development of a tetravalent anti-gpa33/ anti-cd3 ... · bsab, and especially one targeting gpa33...

Large Molecule Therapeutics

Development of a Tetravalent Anti-GPA33/Anti-CD3 Bispecific Antibody for ColorectalCancersZhihao Wu, Hong-Fen Guo, Hong Xu, and Nai-Kong V. Cheung

Abstract

Despite progress in the treatment of colorectal cancer, curingmetastatic colorectal cancer remains a major unmet medicalneedworldwide.Here,wedescribe a T-cell–engaging bispecificantibody (T-BsAb) to redirect polyclonal cytotoxic T cells toeradicate colorectal cancer. A33, amurine antibody specific forGPA33, was humanized to huA33 and reformatted to huA33-BsAb, based on a novel IgG(L)–scFv platform by linking theanti-CD3 huOKT3 scFv to the carboxyl end of the light chain.This T-BsAbwas stably expressed inCHOcells andpurified as astable monomer by HPLC, retaining immunoreactivity byFACS through 30 days of incubation at 37�C. In vitro, itinduced activation and expansion of unstimulated T cells

and elicited potent T-cell–dependent cell-mediated cytotoxic-ity against colon and gastric cancer cells in an antigen-specificmanner. In vivo, huA33-BsAb inhibited the colon and gastriccancer xenografts, in both subcutaneous and intraperitonealtumormodels. More importantly, bothmicrosatellite instableand microsatellite stable colorectal cancer were effectivelyeliminated by huA33-BsAb. These preclinical results providefurther support for the use of IgG(L)–scFv platform to buildBsAb, and especially one targeting GPA33 for colorectalcancer. These preclinical results also support further develop-ment of huA33-BsAb as a potential immunotherapeutic.Mol Cancer Ther; 17(10); 2164–75. �2018 AACR.

IntroductionColorectal cancer is the third leading cause of death among

cancers in the United States (1) and accounts for 10% of allcancers in men and 9.2% in women worldwide (2). Althoughthere has been steady yearly 3% decline of incidences from2004 to 2013, around 135,000 new cases are expected annuallyin the United States alone. Although localized and regionaldiseases are often curable, the prognosis of metastatic colo-rectal cancers (mCRC) is poor, with a 5-year survival rate ofonly 14% (1).

The standard treatment of mCRC consists of chemotherapyin combination with monoclonal antibodies that block tumorsignaling or angiogenesis. Currently, four monoclonal anti-body drugs have been approved by the FDA for colorectalcancer treatment. Bevacizumab and ramucirumab target theVEGF–VEGFR angiogenesis pathway, whereas cetuximab andpanitumumab target the EGFR pathway. However, cetuximaband panitumumab do not provide clinical benefits in mCRCpatients with RAS mutations (3), which account for around40% of mCRC (4, 5). Moreover, improvement in survival withthese antibodies is generally modest and can be associated withsevere side effects (5–7).

Given the clinical success in melanoma, renal cell carcinoma,and lung cancer, immune checkpoint inhibitors (ICI) haverevolutionized modern immunotherapy and have become thebackbones of numerous ongoing trials in a broad spectrum ofadult cancers (8, 9). Unfortunately, for colorectal cancer, ben-efits so far have been restricted to the subset of patients withmicrosatellite instability (MSI), while the majority of mCRC aremicrosatellite stable (MSS) and are not expected to benefit fromICI monotherapy (10–12).

More effective strategies to take advantage of T cells totreat mCRC are needed. T-cell–engaging bispecific antibodies(T-BsAb), which function by actively redirecting polyclonalcytotoxic T cells to kill cancer cells, are promising candi-dates. Their efficacy in leukemia has been proven for blina-tumomab, an anti-CD19 T-BsAb approved by the FDA in2014. For solid tumors, numerous preclinical studies havedocumented their potent antitumor activities and several arein clinical trials.

We hypothesized that T-BsAbs could expand the potential ofICIs and overcome genetic barriers in the treatment of mCRC.We now report a T-BsAb that targets the protein GPA33, anantigen expressed in >95% of colon cancers and a subset ofgastric cancers (13). This target has been extensively exploredin radioimmunodiagnosis and radioimmunotherapy, notablyusing a murine-derived antibody A33 against GPA33 (14–17)and its humanized version, which have been relatively welltolerated (15, 18, 19), despite the presence of human anti-mouse antibodies (HAMA; refs. 18, 19), and human antihu-man antibodies (HAHA; refs. 15, 20). Using a rehumanizedversion of the mouse A33, we successfully built a tetravalentT-BsAb and characterized its in vitro physicochemical propertiesand functional activities, as well as its in vivo antitumor pro-perties in multiple xenograft models. We showed that its

Memorial Sloan Kettering Cancer Center, New York, New York.

Note: Supplementary data for this article are available at Molecular CancerTherapeutics Online (http://mct.aacrjournals.org/).

Corresponding Author: Nai-Kong V. Cheung, Memorial Sloan Kettering CancerCenter, 1275 York Avenue, Box 170, Z-1403, New York, NY 10065. Phone: 646-888-2313; Fax: 646-422-0452; E-mail: [email protected]

doi: 10.1158/1535-7163.MCT-18-0026

�2018 American Association for Cancer Research.

MolecularCancerTherapeutics

Mol Cancer Ther; 17(10) October 20182164

http://crossmark.crossref.org/dialog/?doi=10.1158/1535-7163.MCT-18-0026&domain=pdf&date_stamp=2018-9-13

efficacy was independent of the MSI status of colorectal cancertumors, a necessary attribute if this T-BsAb is meant to functionagainst all mCRCs.

Materials and MethodsCell lines and human white cells

Cell lines used in this study and their sources were summa-rized in Supplementary Table S1. Luciferase-expressing celllines were established using lentiviral transduction. All cellswere authenticated by STR typing within 6 months of experi-ments. Cells were maintained in RPMI medium supplement-ed with 10% FBS (Sigma), 0.03% L-glutamine (Gibco), andpenicillin–streptomycin (Gibco). Buffy coats from healthydonors were purchased from New York Blood Center, andhuman peripheral blood mononuclear cells (PBMC) wereisolated by Ficoll gradient of Buffy coats.

SEC-HPLC analysisSize purity of T-BsAb was analyzed using HPLC system

(Shimadzu Scientific Instruments Inc.). Monomeric species wasidentified using molecular weight standard (Bio-Rad) and per-centage of monomers was calculated based on the relative areaunder curve (AUC) of different non-buffer peaks.

Humanization of murine A33 and construction ofhuA33-BsAb bispecific antibody

Using CDR grafting, mouse A33 was humanized as IgG1. Twodifferent heavy chain variable (VH) and light chain variable(VL) sequences were combined to generate 4 different human-ized A33 antibodies. Binding kinetics was compared with thatof chimeric antibody chA33 using surface plasmon resonance(SPR) analysis. The heavy chain of selected sequence is

EVQLVESGGGLVKPGGSLRLSCAASGFAFSTYDMSWVRQAP-GKRLEWVATISSGGSYTYYLDSVKGRFTISRDNAKNSLYLQMNS-LRAEDTAVYYCAPTTVVPFAYWGQGTLVTVSS

and the light chain sequence is

DIQMTQSQSSLSTSVGDRVTITCKASQNVRTVVAWYQQKP-GKSPKTLIYLASNRHTGVPSRFSGSGSGTEFTLTISNVQPEDFAD-YFCLQHWSYPLTFGSGTKLEIKR.

HuA33-BsAb was constructed by fusing the humanizedOKT3 scFv onto the C-terminus of the light chain of huA33antibody via a (G4S)3 linker as previously described (21, 22).The DNA construct was then transfected into CHO-S cells andstable clones were selected for high levels of antibody produc-tion. For larger-scale antibody purification, selected stableclone was expanded in shaker flasks. Bispecific antibody waspurified from supernatant using one-step protein A affinitychromatography.

SPR analysisHuman GPA33 (Novoprotein) was immobilized on CM5

chips. Five concentrations of 2-fold serially diluted huA33 IgG1or T-BsAbs (starting at 20 nmol/L)wereflowedover the chip usingaBiacore T100 system. Binding kinetics of huA33wasmeasured at25�C and those of T-BsAbs at both 25�C and 37�C. The sensor-grams were fitted with a 1:1 binding model for both to derivekinetic parameters.

IHCIHC staining of human tissue was done using ABC immu-

noperoxidase (Vector Laboratories). Five-micrometer thick sec-tions were fixed in cold acetone for 30 minutes at �20�C,followed by rinsing in phosphate-buffered saline (PBS), andimmersed in 0.1% hydrogen peroxide for 15 minutes at roomtemperature (RT). All the following steps were performed atRT. Avidin and biotin solutions were added sequentially for20 minutes each. After blocking with 10% horse serum in PBSfor 1 hour, each subsequent step was followed by washing ofslides with PBS. Primary huA33-BsAb or control T-BsAb wasadded to the tissue sections at 1.0 mg/mL for 45 minutes,followed by anti-OKT3-mFc at 0.1 mg/mL for 30 minutes.Lastly, biotinylated horse anti-mouse IgG (H þ L) antibodyat 1:500 dilution was added for 30 minutes, followed by100 mL of ABC for 30 minutes. Peroxidase substrate was thenadded to each tissue section for 2 minutes, rinsed with runningtap water for 5 minutes, and counterstained with Myer's hema-toxylin. Stained slides were dehydrated by dipping sequenti-ally in 75%, 95%, and 100% ethanol, and finally in xylene.Dehydrated slides were mounted with one drop of Cytoseal(Richard-Allan Scientific) for examination.

Human CD3 IHC was performed at the Molecular CytologyCore Facility of Memorial Sloan Kettering Cancer Center(MSKCC) using the Discovery XT processor (Ventana MedicalSystems). Anti-CD3 antibody (DAKO; cat. #A0452, 1.2 mg/mL)was applied to deparaffinized sections and incubated for5 hours, followed by 60 minutes of incubation with biotiny-lated goat anti-rabbit IgG (Vector Labs, cat. #PK6101) anddetected with DAB detection kit (Ventana Medical Systems)according to the manufacturer's instruction. Slides were coun-terstained with hematoxylin and coverslipped with Permount(Fisher Scientific). All images were captured using NikonECLIPSE Ni-U microscope and NIS-Elements 4.0 imaging soft-ware (Nikon Instruments, Inc.).

T-cell–dependent cytotoxicity (TDCC) assaysCytotoxicity assays were performed using both 51Cr-release

assay and LDH-release assay (Pierce). For both assays, we usedanti-CD3/anti-CD28 Dynabeads activated T cells (ATC) for14 days as effector cells, except for sorted cells from PBMCs,which were used for TDCC assay without prior stimulation.51Cr assay was performed as previously described (23). LDHassay was done following the manufacturer's instructions. EC50

values were calculated by fitting the curves to a 4-parameternonlinear regression model using GraphPad Prism.

Cytokines and cytolytic molecules release assayT cells were purified from PBMCs using pan–T-cell isolation

kit (Miltenyi Biotec). Target and control tumor cells wereincubated with 2 � 106 cells/well at an effector-to-target ratioof 5:1 in 24-well plates in triplicates, with 2 mL total volumeper well. Culture supernatants were collected at 24, 48, 72, and96 hours, and cytokine levels were measured with flow cyto-metry using LEGENDplex human CD8/NK Panel (BioLegend),following the manufacturer's protocol.

T-cell proliferation assayFresh PBMCs from healthy donors were labeled with

2.5 mmol/L CFSE (Life Technologies) for 5 minutes at RT,followed by neutralization using PBS containing 5% FBS.

Development of an Anti-GPA33 � Anti-CD3 T-BsAb

www.aacrjournals.org Mol Cancer Ther; 17(10) October 2018 2165

Target cells were then incubated with labeled PBMCs underdifferent conditions at 37�C before analyzing expressionof surface activation markers and T-cell proliferation at 24 and96 hours.

In vivo tumor therapyAll experiments have been conducted in accordance with and

approved by the Institutional Animal Care andUse Committee inMSKCC. For subcutaneous (s.c.) tumor models, 5 � 106 LS174T,3 � 106 Colo205, or 5 � 106 SNU16 cells were combinedwith fresh PBMCs at a 1:1 ratio, mixed with Matrigel (volumeof cell:gel ¼ 1:2), and implanted with 100 mL/mouse at the flankof Balb/c Rag2�/�IL2Rg�/� (DKO) mice (derived from colony ofDr. Mamoru Ito, CIEA, Kawasaki, Japan, and now commerciallyavailable fromTaconic asCIEABRGmice). Treatment started afterconfirmation of tumor presence with a twice a week (BIW)schedule at 100 mg/mouse and continued for 3 to 4 weeks. Tumorgrowth was monitored by weekly measurement of tumor volumeusing a caliper or a digital device Peira TM900 Scanner (PeiraScientific Instruments).When caliperwas used, the tumor volumewas calculated as length � width � width/2.

For intraperitoneal (i.p.) tumor models, 3 � 106 or 1 � 106

luciferase-expressing LS174T-luc or SW1222-luc cells, respec-tively, were resuspended in RPMI medium and injected i.p.into DKO mice. Human effector cells were injected i.v. as20 � 106 ATC per dose for LS174T-luc tumors or 5 � 106 ATCper dose for SW1222-luc tumors, whereas antibody wasinjected at 50 mg/mouse/dose for LS174T-luc tumors or100 mg/mouse/dose for SW1222-luc tumors. Mice were treatedwith 2 to 3 weekly cycles, with each component separated by3 to 4 days. Growth of tumors was followed weekly by mea-suring luminescence signals on an IVIS Spectrum in vivo imag-ing system (PerkinElmer) after injection of 3 mg/mouse ofluciferin. Luminescence signals were analyzed and quantifiedusing Living Image Software (PerkinElmer).

Antibodies and flow cytometryAnti-hCD25-PE, anti-CD69-PE, anti-hCD8-APC, anti-hCD45-

PECy7, anti-hCD4-PE, anti-hCD4-BV421, anti-hCD62L-PercpCy5.5,anti-hPD1-BV421, and anti-hCD45RO-PECy7 were fromBioLegend. Goat antihuman IgG-PE was from SouthernBiotech.Streptavidin-PE, anti-hCD4-APC, and anti-hCD25-APC werefrom BD Biosciences. All FACS analyses were carried out usingFACSCalibur or LSRII system (BD Biosciences) and analyzedusing FlowJo (FlowJo, Inc.).

Affinity maturation using yeast displayParental huA33 was converted into scFv format with a

20–amino acid (G4S)4 linker and cloned into a yeast displayvector. HuA33 scFv was randomly mutated using GeneMorph IImutagenesis kit (Agilent Technologies). PCR product was elec-troporated together with linearized vector into yeast and thelibrary was subjected to 4 rounds of sorting using biotinylatedGPA33. Individual clones from the last round were PCRamplified and sequenced to analyze the mutation pattern.Conversion of selected scFv clones into T-BsAb format wasdone using a one-step 4-fragment ligation method with 50 nglinearized vector and a 1:3 vector to insert molar ratio for theother 3 components. Ligation was done with a Rapid DNAligation kit (Thermo Fisher Scientific) at RT for 1 hour. Type IIrestriction enzyme SapI (NEB) was used to ensure seamless

linkage among the different components (SupplementaryFig. S3). Selected clones were transiently expressed using theExpi293 expression system (Thermo Fisher Scientific) followingthe manufacturer's instructions. Supernatant from Expi293cells after 4 to 5 days of culture in shaking flasks was used topurify antibodies using MabSelect SuRe (GE Healthcare) anddialyzed against pH8.0 citrate buffer in dialysis membrane(SpectrumLabs.com).

Statistical analysisDifferences in tumor sizeswere testedusing two-tailed Student t

test. Kaplan–Meier survival analysis and log-rank test were used tocompare survival curves. When P < 0.05, the differences wereconsidered statistically significant.

ResultsHuA33 retained affinity for GPA33

When compared with chimeric A33, all four newly humanizedA33 antibodies have slightly improved koff in binding to immo-bilized GPA33 in SPR analysis (Fig. 1A). Based on theKD, stabilityat 37�C and T20 humanness score (24), one huA33 clone waschosen for further development.

HuA33-BsAb was highly stable and bound to antigens withhigh affinity and specificity

The huA33 antibody was reformatted into the 2þ2 bispeci-fic format (25) by fusing scFv of humanized OKT3 to theC-terminus of light chain via a flexible GS linker (Fig. 1B). TheDNA construct was used to establish a CHO-S stable cellline producing 50 mg/L to 100 mg/L of protein without exten-sive optimization. Slightly lower yields were observed usingExpi293 transient expression system (around 33 mg/L).One-step protein A purification routinely produced proteinwith purity above 90%, as measured by SEC-HPLC. Afterincubating the protein at 37�C for 4 weeks, there was onlyminimal decrease in the percentage of monomers, as shown inFig. 1C. These data suggest that huA33-BsAb had good solu-bility, purity, and thermal stability, all of which are criticalcharacteristics for further downstream development.

We measured the avidities of huA33-BsAb toward GPA33 atboth 25�C and 37�C using GPA33-immobilized CM5 chips. Asshown in Fig. 1D, huA33-BsAb bound GPA33 with a highapparent affinity of around 0.2 nmol/L, which is slightly lowerthan 0.13 nmol/L obtained for parental huA33. FACS analysisof a panel of cell lines derived from different cancers showedthat huA33-BsAb stained colon cancer cell lines and one gastriccancer cell line but not GPA33� neuroblastoma cell line IMR32,osteosarcoma cell line TC32 or melanoma cell line SKMEL5(Fig. 1E; Supplementary Table S1), suggesting that huA33-BsAbretained the specificity of parental antibody A33 in binding totarget antigens on colon cancer cells and a subset of gastriccancer cells. Specific expression of GPA33 on colon tissues wasalso confirmed by IHC (Supplementary Fig. S1). Staining ofATC also showed that huA33-BsAb bound to CD3 on T-cellsurface (Fig. 1E).

HuA33-BsAb activated and induced cell-cycle entry of freshT cells

To test the ability of huA33-BsAb to activate unstimulatedT cells, CFSE-labeled PBMCs were mixed with Colo205 cells

Wu et al.

Mol Cancer Ther; 17(10) October 2018 Molecular Cancer Therapeutics2166

at an effector-to-target ratio of 5:1 (E:T ¼ 5:1) and cultured inthe presence of huA33-BsAb (1 mg/mL). As negative controls,we used huA33-C825 that carried an irrelevant scFv (26)instead of the anti-CD3 scFv, as well as a control T-BsAbantibody that did not bind to Colo205 by FACS. After 24 and96 hours, cells were stained with different T-cell activationmarkers to assess T-cell activation status and proliferation. Asearly as 24 hours, huA33-BsAb caused activation of both CD4þ

and CD8þ T cells, as shown by the upregulation of CD25 andCD69 markers on cell surface (Fig. 2A). In contrast, huA33-C825 and control T-BsAb caused only minimal upregulation ofCD25. Control T-BsAb did increase the expression of CD69,especially in CD4þ T cells. Similarly, PD-1 upregulation wasobserved after 24 hours and persisted until 96 hours (Supple-mentary Fig. S2A). Cell division, as measured by CFSE dyedilution, was observed in both CD4þ and CD8þ T cells after 96hours (Fig. 2B). CD8þ T cells showed higher cell division cycles,suggesting that CD8þ T cells divided faster than CD4þ T cells(Fig. 2C). Control antibody huA33-C825 did not stimulatesignificant amount of cell division in either T-cell subset,confirming the requirement of CD3 activation for T-cell divi-

sion. A low level of cell division was induced by control T-BsAb,consistent with the low level of activation observed above.When GPA33� SKMEL5 targets were used, huA33-BsAb didnot activate T cells (Supplementary Fig. S2B). These datasuggested that activation of T cells by huA33-BsAb dependedon the presence of cognate antigens on tumor cells. We alsoobserved that cell division was associated with expansion ofCD45ROþ effector/memory cells (Fig. 2D), suggesting theimportance of this subset in mediating T-BsAb activity, whichwas confirmed below. These assays were repeated using adifferent colon cancer cell line LS174T with similar conclusions(Supplementary Fig. S2C).

To see if T cells can be activated in vivo, we conducted an in vivoproliferation assay bymixing CFSE-labeled PBMCs with Colo205cells and implanted the mixture subcutaneously onto DKOmice.HuA33-BsAb was injected the next day, and the tumors wereisolated after another 4 days and analyzed by FACS. As shownin Fig. 2E, around 25% of CD4þ and CD8þ T cells upregulatedCD25 expression while undergoing cell division (progressivehalving of CFSE fluorescence), suggesting that huA33-BsAb wasable to stimulate T-cell activation and proliferation in vivo as well.

Figure 1.

Stability and binding characteristics ofthe huA33-BsAb antibody. A, SPRanalysis of 4 versions of humanizedA33. All antibodies were in IgG1 format.3A3-H1L1, 3A3-H1L2, 3A3-H2L1, and3A3-H2L2 were 4 versions ofhumanized 3A3 (monospecific). 3A3-chA33was chimeric 3A3. B,Design andconstruction of huA33-BsAb. C,Accelerated stability test of purifiedhuA33-BsAb at 37�C over 4 weeks;monomer % represents the percentageof monomers in SEC profile for eachtime point, based on AUC analysisexcluding buffer peak. D, SPR analysisof huA33-BsAb at 25�C and 37�Caccording to conditions in Materialsand Methods. Data were fit to a 1:1binding model. E, FACS staining ofdifferent tumor cell lines and ATC. MFIvalues were geometric means.



HuA33-BsAb induced secretion of inflammatory cytokines andcytolytic molecules

We next examined the secreted cytokine profile of T cellsactivated by huA33-BsAb in the presence of target tumors.Total T cells were purified from PBMCs and cultured in thepresence of Colo205 tumor cells at E:T ¼ 5:1; to estimatenonspecific activation, we used SKMEL5 as a negative control.Cell culture supernatants were collected daily over 4 days andthe levels of cytokines and cytotoxic molecules were measuredusing a flow cytometry–based multiplex method. As shownin Fig. 3, both Th1 cytokines (IL2, IFNg , and TNFa) and Th2cytokines (IL4 and IL10) were secreted by ATCs, although IL4was secreted at a much lower level than other cytokines.Similarly, significant amount of IL6 was secreted by ATCs.Interestingly, Th17 cytokine IL17a was also secreted at highlevels. As expected, cytotoxic components sFasL, Granzyme A,Granzyme B, Perforin, and Granulysin were all released in thesupernatant.

HuA33-BsAb redirected T cells to specifically kill coloncancer and gastric cancer cells

We next tested if huA33-BsAb could redirect T cells to killcancer cells. Based on the survey of GPA33 expression onmultiple human cancer cell lines (Supplementary Table S1),we selected three colon cancer cell lines (LS174T, SW1222, andColo205) and one gastric cancer cell line (SNU16) for cyto-toxicity studies. These cell lines were classified as MSI (LS174T)subtype or MSS (SW1222, Colo205, SNU16) subtype based ontheir MSI profile (27–29). Moreover, LS174T cells carried aKRAS G12D mutation while the others carried TP53 mutationsor deletions (30–33). Cancer cells were incubated with ATCat E:T ¼ 10:1 in the presence of 10-fold serial dilutions ofhuA33-BsAb. GPA33� melanoma cell line SKMEL5 and oste-osarcoma cell line TC32 were used as negative controls. Asshown in Fig. 4A, huA33-BsAb redirected T cells to specificallykill all GPA33-expressing cancer cells regardless of their geneticbackgrounds, while sparing SKMEL5 and TC32, confirming the

Figure 2.

In vitro and in vivo activation of humanPBMCsbyhuA33-BsAb.A,Activation ofCD25 andCD69 on T cells 24 hours afterincubation with Colo205 cells anddifferent antibodies. B, Quantificationof dividing cells based on CFSE dyedilution 96 hours after incubation withColo205 cells and different antibodies.C, Representative images for (B). D,Staining of CD45RO 96 hours afterincubation with Colo205 cells andhuA33-BsAb in an independentexperiment. E, In vivo activation andproliferation of T cells by huA33-BsAb.

Wu et al.


antigen specificity of huA33-BsAb–mediated TDCC. Maximallevel of cytotoxicity seemed to correlate with the level of FACSstaining (Figs. 1E and 4A). Moreover, TDCC induced by huA33-

BsAb was potent, with EC50 values in the picomolar range. Wealso confirmed the specificity of TDCC mediated by huA33-BsAb by replacing either the huA33 arm with control BsAb or

Figure 3.

Profile of secreted cytokines and cytotoxic components by huA33-BsAb ATC in the presence of target tumor. Kinetics of production of cytokines andcytolytic molecules by T cells in the presence of huA33-BsAb and target cell Colo205 or negative control cell SKMEL5 over 4 days. SKMEL5 itselfsecreted copious amount of IL6; therefore, only supernatant from T cells incubated with Colo205 cells in the absence of antibody was used as control.IFNg, granzyme A, granzyme B, and perforin were from a second independent experiment.



OKT3 arm with irrelevant C825; replacement of either armcompletely abolished TDCC activity (Supplementary Fig. S3).

To see which subsets of T cells were mobilized by huA33-BsAb,we sorted CD45RAþCD62Lþ and CD45RA�CD45ROþCD62L�

memory subsets from both CD4þ and CD8þ T cells. Sortedcells were cultured in the presence of Colo205 tumors cells atE:T ¼ 5:1 for 48 hours before measuring cytotoxicity. As shownin Fig. 4B, both CD4þ and CD8þ memory T-cell subsets werecapable of inducing cytotoxicity, with CD8þ memory T cellsmediating more efficient killing at higher concentration. Inter-estingly, CD45RAþCD62Lþ subsets of both CD4þ and CD8þ

populations, with themajority being na€�ve T cells, were capable ofinducing cytotoxicity after 48 hours, although the potency wasless compared with memory T cells. When T cells from the TDCCassay were stained with CD45RO and CD25, it was found thatCD25 expression was upregulated in the presence of huA33-BsAb, in both CD45ROþ and CD45RO� fractions (Supplemen-tary Fig. S4), confirming that both na€�ve and memory T cellscould be activated by T-BsAb. The majority of CD45RAþCD62Lþ

T cells stayed CD45RO� after incubation, with a small butsignificant population increasing their CD45RO expression, espe-cially among the CD8þ cells. However, this population couldhave been derived from either the maturation of na€�ve T cells or

from expansion of rare CD45ROþ cells present in the initialculture. Overall, these data demonstrated that huA33-BsAb couldinduce potent cytotoxicity against colon and gastric cancer cells ina GPA33 dependentmanner, bymobilizing both CD4 and CD8 Tcells, especially those of the memory phenotype.

Affinity maturation of huA33-BsAb by yeast displayIn an attempt to further improve the potency of huA33-BsAb,

we used yeast displaymethod to affinitymature scFv derived fromhuA33. Because scFv tends to aggregate easily, we developed amethod to rapidly reformat scFv to T-BsAb (SupplementaryFig. S5A), which is a more relevant format and could be readilyproduced in high yield and purity.

From sequence analysis of 60 single clones, we selected 7clones (Supplementary Fig. S5B) for further characterization bySPR and TDCC assay. All 7 clones showed increased bindingaffinity (4.3- to 51-fold) as compared with parental antibody(Supplementary Fig. S5C). The majority of improvement wascontributed by slower off-rate. When tested in the TDCC assay,all clones showed slight improvement in maximal killing(Supplementary Fig. S5D); however, no significant enhance-ment in EC50 was observed (Supplementary Fig. S5D).

In vivo therapy studies using huA33-BsAbEfficacy of huA33-BsAb was tested in vivo with two xenograft

models in DKO mice adoptively transferred with human Tcells: (i) s.c. tumor plus s.c. effectors and (ii) i.p. tumor plusi.v. effectors.

HuA33-BsAb cured MSI tumor LS174T in an s.c. xenograft modeland suppressed tumor growth in an i.p. model. LS174T cells weremixedwith PBMCs at a 1:1 ratio and implanted subcutaneously inDKO mice. As shown in Fig. 5A, without i.v. antibody treatment,both tumor-only and tumorþ PBMCgroups showed rapid tumorgrowth. Mice from tumor þ PBMC group developed tumorulceration and had to be euthanized. In contrast, 6 doses of i.v.huA33-BsAb over 3 weeks effectively cured the mice, whichremained tumor free for at least 120 days.

To simulate malignant ascites, a common occurrence in coloncancers, luciferase-expressing LS174T cells were planted i.p. intoDKO mice. When tumor growth was confirmed by lumines-cence, mice were randomized into different treatment groups:ATC only, ATCþ huA33-C825, ATC þ control BsAb, and ATCþhuA33-BsAb. Antibody treatment started on day 8 after tumorimplantation and consisted of 6 doses i.p. over 3 weeks,whereas T-cell treatment consisted of weekly injection of ATCthrough retro-orbital route over 3 weeks. As shown in Fig. 5Band C, all three control groups showed exponential growth oftumor in the abdomen. In contrast, i.p. huA33-BsAb signifi-cantly suppressed metastatic growth of LS174T tumor in theabdominal areas of treated mice, accompanied by substantialamount of T-cell tumor infiltration (Fig. 5D). All mice in thisgroup remained alive until at least 55 days without furthertreatment, while all mice in control groups succumbed totumors by day 43 (Fig. 5B, right). This is consistent with ourprevious results showing that huA33-C825 had no direct anti-tumor effect on LS174T tumor (26, 34).

These data suggested that huA33-BsAb was effective againstcolorectal cancer tumors with an MSI genotype. However, asmentioned before, MSI tumors account for only a minority ofcolorectal cancer patients. A majority of colorectal cancer patients

Figure 4.

In vitro cytotoxicity induced by huA33-BsAb. A, Cytotoxicity elicited byhuA33-BsAb against different target tumor cell lines and control cells.ATCs were used as effector cells at an E:T of 10:1. Cells were incubated for16 hours before reading. B, Cytotoxicity induced by huA33-BsAb againstColo205 cells. Sorted fresh T cell subsets were used as effector cells(E:T ¼ 5:1). Cells were incubated for 48 hours before reading. EC50 forCD4 memory T cells was 25 pmol/L.

Wu et al.


Figure 5.

huA33-BsAb eliminated MSI colorectal cancer tumors in the xenograft model. A, Tumor growth of s.c. LS174T tumors in the treatment group and controlgroups. PBMCs were planted s.c. Antibodies were injected i.v. Tumor sizes were assessed by volume (P ¼ 0.0133 for tumor þ scPBMC versus tumor þscPBMC þ huA33-BsAb; P ¼ 0.006 for tumor-only versus tumor þ scPBMC þ huA33-BsAb). B, Left, Growth of i.p. LS174T tumor in treatment groupand control groups as assessed by bioluminescence. � , P < 0.01 determined by Student t test when treatment group (ATC þ huA33-BsAb) wascompared with the corresponding control groups, respectively. Right, survival of mice in treatment group and control groups. C, Representativebioluminescent images of abdominal LS174T tumors from different groups in B were shown. D, IHC images of CD3 staining. Representative images (200�magnifications) of IHC staining of abdominal LS174T tumor sections collected 5 days after i.v. ATC were shown.



are MSS. Therefore, efficacy of huA33-BsAb was tested furtherusing MSS tumors.

HuA33-BsAb cured MSS tumor COLO205 in an s.c. xenograftmodel and suppressed tumor growth of metastatic MSS tumorSW1222. Two MSS colon cancer cell lines, Colo205 and

SW1222, were tested. For Colo205 cells, tumors were implant-ed subcutaneously with PBMCs as effector cells. In this model,4 doses of i.v. huA33-BsAb were enough to completely era-dicate Colo205 tumors and the mice remained tumor free forat least 4 months after a total of 6 doses of antibody treatment(Fig. 6A and B).

Figure 6.

huA33-BsAb cured MSS colorectal cancer tumors in s.c. xenograft model (A and B) and suppressed tumor growth in an i.p. metastasis model (C and D).A, Luminescence images showing growth of s.c. Colo205 tumors in different groups. ATCs were injected i.v.; antibodies were injected i.p. B, Left,Quantification of signals; right, survival of mice among different treatment groups. Mice treated with 4 doses i.v. huA33-BsAb showed complete tumoreradication. C, Luminescence images of i.p. SW1222 tumor, where ATCs were injected i.v. weekly over 2 weeks, and 6 doses of huA33-BsAb i.p. Onday 21, tumor growth was significantly different between huA33-BsAb þ ATC group and no-treatment group (P ¼ 0.0051) or ATC-only group (P ¼ 0.0233).By day 30, tumor growth in the huA33-BsAb þ ATC group was still significantly suppressed when compared with the no-treatment group (P ¼ 0.0037).Mouse #1 from the tumor-only group and mouse #3 from the tumor þ ATC group, which did not take tumor after 21 days, were excluded fromimaging in later time points and not included in tumor growth comparison nor survival analysis in D. Survival of mice with i.p. SW1222 tumor treatedwith ATC with or without huA33-BsAb.

Wu et al.


For the SW1222 cell line, tumors were planted i.p. and treatedwith 6 doses of i.p. antibody over 3 weeks, and i.v. T cell weeklyover 2 weeks. As shown in Fig. 6C and D, the two control groupshad tumor growth spread throughout the whole abdominal area,whereas treatment with huA33-BsAb suppressed tumor growthand significantly prolonged mice survival.

These results together suggested that huA33-BsAb had similarefficacy in MSS tumors as that in MSI tumors. We expected moretreatment cycles could eradicate the growth of tumors moreefficiently, as shown in subcutaneous models.

HuA33-BsAb inhibited growth of gastric cancers in an s.c. xenograftmodel. GPA33 is expressed in a subset of gastric cancer cells (13).By FACS analysis and in vitro TDCC assays, SNU16 expressedGPA33 and was sensitive to huA33-BsAb redirected T cell killing.SNU16 cells were xenografted subcutaneously after mixing withhuman PBMCs. As expected, i.v. huA33-BsAb effectively cured themice of s.c. tumors (Supplementary Fig. S6) and significantlyprolonged survival.

DiscussionCuring mCRC is an urgent unmet medical need. In this study,

we demonstrated a potential strategy using a T-BsAb, huA33-BsAb, in vitro and in vivo, to redirect polyclonal cytotoxic T cells tokill tumors that expressed GPA33 antigen. Besides systemicspread, peritoneal carcinomatosis has also been difficult to con-trol. Unlike metastasis to liver and lung, it is usually unresectableand unresponsive to chemotherapy and radiation, resulting insignificant morbidity. Current treatment is mostly palliative,consisting of cytoreductive surgery or hyperthermic chemother-apy (HIPEC), which are effective in only a small percentage ofpatients with small volume disease. Development of effectiveimmunotherapy using i.p. T-BsAbs for malignant ascites, asdemonstrated in this study, is an attractive alternative for severalreasons. First, the peritoneal cavity is an immunocompetentcompartment harboring immune cells, the majority (>40%)being T cells and monocytes/macrophage (>40%; ref. 35). Sec-ond, i.p. administered antibodies are placed in immediate contactwith cancer cells, facilitating tumor binding through high localdrug concentrations. Third, rapid uptake of antibodies by tumorsreduces systemic drug exposure, hence limiting the toxicity ofT-cell activating biologics commonly seen with i.v. injections. Itwas not surprising that the first T-BsAb drug approved for solidtumor, the anti-EpCAM antibody catumaxomab, was adminis-tered i.p. for malignant ascites. Besides targeting the peritonealtumor compartment, huA33-BsAb could suppress tumor growthand prolong survival in subcutaneous xenograft models ofaggressive mCRC. Furthermore, huA33-BsAb could recruit cyto-toxic T cells from the systemic circulation into the peritonealcavity for effective tumor kill—an important first step in T-cell–based immunotherapy. Unlike the robust response of s.c.models,i.p. tumors were not cured in all mice. However, consideringthe suboptimal engraftment of human cells in Balb/c derivedDKO mice without exogenous cytokines (36, 37), the efficacy ofhuA33-BsAb could be underestimated. One could even proposecombining injections of i.p. huA33-BsAb with i.p. T cells or withi.p. PBMC to enhance the antitumor effects.

T-BsAbs generally have higher tumor killing potency thanADCC-dependent IgG antibodies, possibly because of the pres-ence of competing IgGs present in the medium and/or the

intrinsic differences between cytotoxic T cells and NK cells. Ourin vitro TDCC assay showed that huA33-BsAb killed colon cancercell lines with an EC50 of around 4 to 66 pmol/L and 25 pmol/Lfor purified CD4þ memory T cells (Fig. 4B), which was 2 to40 times lower than the apparent affinity of the molecule. Thispotency is comparable to MEDI-565 (38), which has an EC50

of 91 pmol/L and higher than RG7802, which has an EC50 of0.24 to 6.28 nmol/L (39). One of the most potent T-BsAbsreported so far by Xu and colleagues (21) that targeted GD2 hasEC50s in the fM range. These differences may be attributed tothe differences in the intrinsic properties of antigens or in theparticular epitopes targeted by T-BsAb. Nevertheless, it is stillnot clear if high potency translates into clinical benefit becausemore efficient killing of tumor cells can be accompanied by moreon-target off-tumor toxicities, unless the therapeutic window isimproved (21). A tradeoff between potency and toxicity has to becarefully considered, especially for tumor antigens that are alsoexpressed in normal tissues.

Our analyses of cytokines and cytotoxic molecules in culturesupernatants provided a mechanistic basis for the action ofhuA33-BsAb. Cytokines belonging to Th1, Th2, and Th17 typeswere all involved, althoughwith different kinetics. Cytokines playa major role in modifying the immunologic landscape of thetumor microenvironment and may be important for the long-term efficacy and toxicity of T cell–based immunotherapy. Anextreme example is the uncontrolled simultaneous release of amyriad of cytokines that results in "cytokine storm," a potentiallyfatal condition, of which IL6 is one of the main components.Understanding how a tumoricidal cytokine pattern transitionsinto systematic toxicity will undoubtedly help improve current Tcell–based immunotherapy.

We also tested if increasing affinity of huA33-BsAb to3.2 pmol/L could improve T-BsAb potency. Although the facileyeast system was efficient in identifying novel mutations toimprove affinity, no significant improvement in EC50 amongthese variants was observed in TDCC assay. There appears to bean affinity ceiling of subnanomolar, beyond which furtherincrease in affinity will not significantly improve the potencyof T-BsAbs, as was also observed using other T cell–engagingmolecules (40). The molecular basis of this phenomenon isunclear. It is also possible that antibodies with these extremelyhigh-affinity molecules became irreversibly trapped by deadcells and no longer available during cytotoxicity assays.

Besides GPA33, both EpCAM and CEA are being tested astherapeutic targets for mCRC. EpCAM has the advantage ofbeing restricted to epithelial cells, which are not present inhealthy peritoneal cavity because it is of mesenchymal origin.CEA has the advantage of having a polarized expression patternwith expression restricted to the apical surface of normal colontissues. However, both proteins are expressed in a wide varietyof normal tissues and could potentially raise toxicity concerns.In contrast, the expression of GPA33 is specifically restrictedto colon tissues and no other organs, which could be advan-tageous in limiting its toxicity. Because the huA33 antibody isnot cross-reactive with mouse GPA33, it is impossible for us totest systemic toxicity in the murine xenograft models used.However, we reasoned that healthy normal colon cells could beshielded from the peritoneal cavity by the physical barriers ofserosa and muscular layers wrapping around colons. In addi-tion, on the luminal side, the continuous turnover of the gutepithelium carrying with it GPA33 could partially alleviate



this cross-reactivity issue (41). It is conceivable that removingFcR(n) affinity could further reduce antibody transport intothe gut lumen (42). More careful investigations using anti-bodies against this antigen in nonhuman primates or inhumans will be needed to address these issues. The resultsfrom ongoing Macrogenics' phase I clinical trial of MGD007(Clinicaltrials.gov, NCT02248805), a bivalent (1þ1) T-BsAbdirected at GPA33, with antitumor efficacy in a Colo205subcutaneous xenograft model, should provide valuable infor-mation on the clinical safety of such approaches.

Disclosure of Potential Conflicts of InterestZ. Wu, H. Xu, and NK Cheung have ownership interest as co-inventors on a

patent application for humanized anti-GPA33 antibodies huA33 filed byMemorial Sloan Kettering Cancer Center. NK Cheung reports receiving othercommercial research support from Ymabs and Abpro, has ownership interest(including stock, patents, etc.) inYmabs, Abpro, andEureka, and is a consultant/advisory boardmember for Abpro and Eureka. No potential conflicts of interestwere disclosed by the other author.

Authors' ContributionsConception and design: Z. Wu, H. Xu, NK CheungDevelopment of methodology: Z. Wu, H. Xu, NK Cheung

Acquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): Z. Wu, H.-F. Guo, H. XuAnalysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): Z. Wu, H.-F. Guo, H. Xu, NK CheungWriting, review, and/or revision of the manuscript: Z. Wu, H. Xu,NK CheungAdministrative, technical, or material support (i.e., reporting or organizingdata, constructing databases): Z. Wu, H.–F. Guo, H. Xu, NK CheungStudy supervision: NK Cheung

AcknowledgmentsWe thank Dr. Mamoru Ito of Central Institute for Experimental Animals,

Kawasaki, Japan, for providing the DKO mice. Z. Wu, H.F. Guo, H. Xu, andNK Cheung were partly supported by Enid A. Haupt Endowed Chair andthe Robert Steel Foundation. Technical service provided by the MSKAnimal Imaging Core Facility was supported in part by the NCI Cancer CenterSupport Grant P30 CA008748.

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 January 22, 2018; revised June 17, 2018; accepted July 31, 2018;published first August 6, 2018.

References1. Siegel RL, Miller KD, Jemal A. Cancer statistics, 2017. CA Cancer J Clin

2017;67:7–30.2. Ferlay J, Soerjomataram I, Dikshit R, Eser S, Mathers C, Rebelo M, et al.

Cancer incidence and mortality worldwide: sources, methods and majorpatterns in GLOBOCAN 2012. Int J Cancer 2015;136:E359–E86.

3. Van Cutsem E, Cervantes A, Adam R, Sobrero A, Van Krieken JH, Aderka D,et al. ESMO consensus guidelines for the management of patients withmetastatic colorectal cancer. Ann Oncol 2016;27:1386–422.

4. Bencsikova B, Bortlicek Z, Halamkova J, Ostrizkova L, Kiss I, Melichar B,et al. Efficacy of bevacizumab and chemotherapy in the first-line treatmentof metastatic colorectal cancer: broadening KRAS-focused clinical view.BMC Gastroenterol 2015;15:37.

5. Peeters M, Price TJ, Cervantes A, Sobrero AF, Ducreux M, Hotko Y, et al.Randomized phase III study of panitumumab with fluorouracil, leucov-orin, and irinotecan (FOLFIRI) compared with FOLFIRI alone as second-line treatment in patients with metastatic colorectal cancer. J Clin Oncol2010;28:4706–13.

6. Cutsem EV, K€ohne CH, L�ang I, Folprecht G, Nowacki MP, Cascinu S, et al.Cetuximab plus irinotecan, fluorouracil, and leucovorin as first-line treat-ment for metastatic colorectal cancer: updated analysis of overall survivalaccording to tumor KRAS and BRAF mutation status. J Clin Oncol 2011;29:2011–9.

7. Saltz LB, Clarke S, Díaz-Rubio E, Scheithauer W, Figer A, Wong R, et al.Bevacizumab in combination with oxaliplatin-based chemotherapy asfirst-line therapy in metastatic colorectal cancer: a randomized phase IIIstudy. J Clin Oncol 2008;26:2013–9.

8. Gaillard SL, Secord AA,Monk B. The role of immune checkpoint inhibitionin the treatment of ovarian cancer. Gynecol Oncol Res Pract 2016;3:11.

9. Galanina N, Kline J, Bishop MR. Emerging role of checkpoint blockadetherapy in lymphoma. Thera Adv Hematol 2017;8:81–90.

10. Overman MJ, Kopetz S, McDermott RS, Leach J, Lonardi S, Lenz H-J, et al.Nivolumab � ipilimumab in treatment (tx) of patients (pts) with meta-static colorectal cancer (mCRC) with and without high microsatelliteinstability (MSI-H): CheckMate-142 interim results. J Clin Oncol 2016;34 (15 suppl):3501.

11. Le DT, Uram JN, Wang H, Bartlett B, Kemberling H, Eyring A, et al.Programmed death-1 blockade in mismatch repair deficient colorectalcancer. J Clin Oncol 2016;34 (15 suppl):3003.

12. Le DT, Uram JN,WangH, Bartlett BR, Kemberling H, Eyring AD, et al. PD-1blockade in tumors with mismatch-repair deficiency. N Engl J Med2015;372:2509–20.

13. GarinChesa P, Sakamoto J, Welt S, Real F, Rettig W, Old L. Organ-specific expression of the colon cancer antigen A33, a cell surface targetfor antibody-based therapy. Int J Oncol 1996;9:465–71.

14. Welt S, Scott AM, Divgi CR, Kemeny NE, Finn RD, Daghighian F,et al. Phase I/II study of iodine 125-labeled monoclonal antibodyA33 in patients with advanced colon cancer. J Clin Oncol 1996;14:1787–97.

15. Welt S, Ritter G, Williams C, Cohen LS, John M, Jungbluth A, et al. Phase Istudy of anticolon cancer humanized antibody A33. Clin Cancer Res2003;9:1338–46.

16. Chong G, Lee FT, Hopkins W, Tebbutt N, Cebon JS, Mountain AJ, et al.Phase I Trial of 131I-huA33 in patients with advanced colorectal carcinoma.Clin Cancer Res 2005;11:4818–26.

17. Ciprotti M, Chong G, Gan HK, Chan A, Murone C, MacGregor D, et al.Quantitative intratumoural microdistribution and kinetics of (131)I-huA33 antibody in patients with colorectal carcinoma. EJNMMI Res2014;4:22 eCollection 2014.

18. Welt S, Divgi CR, KemenyN, Finn RD, Scott AM,GrahamM, et al. Phase I/IIstudy of iodine 131-labeled monoclonal antibody A33 in patients withadvanced colon cancer. J Clin Oncol 1994;12:1561–71.

19. Welt S, Divgi CR, Real FX, Yeh SD, Garin-Chesa P, Finstad CL, et al.Quantitative analysis of antibody localization in human metastatic coloncancer: a phase I study of monoclonal antibody A33. J Clin Oncol 1990;8:1894–906.

20. Ritter G, Cohen LS, Williams C, Richards EC, Old LJ, Welt S. Serologicalanalysis of human anti-humanantibody responses in colon cancer patientstreated with repeated doses of humanized monoclonal antibody A33.Cancer Res 2001;61:6851–9.

21. XuH, ChengM,GuoH, Chen Y,HuseM, CheungNK. Retargeting T cells toGD2 pentasaccharide on human tumors using bispecific humanizedantibody. Cancer Immunol Res 2015;3:266–77.

22. Lopez-Albaitero A, Xu H, Guo H,Wang L, Wu Z, Tran H, et al. Overcomingresistance to HER2-targeted therapy with a novel HER2/CD3 bispecificantibody. OncoImmunology 2017;6:e1267891.

23. ChengM,AhmedM,XuH,CheungNK. Structural designof disialoganglio-side GD2 and CD3-bispecific antibodies to redirect T cells for tumortherapy. Int J Cancer 2015;136:476–86.

24. Gao SH, Huang K, Tu H, Adler AS. Monoclonal antibody humanness scoreand its applications. BMC Biotech 2013;13:55.

25. Wu Z, Cheung NV. T cell engaging bispecific antibody (T-BsAb): fromtechnology to therapeutics. Pharmacol Ther 2018;182:161–75.

Wu et al.


26. Cheal SM, Xu H, Guo HF, Lee SG, Punzalan B, Chalasani S, et al. Ther-anostic pretargeted radioimmunotherapy of colorectal cancer xenografts inmice using picomolar affinity (8)(6)Y- or (1)(7)(7)Lu-DOTA-Bn bindingscFv C825/GPA33 IgG bispecific immunoconjugates. Eur J Nucl Med MolImaging 2016;43:925–37.

27. Williams DS, Bird MJ, Jorissen RN, Yu YL, Walker F, Zhang HH, et al.Nonsense mediated decay resistant mutations are a source of expressedmutant proteins in colon cancer cell lines with microsatellite instability.PLoS One 2011;5:e16012.

28. Yoon K, Lee S, Han TS, Moon SY, Yun SM, Kong SH, et al. Comprehensivegenome- and transcriptome-wide analyses of mutations associated withmicrosatellite instability in Korean gastric cancers. Genome Res 2013;23:1109–17.

29. Suter CM, Norrie M, Ku SL, Cheong KF, Tomlinson I, Ward RL. CpG islandmethylation is a common finding in colorectal cancer cell lines. Br J Cancer2003;88:413–9.

30. Ahmed D, Eide PW, Eilertsen IA, Danielsen SA, Eknas M, Hektoen M, et al.Epigenetic and genetic features of 24 colon cancer cell lines. Oncogenesis2013;2:e71.

31. Liu Y, Bodmer WF. Analysis of P53 mutations and their expression in56 colorectal cancer cell lines. PNAS 2006;103:976–81.

32. Ku JL, Park JG. Biology of SNU cell lines. Cancer Res Treat 2005;37:1–19.33. Ikediobi ON, Davies H, Bignell G, Edkins S, Stevens C, O'Meara S, et al.

Mutation analysis of 24 known cancer genes in the NCI-60 cell line set.Mol Cancer Ther 2006;5:2606–12.

34. Cheal SM, Fung EK, Patel MV, Punzalan B, Xu H, Guo HF, et al. Curativemulti-cycle radioimmunotherapy monitored by quantitative SPECT/CT-based theranostics, using bispecific antibody pretargeting strategyin colorectal cancer. J Nucl Med 2017;58:1735–42.

35. Kubicka U, Olszewski WL, Tarnowski W, Bielecki K, Zi �OŁKowska A,Wierzbicki Z. Normal human immune peritoneal cells: subpopulationsand functional characteristics. Scand J Immunol 1996;44:157–63.

36. Strowig T, Rongvaux A, Rathinam C, Takizawa H, Borsotti C, Philbrick W,et al. Transgenic expression of human signal regulatory protein alpha inRag2(�/�)g(c)(�/�) mice improves engraftment of human hematopoi-etic cells in humanized mice. PNAS 2011;108:13218–23.

37. Iwamoto C, Takenaka K, Urata S, Yamauchi T, Shima T, Kuriyama T, et al.The BALB/c-specific polymorphic SIRPA enhances its affinity for humanCD47, inhibiting phagocytosis against human cells to promote xenogeneicengraftment. Exp Hematol 2014;42:163–71.e1.

38. Oberst MD, Fuhrmann S, Mulgrew K, Amann M, Cheng L, LutterbueseP, et al. CEA/CD3 bispecific antibody MEDI-565/AMG 211 activation ofT cells and subsequent killing of human tumors is independent ofmutations commonly found in colorectal adenocarcinomas. mAbs2014;6:1571–84.

39. Bacac M, Fauti T, Sam J, Colombetti S, Weinzierl T, Ouaret D, et al. A novelcarcinoembryonic antigen T-cell bispecific antibody (CEA TCB) for thetreatment of solid tumors. Clin Cancer Res 2016;22:3286–97.

40. Liddy N, Bossi G, Adams KJ, Lissina A, Mahon TM, Hassan NJ, et al.Monoclonal TCR-redirected tumor cell killing. Nat Med 2012;18:980–7.

41. Scott AM, Lee FT, Jones R, Hopkins W, MacGregor D, Cebon JS, et al. Aphase I trial of humanized monoclonal antibody A33 in patients withcolorectal carcinoma: biodistribution, pharmacokinetics, and quantitativetumor uptake. Clin Cancer Res 2005;11:4810–7.

42. Yoshida M, Claypool SM, Wagner JS, Mizoguchi E, Mizoguchi A,Roopenian DC, et al. Human neonatal Fc receptor mediates transportof IgG into luminal secretions for delivery of antigens to mucosal den-dritic cells. Immunity 2004;20:769–83.



development of a tetravalent anti-gpa33/ anti-cd3 ... · bsab, and especially one targeting gpa33...

Documents