TCellsExpressingCD19/CD20BispeciﬁcChimeric Antigen ... · chimericantigenreceptors(CARs)havecurativepotentialagainst relapsed B-cell malignancies (3–9). However, these trials

Research Article

TCellsExpressingCD19/CD20BispecificChimericAntigen Receptors Prevent Antigen Escape byMalignant B CellsEugenia Zah1, Meng-Yin Lin1, Anne Silva-Benedict2,3, Michael C. Jensen2,4,5, andYvonne Y. Chen1

Abstract

The adoptive transfer of T cells expressing anti-CD19 chimericantigen receptors (CARs) has shown remarkable curative potentialagainst advanced B-cell malignancies, butmultiple trials have alsoreported patient relapses due to the emergence of CD19-negativeleukemic cells. Here, we report the design and optimization ofsingle-chain, bispecificCARs that trigger robust cytotoxicity againsttarget cells expressing either CD19 or CD20, two clinically vali-dated targets for B-cell malignancies.We determined the structuralparameters required for efficient dual-antigen recognition, and wedemonstrate thatoptimizedbispecificCARs can controlbothwild-type B-cell lymphoma and CD19– mutants with equal efficiency

in vivo. To our knowledge, this is the first bispecific CAR capable ofpreventing antigen escape by performing true OR-gate signalcomputation on a clinically relevant pair of tumor-associatedantigens. The CD19-OR-CD20 CAR is fully compatible withexisting T-cell manufacturing procedures and implementableby current clinical protocols. These results present an effectivesolution to the challenge of antigen escape in CD19 CAR T-celltherapy, and they highlight the utility of structure-based rationaldesign in thedevelopmentof receptorswithhigher-level complexity.Cancer Immunol Res; 4(6); 498–508. �2016 AACR.

See related Spotlight by Sadelain, p. 473.

IntroductionAdoptive T-cell therapy has demonstrated clinical efficacy

against advanced cancers (1, 2). In particular, multiple clinicaltrials have shown that T cells programmed to express anti-CD19chimeric antigen receptors (CARs) have curative potential againstrelapsed B-cell malignancies (3–9). However, these trials havealso revealed critical vulnerabilities in current CAR technology,including susceptibility to antigen escape by tumor cells (7, 10).For example, a recent trial of CD19 CAR T-cell therapy revealedthat 90% of patients achieved a complete response, but 11% ofthose patients eventually relapsed with CD19-negative tumors(10). Antigen escape has also been observed in the adoptivetransfer of T cells expressing NY-ESO1–specific T-cell receptors(11) and in cancer vaccine therapy for melanoma (12, 13).

The probability of antigen escape by spontaneous mutationand selective expansion of antigen-negative tumor cells decreaseswith each additional antigen that can be recognized by the CAR Tcells. Therefore, a potential prophylaxis against antigen escape isto generate T cells capable of recognizing multiple antigens. Here,we present the rational design and systematic optimization ofsingle-chain, bispecific CARs that efficiently trigger T-cell activa-tion when either of two pan–B-cell markers, CD19 or CD20, ispresent on the target cell. These CARs can be efficiently integratedinto primary human T cells and administered in the samemanneras CAR T cells currently under evaluation in the clinic.

A single-chain, bispecific CAR targeting CD19 and the humanepidermal growth factor receptor 2 (HER2/neu) was previouslyreported by Grada and colleagues (14). This bispecific receptor,termed TanCAR, efficiently triggered T-cell activation in responseto either CD19 or HER2. However, because CD19 and HER2 arenot typically expressed on the same cell, the TanCAR remains aproof-of-concept design. Importantly, Grada and colleagueshighlighted that the TanCAR is less strongly activated both invitro and in vivo when challenged with target cells that expressHER2 alone, as compared with HER2/CD19 double-positivetargets (14). This observation suggests that tumor cells can stillescape TanCAR detection by eliminating CD19 expression.

To effectively prevent antigen escape, the bispecific CAR mustnot only recognize two antigens, but also process both signals in atrue Boolean OR-gate fashion—i.e., either antigen input shouldbe sufficient to trigger robust T-cell output. We thus refer to thisparticular type of bispecific receptors as "OR-gate CARs." Here, wereport on the development of CD19-OR-CD20 CARs, whichtrigger robust T-cell–mediated cytokine production and cytotox-icity when either CD19 or CD20 is present on the target cell. Wedemonstrate that the size and rigidity of CAR molecules can be

1Department of Chemical andBiomolecular Engineering, University ofCalifornia, LosAngeles, LosAngeles,California. 2BenTowneCenter forChildhood Cancer Research, Seattle Children's Research Institute,Seattle, Washington. 3Department of Oncology and Hematology,St. Luke's Regional Cancer Center, Duluth, Minnesota. 4Division ofPediatric Hematology-Oncology, University of Washington School ofMedicine, Seattle, Washington. 5Program in Immunology, ClinicalResearch Division, Fred Hutchinson Cancer Research Center, Seattle,Washington.

Note: Supplementary data for this article are available at Cancer ImmunologyResearch Online (http://cancerimmunolres.aacrjournals.org/).

Corresponding Author: Yvonne Y. Chen, University of California, Los Angeles,420 Westwood Plaza, Boelter Hall 5531, Los Angeles, CA 90095. Phone:310-825-2816. Fax: 310-206-4107. E-mail: [email protected].

doi: 10.1158/2326-6066.CIR-15-0231

�2016 American Association for Cancer Research.

CancerImmunologyResearch

Cancer Immunol Res; 4(6) June 2016498

on June 29, 2020. © 2016 American Association for Cancer Research. cancerimmunolres.aacrjournals.org Downloaded from

Published OnlineFirst April 8, 2016; DOI: 10.1158/2326-6066.CIR-15-0231









http://cancerimmunolres.aacrjournals.org/





calibrated tomatch the specific antigens targeted, and the optimalOR-gate CAR structure can be deduced from known structuralrequirements for single-input CARs. Finally, we show that theCD19-OR-CD20 CARs can control both wild-type and CD19–

mutant B-cell lymphomas with equal efficiency in vivo, thusproviding an effective safeguard against antigen escape in adop-tive T-cell therapy for B-cell malignancies.

Materials and MethodsPlasmid construction

Bispecific CD19-CD20 CARs were constructed by isothermalassembly (15) ofDNA fragments encoding the CD19 single-chainvariable fragment (scFv) derived from the FMC63 mAb (16), theCD20 scFv derived from the Leu-16 mAb (17), an IgG4-basedextracellular spacer, the CD28 transmembrane domain, and thecytoplasmic domains of 4-1BB and CD3 zeta. Sequences ofextracellular spacers and linkers connecting scFv domains arelisted in Supplementary Table S1, and sequences of all OR-gateCARs are available in GenBank (accession numbers KX000905 toKX000911). All CARs were fused to a truncated epidermal growthfactor receptor (EGFRt) via a T2A peptide to facilitate antibodystaining and sorting of CAR-expressing cells (18).

Cell line generation and maintenanceParental Raji cells were obtained from the ATCC in 2003, and

parental K562 cells were a gift fromDr. Laurence Cooper in 2001.Both cell lines were authenticated by short tandem repeat profil-ing at the University of Arizona Genetics Core in 2015. TheEpstein–Barr virus-transformed lymphoblastoid cell line (TM-LCL) was made from peripheral blood mononuclear cells aspreviously described (19). CD19þ, CD20þ, and CD19þ/CD20þ

K562 cells were generated by lentivirally transducing parentalK562 cells with CD19 and/or CD20 constructs. CD19– Raji cellswere generated by CRISPR/Cas9-mediated gene editing. Raji cells(2 � 106) were transiently transfected with CD19-CRISPR-T2A-NM plasmid (2 mg) using the Amaxa Nucleofection Kit V (Lonza)and the Amaxa Nucleofector 2b Device (Lonza). After 24 hours,cells were coincubated with G418 sulfate (1.5 mg/mL; Enzo LifeSciences) for 120 hours and further expanded in the absence ofantibiotics for 11 days. CD19– cells were first enriched by stainingcells with anti–CD19-APC (allophycocyanin) followed by anti-APC microbeads (Miltenyi), and then removing labeled cells viamagnetic bead–based cell sorting. A pure CD19– population wassubsequently obtained by fluorescence-activated cell sortingusing the FACSAria (II) at theUCLAFlowCytometry Core Facility.All TM-LCL, Raji, andK562 cell linesweremaintained in completeT-cell medium [RPMI-1640 (Lonza) with 10% heat-inactivatedFBS (HI-FBS; Life Technologies)].Human embryonic kidney293Tcells (ATCC) were cultured in DMEM (HyClone) supplementedwith 10% HI-FBS.

Generation of CAR-expressing primary human T cellsCD8þ/CD45RA–/CD62Lþ T cells were isolated from healthy

donor whole blood obtained from the UCLA Blood and PlateletCenter, stimulated with CD3/CD28 T-cell activation Dynabeads(Life Technologies) at a 1:1 bead:cell ratio, and lentivirally trans-duced 72 hours later at amultiplicity of infection of 1.5. All T cellswere expanded in complete T-cell medium supplemented withpenicillin–streptomycin (100 U/mL; Life Technologies) and fedIL2 (50 U/mL; Life Technologies) and IL15 (1 ng/mL; Miltenyi)

every 48 hours. Dynabeads were removed 9 or 10 days afterisolation. CARþ cells were enriched by magnetic bead–basedsorting (Miltenyi) and expanded by stimulation with irradiatedTM-LCLs at a T-cell:TM-LCL ratio of 1:7. Mock-transduced T cellswere stimulated using the rapid expansion protocol as previouslydescribed (20). Each in vitro experiment was repeated with T cellsfrom different donors (T cells were never pooled). See Supple-mentary Materials and Methods for additional details.

Cytotoxicity assayTarget cells (K562 cells) seeded at 1� 104 cells/well in a 96-well

plate were coincubated with effector cells at varying effector-to-target (E:T) ratios in completemediawithout phenol red andwith5%HI-FBS for 4 hours. Supernatants were harvested and analyzedusing the CytoTox 96 Non-Radioactive Cytotoxicity Assay kit(Promega).

Cytokine production quantificationTarget cells were seeded at 5 � 104 cells/well in a 96-well plate

and coincubated with effector cells at an E:T ratio of 2:1 for 24hours. Cytokine concentrations in the culture supernatant weremeasured with the BD Cytometric Bead Array Human Th1/Th2Cytokine Kit II (BD Biosciences).

In vivo xenograft studies in NOD/SCID/gc�/� (NSG) miceAll in vivo experimentswere approvedby theUCLA Institutional

Animal Care and Use Committee. Six- to eight-week-old femaleNSG mice were bred in house by the UCLA Department ofRadiation andOncology. EGFPþ, firefly luciferase (ffLuc)–expres-sing Raji cells (5 � 105) were administered to NSG mice via tail-vein injection. Seven days later, mice bearing engrafted tumorswere treated with 10 � 106 mock-transduced or CARþ/EGFRtþ

cells via tail-vein injection. Tumor progression was monitored bybioluminescence imaging using an IVIS Lumina III LT ImagingSystem (PerkinElmer). Peripheral blood was obtained by retro-orbital bleeding 10 days and 20 days after tumor-cell injection,and samples were analyzed by flow cytometry.

Statistical analysisStatistical significance of in vitro results was analyzed using two-

tailed, unpaired, homoscedastic Student t test. Survival curveswere evaluated by the log-rank test, and statistical significancewasdetermined by comparing log-rank test statistics against a c2 tablewith the degree of freedom equal to one.

ResultsCD19 and CD20 CARs require distinct extracellularspacer lengths

Conventional CAR molecules comprise four main domains:an scFv, an extracellular spacer, a transmembrane domain, and acytoplasmic tail including costimulatory signals and the CD3zchain (Fig. 1A). The CD19 CAR has superior activity whenconstructed with a short extracellular spacer (21), but no sys-tematic study has been reported for CD20 CARs. Therefore, weconstructed and characterized a series of CD20 CARs incorpo-rating long (IgG4 hinge-CH2-CH3; 229 aa), medium (IgG4hinge-CH3; 119 aa), or short (IgG4 hinge; 12 aa) extracellularspacers connecting the Leu-16 scFv to the CD28 transmembranedomain (Supplementary Fig. S1A). All three CARs were efficient-ly expressed in primary CD8þ human T cells (Supplementary

Bispecific T Cells Prevent B-cell Antigen Escape

www.aacrjournals.org Cancer Immunol Res; 4(6) June 2016 499




Fig. S1B). When challenged with CD20þ target cells, T cellsexpressing the long-spacer CAR consistently showed greatertarget-cell lysis, cytokine production, and T-cell proliferation(Supplementary Fig. S1C–S1E), indicating functional superior-ity of the long-spacer CAR for CD20 detection.

Bispecific and single-input CARs share antigen-specificstructural requirements

Second-generation, CD19-OR-CD20 CARs were constructed bytaking the standard four-domain CAR architecture and incorpo-rating two scFvs connected in tandem via a G4S flexible linker(Fig. 1B). In light of the different preferences in extracellular spacerlength for CD19 and CD20 single-input CARs, a panel of CD19-OR-CD20 CARs was constructed to systematically evaluate theeffects of spacer length and the ordering of the two scFv domains(Fig. 1C). Each CAR was constructed in the scFv #1 (VL-VH) – scFv#2 (VH-VL) orientation tominimize potentialmispairing of VL andVH domains between the two scFvs. All four OR-gate CARs wereproduced at full length and expressed on the surface of primaryhuman CD8þ T cells (Supplementary Fig. S2A and S2B).

When challenged with CD19þ target cells, the OR-gate CARsshowed superior cytokine production and target-cell lysis when ashort extracellular spacer was used, regardless of scFv domainorder (Fig. 2). Both long-spacer CARs failed to produce IL2(Fig. 2A) and had minimal lysis activity against CD19þ targets(Fig. 2B). Thus, short-spacer CARs weremore effective in targeting

CD19, consistent with previous reports of single-input CD19CARs (21).

Although the long-spacer CARs did not target CD19, they couldrespond toCD20 (Fig. 2A). The20-19 shortCAR,which containeda short spacer and theCD20 scFv in themembrane-distal position,also performed well in cytokine production and was the mostefficient at lysing CD20þ targets (Fig. 2). Given the scFv orienta-tion in this construct, the CD19 scFv and the G4S linker betweenthe two scFv domains could serve as a proxy spacer that projectsthe CD20 scFv away from the T-cell membrane, to a positionconducive for CD20 binding. We thus hypothesized that modi-fying the G4S linker while keeping the 20-19 short CAR config-urationmight further enhanceCD20 responsewithout compromis-ing the receptor's sensitivity toward CD19þ target cells.

Sequence modifications on scFv linkers enable effectivetargeting of disparate antigens

A new panel of 20-19 short CARs was constructed with variouslinker sequences inserted between the two scFv domains (Fig. 3A).The original, short (G4S)1 flexible linker was compared against along (G4S)4 flexible linker, a short (EAAAK)1 rigid linker, and along (EAAAK)3 rigid linker (22, 23). CARs with modified linkersequences were expressed at full length and localized to the cellsurface (Supplementary Fig. S2C and S2D).

To evaluate the utility of OR-gate CARs in preventing antigenescape, a mutant CD19– lymphoma cell line was generated by

A B

C

ScFv #2

IgG4 Spacer

Cell membrane

ζ ζ

CD28 tm

4-1BB

CD3ζ Signaling domain

ScFv #1 VH

VH VL G4S Linker

VL

VH VL

ScFv

VL

VH

CD3ζ Signaling domain

Transmembrane domain

Costimulatory domain

Cell membrane

Extracellular spacer

Figure 1.Schematics of single-input and bispecific CARs. A, a second-generation, single-input CAR. B, a bispecific, OR-gate CAR. C, schematic of four OR-gate CARscontaining variations in extracellular spacer length and ordering of scFv domains. Hinge, CH2, and CH3 are domains within human IgG4. CD28tm is the CD28transmembrane domain. EGFRt was fused to all CAR constructs via a T2A cleavage peptide to facilitate staining for CAR expression and sorting.

Zah et al.

Cancer Immunol Res; 4(6) June 2016 Cancer Immunology Research500




CRISPR/Cas9-mediated genome editing of Raji lymphomacells (Supplementary Fig. S3). As expected, the single-input CD19CAR-T cells showedno response toCD19– target cells (Fig. 3B–D).In contrast, T cells expressing OR-gate CARs efficiently lysed bothwild-type (WT; CD19þ/CD20þ) and CD19– target cells (Fig. 3D).

The original OR-gate CAR with a (G4S)1 linker had lowertoxicity against mutant (CD19–/CD20þ) Raji compared withWT Raji, indicating suboptimal CD20 targeting. Increasingthe length and/or rigidity of the linker sequence improvedthe OR-gate CARs' ability to recognize CD20, resulting in

A

B

Parental K562

CD19 K562

CD20K562

Parental K562

CD19 K562

CD20K562

Parental K562

CD19 K562

CD20K562

IFNγ TNFα IL2

pg/m

L19-20 Long 20-19 Long 19-20 Short 20-19 Short Mock CD19 Short CD20 Long

0 500

1,000 1,500 2,000

8,000 14,000

0 2,000 4,000 6,000 8,000

10,000 12,000

0

1,000

2,000

3,000

4,000

5,000

–10

10

30

50

70

1:1 3:1 10:1

19–20 Long 20–19 Long 19–20 Short 20–19 Short

E:T RatioE:T RatioE:T Ratio

% L

ysis

–10

10

30

50

70

1:1 3:1 10:1 –10

10

30

50

70

1:1 3:1 10:1

Parental K562 CD19 K562 CD20 K562

Figure 2.Functional CD19-OR-CD20 CARs can beconstructed by linking scFv domains intandem. A, IFNg , TNFa, and IL2 productionby T cells expressing single-input or OR-gate CARs. Cytokine levels in media weremeasured after a 24-hour coincubationwith K562 target cells. B, target-cell lysisactivity of CD19-OR-CD20 CARs followinga 4-hour coincubation with K562 targetcells. Reported values are the mean oftriplicates, with error bars indicating oneSD. Results are representative of twoindependent experiments performed withCAR T cells derived from two differentdonors.

0

500

1,000

1,500

2,000

0

2,000

4,000

6,000

0

1,000

2,000

3,000 ******

******

******

0

100

200

300

400

0

500

1,000

1,500

2,000

0 500

1,000 1,500 2,000 2,500 3,000

******

******

****

*

Linker = i. (G4S)1 ii. (G4S)4 iii. (EAAAK)1 iv. (EAAAK)3

CD19 scFv CD20 scFv 20–19 Short Linker Hinge CD3ζ EGFRt T2A 4-1BB CD28 tm

Mock CD19 ShortCD20 Long (G4S)1 (G4S)4 (EAAAK)1 (EAAAK)3

A

B

D

C IFNγ IL2TNFα

pg/m

L

MFI

CD69 CD107aCD137

–20 0

20 40 60 80

100

1:1 3:1 10:1 E:T Ratio

% L

ysis

–20

0 20 40 60 80

100

1:1 3:1 10:1 –20 0

20 40 60 80

100

1:1 3:1 10:1

CD20 LongCD19 Short

E:T Ratio

(G4S)1

–20 0

20 40 60 80

100

1:1 3:1 10:1 –20 0

20 40 60 80

100

1:1 3:1 10:1 –20 0

20 40 60 80

100

1:1 3:1 10:1

WT Raji

CD19- Raji

(G4S)4 (EAAAK)3(EAAAK)1

E:T Ratio E:T Ratio E:T Ratio E:T Ratio

Figure 3.Rational structural modifications improve OR-gate CAR activity against mutant tumor cells. A, design of 20-19 short CARs incorporating linkers with increasedlength and/or rigidity. B, CD69, CD137, and CD107a surface expression (MFI) by CAR T cells after a 24-hour coincubation with CD19– Raji cells. C, IFNg , TNFa,and IL2 production by the CAR T cells in B. D, cell lysis by single-input and OR-gate CAR T cells after 4-hour coincubation with WT and CD19– Raji cells.Reported values are themeanof triplicates,with error bars indicatingone SD.P valueswere calculated by two-tailed Student t test; � ,P<0.05; �� ,P<0.01. Data in B–Eare representative of two independent experiments performed with CAR T cells derived from two different donors.






equally efficient elimination of both WT and CD19– Raji targetcells (Fig. 3D).

In addition to enhanced cytotoxicity, modified OR-gate CARsexpressed more activation and degranulation markers, and theyproduced significantly more IFNg , TNFa, and IL2 compared withtheoriginalCARwith a (G4S)1 linker (Fig. 3B andC). TheOR-gateCARwith a (G4S)4 linker showed similar levels of effector outputcompared with the single-input CD20 CAR (Fig. 3B–D). Thus,linker modifications successfully compensated for impairmentsin CD20 targeting imposed by the short extracellular spacer,which was necessary for efficient CD19 targeting.

The increase in CD20 targeting efficiency obtained by linkermodifications did not compromise CD19 targeting capability. All20-19 short CARs, regardless of linker type, showed robust CD69,CD137, andCD107a expressionwhen challengedwithWTRaji orCD19þ/CD20– K562 target cells (Supplementary Fig. S4A). Com-pared with the single-input CD19 CAR, OR-gate CARs triggeredcomparable expression of activation and degranulation markers,as well as IFNg production in response to CD19 stimulation(Supplementary Fig. S4A and S4B). The OR-gate CAR T cellscould lyse CD19þ target cells as efficiently as single-input CD19CAR T cells (Supplementary Fig. S4C). Taken together, these

results demonstrate that OR-gate CARs can efficiently detect andlyse CD19– escape mutants in vitro, and that rational modifica-tions to the CAR structure successfully improved CD20 targetingwhile maintaining high CD19 sensitivity.

Bispecificity does not affect CAR T-cell growth, differentiation,exhaustion profile, or lytic capability in vitro

Despite comparable activation marker expression and lyticcapabilities, CD19 and OR-gate CAR T cells showed disparateIL2 production upon antigen stimulation (Supplementary Fig.S4B). To investigate potential effects of this difference in cytokineproduction, we examined the cell differentiation pattern andproliferation rate of the various CAR T-cell lines during extendedcoincubation with Raji target cells in the absence of exogenouscytokines.

Across multiple donors, we consistently observed a slight butstatistically significant elevation in the central memory (TCM)population among cultured CD19 CAR T cells compared withOR-gate CAR T cells prior to antigen stimulation (Fig. 4A).However, this statistical difference disappeared after 6 days ofcoincubation with WT Raji targets, with no major fluctuationin cell-type distribution (Fig. 4A), indicating that the level of

A

B C

0

20

40

60

80

100

120

140

Mock

CD19Sho

rt

(G4S

)4

(EAAAK)3

n.s. n.s.

n.s.

0

20

40

60

80

100

120

140

Mock

CD19Sho

rt

(G4S

)4

(EAAAK)3

* **

*

6 days post-antigen stimulationPre-antigen stimulation

% T

-cel

l Sub

type

TCM (CD45RA−, CCR7+)

TE/EM (CD45RA−, CCR7−)

TEMRA (CD45RA+, CCR7−)

Tnaïve (CD45RA+, CCR7+)

0

0.5

1

1.5

2

2.5

0 2 4 6 8 0 2 4 6 8

Fold

exp

ansi

on

Days post–target cell addition

Mock CD19 Short CD20 Long

(G4S)4 (EAAAK)3

0

1

2

3

4

5

Fold

exp

ansi

on

Days post–target cell addition

*** **

**

Figure 4.CD19 and OR-gate CAR T cells exhibit similardifferentiation and cell proliferation followingantigen stimulation. A, distribution of T-cellsubtypes prior to coincubation with WT Rajicell targets and 6 days after target-celladdition. Cells were surface-stained forCD45RA and CCR7 expression. B, T-cellproliferation after coincubation with WT Rajicells. CD19 and OR-gate CAR T cells showsimilar proliferation, whereas CD20 CAR Tcells show significantly less proliferationcompared with CD19 CAR T cells (P < 0.01 forall time points except for day 6). C, expansionof mixed CD19 and CD20 CAR T cellsstimulated with WT Raji cells over 8 days.Values shown are the mean of triplicates, witherror bars indicating one SD. P values werecalculated by two-tailed Student t test; n.s.,not statistically significant (P>0.1); � ,P<0.05;�� ,P <0.01. Results are representative of threeindependent experiments using CAR T cellsderived from three different donors.

Zah et al.





antigen-stimulated IL2 production does not have a major impacton cell-type differentiation in vitro. Furthermore, CD19 andOR-gate CAR T cells had similar proliferation rates upon antigenstimulation (Fig. 4B). Thus, the amount of IL2 produced byOR-gate CAR T cells is sufficient to support robust T-cell prolif-eration, an observation that was later confirmed in vivo.

Despite producing similar IL2 concentrations asOR-gate CAR Tcells, CD20 CAR T cells showed significantly weaker proliferationcompared with CD19 CAR T cells (Fig. 4B). When single-inputCD19 andCD20CAR T-cell lines weremixed and cocultured withWTRaji target cells, theCD19CART cells proliferated significantlymore than CD20 CAR T cells, ultimately leading to a net decrease

in the CD20 CAR T-cell population (Fig. 4C). Thus, simultaneousadministration of a mixture of two CAR T-cell lines can result inthe selective expansion of one at the expense of the other. Thisbehavior highlights an important benefit of using a single, bis-pecific CAR T-cell product that can ensure the maintenance ofboth CD19 and CD20 recognition capabilities while avoidinggrowth competition between multiple T-cell products.

We next investigated whether the tumor-lysis capability andexhaustion profile of the OR-gate CARs differ from those of thesingle-input CARs in response to initial antigen stimulation andsubsequent emergence of antigen-escape mutants. At a 2:1 E:Tratio, all CAR T-cell lines eliminated WT Raji cells after 6 days

0

0.5

1

1.5

6 4 2 0 0

0.5

1

1.5

6 4 2 0

Fold

cha

nge

in s

urvi

val

BA

Mock CD19 Short CD20 Long

Fold

cha

nge

in s

urvi

val

(G4S)4 (EAAAK)3

Days post–target cell addition Days post–target cell addition

020406080

100120

CD19Sho

rt

CD20Lo

ng

(G4S

)4

(EAAAK)3

020 40 60 80

100 120

Mock

CD19Sho

rt

CD20Lo

ng

(G4S

)4

(EAAAK)3

No marker One marker Two markers Three markers

C%

Pos

itive

Day 8:

48 h post-stimulationwith CD19-Raji

Day 12: 6 days post-stimulation

with CD19-Raji

Day 2: 48 h post-stimulation

with WT Raji

Pre-antigen stimulation

% P

ositi

ve

020406080

100120

CD19Sho

rt

CD20Lo

ng

(G4S

)4

(EAAAK)3

020406080

100120

Mock

CD19Sho

rt

CD20Lo

ng

(G4S

)4

(EAAAK)3

Figure 5.OR-gate CAR T cells maintain robust lysis capabilitythrough repeated antigen stimulation. Mock-transduced and CAR T cells were coincubated withWT Raji targets for 6 days, and then coincubatedwith CD19– mutant Raji targets for another 6 days.A, survival ofWT Raji cells coincubated with effectorT cells. B, survival of CD19– Raji cells coincubatedwith T cells that were previously challengedwithWTRaji cells. (Note: Mock-transduced T cells had beenoverwhelmed by WT Raji by Day 6 and were notrechallenged with mutant Raji.) C, exhaustionmarker staining of CAR T cells before antigenstimulation, 48 hours after coincubation with WTRaji (Day 2), and 48 hours or 6 days after subsequentcoincubation with CD19– Raji cells (Days 8 and 12,respectively). At each time point, cells weresurface-stained for Lag-3, Tim-3, and PD-1 and thenanalyzed for the simultaneous expression of one,two, or threemarkers. Values shown are themean oftriplicates, with error bars indicating one SD.Results are representative of three independentexperiments using CAR T cells derived from threedifferent donors.






(Fig. 5A), at which point each sample was rechallenged withCD19– mutant Raji cells at a 2:1 E:T ratio. Target-cell countindicated that both OR-gate and CD20 CAR T cells efficientlyeliminated CD19– targets, whereas CD19 CAR T cells failed tocurb the outgrowth of escape mutants (Fig. 5B). Surface antibodystaining revealed consistent upregulation of PD-1, Tim-3, andLag-3 upon antigen stimulation across all CAR constructs (Fig. 5C;Supplementary Fig. S5). Upon subsequent challenge with CD19–

target cells, exhaustion marker expression was sustained in OR-gate and CD20 CAR T cells but declined in CD19 CAR T cells (Fig.5C; Supplementary Fig. S5). Taken together, these results indicatethat OR-gate CAR T cells retain robust effector function uponrepeated antigen presentation, and they exhibit equivalent lysiscapabilities and exhaustion marker expression patterns as CD20CAR T cells in response to CD19– escape mutants.

OR-gate CARs prevent tumor antigen escape in vivoTo determine the in vivo functionality of OR-gate CARs, NOD/

SCID/gc�/� (NSG) mice were injected with either a pure popu-

lation ofWTRaji cells or a 3:1mixture ofWT andCD19–Raji cells.Mice bearing established xenografts were then treated with CD8þ

CAR T cells. As expected, single-input CD19 CAR T cells signif-icantly extended the survival of animals engrafted with WT Raji

cells but failed to control the mixed Raji population (Fig. 6;Supplementary Fig. S6A). Postmortem analysis revealed the out-growth of CD19–mutants in themixed Raji xenograft, confirmingantigen escape from single-input CAR T-cell therapy (Supplemen-tary Fig. S6B).

In contrast with the single-input CD19 CAR, OR-gate CARscould target WT and mixed Raji tumors with equal efficiency(Fig. 6; Supplementary Fig. S6C), demonstrating that OR-gateCAR T cells were unaffected by the loss of CD19 expression ontumor cells and only require a single antigen to trigger robustantitumor functions. The OR-gate CARs were as effective as thesingle-input CD19 CAR in controlling the growth of WT Raji cells(Supplementary Fig. S6D), confirming that the broadening ofantigen specificity did not compromise in vivo antitumor efficacyagainst CD19þ targets. Taken together, these results demonstratethat OR-gate CARs can efficiently target malignant B cells andabrogate the effects of tumor antigen loss in vivo.

Although OR-gate CARs were unaffected by the loss of CD19antigen, none of the CAR T-cell lines—including second-gener-ation CD19 CAR T cells challenged with WT Raji cells—couldcompletely eradicate engrafted tumors under the conditionstested. Analysis of retro-orbital blood samples obtained 3 daysafter T-cell injection confirmed the presence of CAR T cells

A

B

n.s.**

**

%S

urvi

val

Days after tumor inoculation

n.s.

CD19 Short + Mixed Raji

(G4S)4 + Mixed Raji

(EAAAK)3 +Mixed Raji

(EAAAK)3 +WT Raji

d6

d18

d21

Min = 1.5e5Max = 1.0e8

Min = 2.0e7Max = 2.0e8

Min = 2.0e7Max = 2.0e8

Figure 6.OR-gate CARs abrogate the effects of antigen escape in vivo. A, tumor progression in NSG mice bearing WT or mixed (75% WT, 25% CD19–) Raji xenografts.Bioluminescence imaging was performed on days 6, 18, and 21 after tumor injection (T cells were injected on day 7). B, survival of mice bearing mixed Rajitumor xenografts and treated with T cells expressing no CAR, the single-input CD19 CAR, or OR-gate CARs. n ¼ 5 in all test groups. P values were calculated bylog-rank test analysis; n.s., not statistically significant (P > 0.1); � , P < 0.1; �� , P < 0.05. Results represent one independent trial.

Zah et al.





(Fig. 7A). Both OR-gate CAR T-cell lines showed expansion in vivo(Supplementary Fig. S6E). Finally, CAR-expressing T cells couldbe found in the bonemarrow of all animals at the time of sacrifice(Fig. 7B), and all Raji cells retained either CD19 expression orCD19/CD20 dual expression (Supplementary Fig. S7). Therefore,inability to clear the tumor burdenwas not due to absence of CART-cell proliferation or the loss of both targeted antigens.

We next investigated the cell-type differentiation and exhaus-tionmarker expression patterns of theCART cells. Results indicatea significant increase in PD-1 expression but little change in cell-type distribution after in vivo antigen exposure (Fig. 7C and D).PD-1 expression was significantly higher for CD19 CAR T cellscompared with OR-gate CAR T cells (Fig. 7C). However, PD-1 isknown to be upregulated upon T-cell activation, and prolongedPD-1 does not necessarily indicate lack of effector function inCAR T cells (24). Given the proven clinical efficacy of the second-generation CD19 CAR, the inability to achieve tumor clearancehere is likely a result of the aggressive tumor line and specifictumor and T-cell dosages evaluated in this animal study.

DiscussionCD19 CAR T-cell therapy has yielded remarkable clinical

outcomes in the treatment of acute and chronic B-cell malig-nancies (3–9). However, this treatment strategy's vulnerability toantigen escape has been highlighted by multiple cases of relapse

resulting from the emergence of CD19– tumor cells (7, 10). Toarm CD19 CAR T cells against antigen escape, we constructednovel CARs capable of OR-gate signal processing—i.e., receptorsthat can trigger robust T-cell responses as long as the target cellsexpress either CD19 or CD20. Results of our in vitro and in vivocharacterization experiments indicate that the CD19-OR-CD20CARs can safeguard against the effects of antigen escape bytargeting malignant B cells through CD20 when CD19 expres-sion has been lost.

The choice of CD19 and CD20 as the target-antigen pairprovides several important advantages. First, CD19 and CD20are both clinically validated B-cell antigens expressed on the vastmajority of malignant B cells (10, 25, 26), making the CD19-OR-CD20 CAR a clinically applicable construct that addresses a realchallenge facing adoptive T-cell therapy. Second, efforts to broad-en the recognition capability of CAR T cells are oftenmet with theundesirable side effect of increased on-target, off-tumor toxicity.However, this trade-off does not apply to the case of CD19 andCD20, because both are exclusively expressed on B cells andhave the same off-tumor toxicity profile. Finally, the ubiquitousexpression of CD19 and CD20 on B cells and their known andpredicted roles in promoting B-cell survival (27, 28) suggest thatsimultaneous loss of both antigens would be a very low-prob-ability event. Therefore, targeting CD19 and CD20 is expectedto provide an effective safeguard against antigen escape bymalignant B cells.

**

0

20

40

60

80

100

120

CD19Sho

rt

(G4S

)4

(EAAAK)3

0

20

40

60

80

100

120

CD19Sho

rt

(G4S

)4

(EAAAK)3

0

20

40

60

80

100

120

CD19Sho

rt

(G4S

)4

(EAAAK)3

BA C

D

% T

-cel

l Sub

type

CD19Sho

rt

(G4S

)4

(EAAAK)3

CD19Sho

rt

(G4S

)4

(EAAAK)3

CD19Sho

rt

(G4S

)4

(EAAAK)3

CD19Sho

rt

(G4S

)4

(EAAAK)3

CD19Sho

rt

(G4S

)4

(EAAAK)3

CD19Sho

rt

(G4S

)4

(EAAAK)3

% C

D8+

EG

FRt+

WT Raji

WT Raji Preinjection

Mixed Raji

Mixed Raji Postmortem

% C

D8+

EG

FRt+

% P

D-1

+

Preinjection WT Raji Mixed RajiPostmortem

*

**

*

** *

***

*

*

TCM (CD45RA−, CCR7+)

TE/EM (CD45RA−, CCR7−)

TEMRA (CD45RA+, CCR7−)

Tnaïve (CD45RA+, CCR7+)

0

0.1

0.2

0.3

0.4

0.5

0 05

1015202530

0 0

20

40

60

80

100

0

20

40

60

80

100

51015202530

0.1

0.2

0.3

0.4

0.5

Figure 7.CD19 and OR-gate CAR T cells persist, upregulate PD-1, and exhibit similar differentiation subtypes in vivo. A, presence of CAR T cells in peripheral blood3 days after T-cell injection. Cells from retro-orbital blood were stained for the expression of human CD8 and EGFRt, which was cotranslated with each CAR asa T2A fusion. B, percentage of CAR T cells present in the bone marrow collected at the time of sacrifice. C, PD-1 expression of CAR T cells in the bone marrowof animals bearing WT Raji tumors. D, preinjection and postmortem T-cell subtype distributions. "WT Raji" and "Mixed Raji" labels indicate the type oftumor engrafted in the animal from which the T cells were harvested. P values were calculated by two-tailed Student t test; n.s., not significant (P > 0.1);� , P < 0.05; �� , P < 0.01.






In principle, multiple receptor configurations can be adoptedto achieve bispecific signal computation, such as coexpressingtwo different CARs in one T cell (29) or mixing two CAR T-celllines, each targeting a different antigen (30). Instead, we choseto engineer dual-antigen recognition capability into a singleCAR molecule due to a number of advantages. First, comparedwith expressing two separate single-input CARs in one T cell,the bispecific CAR has a significantly smaller DNA footprint(reduced by �40% in DNA length), and previous studies haveshown that construct size significantly affects viral vector pack-aging and transduction efficiency (31, 32). Most clinical T-cellproducts are administered as a polyclonal population withoutprior sorting for CARþ cells (3, 5, 6), thus the ability to achievehigh transduction efficiency has a direct impact on clinicalefficacy. Genetic compactness becomes particularly criticalwhen features such as suicide genes (33) and additional sig-nal-processing units (e.g., inhibitory receptors; ref. 34) alsoneed to be integrated into the CAR T cells to ensure safety orenhance efficacy. Second, compared with mixing two differentsingle-input CAR T-cell lines, the ability to produce and admin-ister a single, bispecific CAR T-cell product significantly reducestreatment costs and increases the probability of successful T-cellproduction within a short clinical timeframe. Third, the CD19CAR has outperformed all other CARs evaluated in the clinic todate, including the CD20 CAR (35, 36). Our data demonstratethat in a coculture, CD19 CAR T cells have a significant growthadvantage over CD20 CAR T cells, resulting in a net decline inCD20 CAR T-cell count despite the presence of CD20 antigen(Fig. 4C). Therefore, it is probable that coadministering twosingle-input CAR T-cell populations would result in the dis-proportionate expansion of CD19 CAR T cells at the expense ofCD20 CAR T cells, thereby compromising this strategy's abilityto safeguard against CD19– mutants when they emerge later inthe treatment period. By making each T cell capable of bispe-cific antigen recognition, the OR-gate CAR design maximizesthe number of T cells that can recognize an escape mutant whenit appears. For these reasons, we chose to engineer a single CARmolecule capable of dual-antigen recognition by attaching twotandem scFv domains to the standard CAR chassis. Our dataindicate that T cells expressing OR-gate CARs are indeed insen-sitive to the loss of CD19 on target cells, proliferate robustly inresponse to either CD19 or CD20 stimulation, do not exhibitaltered cell-type differentiation patterns compared with single-input CAR T cells, and retain robust target-cell lysis capabilityupon repeated antigen stimulation.

CARs are frequently thought of as modular proteins withsensor domains (i.e., the scFv) that can be easily changed toalter receptor specificity. However, results from a previous report(21) and our own investigation (Supplementary Fig. S1)revealed that efficient CD19 and CD20 targeting by single-inputCARs requires distinct receptor structures. These divergent pre-ferences may be a consequence of structural differences betweenthe two antigens: CD19 is an immunoglobulin-like moleculethat belongs to a family of single-pass transmembrane proteinsthat project outward from the cell membrane (37); CD20 is amultipass transmembrane protein that lies close to the cellsurface (38). To achieve the optimal conjugation distancebetween a T cell and its target, the receptor needs to be adjustedto match the size of the target antigen, resulting in the need for ashorter CAR when targeting antigens with extensive extracellulardomains and vice versa. Although this hypothesis is consistent

with our observations, quantitative imaging studies would berequired to confirm its accuracy.

Based on our understanding of the structural requirementsfor productive CAR/antigen interactions, we arrived at theoptimal structure for a bispecific CAR through systematic,rational design. By adjusting the extracellular spacer lengthand linker sequence between the two scFv domains, we wereable to independently optimize CD19 and CD20 targetingwithout compromising the CAR's ability to recognize eitherantigen. In vitro and in vivo results demonstrate that the OR-gateCARs are as efficient in CD19 targeting as the clinically suc-cessful single-input CD19 CAR, and the OR-gate CARs succeedin controlling CD19– mutant lymphoma cells where the single-input CAR fails.

A potential concern with the design of bispecific CARs is theincreased number of peptide fusion points in the receptorprotein. However, almost all of the components used in theOR-gate CARs presented in this study have been tested in theclinic, including the two scFv domains, the long and shortextracellular spacers, as well as the transmembrane and cyto-plasmic signaling domains (3–5, 7, 8, 39). Long, flexible linkerpeptides consisting of glycine and serine residues have beenused in fusion peptides such as rVIIa-FP, which is now beingevaluated in a clinical trial (40, 41). It remains possible thatrigid peptide linkers or new combinations of previously testedpeptide components could result in immunogenicity. However,available evidence suggests that the probability is low and willhave to be verified through extensive testing in preclinical andclinical studies.

Several studies have explored the contribution of individualCAR components to T-cell functionality, with results suggestingthat the scFv framework regions, extracellular spacer length, andcostimulatory signals all play important and nonobvious roles inenabling robust T-cell–mediated response to tumor antigens(21, 36, 42–45). It remains possible that additionalmodificationsto each of these domains could further increase theOR-gate CAR'sfunctionality. Increasingly detailed mechanistic understanding ofCAR signaling and the systematic incorporation of such knowl-edge into the rational design of next-generation CAR moleculeswill continue to facilitate the development of more effective andpredictable cell-based therapeutics.

Adoptive T-cell therapy has been hailed as one of the mostpromising advancements in cancer therapy in recent years, andCD19 CAR T-cell therapy holds the spotlight as the mostsuccessful adoptive T-cell therapy to date. The work reportedhere presents an effective and clinically applicable solution tothe challenge of antigen escape, which has been observed inmultiple clinical trials of CD19 CAR T-cell. The CD19-OR-CD20 CAR is fully compatible with current T-cell manufactur-ing processes, does not impose extra burden in the form of largeviral packaging and transduction payloads, and enables a singleT-cell product to target two clinically validated antigens asso-ciated with B-cell leukemia and lymphoma. Finally, the designprinciples highlighted in this study can be further utilized toconstruct novel CARs for additional antigen pairs to broadenthe applicability and increase the efficacy of T-cell therapy forcancer.

Disclosure of Potential Conflicts of InterestM.C. Jensen reports serving as a scientific advisory board member and as a

consultant for Juno Therapeutics, Inc., of which he is a scientific cofounder with


Zah et al.




ownership interest (including patents). No potential conflicts of interest weredisclosed by the other authors.

Authors' ContributionsConception and design: M.C. Jensen, Y.Y. ChenDevelopment of methodology: E. Zah, A. Silva-Benedict, M.C. Jensen, Y.Y. ChenAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): E. Zah, M.-Y. Lin, Y.Y. ChenAnalysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): E. Zah, M.C. Jensen, Y.Y. ChenWriting, review, and/or revision of the manuscript: E. Zah, M.C. Jensen,Y.Y. ChenAdministrative, technical, or material support (i.e., reporting or organizingdata, constructing databases): M.C. Jensen, Y.Y. ChenStudy supervision: Y.Y. Chen

AcknowledgmentsThe authors thank Lisa Rolczynski for her technical assistance.

Grant SupportThis research was funded by the NIH (5DP5OD012133, grant to Y.Y. Chen;

2R01CA136551-06A1, grant to M.C. Jensen).The costs of publication of this article were defrayed in part by the

payment of page charges. This article must therefore be hereby markedadvertisement in accordance with 18 U.S.C. Section 1734 solely to indicatethis fact.

Received September 10, 2015; revised February 19, 2016; accepted March 6,2016; published OnlineFirst April 8, 2016.

References1. KalosM, JuneCH. Adoptive T cell transfer for cancer immunotherapy in the

era of synthetic biology. Immunity 2013;39:49–60.2. Rosenberg SA, Restifo NP. Adoptive cell transfer as personalized immu-

notherapy for human cancer. Science 2015;348:62–8.3. Kalos M, Levine BL, Porter DL, Katz S, Grupp SA, Bagg A, et al. T cells with

chimeric antigen receptors have potent antitumor effects and can establishmemory in patients with advanced leukemia. Sci Transl Med 2011;3:95ra73.

4. Porter DL, Levine BL, KalosM, Bagg A, June CH. Chimeric antigen receptor-modified T cells in chronic lymphoid leukemia. N Engl J Med 2011;365:725–33.

5. Brentjens RJ, Davila ML, Riviere I, Park J, Wang X, Cowell LG, et al. CD19-targeted T cells rapidly induce molecular remissions in adults with che-motherapy-refractory acute lymphoblastic leukemia. Sci Transl Med 2013;5:177ra38.

6. Kochenderfer JN,DudleyME, Feldman SA,WilsonWH, SpanerDE,Maric I,et al. B-cell depletion and remissions of malignancy along with cytokine-associated toxicity in a clinical trial of anti-CD19 chimeric-antigen-recep-tor-transduced T cells. Blood 2012;119:2709–20.

7. Grupp SA, Kalos M, Barrett D, Aplenc R, Porter DL, Rheingold SR, et al.Chimeric antigen receptor-modified T cells for acute lymphoid leukemia.N Engl J Med 2013;368:1509–18.

8. Davila ML, Riviere I, Wang X, Bartido S, Park J, Curran K, et al. Efficacy andtoxicity management of 19–28z CAR T cell therapy in B cell acute lym-phoblastic leukemia. Sci Transl Med 2014;6:224ra25.

9. Kochenderfer JN, Dudley ME, Kassim SH, Somerville RP, Carpenter RO,Stetler-Stevenson M, et al. Chemotherapy-refractory diffuse large B-celllymphoma and indolent B-cell malignancies can be effectively treated withautologous T cells expressing an anti-CD19 chimeric antigen receptor.J Clin Oncol 2015;33:540–9.

10. Maude SL, FreyN, ShawPA, Aplenc R, Barrett DM, BuninNJ, et al. Chimericantigen receptor T cells for sustained remissions in leukemia. N Engl J Med2014;371:1507–17.

11. Rapoport AP, Stadtmauer EA, Binder-Scholl GK, Goloubeva O, Vogl DT,Lacey SF, et al. NY-ESO-1-specific TCR-engineered T cells mediatesustained antigen-specific antitumor effects in myeloma. Nat Med2015;21:914–21.

12. Jager E, Ringhoffer M, Karbach J, Arand M, Oesch F, Knuth A. Inverserelationship of melanocyte differentiation antigen expression in melano-ma tissues and CD8þ cytotoxic-T-cell responses: evidence for immuno-selection of antigen-loss variants in vivo. Int J Cancer 1996;66:470–6.

13. Riker A, Cormier J, Panelli M, Kammula U, Wang E, Abati A, et al. Immuneselection after antigen-specific immunotherapy of melanoma. Surgery1999;126:112–20.

14. GradaZ,HegdeM,Byrd T, ShafferDR,Ghazi A, BrawleyVS, et al. TanCAR:Anovel bispecific chimeric antigen receptor for cancer immunotherapy. MolTher Nucleic Acids 2013;2:e105.

15. GibsonDG, Young L, Chuang RY, Venter JC,Hutchison CA 3rd, SmithHO.Enzymatic assembly of DNA molecules up to several hundred kilobases.Nat Methods 2009;6:343–5.

16. Nicholson IC, Lenton KA, Little DJ, Decorso T, Lee FT, Scott AM, et al.Construction and characterisation of a functional CD19 specific single

chain Fv fragment for immunotherapy of B lineage leukaemia and lym-phoma. Mol Immunol 1997;34:1157–65.

17. Jensen M, Tan G, Forman S, Wu AM, Raubitschek A. CD20 is a moleculartarget for scFvFc:zeta receptor redirected T cells: implications for cellularimmunotherapy of CD20þ malignancy. Biol Blood Marrow Transplant1998;4:75–83.

18. Wang X, ChangWC,Wong CW, Colcher D, ShermanM,Ostberg JR, et al. Atransgene-encoded cell surface polypeptide for selection, in vivo tracking,and ablation of engineered cells. Blood 2011;118:1255–63.

19. Pelloquin F, Lamelin JP, Lenoir GM. Human B lymphocytes immortali-zation by Epstein-Barr virus in the presence of cyclosporin A. In Vitro CellDev Biol 1986;22:689–94.

20. Riddell SR,Greenberg PD. Theuseof anti-CD3 and anti-CD28monoclonalantibodies to clone and expand human antigen-specific T cells. J ImmunolMethods 1990;128:189–201.

21. HudecekM, Sommermeyer D, Kosasih PL, Silva-Benedict A, Liu L, Rader C,et al. The nonsignaling extracellular spacer domain of chimeric antigenreceptors is decisive for in vivo antitumor activity. Cancer Immunol Res2015;3:125–35.

22. Arai R, Ueda H, Kitayama A, Kamiya N, Nagamune T. Design of the linkerswhich effectively separate domains of a bifunctional fusion protein.Protein Eng 2001;14:529–32.

23. Arai R, WriggersW, Nishikawa Y, Nagamune T, Fujisawa T. Conformationsof variably linked chimeric proteins evaluated by synchrotron X-ray small-angle scattering. Proteins 2004;57:829–38.

24. Chang ZL, Silver PA, Chen YY. Identification and selective expansion offunctionally superior T cells expressing chimeric antigen receptors. J TranslMed 2015;13:161.

25. Boye J, Elter T, Engert A. An overview of the current clinical use of the anti-CD20 monoclonal antibody rituximab. Ann Oncol 2003;14:520–35.

26. Till BG, Jensen MC, Wang J, Qian X, Gopal AK, Maloney DG, et al. CD20-specific adoptive immunotherapy for lymphoma using a chimeric antigenreceptor with both CD28 and 4-1BB domains: pilot clinical trial results.Blood 2012;119:3940–50.

27. Otero DC, Anzelon AN, Rickert RC. CD19 function in early and late Bcell development: I. Maintenance of follicular and marginal zone Bcells requires CD19-dependent survival signals. J Immunol 2003;170:73–83.

28. Tedder TF, Engel P. CD20: A regulator of cell-cycle progression of Blymphocytes. Immunol Today 1994;15:450–4.

29. Hegde M, Corder A, Chow KK, Mukherjee M, Ashoori A, Kew Y, et al.Combinational targeting offsets antigen escape and enhances effectorfunctions of adoptively transferred T cells in glioblastoma. Mol Ther2013;21:2087–101.

30. AnurathapanU,ChanRC,HindiHF,Mucharla R, Bajgain P,Hayes BC, et al.Kinetics of tumor destruction by chimeric antigen receptor-modified Tcells. Mol Ther 2014;22:623–33.

31. Bos TJ, De Bruyne E, Van Lint S, Heirman C, Vanderkerken K. Large doublecopy vectors are functional but show a size-dependent decline in trans-duction efficiency. J Biotechnol 2010;150:37–40.

32. KumarM, Keller B,MakalouN, Sutton RE. Systematic determination of thepackaging limit of lentiviral vectors. Hum Gene Ther 2001;12:1893–905.






33. Zhou X, Dotti G, Krance RA, Martinez CA, Naik S, Kamble RT, et al.Inducible caspase-9 suicide gene controls adverse effects from allorepleteT cells after haploidentical stem cell transplantation. Blood 2015;125:4103–13.

34. Fedorov VD, Themeli M, Sadelain M. PD-1- and CTLA-4-based inhibitorychimeric antigen receptors (iCARs) divert off-target immunotherapyresponses. Sci Transl Med 2013;5:215ra172.

35. Gill S, June CH. Going viral: Chimeric antigen receptor T-cell therapy forhematological malignancies. Immunol Rev 2015;263:68–89.

36. Long AH, Haso WM, Shern JF, Wanhainen KM, Murgai M, Ingaramo M,et al. 4-1BB costimulation ameliorates T cell exhaustion induced by tonicsignaling of chimeric antigen receptors. Nat Med 2015;21:581–90.

37. Tedder TF, Isaacs CM. Isolation of cDNAs encoding the CD19 antigen ofhuman andmouse B lymphocytes. A newmember of the immunoglobulinsuperfamily. J Immunol 1989;143:712–7.

38. Einfeld DA, Brown JP, Valentine MA, Clark EA, Ledbetter JA. Molecularcloning of the human B cell CD20 receptor predicts a hydrophobic proteinwith multiple transmembrane domains. EMBO J 1988;7:711–7.

39. Till BG, Jensen MC, Wang J, Chen EY, Wood BL, Greisman HA, et al.Adoptive immunotherapy for indolent non-Hodgkin lymphoma andmantle cell lymphoma using genetically modified autologous CD20-specific T cells. Blood 2008;112:2261–71.

40. Weimer T, Wormsbacher W, Kronthaler U, Lang W, Liebing U, Schulte S.Prolonged in-vivo half-life of factor VIIa by fusion to albumin. ThrombHaemost 2008;99:659–67.

41. Zollner S, Schuermann D, Raquet E, Mueller-Cohrs J, Weimer T, Pragst I,et al. Pharmacological characteristics of a novel, recombinant fusionprotein linking coagulation factor VIIa with albumin (rVIIa-FP). J ThrombHaemost 2014;12:220–8.

42. Guest RD, Hawkins RE, Kirillova N, Cheadle EJ, Arnold J, O'Neill A, et al.The role of extracellular spacer regions in the optimal design of chimericimmune receptors: evaluation of four different scFvs and antigens.J Immunother 2005;28:203–11.

43. Hudecek M, Lupo-Stanghellini MT, Kosasih PL, Sommermeyer D, JensenMC, Rader C, et al. Receptor affinity and extracellular domain modifica-tions affect tumor recognition by ROR1-specific chimeric antigen receptorT cells. Clin Cancer Res 2013;19:3153–64.

44. Frigault MJ, Lee J, Basil MC, Carpenito C, Motohashi S, Scholler J, et al.Identification of chimeric antigen receptors that mediate constitutive orinducible proliferation of T cells. Cancer Immunol Res 2015;3:356–67.

45. Kunkele A, Johnson AJ, Rolczynski LS, Chang CA, Hoglund V, Kelly-SprattKS, et al. Functional tuning of CARs reveals signaling threshold abovewhich CD8þ CTL antitumor potency is attenuated due to cell fas-FasL-Dependent AICD. Cancer Immunol Res 2015;3:368–79.


Zah et al.




Addendum

ADDENDUM: T Cells Expressing CD19/CD20Bispecific Chimeric Antigen Receptors PreventAntigen Escape by Malignant B CellsEugenia Zah, Meng-Yin Lin, Anne Silva-Benedict, Michael C. Jensen, andYvonne Y. Chen

In this article (Cancer Immunol Res 2016;4:498–508), whichappeared in the June 2016 issue ofCancer Immunology Research, wereported the design and optimization of bispecific, OR-gatechimeric antigen receptors (CARs) that can trigger robust T-cellactivation in response to target cells that present either CD19 orCD20 (1), thus preventing malignant B cells from escaping T-celltherapy through loss of CD19 expression. In our original study,although OR-gate CAR T cells could significantly delay tumorexpansion and prolonged survival, tumors were not eradicated.Tumor persistence was not due to the loss of both CD19 andCD20 antigens, nor due to the inability of the adoptively trans-ferred T cells to survive or expand in vivo (1). We concluded thatthe failure to eradicate tumors was due to suboptimal tumor or T-cell dosing and/or the aggressiveness of the Raji tumor model.

As the Raji tumor model had been used in other studies atdosages comparable with those employed in our own experi-ments (2), complete tumor clearance should be possible in thisxenograft model using adoptively transferred CAR T cells. Here,we report follow-up studies showing that the viability levels ofCAR T cells, both immediately after thawing and after 1–2 days ofin vitro culture, are an important indicator of the CAR T cells'antitumor capability in vivo. Healthy T cells with OR-gate CARscould efficiently eradicate established lymphoma xenografts evenif the lymphoma cells had spontaneously lost CD19 expression,whereas single-input CD19CART cells succumbed to the selectiveexpansion of CD19� mutants.

Stability of CAR T-cell viability varies across donorsThe in vivo protocol employed in our study called for the

tail-vein injection of luciferase-expressing Raji tumor cells, fol-lowed by the tail-vein injection of CAR T cells upon confirmationof tumor establishment, which was defined as clear tumor signalobserved on two consecutive days via bioluminescence imaging(BLI). The CAR T cells used in these studies were previously frozenand thawed on the day of injection, and cells with >70% viabilitywere considered suitable for adoptive transfer. To better charac-terize the CAR T cells (from donor A) used in our initial animalstudy reported in (1), we thawed additional stocks of the samedonor's cells (donor A) and compared them with CAR T cellsderived from other healthy donor blood samples (donors B andC) for cell viability, T-cell subtype distribution, and exhaustionmarker expression. T-cell subtype distribution, exhaustionmarkerexpression, and viability at the time of thawingwere similar for alldonors (Fig. 1). However, cell viability of CAR T cells fromdonorsA and B drastically declined after 24 hours in culture, whereasdonor C cells remained >70% viable, indicating that donor A cellsmay not have been of optimal quality (Fig. 1C). Although we had

confirmed that donor A CAR T cells could survive and expandin vivo (1), they may not have proliferated well enough tocompletely eliminate the tumor xenograft.

CAR T cells with high post-thaw viability efficiently eradicateestablished tumor xenografts

To investigate this hypothesis, we repeated the in vivo studyusing donor C cells, which showed superior viability in cultureafter thawing (Fig. 1C). As in the previous study, 5 � 105 Rajicells were delivered by tail-vein injection and allowed to engraftprior to treatment with 10 � 106 CAR T cells from donor C.Animals were injected with either purely wild-type (WT;CD19þ/CD20þ) Raji cells or a mixture of 75% WT and 25%CD19�-mutant Raji tumors. In contrast to the results obtainedwith donor A cells, we observed complete clearance of WTtumors treated with both single-input and OR-gate CAR T cells(Fig. 2A and B). Only OR-gate CAR T cells, however, coulderadicate mixed (75% WT, 25% CD19�) Raji tumors (Fig. 2Aand B). Animals engrafted with mixed Raji cells and treated withsingle-input CD19 CAR T cells eventually succumbed to tumorgrowths that consisted almost exclusively of the CD19�-mutantphenotype (Fig. 2C), confirming the selective expansion ofantigen-negative clones when treated with single-input CAR Tcells. Thus, unlike single-input CD19 CAR T cells, OR-gate CAR Tcells limited the escape of CD19�-mutant tumors.

OR-gate CARs prevent the spontaneous emergence of CD19�

mutant tumorsHaving confirmed the ability to eradicate both WT and

CD19-knockout tumors using high-quality CAR T cells, we nextinvestigated whether OR-gate CAR T cells would have superiorresistance compared with single-input CD19 CAR T cells againstthe spontaneous loss of CD19 expression by tumor cells. Specif-ically, we investigated whether antigen loss is dependent on thetiming of CAR T-cell treatment relative to the size of tumorburden, and whether the OR-gate CAR T cells could safeguardagainst the spontaneous emergence of CD19� mutants.

Mice were injected with 5 � 105 WT Raji cells, which wereallowed to grow either until first confirmation of engraftment(i.e., yielding clear tumor signal observed on 2 consecutivedays via BLI, which occurred on day 6 in this set of studies), oruntil the tumor signal reached a 6-fold increase in radianceintensity compared with the first group (which occurred on day9). Animals were then treated with 10 � 106 donor C T cellsexpressing either the single-input CD19 CAR or the OR-gateCAR. Consistent with previous results, animals treated at theearlier time point with either single-input or OR-gate CART cells were cleared of tumor engraftment within 6 days ofT-cell injection (Fig. 3A). In contrast, all animals treated withCAR T cells on day 9 after tumor injection showed continuedtumor progression for 3 days before tumor size began to

doi: 10.1158/2326-6066.CIR-16-0108

�2016 American Association for Cancer Research.

CancerImmunologyResearch

www.aacrjournals.org 639

0

20

40

60

80

100

120

Donor Donor Donor

% o

f Ove

rall

popu

latio

n

B A C

A

TCM (CD45RA-, CCR7+)

TE/EM (CD45RA-, CCR7-)

B

C

% P

ositi

ve

0

20

40

60

80

100

120

Donor Donor Donor B A C

TEMRA (CD45RA+, CCR7-)

Tnaive (CD45RA+, CCR7+)

No marker

One marker

Two markers

Three markers

% V

iabl

e

0 10 20 30 40 50 60 70 80 90

48 24 0

Donor A

Donor B

Donor C

Hours post thaw

Figure 1.

Comparison of CAR T-cell preparations from multiple donors. A, T-cell subtype distribution patterns. B, exhaustion markers. Cells were stained for Tim-3,Lag-3, and PD-1. Values shown in A and B are means of three technical replicates �1 SD. C, viability changes during cell culture. Data shown are representativeof at least two independent experiments performed on the cells from each of the three donors.

A

B

Day 5

Day 8

Day 11

WT Raji Mixed Raji WT Raji Mixed Raji WT Raji Mixed Raji Mock OR-gate CAR CD19 CAR

Ave

rage

radi

ance

Day post tumor injection

C

%C

D19

− R

aji/t

otal

Raj

i

Ratio at time of tumor injection

Mock

CD19 C

AR 0

25

50

75

100

Figure 2.

Wild-type (WT) and CD19�-mutant lymphoma growth in vivo after injection of OR-gate CAR or single-input CD19 CAR T cells. A, tumor progression inNSG mice bearing WT or mixed (75% WT, 25% CD19�) Raji xenografts. T cells were injected 5 days after tumor injection. B, radiance intensity of tumors engraftedin NSG mice; n ¼ 5 in all test groups. Reported values are the means of all surviving animals in each test group, with error bars indicating �1 SD. C, femoralbone marrow of mice bearing mixed Raji tumors and treated with mock-transduced or single-input CD19 CAR T cells analyzed by flow cytometry. Raji cellswere identifiedby the expressionof EGFP,whichwas stably integrated into bothRaji cell lines, andCD19 stainingwasperformed todistinguishWTandCD19�-mutantRaji populations. Results represent one independent trial.

Cancer Immunol Res; 4(7) July 2016 Cancer Immunology Research640

Zah et al.

decline (Fig. 3B). Animals treated with OR-gate CAR T cellscompletely cleared their tumors and showed no sign of relapseat the time of this writing (through day 109). In contrast,animals treated with single-input CD19 CAR T cells experi-enced only temporary tumor regression before relapsing, ulti-mately succumbing to multifocal cancer growth (Fig. 3B).Postmortem analysis revealed tumor engraftments in the liverand the chest cavity (Fig. 3C). The recovered Raji tumors,which were identified by their luciferase and EGFP expression(Fig. 3C), were CD20þ but had either no CD19 expression orsignificantly reduced CD19 expression (Fig. 3D), highlighting

the risk of antigen escape with single-input CD19 CAR T cellsand the OR-gate CAR T cells' ability to safeguard against thistumor defense mechanism.

Patient relapse due to loss of CD19 antigen expression has beenobserved in multiple clinical trials, and antigen escape has beenidentified as a potential pitfall of CD19-directed therapies for B-cellmalignancies (3, 4). The CD19/CD20 bispecific, OR-gate CARdemonstrates the ability to address this critical medical challengeand increase the efficacy of adoptive T-cell therapy for cancer.

Published online July 1, 2016.

References1. Zah E, Lin MY, Silva-Benedict A, Jensen MC, Chen YY. T cells expressing

CD19/CD20 bi-specific chimeric antigen receptors prevent antigen escapeby malignant B cells. Cancer Immunol Res 2016;4:498–508.

2. Hudecek M, Sommermeyer D, Kosasih PL, Silva-Benedict A, Liu L, Rader C,et al. The nonsignaling extracellular spacer domain of chimeric antigenreceptors is decisive for invivo antitumor activity. Cancer Immunol Res2015;3:125–35.

3. Lee DW, Kochenderfer JN, Stetler-Stevenson M, Cui YK, Delbrook C, Feld-man SA, et al. T cells expressing CD19 chimeric antigen receptors for acutelymphoblastic leukaemia in children and young adults: a phase 1 dose-escalation trial. Lancet 2015;385:517–28.

4. Maude SL, Frey N, Shaw PA, Aplenc R, Barrett DM, Bunin NJ, et al. Chimericantigen receptor T cells for sustained remissions in leukemia. N Engl J Med2014;371:1507–17.

AAv

erag

e ra

dian

ce


CD19 CAR

OR-gate CAR

9 12 24 34


CD19 CAR

OR-gate CAR

12 9 6 Day post tumor injection

B

Day post tumor injection A

vera

ge ra

dian

ce

C

Chest

Liver

EG

FP

ffLuc

D

CD

19

CD20

Chest Unstained Stained

39 61

Min: 1E7 Max: 2E8 Scaling: Min: 1E5 Max: 2E6

Liver

Pre-injection Tumor

97.0 24.5

68.5

25.9

73.1

Min: 2E4 Max: 1E5

1.0E+03

1.0E+04

1.0E+05

CD19 CAR

OR-gate CAR

30 20 10 –1 40

T-cellinjec�on T-cell

injec�on

Figure 3.

Prevention of spontaneous CD19�-mutant tumor outgrowth by OR-gate CAR T cells. A and B, NSG mice bearing WT Raji xenografts were treated with CAR T cellseither (A) at the first sign of tumor engraftment (day 6 after tumor injection in this trial) or (B) after the tumor had expanded by 6 folds (day 9 after tumor injection).C, Raji tumor nodules were recovered from relapsed animals and their identity was verified by EGFP and firefly luciferase (ffLuc) expression via IVIS imaging.D, tumor nodule recovery fromanimals treatedwith CD19 CART cells. n¼ 2 in all test groups. Radiance values reported in A andB are themeans of twoanimalswith errorbars indicating the data range. Signal intensity scaling in images in A and B were adjusted as indicated for visual clarity. Results represent one independent trial.

www.aacrjournals.org Cancer Immunol Res; 4(7) July 2016 641


2016;4:498-508. Published OnlineFirst April 8, 2016.Cancer Immunol Res Eugenia Zah, Meng-Yin Lin, Anne Silva-Benedict, et al. Receptors Prevent Antigen Escape by Malignant B CellsT Cells Expressing CD19/CD20 Bispecific Chimeric Antigen

Updated version

10.1158/2326-6066.CIR-15-0231doi:

Access the most recent version of this article at:

Material

Supplementary

http://cancerimmunolres.aacrjournals.org/content/suppl/2017/07/25/2326-6066.CIR-15-0231.DC2

http://cancerimmunolres.aacrjournals.org/content/suppl/2016/04/08/2326-6066.CIR-15-0231.DC1Access the most recent supplemental material at:

Cited articles

http://cancerimmunolres.aacrjournals.org/content/4/6/498.full#ref-list-1

This article cites 45 articles, 17 of which you can access for free at:

Citing articles

http://cancerimmunolres.aacrjournals.org/content/4/6/498.full#related-urls

This article has been cited by 23 HighWire-hosted articles. Access the articles at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected]

To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://cancerimmunolres.aacrjournals.org/content/4/6/498To request permission to re-use all or part of this article, use this link



http://cancerimmunolres.aacrjournals.org/lookup/doi/10.1158/2326-6066.CIR-15-0231

http://cancerimmunolres.aacrjournals.org/content/suppl/2016/04/08/2326-6066.CIR-15-0231.DC1 http://cancerimmunolres.aacrjournals.org/content/suppl/2017/07/25/2326-6066.CIR-15-0231.DC2

http://cancerimmunolres.aacrjournals.org/content/suppl/2016/04/08/2326-6066.CIR-15-0231.DC1 http://cancerimmunolres.aacrjournals.org/content/suppl/2017/07/25/2326-6066.CIR-15-0231.DC2

http://cancerimmunolres.aacrjournals.org/content/4/6/498.full#ref-list-1

http://cancerimmunolres.aacrjournals.org/content/4/6/498.full#related-urls

http://cancerimmunolres.aacrjournals.org/cgi/alerts

mailto:[email protected]

http://cancerimmunolres.aacrjournals.org/content/4/6/498


Documents

TCellsExpressingCD19/CD20BispeciﬁcChimeric Antigen ... · chimericantigenreceptors(CARs)havecurativepotentialagainst relapsed B-cell malignancies (3–9). However, these trials