especiallyimportantforthepatientpopulations ofchildrenandyoungadults.Asmore

infor-mationabouttreatmentoptionsbecomes available,patients,physicians,regulatorsand payersarereassessinghowtheybalancethe possiblebenefitsandrisksoftherapeutic options[4].Theoretically,nopatientsexpectany treatmentofprocedurethatis disproportion-atelycostly,burdensomeorpainful[62]. How-ever,practically,whendoctorstreatpatients withlife-threateningconditions(e.g.lethal cancers),themajorfocuswouldoftenbequickly directedtowardinstitutingtherapeutic mea-surestopreservelife(lifesaving),andoftenthey areunabletoaddresstheimpactofmedicalcare onQOLuntilafterthelifesavingintervention

[63].KymriahTMandYescartaTMwereregarded asalifesavingtreatment(alast-resorttreatment)

[1,2]andfallwithinthescopeofaformaldebate inthisregard.Ironically,whereadvancesin technologyandknowledgehavegivendoctors anincreasedcapacitytopreserveandprolong life,somefundamentalethicalquestionscould beraisedinparallel:shoulddoctorsbe con-cernedonlywithcuringdisease(lifesaving)?Do theyhavearesponsibilitytogivethepatients thebestpossibleQOLwhilebeingphysicallyor fiscallyreasonable[63]?Theseethicaldilemmas mighthavetobeaddressedattheclinicdoor thatimpactsindividualpatientsbya participa-tivemanagementinvolvingpatients,doctors andotherstakeholders.Inthiscontext,an eth-icalimperativerequiresclassificationofthe medicalsignificanceofanintervention espe-ciallywhentheinterventionremains contro-versialandunderexamined,whichwillbenefit fromdecreasingtheuncertaintyassociatedwith theintervention.

Concludingremarksandfuture perspectives

KymriahTMandYescartaTMgainedgroundas last-resorttreatmentsforR/RpediatricALLand R/RadultBcelllymphoma,respectively,owing totheirlifesavingpotentials.Thebroad appli-cationsremainchallengingbecauseofacute lethaltoxicitiesandalsouncertainlong-term impacts.Post-approvalpharmacovigilanceis crucialasoneofthefirstconsiderationsforrisk mitigationoftheseknownshort-termtoxicities. Long-termfollow-upfordurableefficacyand safetyconcernsispendingfurtherprogress. Furthermore,advancesinmanufacturing pro-cessescouldrevealthebetterversionof T-cell-basedtherapies,evenbeyondcancertherapy,to extrapolatetheapproachtotreatmentof in-fectiousandautoimmunediseases.Tothisend, alleffortsshouldbechanneledintoturningthe icebucketchallengesintosolutionsand opportunities.

PERSPECTIVE DrugDiscoveryTodayVolume00,Number00December2018

Features



Conflictsofinterest

Theauthorshavenoconflictsofinterestto declare

Acknowledgements

P.P.Z.conceivedtheideas,organizedthestudy anddraftedthemanuscript.Alltheauthors reviewedandapprovedthesubmission.

References

1 Novartis(2017)PrescribingInformation(KymriahTM ). Availableat: https://www.pharma.us.novartis.com/ sites/www.pharma.us.novartis.com/files/kymriah.pdf 2 YescartaTM

(2017)PrescribingInformation. Availableat: https://www.fda.gov/downloads/

BiologicsBloodVaccines/CellularGeneTherapyProducts/ ApprovedProducts/UCM581226.pdf

3 Sadelain,M.(2017)CD19CARTcells.Cell171,1471 4 Juhn,P.etal.(2007)Balancingmodernmedicalbenefits

andrisks.HealthAff.26,647–652

5 Zheng,P.P.etal.(2018)Breakthroughsinmoderncancer therapyandelusivecardiotoxicity:critical research-practicegaps,challenges,andinsights.Med.Res.Rev.38, 325–376.http://dx.doi.org/10.1002/med.21463 6 Ruella, M. and Maus, M.V. (2016) Catch me if you can: leukemia

escapeafterCD19-directedTcellimmunotherapies.Comput. Struct.Biotechnol.J.14,357–362

7 FDABriefingDocumentOncologicDrugsAdvisory CommitteeMeetingBLA125646Tisagenlecleucel NovartisPharmaceuticalsCorporation.

8 NationalCancerInstitute(2017)WithFDAApprovalfor AdvancedLymphoma,SecondCART-CellTherapyMoves totheClinic. Availableat: https://www.cancer.gov/ news-events/cancer-currents-blog/2017/

yescarta-fda-lymphoma

9 Bonavida,B.andChouaib,S.(2017)Resistanceto anticancerimmunityincancerpatients:potential strategiestoreverseresistance.Ann.Oncol.28,457–467 10Lu,X.V.(2017)SummaryBasisforRegulatoryAction.

Availableat: https://www.fda.gov/downloads/ BiologicsBloodVaccines/

CellularGeneTherapyProducts/ApprovedProducts/ UCM577221.pdf

11Turtle,C.J.etal.(2016)CD19CAR-TcellsofdefinedCD4 +:CD8+compositioninadultBcellALLpatients.J.Clin. Invest.126,2123–2138

12Zhao,Z.etal.(2015)Structuraldesignofengineered costimulationdeterminestumorrejectionkineticsand persistenceofCARTcells.CancerCell28,415–428 13Kawalekar,O.U.etal.(2016)Distinctsignalingof

coreceptorsregulatesspecificmetabolismpathways andimpactsmemorydevelopmentinCARTcells. Immunity44,380–390

14Faitschuk,E.etal.(2016)ChimericantigenreceptorTcells targetingFcmureceptorselectivelyeliminateCLLcells whilesparinghealthyBcells.Blood128,1711–1722 15Sommermeyer,D.etal.(2016)Chimericantigen

receptor-modifiedTcellsderivedfromdefinedCD8+ andCD4+subsetsconfersuperiorantitumorreactivity invivo.Leukemia30,492–500

16Almasbak,H.etal.(2015)InclusionofanIgG1-Fcspacer abrogatesefficacyofCD19CARTcellsinaxenograft mousemodel.GeneTher.22,391–403

17Watanabe,N.etal.(2016)Fine-tuningtheCARspacer improvesT-cellpotency.Oncoimmunology5,e1253656 18Qasim,W.etal.(2017)Molecularremissionofinfant B-ALLafterinfusionofuniversalTALENgene-editedCAR

Tcells.Sci.Transl.Med.9http://dx.doi.org/10.1126/ scitranslmed.aaj2013

19Golubovskaya,V.andWu,L.(2016)Differentsubsetsof Tcells,memory,effectorfunctions,andCAR-T immunotherapy.Cancers8,E36

20Johnson,L.A.andJune,C.H.(2017)Driving gene-engineeredTcellimmunotherapyofcancer.CellRes. 27,38–58

21Raphael,I.etal.(2015)Tcellsubsetsandtheirsignature cytokinesinautoimmuneandinflammatorydiseases. Cytokine74,5–17

22Sadelain,M.etal.(2017)TherapeuticTcellengineering. Nature545,423–431

23Gill,S.etal.(2016)ChimericantigenreceptorTcell therapy:25yearsinthemaking.BloodRev.30,157–167 24Schietinger,A.andGreenberg,P.D.(2014)Tolerance

andexhaustion:definingmechanismsofTcell dysfunction.TrendsImmunol.35,51–60

25Nellan,A.andLee,D.W.(2015)Pavingtheroadahead forCD19CART-celltherapy.Curr.Opin.Hematol.22, 516–520

26Vinay,D.S.etal.(2015)Immuneevasionincancer: mechanisticbasisandtherapeuticstrategies.Semin. CancerBiol.35(Suppl),185–198

27Moon,E.K.etal.(2014)MultifactorialT-cell hypofunctionthatisreversiblecanlimittheefficacyof chimericantigenreceptor-transducedhumanTcellsin solidtumors.Clin.CancerRes.20,4262–4273 28Jaspers,J.E.andBrentjens,R.J.(2017)Developmentof

CARTcellsdesignedtoimproveantitumorefficacyand safety.Pharmacol.Ther.178,83–91

29Wang,J.etal.(2017)Acutelymphoblasticleukemia relapseafterCD19-targetedchimericantigenreceptor Tcelltherapy.J.Leukoc.Biol.102,1347–1356 30Jackson,H.J.andBrentjens,R.J.(2015)Overcoming

antigenescapewithCART-celltherapy.CancerDiscov. 5,1238–1240

31Onea,A.S.andJazirehi,A.R.(2016)CD19chimeric antigenreceptor(CD19CAR)-redirectedadoptiveT-cell immunotherapyforthetreatmentofrelapsedor refractoryB-cellnon-Hodgkin'slymphomas.Am.J. CancerRes.6,403–424

32Sotillo,E.etal.(2015)AlternativesplicingofCD19 mRNAinleukemiasescapingCART-19immunotherapy eliminatesthecognateepitopeandcontributesto treatmentfailure.CancerRes.75 (abstract3143) 33Wei,G.etal.(2017)AdvancesofCD19-directed

chimericantigenreceptor-modifiedTcellsin refractory/relapsedacutelymphoblasticleukemia.Exp. Hematol.Oncol.6,10

34vanLoenen,M.M.etal.(2010)MixedTcellreceptor dimersharborpotentiallyharmfulneoreactivity.Proc. Natl.Acad.Sci.U.S.A.107,10972–10977

35Eyquem,J.etal.(2017)TargetingaCARtotheTRAC locuswithCRISPR/Cas9enhancestumourrejection. Nature543,113–117

36Wherry,E.J.andKurachi,M.(2015)Molecularand cellularinsightsintoTcellexhaustion.Nat.Rev. Immunol.15,486–499

37Gattinoni,L.etal.(2012)Pathstostemness:building theultimateantitumourTcell.Nat.Rev.Cancer12,671– 684

38Busch,D.H.etal.(2016)RoleofmemoryTcellsubsetsfor adoptiveimmunotherapy.Semin.Immunol.28,28–34 39Brudno,J.N.andKochenderfer,J.N.(2016)Toxicitiesof

chimericantigenreceptorTcells:recognitionand management.Blood127,3321–3330

40Pardridge,W.M.(1998)CNSdrugdesignbasedon principlesofblood–brainbarriertransport.J. Neurochem.70,1781–1792

41Davila,M.L.etal.(2014)Efficacyandtoxicity managementof19–28zCARTcelltherapyinBcellacute lymphoblasticleukemia.Sci.Transl.Med.6,224ra225 42Davis,K.L.andMackall,C.L.(2016)Immunotherapyfor

acutelymphoblasticleukemia:fromfaminetofeast. BloodAdvances1,265–269

43Krishnan,S.etal.(2010)Temporalchangesinthe incidenceandpatternofcentralnervoussystemrelapses inchildrenwithacutelymphoblasticleukaemiatreated onfourconsecutiveMedicalResearchCounciltrials, 1985–2001.Leukemia24,450–459

44Alsadeq,A.andSchewe,D.M.(2017)Acute lymphoblasticleukemiaofthecentralnervoussystem: ontheroleofPBX1.Haematologica102,611–613 45Burger,B.etal.(2003)Diagnosticcerebrospinalfluid

examinationinchildrenwithacutelymphoblastic leukemia:significanceoflowleukocytecountswith blastsortraumaticlumbarpuncture.J.Clin.Oncol.21, 184–188

46Maude,S.L.etal.(2014)ChimericantigenreceptorT cellsforsustainedremissionsinleukemia.N.Engl.J. Med.371,1507–1517

47Grupp,S.A.etal.(2013)Chimericantigen receptor-modifiedTcellsforacutelymphoidleukemia.N.Engl.J. Med.368,1509–1518

48Abramson,J.S.etal.(2017)Anti-CD19CARTcellsin CNSdiffuselarge-B-celllymphoma.N.Engl.J.Med.377, 783–784

49Lee,D.W.etal.(2015)TcellsexpressingCD19chimeric antigenreceptorsforacutelymphoblasticleukaemiain childrenandyoungadults:aPhase1dose-escalation trial.Lancet385,517–528

50Kochenderfer,J.N.etal.(2015) Chemotherapy-refractorydiffuselargeB-celllymphomaandindolent B-cellmalignanciescanbeeffectivelytreatedwith autologousTcellsexpressingananti-CD19chimeric antigenreceptor.J.Clin.Oncol.33,540–549 51Pizzo,P.A.andPoplack,D.G.,eds(2015)Principlesand

PracticeofPediatricOncology,LippincottWilliams& Wilkins

52Neelapu,S.S.etal.(2018)Chimericantigenreceptor T-celltherapy–assessmentandmanagementof toxicities.Nat.Rev.Clin.Oncol.15,47–62

53Gust,J.etal.(2017)Endothelialactivationandblood– brainbarrierdisruptioninneurotoxicityafteradoptive immunotherapywithCD19CAR-Tcells.CancerDiscov. 7,1404–1419

54Tocilizumab(2010)PrescribingInformation. Available at: https://www.accessdata.fda.gov/drugsatfda_docs/ label/2013/125276s092lbl.pdf

55Ruella,M.etal.(2017)Kinaseinhibitoribrutinibto preventcytokine-releasesyndromeafteranti-CD19 chimericantigenreceptorTcellsforB-cellneoplasms. Leukemia31,246–248

56Cheson,B.D.etal.(2017)Tumorlysissyndromein chroniclymphocyticleukemiawithnoveltargeted agents.Oncologist22,1283–1291

57Kaur,V.etal.(2014)Ibrutinib-associatedtumorlysis syndromeinapatientwithchroniclymphocytic leukemia.Blood124,3503–3505

58Bach,P.B.etal.(2017)FDAapprovaloftisagenlecleucel: promiseandcomplexitiesofa$475000cancerdrug. JAMA318,1861–1862

59Hernandez,C.etal.(2016)Damage-associated molecularpatternsincancer:adouble-edgedsword. Oncogene35,5931–5941

DrugDiscoveryTodayVolume00,Number00December2018 PERSPECTIVE

DRUDIS-2196;NoofPages8

Pleasecitethisarticleinpressas:Zheng,P.P.etal.ApprovedCARTcelltherapies:icebucketchallengesonglaringsafetyrisksandlong-termimpacts,DrugDiscovToday(2018),https:// doi.org/10.1016/j.drudis.2018.02.012

www.drugdiscoverytoday.com 7

Features

60Alvarez,K.andVasquez,G.(2017)Damage-associated molecularpatternsandtheirroleasinitiatorsof inflammatoryandauto-immunesignalsinsystemic lupuserythematosus.Int.Rev.Immunol.36,259–270 61Yatim,N.etal.(2017)Dyingcellsactivelyregulate

adaptiveimmuneresponses.Nat.Rev.Immunol.17, 262–275

62Paris,J.J.andPoorman,M.(1995)``PlayingGod”and theremovaloflife-prolongingtherapy.J.Med.Philos. 20,403–418

63Rosoff,P.M.(2008)Physicianresponsibilityforpatients’ qualityoflife.VirtualMentor10,401–403

Ping-PinZheng1,2,* JohanM.Kros2 JinLi3

1

DrugToxicityResearch,ErasmusMedicalCenter, TheNetherlands

2_{DepartmentofPathology,ErasmusMedical} Cen-ter,TheNetherlands

3

DepartmentofOncology,ShanghaiEast Hospi-tal,TongjiUniversitySchoolofMedicine, Shang-hai,China

*Correspondingauthor:

PERSPECTIVE DrugDiscoveryTodayVolume00,Number00December2018

Features