APG350 induces superior clustering of TRAIL-Receptors and … · Introduction Page 6 of 26 of FcγR-mediated cross-linking events. In the final layout, two single-chain TRAIL-R binding

Page 1 of 26

APG350 induces superior clustering of TRAIL-Receptors and shows therapeutic anti-tumor efficacy independent of cross-linking via Fcγ-

Receptors

Christian Gieffers1, Michael Kluge1, Christian Merz1 Jaromir Sykora1, Meinolf Thiemann1, René Schaal1, Carmen Fischer1, Marcus Branschädel1;4, Behnaz Ahangarian Abhari2, Peter Hohenberger3, Simone Fulda2, Harald Fricke1 and Oliver Hill1

Running title: Hexavalent TRAIL-Receptor agonists

Keywords: single-chain-TRAIL-receptor-binding-domain, TRAIL, TRAILR1, TRAILR2,

multivalent agonist, DR5

Conflict of interest: C. Gieffers, M. Kluge, C. Merz, J. Sykora, M. Thiemann, R. Schaal, C.

Fischer, M. Branschädel, H. Fricke and O. Hill are employees and/or stockholders of the

Apogenix GmbH, Germany

Grant support: This work was supported by the Federal Ministry of Education and Research

(BMBF); Germany. Grant: BioChancePLUS; Grant number: 0315164 (Title: Liganden der TNF

Rezeptor Superfamilie als innovative Biopharmazeutika). The grant was awarded bei the

Apogenix GmbH as company and therefore C. Gieffers, M. Kluge, C. Merz, J. Sykora, M.

Thiemann, R. Schaal, C. Fischer, M. Branschädel, H. Fricke and O. Hill as employees and/or

stockholders of the Apogenix GmbH have to be stated as recipients of financial support

regarding this publication.

Authors`Affiliations:

1: Apogenix GmbH, Im Neuenheimer Feld 584, 69120 Heidelberg, Germany

2: Institute for Experimenal Cancer Research in Pediatrics, Goethe-University Frankfurt, Komturstraße 3a, 60528 Frankfurt am Main, Germany

3: Division of Surgical Oncology, Mannheim University Medical Center, University of Heidelberg, Theodor-Kutzer-Ufer 1-3, 68167 Mannheim, Germany

4: Present adress: Boehringer Ingelheim Pharma GmbH & Co. KG, Birkendorfer Straße 65, 88400 Biberach an der Riß

on August 18, 2020. © 2013 American Association for Cancer Research. mct.aacrjournals.org Downloaded from

Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on October 7, 2013; DOI: 10.1158/1535-7163.MCT-13-0323

http://mct.aacrjournals.org/

Corresponding Author

Page 2 of 26

Corresponding Author

Oliver Hill, Apogenix GmbH, Im Neuenheimer Feld 584, 69120 Heidelberg, Germany

Phone: +49-6221-58608-18

Fax: +49-6221-58608-10

E-Mail: [email protected]




Abstract

Page 3 of 26

Abstract

Cancer cells can be specifically driven into apoptosis by activating Death-receptor-4 (DR4;

TRAIL-R1) and/or Death-receptor-5 (DR5; TRAIL-R2). Albeit showing promising preclinical

efficacy, first generation protein therapeutics addressing this pathway, especially agonistic

anti-DR4/DR5-monoclonal antibodies, have not been clinically successful to date. Due to

their bivalent binding mode, effective apoptosis induction by agonistic TRAIL-R antibodies is

achieved only upon additional events leading to antibody-multimer formation. The binding

of these multimers to their target subsequently leads to effective receptor-clustering on

cancer cells. The research results presented here report on a new class of TRAIL-receptor

agonists overcoming this intrinsic limitation observed for antibodies in general. The main

feature of these agonists is a TRAIL-mimic consisting of three TRAIL-protomer subsequences

combined in one polypeptide chain, termed single-chain-TRAIL-receptor-binding-domain

(scTRAIL-RBD). In the active compounds, two scTRAIL-RBDs with three receptor binding sites

each are brought molecularly in close proximity resulting in a fusion protein with a

hexavalent binding mode. In the case of APG350 - the prototype of this engineering concept

- this is achieved by fusing the Fc-part of a human IgG1-mutein C-terminally to the scTRAIL-

RBD-polypeptide, thereby creating six receptor binding sites per drug-molecule. In vitro,

APG350 is a potent inducer of apoptosis on human tumor cell-lines and primary tumor cells.

In vivo, treatment of mice bearing Colo205-xenograft tumors with APG350 showed a dose

dependent anti-tumor efficacy. By dedicated muteins we confirmed, that the observed in

vivo-efficacy of the hexavalent scTRAIL-RBD fusion proteins is - in contrast to agonistic

antibodies - independent of FcγRs based cross-linking events.

.




Introduction

Page 4 of 26

Introduction

Tumor Necrosis Factor (TNF)-Related Apoptosis Inducing Ligand (TRAIL; Apo2L) is a member

of the TNF Superfamily (SF) of cytokines (1, 2). TRAIL induces apoptosis through binding to

two closely related cell surface receptors, TRAIL-R1 (DR4) and TRAIL-R2 (DR5). Binding of

membrane bound TRAIL induces receptor oligomerization and subsequent formation of the

death inducing signaling complex (DISC) by recruitment of FADD and caspase-8 to the

intracellular TRAIL receptor chains. Auto-activation of caspase-8 within the DISC then causes

activation of executioner caspases-3 and -7, which exerts pro-apoptotic functions by the

cleavage of intracellular target proteins ultimately triggering apoptotic cell death (3, 4).

TRAIL is known to selectively induce apoptosis in cancer cells, while normal cells are

relatively resistant to TRAIL-induced apoptosis. Based on this unique activity profile,

Apo2L/TRAIL variants (5, 6) and numerous agonistic TRAIL-R antibodies towards DR4 and/or

DR5 were developed and evaluated for tumor treatment (3). Preclinical data indicate that

both classes of agonistic molecules possess significant anti-tumor efficacy without showing

toxicity (7, 8). Treatment of mice bearing xenografted human tumors derived from

colorectal, lung, pancreatic or breast cancer with recombinant human (rh)Apo2L/TRAIL

induced tumor cell apoptosis, suppressed tumor progression, and improved survival (5, 6, 8,

9). Agonistic anti-DR4 and -DR5 antibodies exerted potent anti-tumor activity as a single

agent or in combination with chemotherapy in human tumor xenograft models in mice,

including those based on human renal, colorectal, non-small cell lung and pancreatic cancer

cell lines (7, 10, 11).

There is a sound scientific rationale to use TRAIL-R agonists for the treatment of cancer.

Based on promising preclinical data, recombinant TRAIL and specific agonistic antibodies

recognizing either DR4 or DR5, respectively, were investigated in clinical studies on a diverse

subset of tumor entities. These clinical studies demonstrated in line with preclinical results

that the therapeutic concept of apoptosis induction via DR4 and/or DR5 is generally well

tolerated. Whereas anti-tumor activity was observed for individual patients, the overall

response rates were disappointing and could not confirm the promising preclinical results for

the different TRAIL-R agonists (12-21). Despite potent anti-tumor efficacy of all TRAIL-R

agonists in xenograft tumor models derived from various human cancer cell lines during




Introduction

Page 5 of 26

preclinical development, translation of the preclinical anti-tumor efficacy into the clinical

setting has not yet been successful.

Clustering of receptor chains upon ligand binding is a general feature of the TNF-SF for

executing biological functions. For TRAIL-R agonists efficient clustering via the respective

receptors DR4 and/or DR5 is an essential prerequisite for apoptosis induction. However, for

most of the agonistic DR4 and DR5 antibodies, robust apoptosis induction in vitro is achieved

only upon additional cross-linking via the Fc-part of the antibodies, suggesting that

monomeric antibodies are not capable to cluster the critical number of receptors in order to

trigger apoptosis (7, 11, 22, 23). Although cross-linking is important to achieve sufficient

potency in vitro, agonistic antibodies show excellent anti-tumor activity in vivo on human

tumor xenografts. It was therefore assumed that the critical level of receptor clustering

required for effective apoptosis induction by agonistic DR4 and DR5 antibodies in vivo is

mediated by additional mechanisms. In fact, a recent publication by Wilson et al. (2011) (24)

underscored the dependence of anti-tumor efficacy of an agonistic DR5 antibody

(Drozitumab) on FcγR mediated clustering. In vitro, Drozitumab-induced caspase-3 activation

and cell death in diverse cancer cell lines were augmented by cross-linking with an anti-

human Fc-specific F(ab’)2 reagent. In vivo, Drozitumab (with wild-type Fc-part) displayed

potent anti-tumor activity in the absence of any exogenous cross-linking reagent. However, a

Drozitumab variant encoding a mutant Fc-part that impedes FcγR interaction was essentially

inactive (24). Based on the data set obtained with DR5-specific Drozitumab and the variant

antibody it was concluded, that agonistic TRAIL-R antibodies in general require further

clustering via their Fc functionality by leukocyte FcγRs to show full anti-tumor efficacy in

vivo. The published data insinuates a reason, why agonistic DR4/DR5 antibodies failed in

clinical trials to date. In cancer patients, TRAIL-R antibodies have to compete with

endogenous IgG for FcγR interaction at physiological concentrations, which marks a

fundamental difference to preclinical immune-deficient mouse xenograft models generally

having very low or undetectable levels of immunglobulins in serum. If FcγR mediated cross-

linking of agonistic DR4/DR5 antibodies in cancer patients is required for tumor-specific

apoptosis induction, these events are probably rare finally leading to a limited anti-tumor

efficacy during therapy.

Our intention was to design an agonistic fusion protein which resembles the natural

DR4/DR5 interaction sites of TRAIL and whose in vivo potency on tumor cells is independent




Introduction

Page 6 of 26

of FcγR-mediated cross-linking events. In the final layout, two single-chain TRAIL-R binding

domains (each comprising three TRAIL-R binding sites) are dimerized by a Fc-part of human

IgG1 with lack of / reduced FcγR binding capability.

Therefore, APG350, which is the prototype of this newly designed family of TRAIL-R agonists,

simultaneously binds up to six TRAIL-Receptors, and its molecularly improved ability to form

TRAIL-R clusters on target cells causes efficient apoptosis induction. In advantage to

agonistic TRAIL-R antibodies, induction of apoptosis by APG350 (and other hexavalent

scTRAIL-RBD based agonists) on tumor cells in vitro does not require additional cross-linking

of the molecule. We furthermore present evidence that the APG350 based anti-tumor

activity on human xenograft tumors in mice does not require FcγR mediated cross-linking in

vivo. APG350 is a pro-apoptotic fully human TRAIL-R agonist that has the capacity to bridge

the gap between preclinical and clinical anti-tumor efficacy observed for the current clinical

development candidates in this pathway.




Material and Methods

Page 7 of 26


Construction of APG350

To engineer an Apo2L/TRAIL mimetic we combined three Apo2L/TRAIL protomer

subsequences in one polypeptide chain. This trivalent single-chain-TRAIL-receptor-binding-

domain (scTRAIL-RBD; for engineering details see supplement and Figure S1) was fused to

the Fc-part of a human IgG1-mutein to create a hexavalent scTRAIL-RBD dimer. APG350

constitutes the prototype of this engineering concept. The APG350 surrogate molecules

APG808 and APG780 were specifically designed to exclude binding to Fcγ receptors. APG802

contains wt-Fc part and was used as a positive control for Fcγ receptor binding. They share

the scTRAIL-RBD, but have a modified Fc-domain (APG808) or are dimerized by a different

functional motif (APG780). An overview on structural and functional properties of all TRAIL-R

agonists used is shown in the Supplemental Table S1.

Expression and purification APG350

For the expression of APG350, a synthetic DNA cassette was inserted into an eukaryotic

expression vector that was subsequently introduced to suspension adapted Chinese Hamster

Ovary cells (CHO-S, Invitrogen). APG350 expression was analysed by ELISA and the best

expressing cell pools were subsequently expanded for protein production. For production,

cells were cultured for 7-9 days in chemically defined medium (PowerCHO2-CD, Lonza) at

37°C and 7% CO2 atmosphere in an orbital shaker incubator using 250 ml Erlenmeyer shake

flasks (Corning). Cells were harvested when the cell viability dropped below 70% and the

supernatant was clarified by centrifugation and filtration prior to purification. The APG350

expression levels measured in cell-culture supernatants from CHO-S pools at lab scale

reached concentrations of above 250 mg/l. APG350 was purified by a two-step purification

process combining a Streptactin-affinity purification followed by a preparative SEC, both

performed under physiological buffer conditions in PBS, pH7.4.

Apo2L/TRAIL

Human recombinant Apo2L/TRAIL (aa114 – aa281) produced in E. coli was purchased from

PeproTech (catalogue number 310 04). Based on the information provided by the vendor,





Page 8 of 26

the amino acid sequence of this Apo2L/TRAIL version is identical to Dulanermin, a human

recombinant Apo2L/TRAIL-variant that has been tested in clinical trials for the treatment of

cancer. We analysed the commercially available Apo2L/TRAIL-variant with regard to identity,

integrity, purity, and biological activty and confirmed that all results were in accordance with

those reported for Dulanermin (6, 8) (data not shown).

Anti-DR5 antibody

An agonistic anti-DR5 antibody (clone 71903, catalogue number MAB631) was purchased

from R&D Systems. Fractionation of the “bulk” antibody preparation was done on a

Superdex 200 10/300 GL column (GE Healthcare) equilibrated with PBS, pH 7.4. Separated

antibody fractions were collected and analysed by SDS-PAGE/Silver staining (see Figure S2).

Cell lines and primary tumor cells

WIDR-colon carcinoma, T98G-glioblastoma and MDA-MB231-breast cancer cells were

purchased from the American Type Culture Collection (ATCC). A549-lung carcinoma cells

were purchased from the German Resource Center for Biological Material. Colo205-colon

carcinoma cells were purchased from CLS (Cell Lines Service, Heidelberg, Germany). Huh-7

hepatocyte derived cellular carcinoma cells were a gift from Prof. Nüssler, BGU Tübingen,

Germany. Cell lines were passaged for less than 4 months and maintained in either

Dulbecco’s Modified Eagle’s Media (DMEM) or RPMI-1640 media supplemented with 10%

FBS and 1% nonessential amino acids at 37°C in 5% CO2. No further authentication was done

on these lines. All cell lines were tested negative for mycoplasma contamination. Primary

tumor cells were isolated (see supplement for details) from fresh surgical tumor samples

obtained from P. Hohenberger, MD (Div. of Surgical Oncology, Mannheim University Medical

Center, University of Heidelberg) with informed consent and approval by the local Ethics

Committee.

In vitro assays of cytotoxicity and apoptosis detection

Cytotoxicity assay

The biological activity of TRAIL-R agonists was evaluated by using the One Solution Cell

Proliferation Assay (MTS, Promega). Incubation of respective cells was done for 24 hours





Page 9 of 26

with varying concentrations of TRAIL-R agonists in a 96-well plate format. Cell viability was

assessed by measuring the MTS absorption at 492 nm. In some experiments, cross-linking of

anti-DR5 antibody was performed at room temperature for 20 min by co-incubation with an

anti-mouse polyclonal antibody (anti-mouse IgG1 from goat; Novus Biologicals) prior to

incubation with the cells. Cross-linking of APG350 was done using a mouse-anti-human IgG

heavy chain antibody (ab9243, Abcam).

Western Blot analysis of apoptosis induction in cell lysates

For the analysis of apoptosis induction, T98G cells were incubated with different TRAIL-R

agonists. After 3, 6, 9, 12 and 24 hours of incubation, cells were lysed in 50 mM TrisHCl, 1%

(v/v) Triton-X 100, 150 mM NaCl, protease inhibitor cocktail (Roche, Mannheim, Germany).

Western blot analysis was performed as described previously (25) using the following

antibodies: mouse anti-caspase-8 (Alexis Biochemicals, Grünberg, Germany), rabbit anti-Bid,

rabbit anti-caspase-3, mouse anti-caspase-9 (Cell Signaling, Beverly, MA), rabbit anti-DR5

(Chemicon, Billerica, MA); mouse anti-β-actin (Sigma) was used for loading control. Goat

anti-mouse IgG, goat anti-rabbit IgG conjugated to horseradish peroxidase (Santa Cruz

Biotechnology) and goat anti-mouse IgG1 or goat anti-mouse IgG2b (Southern Biotech,

Birmingham, AL) conjugated to horseradish peroxidase were used as secondary antibodies.

Enhanced chemiluminescence was used for detection (Amersham Bioscience, Freiburg,

Germany).

Binding of TRAIL receptor agonists to Fcγ Receptors

The assessment of binding to FcγRs (CD16, CD32a/b and CD64) was performed employing

kits from CisBio (catalogue no. 62C16PAG) according to the manufacturer´s instructions. In

brief, FcγRs and gamma chain are co-expressed in HEK293-cells and fused with the SNAP-tag.

Cells are labeled with the SNAP-Tb substrate, frozen, and thawed just before the

experiment. IgG labeled with the d2 acceptor was used as the second assay partner. The

binding of the acceptor conjugate to the Lumi4®-Tb cryptate pre-labeled cells generates a

specific FRET signal, which can subsequently be quenched by displacement through the

unlabeled molecules (APG350, APG780, APG802 and APG808). A reduction of the starting

signal indicates binding of the tested compounds to the respective FcγR. The HTRF signals

were then read on the HTRF compatible reader Tecan Infinite F500.





Page 10 of 26

In vivo xenograft models

Female NMRI athymic (nu+/nu+) mice (Janvier, Le Genest St. Isle, France) were inoculated

subcutaneously with 2x106 human Colo205 colon cancer cells per animal. Tumor size was

measured in two dimensions with a caliper (in mm) and the volume was calculated by the

formula (a*b2)/2 [mm3] where “a” is the length and “b” the width of the tumor dimension.

When the subcutaneous tumors had reached an appropriate volume, the animals were

distributed between the different treatment groups. The tumor volume was determined at

the start of treatment (at group distribution) and then 3 times per week throughout the

study. For details about animal care see Supplementary statement. All test compounds were

dissolved in PBS. With exception of the dose-response experiment (see Figure 4A) a fixed

dose level (1 mg/kg bw) of APG350, Apo2L/TRAIL or murine anti-human DR5 antibody was

used for a direct comparison of their anti-tumor efficacy. Apo2L/TRAIL was injected i.v. on 5

consecutive days and the anti-DR5 antibody was injected only once i.p. due to the long half-

life of murine antibodies in mice. APG350 and the surrogate molecules (APG808 and

APG780) were injected i.v. on 5 consecutive days at the same dose level (1 mg/kg bw).

Immunohistochemistry of xenograft tumors

Mice bearing Colo205-cell derived xenograft tumors were treated with a single i.v. dose of

TRAIL-R agonists. For immunohistochemical (IHC) analysis of the xenograft tumors, the mice

were killed 4 hours after treatment and tumors were excised, fixed in 4% Paraformaldehyde

(PFA) in PBS. The PFA-fixed tumors were dehydrated and embedded in paraffin.

Subsequently, 3 µm tumor sections were prepared and placed on a microscope slide. Prior

to incubation with primary antibody, paraffin sections were dewaxed in xylene and

rehydrated in ethanol/water. For antigen retrieval, the tumor sections were treated at 99°C

for 25 min in citrate buffer (Target Retrieval Solution, pH 6.0, DAKO). The following primary

antibodies were used for analysis of apoptotic signaling: rabbit anti-cleaved caspase 8 (Cell

signaling technology), rabbit anti-cleaved caspase 3 (BD-Bioscience) and rabbit anti-cleaved

PARP (Abcam). The tumor sections were incubated with primary rabbit antibodies diluted

1:100 in blocking buffer (PBS + 20 mg/ml BSA + 1 mg/ml human IgG) for 60 min at room

temperature. After a PBS washing step, specific binding of the primary antibody was

visualized using an anti-rabbit biotinylated secondary antibody (Southern Biotech) and

streptavidin alkaline phosphatase (BioGenex). The FAST-Red substrate system (DAKO) was





Page 11 of 26

used as substrate for the alkaline phosphatase, which produced a red precipitate at antibody

binding sites. Sections were then counterstained with Mayer´s hematoxylin and mounted

with glycerin-gelatin. A rabbit isotype control antibody was used for control staining.




Results and Discussion

Page 12 of 26


Structure, design and characterization of APG350

APG350 is a newly designed fully human TRAIL-R agonist. It comprises two single-chain

TRAIL-R-binding-domains (scTRAIL-RBD) each consisting of three covalently linked TRAIL

protomer subsequences (see Figure S1 for construction details). Like its endogenous

counterpart, the scTRAIL-RBD forms three binding sites for receptor interaction and induces

apoptosis efficiently only upon multimerisation (Figure S3). Dimerization of two scTRAIL-

RBDs via an engineered Fc-part of a human IgG1-mutein creates as a functional unit a

molecule with six TRAIL-R binding sites with lack of FcγR binding, respectively. A schematic

representation of APG350 which is the prototype of this engineering concept, is shown in

Figure 1A. The dimeric molecule has a theoretical MW of about 170 kDa and exhibits an

overall antibody-like shape and volume.

For the manufacturing of APG350, a robust, high yield production process has been

established that is based on expression in CHO cells followed by two chromatographic

purification steps. This process results in a final APG350 product that shows excellent

solubility in physiological buffers (e.g. PBS, pH 7.4), which is devoid of aggregated protein

species and lacks contaminating proteins. Analysis of different purified APG350 lots revealed

a single peak as demonstrated by SEC (Figure 1B) and a single band as shown by SDS-PAGE

under non-reducing and reducing conditions (Figure 1C). Different APG350 lots show

reproducible, high cellular cytotoxicity on Colo205 cells (Figure 1D) in vitro with an EC50 of

about 10 ng/ml (~60 pM). APG350 binds to all four human TRAIL-Rs and to the mouse TRAIL-

R with high affinity in the low nM range and is comparable to the affinity profile of

Apo2L/TRAIL (Supplemental Table S2). Direct comparison of the PK for APG350 and

Apo2L/TRAIL in mice revealed a superior terminal half life of 28h for APG350 in comparison

to 1h determined for Apo2L/TRAIL (Supplemental Table S1). In summary, the analysis

revealed that dimerization of the scTRAIL-RBD by the Fc-part of human IgG1 (like APG350)

results in a fully human molecule with favorable functional features for clinical development.





Page 13 of 26

In vitro properties of APG350 compared to anti-DR5 antibody

To compare the biological activity of APG350 to an agonistic anti-DR5 antibody, we have

chosen a commercially available anti-DR5 antibody that shows high apoptosis-inducing

activity and binds with sub-nanomolar affinity to human DR5 (KD = 0.7 nM, data not shown).

This anti-DR5 antibody has similar functional features as described for agonistic anti-DR5

antibodies under clinical development (24) and was used as a surrogate molecule for them in

our investigation.

It is published that for most agonistic TRAIL-R antibodies their capacity to induce apoptosis

in vitro is significantly enhanced by cross-linking agents (7, 11, 22, 23). When we assessed in

vitro cytotoxicity of the “bulk” anti-DR5 antibody preparation (as supplied by the vendor) on

Colo205 cells, we noticed that the addition of a cross-linking antibody leads to a more than

10-fold increase in potency as the EC50 value drops from 71 ng/ml to 6 ng/ml (Figure 2A).

However, the cytotoxicity of APG350 could only be increased marginally upon addition of a

cross-linking antibody from an EC50 value of 10 ng/ml to 5 ng/ml (Figure 2B).

Therapeutic antibody-preparations are well known to contain small amounts of process

related aggregates consisting in part of antibody-multimers with preserved native IgG-

structure (26-28). It is therefore possible that the in vitro cytotoxicity observed for the anti-

DR5 antibody on Colo205 cells is to some extent an effect of functional, multivalent anti-DR5

oligomers present in the “bulk” antibody preparation. To prove this hypotheses we

fractionated the “bulk” anti-DR5 antibody by SEC and obtained two minor high molecular

weight fractions (HMW) and a major fraction of monomeric antibody (95.7% of input)

(Figure S2). Fraction “HMW-1” (3.7% of input), based on its calculated apparent MW most

likely contains antibody dimers whereas fraction “HMW-2” (0.6% of input) corresponds to

larger antibody multimers (Figure S2). Direct comparison of the cellular in vitro cytotoxicity

of these antibody fractions on Colo205 cells revealed that the purified monomeric anti-DR5

antibody showed no relevant cytotoxic activity (EC50 >3000 ng/ml, see Figure 2C). However,

the dimeric anti-DR5 fraction “HMW-1” showed a strong induction of cytotoxicity with an

EC50 of 3.3 ng/ml and fraction “HMW 2” (larger multimers) had an even higher activity with

an EC50 below 0.2 ng/ml on Colo205 cells (Figure 2C). Consistently, cross-linking increased

apoptosis induction of the isolated monomeric DR5 antibody more than 500-fold (EC50

>3000 ng/ml to 6 ng/ml; Figure 2D).





Page 14 of 26

These results indicate that a monomeric preparation of anti-DR5 antibody that binds two

receptor chains at the same time is not capable to induce DR5-mediated apoptosis on

Colo205 cells. Our observation is in accordance with published data on agonistic DR4/DR5

specific antibodies (7, 11, 22, 23) and confirms that cross-linking strongly potentiates the

induction of apoptosis. Furthermore, our data suggest that in vitro cytotoxic activity seen in

bulk antibody preparations of agonistic TRAIL-R antibodies are probably due to intrinsic

portions of functional antibody multimers. In contrast to the anti-DR5 antibody, APG350

requires no cross-linking for efficient apoptosis induction. Furthermore, we could observe

that small fractions of multimeric APG350 that are removed during the purification process

are slightly less active with regard to apoptosis induction than the purified monomeric

molecule (data not shown).

To confirm the differences of APG350 and the anti-DR5 antibody with respect to receptor

clustering and apoptosis induction, we analysed the kinetics of apoptosis induction in T98G

glioblastoma cells. For this purpose, T98G cells were incubated with 0.1 nM of APG350, anti-

DR5 or cross-linked anti-DR5, respectively. Cell lysates prepared after 3, 6, 9, 12 and 24h

were analysed for activation of caspases and BID by western blot analysis (Figure 2E).

APG350 shows fast and effective apoptosis induction on T98G cells. Already 3h after

incubation with APG350, strong activation of caspase 8 (Figure 2E, signals at 43/41 and 18

kDa), caspase 3 (Figure 2E, signals at 19/17/12 kDa) and caspase 9 (Figure 2E, signals at

37/35 kDa) were detectable, indicating activation of extrinsic and intrinsic apoptosis

signaling. Upon APG350 incubation, caspase activation proceeds until depletion of the

intrinsic pro-caspase pools was observed. The activation of caspase 8 was confirmed by

cleavage of the pro-apoptotic caspase 8 substrate BID (29) (Figure 2E, signals at 15 kDa). In

contrast to APG350, the monomeric anti-DR5 antibody alone shows no activation of

caspases and no apoptosis induction in T98G cells within the 24 h observation period. Upon

cross-linking of the anti-DR5 antibody an effective caspase activation comparable to APG350

was observed, however, depletion of internal pro-caspase pools was not detectable (Figure

2E). In summary, APG350 induces a fast apoptotic signaling in T98G cells, whereas DR5-

antibody strongly depends on additional cross-linking for apoptosis induction. These results

were further confirmed by immunoprecipitation of the DISC upon TRAIL-R activation.

Formation of the DISC is shown upon incubation of T98G-cells with anti-DR5 antibody only in

the presence of a cross-linking antibody. For APG350, DISC formation is demonstrated





Page 15 of 26

independently of cross-linking agents (Supplemental Figure S4). The enhanced DISC-forming

capacitiy of our hexavalent TRAIL-mimetic is in line with the observations of Kim et al. (2004)

who analysed differently refolded TRAIL(114-281)-preparations, thereby discovering that

TRAIL(114-281)-hexa- and nonomeres were far more potent than the trimeric TRAIL(114-

281) (30). Similar observations regarding oligomeric state and efficient DISC formation were

made for recombinant hexameric forms of the CD95-ligand (31).

With respect to apoptosis induction, the fundamental difference of APG350 compared to

agonistic TRAIL-R antibodies is the enhanced clustering efficiency of TRAIL-Rs on the cell

surface of the target cells. APG350 binds up to six TRAIL-Rs at the same time and thus

clusters a critical number of TRAIL-Rs required for efficient intracellular DISC assembly,

caspase activation and induction of apoptotic cell death. Agonistic monomeric TRAIL-R

antibodies capable of clustering only two TRAIL-Rs trigger a signal that is below the critical

threshold required for DISC assembly. Only upon cross-linking or multimerization of the

antibodies a critical number of TRAIL-Rs can be clustered in order to create a signal that is

sufficient for DISC assembly and apoptosis induction.

APG350 shows superior anti-tumor activity in vitro

In order to obtain an overall impression of the anti-tumor activity of APG350 in comparison

to Apo2L/TRAIL and anti-DR5 antibodies, we compared the in vitro cytotoxicity in a variety of

established human tumor cell lines and primary human tumor cells. For each cell type

examined, the biological activities of equimolar dose ranges of APG350, bulk anti-DR5

antibody (as supplied commercially) and Apo2L/TRAIL were compared. Bulk anti-DR5

antibody was used in these studies to allow comparison with previous studies and literature

reports. APG350 showed superior induction of apoptosis (decreased viability) on established

human tumor cells (Figure 3A) and primary human tumor cells (Figure 3B) in comparison to

recombinant Apo2L/TRAIL and the anti DR5 agonistic antibody. This activity was observed

for a wide subset of human tumor types and indicates a potential therapeutic benefit for

APG350 in many different tumor entities.





Page 16 of 26

In vivo studies: investigation of anti-tumor efficacy in a tumor

xenograft model

Next, we tested the in vivo anti-tumor efficacy of APG350 using the Colo205 xenograft

model. Prior to xenograft experiments in mice, the pharmacokinetics and the toxicological

profile in mice and monkeys was analysed for APG350. Pharmacokinetic data for APG350

were determined in mice and a Cynomolgus monkey after single dose administration and

compared to Apo2L/TRAIL. APG350 showed an extended half-life in mice compared to

recombinant Apo2L/TRAIL. The terminal half-life of APG350 in mice was calculated to be 28

hours compared to 1 hour for Apo2L/TRAIL. Similar PK data for APG350 was also confirmed

in a Cynomolgus monkey (data not shown). Toxicology studies conducted in mice and

Cynomolgus monkeys have not shown unwanted side effects. In particular, evidence of liver

toxicity was not observed and the overall results of these studies indicated a good

tolerability for APG350 (data not shown).

Anti-tumor efficacy of APG350

The anti-tumor efficacy of APG350 was first tested with a dose-response experiment

employing Colo205 xenograft tumor-bearing mice. When tumors were established, vehicle

(PBS) or different doses of APG350 (0.3, 1 and 3 mg/kg bw) were administered once daily for

5 consecutive days and tumor sizes were analysed. One treatment cycle with APG350

resulted in a highly efficient, dose-dependent reduction of tumor size (Figure 4A). Tumor

regression below the initial tumor volumes was observed for all animals even in the low dose

group (0.3 mg/kg bw), albeit no cases of complete response were observed. At the higher

doses (1 and 3 mg/kg bw) the tumor volumes decreased continually and in both groups all

animals showed complete responses (Figure 4A). In addition, the highest dose of APG350 (3

mg/kg bw) was also applied to animals with large Colo205 tumor xenografts (average tumor

volume ca. 650 mm³). Even these large tumors could be treated successfully with APG350

and complete responses were achieved in 6 out of 7 animals (Supplemental Figure S5).

During the period of the study, APG350 was generally well tolerated, thereby, confirming the

data of the toxicology studies. Median body weights in the control group decreased

continually during the study, whilst no such tumor cachexia was observed in the APG350-

treated groups (data not shown).





Page 17 of 26

Anti-tumor efficacy of APG350 compared to anti-DR5 antibody and Apo2L/TRAIL Next, we compared the anti-tumor efficacy of APG350 to anti-DR5 antibody and

Apo2L/TRAIL in NMRI/nude mice with established Colo205 tumors. APG350 or Apo2L/TRAIL

was administered, each at a dose of 1 mg/kg bw on 5 consecutive days. Monomeric anti-DR5

antibody (see Supplemental Figure S2) was injected at a single dose of 1mg/kg bw, due to

the long physiological half-life (7days) of this murine antibody in mice. APG350 showed

superior efficacy in the Colo205 xenograft model compared to Apo2L/TRAIL and monomeric

anti-DR5 antibody at the same dose level of 1 mg/kg bw (Figure 4B). Apo2L/TRAIL, although

clearly active on Colo205 cells in vitro (see Figure 3A), had no anti-tumor activity at a dose of

1 mg/kg bw and the tumor growth curve was indistinguishable from the control group. The

complete absence of an anti-tumor response for Apo2L/TRAIL is most probably due to the

combination of a low dose (1mg/kg bw) and the very short terminal half life of about 60 min

in mice (8). Observed in all animals, the monomeric anti-DR5 antibody led to moderate

tumor regression, however, anti-tumoral action was somewhat delayed and no animal

became tumor-free. In the APG350 group, the tumor volume decreased continually, all

animals were tumor free by day 30 and even without further treatment did not relapse

during the follow-up period.

Apoptotic signaling was analyzed in Colo205 cell-derived xenograft tumors by

immunohistochemistry (IHC). Colo205 tumors treated with a single dose of APG350,

Apo2L/TRAIL or monomeric anti-DR5 antibody (each at 1mg/kg bw) were excised 4 hours

after treatment and analyzed by IHC for markers of apoptosis induction (i.e. active caspases-

3 and -8 and cleaved PARP). Already 4h after the injection of a single dose of APG350

extensive apoptosis induction was detectable, indicating fast tumor penetration for APG350

combined with a potent anti-tumor efficacy (Figure 4C). Apo2L/TRAIL and the monomeric

anti-DR5 antibody showed only moderate apoptosis induction (Figure 4C) indicating an

insufficient anti-tumor efficacy, in line with the anti-tumor efficacy shown in Figure 4B.

Hexavalent scTRAIL-RBD fusion proteins do not require cross-linking through

binding to Fcγ-receptors for anti-tumor efficacy in vivo

The role of Fcγ-receptor (FcγR)-positive immune cells on TRAIL-R clustering and anti-tumor

activity in vivo was reported for anti-DR5 specific Drozitumab (24). To study these FcγR-

mediated effects, we conducted a head-to-head comparison on Colo205 xenograft tumors





Page 18 of 26

with specifically designed dimerized scTRAIL-RBD derivates that differ in their respective

binding potential to FcγRs:

APG350 contains a modified IgG1 Fc-part which is reported to exert reduced FcγR binding

(32). APG808 lacks the N-linked glycosylation in the CH2 domain of the IgG1 Fc-part which is

required for high affinity FcγR binding (33). APG780 is dimerized via a novel scaffold that

cannot bind to FcγRs. APG802 contains the wild-type Fc part of human IgG1 and serves as a

positive control regarding FcγR binding. The interaction analysis of these scTRAIL-RBD

constructs with human FcγRs CD16, CD32a/b and CD64 is shown in Figure 5A. For APG808

(de-glyco) and APG780 no interaction could be observed to any of the FcγRs. APG350 did not

bind to CD16 and CD32b and showed only a weak interaction to CD32a and to CD64 at high

concentrations. A binding to all analysed FcγRs could be observed with the positive control

APG802 (Figure 5A). Explicit interaction analysis with the respective murine FcγRs was not

performed. However, based on a recent publication we can suppose that the Fc-part of

human IgG1 exhibits a comparable binding to murine FcγRs and mediates comparable

effects on murine and human effector cells (34).

To analyse if different FcγR binding properties of these scTRAIL-RBD constructs affect the

anti-tumor efficacy, NMRI/nude mice with established Colo205 tumors were administered

once daily for 5 consecutive days with APG350, APG808 and APG780 (each at 1 mg/kg bw).

Monomeric anti-DR5 (1 mg/kg bw) was administered as a single dose on the first day of

dosing. Tumor volumes were measured over a 21 day period.

All tested dimerized scTRAIL-RBD constructs showed efficient reduction of Colo205 derived

tumors, independent of their capability to interact with FcγRs (Figure 5B). The anti-tumor

efficacy of APG350 (clearly reduced FcγR binding) was comparable to that of APG808 (no

detectable FcγR binding). At later time points (data not shown) all animals in the APG350

group were found to be free of tumors. In the APG808 group, 7 out of 8 mice were tumor-

free and the 8th very small residual tumor had a volume of about 10 to 20 mm³. Although

APG780 (no FcγR binding) showed the same anti-tumor efficacy as APG350 and APG808

during the first 10 days of treatment (Figure 5B), slow re-growth of tumors was observed

beyond treatment day 10. This reduced anti-tumor efficacy is most likely caused by the

shorter PK of APG780 in mice and is not an effect of the missing FcγR interaction. Compared

to APG350 and APG808 with half lives of 28 h and 25 h, respectively, APG780 has a half life





Page 19 of 26

of 7 h only. The results obtained with the scTRAIL-RBD constructs lacking FcγR-binding

(APG808, APG780) indicate that APG350 and other dimerized scTRAIL-RBD constructs do not

require cross-linking through binding to FcγRs of immune cells for efficient tumor reduction

in vivo.

Figure 6 illustrates the mode of action for APG350 in comparison to agonistic TRAIL-R

antibodies and depicts the fundamental differences with respect to FcγR-dependent cross-

linking. Agonistic antibodies are not capable of inducing apoptosis without further

amplification of TRAIL-R clustering. In vitro, this can be achieved by cross-linking agents or by

small amounts of antibody aggregates (see also Figure 2). In vivo, anti-tumor efficacy of

monomeric agonistic anti-DR4/DR5 antibodies is strongly dependent upon cross-linking by

FcγRs on tumor infiltrating immune cells. In xenograft tumor models that use immune-

deficient mice, tumors contain cell populations expressing FcγRs but have very low levels of

endogenous IgG. In this case, cross-linking by FcγRs effectively amplifies the anti-tumor

response of agonistic anti-DR4/DR5 antibodies. With respect to clinical efficacy, the

dependency on FcγR cross-linking most probably impacts the efficacy of agonistic TRAIL-R

antibodies. In cancer patients, FcγR-mediated cross-linking is strongly impaired in the

presence of endogenous IgGs which compete for FcγR interaction (Figure 6). The reduced

capacity for FcγR-mediated cross-linking ultimately impairs efficient apoptosis induction and

results in a poor anti-tumor response. This lack of/reduced cross-linking in the presence of

physiological IgG is very likely responsible for the disappointing anti-tumor efficacy seen for

many agonistic TRAIL-R antibodies during clinical trials and is a plausible reason why most

agonistic DR4 and DR5 antibodies have been discontinued in clinical development.

Here we present data on APG350, a specifically designed hexavalent pro-apoptotic fully

human TRAIL-R (DR-4 and DR-5) agonist with an improved and distinct mode of action

compared with current clinical development candidates. APG350 induces superior clustering

of TRAIL-Rs and in contrast to agonistic TRAIL-R antibodies does not require cross-linking via

FcγRs for its potent anti-tumor efficacy. The favorable efficacy profile of APG350 is

completed by its excellent tolerability in mice and Cynomolgus monkeys, by a PK profile that

is well suited for clinical development and by a high yield CHO-based manufacturing process.

Dimeric scTRAIL-RBD formats like APG350 may therefore, have the potential to transfer their

proven promising preclinical anti-tumor efficacy to clinical efficacy.




References

Page 20 of 26

REFERENCES

1. Pitti RM, Marsters SA, Ruppert S, Donahue CJ, Moore A, Ashkenazi A. Induction of

apoptosis by Apo-2 ligand, a new member of the tumor necrosis factor cytokine family. The

Journal of biological chemistry. 1996;271:12687-90.

2. Wiley SR, Schooley K, Smolak PJ, Din WS, Huang CP, Nicholl JK, et al. Identification

and characterization of a new member of the TNF family that induces apoptosis. Immunity.

1995;3:673-82.

3. Ashkenazi A, Herbst RS. To kill a tumor cell: the potential of proapoptotic receptor

agonists. J Clin Invest. 2008;118:1979-90.

4. Johnstone RW, Frew AJ, Smyth MJ. The TRAIL apoptotic pathway in cancer onset,

progression and therapy. Nat Rev Cancer. 2008;8:782-98.

5. Walczak H, Miller RE, Ariail K, Gliniak B, Griffith TS, Kubin M, et al. Tumoricidal activity

of tumor necrosis factor-related apoptosis-inducing ligand in vivo. Nature medicine.

1999;5:157-63.

6. Ashkenazi A, Pai RC, Fong S, Leung S, Lawrence DA, Marsters SA, et al. Safety and

antitumor activity of recombinant soluble Apo2 ligand. J Clin Invest. 1999;104:155-62.

7. Adams C, Totpal K, Lawrence D, Marsters S, Pitti R, Yee S, et al. Structural and

functional analysis of the interaction between the agonistic monoclonal antibody Apomab

and the proapoptotic receptor DR5. Cell Death Differ. 2008;15:751-61.

8. Kelley SK, Harris LA, Xie D, DeForge L, Totpal K, Bussiere J, et al. Preclinical studies to

predict the disposition of Apo2L/tumor necrosis factor-related apoptosis-inducing ligand in

humans: characterization of in vivo efficacy, pharmacokinetics, and safety. J Pharmacol Exp

Ther. 2001;299:31-8.

9. Nozawa F, Itami A, Saruc M, Kim M, Standop J, Picha KS, et al. The combination of

tumor necrosis factor-related apoptosis-inducing ligand (TRAIL/Apo2L) and Genistein is

effective in inhibiting pancreatic cancer growth. Pancreas. 2004;29:45-52.




References

Page 21 of 26

10. Pukac L, Kanakaraj P, Humphreys R, Alderson R, Bloom M, Sung C, et al. HGS-ETR1, a

fully human TRAIL-receptor 1 monoclonal antibody, induces cell death in multiple tumour

types in vitro and in vivo. Br J Cancer. 2005;92:1430-41.

11. Yada A, Yazawa M, Ishida S, Yoshida H, Ichikawa K, Kurakata S, et al. A novel

humanized anti-human death receptor 5 antibody CS-1008 induces apoptosis in tumor cells

without toxicity in hepatocytes. Ann Oncol. 2008;19:1060-7.

12. Subbiah V, Brown RE, Buryanek J, Trent JC, Ashkenzai A, Herbst RS, et al. Targeting

the Apoptotic Pathway in Chondrosarcoma Using Recombinant Human Apo2L/TRAIL

(dulanermin), a Dual Pro-apoptotic Receptor (DR4/DR5) Agonist. Molecular cancer

therapeutics. 2012.

13. Soria JC, Mark Z, Zatloukal P, Szima B, Albert I, Juhasz E, et al. Randomized phase II

study of dulanermin in combination with paclitaxel, carboplatin, and bevacizumab in

advanced non-small-cell lung cancer. J Clin Oncol. 2011;29:4442-51.

14. Trarbach T, Moehler M, Heinemann V, Köhne C-H, Przyborek M, Schulz C, et al. Phase

II trial of mapatumumab, a fully human agonistic monoclonal antibody that targets and

activates the tumour necrosis factor apoptosis-inducing ligand receptor-1 (TRAIL-R1), in

patients with refractory colorectal cancer. Br J Cancer. 2010;102:506-12.

15. Younes A, Vose JM, Zelenetz AD, Smith MR, Burris HA, Ansell SM, et al. A Phase 1b/2

trial of mapatumumab in patients with relapsed/refractory non-Hodgkin's lymphoma. Br J

Cancer. 2010;103:1783-7.

16. Doi T, Murakami H, Ohtsu A, Fuse N, Yoshino T, Yamamoto N, et al. Phase 1 study of

conatumumab, a pro-apoptotic death receptor 5 agonist antibody, in Japanese patients with

advanced solid tumors. Cancer chemotherapy and pharmacology. 2010;68:733-41.

17. Forero-Torres A, Shah J, Wood T, Posey J, Carlisle R, Copigneaux C, et al. Phase I trial

of weekly tigatuzumab, an agonistic humanized monoclonal antibody targeting death

receptor 5 (DR5). Cancer Biother Radiopharm. 2010;25:13-9.

18. Herbst RS, Eckhardt SG, Kurzrock R, Ebbinghaus S, O'Dwyer PJ, Gordon MS, et al.

Phase I dose-escalation study of recombinant human Apo2L/TRAIL, a dual proapoptotic

receptor agonist, in patients with advanced cancer. J Clin Oncol. 2010;28:2839-46.




References

Page 22 of 26

19. Herbst RS, Kurzrock R, Hong DS, Valdivieso M, Hsu CP, Goyal L, et al. A first-in-human

study of conatumumab in adult patients with advanced solid tumors. Clin Cancer Res.

2010;16:5883-91.

20. Plummer R, Attard G, Pacey S, Li L, Razak A, Perrett R, et al. Phase 1 and

pharmacokinetic study of lexatumumab in patients with advanced cancers. Clin Cancer Res.

2007;13:6187-94.

21. Soria J-C, Smit E, Khayat D, Besse B, Yang X, Hsu C-P, et al. Phase 1b study of

dulanermin (recombinant human Apo2L/TRAIL) in combination with paclitaxel, carboplatin,

and bevacizumab in patients with advanced non-squamous non-small-cell lung cancer. J Clin

Oncol. 2010;28:1527-33.

22. Kaplan-Lefko PJ, Graves JD, Zoog SJ, Pan Y, Wall J, Branstetter DG, et al.

Conatumumab, a fully human agonist antibody to death receptor 5, induces apoptosis via

caspase activation in multiple tumor types. Cancer biology & therapy. 2010;9:618-31.

23. Natoni A, MacFarlane M, Inoue S, Walewska R, Majid A, Knee D, et al. TRAIL signals to

apoptosis in chronic lymphocytic leukaemia cells primarily through TRAIL-R1 whereas cross-

linked agonistic TRAIL-R2 antibodies facilitate signalling via TRAIL-R2. Br J Haematol.

2007;139:568-77.

24. Wilson NS, Yang B, Yang A, Loeser S, Marsters S, Lawrence D, et al. An Fcγ receptor-

dependent mechanism drives antibody-mediated target-receptor signaling in cancer cells.

Cancer Cell. 2011;19:101-13.

25. Fulda S, Friesen C, Los M, Scaffidi C, Mier W, Benedict M, et al. Betulinic acid triggers

CD95 (APO-1/Fas)- and p53-independent apoptosis via activation of caspases in

neuroectodermal tumors. Cancer Res. 1997;57:4956-64.

26. Hawe A, Kasper JC, Friess W, Jiskoot W. Structural properties of monoclonal antibody

aggregates induced by freeze-thawing and thermal stress. Eur J Pharm Sci. 2009;38:79-87.

27. Joubert MK, Luo Q, Nashed-Samuel Y, Wypych J, Narhi LO. Classification and

characterization of therapeutic antibody aggregates. The Journal of biological chemistry.

2011;286:25118-33.




References

Page 23 of 26

28. Luo Y, Lu Z, Raso SW, Entrican C, Tangarone B. Dimers and multimers of monoclonal

IgG1 exhibit higher in vitro binding affinities to Fcgamma receptors. mAbs. 2009;1:491-504.

29. Li H, Zhu H, Xu CJ, Yuan J. Cleavage of BID by caspase 8 mediates the mitochondrial

damage in the Fas pathway of apoptosis. Cell. 1998;94:491-501.

30. Kim SH, Kim K, Kwagh JG, Dicker DT, Herlyn M, Rustgi AK, et al. Death induction by

recombinant native TRAIL and its prevention by a caspase 9 inhibitor in primary human

esophageal epithelial cells. The Journal of biological chemistry. 2004;279:40044-52.

31. Holler N, Tardivel A, Kovacsovics-Bankowski M, Hertig S, Gaide O, Martinon F, et al.

Two adjacent trimeric Fas ligands are required for Fas signaling and formation of a death-

inducing signaling complex. Mol Cell Biol. 2003;23:1428-40.

32. Armour KL, van de Winkel JG, Williamson LM, Clark MR. Differential binding to

human FcgammaRIIa and FcgammaRIIb receptors by human IgG wildtype and mutant

antibodies. Molecular immunology. 2003;40:585-93.

33. Jefferis R. Glycosylation of recombinant antibody therapeutics. Biotechnology

progress. 2005;21:11-6.

34. Overdijk MB, Verploegen S, Ortiz Buijsse A, Vink T, Leusen JH, Bleeker WK, et al.

Crosstalk between human IgG isotypes and murine effector cells. J Immunol. 2012;189:3430-

8.




References

Page 24 of 26

Figure Legends

Figure 1: Structure and characterization of APG350. (A) Schematic depiction of APG350

structure. C-terminal fusion of an engineered IgG1-Fc to the single-chain TRAIL receptor

binding domain (scTRAIL-RBD) resulted in a dimeric molecule with six TRAIL-R binding sites.

(B) Size exclusion chromatography of purified APG350. Protein was detected by on-line

measurement of absorption at 210 nm. (C) SDS polyacrylamide gel electrophoresis of two

batches of purified APG350 under non-reducing and reducing conditions. (D) Efficacy of two

different batches of purified APG350 on Colo205 cells as demonstrated by their cytotoxicity

(MTS assay).

Figure 2: Comparison of in vitro cytotoxicity of APG350 and (fractionated) anti-DR5 antibody.

Cytotoxicity in vitro was assessed using the MTS-assay on Colo205 cells (A-D). (A) In vitro

cytotoxicity of bulk anti-DR5 with and without cross-linking. Activity of “bulk” anti-DR5

antibody alone (; EC50 = 71 ng/ml) and with cross-linking (; EC50 = 6 ng/ml). (B) In vitro

cytotoxicity of APG350 alone (; EC50 =10 ng/ml) and in the presence of a cross-linking

antibody (; EC50 = 5 ng/ml). (C) In vitro cytotoxicity of the different fractions of anti-DR5

antibody bulk preparation. Monomeric anti-DR5 antibody (; EC50 = >3000 ng/ml), anti-DR5

dimers “HMW1” (; EC50 = 3.3 ng/ml), anti-DR5 multimers “HMW2” (; EC50 < 0.2 ng/ml).

(D) In vitro cytotoxicity of monomeric anti-DR5 antibody alone (; EC50 = >3000 ng/ml) and

in the presence of a cross-linking antibody (; EC50 = 6 ng/ml). (E) Kinetics of apoptosis

induction in T98G glioblastoma cells. T98G cells were treated with 0.1nM of APG350 or an

agonistic anti-DR5-antibody that was either applied monomeric or cross-linked as indicated.

Apoptosis induction indicated by caspase activation or Bid cleavage was analyzed by

Western-blot on cell lysates prepared at the time points indicated.

Figure 3: Comparison of in vitro cytotoxicity of APG350 to TRAIL and an anti-DR5 antibody in

established (A) and primary (B) human tumor cells. In vitro cytotoxicity was evaluated after

24 h of treatment with TRAIL-R agonists using the MTS assay. For each cell type examined,

the activity of APG350 was compared to bulk anti-DR5 antibody and Apo2L/TRAIL. Bulk anti-

DR5 antibody, as commercially supplied, was used in these studies to allow comparison to

previous studies and literature data. For comparative purposes, the TRAIL-R agonists were

applied at equimolar concentrations over concentration ranges indicated in subfigures for

established tumor cell lines (A) and for primary tumor cells (B). The TRAIL receptor surface




References

Page 25 of 26

expression analysis of the tumor cells used as determined by flow cytometry is summarized

in Supplemental Table S3.

Figure 4: Comparison of anti-tumor efficacy of APG350 to TRAIL and anti-DR5 antibody on

Colo205 cell derived mouse xenograft tumors in vivo. NMRI/nude mice (8-9 per group)

bearing Colo205 xenograft tumors were treated (i.v.) with respective TRAIL-R agonists

APG350, monomeric anti-DR5 antibody and Apo2L/TRAIL, tumor volumes were measured at

time points indicated. (A) Dose-dependent anti-tumor activity of APG350. The number of

complete responses was 0/9 in the control group, 0/8 at 0.3 mg/kg bw APG350, 8/8 at 1

mg/kg bw APG350, and 8/8 at 3 mg/kg bw APG350. Control ( black), APG350 0.3 mg/kg

(, grey), 1 mg/kg (, red), 3 mg/kg (, green). Therapy: administration of APG350 i.v on

five consecutive days (). (B) Anti-tumor activity of 1 mg/kg APG350 (, red), monomeric

anti-DR5 (, blue) and Apo2L/TRAIL (, violett) compared to control ( black). Therapy:

administration of APG350 or Apo2L/TRAIL i.v. on five consecutive days (); administration

of monomeric anti-DR5 i.v. on day 0 (). (C) NMRI/nude mice with established Colo205

tumor xenografts received a single dose (1 mg/kg bw) of each TRAIL receptor agonist or PBS

as indicated. Tumors were excised 4 hours after treatment and analyzed by

immunohistochemistry for markers of apoptosis induction (active caspase 3/8 and cleaved

PARP). Please note, that the data presented in figures 4 and 5 belong to one study.

Therefore, in Fig.4b and Fig.5b the same control- and anti-DR5 mice were included in the

graphs. Mean tumor volumes and SEM values of all groups are listed in the supplement

(Supplemental Tables S4 and S5).

Figure 5: The anti-tumor effect of APG350 is independent of clustering by Fcγ Receptors. (A)

Binding to Fcγ Receptors (CD16, CD32a/b and CD64). The binding of APG350, APG780,

APG802 and APG808 to FcγRs (as indicated, see Supplemental Table S1 for molecular design)

was determined employing an HTRF assay. In the assay, unlabeled TRAIL-R agonists compete

with a labeled human IgG for binding to the respective FcγR. The resulting decrease in FRET

signal is proportional to the sample´s binding affinity. (B) Anti-tumor efficacy in a tumor

xenograft model. NMRI/nude mice (8 per group) bearing Colo205 xenograft tumors were

treated with the respective TRAIL-R agonists. APG350, APG808 and APG780 (each at 1 mg/kg

bw) were administered once daily on 5 consecutive days. Monomeric anti-DR5 (1 mg/kg bw)

was administered as a single dose on the first day of dosing. Mean tumor volume was

measured over a 21-day period. Please note, that the the data presented in figures 4 and 5




References

Page 26 of 26

belong to one study. Therefore, in Fig.4b and Fig.5b the same control- and anti-DR5 mice

were included in the graphs. Mean tumor volumes and SEM values of all groups are listed in

the supplement (Supplemental Tables S4 and S5).

Figure 6: Hypothetical mode of action of APG350 compared to agonistic antibodies. For

agonistic monoclonal antibodies (which bind 2 TRAIL receptors) clustering via FcγRs is

required for efficient anti-tumor response in vivo in xenograft models. In cancer patients an

efficient clustering of agonistic antibodies is inhibited by high concentrations of endogenous

IgG that competes for FcγR binding. No FcγR dependent clustering is required for full anti-

tumor activity of APG350 (six TRAIL receptor binding sites) preclinically and very likely also

for clinical efficacy.




Figure 1

COOH BACH3

CH2 CH2

CH3

COOHCOOH

2000

2500

mAU

NH2 NH2

IIIIII

1000

1500

2000

Module I Module II IgG1-FcModule III IgG1-Fc

OD 210nm0

500

0.0 5.0 10.0 15.0 20.0 25.0 ml

C D

100

125APG350 Lot3149APG350 Lot3179[%

]

70

100

150

25

50

75C

ell v

iabi

lity

[

non reduced reduced50

00.1 1 10 100 1000 10000

Concentration [ng/ml]

on August 18, 2020. ©

2013 Am

erican Association for C

ancer Research.

mct.aacrjournals.org

Dow

nloaded from

Author m

anuscripts have been peer reviewed and accepted for publication but have not yet been edited.

Author M

anuscript Published O

nlineFirst on O

ctober 7, 2013; DO

I: 10.1158/1535-7163.MC

T-13-0323


Figure 2

BA

25

50

75

100

125

APG350 + x-link

APG350

Cell v

iabilli

ty [%

]

BA

25

50

75

100

125

anti-DR5 (bulk)

anti-DR5 (bulk)+ x-link

Cel

l vi

abil

ity

[%]

75

100

125

ity

[%]

75

100

125

ty [

%]

00.1 1 10 100 1000 10000

Concentration [ng/ml]C D

00.1 1 10 100 1000 10000


0

25

50 anti-DR5 (monomer)

0.1 1 10 100 1000 10000

anti-DR5 (HMW-1)

anti-DR5 (HMW-2)


Cel

l vi

abil

0

25

50

75

anti-DR5 (monomer)

anti-DR5 + x-link

0.1 1 10 100 1000 10000


Cel

l vi

abil

it

E

Anti-DR5 APG350

Anti-DR5+

x-link

63 9 12 24 63 9 12 24 63 9 12 24Time [h] Ctr

l

Ctr

l +

x-l

ink

55/53

kDa

Caspase 8

Bid

55/53

43/41

18

22

Caspase 3

Bid

15

35

19/17/12

β-Actin

Caspase 943

37/35

41




A100

non-small-cell lung cancer cell line A549

100

colon adenocarcinoma cell line WiDr

Figure 3

0

25

50

75

0,1 1 10 100

viab

ility

[%]

[ M]

0

25

50

75

0,1 1 10 100

viab

ility

[%]

conc [nM]conc. [nM]

50

75

100

y [%

]

metastatic breast adenocarcinoma cell line MDA-MB-231

50

75

100

y [%

]

glioblastoma cell line T98G

conc. [nM]

0

25

50

0,1 1 10 100

viab

ility

conc. [nM]

0

25

50

0,001 0,01 0,1 1 10

viab

ility

conc. [nM]

APG350 Apo2L/TRAIL anti-DR5

B

75

100

y [%

]

primary ovarian carcinoma cells (T0997)

75

100

ility

[%]

primary metastatic prostate carcinoma cells (T1111)

0

25

50

0,1 1 10 100

viab

ility

conc. [nM]

0

25

50

0,1 1 10 100

viab

conc. [nM]

50

75

100

bilit

y [%

]

primary systemic mastocytosis cells (T1034)

50

75

100

viab

ility

[%]

primary colon adenocarcinoma cells (T1110)

APG350 Apo2L/TRAIL anti-DR5

0

25

0,1 1 10 100

via

conc. [nM]

0

25

0,1 1 10 100conc. [nM]




1200 1.200 BA

400

600

800

1000m

or v

olum

e (m

m³)

400

600

800

1.000

umor

vol

ume

(mm

3 )

0

200

400

0 5 10 15 20 25 30

Mea

n tu

m

Time (days)

0

200

400

0 10 20 30 40 50

Mea

n tu

Time (days)

C

Control APG350 0.3 mg/kg APG350 1.0 mg/kgAPG350 3.0 mg/kg Therapy

Control Apo2L/TRAIL 1mg/kg Anti-DR5 1mg/kg

APG350 1mg/kg Therapy Therapy Anti-DR5

C

Figure 4

on August 18, 2020. ©

2013 Am


ancer Research.


Dow

nloaded from

Author m


Author M


nlineFirst on O

ctober 7, 2013; DO

I: 10.1158/1535-7163.MC

T-13-0323


Figure 5

CD16 (FcγRIII) CD32 (FcγRIIB)A

HT

RF

rat

io [

RF

U]

-1 0 1 2 3 4

2000

4000

6000

8000

10000

12000

14000

HT

RF

rat

io [

RF

U]

-1 0 1 2 3 4

2000

4000

6000

8000

10000

12000

Concentration [log µM]

APG350 APG802 APG808APG780

CD32 (FcγRIIA)

tio

[R

FU

]

8000

10000

12000



CD64 (FcγRI)

tio

[R

FU

]

8000

10000

12000


HT

RF

ra

-1 0 1 2 3 4

2000

4000

6000



HT

RF

ra

-2 -1 0 1 2 3 4

2000

4000

6000


300

400

500

e (m

m3)

B

100

200

300

Me

an

tum

or v

olu

me

00 5 10 15 20 25

M

Time (days)

Control APG780 1mg/kg APG808 1mg/kgAnti-DR5 1mg/kg APG350 1mg/kg TherapyTherapy Anti-DR5




Agonistic TRAIL-R specific antibodies Hexavalent scTRAIL-RBD agonists

In cancerpatients

• Fcγ receptors• High IgG concentration

In vitro In vivo xenograft

• Fcγ receptors• Low IgG concentration

• No Fcγ receptors• No IgGs

In vivo xenograftIn cancer patients

• Fcγ receptors• High IgG concentration

Human endogenous

Fcγ receptor

Fcγ receptor

AgonisticHuman

endogenous

Fcγ receptor

High IgG concentrationLow IgG concentrationNo IgGs High IgG concentration

endogenousIgGs

gantibody

TRAIL-R

endogenousIgGsAPG350

Low clusteringcapacity

Insufficient

High clusteringcapacity

A i

Low clusteringcapacity

Insufficient

High clusteringcapacity

ApoptosisApoptosisApoptosis Apoptosis

Figure 6

on August 18, 2020. ©

2013 Am


ancer Research.


Dow

nloaded from

Author m


Author M


nlineFirst on O

ctober 7, 2013; DO

I: 10.1158/1535-7163.MC

T-13-0323


Published OnlineFirst October 7, 2013.Mol Cancer Ther Christian Gieffers, Michael Kluge, Christian Merz, et al.

-Receptorsγcross-linking via Fcshows therapeutic anti-tumor efficacy independent of APG350 induces superior clustering of TRAIL-Receptors and

Updated version

10.1158/1535-7163.MCT-13-0323doi:

Access the most recent version of this article at:

Material

Supplementary

http://mct.aacrjournals.org/content/suppl/2013/10/07/1535-7163.MCT-13-0323.DC1

Access the most recent supplemental material at:

Manuscript

Authoredited. Author manuscripts have been peer reviewed and accepted for publication but have not yet been

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

Subscriptions

Reprints and

[email protected] at

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

Permissions

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

.http://mct.aacrjournals.org/content/early/2013/10/05/1535-7163.MCT-13-0323To request permission to re-use all or part of this article, use this link



http://mct.aacrjournals.org/lookup/doi/10.1158/1535-7163.MCT-13-0323

http://mct.aacrjournals.org/content/suppl/2013/10/07/1535-7163.MCT-13-0323.DC1

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

mailto:[email protected]

http://mct.aacrjournals.org/content/early/2013/10/05/1535-7163.MCT-13-0323


APG350 induces superior clustering of TRAIL-Receptors and … · Introduction Page 6 of 26 of FcγR-mediated cross-linking events. In the final layout, two single-chain TRAIL-R binding

Documents