bispecific antibody platforms for cancer immunotherapy.pdf

Critical Reviews in Oncology/Hematology 92 (2014) 153165

Bispecific antibody platforms for cancer immunotherapyRoeland Lameris a, Rene C.G. de Bruin a, Famke L. Schneiders a,

Paul M.P. van Bergen en Henegouwen b, Henk M.W. Verheul a,Tanja D. de Gruijl , Hans J. van der Vliet

a Department of Medical Oncology, VU University Medical Center, De Boelelaan 1117, 1081 HV Amsterdam, The Netherlandsb Division of Cell Biology, Department of Biology, Faculty of Science, Utrecht University, Padualaan 8, 3584 CH Utrecht, The Netherlands

Accepted 8 August 2014

Contents

1. Introdu2. Curren

2.1. T

2.2. S

2.3. O3. Limitat

3.1. I3.2. S3.3. A

4. Perspec5. Targeti

5.1. 5.2. I5.3. N

6. ConcluConflicAcknowReferenBiograp

Abstract

Over the bispecific anstudies of va

Abbreviatchain variablediabody; scDb

CorresponE-mail ad

http://dx.doi.o1040-8428/ction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 154tly available bispecific antibody platforms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 154

rifunctional hybrid antibodies (Triomab) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1542.1.1. Catumaxomab . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1542.1.2. Ertumaxomab . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1552.1.3. FBTA05 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 156

ingle chain variable fragment (scFv) based platforms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1562.2.1. Tandem scFv (TaFv). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1562.2.2. Bispecific diabodies (bsDb) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157

ther bispecific antibody based platforms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157ions of the type of bsAb constructs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157mmunogenicity of bispecific antibody constructs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157ize of the construct: Pay off between circulation half-life time and tumor penetration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 158ntibody construct stability and manufacturing difficulties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 158

tives related to the bsAb construct . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 159ng specific lymphocyte subsets to maximize efficacy. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 159 T-cells . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 160nvariant natural killer T-cells (iNKT) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 160atural killer cells (NK) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 160

ding remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161t of interest statement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161ledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161ces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161hy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 165

past decades advances in bioengineering and expanded insight in tumor immunology have resulted in the emergence of noveltibody (bsAb) constructs that are capable of redirecting immune effector cells to the tumor microenvironment. (Pre-) clinicalrious bsAb constructs have shown impressive results in terms of immune effector cell retargeting, target dependent activation

ions: VH, variable heavy chain domain; VL, variable light chain domain; mAb, monoclonal antibody; bsAb, bispecific antibody; scFv, single fragment; Triomab, trifunctional hybrid antibody; TaFv, tandem single chain variable fragment; BiTE, bispecific T-cell engager; bsDb, bispecific, single chain diabody; DART, dual affinity retargeting molecule; VHH, variable domain of heavy chain-only Ab.ding author. Tel.: +31 20 4441295; fax: +31 20 4444355.dress: [email protected] (H.J. van der Vliet).

rg/10.1016/j.critrevonc.2014.08.003 2014 Elsevier Ireland Ltd. All rights reserved.a a,

154 R. Lameris et al. / Critical Reviews in Oncology/Hematology 92 (2014) 153165

and the induction of anti-tumor responses. This review summarizes recent advances in the field of bsAb-therapy and limitations that wereencountered n be expected to take the bsAb approach successfully forward. 2014 Else

Keywords: B -cancer

1. Introdu

Several bodies (mkilling thrdependent cellular cytto its tumoreceptors (killer (NKin CDC anity and/or demonstratious mAbs(anti-humacetuximab [1]. Althoube influenctherapeuticsecondary exploit the e.g. hampeing for Fcinadequatelarge size (FcR on immAb-tumoefficacy [4]

Bispecifisimultaneotions and cimmune rechemical cas a result hybridomanone of thtrials, as aation of hehuman anthampered that bsAb rfused mouvariable fra

Here, wcurrently being developed for tumor retargeting of immunecells and discuss thus far obtained (pre-)clinical results andencountered limitations. Furthermore, we will elaborate onpotential ways to take the bsAb approach forward.

urren

Trifun

troducm ass

uadromct antfic ratermedes-pre

was gortion eclinision melial cd redNK ceation wtumoricomoortion ice treisting ng (Fa

TriomL/6 assing, demnationth pa% sun agation of tum

injectab trral im

estive lopmeding ce disc

. CatuCatumaxomab, an anti-EpCAM-anti-CD3 Triomab, was

the first Triomab studied in patients. A phase-I trial of a sin-gle intravenous (i.v.) dose of catumaxomab in patients withnon-small cell lung cancer (NSCLC), established a maximum. Furthermore, we will discuss potential future developments that cavier Ireland Ltd. All rights reserved.

i-specific antibodies; Dual specific retargeting; Immune effector cells; Anti

ction

clinically available therapeutic monoclonal anti-Abs) can induce immune-mediated tumor cellough mechanisms that include complement-cytotoxicity (CDC) and antibody-dependent

otoxicity (ADCC). Following binding of the mAbr target, interactions of the Fc-portion with Fc-FcR) expressed by effector cells (e.g. natural) cells, macrophages and T-cells) may resultd ADCC and subsequent antitumor cytotoxic-

phagocytosis. In clinical series, ADCC has beened to significantly enhance the efficacy of var-, including rituximab (anti-CD20), trastuzumabn-epidermal-growth-factor receptor 2 (Her2)) and(anti-epidermal-growth-factor receptor (EGFR))gh data are inconsistent, clinical responses mayed by FcR polymorphisms [2]. Granting their

efficacy can in part be attributed to beneficialimmune effects, it is clear that mAbs still do notfull potential of the immune system as effects arered by circulating immunoglobulins (Ig) compet-R binding spots on immune effector cells, and

tumor-target penetration due to their relatively150 kDa) [3]. Furthermore, binding to inhibitorymune cells may result in internalization of the

r target-FcR complex reducing its therapeutic.

c Abs (bsAbs), capable of binding two targetsusly, lack several of the above described limita-an potentially induce a more powerful anti-tumorsponse. The first bsAbs were engineered by eitherrosslinking or exchange of different heavy chainsof fusion of two hybridoma cell lines (i.e. hybrids or quadromas). Despite some clinical effects,ese bsAbs made it to advanced stage clinical

low production yield owing to random associ-avy/light chains and immunogenicity caused byi-mouse/rat antibodies (HAMA/HARA) severelyclinical applicability [5]. It was not until 1995esearch was sparked again by the introduction ofse-rat hybrid antibodies and tandem single-chaingments (TaFv) [6,7].e will review the main bsAb formats that are

2. C

2.1.

Inrandosic qdistinspeciand tspeciationFc-p

Pradheepithstrate(e.g. activated the pFc-pin mcons

bindiof theC57Bexpretivelyelimiof boa 100tectiodepleloss omalsTriomhumosuggdeveincluwill b

2.1.1therapy

tly available bispecic antibody platforms

ctional hybrid antibodies (Triomab)

ed in 1995, this platform offered a solution to theociation of heavy/light chains observed in clas-a technology. By combining the halves of two

ibodies, a tumor-specific mouse IgG2a and a CD3- IgG2b, a full-size functional mAb was engineered

Triomab (Trion pharma Inc.) (Fig. 1b). Due toferential heavy/light chain pairing random associ-reatly reduced. Interestingly, the hybrid mouse/ratwas able to activate FcR+ accessory cells [6,8,9].cal studies with a Triomab targeting epithelial cellolecule (EpCAM), expressed by the majority ofancers, and CD3 expressed by T-cells, demon-

irection and activation of T- and accessory cellslls, dendritic cells (DC) and macrophages). T-cellas complemented by the induction of T-cell medi-

lysis, cytotoxic cytokine release, and ADCC inlar range [8,9]. The additive value of a functionalwas underscored by enhanced tumor protectionated with a Triomab compared to a similar bsAbof two chemically cross-linked fragment antigenb) regions (F(ab)2) (Fig. 1c). In vivo assessmentab injected intraperitoneally (i.p.) in a syngeneic

nd BALB/c mouse model using (human) EpCAM B16-melanoma and A20-lymphoma cells, respec-onstrated a significantly improved tumor cell

compared with the simultaneous administrationrental antibodies. In the Triomab treated grouprvival rate, complete tumor eradication and pro-inst tumor rechallenge was reported. Selectivef both CD4+ and CD8+ T-cells resulted in a markedor protection and survival. Of note, only the ani-ed with human EpCAM expressing tumors andeatment developed strong anti-EpCAM specificmune responses [10], with additional evidence

of epitope spreading. These data resulted in thent and clinical evaluation of a number of Triomabs,atumaxomab, ertumaxomab and FBTA05 whichussed below.

maxomab

R. Lameris et al. / Critical Reviews in Oncology/Hematology 92 (2014) 153165 155

Fig. 1. Variabdark orange, and light oranPeptide linkerTriomab, bisplinked fragmefragment; (e)pecific diabodretargeting mvariable domaences to colorof this article.

tolerated dofavorable sease survivHAMA or

More extion of cadue to varphase II/IIIsymptomatan increas[12,13]. An

vascular endothelial growth factor (VEGF), increased levelsof activated CD4+ and CD8+ T-cells as well as an eliminationof CD133+/EpCAM+ cancer stem cells (CSC) compared to

baselirtantlyted in

(medto nexpproxise, s

ly redovemes 68 d

ignificboth Imporesulvivaltime ing aLikewicantimprversu

cal sle heavy chain domains (VH ) are depicted in dark blue andvariable light chain domains (VL) are depicted in light bluege. Orange and blue indicate arms with different specificities.s are shown as gray lines. (a) mAb, monoclonal antibody; (b)ecific rat/mouse antibody; (c) F(ab)2, two chemically cross-nt antigen binding (Fab) regions; (d) scFv, single chain variable

TaFv, tandem single chain variable fragment; (f) bsDb, bis-y; (g) scDb, single chain diabody; (h) DART, dual affinity

olecule; (i) Heavy chain-only antibody; (j) bsVHH, bispecificins of heavy chain-only Ab. (For interpretation of the refer-

in this figure legend, the reader is referred to the web version)

se (5 g with 40 mg dexamethasone) and reportedurvival in several patients with advanced-stage dis-ing past 26 months. None of the patients developedHARA within 28 days after infusion [11].tensive studies have evaluated i.p. administra-

tumaxomab in patients with malignant ascitesious EpCAM positive tumors. In a randomized

study, involving heavily pre-treated patients withic malignant ascites, infusions were given withing dose (10150 g) on days 0, 3, 7 and 10alyses of ascites demonstrated decreased levels of

71 vs 44 ddevelopmepoints, waafter the ladetail in Secommonlyvomiting), Common Tered to be tand lymphileus (3.2%on these dpean Medimalignant mas wherefeasible.

2.1.2. ErtuErtumax

demonstrattoxicity agand to effiwhere trastineffective of 3 ascendays 1, 7 cer induceresponses AEs were immune rement relate(5.9%), syand, aggragrade 3 topenia (76note, the nAE seemefirst dose, of patientsnated premplans.ne levels and non-treated control patients [14]., compared to control patients catumaxomab

a significant prolongation of puncture-free sur-ian 11 vs 46 days, p < 0.0001) and an increasedt paracentesis (median 13 vs 77 days), preclud-imately 5 therapeutic paracentesis procedures.ymptoms and signs related to ascites were signif-uced. Although a positive trend was observed innt of median overall survival (72 (95% CI: 6198)ays (95% CI: 4981), this only reached statisti-ance in patients with gastric carcinoma (medianays; p = 0.0313) [15]. Notably, HAMA/HARA

nt, which strongly correlated with clinical end-s observed in the majority of patients (>70%)st treatment cycle and will be discussed in morection 3.1 [16]. Adverse effects (AE) were most

cytokine release-related (i.e. pyrexia, nausea andmanageable and generally reversible. Reportedoxicity Criteria (CTC) grade 3 AE consid-

reatment related included abdominal pain (9.6%)openia (7.0%). Reported serious AE included) and gastric hemorrhage (0.6%) [15]. Based

ata, catumaxomab was approved by the Euro-cine Agency (EMA) for the i.p. treatment of

ascites in patients with EpCAM-positive carcino- standard therapy is not available or no longer

maxomabomab, an anti-Her2-anti-CD3 Triomab, wased to initiate effective immune mediated cyto-ainst Her2 expressing tumor cell lines in vitrociently lyse low Her2 expressing target cellsuzumab, a mAb targeting Her2, was completely[17]. In a phase I trial, i.v. administrationding doses (10200 g) of ertumaxomab on

and 13 to patients with metastatic breast can-d strong inflammatory reactions with clinicalin some patients. The severity and number ofdose related, with dose limiting events being

lated. Reported serious AE classified to be treat-d included fully reversible severe hypotension

stemic inflammatory response syndrome (5.9%)vation of congestive heart failure (5.9%). CTCtoxicities included fully reversible lymphocy-%) and elevation of liver enzymes (47%). Ofumber of CTC grade 3 toxicities and seriousd to be dose related. Within 6 weeks after theHAMA/HARA were observed in 25 and 31%, respectively [18]. A phase II trial was termi-aturely due to changes in company development


2.1.3. FBTA05FBTA05, an anti-CD20-anti-CD3 Triomab, demonstrated

effective CD20+ lymphoma killing in vitro. A pilot studyinvolving lymphomatreated withsion or perprompt butI/II study is

2.2. Singleplatforms

Some ofbe overcomdomains. Avast amounantibody frFv (scFv), of heavy (V(Fig. 1d) [2ied extensidiabodies (

2.2.1. TandBy cova

tide linker formed (Fibetween thget ligationtumor pene

One typtumor assolinker (GGtermed bispeffectively resulting inlar concentVideo microf target ction of targ24 times lcell lysis inbetween taof cytotoxiB) [24]. Asing of BIToccurred in

2.2.1.1. Blmost advamurine scFclinical stucell NHL aresponses 15 g/m2/2

weeks. Importantly, continuous infusion was required toensure sustained effective serum levels due to a very shortserum half-life time of less than 2 h due to renal excretion

7].ased otients D) ponts wee of 15itted t

patie, 12 oreatmenstrat

patiencover

nt peiently ounts

therap resul

reporsed/reeducenteen tologiost cotoms r ongoated Cines lifferenpenia

ly revencephctivelently . Also) pre

descrr startrticoid

ent cultipleent s

ngoingCT01

.2. Alg EpCA) hnd pr01723re-cli

noembprostapatients with recurrent/refractory non-Hodgkin (NHL) or chronic lymphatic leukemia (CLL)

i.v. FBTA05 followed by donor lymphocyte infu-ipheral blood stem cell transplantation showed a

transient response in several patients [19]. A phase ongoing (NCT01138579).

chain variable fragment (scFv) based

the mentioned limitations of whole antibodies cane by reducing antibodies to their minimal bindingdvances in genetic engineering have provided at of such antibody like constructs, including smallagments. One such fragment, termed single-chainis made by the association of the variable partsH) and light (VL) chains through a peptide linker0]. Two formats based on scFvs have been stud-

vely, namely tandem scFv (TaFv) and bispecificbsDb).

em scFv (TaFv)lent bonding of two scFvs with a flexible pep-in a tandem orientation, a bispecific TaFv can beg. 1e). It is expected that the flexible orientatione two binding domains enhances simultaneous tar-, while their small size (5560 kDa) allows rapidtration [3].e of TaFv, made by fusing an anti-CD3 to an anti-ciated antigen (TAA) scFv via a 5 residue peptideGGS), has been extensively studied. These TaFv,ecific T-cell engagers (BiTE) (Micromet Inc.) canredirect and activate polyclonal T-cells in vitro

a highly cytotoxic response at pico- to femtomo-rations in the absence of co-stimulation [21,22].oscopy revealed that BiTEs allowed serial killingells by engaged T-cells with complete elimina-et cells at effector-to-target ratios of 1:5, which isower than ratios needed for Triomabs [23]. Targetvolved the formation of tight cytolytic synapses

rget- and effector cells and the subsequent releasec effector molecules (e.g. perforin and granzyme

T-cell activation depended on multivalent bind-Es due to their low affinity for CD3, activation

a strictly target dependent fashion [25].

inatumomab. Blinatumomab, being the first andnced BiTE in clinical studies, is derived fromvs targeting human CD19 and CD3. Several

dies evaluated its efficacy in various types of B-nd leukemia. Complete (CR) and partial tumor

(PR) were demonstrated upwards of a dose of4 h, given as continuous infusion over 4 or 8

[26,2B

in pa(MRPatiea dospermof 20cycleous tdemoof 20but renificatranscell cafterlikelywere

relapwas r

Sevehema

Msympundeactivcytokthe dleukopleteand respefrequation(HLHbeenlowecoco

treatmM

treatmare o

and N

2.2.1getin(PSMlial aNCTated pcarcianti-n results in B-NHL, a phase II study was initiatedwith persistent/relapsed minimal residual diseasesitive B-cell acute lymphatic leukemia (B-ALL).re treated with a continuous 4-week i.v. infusion at

g/m2/24 h every 6 weeks with responders beingo receive up to 3 consolidation cycles. Sixteen outnts became MRD-negative at the end of the firstf which had molecular refractory disease to previ-nt [28]. Long term follow-up (median 33 months)ed a hematologic relapse-free survival in 12 outts [29]. In all 20 patients T-cells declined rapidly

ed within days to above baseline levels, with a sig-rcentage of reappearing CD8+ and CD4+ T-cellsexpressing the early activation marker CD69. B-

dropped below 1 cell/L and did not recover untily cessation. The decline in B-cell counts most

ted from redirected cell lysis [30]. Similar resultsted in an interim analysis of a phase II trial infractory B-ALL. In this study the starting dosed (5 g/m2/24 h in the first week) to prevent AE.out of 25 patients achieved CR or CR with partialcal recovery [31].mmon AEs with blinatumomab included flu-likeduring the first days of treatment which resolveding infusion. AEs correlated with peak levels ofD8+ and CD4+ T-cells and with increased serumevels. Commonly reported CTC grade 3 AE int studies included lymphopenia (up to 77%) and

(up to 47%). More serious AEs, although com-ersible, included neurological AE (e.g. seizuresalopathy in up to 4.8% and 8.6% of patients,

y) presumably caused by released cytokines andresulted in treatment interruption or discontinu-, cases of hemophagocytic lymphohistiocytosis

sumably caused by blinatumomab treatment haveibed [2630,32]. Mitigating measures, includinging doses, a double-step dose increase and glu-

administration, were protective and allowed forontinuation or resumption [2630].

trials designed to determine the optimal dose,chedule and clinical efficacy of blinatumomab

(NCT01741792, NCT01471782, NCT01207388466179).

ternative TaFv constructs. TaFv constructs tar-AM and prostate specific membrane antigen

ave entered phase I clinical trials in epithe-ostate cancer, respectively (NCT00635596 and475). A multitude of other TaFvs is being evalu-nically, including constructs targeting EGFR [33],ryonic antigen (CEA) and CD33. Of interest, an

te specific cell antigen (PSCA)-anti-CD3 TaFv


was recently successfully humanized without compromisingits in vitro and in vivo characteristics. Humanization mayprohibit HAMA associated AE [34].

Efforts ing site toIndeed, anstrated enhcells in vitran anti-CD

2.2.2. BispIn contr

associationfrom differlinker precbsDbs havefashion andindividual binding inbility of thdimerizatioin part, belinker betw(Fig. 1g) [be achieveC-terminusgeting (DAextended stand demon

Notablytide linker formed [37dissociatioin vivo expeto bsDbs [4(114 kDamay howev

2.2.2.1. PrbsDb provspecificallycer cells in in xenograan anti-CDbsDb [46].bsDb withaffinity, buTaFv [47]. cross-linkinof the varorder) can to the TaFmay also 37 C [48].

The potside-by-sid

bearing identical CD3 and CD19 Fv-domains. The DARTmolecule outperformed the BiTE molecule with respectto both the induction of B-cell lysis and stimulation of

l activrring s. Basecular mrm th

of fixaved hhave r

more r

ed prorently ted.

Other

lthougned feeredal stugineellow,

ange t to a hfic for3], a fhe hum2D ligne ef

minuscific gates

I mol showls in vd as cy da

imitat

Immu

istoriciated nistratA/HAine red, i.v.nd, HA

ent cmit anletelyNF-have also focused on adding a third bind- TaFv, thereby creating triple bodies (scTb).

anti-CD16-anti-CD33-anti-CD19 scTb demon-anced lysis of CD33-CD19-positive leukemico, compared to both an anti-CD16-anti-CD19 and16-CD33 TaFv [35].

ecic diabodies (bsDb)ast to TaFvs, bsDbs are formed by non-covalent

of two scFvs, consisting of a VH and VL domainent parent Abs, connected with a very short peptideluding intra-chain domain interactions (Fig. 1f).

their variable domains orientated in an opposed are very rigid and small (60 kDa) [36]. Becausechains are not covalently associated, VH and VLteractions remain a limiting factor in the sta-e molecule, possibly resulting in disintegration,n and aggregate formation. This can, at least

prevented by introducing an additional peptideeen the scFv, forming a single chain bsDb (scDb)3739]. Alternatively, an increased stability cand by introducing a disulfide bond between the

of each scFv, resulting in a Dual-Affinity ReTar-RT) molecule (Fig. 1h). DART molecules have anorage and serum stability, do not form aggregatesstrate potent in vitro and in vivo activity [40]., by varying the length of the additional pep-in scDbs, tetramers and tandem bsDbs can be,4143] with higher affinity, a reduced target cell-n rate and improved storage stability. Furthermore,riments showed higher serum retention compared1], possibly due to their increased affinity and size) reducing renal clearance. Their increased sizeer decrease effective tumor penetration.

e-clinical studies. An anti-PSMA-anti-CD3ed to be potent for CD3+ T-cell tumor targeting

inducing lysis of PSMA-expressing prostate can-vitro and inhibiting human prostate cancer growthfted mice [42]. Similar results were reported for19-anti-CD3 [44,45] and an anti-EGFR-anti-CD3

Direct comparison of an anti-EGFRanti-CD3 its TaFv equivalent indicated an equal bindingt enhanced in vitro cytotoxicity in favor of theAs the compact structure of bsDbs might hinderg, it is interesting to note that rearrangementiable domains (e.g. varying the VH-linker-VLlead to superior in vivo results when comparedv equivalent, though this in vivo superiority

result from the enhanced stability of bsDbs at

ency of DARTs was demonstrated in an in vitroe comparison of blinatumomab and a DART

T-celoccu

matsmoleplatfolevelobsermay ing ayieldapparepor

2.3.

Adesigenginclinicto, enthat aexchfusedspeci[52,5and tNKGimmuC-terbispeconjuclassbeenT-celcusse

effica

3. L

3.1.

Hassoc

admiHAMcytokcusse

AE atreatmto licompanti-Tation, with T-cell activation and proliferationolely in a target-cell dependent manner in both for-d on identical recognition specificities and similar

asses, differences between the DART and BiTEus resulted from structural differences, e.g. in thetion between the two opposing binding sites. The

igher association rate and target affinity of DARTsesulted in prolonged intercellular contacts creat-obust cytotoxic T-cell response [49]. While bsDbsmising results in pre-clinical studies, with someoutperforming TaFv, no clinical data have yet been

bispecic antibody based platforms

h a number of alternative bispecific constructsor tumor targeting of effector cells have been, to date only a few have been evaluated in pre-dies [20,50]. These include, but are not limitedred human monospecific IgG1 and IgG2 subtypesunder appropriate redox conditions, heavy chaino form full-length bsAb [51], an anti-CD3 scFvigh-affinity monoclonal T-cell receptor (mTCR)

a tumor-associated MHC-1 antigen (ImmTAC)usion protein consisting of a scFv targeting CEAan UL16 binding protein 2 (ULBP2), which as anand can be used for targeting NKG2D-expressing

fector cells [54] and, a scFv fused to the N- and of a trimerizing scaffold domain, resulting in ahexavalent trimerbody [55]. In, addition fusion

consisting of an anti-TAA coupled to an MHCecule containing a selected antigenic peptide haven to be able to effectively redirect and activateivo [56,57]. These platforms are not further dis-limited preclinical efficacy data and no clinicalta are presently available.

ions of the type of bsAb constructs

nogenicity of bispecic antibody constructs

ally, murine- and rat-derived mAbs have beenwith short serum half-life time upon repeatedion due to the formation of neutralizingRA and the related susceptibility to trigger a

lease syndrome (CRS) in patients. Indeed, as dis- dosing of Triomabs was limited by immunological

MA/HARA formation may restrict repetition ofycles. Though humanization is generally thoughtti-drug antibody (ADA) formation, it cannot

circumvent this, as was demonstrated by the mAb adalimumab, where up to 89% of patients


developed neutralizing human anti-human Abs (HAHA). Asdirect ligation of effector cells through e.g. FcR may alsotrigger CRS, immune-related AEs of mAbs cannot alwaysbe preventbeen showactually prdevelopmecatumaxomvival [16]. Asimply refl

As indidestructionof an anti(FcRIIIAamplified ational Fc-tain vitro stuof FcR+ cand activat[17,63] theing of thescross-link e

By elimto their mand it is be[38]. In supderived) blsteady-statlow immunADA mightion. BsDbcompact si

Concerntion duringuncontrolleMoreover, though theing treatme[30].

3.2. Size ohalf-life tim

The Fc-(to >10 dae.g. endothresult in swbsAb fragmallow for etion due toof 4050BiTEs wasmaximize is availableeither treatless prolon

for optimal therapeutic efficacy [58]. Hence various meth-ods have been successfully deployed to extend the serumhalf-life time of small-sized constructs including PEGyla-

N-glysing a6 Ab f

to anlife timgrafteeduce

hindrted fo5].ompar

ents homoo theiaviditumor rgisticking t

tumotrationnced

constgly, raspecifiood ra

Antiboulties

s meniomabared turificaared fae aftertated ss, as ed as

can bcules orrectn cellisadvxpressuced aditionRTs)

e exprria [3ough

l engins in hDb med [38,58,59]. Despite these drawbacks, it hasn for various mAbs that both ADA and CRS canomote tumor destruction [16,60]. Indeed, HAMAnt in patients with malignant ascites treated withab correlated with prolonged puncture-free sur-lthough tentative, HAMA development may also

ect a less suppressed immune system.cated, targeting the FcR can enhance tumor

through ADCC. Indeed, the combination-CD19-anti-CD3 and an anti-CD19-anti-CD16) bsDb had a synergistic effect [61]. Similarly annti-tumor response was seen after adding a func-il to a scDb [62]. However, as was observed indies with catumaxomab, simultaneous targetingells and T-cells may also result in cross-linkage

ion of these cells in the absence of target antigenreby limiting the intended efficacy of tumor target-e immune cells. Other tri-specific constructs thatffector cells may exhibit similar limitations.inating the Fc-region Abs are reduced in size

inimal binding domain (e.g. TaFvs and bsDbs)lieved that they are therefore less immunogenicport of this, patients treated with (murine mAb-

inatumomab, developed no HAMA and had stablee levels over the course of treatment, indicatingogenicity [27,28]. Of note, in this case lack oft also have been due to effective B-cell deple-

s may be even less immunogenic due to their veryze [37,42,45].s have been raised about ongoing T-cell activa-

BiTE-treatment, with several patients developingd immune activation (i.e. CRS and HLH) [32].persistent T-cell activation may induce anergy,

observed continued B-cell suppression dur-nt essentially reflects preserved T-cell function

f the construct: Pay off between circulatione and tumor penetration

domain of Abs enhances the serum half-life timeys) through neonatal-FcR mediated recycling byelial cells; hence absence of the Fc-tail mayift clearance from the circulation. Small sizedents (5060 kDa), such as TaFv and bsDb, alsoasy extravasation, renal filtration and degrada-

a glomerular filtration barrier (GBM) threshold kDa [3,38,64]. Indeed, continuous infusion of

required to maintain stable plasma levels andclinical efficacy [26]. At present no information

on the impact of such continuous infusions onment costs or patient quality of live, neverthe-ged exposure time is often considered critical

tion, or u

CD1scDbhalf-xeno

ting rstericrepor[64,6

Cfragmmore

due thigh rior tsyne

TabulkypeneenhabsAbestinnon-

to-bl

3.3. difc

Ain Trcompgle pappeactivfacilitheleneedscFvmoleto incmaliathis dare e

prodan adin DAcan bbacte

ThtionaTaFvof sccosylation, fusion to human albumin (covalentlylbumin-binding domains), or linkage to anti-ragments [64]. Linkage of an anti-CEA-anti-CD3

albumin-binding domain resulted in a prolongede and a 5-fold increase in accumulation in

d CEA+ tumors in mice. Nevertheless, in this set-d cytotoxicity was observed [65], possibly due toance. Similar reductions in cytotoxicity have beenr other serum half-life time extending strategies

ed to large IgG (150 kDa) molecules, small bsAbexhibit improved tumor penetration and display ageneous distribution within tumors; furthermore,

r multivalent nature bsAb fragments tend to havey with prolonged target retention [3]. Both supe-penetration and target retention may result in a

effect on tumor destruction.he above into consideration, especially solid andrs may require strong and homogeneous tumor

provided by small sized constructs. Due to theirtumor penetration, serum levels of small-sizedructs might not necessarily reflect efficacy. Inter-pid clearance from the circulation may reducec (off-target) cytotoxicity due to favorable tumor-tios.

dy construct stability and manufacturing

tioned, species restricted heavy-light chain parings resulted in a 3.5-fold higher production yieldo conventional quadromas and allowed for a sin-tion step. The in vivo stability of catumaxomabvorable, with up to 100% being immunologically

three days in ascites [66]. These properties greatlyproduction and clinical development [6]. Never-with all whole mAbs, mammalian cells are oftena host, increasing cost and production time [67].e expressed in bacteria, but tandem (i.e. TaFv)tend to form insoluble aggregates in bacteria due

folding and therefore require production in mam-s for high production yields [7,21,41]. bsDbs lackantage, but since two different polypeptide chainsed within one cell, inactive homodimers can belongside the active heterodimers. By introducingal peptide linker (as in scDbs) or disulfide bond (as

homodimerization can be prevented. While scDbessed in bacteria, disulfide bonds reduce yields in

7,38,40,68]. long-term stability of TaFvs may require addi-eering [40,69], scDb are 2-3-fold more stable than

uman plasma at 37 C [70]. Biological propertiesay however be strongly affected by even modest


variations in their composition (e.g. differences in linkerlengths or the order of variable domains) [38].

4. Perspec

As outliretargetingpeutic respdifficultiescussed limibe recognirelatively lbility, (iii) and homogficiently lo(v) have hibut (vi) safinally, (viibsAb constextent.

The varnaturally oing llama, if not all,domains o(Ablynx Inand specifitional Absincluding ctopes, due[3,73]. Spephage-dispimmunizedbacteria, ygation andscFvs, areThe relativVHH into mers that r[7578].

Owing circulationa VHH tarBio-distribanti-EGFRhalf-life tithe anti-EGhomogeneoEGFR-anti60% of tumhigh homomost likelyStill, VHHtion [75] wADA devel

evaluated for immune effector cell retargeting, however basedon the above listed properties, bsVHHs may prove to bevery efficient. Future studies will have to determine if these

rties w

rgeticy

ordernduce-portioue to ngly distic C

in hedies affecto

rgetin of T

rent imts. T-hry celeas Tatory ple, Tffect

Trege tum

of cantly it

scDbtor T-cof the

suppof CDabrogd Th-

espited that in thry antimay tor T-ited atls werOD/Ss [69]s of yD3-exever, itial Ag to spts.tives related to the bsAb construct

ned, currently used bsAb constructs can induce of immune effector cells and can result in thera-onses, yet improvements to overcome encountered

are required and appear feasible. Based on dis-tations, several ideal characteristics of bsAbs canzed: they should be (i) easy to manufacture atow cost, (ii) have a high serum and storage sta-be small (5060 kDa or less) to allow for rapideneous tumor penetration, yet (iv) have a suf-ng half-life time to induce a therapeutic effect,gh target-cell affinity to ensure tumor retention,feguard strictly target-dependent activation and,) be non-immunogenic. All previously discussedructs lack several of these characteristics to some

iable domain of heavy chain-only Abs (Fig. 1i),ccurring in the family of Camelidae, includ-camel and dromedary [71,72], may fulfill many,

of these ideal characteristics. These variablef heavy chain-only Abs (VHH) or Nanobodiesc.) (Fig. 1j) share the same large, diversifiedc repertoire of Ag binding sites as conven-

but also allow specificity to unique epitopes,ryptic and not otherwise easily accessible epi-

to their distinctive three-dimensional structurecific VHHs are easily retrieved after panning of alayed rearranged VHH-gene pool cloned from an

camelid, and can inexpensively be produced ineast or mammalian cells [74]. Solubility, aggre-

degradation problems often encountered with prevented due to their single domain nature.ely small size (15 kDa) permits easy linkage ofdimers (35 kDa), trimers (50 kDa) or multi-etain rapid and homogeneous tumor penetration

to their size VHH are rapidly cleared from the. This can, however, be prevented by fusion togeting e.g. mouse/human albumin [79] or FcR.ution studies in mice with a radiolabeled bivalent-anti-albumin VHH demonstrated an extendedme. Blood clearance was similar to that ofFR mAb cetuximab, whereas more deep andus tumor penetration was observed using the anti-

-albumin VHH (Ab-construct binding to 100% vsor cells) [79]. VHH amino acid sequences share

logy with the human type 3 VH domain (VH3), accounting for their low immunogenicity [72,78].

are usually humanized before clinical applica-ith preliminary clinical data thus far reporting noopment [80]. As yet no bispecific VHH has been

prope

5. Taefca

Inand iCD3tive dstrikiagonstormtrageent e[59].

Tarangediffesubsematowherregulexam

ous e

Highin thtypesReceCD3effection Tregstion and recte[87].

DstrateresulmatoThis effecrecru

T-celin a NBiTE

Aall CHowpotengetineffecill translate to the clinic.

ng specic lymphocyte subsets to maximize

to effectively recruit and activate all T-cell subsets tumor cell lysis, most bsAbs target the uniformn of the TCR complex. Targeting T-cells is attrac-

their destructive potential, but carries a risk as wasemonstrated by TGN1412, a monospecific super-D28-mAb that induced a life threatening cytokinealthy volunteers [81]. Measures to preclude suchre required, and include strictly target depend-r cell activation as a means of improving safety

g of CD3 results in the recruitment of a wide-cells, including CD4+, CD8+, T-cells andmunoregulatory and immunosuppressive T-cellelper 1 (Th1-)cells are considered to be proinflam-

ls that play an important role in tumor immunity,h2-cells and most notably CD4+CD25+ Foxp3+T-cells (Tregs) may promote tumor growth. Forregs have been recognized for their deleteri-

in malignancy by inducing immunosuppression. numbers have been detected systemically andor microenvironment in patients with variousncer and correlate with poor survival [8286].

has been demonstrated that an anti-PSCA-anti- not only effectively re-directed and activatedells, but also Tregs resulting in an increased secre-

immunosuppressive cytokine IL-10. Redirectedressed the proliferation and cytokine produc-4+ effector T-cells both in vitro and in vivo

ated the antitumor effector function of redi-cells thereby promoting tumor growth in vivo

these drawbacks, multiple studies have demon-t treatment with anti-TAA-anti-CD3 bsAbs cane induction of an overall effective proinflam-tumor immune response and in clinical responses.be explained by an initial recruitment of CD8+cells, with CD4+ T-cells, including Tregs, being

a later stage [21,69]. Furthermore, tumor residente sufficient to induce substantial tumor reductionCID xenotransplanted mouse model treated with.

et it remains unclear whether tumor targeting ofpressing T-cells has significant clinical impact.n order to maximize clinical effect and minimizeE it seems more attractive to restrict tumor tar-ecific effector cell subsets with known antitumor


Several specific immune effector cell subsets thatcan be considered attractive candidates for tumor tar-geting (i.e. T-cells, invariant natural killer T-cells(iNKT), and natural killer cells (NK)) are discussedbelow.

5.1. T-c

T-celsubset en roa unique plyse stressecytokines aimmune ce[88].

V9V2subset in hactivated angens (pAg)counterparand hydrorespectivel(e.g. zoledcell killingof endogen[89,90]. Re3A1 is requ[91,92].

In stressupregulatedIndeed, Vto recognitiple tumoNHL, prosand qualitaare noted impact dis[93]. Impoare reversicancers trmoted theT-cells towaugmentedgression. Otransient flious maligfollowing nates or mresponses ated lysis enhanced bclinical dacate that exin cancer toxicity.

5.2. Invariant natural killer T-cells (iNKT)

iNKT represent a distinct population of lymphocytescharacteriz

l T-cepolym

rangetor (endent KT crs androle in,101]. isplayers coteracte iNKilitatiderived -ga04,10e tum

with anti-tr grafce [10es evaor wit

report112].

admin-arteriies pronumbelated

summ

rscoreillanc

ncer trtially

Natur

K celhoid cignalsggered

I molectivatiMHCced se

ation, rin- anflammeir romioloells

ls, once regarded an evolutionary redundant T-cellute to extinction, have been demonstrated to hold

osition in the immune system. They can directlyd or infected cells, produce a diversified set ofnd chemokines to regulate both immune and non-lls, and can present antigens for T-cell priming

T-cells constitute the predominant T-celluman peripheral blood. V9V2 T-cells can bed expanded by non-peptidic pyrophosphate anti-

, of which there are both host and microbe-derivedts, typified by isopentenyl pyrophosphate (IPP)xymethyl-but-2-enyl-pyrophosphate (HMBPP),y. Furthermore, aminobisphosphonate compoundsronic acid) sensitize target cells to V9V2 T-

by promoting the intracellular accumulationous IPP by inhibiting mevalonate metabolismcently, it was reported that butyrophilin (BTN)ired for the presentation of pAg to V9V2 T-cells

ed/malignant cells pAg production is frequently allowing discrimination from normal tissue.9V2 T-cells have been shown to be ableze and eliminate malignant cells from mul-rs types, including multiple myeloma (MM),tate-, renal cell- and colon cancer. Quantitativetive defects in the V9V2 T-cell populationin various malignancies [90] and negativelyease-free survival, e.g. in ovarian carcinomartantly, these functional V9V2 T-cell defectsble [90]. In patients with various metastaticeatment with zoledronic acid and IL-2 pro-

differentiation of peripheral blood V9V2ard an effector/memory-like phenotype with

numbers correlating with arrested disease pro-bserved toxicities were minor and limited to

u-like symptoms [94,95]. In patients with var-nancies, adoptive transfer of V9V2 T-cellsex vivo expansion by pAg, aminobisphospho-Abs combined with IL-2, resulted in clinical

[90]. Strikingly, in vitro V9V2 T-cell medi-of hepatic tumor cell lines was significantlyy an anti-EpCAM-anti-CD3 BiTE [96]. Althoughta are scarce, preliminary findings clearly indi-ploiting the natural abilities of V9V2 T-cellsimmunotherapy is feasible and carries low

tionanon-

wideeffecdepe

iNtumootal [100cer dnumbcal inin threhabcyte ligan12 [1at thmentto antumoin mistudiwith have[108cosalintraarterThe corre

Inundesurve

in capoten

5.3.

Nlymping sbe triclassthe a(e.g. induactivperfopro-i

Thepideed by a (semi-)invariant TCR. Unlike conven-lls, iNKT recognize (glyco-)lipids presented byorphic CD1d molecules and rapidly secrete a

of cytokines upon stimulation thereby inducing.g. NK and CTL) cell activation in an IFN-manner [9799].ontribute to immune surveillance in early-stage

chemically induced cancers and play a piv- controlling different forms of cancer in mice

Moreover, iNKT from patients with advanced can- quantitative and qualitative defects and circulatingrrelate with patient survival [100103]. Recipro-ions between DC and iNKT can reverse defectsT population. Indeed, in vitro results indicate

on of iNKT function after stimulation with mono-d DC (moDC) pulsed with the agonistic CD1dlactosylceramide (-GalCer) and exogenous IL-

5]. It was shown that sustained activation of iNKTor site could be induced after systemic treat--GalCer loaded soluble CD1d-molecules fusedumor scFv. Potent tumor inhibition of aggressivets expressing the targeted antigen was observed6,107]. Although less clear than in mice, clinicalluating administration of -GalCer-pulsed moDChout adoptive transfer of ex vivo expanded iNKTed objective tumor regressions in several patients

In patients with recurrent HNSCC, nasal submu-istration of GalCer-pulsed APCs combined with

al infusion of activated iNKT via tumor-feedingduced objective responses in 5 out of 10 patients.r of infiltrating iNKT in extirpated tumor tissue

with clinical outcome [112].ary, extensive preclinical and early clinical data

the important role of iNKT in tumor immuno-e and indicate beneficial effects with low toxicityeatment, therefore redirecting this invariant subset

constitutes a valuable approach.

al killer cells (NK)

ls represent a major subset of innate cytotoxicells tightly regulated by inhibitory and activat-

sensed via cell surface receptors. Activation can by a lack of inhibitory signals delivered by MHCcule engagement missing self via ligation of

ng receptor NKG2D by stress-induced molecules class I chain-related genes (MIC) A and B) lf and by FcRIIIA ligation (i.e. ADCC). UponNK cells become highly cytotoxic (e.g. throughnd granzyme mediated mechanisms) and secrete

atory cytokines (e.g. IFN-) [113,114].le in tumor immunosurveillance is underscored bygical studies correlating disease prognosis with


tumor infiltrating NK cells and diminished cytotoxicity withan increased risk of cancer development [114116]. Addi-tionally, NK cells were shown to contribute to mAb basedcancer treanition and could be stions result[117].

Given thmalignanciin clinical tactivated ausignificant tolerance aof allogenetive. An enpatients wiwith succecal related note, a subcombined Itheir therap

Thoughmodest ancancer typvation stimtumors maanti-tumor structs (e.gan anti-CDtumor destrection anddampened NKG2D-liand NK ceat least ex anti-CEA sNK activat[54]. Redirtargeting tuhighly effe

In conccyte subseV9V2-Tallow for rintrinsic anredirectionconserved treatment.

6. Conclu

Advancof Ab baseof construc

killing by engaged effector cells. Indeed, (pre-)clinical trialshave demonstrated results that are not just an enhancementof those accomplished by conventional mAbs, but represent

ole ne in terovemehalf-lrespecrgetedture

ct canverco

ct rediinical eld w

in the

ict of

ll auth

owled

is woerland

(ZonMty (K

nation

rence

Scott ACancerMellorical revthe res2013;6Holligeof singVaughaet al. Inpeutic to antibChameat the eLindhospeciesImplicaImmunMack Mstruct tumor 70215ZeidlerB, et anew clakilling.tment though FcRIIIA (CD16A) mediated recog-elimination of mAb coated tumor cells. ADCCignificantly enhanced through Fc-tail modifica-ing in increased FcRIIIA affinity, at least in vivo

e role NK cells play in first-line defense againstes, their therapeutic use has been widely exploredrials. Adoptively transferred ex vivo expanded andtologous NK cells proved to be safe, however no

clinical responses were observed. To break self-ssociated with autologous NK, adoptive transferic NK cells has been investigated as an alterna-hanced leukemia clearance rate was reported in

th poor-prognosis acute myeloid leukemia (AML)ssful in vivo NK cell expansion after haploidenti-donor infusion and daily i.v. IL-2 [116,118]. Ofstantial increase in host Tregs was reported afterL-2 and NK cell therapy [119], possibly limitingeutic efficacy.

proven safe, NK cell therapy has thus far inducedd inconsistent clinical responses across variouses and it has been argued that additional acti-uli (e.g. FcRIIIa ligation) and/or homing to

y be required for the induction of more robustresponses [116,118]. As noted, several bsAb con-. Triomab, anti-CD19-anti-CD16 bsDb/TaFv and33-anti-CD16 TaFv) are capable of enhancingruction through effective FcR-mediated redi-

activation of NK cells [10,61,120]. Moreover,NK cell responses through tumor derived solublegands were reversible upon cross-linking of tumorlls by an anti-CD30-anti-CD16A bsAb construct,vivo [115]. Similarly, the fusion protein ULBP2-cFv induced an effective anti-tumor response andion in a syngeneic colon carcinoma mouse modelection of adoptively transferred NK cells by bsAbsmor cells and NK cells may therefore prove to bective.lusion, targeting invariant/conserved lympho-

ts through their highly conserved receptor (e.g.CR, iNKT-TCR, NK-receptors or FcRIIIA)edirection of specific lymphocyte subsets withti-tumor efficacy and can prevent the simultaneous

of immunosuppressive T-cells. Importantly, theirreceptors preclude the need for individualized

ding remarks

es in bsAb engineering have marked a new erad cancer treatment and have resulted in an arrayts shown to be effective in inducing target cell

a whwardimprtheir with be ta

Fuimpacan o

impaon clthis fitrials

Con

A

Ackn

ThNethmentSocieInter

Refe

[1]

[2]

[3]

[4]

[5]

[6]

[7]

[8] w therapeutic repertoire and embody a leap for-ms of therapeutic efficacy. Nevertheless, furthernts of the constructs with respect to increasingife, stability, and tumor penetration as well ast to the specific immune effector cell subset to

are required.research will reveal to what extent bsAbs willcer treatment, whether new formats such as VHHme shortcomings of existing constructs and whatrection of different effector cell subsets will haveoutcome. Undoubtedly, the expanding interest inill result in multiple compounds entering clinical

near future shedding light on these issues.

interest statement

ors have no conflicts of interest to declare.

gments

rk is supported by grant no. 90700309 from Thes Organization for Health Research and Develop-

w), grant VU 2010-4728 from the Dutch CancerWF), and grant 14-0343 from the Association foral Cancer Research (AICR).

s

M, Wolchok JD, Old LJ. Antibody therapy of cancer. Nat Rev 2012;12(Apr (4)):27887.

JD, Brown MP, Irving HR, Zalcberg JR, Dobrovic A. A crit-iew of the role of Fc gamma receptor polymorphisms in

ponse to monoclonal antibodies in cancer. J Hematol Oncol:1.r P, Hudson PJ. Engineered antibody fragments and the rise

le domains. Nat Biotechnol 2005;23(Sep (9)):112636.n AT, Iriyama C, Beers SA, Chan CH, Lim SH, Williams EL,hibitory FcgammaRIIb (CD32b) becomes activated by thera-

mAb in both cis and trans and drives internalization accordingody specificity. Blood 2014;123(Jan (5)):66977.s P, Baty D. Bispecific antibodies for cancer therapy: the lightnd of the tunnel? MAbs 2009;1(Nov (6)):53947.fer H, Mocikat R, Steipe B, Thierfelder S. Preferential-restricted heavy/light chain pairing in rat/mouse quadromas.tions for a single-step purification of bispecific antibodies. Jol 1995;155(Jul (1)):21925.

, Riethmuller G, Kufer P. A small bispecific antibody con-expressed as a functional single-chain molecule with highcell cytotoxicity. Proc Natl Acad Sci USA 1995;92(Jul (15)):.

R, Reisbach G, Wollenberg B, Lang S, Chaubal S, Schmittl. Simultaneous activation of T cells and accessory cells by ass of intact bispecific antibody results in efficient tumor cell

J Immunol 1999;163(Aug (3)):124652.


[9] Zeidler R, Mysliwietz J, Csanady M, Walz A, Ziegler I, Schmitt B,et al. The Fc-region of a new class of intact bispecific antibody medi-ates acphagoc

[10] Ruf P, La trifun

[11] SebastiKanniewith thEpCAM2007;5

[12] Linke Rand fut

[13] BurgesSchmawith adCD3 a(13)):3

[14] Jager Met al. Imascites(anti-E

[15] Heiss Ivanchthe treaa prosp(9)):22

[16] Ott MGR, et acal outmalign

[17] Jager Mantibodhuman(10)):4

[18] Kiewe Phase Imaxom

(10)):3[19] Buhma

et al. Imwith Band do(5)):38

[20] KonterMAbs

[21] LofflerbecherCD19 toxicity(6)):20

[22] Dreier F, et acytotoxa sing(6)):69

[23] HoffmaBargourected Int J C

[24] Haas CSchlereantibod

[25] Baeuercancer

[26] Nagorstorical

[27] Bargou R, Leo E, Zugmaier G, Klinger M, Goebeler M, Knop S,et al. Tumor regression in cancer patients by very low doses of a Tcell-enTopp Mmann Sblinatuin B-lirespon2011;2Topp MKlingesurvivaB-lineaKlingeet al. Iacute lengagi2012;1Topp NeumacompleALL: ASH aash/201TeacheSeif Atreatmerated w51547Choi BSystemsuccess

2013;1FeldmaM, et amediatby CD8SchubeA singlates effMAbs Perisicture of (12)):1KipriyaCochloof bactBiol 20Konterapy. AcBaum VWW, erected T(1)):27Johnsotor celproteinMol BiKipriyader LietherapyBiol 19BuhlerSeemaspecificof pros(1)):43tivation of accessory cells and NK cells and induces directytosis of tumour cells. Br J Cancer 2000;83(Jul (2)):2616.indhofer H. Induction of a long-lasting antitumor immunity byctional bispecific antibody. Blood 2001;98(Oct (8)):252634.an M, Passlick B, Friccius-Quecke H, Jager M, Lindhofer H,ss F, et al. Treatment of non-small cell lung cancer patientse trifunctional monoclonal antibody catumaxomab (anti-

anti-CD3): a phase I study. Cancer Immunol Immunother6(Oct (10)):163744.

, Klein A, Seimetz D. Catumaxomab: clinical developmenture directions. MAbs 2010;2(Mar (2)):12936.

A, Wimberger P, Kumper C, Gorbounova V, Sommer H,lfeldt B, et al. Effective relief of malignant ascites in patientsvanced ovarian cancer by a trifunctional anti-EpCAM anti-ntibody: a phase I/II study. Clin Cancer Res 2007;13(Jul899905.

, Schoberth A, Ruf P, Hess J, Hennig M, Schmalfeldt B,munomonitoring results of a phase II/III study of malignant

patients treated with the trifunctional antibody catumaxomabpCAM anti-CD3). Cancer Res 2012;72(Jan (1)):2432.MM, Murawa P, Koralewski P, Kutarska E, Kolesnik OO,enko VV, et al. The trifunctional antibody catumaxomab fortment of malignant ascites due to epithelial cancer: results ofective randomized phase II/III trial. Int J Cancer 2010;127(Nov0921., Marme F, Moldenhauer G, Lindhofer H, Hennig M, Spannagll. Humoral response to catumaxomab correlates with clini-come: results of the pivotal phase II/III study in patients withant ascites. Int J Cancer 2012;130(May (9)):2195203., Schoberth A, Ruf P, Hess J, Lindhofer H. The trifunctional

y ertumaxomab destroys tumor cells that express low levels of epidermal growth factor receptor 2. Cancer Res 2009;69(May2706.P, Hasmuller S, Kahlert S, Heinrigs M, Rack B, Marme A, et al.

trial of the trifunctional anti-HER2 anti-CD3 antibody ertu-ab in metastatic breast cancer. Clin Cancer Res 2006;12(May08591.nn R, Simoes B, Stanglmaier M, Yang T, Faltin M, Bund D,munotherapy of recurrent B-cell malignancies after allo-SCT

i20 (FBTA05), a trifunctional anti-CD3 anti-CD20 antibodynor lymphocyte infusion. Bone Marrow Transpl 2009;43(Mar397.mann R. Dual targeting strategies with bispecific antibodies.2012;4(Mar (2)):18297.

A, Kufer P, Lutterbuse R, Zettl F, Daniel PT, Schwenken- JM, et al. A recombinant bispecific single-chain antibody,

CD3, induces rapid and high lymphoma-directed cyto- by unstimulated T lymphocytes. Blood 2000;95(Mar

98103.T, Lorenczewski G, Brandl C, Hoffmann P, Syring U, Hanakaml. Extremely potent, rapid and costimulation-independentic T-cell response against lymphoma cells catalyzed by

le-chain bispecific antibody. Int J Cancer 2002;100(Aug07.nn P, Hofmeister R, Brischwein K, Brandl C, Crommer S,

R, et al. Serial killing of tumor cells by cytotoxic T cells redi-with a CD19-/CD3-bispecific single-chain antibody construct.ancer 2005;115(May (1)):98104.

, Krinner E, Brischwein K, Hoffmann P, Lutterbuse R,th B, et al. Mode of cytotoxic action of T cell-engaging BiTEy MT110. Immunobiology 2009;214(6):44153.le PA, Reinhardt C. Bispecific T-cell engaging antibodies fortherapy. Cancer Res 2009;69(Jun (12)):49414.en D, Kufer P, Baeuerle PA, Bargou R. Blinatumomab: a his-perspective. Pharmacol Ther 2012;136(Dec (3)):33442.

[28]

[29]

[30]

[31]

[32]

[33]

[34]

[35]

[36]

[37]

[38]

[39]

[40]

[41]

[42] gaging antibody. Science 2008;321(Aug (5891)):9747.S, Kufer P, Gokbuget N, Goebeler M, Klinger M, Neu-, et al. Targeted therapy with the T-cell-engaging antibody

momab of chemotherapy-refractory minimal residual diseaseneage acute lymphoblastic leukemia patients results in highse rate and prolonged leukemia-free survival. J Clin Oncol9(Jun (18)):24938.

S, Gokbuget N, Zugmaier G, Degenhard E, Goebeler ME,r M, et al. Long-term follow-up of hematologic relapse-freel in a phase 2 study of blinatumomab in patients with MRD inge ALL. Blood 2012;120(Dec (26)):51857.r M, Brandl C, Zugmaier G, Hijazi Y, Bargou RC, Topp MS,mmunopharmacologic response of patients with B-lineageymphoblastic leukemia to continuous infusion of T cell-

ng CD19/CD3-bispecific BiTE antibody blinatumomab. Blood19(Jun (26)):622633.MS, Gokbuget N, Zugmaier G, Viardot A, Stelljes M,nn S, et al. Anti-CD19 BiTE blinatumomab induces highte remission rate in adult patients with relapsed B-precursorupdated results of an ongoing phase II trial. In: 53rdnnual meeting and exposition. 2011. https://ash.confex.com/1/webprogram/Paper36294.html.

y DT, Rheingold SR, Maude SL, Zugmaier G, Barrett DM,E, et al. Cytokine release syndrome after blinatumomabnt related to abnormal macrophage activation and amelio-ith cytokine-directed therapy. Blood 2013;121(Jun (26)):

.D, Kuan CT, Cai M, Archer GE, Mitchell DA, Gedeon PC, et al.ic administration of a bispecific antibody targeting EGFRvIIIfully treats intracerebral glioma. Proc Natl Acad Sci USA10(Jan (1)):2705.nn A, Arndt C, Topfer K, Stamova S, Krone F, Cartellieril. Novel humanized and highly efficient bispecific antibodiese killing of prostate stem cell antigen-expressing tumor cells

+ and CD4+ T cells. J Immunol 2012;189(Sep (6)):324959.rt I, Kellner C, Stein C, Kugler M, Schwenkert M, Saul D, et al.e-chain triplebody with specificity for CD19 and CD33 medi-ective lysis of mixed lineage leukemia cells by dual targeting.2011;3(Jan (1)):2130.

O, Webb PA, Holliger P, Winter G, Williams RL. Crystal struc-a diabody, a bivalent antibody fragment. Structure 1994;2(Dec21726.nov SM, Moldenhauer G, Braunagel M, Reusch U,vius B, Le GF, et al. Effect of domain order on the activityerially produced bispecific single-chain Fv antibodies. J Mol03;330(Jun (1)):99111.mann RE. Recombinant bispecific antibodies for cancer ther-ta Pharmacol Sin 2005;26(Jan (1)):19., Buhler P, Gierschner D, Herchenbach D, Fiala GJ, Schamel

t al. Antitumor activities of PSMAxCD3 diabodies by redi--cell lysis of prostate cancer cells. Immunotherapy 2013;5(Jan

38.n S, Burke S, Huang L, Gorlatov S, Li H, Wang W, et al. Effec-l recruitment with novel Fv-based dual-affinity re-targeting

leads to potent tumor cytolysis and in vivo B-cell depletion. Jol 2010;399(Jun (3)):43649.nov SM, Moldenhauer G, Schuhmacher J. Cochlovius B, Vonth CW, Matys ER, et al. Bispecific tandem diabody for tumor

with improved antigen binding and pharmacokinetics. J Mol99;293(Oct (1)):4156.

P, Wolf P, Gierschner D, Schaber I, Katzenwadel A, Schultze-nn W, et al. A bispecific diabody directed against prostate-

membrane antigen and CD3 induces T-cell mediated lysistate cancer cells. Cancer Immunol Immunother 2008;57(Jan52.


[43] Volkel T, Korn T, Bach M, Muller R, Kontermann RE. Optimizedlinker sequences for the expression of monomeric and dimeric bispe-cific sin

[44] CochloA, Molbined itandem(16)):4

[45] CochloJ, Strauwith a (2)):88

[46] Asano et al. Htor CCancer

[47] Asano M, et mat cothe cas18128

[48] Asano Domaitumor hEx3 d

[49] Moore Applicredirec(17)):4

[50] DhimoSeptem

[51] Strop PSE, et aany ant

[52] Liddy NMonoc(6)):98

[53] Oates Jgeted c

[54] Rothe J, Reinredirectumor (12)):2

[55] BlancoCuestation of2013;5

[56] Donda In vivoclass I cestablis

[57] King BRR, etcognate2013;6

[58] Getts Dwe ove

2010;2[59] Tranter

antibodPharma

[60] DeNardresponsImmun

[61] KipriyaLe GF,

and CD19 CD16 diabodies in a preclinical model of non-Hodgkinslymphoma. J Immunol 2002;169(Jul (1)):13744.Asano R, Kawaguchi H, Watanabe Y, Nakanishi T, Umetsu M,HayashIgG-likto tumoRuf P, new trmelanoKontertherapeStork RtributioderivatRuf P, JPharmatered th(15S)):Jeong Kneering2011;6Olafsenet al. Cconjugtein EnBrischwR, et awith hi2006;4Korn Tbinant CEA-eally ex2004;6HamerHamerlight chRooverNanobo2007;9Conratand evoComp De Gebody r2006;3Oliveirgouwenand theRooverKM, esolid tuHarmsetions oBiotechCortez-Conratdomain(3)):45Tijink Haard factor rtage of(8)):22Holz Jojkovicanti-ILgle-chain diabodies. Protein Eng 2001;14(Oct (10)):81523.vius B, Kipriyanov SM, Stassar MJ, Schuhmacher J, Bennerdenhauer G, et al. Cure of Burkitts lymphoma in severe com-mmunodeficiency mice by T cells, tetravalent CD3 CD19

diabody, and CD28 costimulation. Cancer Res 2000;60(Aug33641.vius B, Kipriyanov SM, Stassar MJ, Christ O, Schuhmacherss G, et al. Treatment of human B cell lymphoma xenograftsCD3 CD19 diabody and T cells. J Immunol 2000;165(Jul895.R, Sone Y, Makabe K, Tsumoto K, Hayashi H, Katayose Y,umanization of the bispecific epidermal growth factor recep-D3 diabody and its efficacy as a potential clinical reagent. Clin

Res 2006;12(Jul (13)):403642.R, Ikoma K, Shimomura I, Taki S, Nakanishi T, Umetsu

al. Cytotoxic enhancement of a bispecific diabody by for-nversion to tandem single-chain variable fragment (taFv):e of the hEx3 diabody. J Biol Chem 2011;286(Jan (3)):.R, Kumagai T, Nagai K, Taki S, Shimomura I, Arai K, et al.n order of a bispecific diabody dramatically enhances its anti-activity beyond structural format conversion: the case of theiabody. Protein Eng Des Sel 2013;26(May (5)):35967.PA, Zhang W, Rainey GJ, Burke S, Li H, Huang L, et al.

ation of dual affinity retargeting molecules to achieve optimalted T-cell killing of B-cell lymphoma. Blood 2011;117(Apr54251.lea E, Reichert JM. World bispecific antibody summit,ber 2728, 2011, Boston, MA. MAbs 2012;4(Jan (1)):413., Ho WH, Boustany LM, Abdiche YN, Lindquist KC, Fariasl. Generating bispecific human IgG1 and IgG2 antibodies fromibody pair. J Mol Biol 2012;420(Jul (3)):20419., Bossi G, Adams KJ, Lissina A, Mahon TM, Hassan NJ, et al.

lonal TCR-redirected tumor cell killing. Nat Med 2012;18(Jun07., Jakobsen BK. ImmTACs. Novel bi-specific agents for tar-ancer therapy. Oncoimmunology 2013;2(Feb (2)):e22891.A, Jachimowicz RD, Borchmann S, Madlener M, Kesslerers KS, et al. The bispecific immunoligand ULBP2-aCEAts natural killer cells to tumor cells and reveals potent anti-activity against colon carcinoma. Int J Cancer 2014;134(Jun82940.-Toribio A, Sainz-Pastor N, Alvarez-Cienfuegos A, Merino N,

AM, Sanchez-Martin D, et al. Generation and characteriza- monospecific and bispecific hexavalent trimerbodies. MAbs(Jan (1)):709.A, Cesson V, Mach JP, Corradin G, Primus FJ, Robert B.

targeting of an anti-tumor antibody coupled to antigenic MHComplexes induces specific growth inhibition and regression ofhed syngeneic tumor grafts. Cancer Immun 2003;Aug (3):11.C, Hamblin AD, Savage PM, Douglas LR, Hansen TH, French

al. Antibody-peptide-MHC fusion conjugates target non- T cells to kill tumour cells. Cancer Immunol Immunother

2(Jun (6)):1093105.R, Getts MT, McCarthy DP, Chastain EM, Miller SD. Haverestimated the benefit of human(ized) antibodies? MAbs

(Nov (6)):68294. E, Peters G, Boyce M, Warrington S. Giving monoclonalies to healthy volunteers in phase 1 trials: is it safe? Br J Clincol 2013;76(Aug (2)):16472.o GL, Bradt BM, Mirick GR, DeNardo S. Human antiglobuline to foreign antibodies: therapeutic benefit? Cancer Immunolother 2003;52(May (5)):30916.nov SM, Cochlovius B, Schafer HJ, Moldenhauer G, Bahre A,

et al. Synergistic antitumor effect of bispecific CD19 CD3

[62]

[63]

[64]

[65]

[66]

[67]

[68]

[69]

[70]

[71]

[72]

[73]

[74]

[75]

[76]

[77]

[78]

[79]

[80] i H, et al. Diabody-based recombinant formats of humanizede bispecific antibody with effective retargeting of lymphocytesr cells. J Immunother 2008;31(Oct (8)):75261.Jager M, Ellwart J, Wosch S, Kusterer E, Lindhofer H. Twoifunctional antibodies for the therapy of human malignantma. Int J Cancer 2004;108(Feb (5)):72532.mann RE. Strategies for extended serum half-life of proteinutics. Curr Opin Biotechnol 2011;22(Dec (6)):86876., Campigna E, Robert B, Muller D, Kontermann RE. Biodis-n of a bispecific single-chain diabody and its half-life extendedives. J Biol Chem 2009;284(Sep (38)):256129.ager M, Volovat C, Burges A, Heiss MM, Wimberger P, et al.cokinetics and in vivo stability of intraperitoneally adminis-erapeutic antibody catumaxomab. J Clin Oncol 2008;26(May14006.

J, Jang SH, Velmurugan N. Recombinant antibodies: engi- and production in yeast and bacterial hosts. Biotechnol J(Jan (1)):1627.

T, Cheung CW, Yazaki PJ, Li L, Sundaresan G, Gambhir SS,ovalent disulfide-linked anti-CEA diabody allows site-specification and radiolabeling for tumor targeting applications. Pro-g Des Sel 2004;17(Jan (1)):217.ein K, Schlereth B, Guller B, Steiger C, Wolf A, Lutterbuese

l. MT110: a novel bispecific single-chain antibody constructgh efficacy in eradicating established tumors. Mol Immunol3(Mar (8)):112943., Nettelbeck DM, Volkel T, Muller R, Kontermann RE. Recom-bispecific antibodies for the targeting of adenoviruses to

xpressing tumour cells: a comparative analysis of bacteri-pressed single-chain diabody and tandem scFv. J Gene Med(Jun (6)):64251.s-Casterman C, Atarhouch T, Muyldermans S, Robinson G,s C, Songa EB, et al. Naturally occurring antibodies devoid ofains. Nature 1993;363(Jun (6428)):4468.s RC, van Dongen GA, van Bergen en Henegouwen PM.dies in therapeutic applications. Curr Opin Mol Ther

(Aug (4)):32735.h KE, Wernery U, Muyldermans S, Nguyen VK. Emergencelution of functional heavy-chain antibodies in Camelidae. Dev

Immunol 2003;27(Feb (2)):87103.nst E, Saerens D, Muyldermans S, Conrath K. Anti-epertoire development in camelids. Dev Comp Immunol0(12):18798.a S, Heukers R, Sornkom J, Kok RJ, van Bergen en Hene-

PM. Targeting tumors with nanobodies for cancer imagingrapy. J Control Release 2013;28(Dec):1723.s RC, Vosjan MJ, Laeremans T, el KR, de Bruin RC, Fergusont al. A biparatopic anti-EGFR nanobody efficiently inhibitsmour growth. Int J Cancer 2011;129(Oct (8)):201324.n MM, De Haard HJ. Properties, production, and applica-

f camelid single-domain antibody fragments. Appl Microbiolnol 2007;77(Nov (1)):1322.Retamozo V, Lauwereys M, Hassanzadeh GG, Gobert M,

h K, Muyldermans S, et al. Efficient tumor targeting by single- antibody fragments of camels. Int J Cancer 2002;98(Mar662.BM, Laeremans T, Budde M, Stigter-van WM, Dreier T, DeHJ, et al. Improved tumor targeting of anti-epidermal growtheceptor nanobodies through albumin binding: taking advan-

modular nanobody technology. Mol Cancer Ther 2008;7(Aug8897.-B, Sargentini-Maier L, De Bruyn S, Gachlyi B, dvaros I,h B, et al. Twenty-four weeks of treatment with a novel-6 receptor Nanobody (ALX-0061) resulted in 84% ACR20


improvement and 58% DAS28 remission in a phase I/II study in RA.Ann Rheum Dis 2013;72(Suppl 3):64.

[81] SunthaBrunneCD28 m(10)):1

[82] Finn O(25)):2

[83] NishikInt J C

[84] Zou WNat Re

[85] Colomtion: th2007;7

[86] PaluckNat Re

[87] KoristkS, et abispeci

[88] Vantougamma(2)):88

[89] SchneiSchepecell analpha.

[90] Braza VgammJ Haem

[91] VavassEl DS,stimula(9)):90

[92] Harly Monkk(BTN3T-cell s

[93] ThedreA quanis a negOne 20

[94] MeraviG, et aldronatebreast c

[95] Dieli FG, et aand intcancer.

[96] Hoh A,The accells an

[97] van deW, vanrestrict(1)):8

[98] GodfreNKT c(3)):23

[99] Godfretion vi(10)):1

[100] BellonNapolinoma i2010;5

[101] Swann JB, Uldrich AP, van DS, Sharkey J, Murray WK, Godfrey DI,et al. Type I natural killer T cells suppress tumors caused by p53 lossin miceExley SP. Desubsets2011;1Schneimans CT-cell carcino2012;3Tahir Set al. Ladvancvan deW, Porkiller Tgalactocells. CStirnemet al. Sa CD1mice. JCorgnaM, et tumor aImmunSchneiJanssenalpha-gcer and(2)):13UchidaS, et al.sentingor recu

2008;5Chang M, et T cellsdendrit15031Nieda Therapsubjectand innYamasH, et acer tissImmunReineret al. Solympho2013;1Vivier Eural kil2012;1Reineret al. Rbispeci(4)):89Cheng apy fo(3)):23ZalevskJR, et ralingam G, Perry MR, Ward S, Brett SJ, Castello-Cortes A,r MD, et al. Cytokine storm in a phase 1 trial of the anti-onoclonal antibody TGN1412. N Engl J Med 2006;355(Sep

01828.J. Cancer immunology. N Engl J Med 2008;358(Jun

70415.awa H, Sakaguchi S, Regulatory. T cells in tumor immunity.ancer 2010;127(Aug (4)):75967., Regulatory. T cells, tumour immunity and immunotherapy.v Immunol 2006;6(Apr (4)):295307.bo MP, Piconese S. Regulatory-T-cell inhibition versus deple-e right choice in cancer immunotherapy. Nat Rev Cancer

(Nov (11)):8807.a K, Banchereau J. Cancer immunotherapy via dendritic cells.v Cancer 2012;12(Apr (4)):26577.a S, Cartellieri M, Theil A, Feldmann A, Arndt C, Stamoval. Retargeting of human regulatory T cells by single-chainfic antibodies. J Immunol 2012;188(Feb (3)):15518.rout P, Hayday A. Six-of-the-best: unique contributions ofdelta T cells to immunology. Nat Rev Immunol 2013;13(Feb100.ders FL, de Bruin RC, Santegoets SJ, Bonneville M, Scotet E,r RJ, et al. Activated iNKT cells promote Vgamma9Vdelta2-Tti-tumor effector functions through the production of TNF-Clin Immunol 2012;142(Feb (2)):194200.

MS, Klein B. Anti-tumour immunotherapy witha9Vdelta2 T-lymphocytes: from the bench to the bedside. Bratol 2013;160(Jan (2)):12332.

ori S, Kumar A, Wan GS, Ramanjaneyulu GS, Cavallari M, et al. Butyrophilin 3A1 binds phosphorylated antigens andtes human gammadelta T cells. Nat Immunol 2013;14(Sep816.C, Guillaume Y, Nedellec S, Peigne CM, Monkkonen H,onen J, et al. Key implication of CD277/butyrophilin-3A) in cellular stress sensing by a major human gammadeltaubset. Blood 2012;120(Sep (11)):226979.z A, Lavoue V, Dessarthe B, Daniel P, Henno S, Jaffre I, et al.titative deficiency in peripheral blood Vgamma9Vdelta2 cellsative prognostic biomarker in ovarian cancer patients. PLoS13;8(5):e63322.glia S, Eberl M, Vermijlen D, Todaro M, Buccheri S, Cicero. In vivo manipulation of Vgamma9Vdelta2 T cells with zole-

and low-dose interleukin-2 for immunotherapy of advancedancer patients. Clin Exp Immunol 2010;161(Aug (2)):2907., Vermijlen D, Fulfaro F, Caccamo N, Meraviglia S, Cicerol. Targeting human {gamma}delta} T cells with zoledronateerleukin-2 for immunotherapy of hormone-refractory prostate

Cancer Res 2007;67(Aug (15)):74507. Dewerth A, Vogt F, Wenz J, Baeuerle PA, Warmann SW, et al.tivity of gammadelta T cells against paediatric liver tumourd spheroids in cell culture. Liver Int 2013;33(Jan (1)):12736.

r Vliet HJ, Molling JW, von Blomberg BM, Nishi N, Kolgen den Eertwegh AJ, et al. The immunoregulatory role of CD1d-ed natural killer T cells in disease. Clin Immunol 2004;112(Jul23.y DI, MacDonald HR, Kronenberg M, Smyth MJ, Van KL.ells: whats in a name? Nat Rev Immunol 2004;4(Mar

17.y DI, Kronenberg M. Going both ways: immune regula-a CD1d-dependent NKT cells. J Clin Invest 2004;114(Nov37988.e M, Ceccon M, Grioni M, Jachetti E, Calcinotto A,tano A, et al. iNKT cells control mouse spontaneous carci-ndependently of tumor-specific cytotoxic T cells. PLoS One(1):e8646.

[102]

[103]

[104]

[105]

[106]

[107]

[108]

[109]

[110]

[111]

[112]

[113]

[114]

[115]

[116]

[117] . Blood 2009;113(Jun (25)):63825.MA, Lynch L, Varghese B, Nowak M, Alatrakchi N, Balkveloping understanding of the roles of CD1d-restricted T cell

in cancer: reversing tumor-induced defects. Clin Immunol40(Aug (2)):18495.ders FL, de Bruin RC, van den Eertwegh AJ, Scheper RJ, Lee-R, Brakenhoff RH, et al. Circulating invariant natural killer

numbers predict outcome in head and neck squamous cellma: updated analysis with 10-year follow-up. J Clin Oncol0(Feb (5)):56770.M, Cheng O, Shaulov A, Koezuka Y, Bubley GJ, Wilson SB,oss of IFN-gamma production by invariant NK T cells in

ed cancer. J Immunol 2001;167(Oct (7)):404650.r Vliet HJ, Molling JW, Nishi N, Masterson AJ, Kolgencelli SA, et al. Polarization of Valpha24+ Vbeta11+ natural

cells of healthy volunteers and cancer patients using alpha-sylceramide-loaded and environmentally instructed dendriticancer Res 2003;63(Jul (14)):41016.ann K, Romero JF, Baldi L, Robert B, Cesson V, Besra GS,ustained activation and tumor targeting of NKT cells using

d-anti-HER2-scFv fusion protein induce antitumor effects in Clin Invest 2008;118(Mar (3)):9941005.c S, Perret R, Derre L, Zhang L, Stirnemann K, Zaudereral. CD1d-antibody fusion proteins target iNKT cells to thend trigger long-term therapeutic responses. Cancer Immunolother 2013;62(Apr (4)):74760.ders FL, Scheper RJ, von Blomberg BM, Woltman AM,

HL, van den Eertwegh AJ, et al. Clinical experience withalactosylceramide (KRN7000) in patients with advanced can-

chronic hepatitis B/C infection. Clin Immunol 2011;140(Aug041.

T, Horiguchi S, Tanaka Y, Yamamoto H, Kunii N, Motohashi Phase I study of alpha-galactosylceramide-pulsed antigen pre-

cells administration to the nasal submucosa in unresectablerrent head and neck cancer. Cancer Immunol Immunother7(Mar (3)):33745.DH, Osman K, Connolly J, Kukreja A, Krasovsky J, Packal. Sustained expansion of NKT cells and antigen-specific

after injection of alpha-galactosyl-ceramide loaded matureic cells in cancer patients. J Exp Med 2005;201(May (9)):7.

M, Okai M, Tazbirkova A, Lin H, Yamaura A, Ide K, et al.eutic activation of Valpha24+Vbeta11+ NKT cells in humans results in highly coordinated secondary activation of acquiredate immunity. Blood 2004;103(Jan (2)):3839.

aki K, Horiguchi S, Kurosaki M, Kunii N, Nagato K, Hanaokal. Induction of NKT cell-specific immune responses in can-ues after NKT cell-targeted adoptive immunotherapy. Clinol 2011;138(Mar (3)):25565.s KS, Topolar D, Henke A, Simhadri VR, Kessler J, Sauer M,luble ligands for NK cell receptors promote evasion of chroniccytic leukemia cells from NK cell anti-tumor activity. Blood

21(May (18)):365865., Ugolini S, Blaise D, Chabannon C, Brossay L. Targeting nat-

ler cells and natural killer T cells in cancer. Nat Rev Immunol2(Apr (4)):23952.s KS, Kessler J, Sauer M, Rothe A, Hansen HP, Reusch U,escue of impaired NK cell activity in hodgkin lymphoma withfic antibodies in vitro and in patients. Mol Ther 2013;21(Apr5903.M, Chen Y, Xiao W, Sun R, Tian Z. NK cell-based immunother-r malignant diseases. Cell Mol Immunol 2013;10(May052.y J, Leung IW, Karki S, Chu SY, Zhukovsky EA, Desjarlais

al. The impact of Fc engineering on an anti-CD19 antibody:


increased Fcgamma receptor affinity enhances B-cell clearing in non-human primates. Blood 2009;113(Apr (16)):373543.

[118] Miller JS. Therapeutic applications: natural killer cells in the clinic.Hematol Am Soc Hematol Educ Program 2013;2013:24753.

[119] Bachanova V, Burns LJ, McKenna DH, Curtsinger J, Panoskaltsis-Mortari A, Lindgren BR, et al. Allogeneic natural killer cells forrefractory lymphoma. Cancer Immunol Immunother 2010;59(Nov(11)):173944.

[120] Wiernik A, Foley B, Zhang B, Verneris MR, Warlick E, GleasonMK, et al. Targeting natural killer cells to acute myeloid leukemiain vitro with a CD16 33 bispecific killer cell engager and ADAM17inhibition. Clin Cancer Res 2013;19(Jul (14)):384455.

Biography

Hans J. van der Vliet is a medical oncologist at theDepartment of Medical Oncology of the VU University Med-ical Center and Cancer Center Amsterdam. His translationalresearch focuses on cancer immunotherapy with a spe-cial emphasis on conserved immunoregulatory and immuneeffector cell subsets.

Bispecific antibody platforms for cancer immunotherapy1 Introduction2 Currently available bispecific antibody platforms2.1 Trifunctional hybrid antibodies (Triomab)2.1.1 Catumaxomab2.1.2 Ertumaxomab2.1.3 FBTA05

2.2 Single chain variable fragment (scFv) based platforms2.2.1 Tandem scFv (TaFv)2.2.1.1 Blinatumomab2.2.1.2 Alternative TaFv constructs

2.2.2 Bispecific diabodies (bsDb)2.2.2.1 Pre-clinical studies

2.3 Other bispecific antibody based platforms

3 Limitations of the type of bsAb constructs3.1 Immunogenicity of bispecific antibody constructs3.2 Size of the construct: Pay off between circulation half-life time and tumor penetration3.3 Antibody construct stability and manufacturing difficulties

4 Perspectives related to the bsAb construct5 Targeting specific lymphocyte subsets to maximize efficacy5.1 T-cells5.2 Invariant natural killer T-cells (iNKT)5.3 Natural killer cells (NK)

6 Concluding remarksConflict of interest statementAcknowledgmentsReferences

Biography

bispecific antibody platforms for cancer immunotherapy.pdf

Documents