therapeuticefﬁcacyofanfc-enhancedtcr-likeantibodyto the ......for acute lymphoblastic leukemia...

Cancer Therapy: Preclinical

TherapeuticEfficacyof anFc-EnhancedTCR-likeAntibody tothe Intracellular WT1 Oncoprotein

Nicholas Veomett1,2, Tao Dao1, Hong Liu3, Jingyi Xiang3, Dmitry Pankov1, Leonid Dubrovsky1,Joseph A. Whitten4, Sun-Mi Park1, Tatyana Korontsvit1, Victoria Zakhaleva1, Emily Casey1, Michael Curcio1,Michael G. Kharas1, Richard J. O'Reilly1, Cheng Liu3, and David A. Scheinberg1,2

AbstractPurpose: RMFPNAPYL (RMF), a Wilms’ tumor gene 1 (WT1)–derived CD8 T-cell epitope presented by

HLA-A�02:01, is a validated target for T-cell–based immunotherapy. We previously reported ESK1, a high

avidity (Kd < 0.2 nmol/L), fully-human monoclonal antibody (mAb) specific for the WT1 RMF peptide/

HLA-A�02:01 complex, which selectively bound and killed WT1þ and HLA-A�02:01þ leukemia and solid

tumor cell lines.

Experimental Design:We engineered a second-generation mAb, ESKM, to have enhanced antibody-

dependent cell-mediated cytotoxicity (ADCC) function due to altered Fc glycosylation. ESKM was

compared with native ESK1 in binding assays, in vitro ADCC assays, and mesothelioma and leukemia

therapeutic models and pharmacokinetic studies in mice. ESKM toxicity was assessed in HLA-A�02:01þ

transgenic mice.

Results: ESK antibodies mediated ADCC against hematopoietic and solid tumor cells at concentrations

below 1 mg/mL, but ESKM was about 5- to 10-fold more potent in vitro against multiple cancer cell lines.

ESKM was more potent in vivo against JMN mesothelioma, and effective against SET2 AML and fresh ALL

xenografts. ESKMhad a shortened half-life (4.9 days vs. 6.5 days), but an identical biodistribution pattern in

C57BL/6Jmice. At therapeutic doses of ESKM, there was no difference in half-life or biodistribution inHLA-

A�02:01þ transgenicmice comparedwith the parent strain. Importantly, therapeutic doses of ESKM in these

mice caused no depletion of total WBCs or hematopoetic stem cells, or pathologic tissue damage.

Conclusions: The data provide proof of concept that an Fc-enhancedmAb can improve efficacy against a

low-density, tumor-specific, peptide/MHC target, and support further development of this mAb against an

important intracellular oncogenic protein. Clin Cancer Res; 20(15); 4036–46. �2014 AACR.

IntroductionTherapeutic monoclonal antibodies (mAb) are highly

specific and effective drugs, with pharmacokinetics suitablefor infrequent dosing. However, all current marketed ther-apeutic anticancer mAbs target extracellular or cell-surfacemolecules, whereas many of the most important tumor-associated and oncogenic proteins are nuclear or cyto-plasmic (1, 2). Intracellular proteins are processed by theproteasome and presented on the cell surface as smallpeptides in the pocket ofmajor histocompatibility complex

(MHC) class I molecules (HLA in humans), allowing rec-ognition by T-cell receptors (TCR; refs. 3 and 4). Therefore,mAbs that mimic the specificity of TCRs (i.e., recognizing apeptide presented in the context of a specific HLA-type) canbind cell-surface complexes with specificity for an intracel-lular protein. A "TCR-mimic" (TCRm) antibody was firstreported by Andersen et al. (5), and several have since beendeveloped by various groups (6–11).

We recently reported the first fully human TCRm mAb,called ESK1, that specifically targets RMFPNAPYL (RMF), apeptide derived from Wilms’ tumor gene 1 (WT1), pre-sented in the context of HLA-A�02:01 (RMF/A2; ref. 12).WT1 is an important, immunologically validated oncogenictarget that has been the focus of many vaccine trials (13).WT1 is a zinc finger transcription factor with limited expres-sion in normal adult tissues, but is overexpressed in themajority of leukemias and a wide range of solid tumors,especially mesothelioma and ovarian cancer (14–16). WT1was ranked as the top cancer antigenic target for immuno-therapy by anNIH-convened panel (17); furthermore,WT1expression is a biomarker and a prognostic indicatorin leukemia (18, 19). ESK1 mAb specifically bound to

Authors' Affiliations: 1Sloan Kettering Institute; 2Weill Cornell MedicalCollege, New York, New York; 3Eureka Therapeutics Inc., Emeryville,California; and 4Temple University, Philadelphia, Pennsylvania

Note: Supplementary data for this article are available at Clinical CancerResearch Online (http://clincancerres.aacrjournals.org/).

Corresponding Author: David A. Scheinberg, Memorial Sloan KetteringCancer Center, 1275 York Avenue, New York, NY 10021. Phone:6468882190; Fax: 6464220296; E-mail: [email protected]

doi: 10.1158/1078-0432.CCR-13-2756

�2014 American Association for Cancer Research.

ClinicalCancer

Research

Clin Cancer Res; 20(15) August 1, 20144036

on August 19, 2021. © 2014 American Association for Cancer Research. clincancerres.aacrjournals.org Downloaded from

Published OnlineFirst May 21, 2014; DOI: 10.1158/1078-0432.CCR-13-2756

http://clincancerres.aacrjournals.org/

leukemias and solid tumor cell lines that are bothWT1þ andHLA-A�02:01þ and showed efficacy inmousemodels in vivoagainst severalWT1þHLA-A�02:01þ leukemias (12). There-fore, ESK1 is a useful therapeutic platform for furtherclinical development, and improvements to the nativeantibody could help potentiate its effect and improve clin-ical efficacy.The mechanisms of action of mAbs can be enhanced

through Fc region protein engineering (20), or by modifi-cation of Fc-region glycosylation (21, 22). Removal offucose from the carbohydrate chain increases mAb bindingaffinity for the activating FcgRIIIa receptor and enhancesantibody-dependent cellular cytotoxicity (ADCC; refs. 23–26). The addition of bisecting N-acetyl-D-glucosamine(GlcNAc) can also significantly enhance ADCC (26–28).However, removal or replacement of the terminal galactoseresidues present on endogenous IgG reduces complement-dependent cytotoxicity (CDC) activity (22, 29).TCRmantibodies are potentially limited by the extremely

low number of epitopes presented on the target cell, whichmay be as few as several hundred sites (30). Therefore,mechanisms to enhance potency may be essential to theirsuccess in humans as therapeutic agents against cancer. TheESK1mAbworks primarily throughADCCand thereforewehypothesized that an Fc-glycosylation altered version of theantibody would improve efficacy in vivo. An Fc-modifiedantibody was generated by expressing the ESK1 construct inMAGE 1.5 CHO cells (Eureka Therapeutics, Inc.), resultingin a consistent pattern of defucosylation and exposed ter-minal hexose (mannose and/or glucose), allowing higheraffinity for activating human FcgRIIIa and murine FcgRIVwhile decreasing affinity for inhibitory FcgRIIb. The mod-ified antibody, "ESKM," mediated ADCC at lower dosesthan native ESK1 and was more potent in human tumor

models in vivo. Furthermore, ESKM had similar pharmaco-kinetics and biodistribution to the native antibody. ESKMshowed no observable off-target tissue sink in wild-typemice, and at therapeutic doses there was no difference inhalf-life or biodistribution in HLA-A2.1þ transgenic micecompared with the parent strain. Importantly, therapeuticdoses of ESKM in these mice caused no depletion of totalWBCs or hematopoietic stem cells (HSC), or pathologictissue damage. The retained specificity, enhanced potency,favorable pharmacokinetics and distribution, and lack oftoxicity in these models support ESKM as a promising leadclinical drug candidate to treat a wide variety of cancers andleukemias.

Materials and MethodsOligosaccharide analysis and FcR binding assays

N-Glycan fromESK1 or ESKMwas cleaved from antibodyby PNGase F, and measured by HPAEC-PAD using PA200column. Binding of ESK1/ESKM to mouse FcgR4 andmouse FcgRIIb were measured by ELISA. Briefly, 2 mg/mLrecombinant mouse FcgR4 or FcgRIIb were coated ontoELISA plate. Various concentrations of ESK1 or ESKM anti-bodies were added to the wells for 1 hour at room temper-ature, then detected by secondary antibody (HRP-conjugat-ed antihuman IgG Fab’2 fragment). Binding of ESK1/ESKMto human FcgRs was measured by flow cytometry (GuavaeasyCyte HT; Millipore) against CHO cells expressingappropriate human FcgR. Binding of ESK1/ESKM to humanFcgRI, FcgRIIa, FcgRIIIa-158V, FcgRIIIa-158F, and humanFcRn were measured directly using ESK1 or ESK1 antibody,followed by the second antibody (FITC-conjugated Fab’2fragment anti-human IgG Fab’2). For human FcgRIIb,dimers were formed first by mixing ESK1 or ESKM to aphycoerythrin-conjugated Fab’2 fragment antihuman Fab’2at 2:1 ratio at RT for 2 hours. Binding of dimeric complex ofESK1 or ESKM to human FcgRIIb weremeasured directly byflow cytometry using the immunocomplex.

Cells and reagentsCell lines were from laboratory stocks, andwere passaged

in RPMI with 10% FBS for less than 1 month beforeexperiments. Cell lines were from the same stocks as pub-lished previously (12), and were not tested and authenti-cated. An ovarian patient sample (WT1 positive by immu-nohistochemistry, and strongly positive for HLA-A2 andESK mAb binding by flow cytometry) was obtained underMSKCC IRB approved protocols. For acute lymphoblasticleukemia (ALL) leukemia animal studies, fresh pre-B-cellALL cells were obtained under MSKCC IRB approved pro-tocols from the CNS relapse of a female pediatric patientafter treatmentwith a chemotherapy induction regimenandbone-marrow transplant. Leukemia cells were transducedwith a lentiviral vector containing a plasmid encodingluciferase/GFP (kindly provided by Vladimir Ponomarev,MSKCC). Luciferaseþ/GFPþ leukemiawas then expanded inNSG mice, luciferase signal was confirmed by biolumines-cent imaging, and tumor cells were harvested and sorted forCD45. Peptides for T2 pulsing assays were purchased and

Translational RelevanceWilms’ tumor protein (WT1) is an intracellular, onco-

genic transcription factor that is overexpressed in a widerange of cancers, but has limited expression in normaladult tissues. ESK1, a human IgG1 mAb, mimics T-cellreceptor specificity for a WT1-derived peptide (RMF)presented by HLA-A�02:01 and kills cancer cells viaantibody-dependent cellular cytotoxicity (ADCC), sug-gesting that it could treat a wide range of cancers whilesparing normal tissue. However, RMF/HLA-A�02:01 epi-tope levels are far lower than other therapeutic mAbtargets, so enhanced ADCC activity should be morevaluable clinically. ESKM, a construct with altered Fcglycosylation, demonstrated increased ADCC activity invitro and greater potency and efficacy in mice. ESKMwasnot toxic to human HLA-A�02:01 transgenic mice. Thisstudyprovides proof of concept that anADCC-enhancedmAb construct can effectively treat cancers while target-ing a low-density peptide/HLA epitope (500–6,000 percell), and supports the clinical utility of ESKM.

Fc-Enhanced TCR-like Antibody against WT1 Peptide/HLA-A2

www.aacrjournals.org Clin Cancer Res; 20(15) August 1, 2014 4037




synthesized by Genemed Synthesis, Inc. Peptides were>90% pure. GFPþ luciferase-expressing SET2 and JMN cellswere generated as described previously (12). All cells wereHLA typed by the Department of Cellular Immunology atMemorial Sloan-Kettering Cancer Center.

AnimalsC57BL/6J and C57BL/-Tg (HLA-A2.1) 1 Enge/J (6- to 8-

week-old male) and NOD.Cg-PrkdcscidIl2rgtm1Wjl/SzJ mice(6- to 8-week-old male), known as NOD SCID g (NSG),were purchased from Jackson Laboratory. Functionallyequivalent NOD.Cg-PrkdcscidIl2rgtm1Sug/JicTac (6- to 8-week-old male), known as NOG, and C.B-Igh-1b/IcrTac-Prkdcscid (6- to 8-week-old male), known as CB17 SCID,were purchased fromTaconic. All studies were conducted inaccordance with IACUC-approved protocols.

Antibody-dependent cellular cytotoxicityAfter informed consent on Memorial Sloan-Kettering

Cancer Center Institutional Review Board (MSKCC IRB)approved protocols, peripheral blood mononuclear cells(PBMC) from healthy donors were obtained by Ficolldensity centrifugation. Target cells used for ADCC wereT2 cells pulsedwith orwithoutWT1 or RHAMM-3 peptides,and cancer cell lines or primary ovarian cancer samplewithout peptide pulsing. ESK1, ESKM, or isotype controlhuman IgG1 (Eureka Therapeutics, Inc.) at various concen-trations were incubated with target cells and fresh PBMCs atdifferent effector:target (E:T) ratio. Cytotoxicity was mea-sured by standard 4-hour 51Cr-release assay.

Therapy of ESK1 and ESKM in human mesothelioma,acute myeloid leukemia, and ALL xenograft mousemodels

Luciferase-expressing JMN cells (3 � 105) were injectedinto the intraperitoneal cavity of CB17 SCID mice. On day4, tumor engraftment was confirmed by luciferase imaging,signal was quantified with Living Image software (Xeno-gen), andmice were sorted into groups with similar averagesignal from the supine position. Mice were injected intra-peritoneally with 50 mg ESK1, ESKM, or human isotypeIgG1 antibody twice weekly beginning on day 4.

For acute myeloid leukemia (AML) leukemia studies,luciferase-expressing SET2 (AML) cells (3 � 106) wereinjected intravenously via tail vein into NSGmice. Animalswere sorted and, where indicated, treated with intraperito-neal injections of 100 mg ESKM twice weekly beginning onday 6. For ALL studies, fresh leukemia cells were obtained asdescribe above (cell lines and reagents) then injected intra-venously into NSG mice (55 � 106/animal), and engraft-ment was confirmed by bioluminescent imaging on day 2after injection. Animals were sorted into 2 groups (n ¼ 5each) so that average signal in each group was equal. ESKMor isotype control antibody (100 mg/animal) was adminis-tered via retro-orbital injection on days 2, 5, 9, 12, 14, and23, and leukemia growth was followed by bioluminescentimaging. On day 41, animals were sacrificed and bonemarrow cells were harvested and pooled. After dissection

and homogenization, cells were centrifuged, subjected toFicoll density centrifugation, and counted after red bloodcell lysis (acetic acid). An equal number of cells from eachtreatment group was resuspended in matrigel (200 mL/injection) and engrafted subcutaneously into the oppositeshoulders of NSG mice (n ¼ 4). No further treatment wasgiven, and tumor growth was followed by bioluminescentimaging.

Pharmacokinetic and biodistribution studiesAntibody was labeled with 125I (PerkinElmer) using the

chloramine-T method. One hundred micrograms of anti-body was reacted with 1mCi 125I and 20 mg chloramine-T,quenched with 200 mg Na metabisulfite, then separatedfrom free 125I using a 10DG column equilibrated with 2%bovine serumalbumin in PBS. Specific activities of productswere in the range of 4 to 8 mCi/mg. Radiolabeled mAb wasinjected into mice retro-orbitally, and blood and/or organswere collected at various time points, weighed and mea-sured on a g counter.

Toxicity studiesFor isolated cell binding studies, C57BL/6J or HLA-A2.1þ

transgenic mice were sacrificed, and cells were harvestedfrom spleen, thymus, and bone-marrow. After red bloodcell lysis, cells (106 per tube, in duplicate) were incubatedwith 125I-labeled ESK1 (1 mg/mL) for 45 minutes on ice,then washed extensively with 1% bovine serum albumin inPBS on ice. To determine specific binding, a set of cells wasassayed after preincubation in the presence of 50-fold excessunlabeled ESK1 for 20 minutes on ice. Bound radioactivitywas measured by a g counter, specific binding was deter-mined, and the number of bound antibodies per cell wascalculated from specific activity.

For toxicity studies, 100 mg of ESKM or isotype controlmAb was injected into human HLA-A�02:01 transgenicmice (Jackson Labs) on days 0 and 4, to mimic the max-imum dose and therapeutic schedule used in the therapyexperiments. Mice were sacrificed on day 5 for collectionand analysis of whole blood and bone marrow leukocytes.Whole blood was analyzed with a Hemavet system (DrewScientific). Bone marrow cells were harvested from bothfemurs and tibias of mice and subjected to red blood celllysis, then analyzed by flow cytometry (see Antibodies andflow cytometry analysis).

Antibodies and flow cytometry analysisFor cell surface staining, cells were blocked with human

FcR blocking reagent (Miltenyi Biotec), then incubatedwithappropriate mAbs for 30 to 60 minutes on ice. Flow cyto-metry data were collected on a FACS Calibur or LSRFortessa(Becton Dickinson) and analyzed with FlowJo software(TreeStar). APC-labeled ESK1 and hIgG1 isotype (EurekaTherapeutics, Catalog No. ET901) were generated withLightning-Link Kit (Innova Biosciences). For human leu-kemia stem cell studies, cells were stained with the follow-ing antibodies, and corresponding fully stained minus onecontrols: PE-Cy5 anti-human lineage cocktail (BioLegend;

Veomett et al.

Clin Cancer Res; 20(15) August 1, 2014 Clinical Cancer Research4038




anti-CD3, CD14, CD16, CD19, CD20, CD56), FITC-labeled anti-CD38, PE-labeled anti-CD34, PE-Cy7-labeledBB7.2, and APC-labeled ESK1.For HSC toxicity studies, mouse bone marrow cells were

stainedwith the following antibodies: (Lineage; CD3, CD4,CD8, Gr1, B220, CD19, TER119, all conjugated with PE-Cy5), Sca-Pacific Blue, CD34-FITC, SLAM-APC, CD48-PE,and c-KIT-Alexa Fluor 780. The stained cells were analyzedfor flow cytometry on the BD LSRII instrument.For mouse immunophenotyping, cells were isolated

from the intraperitoneal cavity by washing with completemedia, or from spleen by dissection and red blood cell lysis(RBC Lysis Solution; Qiagen). Samples were then analyzedby flow cytometry after multicolor staining with well-char-acterized lineage-specific markers: CD335 (NKp46)-PE andF4/80 (BM8)-Alexa Fluor 700 (BioLegend), CD49b/VLA-2a(DX5)-FITC (Life Technologies), CD3e-PE-Cy7 and Gr-1/Ly-6G/Ly-6C (RB6-8C5)-PerCP-Cy5.5 (BD Pharmingen).

ResultsESKM antibody has enhanced binding affinity forFcgRIIIa and reduced affinity for FcgRIIbESKM mAb was produced in MAGE 1.5 CHO cells, with

the homogeneous oligosaccharide structure (Supplemen-tary Fig. S1A) and no detectable fucose or galactose. ESKMhad 80% higher affinity for activating human FcgRIIIa (158Vvariant), 3.5-fold higher affinity for the FcgRIIIa 158Fvariant, and 50% reduced affinity for inhibitory FcgRIIb.Importantly, ESKM affinity for FcRn was unchanged (Table1 and Supplementary Fig. S1B). Similarly, ESKM had 51%higher affinity for activating mouse FcgRIV, and half theaffinity for inactivating mouse FcgRIIb (Table 1 and Sup-plementary Fig. S1C). Changes in Fc glycosylation pattern

did not affect antigen binding, as avidity of ESKM againstWT1þ HLA-A�02:01þ JMN cells was nearly identical to thenative ESK1 (0.2–0.4 nmol/L; Table 1 and SupplementaryFig. S1D).

ADCC mediated by ESKM in vitroWe investigated the relationship of cell surface antigen

density with ESKM ADCC efficacy using T2, a TAP-deficientcell line that expresses HLA-A�02:01 but does not presentpeptides through the ER pathway, and thus can be loadedwith exogenous peptides for presentation in a dose-depen-dent manner. To determine whether ESKM could bettermediate ADCC against cells with low antigen density, wefixed the dose of ESK1 and ESKM mAbs and tested themagainst T2 cells loaded with titrated RMF peptide. Bothantibodies were effective against T2 cells pulsed with highpeptide concentrations, but ESKM was able to mediategreater ADCC against cells with fewer RMF/A2 complexes(Fig. 1A).

We next determined the in vitro ADCC activity of ESK1and ESKM against cell lines presenting a range of levels ofcell surface RMF/A2 (12). ESKM showed both increasedpotency and efficacy against 6 leukemia and mesotheliomacell lines in an HLA-restricted manner. ESKM effectivelymediated ADCC against BA-25, an ALL cell line expressingapproximately 1,000 to 2,000 RMF/A2 targets per cell; bothantibodies were similarly effective at ADCC at concentra-tions above 1 mg/mL, but ESKM was more potent at con-centrations down to 100 ng/mL of mAb (Fig 1B). AgainstAML-14 and SET2 AML cell lines, which both bind �5,000mAb per cell, ESKMmediated higher cell lysis than ESK1 atthe highest antibody concentrations, and showed cytolyticefficacy down to doses as low as 100 ng/mL (Fig. 1B). As we

Table 1. Summary of ESK1 and ESKM binding to mouse and human FcgRs, and RMF/A2þ target cells

Kd � SD (nmol/L)

Receptor ESK1 ESKMRatio of affinityconstants (ESKM/ESK1)

MouseFcgRIIb 32.0 � 0.454 62.3 � 7.27 0.51FcgRIV 3.34 � 0.193 2.21 � 0.153 1.51

HumanFcgRI 0.581 � 0.113 0.680 � 0.125 N.C.FcgRIIa 105 � 15.5 58.3 � 8.80 1.81FcgRIIb 1338 � 253 2644 � 438 0.51FcgRIIIa (158V) 92.6 � 15.0 50.4 � 8.35 1.84FcgRIIIa (158F) 19.0 � 2.38 5.53 � 0.741 3.45FcRn 824 � 102 780 � 97.5 N.C.

Cell lineJMN 0.2 0.4

NOTE: Anti-mouse FcR binding was assessed by ELISA, whereas anti-human FcR binding was determined by FCM titration on FcgR-expressing CHOcells. N.C., no change. To determine affinity for the RMF/A2 epitope, 125I-labeled ESK1 and ESKMmAbswere titratedagainst JMN cells. All curves were fit with a nonlinear single-site total binding saturation curve, and Kd was calculated using Prismsoftware.






have shown previously for ESK1 (12), ESKM did not killleukemia cells not expressing HLA-A�02:01 (Fig. 1C). Fur-thermore, ESKM mediated higher specific lysis at nearly alldoses tested against 3 HLA-A�02:01þ mesothelioma celllines: JMN, Meso-37, and Meso-56 (Fig. 1D). Importantly,ESKMwas alsomore potent against aWT1þ, HLA-A�02:01þ

primary fresh patient ovarian cancer sample (Fig 1D), whichwas positive for ESK binding by flow cytometry. To assessthe human effector cell populations responsible forADCC invitro, we separated whole blood from healthy donors intoneutrophils and PBMCs, which were further sorted intomacrophage/monocyte and NK-cell populations (see Sup-plementary Methods). Only isolated NK cells were capableof ADCC directed by ESKM in vitro against BV173 targetcells; Supplementary Fig. S2). These data showed that ESKM

was bothmore potent—as illustrated by its ability to kill cellswith lower mAb concentrations and fewer cell surfacetargets—and more effective than ESK1, as demonstrated byhigher specific lysis attained at the highest concentrations.

Potency of ESKM against human mesothelioma andleukemia models in mice

Data from several experiments in vitro and in vivo pro-vided strong evidence that ADCC was the dominant mech-anism of therapeutic action of the ESK1mAb (12). Here, weutilized a SCID mouse model to investigate whether ESKMoffered a consistent and significant improvement overnative ESK1 in vivo in mice with more available effectorcells capable of ADCC than NSG/NOG mice (12). ESKMmore effectively engages activating murine FcgRIIa and

Figure 1. ESKM is more efficacious and potent in ADCC assays with human PBMC effectors at the indicated mAb concentrations and effector/target (E:T)ratios. Cytotoxicitywasmeasuredby 4-hour 51Cr release assay. A, T2 cells were pulsedwithRMFpeptide and incubatedwith 3mg/mLmAb. B,HLA-A�02:01þ

human leukemia cell lines: BA25 ALL, AML14, and SET2 AML. C,HLA-A�02:01� HL60 promyelocytic leukemia. D,HLA-A�02:01þ humanmesothelioma celllines: JMN, Meso37, and Meso56; and a HLA-A�02:01þ primary human ovarian cancer. Data presented are averages of triplicate measurements, withspontaneous 51Cr release subtracted, from representative experiments, all with isolated PBMCs from the same donor. All cell lines, with exception ofMeso37and Meso56, were repeated 3 or more times with multiple donors.

Table 2. Intraperitoneal effector cell populations in BALB/c, SCID, and NOG mice

Cells (% of parent population)

Parent population Cell type Marker phenotype BALB/c CB17 SCID NOG

CD11bþGranulocyte Gr-1þ F4/80� 46.6 � 28.3 7.45 � 2.71 0.215 � 0.137Macrophage Gr-1low F4/80þ 10.5 � 6.87 19.0 � 1.29 96.3 � 1.06Monocyte Gr-1� F4/80� 39.2 � 21.1 70.2 � 2.73 1.25 � 0.562

CD11b�NK cell NKp46þ CD3e� 1.68 � 0.250 36.2 � 12.0 0.278 � 0.413

NOTE: Intraperitoneal cells were isolated from mice (n ¼ 3 each strain) and analyzed by multicolor flow cytometry. Cell type wasdetermined by the indicated markers, and quantified as percentage of total leukocytes isolated.

Veomett et al.





FcgRIV, therefore murine cells should serve as potent effec-tors in vivo. Mice were engrafted with luciferaseþ JMNmesothelioma cells in the intraperitoneal cavity (simulatingthis serosal cavity cancer). To determine the relative abun-dance of murine effector cell populations in the intraperi-toneal cavity, we analyzed extracted cells with commonmurine immunophenotyping markers (31). SCID micecontain intraperitoneal macrophages, neutrophils, and NKcells (Table 2). This flow cytometry analysis also confirmedthe presence of murine monocytes, macrophages, and NKcells, but lack of B and T cells in the spleen and peripheralblood, as expected (32). Biweekly 50 mg treatment withESKM was more effective than ESK1 against intraperitonealJMN, and significantly improved survival over isotype con-trol antibody (Fig. 2A). Furthermore, ESKM was able toreduce tumor burden during the treatment course, whereas

ESK1merely slowed growth (Supplementary Fig. S3A). In athird experiment at the same dose and schedule, neitherantibody construct showed efficacy against intraperitonealJMN in NOG mice (Supplementary Fig. S3B), which lackNK cells and intraperitoneal neutrophils (Table 2), indicat-ing that these cell populations likely play an important rolein efficacy in the intraperitoneal model.

We previously reported that ESKM is effective at a lowdose against disseminated bcr/ablþ BV173 ALL in a NSGmouse model, where FcgRIVþ monomyeloid cells are pres-ent in the spleen (33). We therefore used NSG mice toevaluate ESKM efficacy against disseminated leukemiamodels. We first investigated ESKM against SET2, an AMLthat grew more aggressively than BV173 in the NSG mousemodel (Supplementary Fig. S3C). ESKM significantlyreduced tumor growth (Fig. 2B). ESKM also significantly

70 14 210.0

5.0 × 10

1.0 × 10

1.5 × 10

2.0 × 10

2.5 × 108

8

8

6

6

6

6

6

6

6

6

6

5

6

8

7

hIgG1ESK1ESKM

Day

A

*

0 98 10 11 12 13 14 280

50

100

Weeks

Perc

enta

ge o

f sur

viva

l

hIgG1ESK1ESKM

70 14 21 28 350.0

5.0 × 10

1.0 × 10

1.5 × 10

2.0 × 10

2.5 × 109

9

9

9

8

DayLu

min

esce

nce

(pho

tons

/sec

/cm

2 )

UntreatedESKM

B* *

C

Day

14

Day

25

hIgG1ESKM

ESKM IgG0

1×10

2 × 10

3 × 10

4 × 10

ESKMhIgG1

0.0

5.0 × 10

1.0 × 10

1.5 × 10

2.0 × 10

2.5 × 10

Day

s afte

r eng

raftm

ent

D

Lum

ines

cenc

e(p

hoto

ns/s

ec/c

m)

Day 40 after engraftment

hIgG1ESKM

100 20 300

1 × 10

2 × 10

3 × 10

Day

Lum

ines

cenc

e (p

hoto

ns/s

ec/c

m2 ) IgG

ESKM*

**

**

Day 28 after transplant

Isotype

Lineage+

Lin CD34+ CD38–

ESK

low/-

100

80

60

40

20

00 102 103 104 105

% o

f Max

Figure 2. ESKM is superior to ESK1 in vivo, and is effective against multiple tumor models. Tumor burden was determined by luciferase imaging of mice in thesupine position (n ¼ 5 per group). Data points are averages of each group, and error bars represent SEM. Arrows indicate treatment with mAb.A, ESKM significantly reduced mean tumor growth of intraperitoneal JMN mesothelioma as assessed by total luminescence (�P < 0.05, multiple T testson and after day 18), and also significantly improved survival (P¼ 0.016 vs. isotype, P¼ 0.095 vs, ESK1), with events representing death or terminal morbidityas assessed on protocol by veterinarians. B, ESKM is effective against SET2 AML (�P < 0.05, multiple T tests). C, ESKM is effective against a disseminatedfresh, patient-derived pre-B ALL (�P < 0.05, ��P < 0.01, multiple T tests). D, leukemia burden in the bone marrow was quantitated by imaging on day40, then bone marrow cells were harvested from mice in C and transplanted as subcutaneous tumors into NSG mice. Bone marrow cells from theisotype-treated mice were injected into the right shoulder (viewed from above), whereas an equal number of bone marrow cells from the ESKM-treated micewere injected into the left shoulder. Subcutaneous tumors were then quantitated 28 days after transplantation. In a multicolor flow cytometry analysis of afrozen stock of the same fresh primary ALL sample (before passage in animals), ESK mAb bound the lineagelow/� CD34þ CD38� subset.






reduced initial burden and slowed outgrowth of a patient-derived human pre-B-cell ALL (Fig. 2C). Leukemia relapsedafter treatment was stopped, allowing us to collect leukemiacells from the bone marrow and transplant to new animalsto assess outgrowth from remaining progenitors. Total bonemarrow signal in ESKM-treated mice was lower at time oftransplant, but equal numbers of ESKM-treated and isotype-treated bone marrow cells were engrafted into recipientanimals. Subcutaneous leukemia tumors from isotype-trea-ted leukemia cells grew to 50 times the total signal com-paredwith tumors fromESKM-treated cells (Fig. 2D). In thissame patient ALL sample, ESK bound a lineagelow, CD34þ,CD38� population, characterized as a cancer stem cellpopulation in ALL (34), at nearly the same level aslineageþ populations (Fig. 2D and Supplementary Meth-ods). This suggests that ESKM therapy could target thispopulation, but does not imply that ESK mAbs can differ-entiate between leukemic stem cell and normal HSC popu-lations. Further supporting the hypothesis that ESK mAbscan target leukemic progenitor cells, ESK1 bound CD33þ

/CD34þ cells from a separate HLA-A�02:01þ AML patientwith�8-fold shift in median fluorescence intensity of ESK1compared with isotype control antibody (12).

Pharmacokinetics and biodistribution of ESK1 andESKM

Altering Fc glycosylation could potentially change phar-macokinetic properties of the mAb, thereby affecting itstherapeutic utility. Trace 125I-labeled antibodies wereinjected intravenously into mice and blood levels of mAbwere measured over 7 days. Both ESK1 and ESKM exhibitedbiphasic clearance, with initial tissue distribution and an ahalf-life of 1.1 to 2.4 hours, followed by a slower b half-lifeof several days (Fig. 3A). ESKMhad a shorter b half-life thanESK1 (4.9 days vs. 6.5 days). The biodistribution patterns ofthe antibodies were determined using the same radiola-beled constructs. Both antibodies displayed similar patternsof organ distribution and clearance (Fig. 3A). Althoughincreased FcgR binding or manose receptor binding couldcreate a sink for ESKM, we found no increase in ESKMdistribution to the liver, spleen, thymus, or bone marrowthat could account for the shortened serum half-life.

As the ESKM antibody targets a human-specific epitope,the C57BL/6J mouse model cannot recapitulate possibleon-target binding to normal tissues that could alter anti-body pharmacokinetics and biodistribution. Therefore, weused a transgenic mouse model based on the C57BL/6Jbackground that expresses humanHLA-A�02:01 driven by alymphoid promoter. We confirmed the presence of surfaceHLA-A�02:01 on murine cells by flow cytometry with theBB7.2 antibody. Leukocytes isolated from thymus tissueshowed low but detectable binding, whereas cells isolatedfrom the bone marrow and spleen showed appreciablebinding, with a 5- to 10-fold shift in median fluorescentintensity over isotype control antibody, comparable tosurface HLA-A�02:01 levels observed on human PBMCsfrom healthy donors (12). The 9-mer RMF sequence isidentical in human and mouse, and therefore this trans-

genic model could recapitulate antigen presentation of theRMF/A2 epitope in healthy cells. There was no differencebetween wild-type and HLA-A�02:01 transgenic mice inblood pharmacokinetics of ESKM, indicating that there wasno significant antibody sink. Furthermore, at a therapeuticdose of antibody (100 mg), there was no difference inbiodistribution in transgenic comparedwithwild-typemice(Fig. 3B).

At low doses of trace-labeled ESKM (2 mg), there was asmall yet detectable increase in uptake of antibody in thespleen of transgenic mice compared with wild-type mice(Fig. 3C). The additional splenic binding was small,accounting for only 16 ng of antibody, which could be dueto RMF presentation in HLA-A�02:01þ cells, or to anunknown cross-reacting epitope. Splenic uptake was notbecause of Fc glycosylation pattern alone, as the native ESK1mAb also showed increased uptake in transgenic spleens at24 hours (Supplementary Fig. S4A). In addition, increasedspleen uptake in transgenicmice seemed to be partly relatedto strain differences in clearance, as the isotype controlantibody also showed 60% increased splenic uptake at 24hours in transgenic compared with wild-type mice (Fig. 3Dand Supplementary Fig. S4A). Furthermore, we observed nobinding of ESK1 to cells isolated from transgenic mousespleen cells by either flow cytometry (12) or specific bindingassay with 125I-labeled ESK1 (Supplementary Fig. S4B),suggesting that if a cross-reacting epitope was present, itwas not expressed in detectable amounts on a specific celltype.

Toxicity of ESKM in a HLA-A�02:01 transgenic mousemodel

WT1 is reported to be expressed in hematopoetic stemcells (HSC; ref. 35), so the C57BL/6J transgenic mousemodel with human HLA-A�02:01 driven by a lymphoidpromoter provides an opportunity to assess possible toxic-ity against progenitor cells in the hematopoetic compart-ment that, given the high potency of ESKM, might occureven at low epitope density. White blood cell and bonemarrow cell counts were assessed 1 day after the final of 2therapeutic doses of ESKM or isotype control mAb on thesame schedule as previously described therapy experiments(12). There were no differences in total white blood cellcount, or lymphocyte, neutrophil, monocyte, eosinophil,or basophil cell counts (Fig. 4A). Within the bone marrowcompartment, we found equivalent absolute number andfrequency ofHSCprogenitors (LSK: Lineagelo, c-kitþ, Sca1þ;ref. Fig. 4B) and HSCs (CD150hi, CD48�, LSK; ref. Fig. 4C)in the ESKM treated and isotype control-treated mice.

Finally, we assessed gross and microscopic pathology oflymphoid and major organs of HLA-A�02:01 transgenicmice treated with ESKM or isotype mAb on the sameschedule. No striking differences between ESKM- and iso-type-treated groups were observed by a trained hemato-pathologist (Supplementary Table S1). The bone marrowsections of both groups showed trilineage hematopoiesiswith adequate maturation of the myeloid and erythroidlineages. The megakaryocytes were adequate in number

Veomett et al.





with normal morphology. Thymus sections showed a well-defined cortex and medulla with few Hassall’s corpuscles,which is normal for rodent histology (36). The kidneysections showed no pathologic findings such as glomerulo-sclerosis, congestion, or inflammation. All spleen sectionsshowed a normal distribution of red and white pulp.Occasional scattered megakaryocytes were seen in the redpulp consistent with extramedullary hematopoiesis (37).

DiscussionThe generation of TCRm antibodies allows use of the

mAb to target cell-surface fragments of intracellular pro-

teins, provided that they are processed and presented onMHC class I molecules. ESK1 was the first TCRm antibodyreported against a peptide derived fromWT1, an importantoncogene expressed in a wide variety of cancers, but notnormal adult tissues. WT1 seems to be expressed in leuke-mic stem cells (35), raising the possibility that the mAbcould ultimately eliminate clonogenic leukemia cells inpatients. Other therapeutic TCRm mouse antibodies,human ScFv, and Fab fragments have been previouslydescribed (6–8, 10, 11). However, ESK1 is the first andonly fully human therapeutic TCRm mAb reported. Fea-tures of the RMF/A2 epitope, especially the low levels ofexpression on the cell surface, require selection of a highly

Figure 3. ESKM and native ESK1display similar pharmacokineticsand biodistribution. 125I-labeledmAb was injected intravenouslyand activity wasmeasured in bloodor harvested organs (n ¼ 3 pergroup). Data points areaverages of each group, anderror bars represent SEM.A, pharmacokinetics andbiodistribution of ESKM or nativeESK1 (3 mg each) in C57BL/6Jmice. B, pharmacokinetics of traceESKM (2 mg) and 24-hourbiodistribution of ESKM (100 mg) inC57BL/6J or HLA-A�02:01þ

transgenic mice. C and D, ESKM(C) or hIgG1 isotype control (D)(2 mg each) in C57BL/6J or HLA-A�02:01þ transgenic mice,harvested after 1 day.






potent and effective ESK1 construct. We chose to improvethe ESK1 construct by altering Fc glycosylation as ameans toenhance ADCC, the major mechanism of ESK1 action invitro and in vivo (12). Several ADCC-enhanced mAbs tohighly expressed cell-surface antigens, produced, either byglyco-engineering or pointmutations, are in clinical trials intheUnited Stateswithpromising results (38–40). TheESKMmAb had a homogeneous glycosylation pattern lacking N-linked fucose and with terminal hexose (mannose and/orglucose) structure. This engineering strategy modulatesmAb binding to Fcg receptors in 2 ways: a higher affinityfor activating human FcgRIIIa (and murine FcgRIV)increases ADCC activity, whereas diminished affinity for

both human and murine FcgRIIb should reduce inhibitoryreceptor activation. As expected, ESKM was both morepotent and effective in vitro even at very low epitope density.

ESKM was also more effective than ESK1 in vivo, and wasable to treat peritonealmesothelioma in SCIDmice,model-ing the clinical situation. Furthermore, ESKM significantlyslowed leukemia growth of disseminated SET2, an AML cellline with much more aggressive in vivo leukemia growthkinetics than BV173, and a fresh patient-derived pre-B-cellALL in xenograft models. In the fresh ALL model, tumorrelapsed after mAb therapy was stopped, but leukemia cellsextracted from the bone marrow of ESKM-treated mice andtransplanted as subcutaneous tumors showed minimaloutgrowth. This suggests that ESKM may target a progenitorpopulation of leukemia cells. This supposition is supportedby binding of ESK to a lineagelow/�, CD34þ, CD38� pop-ulation by flow cytometry, and consistent with the hypoth-esis that WT1 expression in HSCs could allow reduction ofthis population. However, cells collected from the bonemarrowafter relapsewere not phenotyped and sorted, sowecannot determine the exact cell population targeted. ESKMwas not curative in the models tested, possibly because ofthe lack of NK cells in NOG/NSGmice and short treatmentcourses. In this system, an incomplete effect is not surpris-ing, and may not accurately model the effect in humans.

ESKM therapy was not effective against peritoneal meso-thelioma in NSG or NOG mice, which lack NK cells andneutrophils, although naked ESK1 did previously showpotent activity against a disseminated leukemia model inthesemice. This discrepancy could be due both to the tumormodel—leukemia cells could have different sensitivity toeffector-mediated cytotoxicity—and to the availability ofintraperitoneal effector cells in the NSG/NOG model: ourassays indicated that the intraperitoneal cavity ofNOGmicecontained predominately macrophages, whereas neutro-phils were present in the blood and spleen, as previouslyreported (33). Themarked improvement in efficacywith theFc-enhancedmAb inSCIDmice indicated thatNK cells and/or monocytes (both with FcgRIV) are important to therapyin the intraperitoneal mesotheliomamodel, whereas splen-ic and circulating neutrophils presumably mediate ADCCagainst disseminated leukemias in NSG/NOG mice.

Altering Fc glycosylation could potentially change phar-macokinetic properties of the mAb through a number ofmechanisms, including altered FcRn binding and antibodyrecycling, modified binding to circulating effector cells, anddifferential engagement with clearance mechanisms, suchas mannose receptors. Similar afucosylated, Fc-modifiedantibodies with improved ADCC have been investigatedin pharmacokinetic studies in vivo (41, 42). ESKM hadnearly identical biodistribution to ESK1, but a shortenedbloodhalf-life.We sawnochange inbiodistributionpatternthat could account for this altered half-life. IgG half-life isregulated by the neonatal Fc receptor, FcRn (43, 44); how-ever, ESKM had identical affinity for FcRn as ESK1. Thealtered pharmacokinetics is possibly because of interactionwith mannose receptor on macrophages, a known mecha-nism of glycoprotein clearance (45–47). ESK1 and ESKM

Figure 4. ESKM treatment does not affect leukocyte or HSC counts inHLA-A�02:01þ transgenic mice. Animals (n ¼ 5 per group) were treatedwith 100 mg ESKM or hIgG1 isotype control on days 0 and 4; blood andbone marrow were harvested on day 5. Data points are averages of eachgroup, and error bars represent SEM. A, total white blood cell (WBC) andWBC subset cell counts. B, absolute number and frequency of lineage�

SCA1þ KITþ (LSK) cells. C, absolute number and frequency of long-termHSCs (Slamf1þ CD34� LSK cells).

Veomett et al.





half-lives could bequite different inhumans—where higheraffinity binding to Fc receptors and clearance throughhuman mannose receptors could have a role—as seen withother high-mannose IgG antibodies in humans (48).Because ESK mAbs target a human HLA-specific epitope,

we utilized the human HLA-A�02:01þ transgenic mousestrain for toxicology studies. WT1 is reportedly expressed inHSCs, yet a therapeutic dose of ESKM that cleared leukemiain our models had no effect on LSK cells or early HSCs.Importantly, ESKM did not affect the architecture or cellcoverage in the bone marrow, thymus, or spleen, whereWT1þ HSCs could be expected, and where HLA-A�02:01expression is highest because the transgene is driven by alymphoid promoter. This transgenic system is the bestavailable for predicting potential off target toxicity, but islimited tohematopoietic tissues.Despite this limitation, thelack of ESKM toxicity against those tissues shown to expresshuman HLA-A�02:01 suggests that ESKM may be welltolerated in humans.Selection of ESKM as a lead construct for human clinical

development required us to consider its features of a mod-erately decreasedhalf-life yet increasedpotency andbroaderapplicability. The potential enhanced efficacy againsttumors expressing fewer RMF/A2 sites could expand thenumber of patients and cancer types eligible for this therapyas well as increase efficacy. In addition, the MAGE 1.5 CHOengineering technology generates mAbs that effectivelyengage FcgRIIIa (CD16), regardless of amino acid 158polymorphism. Carriers of CD16-158F are less responsivethan CD16-158V/V individuals to human IgG1 therapeu-tics such as rituximab and trastuzumab (49, 50). WT1 isexpressed in multiple cancers (14), making RMF/A2 apotential therapeutic target for many indications. Despitethis potentially broad base of cancers, ESKM and mostTCRmantibodies developed to date are directed to peptidespresented by HLA-A�02:01, as this is the most commonallele in the US population. Patients with other epitopesthus will not benefit from ESKM and additional analogousTCRmmAbs would be necessary for these patients. Further-more, ESKM binding depends on the level of HLA-A�02:01expression, and on WT1 processing and peptide presenta-tion on the cell surface. These processes could be aberrant in

cancer cells. In preclinical models, we have shown efficacyagainst bcr/ablþ ALL and B-ALL (12), and here, AML, meso-thelioma, and primary ALL xenografts. In summary, ESKMis a potent therapeutic mAb against a widely expressedoncogenic target with a restricted normal cell expressionprofile, and has shown efficacy against multiple humantumor models in mice.

Disclosure of Potential Conflicts of InterestN. Veomett, T. Dao, L. Dubrovsky, and D.A. Scheinberg are inventors of

intellectual property related to ESKM that is owned by Sloan Kettering andlicensed to Novartis. H. Liu, J. Xiang, and C. Liu have ownership interest(including patents) in Eureka Therapeutics Inc. No potential conflicts ofinterest were disclosed by the other authors.

Authors' ContributionsConception and design: N. Veomett, T. Dao, D. Pankov, C. Liu, D.A.ScheinbergDevelopment of methodology: N. Veomett, T. Dao, H, Liu, J. Xiang, T.Korontsvit, V. Zakhaleva, D.A. ScheinbergAcquisitionofdata (provided animals, acquired andmanagedpatients,provided facilities, etc.): N. Veomett, T. Dao, J. Xiang, D. Pankov, L.Dubrovsky, J.A. Whitten, S.-M. Park, T. Korontsvit, E. Casey, M. Curcio,M.G. Kharas, R.J. O’ReillyAnalysis and interpretation of data (e.g., statistical analysis, biosta-tistics, computational analysis): N. Veomett, T. Dao, S.-M. Park, T.Korontsvit, V. Zakhaleva, M.G. Kharas, C. Liu, D.A. ScheinbergWriting, review, and/or revision of the manuscript:N. Veomett, T. Dao,H, Liu, J. Xiang, J.A. Whitten, M.G. Kharas, R.J. O’Reilly, C. Liu, D.A.ScheinbergAdministrative, technical, or material support (i.e., reporting or orga-nizing data, constructing databases): T. Dao, J.A. Whitten, T. Korontsvit,V. Zakhaleva, M. CurcioStudy supervision: T. Dao, D.A. ScheinbergOther (in vitro study design and method development): T. DaoOther (interpreted the mouse tissue histology): J.A. Whitten

AcknowledgmentsThe authors thank E. Brea and A. Scott for helpful discussions.

Grant SupportThis work was supported by the Leukemia and Lymphoma Society, NIH

R01CA55349 and P01 23766, and T32 CA062948 (N. Veomett), and theMajor Family Fund.

The costs of publication of this article were defrayed in part by thepayment of page charges. This article must therefore be hereby markedadvertisement in accordance with 18 U.S.C. Section 1734 solely to indicatethis fact.

ReceivedOctober 14, 2013; revised April 8, 2014; accepted April 28, 2014;published OnlineFirst May 21, 2014.

References1. Sensi M, Anichini A. Unique tumor antigens: evidence for immune

control of genome integrity and immunogenic targets for Tcell-mediatedpatient-specific immunotherapy. Clin Cancer Res 2006;12:5023–32.

2. Kessler JH, Melief CJ. Identification of T-cell epitopes for cancerimmunotherapy. Leukemia 2007;21:1859–74.

3. Morris E, Hart D, Gao LQ, Tsallios A, Xue SA, Stauss H. Generation oftumor-specific T-cell therapies. Blood Rev 2006;20:61–9.

4. KonigR. InteractionsbetweenMHCmoleculesandco-receptors of theTCR. Curr Opin Immunol 2002;14:75–83.

5. Andersen PS, Stryhn A, Hansen BE, Fugger L, Engberg J, Buus S. Arecombinant antibody with the antigen-specific, major histocompat-ibility complex-restricted specificity of T cells. ProcNatl AcadSci USA1996;93:1820–4.

6. Epel M, Carmi I, Soueid-Baumgarten S, Oh SK, Bera T, Pastan I, et al.Targeting TARP, a novel breast and prostate tumor-associated anti-

gen, with T cell receptor-like human recombinant antibodies. EurJ Immunol 2008;38:1706–20.

7. Wittman VP,Woodburn D, Nguyen T, Neethling FA,Wright S,WeidanzJA. Antibody targeting to a class I MHC-peptide epitope promotestumor cell death. J Immunol 2006;177:4187–95.

8. Klechevsky E, Gallegos M, Denkberg G, Palucka K, Banchereau J,CohenC, et al. Antitumor activity of immunotoxinswith T-cell receptor-like specificity against human melanoma xenografts. Cancer Res2008;68:6360–7.

9. Bhattacharya R, Xu Y, Rahman MA, Couraud PO, Romero IA, WekslerBB, et al. A novel vascular targeting strategy for brain-derived endo-thelial cells using a TCR mimic antibody. J Cell Physiol 2010;225:664–72.

10. Verma B, Neethling FA, Caseltine S, Fabrizio G, Largo S, Duty JA, et al.TCRmimicmonoclonal antibody targets a specific peptide/HLA class I






complex and significantly impedes tumor growth in vivo using breastcancer models. J Immunol 2010;184:2156–65.

11. SergeevaA,AlatrashG,HeH,RuisaardK, LuS,Wygant J, et al. An anti-PR1/HLA-A2 T-cell receptor-like antibody mediates complement-dependent cytotoxicity against acute myeloid leukemia progenitorcells. Blood 2011;117:4262–72.

12. Dao T, Yan S, Veomett N, Pankov D, Zhou L, Korontsvit T, et al.Targeting the intracellular WT1 oncogene product with a therapeutichuman antibody. Sci Transl Med 2013;5:176ra33.

13. Dao T, Scheinberg DA. Peptide vaccines for myeloid leukaemias. BestPract Res Clin Haematol 2008;21:391–404.

14. Tanaka-Harada Y, KawakamiM, Oka Y, Tsuboi A, Katagiri T, ElisseevaOA, et al. Biased usage of BV gene families of T-cell receptors of WT1(Wilms' tumor gene)-specific CD8þ T cells in patients with myeloidmalignancies. Cancer Sci 2010;101:594–600.

15. SoslowRA. Histologic subtypes of ovarian carcinoma: an overview. IntJ Gynecolo Pathol 2008;27:161–74.

16. Raja S, Murthy SC, Mason DP. Malignant pleural mesothelioma. CurrOncol Rep 2011;13:259–64.

17. Cheever MA, Allison JP, Ferris AS, Finn OJ, Hastings BM, Hecht TT,et al. The prioritization of cancer antigens: a national cancer institutepilot project for the acceleration of translational research. Clin CancerRes 2009;15:5323–37.

18. Inoue K, Sugiyama H, Ogawa H, Nakagawa M, Yamagami T, Miwa H,et al. WT1 as a new prognostic factor and a new marker for thedetection of minimal residual disease in acute leukemia. Blood1994;84:3071–9.

19. OgawaH, Tamaki H, IkegameK, Soma T, KawakamiM, Tsuboi A, et al.The usefulness of monitoring WT1 gene transcripts for the predictionand management of relapse following allogeneic stem cell transplan-tation in acute type leukemia. Blood 2003;101:1698–704.

20. Desjarlais JR, Lazar GA, Zhukovsky EA, Chu SY. Optimizing engage-ment of the immune system by anti-tumor antibodies: an engineer'sperspective. Drug Discov Today 2007;12:898–910.

21. Jefferis R. Glycosylation of recombinant antibody therapeutics. Bio-technol Prog 2005;21:11–6.

22. Hodoniczky J, Zheng YZ, James DC. Control of recombinant mono-clonal antibody effector functions by Fc N-glycan remodeling in vitro.Biotechnol Prog 2005;21:1644–52.

23. de Romeuf C, Dutertre CA, Le Garff-Tavernier M, Fournier N, GaucherC, Glacet A, et al. Chronic lymphocytic leukaemia cells are efficientlykilled by an anti-CD20 monoclonal antibody selected for improvedengagement of FcgRIIIA/CD16. Br J Haematol 2008;140:635–43.

24. Masuda K, Kubota T, Kaneko E, Iida S, Wakitani M, Kobayashi-Natsume Y, et al. Enhanced binding affinity for FcgRIIIa of fucose-negative antibody is sufficient to induce maximal antibody-dependentcellular cytotoxicity. Mol Immunol 2007;44:3122–31.

25. Shields RL, Lai J, Keck R, O'Connell LY, Hong K, Meng YG, et al. Lackof fucose on human IgG1 N-linked oligosaccharide improves bindingto human FcgRIII and antibody-dependent cellular toxicity. J BiolChem 2002;277:26733–40.

26. Shinkawa T, Nakamura K, Yamane N, Shoji-Hosaka E, Kanda Y,Sakurada M, et al. The absence of fucose but not the presence ofgalactose or bisecting N-acetylglucosamine of human IgG1 complex-type oligosaccharides shows the critical role of enhancing antibody-dependent cellular cytotoxicity. J Biol Chem 2003;278:3466–73.

27. Davies J, Jiang L, Pan LZ, LaBarreMJ, AndersonD, ReffM. Expressionof GnTIII in a recombinant anti-CD20CHOproduction cell line: expres-sion of antibodies with altered glycoforms leads to an increase inADCC through higher affinity for FcgRIII. Biotechnol Bioeng2001;74:288–94.

28. Umana P, Jean-Mairet J, Moudry R, Amstutz H, Bailey JE. Engineeredglycoforms of an antineuroblastoma IgG1 with optimized antibody-dependent cellular cytotoxic activity. Nat Biotechnol 1999;17:176–80.

29. Boyd PN, Lines AC, Patel AK. The effect of the removal of sialic acid,galactose and total carbohydrate on the functional activity of Cam-path-1H. Mol Immunol 1995;32:1311–8.

30. Veomett N, Dao T, Scheinberg DA. Therapeutic antibodies to intra-cellular targets in cancer therapy. Expert Opin Biol Ther 2013;13:1485–8.

31. Lai L, Alaverdi N,Maltais L,MorseHC3rd.Mouse cell surface antigens:nomenclature and immunophenotyping. J Immunol 1998;160:3861–8.

32. Bosma MJ, Carroll AM. The SCID mouse mutant: definition, charac-terization, and potential uses. Annu Rev Immunol 1991;9:323–50.

33. Shultz LD, Lyons BL, Burzenski LM, Gott B, Chen X, Chaleff S, et al.Human lymphoid and myeloid cell development in NOD/LtSz-scidIL2R gamma null mice engrafted with mobilized human hemopoieticstem cells. J Immunol 2005;174:6477–89.

34. Cox CV, Evely RS, Oakhill A, Pamphilon DH, Goulden NJ, Blair A.Characterization of acute lymphoblastic leukemia progenitor cells.Blood 2004;104:2919–25.

35. Ariyaratana S, Loeb DM. The role of the Wilms tumour gene (WT1) innormal and malignant haematopoiesis. Expert Rev Mol Med 2007;9:1–17.

36. Center TUoTMDAC. Atlas of Laboratory Mouse Histology [cited 2012June 13]. Available from: http://ctrgenpath.net/static/atlas/mousehis-tology/Windows/introduction.html.

37. Cesta MF. Normal structure, function, and histology of the spleen.Toxicol Pathol 2006;34:455–65.

38. Kubota T, Niwa R, SatohM, Akinaga S, Shitara K, Hanai N. Engineeredtherapeutic antibodies with improved effector functions. Cancer Sci2009;100:1566–72.

39. Ishida T, Joh T, Uike N, Yamamoto K, Utsunomiya A, Yoshida S, et al.Defucosylated anti-CCR4 monoclonal antibody (KW-0761) forrelapsedadult T-cell leukemia-lymphoma: amulticenter phase II study.J Clin Oncol 2012;30:837–42.

40. Subramaniam JM, Whiteside G, McKeage K, Croxtall JC. Mogamu-lizumab: first global approval. Drugs 2012;72:1293–8.

41. Gasdaska JR,SherwoodS,Regan JT,Dickey LF. Anafucosylated anti-CD20 monoclonal antibody with greater antibody-dependent cellularcytotoxicity and B-cell depletion and lower complement-dependentcytotoxicity than rituximab. Mol Immunol 2012;50:134–41.

42. Junttila TT, Parsons K, OlssonC, Lu Y, Xin Y, Theriault J, et al. Superiorin vivo efficacy of afucosylated trastuzumab in the treatment of HER2-amplified breast cancer. Cancer Res 2010;70:4481–9.

43. Raghavan M, Bjorkman PJ. Fc receptors and their interactions withimmunoglobulins. Annu Rev Cell Dev Biol 1996;12:181–220.

44. Roopenian DC, Akilesh S. FcRn: the neonatal Fc receptor comes ofage. Nat Rev Immunol 2007;7:715–25.

45. Allavena P, Chieppa M, Monti P, Piemonti L. From pattern recognitionreceptor to regulator of homeostasis: the double-faced macrophagemannose receptor. Crit Rev Immunol 2004;24:179–92.

46. LeeSJ, EversS, RoederD, ParlowAF,Risteli J, Risteli L, et al.Mannosereceptor-mediated regulation of serum glycoprotein homeostasis.Science 2002;295:1898–901.

47. Stahl PD. The mannose receptor and other macrophage lectins. CurrOpin Immunol 1992;4:49–52.

48. Goetze AM, Liu YD, Zhang Z, Shah B, Lee E, Bondarenko PV, et al.High-mannose glycans on the Fc region of therapeutic IgG antibodiesincrease serum clearance in humans. Glycobiology 2011;21:949–59.

49. Cartron G, Dacheux L, Salles G, Solal-Celigny P, Bardos P, ColombatP, et al. Therapeutic activity of humanized anti-CD20 monoclonalantibody and polymorphism in IgG Fc receptor FcgRIIIa gene. Blood2002;99:754–8.

50. MusolinoA,NaldiN,Bortesi B, PezzuoloD,CapellettiM,MissaleG, et al.Immunoglobulin G fragment C receptor polymorphisms and clinicalefficacy of trastuzumab-based therapy in patients with HER-2/neu-positive metastatic breast cancer. J Clin Oncol 2008;26:1789–96.


Veomett et al.




2014;20:4036-4046. Published OnlineFirst May 21, 2014.Clin Cancer Res Nicholas Veomett, Tao Dao, Hong Liu, et al. Intracellular WT1 OncoproteinTherapeutic Efficacy of an Fc-Enhanced TCR-like Antibody to the

Updated version

10.1158/1078-0432.CCR-13-2756doi:

Access the most recent version of this article at:

Material

Supplementary

http://clincancerres.aacrjournals.org/content/suppl/2014/05/23/1078-0432.CCR-13-2756.DC1

Access the most recent supplemental material at:

Cited articles

http://clincancerres.aacrjournals.org/content/20/15/4036.full#ref-list-1

This article cites 49 articles, 20 of which you can access for free at:

Citing articles

http://clincancerres.aacrjournals.org/content/20/15/4036.full#related-urls

This article has been cited by 4 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 at

Permissions

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

.http://clincancerres.aacrjournals.org/content/20/15/4036To request permission to re-use all or part of this article, use this link



http://clincancerres.aacrjournals.org/lookup/doi/10.1158/1078-0432.CCR-13-2756

http://clincancerres.aacrjournals.org/content/suppl/2014/05/23/1078-0432.CCR-13-2756.DC1

http://clincancerres.aacrjournals.org/content/20/15/4036.full#ref-list-1

http://clincancerres.aacrjournals.org/content/20/15/4036.full#related-urls

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

mailto:[email protected]

http://clincancerres.aacrjournals.org/content/20/15/4036


therapeuticefﬁcacyofanfc-enhancedtcr-likeantibodyto the ......for acute lymphoblastic leukemia...

Documents