Characterization of ABT-806, a Humanized Tumor-Speci c ... · development, we humanized mAb 806 to generate ABT-806, a recombinant IgG1/k mAb. In a phase I study in patients with

Large Molecule Therapeutics

Characterization of ABT-806, a HumanizedTumor-Specific Anti-EGFR Monoclonal AntibodyEdwardB. Reilly1, AndrewC. Phillips1, FritzG. Buchanan1,Gillian Kingsbury2,Yumin Zhang1,Jonathan A. Meulbroek1, Todd B. Cole1, Peter J. DeVries1, Hugh D. Falls1, Christine Beam2,Jinming Gu2, Enrico L. Digiammarino1, Joann P. Palma1, Cherrie K. Donawho1,Neal C. Goodwin3, and Andrew M. Scott4

Abstract

Despite clinical efficacy, current approved agents targetingEGFR are associated with on-target toxicities as a consequenceof disrupting normal EGFR function. MAb 806 is a novel EGFRantibody that selectively targets a tumor-selective epitope suggest-ing that a mAb 806-based therapeutic would retain antitumoractivity without the on-target toxicities associated with EGFRinhibition. To enable clinical development, a humanized variantofmAb 806 designated ABT-806was generated and is currently inphase 1 trials. We describe the characterization of binding andfunctional properties of ABT-806 compared with the clinicallyvalidated anti-EGFR antibody cetuximab. ABT-806 binds themutant EGFRvIII with high affinity and, relative to cetuximab,exhibits increased potency against glioblastoma multiforme cellline and patient-derived xenografts expressing this form of thereceptor. ABT-806 also inhibits the growth of squamous cell

carcinoma xenograft models expressing high levels of wild-typeEGFR, associated with inhibition of EGFR signaling, althoughhigher doses of ABT-806 than cetuximab are required forsimilar activity. ABT-806 enhances in vivo potency of stan-dard-of-care therapies used to treat glioblastoma multiformeand head and neck squamous cell carcinoma. An indium-labeled version of ABT-806, [111In]-ABT-806, used to investi-gate the relationship between dose and receptor occupancy,revealed greater receptor occupancy at lowers doses in anEGFRvIII-expressing model and significant uptake in an ortho-topic model. Collectively, these results suggest that ABT-806may have antitumor activity superior to cetuximab in EGFRvIII-expressing tumors, and similar activity to cetuximab in tumorshighly overexpressing wild-type EGFR with reduced toxicity.Mol Cancer Ther; 14(5); 1141–51. �2015 AACR.

IntroductionEGFR is expressed in a majority of human carcinomas and its

frequent overexpression, activation, and mutations in tumors areoften associated with aggressive cancer phenotypes (1). Twomajor classes of EGFR-targeting agents have been developedincluding tyrosine kinase inhibitors, such as gefitinib and erloti-nib, which competitively bind to the ATP pocket of EGFR (2), andmAbs, including cetuximabandpanitumumab, that binddomainIII of EGFR and inhibit ligand binding (3, 4). Despite the successof this class of EGFR antagonists, they are associated with on-target toxicity, most notably a severe papulopustular rash thoughtto be the consequence of disruption of normal EGFR functionin skin (5).

MAb 806 is a novel anti-EGFR antibody that selectively targets aunique epitope of the EGFR which is largely inaccessible whenEGFR is expressed at normal physiologic levels (6, 7). The targeted

epitope is accessible in tumors with wild-type EGFR amplificationor in tumors that express EGFRvIII, the most common deletionmutant of EGFR that lacks the ligand-binding domain of exons2–7 and retains constitutive kinase activity (8–10). ThusmAb806has tumor-specific binding properties and would not be expectedto elicit the side effects, including skin rash, observed with othercetuximab-like EGFR targeting (mAbs) associated with targetingof normal tissues. Consistent with these expectations, in a phase Itrial, a radiolabeled chimeric version of mAb806 (ch806) dem-onstrated tumor uptake in multiple tumor types with minimalevidence of normal tissue uptake (11).

To reduce the risk of immunogenicity and enable clinicaldevelopment, we humanized mAb 806 to generate ABT-806, arecombinant IgG1/k mAb. In a phase I study in patients withadvanced solid tumors, ABT-806 was well tolerated, with a verylow level of cutaneous toxicity and linear pharmacokineticsindicating the absence of extensive EGFR binding in normaltissues (12).

Here, we characterize the binding and functional characteristicsof ABT-806 including assessment of pharmacodynamic tumorchanges in total EGFR tyrosine phosphorylation in response toABT-806. We compared the activity of ABT-806 with cetuximabagainst different human tumor xenografts, including patient-derived xenograft (PDX) models, and correlated outcomes withEGFR expression and genotype. Because its safety profile suggeststhat ABT-806 may be more amenable to higher dosing andcombination with chemotherapy than are other EGFR-directedtherapies, combinations with standard-of-care (SOC) therapy

1AbbVie, Cancer Discovery, North Chicago, Illinois. 2AbbVie Biore-search Center, Worcester, Massachusetts. 3The Jackson Laboratory,Sacramento, California. 4Ludwig Institute for Cancer Research, OliviaNewton-John Cancer Research Institute, and La Trobe University,Melbourne, Victoria, Australia.

Corresponding Author: Edward B. Reilly, AbbVie, Oncology Discovery, R460, 1North Waukegan Road, North Chicago, IL 60064-6099. Phone: 847-937-0815;Fax: 847-938-1336; E-mail: [email protected]

doi: 10.1158/1535-7163.MCT-14-0820

�2015 American Association for Cancer Research.

MolecularCancerTherapeutics

www.aacrjournals.org 1141

on March 20, 2021. © 2015 American Association for Cancer Research. mct.aacrjournals.org Downloaded from

Published OnlineFirst March 2, 2015; DOI: 10.1158/1535-7163.MCT-14-0820

http://mct.aacrjournals.org/

were also investigated. The limited normal tissue-binding prop-erties of ABT-806 also permitted evaluation of [111In]ABT-806tumor uptake in mouse xenograft models to validate its use as anovel molecular imaging technique for patient selection and toinvestigate the relationship between ABT-806 dose and receptoroccupancy.

Materials and MethodsAntibodies and reagents

The soluble ECDs of EGFR were expressed from transientlytransfected HEK293 cells as secreted proteins (inclusive of sig-naling peptide 1–24 that is cleaved during secretion) with a C-terminal LESRGPF-Myc-NMHTG-6His and purified by Ni-IMACand SEC. Details of EGFR protein forms (based on numberingfrom accession# NP_005219.2) are as follows: wild-type EGFRcomprised the entire extracellular binding domain (amino acids1–645); EGFRvIII contained an in-frame deletion generating atruncated protein with a novel glycine (residues 1–29 and G-298-645); EGFR 1-501 (the first 1-525 amino acids inclusive 24 aminoacid leader); and a double mutant EGFRC271A,C283A (containingEGFR 1-645, C295A, C307A) replacing cysteine at 271 and 283 inthe processed protein with alanine (8). ABT-806was produced bytransient transfection of HEK-293-6E cells; cells were grown inFreestyle 293medium (Invitrogen) and transfectedwith plasmidsencoding the light chain andheavy chainof theABT-806 constructas per the Invitrogen Freestyle protocol withminormodifications.Cetuximab (Bristol-Meyer Squibb), 5-FU (GeneraMedix Inc),leucovorin (Bedford Laboratories), cisplatin (Bedford Laborato-ries), temozolomide (Merck & Co.), and human IgG (InnovativeResearch) were purchased.

Cell cultureThe human tumor cell lines A431 (human vulvar squamous

carcinoma), U87MG (human glioma), and U87MGde2-7 (engi-neered fromU87MGtooverexpress EGFRvIII) utilized inpreviouspublished work with mAb806 and ch806 were provided by theLudwig Institute for Cancer Research (Melbourne, Victoria, Aus-tralia) in 2010 (9, 13). SCC15 cells, human head and necksquamous cell carcinoma (HNSCC), were acquired from theATCC in 2002. All cell lineswere expanded in culture upon receiptand cryopreserved to provide cells at similar stage passages for allsubsequent experiments. Cell lineswere not authenticated in the 6months before use; however, their EGFR expression levels wereconfirmed by both FACS and Western blot analysis. A431 cellswere maintained in DMEM/F12 (1:1; Life Technologies) supple-mented with 10% FBS (Life Technologies) and GlutaMAX (LifeTechnologies). U87MG and U87MGde2-7 cells were maintainedin DMEM (Life Technologies) supplemented with 10% FBS andsodium pyruvate (Life Technologies). U87MGde2-7 cells weremaintained under selection with 0.4 mg/mL Geneticin (LifeTechnologies). SCC15 cells were maintained in RPMI-1640 (LifeTechnologies) supplemented with 10% FBS.

FACS analysisCells were harvested from flasks when approximately 80%

confluent usingCellDissociation Buffer (Life Technologies). Cellswerewashedonce inPBS/1%FBS (FACSbuffer) then resuspendedat 2.5� 106 cells/mL in FACS buffer. Of note, 100 mL of cells/wellwere added to a round bottom 96-well plate. Of note, 10 mL ofantibody prepared at 10� concentration (titrated down in half-

log increments from 100 nmol/L) was added, and the plate wasincubated at 4�C for 1 hour. Wells were washed twice with FACSbuffer then resuspended in 50 mL of secondary antibody (AlexaFluor 488, Life Technologies) diluted in 250 mL FACS buffer. Theplate was incubated at 4�C for 1 hour then washed twice withFACS buffer. Cells were then resuspended in 100 mL of PBS/1%formaldehyde and analyzed on a Becton Dickinson LSRII flowcytometer. Data were analyzed using WinList flow cytometryanalysis software (Verity Software House).

Binding ELISAFlat-bottomed, 96-well, high-binding plates (Costar) were

coated with 1 mg/mL of mouse anti-His antibody (Life Technol-ogies) overnight at 4�C. Plates were washed three times with PBSTand blockedwith 250 mL of SuperBlock (Thermo Scientific Pierce)for 2 hour at room temperature. Plates were washed and 100 mL ofHis-tagged human wild-type EGFR, EGFR 1-501, EGFRvIII, orEGFRC271A,C283A were plated at 2 mg/mL for 1 hour at roomtemperature with shaking. After three washes, ABT-806 and cetux-imab were added at 50 and 3 mg/mL, respectively, and 1:3 serialdilutions were prepared. Test antibody was incubated in a 100-mLfinal volume for 1 hour at room temperature with shaking.Following three washes, a secondary goat anti-human IgG-HRPantibody (Thermo Scientific Pierce)was added at 100mL/well andincubated for 1 hour at room temperature with shaking. Plateswere washed three times and developed using 3,30,5,50-tetra-methylbenzidine (TMB) substrate (Thermo Scientific Pierce). Thereaction was stopped with 1N phosphoric acid and OD450 signalwas measured using a SpectraMax (Molecular Devices).

Surface plasmon resonance of antibodiesABiacore T100 surface plasmon resonance instrument (Biacore

Life Sciences) was used to measure binding kinetics of recom-binant soluble EGFR (wild-type, EGFRvIII, EGFR 1-501,EGFRC271A,C283A) protein forms (analytes) binding to anti-EGFRmAbs (ligands). The assay format was Fc-based capture viaimmobilized anti-human (Fc) antibody (Thermo ScientificPierce). A standard amine coupling protocol was used to immo-bilize the capture reagents via primary amines to the carboxy-methyl (CM)dextran surface ofCM5 sensor chips according to themanufacturer's instructions (Biacore Life Sciences). For bindingkinetics measurements, the assay buffer was HBS-EPþ (BiacoreLife Sciences): 10 mmol/L Hepes (pH 7.5), 150 mmol/L NaCl, 3mmol/L EDTA, 0.05% P20. During the assay, all measurementswere referenced against the capture surface alone. Each assaycycle consisted of the following steps: (i) ligand capture; (ii)analyte injection over both reference and test surface, 240 mL at80 mL/minute, after which the dissociation was monitored for 15minutes at 80 mL/minute; and (iii) regeneration of capture surfacewith low pH glycine. For kinetic determinations, analyte injec-tions were randomized 3-fold dilution series from 3 mmol/L to12.35 nmol/L. Buffer only injections were included for secondaryreferencing. Data were processed and fit to a 1:1 binding modelusing Biacore T100 Evaluation Software to determine the bindingkinetic rate constants, ka (on-rate) and kd (off-rate), and the KD

(equilibrium dissociation constant, also referred to as "affinity").

In vivo studiesAll animal studies were reviewed and approved by AbbVie's

Lake County Institutional Animal Care and Use Committee.Animal studieswere conducted in anAAALACaccredited program

Reilly et al.

Mol Cancer Ther; 14(5) May 2015 Molecular Cancer Therapeutics1142




and veterinary care was available to ensure appropriate animalcare. Mice were group-housed 10 per cage with food and wateravailable ad libitum and were acclimated to the animal facilitiesfor a period of at least one week before initiation of experiments.

Tumor cells were mixed with 50% Matrigel (BD Biosciences)and injected s.c. into the flank of 6- to 8-week-old female mice(Charles River Laboratories) as described: SCC15 cells wereinjected at 1 � 106 cells per mouse into SCID Beige mice;A431 cells were injected at 3 � 106 cells per mouse into Nu/Numice; U87MGde 2–7 cells were injected at 3� 106 cells permouseinto Nu/Nu mice. For the SN0199 PDX model (Jackson Labora-tory), 3 to 5mm3passage 3 (P3) tumor fragmentswere s.c. trocharimplanted in the right rear flank of NSG mice.

Tumors were allowed to grow to a predetermined size, at whichtimemicewere allocated by tumor volume into study groups (n¼10 mice/group) so that the mean tumor volumes of the groupswere statistically similar. Mice were then entered into the dosingphase of the study (described below). The PDX tumors wererandomly assigned into study groups (n¼ 7–9mice/group) whenthe individual tumors reached approximately 250 mm3 volume.Tumor volumes were recorded two to three times per week basedon tumor growth rate. Tumor volumes were estimated by theformula V ¼ (L � W2)/2, where V is the volume (mm3), L is thetumor length (mm), andW is the tumorwidth (mm),measured atright angles using a digital caliper. The effect of a treatment ontumor growth inhibition (TGI) was determined as %TGI ¼ 1 �(mean tumor volume of treatment group/tumor volume of treat-ment control group)� 100. Datawere analyzed using the Studentt test for % TGI values [Excel (Windows XP), MicrosoftCorporation]. The tumor growth delay method for efficacy anal-ysis was performed using data achieved with a 60-day studyendpoint or a 1,500 mm3 tumor endpoint for calculating time(days) to tumor endpoint (14). A log-rank test with a two-tailedstatistical analysis with a 95% confidence interval was performedwith GraphPad Prism 6.0 to determine the significance of com-parisons between treatment groups for the time to endpoint (TTE)tumor responses and for plotting Kaplan–Meier survival curves.

Western blot analysisCell lysates (10 mg) were resolved by SDS-PAGE using 4% to

12% Bis-Tris midi gels (Invitrogen, catalog no. WG1402BX10)and transferred to nitrocellulose membranes using an iBlot DryTransfer System (Invitrogen, catalog number IB3010-01). Anti-phosphotyrosine blots were blocked with 3% milk/PBS for 30minutes, washed three times with PBS-T, and then incubatedovernight with anti-phosphotyrosine 4G10 biotin conjugate(Millipore, catalog number16-103; 1:1000) at 4�C. Blots werethen washed twice with water then incubated with 1:1000 strep-tavidin-HRP (KPL, catalog number 474-3000) for 1 hour at roomtemperature, washed twice in water, once in PBS-T, and thenincubated for 3.5 hours in water (changing water several times).Blots were treated with Super Signal West Dura Extended Dura-tion chemiluminescent substrate (Thermo Scientific, productnumber 34076) and were visualized by scanning using an LAS-4000. Total EGFR blots were blocked with 5% milk/TBS-T for 1hour and incubated overnight with rabbit anti-EGFR (LifespanBiosciences, catalog number LS-C6625; 1:2000). Blots were thenwashed three times with TBS-T for 5 minutes and then incubatedwith donkey anti-rabbit IgG-HRP secondary antibody (The Jack-son Laboratory, catalog number 711-0152; 1:2000) for 1 hour atroom temperature. Blots were thenwashed three timeswith TBS-T

for 5 minutes, then treated with chemiluminescent substrate(Thermo Scientific) and were visualized by scanning using anLAS-4000.

IHCIHC analysis of EGFR, phospho-EGFR, phospho-H3, and acti-

vated caspase-3 levels in tumors fromhuman IgG-, cetuximab-, orABT-806-treated mice was evaluated. Briefly, 5 m paraffin sectionswere deparaffinized and rehydrated. Antigen retrieval was per-formed, endogenous peroxidase and nonspecific protein-bindingsites were blocked [Peroxidase Blocking Reagent (Dako); Back-ground Sniper (Biocare)] and subsequently incubated in primaryantibodies (1 hour at room temperature) and detected withEnVisionþ polymer for mouse or rabbit with DAB as chromogen(Dako). Primary antibodies used included: EGFR mAb (LifeTechnologies; catalog number AHR5062); phospho-EGFR pAb(Cell Signaling Technology; catalog number 2234); rabbit anti-human mouse and rat phospho-histone H3 pAb proliferationmarker (Cell Signaling Technology; catalog number 9701); rabbitanti-human/multi-species, and activated caspase-3 pAb apoptosismarker (Cell Signaling Technology; 9661). IHC results wereobtained by semiquantitative assessment of the percent positivecells within viable areas across treatment groups at various timepoints. Data represent four samples per treatment group, per timepoint. Data were analyzed using ANOVA/Fisher PLSD analysis(Statview, SAS Institute).

DTPA conjugationCyclic diethylenetriamine pentaacetic acid dianhydride

(cDTPA) was used as bifunctional chelator for preparation ofDTPA-ABT-806. ABT-806 was transferred into 0.1 mol/L sodiumbicarbonate (pH 8.2) by size-exclusion chromatography andultrafiltration. A 10-fold molar excess of DTPA was added to theantibody and incubated for 1 hour (15). Unbound DTPA wasremoved by size-exclusion chromatography and ultrafiltrationand the DTPA-antibody was transferred into 0.1 mol/L sodiumacetate (pH 7.2). The resultant DTPA antibody was aliquoted andstored at�80�C for radiolabeling. All DTPA-antibody conjugatescontained an average of two to three DTPA groups per antibody.Cetuximab, ABT-806, and human polyclonal IgG (hIgG) wereused as controls and were conjugated to cDTPA in parallel withABT-806 using the same procedure. For DTPA-ABT-806 andcetuximab, the ability to retain a similar affinity as the parentalantibody was verified by FACS-based cell binding assays beforeuse in imaging experiments.

RadiolabelingFor mouse imaging studies, 1 mg of DTPA-antibody (0.4 mL)

wasmixedwith 40mCi 111InCl3 (20–100 mL;MDSNordion). Thereaction vial was swirled for 10 seconds and incubated for 30minutes before analytical testing. The labeling efficiency andradiochemical purity were determined by both radio-sizeexclusion–high-performance liquid chromatography and instantthin layer chromatography with 100 mmol/L acetate buffer/20mmol/L EDTA as the mobile phase. Labeling reactions producedbetween 95% and 99% incorporation of 111In into DTPA-anti-body, and no postlabeling purification was needed.

SPECT/CT imagingA SPECT scanner with a built-in CT dedicated to small animal

imaging was used (nanoSPECT/CT, BioScan). The nanoSPECT

ABT-806: A Tumor-Selective EGFR-Targeting Antibody

www.aacrjournals.org Mol Cancer Ther; 14(5) May 2015 1143




contains four detectors; each fitted with a tungsten collimator anda nine-pinhole aperture for increased spatial resolution (<1 mm)anddetection sensitivity. Tumorswere sizematchedusing calipersto ensure comparable tumor sizeswithin each experimental groupof mice. After injection of [111In]antibody (2 mCi/50 mg), micewere imaged under isoflurane (2%) anesthesia with CT andSPECT sequentially. CT and SPECT images were reconstructed,coregistered and the tumor biodistribution of the [111In]antibodywas measured using InVivoScope software (Bioscan Inc.). Allimages were quantified as percentage of injected dose per cm3

(%ID/cc) or percentage of injected dose per cm3 of the maximumvoxel (%IDmax/cc) within the tumor.

ResultsCharacterization ABT-806–binding properties

A series of binding assays was performed to characterizethe binding of ABT-806 to recombinant EGFR. In an ELISAformat, ABT-806 exhibits higher binding to EGFRvIII andEGFRC271A,C283A, a point mutant known to expose the cryptic806 epitope, relative to truncated wild-type EGFR (1-501) andfull-lengthwild-type EGFRECD(1-621; 8). In contrast, cetuximabdisplays tight binding to all formsof recombinant EGFR (Table 1).

Affinity (KD)measurements using Biacore analysis confirm thisdifferential binding (Table 1). Cetuximab exhibits nearly equiv-alent high-affinity binding to all four EGFR forms (�2nmol/LKD)in Biacore, whereas ABT-806 shows differential binding kineticsfor EGFRvIII, EGFRC271A,C283A, and EGFR-truncated ECD (1-501)and no observable binding to full-length EGFR ECD (1-621).ABT-806 shows highest affinity for EGFRvIII (18 nmol/L KD)followed by EGFRC271A,C283A (170 nmol/L KD) and EGFR1-501(16 mmol/L KD). The high affinity of ABT-806 for EGFRvIII isdue to enhanced on-rates and off-rates compared withEGFRC271A,C283A and EGFR1-501, while these latter two formsdiffer in off-rates only. These observations are consistent withthe EGFRvIII having unique epitope exposure (Table 1). Theincreased ABT-806 binding observed in the ELISA format com-pared with the Biacore format is most likely due to an antibodyavidity effect. Biacore measures intrinsic 1:1 binding of recom-binant soluble EGFR forms to antibody without additional avid-ity type interactions and surface effects.

FACS was used to compare the binding of ABT-806 andcetuximab to A431 cells that express amplified wild-type EGFR

(16). Nonsaturable binding of ABT-806 to A431 cells wasobserved even at high concentrations (1.3 mmol/L; Fig. 1A). Incontrast, cetuximab binding was saturated at single digit nmol/Lconcentrations (Fig. 1A). This binding analysis was extended to apanel of human tumor cell lines expressing wild-type EGFR. In allcases, ABT-806 binding was consistently lower than cetuximabindicating that, at least in vitro, the majority of cell surface wild-type EGFR receptors remain unbound (Fig. 1B and data notshown). In contrast, ABT-806 effectively bound EGFRvIII-expres-sing U87MGde2-7 cells similar to the binding observed forcetuximab (Fig. 1C). Both ABT-806 and cetuximab displayedlimited binding to parental U87MG cells which express low levelsof wild-type EGFR (Fig. 1D).

ABT-806 in vivo potency in EGFRvIII-expressing glioblastomamultiforme tumors

The binding characteristics of ABT-806 suggest that it may beeffective against tumors expressing EGFRvIII. Because glioblas-toma multiforme (GBM) has a high frequency of EGFRvIIIexpression, the activity ABT-806 was compared with cetuximabin the U87MGde2-7 GBM model. This tumor model wasengineered to express EGFRvIII as cell lines do not maintainexpression of endogenous EGFRvIII (17, 18). Growth ofU87MGde2-7 tumors was significantly inhibited by ABT-806treatment (Fig. 2A). In contrast, cetuximab showed only min-imal activity in this model even at the 40 mg/kg high dose leveltested (Fig. 2A).

These studies were extended to the evaluation of ABT-806activity against the EGFRvIII-positive PDX model SN0199. GBMPDX models like SN0199, unlike standard cell line xenograftmodels, can maintain expression of endogenous EGFRvIII andthus may be more clinically relevant than standard cell line-derived xenografts (17, 19, 20). Because of the variable growthrate of the implanted PDX tumors, these studies were performedwith an accrual design with data presented as a Kaplan–Meierplot. A 10 mg/kg dose of cetuximab was selected as the compar-ator since the exposures achieved with this dose would beexpected to exceed those achievable in patients (21). In contrast,ABT-806 was assessed at doses of both 10 and 40 mg/kg (dosedthree times a week for 2 weeks) because reduced normal tissuebinding of ABT-806 compared with cetuximab has been demon-strated thereby allowing higher tolerated doses in patients. Asshown in Fig. 2B, ABT-806 treatment resulted in significant tumor

Table 1. Binding affinity of ABT-806 and cetuximab to recombinant EGFR

ELISA (EC50, nmol/L)a

ABT-806 Cetuximab

EGFR wild-type 0.96 � 0.34 0.010 � 0.000EGFRC271A,C283A 0.09 � 0.04 0.017 � 0.006EGFRvIII 0.12 � 0.04 0.013 � 0.006EGFR 1-501 0.66 � 0.14 0.023 � 0.006

Surface plasmon resonance (binding constants)b

ABT-806 Cetuximabka (1/Ms) kd (1/s) KD (M) ka (1/Ms) kd (1/s) KD (M)

EGFR wild-type No observable binding 4.9Eþ05 (8.8Eþ04) 1.2E�03 (5.8E�06) 2.5E�09 (4.9E�10)EGFRC271A,C283A 7.0Eþ03 (5.8Eþ01) 7.6Eþ04 (6.9E�05) 1.1E�07 (9.5E�09) 4.1Eþ05 (9.3Eþ04) 1.2E�03 (1.2E�06) 3.0E�09 (7.6E�10)EGFRvIII 4.5Eþ04 (3.2Eþ03) 8.5Eþ04 (8.3E�05) 1.9E�08 (3.0E�09) 8.1Eþ05 (1.0Eþ05) 1.4E�03 (1.4E�04) 1.7E�09 (4.0E�11)EGFR 1-501 3.6Eþ03 (2.6Eþ02) 7.1Eþ05 (7.7E�04) 2.0E�06 (1.1E�07) 8.3Eþ05 (1.0Eþ05) 1.4E�03 (1.5E�05) 1.7E�09 (2.2E�10)aEC50 values derived from ELISA in which EGFR ECD was captured on the plate via a His tag. Values are averages of three experiments �SD.bEGFR: 6-point, 3-fold dilution series from 3 mmol/L to 12.35 nmol/L, fit to a 1:1 binding model using Biacore T100 Evaluation Software. Binding constants are theaverage of three experiments (SD in parenthesis).

Reilly et al.





inhibition with all mice surviving beyond the end of the obser-vation period. A second PDX GBM model (SN0207) expressinglow levels of wild-type EGFR was unresponsive to either ABT-806or cetuximab treatment (data not shown).

To investigate whether TGI of the SN0199 model was associ-ated with an inhibition of EGFR signaling, tumor-bearing micewere treated with ABT-806 (40 mg/kg) or cetuximab (10 mg/kg)twice 3 days apart and tumors harvested 72 hours after the finaldose. Total and pEGFR status in the SN0199 tumor was assessedby Western blot analysis (Fig. 2C). ABT-806, but not cetuximab,reduced levels of pEGFR consistent with the potency of ABT-806in EGFRvIII tumor models.

ABT-806 in vivo potency in squamous cell carcinoma tumormodels expressing wild-type EGFR

ABT-806–binding properties and published results with mAb806 and ch806 suggested that ABT-806 may also be effectiveagainst tumor cells overexpressing wild-type EGFR (6, 13, 16).Because EGFR is frequently overexpressed in squamous tumors,the antitumor efficacy of ABT-806 was compared with cetuximabin the A431 squamous xenograft model that expresses amplifiedEGFR (22). Despite poor in vitro binding and an inability toinhibit signaling of wild-type EGFR-expressing cells (data notshown), ABT-806 dosed at 10 mg/kg three times per week for 2weeks exhibited comparable activity to cetuximab dosed in anequivalent manner (Fig. 3A). At higher doses (40 mg/kg), cetux-imabwasmore effective at inhibiting tumor growth thanwasABT-806 dosed (87% TGI for cetuximab and 58% TGI for ABT-806 onday 30; Fig. 3B). These results suggest that higher exposures ofABT-806 relative to cetuximabmay be required to achieve similar

efficacy and are consistent with the higher affinity binding ofcetuximab to wild-type EGFR (Table 1).

ABT-806was also highly effective in aHNSCC xenograft modelSCC15 that overexpresses wild-type EGFR (Fig. 3C). ABT-806treatment resulted in tumor regressions although higher dosingwas required to show activity comparable with that seen withcetuximab (Fig. 3C).

ABT-806 downregulates wild-type EGFR and pEGFR in vivoTo investigate the mechanism of ABT-806–mediated TGI in

wild-type EGFR-expressing tumors, SCC15 tumor-bearing micewere treated with a single dose of ABT-806 or cetuximab andtumors harvested over a time course. IHC analysis of sectionsderived from tumors harvested over 120 hour demonstrated thatABT-806 mediated time-dependent reductions in the level ofpEGFR and total EGFR (Fig. 4A). Consistent with the antitumoreffects observed in Fig. 3C, a more pronounced effect wasobserved following cetuximab treatment. Representative IHCimages of tumor sections collected at the 120-hour time pointare shown in Fig. 4B. Similar levels of human IgG staining wereobserved in tumors treated with either ABT-806 or cetuximab(data not shown). Both ABT-806 and cetuximab-treated samplesalso demonstrated reduced downstream EGFR signaling and cellproliferation, asmeasured by phospho-histoneH3, and increasedapoptosis as measured by caspase-3 cleavage (Fig. 4C).

ABT-806 combinations with chemotherapyThe safety profile of ABT-806 makes it an attractive candi-

date for combination with chemotherapy in the treatment ofEGFR-positive malignancies. To test this premise, ABT-806 in

0.000

001

0.00

001

0.00

010.0

01 0.01

0.11 10 10

00

100

200

300

400

500

Cetuximab

ABT-806

mAb (nmol/L)

Geo

mea

n

U87MGde2-7

0.000

001

0.000

01

0.000

10.

001

0.01 0.1

1 10 100

0

100

200

300

400

500

Cetuximab

ABT-806

mAb (nmol/L)G

eo m

ean

U87MG

0.000

01

0.000

10.0

01 0.01 0.1 1 10 10

00

5,000

10,000

15,000

20,000 Cetuximab

ABT-806

mAb (nmol/L)

Geo

mea

n

A431 SCC15

0.000

001

0.000

01

0.00

010.0

01 0.01

0.11 10 10

00

5,000

10,000

15,000

20,000

25,000CetuximabABT-806

mAb (nmol/L)

Geo

mea

n

A B

C D

Figure 1.FACS analysis of ABT-806 andcetuximab binding to human tumorcell lines. The binding of ABT-806and cetuximab to wild-typeoverexpressing EGFR-positive tumorlines A431 and SCC15 (A and B) and tothe EGFRvIII-expressing cell lineU87MGde2-7 (C) was assessed byFACS analysis. The U87MG tumor cellline (D) expresses low level of wild-type EGFR. Values plotted aregeometric means.






combination with different chemotherapy agents was assessedin both wild-type EGFR and EGFRvIII-expressing tumormodels.

Cetuximab is FDA-approved for use in combination withcisplatin and 5-FU in recurrent or metastatic HNSCC, so thesecombinations with ABT-806 were evaluated in the wild-typeEGFR-expressing HNSCC tumor model SCC15 (23). BecauseABT-806 was highly potent as monotherapy in this model (Fig.2C), it was dosed at a suboptimal level (10 mg/kg) to assess itsability to enhance activity of SOC therapy. The combination ofABT-806 and cisplatin (Fig. 5A) or ABT-806 and 5FU (Fig. 5B)was

more effective in inhibiting SCC15 tumor growth thanwere anyofthe agents tested individually. Increased antitumor activity wasalso observed when ABT-806 was combined with both cisplatinand 5-FU (Fig. 5C).

The ability of ABT-806 to combine with SOC chemotherapywas also evaluated in the EGFRvIII-expressing U87MGde2-7xenograft model. The current SOC therapy for newly diagnosedglioblastoma following surgical debulking is radiation therapy(RT) in combination with temozolomide (24, 25). ABT-806 wasdosed at 10 mg/kg in this model to enable evaluation of combi-nation effects. Addition of ABT-806 to RT resulted in a modest

A431 A431 SCC15

0

500

1,000

1,500

2,000

2,500

3,000

0 5 10 15 20 25 30

Control

Mea

n t

um

or

volu

me

(mm

3 )

Day after tumor cell injection

ABT-806

Cetuximab

0

500

1,000

1,500

2,000

2,500

3,000

0 5 10 15 20 25 30 35 40 45 50 55 60 65 70

Control

Mea

n t

um

or

volu

me

(mm

3 )


ABT-806Cetuximab

0

500

1,000

1,500

2,000

2,500

3,000

5 10 15 20 25 30 35 40 45 50 55 60 65 70

Mea

n t

um

or

volu

me

(mm

3 )


ABT-806 at 40 mkdABT-806 at 20 mkdABT-806 at 10 mkdABT-806 at 2 mkd

ControlCetuximab at 5 mkd

A B C

Figure 3.Antitumor activity of ABT-806 in wild-type EGFR-expressing squamous cell carcinoma xenograft models. The in vivo potency of ABT-806 and cetuximab wasevaluated inmice implantedwithA431 cells (A andB) or SCC15 cells (C). AntibodiesweredosedQD, 3X/week� 2weeks i.p. at 10mg/kg (A) and40mg/kg (B)with alltreatments well tolerated.

A U87MGde2-7

B SN0199

0

500

1,000

1,500

2,000

2,500

3,000

0 5 10 15 20 25 30 35

Mea

n tu

mor

vol

ume

(mm

3 )


ABT-806CetuximabControl

0 5 10 15 20 25 30 35 40 45 50 55 600

102030405060708090

100

Hu IgG control at 40 mkdCetuximab at 10 mkdABT-806 at 10 mkdABT-806 at 40 mkd

Day after initiation of treatment

Perc

ent s

urvi

val

C SN0199

pY EGFR

Tot EGFR

Figure 2.ABT-806 is more potent thancetuximab at inhibiting EGFRvIII tumorgrowth and pEGFR expression.The in vivo potencies of ABT-806 andcetuximab, both dosed at 40mg/kg ina QD, 3X/week� 2-week format, wereevaluated in U87MGde2-7 tumor-bearing mice (A). ABT-806 andcetuximab, both dosed in a 3�/week� 2 week format, were evaluated in aEGFRvIII-expressing PDX tumormodel. The SN0199 tumor model wasperformed with an accrual design anddata are presented as a Kaplan–Meierplot, with a tumor size of 1,500 m3

defined as the end point (B). Finally,total and pEGFR status in SN0199tumors were assessed followingtreatment of mice with ABT-806 at 40mg/kg and cetuximab at 10 mg/kg.Antibodieswere administeredQ3D�2,with tumors harvested 72 hour afterthe final dose (C).

Reilly et al.





increase in TGI (data not shown); however, the addition ofABT-806 to temozolomide produced pronounced increase in TGI(Fig. 5D).

Specific tumor uptake and receptor occupancy measurementsusing [111In]ABT-806

ABT-806 was labeled with 111Indium to evaluate ABT-806tumor uptake and to establish an ABT-806 dose-versus-receptoroccupancy relationship. Mice bearing U87MGde2-7 xenograftswere predosed with unlabeled ABT-806 at 0, 10, 20, 40, and 80mg/kg. One hour later, [111In]ABT-806 was injected into eachanimal. Mice were imaged 72 hours postinjection of tracerand tumor uptake was quantified. Unlabeled ABT-806 pretreat-ment dose dependently inhibited [111In]ABT-806 tumor uptakeas shown in representative SPECT/CT images (Fig. 6A). The%IDmax/cc was plotted against the amount of predosed unlabeledABT-806, revealing significant inhibition of uptake even with asingle dose of 10mg/kg (Fig. 6B). The %IDmax/cc was selected forthis calculation to avoid inclusion of necrotic areas of the tumorvoid of target antigen. A parallel imaging experiment was alsoperformed to establish a cetuximab dose-versus-receptor occu-pancy relationship using theU87MGde2-7model. In comparisonwith the experiment with ABT-806, higher levels of unlabeledcetuximab pretreatment were required to inhibit [111In]cetuxi-mab tumor uptake (Fig. 6C).

To evaluate tumor uptake in a wild-type EGFR-positive model,similar experiments were performed using mice bearing A431xenografts. Tumor uptake of [111In]ABT-806was dose dependent-ly inhibited bypretreatmentwithunlabeledABT-806 (Fig. 6D andE). Similarly tumor uptake of [111In]cetuximab was dose depen-dently inhibited by pretreatment with unlabeled cetuximab (Fig.6F). These results demonstrate that a higher dose of ABT-806would be needed (Fig. 6E, extrapolated dotted line) to reach thesimilar level of inhibition to that of cetuximab in this wild-typeEGFR-positive tumor model (Fig. 6F).

The blood brain barrier (BBB) is a challenge to antibodytherapy of GBM, potentially restricting access to tumors locatedwithin the brain. To evaluate the ability of ABT-806 to penetratethe BBB,micewere injected intracranially withU87MGde2-7 cellsand following tumor development dosed intravenously with[111In]ABT-806. Tumor uptake of the radiolabeled antibody,measured by SPECT/CT imaging was observable at 4 hourwith maximal uptake observed at 120 hour post [111In]ABT-806 dose (Fig. 6G). In contrast, there was no significant uptakeof [111In]ABT-806 in mice that received sham tumor implants(Fig. 6G). These result demonstrates efficient [111In] ABT-806uptake in this glioma model when grown orthotopically and isconsistentwith the utility of ABT-806 treatment in theGBMsetting.

DiscussionABT-806 binds to EGFRvIII with high affinity and exhibits

potent antitumor activity both against glioblastoma cell line andPDX models that express this form of the receptor. PDX modelsmaymore accurately recapitulate the cellular heterogeneity, archi-tectural and molecular characteristics of the primary humantumor compared with standard cell line-passaged xenograft mod-els (26). PDX models are particularly relevant for assessingEGFRvIII-expressing tumors because standard GBM tumor celllines show loss of EGFRvIII expression during passage in cellculture (17–19). In contrast with ABT-806, cetuximab binds toEGFRvIII with similar affinity but does not inhibit signaling andshowed little or no activity in the EGFRvIII-expressing xenograftand PDX models. The different epitopes that these antibodiesrecognize underlie their distinct mechanisms of action and con-sequent efficacies against human xenograft tumors overexpressingEGFRvIII. Previous studies suggest that binding to the ABT-806epitope blocks receptor dimerization and subsequent activation(8). In contrast, cetuximab exerts its antitumor activity, at least inpart, by binding to its epitope on domain III of the EGFR

Figure 4.ABT-806 downregulates tumor wild-type EGFR and pEGFR in SCC15 tumors grown as implants in mice. EGFR, pEGFR, p-H3, and caspase-3 IHC were evaluated intumors harvested at different time periods following a single dose (40 mg/kg) of ABT-806, cetuximab, and control human IgG (A and C). EGFR and pEGFRIHC images of representative samples derived fromharvested tumors 120 hours after treatment (B). IHC resultswere obtained by semiquantitative assessment of thepercent positive cells within viable areas across treatment groups at various time points. Data represent four samples per treatment group, per time point.Data were analyzed by ANOVA (PLSD; � , P < 0.05 vs. human IgG control group; �� , P < 0.05 vs. cetuximab treatment group). Magnification: �400.






preventing ligandbinding (27). Because EGFRvIII is constitutivelyactive, cetuximab does not directly block signaling from thisreceptor. In support of this conclusion ABT-806, but not cetux-imab, blocked EGFRvIII phosphorylation in vitro and in vivo.

EGFR is overexpressed in approximately 50% of primary GBMpatients (28) and the high prevalence of EGFRvIII in GBM (25%–

35%) is well established (29). GBM patients generally have a verypoor prognosis, making EGFR an attractive therapeutic target inthis indication. The EGFR expression profile in GBM suggests thattargeting bothwild-type EGFR andEGFRvIII-expressing tumors inan appropriately stratified GBM patient population may provideclinical benefit particularly when combined with other therapeu-tics. Toward this end, ABT-806 synergizes with SOC temozolo-mide chemotherapy in the U87MGde2-7 tumor model. Theability of ABT-806 to cross the BBB and specifically target glio-blastoma in patients is supported by the selective imaging of[111In]ABT-806 in a GBM orthotopic mouse model. Phase Iclinical observations with both radiolabeled ch806 and ABT-806 also indicate excellent uptake inGBMpatients demonstratingthat this antibody can effectively cross the BBB or that the BBB issufficiently comprised in these patients to allow antibody uptake(11, 30).

In contrast with its high-affinity binding to EGFRvIII, ABT-806displays lower affinity binding to wild-type EGFR with nonsatur-

able binding to cells overexpressing wild-type EGFR even at highconcentrations. Despite its low binding to wild-type EGFR, ABT-806 is efficacious in the wild-type EGFR-expressing A431 andSCC15 squamous carcinoma xenograft models. The reason forefficacy against wild-type EGFR-expressing tumor models in vivodespite low binding in vitro remains unknown but may beexplained by an increased prevalence of the ABT-806 epitope invivo. Although antibody effector functions may contribute to theantitumor activity, ABT-806 is also effective at inhibiting EGFRphosphorylation and downstream signaling in vivo, consistentwith an antitumor effect by inhibition of EGFR signaling.

Cetuximab was more potent than ABT-806 at inhibiting tumorgrowth and downstream signaling in these EGFR wild-type in vivomodels. An important consideration when extrapolating thesepreclinical results to human patients is that reduced normal tissuebinding of ABT-806 compared with cetuximab has been demon-strated thereby allowing higher tolerated doses and subsequenthigher plasma levels of ABT-806 in patients compared withcetuximab. This outcome is supported by phase I studies in whichABT-806 dosed up to 24 mg/kg was well tolerated, with none ofthe characteristic EGFR-inhibitor skin toxicity observed (12, 30).These results suggest that for tumors with high levels of EGFRexpression, ABT-806 may achieve similar potency to cetuximabwith reduced toxicity. ABT-806 was ineffective in several

0

500

1,000

1,500

2,000

2,500

3,000

5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85

Mea

n t

um

or

volu

me

(mm

3 )


ABT-806 + cisplatin

ABT-806CisplatinControl

0

500

1,000

1,500

2,000

2,500

3,000

5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85M

ean

tu

mo

r vo

lum

e (m

m3 )

Mea

n t

um

or

volu

me

(mm

3 )

Mea

n t

um

or

volu

me

(mm

3 )Day after tumor cell injection

ABT-8065-FUControl

ABT-806 + 5-FU

0

500

1,000

1,500

2,000

2,500

3,000

5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85


ABT-8065-FU + CisplatinControl

ABT-806 + 5-FU + Cisplatin

0

500

1,000

1,500

2,000

2,500

3,000

5 10 15 20 25 30 35 40 45

Control


ABT-806 + TemozolomideABT-806Temozolomide

A B

C D

SCC15 SCC15

SCC15 U87MGde2-7

Figure 5.ABT-806 augments the effect ofSOC in SCC15 HNSCC tumor-bearingmice (A–C), ABT-806 wasadministered i.p. at 10 mg/kg (QD,3X/week � 2 weeks). ABT-806 wasdosed in combination with cisplatinat 5 mg/kg (i.v., Q7D � 2; A), incombination with 5-FU at 50mg/kg/d (i.p., QD, 2 days on/5 daysoff, 2 cycles; B) and in combinationwith cisplatin at 5 mg/kg (i.v., QD� 1and 5-FU at 25 mg/kg (i.p., QD � 2;C). In B and C, 5-FU was co-dosed onthe same schedule with leucovorini.p. at 30 and 15 mg/kg, respectively.Against U87MGde2-7 tumor-bearingmice, ABT-806 was combinedwith temozolomide at 5 mg/kg(PO, QD � 5; D).

Reilly et al.





cetuximab-responsivemodels with lower levels of EGFR (data notshown) suggesting that this therapeutic may not be optimal fortreatment of tumors with low EGFR expression.

EGFR overexpression has been well documented in tumorswith squamous histology suggesting that patients with tumorssuch as HNSCC where other EGFR therapeutics have provensuccessful may be well suited for ABT-806 therapy. ABT-806 mayprovide additional clinical benefit in these indications because itsclean toxicity profile suggests that it may be better tolerated thanother EGFR antagonists when combined with chemotherapy inpatients. Results showing additive effects of ABT-806 and SOCchemotherapy regimens in the SCC15 HNSCC tumor modelsupport this strategy.

The tumor specificity of ABT-806 provides a unique opportu-nity to develop a clinical imaging agent for the identification andcharacterization of EGFR-expressing malignancies where other

EGFR antibodies may be limited by their normal tissue bindingproperties. Specific tumor targeting of [111In]ABT-806 of wild-type EGFR and EGFRvIII-expressing tumors implanted into micehas been demonstrated and used to investigate the relationshipbetween ABT-806 and cetuximab dose and receptor occupancy.In the EGFRvIII-expressing xenograft model, a single dose of 10mg/kg ABT-806 or 28mg/kg of cetuximab achieved 50% receptoroccupancy. These results are consistent with the high-affinitybinding of ABT-806 to the EGFRvIII form of the receptor and thepotent antitumor activity of ABT-806observed in theU87MGde2-7 xenograft model. In contrast, ABT-806 required a dose approx-imately twice that of cetuximab to achieve similar receptoroccupancy in the A431 wild-type EGFR-expressing tumormodel. Extrapolation of these results to patients predicts that ahigher dose of ABT-806 may be required to achieve similarreceptor occupancy and efficacy as cetuximab in wild-type

4

0 24 48 72 96 120 1440

20

40

60

80

100

Time after injection (h)

Tum

or u

ptak

e (%

IDm

ax/c

c)

24 72 120 72

Time after injection (h)

U87MGde2-7 implant Sham implant

Rad

ioac

tivity

sca

le

20 40 8010

Cold ABT-806 predose (mg/kg)

0 50 100 150 2000

20

40

60


Tum

or u

ptak

e (%

IDm

ax/

cc)

0 25 50 75 1000

10

20

30

40

50

60

Cold cetuximab predose (mg/kg)

Tum

or u

ptak

e (%

IDm

ax/

cc)

Rad

ioac

tivity

sca

le

0 25 50 75 1000

10

20

30

40

50

60

70

80

Tum

or u

pake

(%

IDm

ax/

cc)


0 25 50 75 1000

10

20

30

40

50

60

70

80

Cold cetuximab predosed (mg/kg)

Tum

or u

pake

(%

IDm

ax/c

c)

20 40 8010


Rad

ioac

tivity

sca

le

A B C

D E

G

F

H

Figure 6.Tumor uptake of [111In]ABT-806 and [111In]cetuximab in U87MGde2-7 and A431 tumor-bearing mice. SPECT/CT images of U87MGde2-7 (A) or A431 (D) tumor-bearing mice predosed with unlabeled ABT-806 before the injection of [111In]ABT-806. Images at 72-hour after [111In]ABT-806 injection. Dose-dependentinhibition of [111In]ABT-806 uptake in U87MGde2-7 (B) or A431 (E) tumors by pretreatment with unlabeled ABT-806. Dose-dependent inhibition of [111In]cetuximabuptake in U87MGde2-7 (C) or A431 (F) tumors by pretreatment with unlabeled cetuximab. SPECT/CT images (G) and quantification of tumor uptake (H) of[111In]ABT-806 in mice with intracranial U87MGde2-7.






EGFR-expressing tumors. Higher tolerated doses of ABT-806 inpatients have been demonstrated in phase I trials supporting thefeasibility of this approach in patients with tumors expressingwild-type EGFR (30)

ABT-806 represents an attractive clinical candidate and it iscurrently under investigation in phase II trials. Furthermore, itssafety profile and functional characteristics, including internali-zation into tumor cells, support its use to deliver a cytotoxicpayload to cancer cells.

Disclosure of Potential Conflicts of InterestN.C. Goodwin is the Vice President of Champions Oncology, Inc. A.M.

Scott reports receiving commercial research support from AbbVie and hasownership interest (including patents) in the Ludwig Institute for CancerResearch. No potential conflicts of interest were disclosed by the otherauthors.

Authors' ContributionsConception and design: E.B. Reilly, A.C. Phillips, F.G. Buchanan, J.A. Meul-broek, T.B. Cole, J. Gu, J.P. Palma, C.K. Donawho, A.M. ScottDevelopment of methodology: E.B. Reilly, A.C. Phillips, F.G. Buchanan,Y. Zhang, J.A. Meulbroek, P.J. DeVries, J. Gu, E.L. Digiammarino, J.P. Palma,C.K. Donawho, A.M. ScottAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): F.G. Buchanan, G. Kingsbury, Y. Zhang, J.A.

Meulbroek, T.B. Cole, P.J. DeVries, H.D. Falls, C. Beam, E.L. Digiammarino,J.P. Palma, C.K. Donawho, N.C. Goodwin, A.M. ScottAnalysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): E.B. Reilly, A.C. Phillips, F.G. Buchanan, G. Kings-bury, Y. Zhang, J.A.Meulbroek, T.B. Cole, P.J. DeVries, J. Gu, E.L.Digiammarino,J.P. Palma, C.K. Donawho, N.C. Goodwin, A.M. ScottWriting, review, and/or revision of the manuscript: E.B. Reilly, A.C. Phillips,F.G. Buchanan, G. Kingsbury, Y. Zhang, J.A. Meulbroek, T.B. Cole, P.J. DeVries,H.D. Falls, C. Beam, J. Gu, E.L. Digiammarino, J.P. Palma, C.K. Donawho,N.C. Goodwin, A.M. ScottAdministrative, technical, or material support (i.e., reporting or organizingdata, constructing databases): E.B. Reilly, P.J. DeVries, C. Beam, J.P. Palma,C.K. DonawhoStudy supervision: E.B. Reilly, A.C. Phillips, F.G. Buchanan, Y. Zhang, J.P.Palma, C.K. Donawho

AcknowledgmentsThe authors thank Lauren Smithee (in vivo studies), Baole Wang, Jerry

Clarin, and Sally Schlessinger (IHC studies) for their excellent technicalassistance.

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.

Received September 24, 2014; revised February 17, 2015; accepted February21, 2015; published OnlineFirst March 2, 2015.

References1. Burgess AW. EGFR family: structure physiology signalling and therapeutic

targets. Growth Factors 2008;26:263–74.2. Mendelsohn J, Baselga J. Epidermal growth factor receptor targeting in

cancer. Semin Oncol 2006;33:369–85.3. Herbst RS, Kim ES, Harari PM. IMC-C225, an anti-epidermal growth factor

receptor monoclonal antibody, for treatment of head and neck cancer.Expert Opin Biol Ther 2001;1:719–32.

4. LynchDH, YangXD. Therapeutic potential of ABX-EGF: a fully human anti-epidermal growth factor receptor monoclonal antibody for cancer treat-ment. Semin Oncol 2002;29:47–50.

5. Li T, Perez-Soler R. Skin toxicities associated with epidermal growth factorreceptor inhibitors. Target Oncol 2009;4:107–19.

6. JungbluthAA, Stockert E,HuangHJ, Collins VP,CoplanK, IversenK, et al. Amonoclonal antibody recognizing human cancers with amplification/overexpression of the human epidermal growth factor receptor. Proc NatlAcad Sci U S A 2003;100:639–44.

7. Gan HK, Burgess AW, Clayton AH, Scott AM. Targeting of a conforma-tionally exposed, tumor-specific epitope of EGFR as a strategy for cancertherapy. Cancer Res 2012;72:2924–30.

8. Garrett TP, Burgess AW, Gan HK, Luwor RB, Cartwright G, Walker F, et al.Antibodies specifically targeting a locally misfolded region of tumorassociated EGFR. Proc Natl Acad Sci U S A 2009;106:5082–7.

9. Johns TG, Stockert E, Ritter G, Jungbluth AA, Huang HJ, Cavenee WK,et al. Novel monoclonal antibody specific for the de2-7 epidermalgrowth factor receptor (EGFR) that also recognizes the EGFR expressedin cells containing amplification of the EGFR gene. Int J Cancer 2002;98:398–408.

10. Batra SK, Castelino-Prabhu S, Wikstrand CJ, Zhu X, Humphrey PA, Fried-man HS, et al. Epidermal growth factor ligand-independent, unregulated,cell-transforming potential of a naturally occurring human mutant EGFR-vIII gene. Cell Growth Differ 1995;6:1251–9.

11. Scott AM, Lee FT, Tebbutt N, Herbertson R, Gill SS, Liu Z, et al. A phase Iclinical trial with monoclonal antibody ch806 targeting transitional stateand mutant epidermal growth factor receptors. Proc Natl Acad Sci U S A2007;104:4071–6.

12. Cleary JM, Yee LK, AzadN,CardcciM,CosgroveD, Limaye S, et al. Aphase 1study of ABT-806, a humanized recombinant anti-EGFR monoclonalantibody, in patientswith advanced solid tumors. AACRMeet AbstrOnline2012;2501.

13. Panousis C, Rayzman VM, Johns TG, Renner C, Liu Z, Cartwright G, et al.Engineering and characterisation of chimeric monoclonal antibody 806(ch806) for targeted immunotherapy of tumours expressing de2-7 EGFRoramplified EGFR. Br J Cancer 2005;92:1069–77.

14. Pastore F, Pastore A,WittmannG,HiddemannW, SpiekermannK. The roleof therapeutic leukapheresis in hyperleukocytotic AML. PLoSONE 2014;9:e95062.

15. Hnatowich DJ, Layne WW, Childs RL, Lanteigne D, Davis MA, Griffin TW,et al. Radioactive labeling of antibody: a simple and efficient method.Science 1983;220:613–5.

16. Luwor RB, Johns TG, Murone C, Huang HJ, Cavenee WK, Ritter G, et al.Monoclonal antibody 806 inhibits the growth of tumor xenograftsexpressing either the de2-7 or amplified epidermal growth factorreceptor (EGFR) but not wild-type EGFR. Cancer Res 2001;61:5355–61.

17. Bigner SH, Humphrey PA, Wong AJ, Vogelstein B, Mark J, Friedman HS,et al. Characterization of the epidermal growth factor receptor in humanglioma cell lines and xenografts. Cancer Res 1990;50:8017–22.

18. Pandita A, Aldape KD, ZadehG, Guha A, James CD. Contrasting in vivo andin vitro fates of glioblastoma cell subpopulations with amplified EGFR.Genes Chromosomes Cancer 2004;39:29–36.

19. Humphrey PA, Wong AJ, Vogelstein B, Friedman HS, Werner MH,Bigner DD, et al. Amplification and expression of the epidermal growthfactor receptor gene in human glioma xenografts. Cancer Res 1988;48:2231–8.

20. NaganeM, Narita Y,Mishima K, Levitzki A, Burgess AW, CaveneeWK, et al.Human glioblastoma xenografts overexpressing a tumor-specific mutantepidermal growth factor receptor sensitized to cisplatin by the AG1478tyrosine kinase inhibitor. J Neurosurg 2001;95:472–9.

21. Luo FR, Yang Z, Dong H, Camuso A, McGlinchey K, Fager K, et al.Correlation of pharmacokinetics with the antitumor activity of Cetuximabin nude mice bearing the GEO human colon carcinoma xenograft. CancerChemother Pharmacol 2005;56:455–64.

22. Merlino GT, Ishii S, Whang-Peng J, Knutsen T, Xu YH, Clark AJ, et al.Structure and localization of genes encoding aberrant and normal epider-mal growth factor receptor RNAs from A431 human carcinoma cells. MolCell Biol 1985;5:1722–34.

23. Cohen MH, Chen H, Shord S, Fuchs C, He K, Zhao H, et al. Approvalsummary: Cetuximab in combination with cisplatin or carboplatin and


Reilly et al.




5-fluorouracil for the first-line treatment of patients with recurrent locor-egional or metastatic squamous cell head and neck cancer. Oncologist2013;18:460–6.

24. Stupp R, Mason WP, van den Bent MJ, Weller M, Fisher B, Taphoorn MJ,et al. Radiotherapy plus concomitant and adjuvant temozolomide forglioblastoma. N Engl J Med 2005;352:987–96.

25. Preusser M, Wohrer A, Stary S, Hoftberger R, Streubel B, Hainfellner JA.Value and limitations of immunohistochemistry and gene sequencingfor detection of the IDH1-R132H mutation in diffuse glioma biopsyspecimens. J Neuropathol Exp Neurol 2011;70:715–23.

26. Garber K. From human to mouse and back: `tumorgraft' models surge inpopularity. J Natl Cancer Inst 2009;101:6–8.

27. Goldstein NI, Prewett M, Zuklys K, Rockwell P, Mendelsohn J. Biologicalefficacy of a chimeric antibody to the epidermal growth factor receptor in ahuman tumor xenograft model. Clin Cancer Res 1995;1:1311–8.

28. Frederick L, Wang XY, Eley G, James CD. Diversity and frequency ofepidermal growth factor receptor mutations in human glioblastomas.Cancer Res 2000;60:1383–7.

29. Fuller GN, Bigner SH. Amplified cellular oncogenes in neoplasms of thehuman central nervous system. Mutat Res 1992;276:299–306.

30. Gan HK, Burge ME, Solomon BJ, Holen KH, Zhang Y, Ciprotti M, et al. Aphase I and biodistribution study of ABT-806i, an 111indium-labeledconjugate of the tumor-specific anti-EGFR antibody ABT-806. J Clin Oncol2013;31.






2015;14:1141-1151. Published OnlineFirst March 2, 2015.Mol Cancer Ther Edward B. Reilly, Andrew C. Phillips, Fritz G. Buchanan, et al. Anti-EGFR Monoclonal AntibodyCharacterization of ABT-806, a Humanized Tumor-Specific

Updated version

10.1158/1535-7163.MCT-14-0820doi:

Access the most recent version of this article at:

Cited articles

http://mct.aacrjournals.org/content/14/5/1141.full#ref-list-1

This article cites 28 articles, 12 of which you can access for free at:

Citing articles

http://mct.aacrjournals.org/content/14/5/1141.full#related-urls

This article has been cited by 8 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://mct.aacrjournals.org/content/14/5/1141To request permission to re-use all or part of this article, use this link



http://mct.aacrjournals.org/lookup/doi/10.1158/1535-7163.MCT-14-0820

http://mct.aacrjournals.org/content/14/5/1141.full#ref-list-1

http://mct.aacrjournals.org/content/14/5/1141.full#related-urls

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

mailto:[email protected]

http://mct.aacrjournals.org/content/14/5/1141


Documents

Characterization of ABT-806, a Humanized Tumor-Speci c ... · development, we humanized mAb 806 to generate ABT-806, a recombinant IgG1/k mAb. In a phase I study in patients with