A potent anti-CD70 antibody-drug conjugate combining a dimeric pyrrolobenzodiazepine drug with site-specific conjugation technology

Article abstract of DOI:10.1021/bc400217g

A highly cytotoxic DNA cross-linking pyrrolobenzodiazepine (PBD) dimer with a valine-alanine dipeptide linker was conjugated to the anti-CD70 h1F6 mAb either through endogenous interchain cysteines or, site-specifically, through engineered cysteines at position 239 of the heavy chains. The h1F6 _239C-PBD conjugation strategy proved to be superior to interchain cysteine conjugation, affording an antibody-drug conjugate (ADC) with high uniformity in drug-loading and low levels of aggregation. In vitro cytotoxicity experiments demonstrated that the h1F6_239C-PBD was potent and immunologically specific on CD70-positive renal cell carcinoma (RCC) and non-Hodgkin lymphoma (NHL) cell lines. The conjugate was resistant to drug loss in plasma and in circulation, and had a pharmacokinetic profile closely matching that of the parental h1F6_239C antibody capped with N-ethylmaleimide (NEM). Evaluation in CD70-positive RCC and NHL mouse xenograft models showed pronounced antitumor activities at single or weekly doses as low as 0.1 mg/kg of ADC. The ADC was tolerated at 2.5 mg/kg. These results demonstrate that PBDs can be effectively used for antibody-targeted therapy.

Full text of DOI:10.1021/bc400217g

Article
pubs.acs.org/bc
A Potent Anti-CD70 Antibody−Drug Conjugate Combining a Dimeric
Pyrrolobenzodiazepine Drug with Site-Specific Conjugation
Technology
Scott C. Jeffrey,* Patrick J. Burke,† Robert P. Lyon,† David W. Meyer,† Django Sussman,†
Martha Anderson,† Joshua H. Hunter,† Chris I. Leiske,† Jamie B. Miyamoto,† Nicole D. Nicholas,†
Nicole M. Okeley,† Russell J. Sanderson,† Ivan J. Stone,† Weiping Zeng,† Stephen J. Gregson,‡
Luke Masterson,‡ Arnaud C. Tiberghien,‡ Philip W. Howard,‡ David E. Thurston,§ Che-Leung Law,†
and Peter D. Senter†
†Department of Research & Translational Medicine, Seattle Genetics, Inc., Bothell, Washington 98021, United States
‡Spirogen Ltd., QMB Innovation Centre, 42 New Road, London E1 2AX, United Kingdom
§Department of Pharmacy, Institute of Pharmaceutical Sciences, King’s College London, Britannia House, 7 Trinity Street, London
SE1 1DB
S
* Supporting Information
ABSTRACT: A highly cytotoxic DNA cross-linking pyrrolobenzodiaze-
pine (PBD) dimer with a valine-alanine dipeptide linker was conjugated to
the anti-CD70 h1F6 mAb either through endogenous interchain cysteines
or, site-specifically, through engineered cysteines at position 239 of the
heavy chains. The h1F6239C-PBD conjugation strategy proved to be
superior to interchain cysteine conjugation, affording an antibody−drug
conjugate (ADC) with high uniformity in drug-loading and low levels of
aggregation. In vitro cytotoxicity experiments demonstrated that the
h1F6239C-PBD was potent and immunologically specific on CD70-positive
renal cell carcinoma (RCC) and non-Hodgkin lymphoma (NHL) cell
lines. The conjugate was resistant to drug loss in plasma and in circulation,
and had a pharmacokinetic profile closely matching that of the parental
h1F6239C antibody capped with N-ethylmaleimide (NEM). Evaluation in
CD70-positive RCC and NHL mouse xenograft models showed pronounced antitumor activities at single or weekly doses as low
as 0.1 mg/kg of ADC. The ADC was tolerated at 2.5 mg/kg. These results demonstrate that PBDs can be effectively used for
antibody-targeted therapy.
tion, and distribution16 have all been identified as important
considerations for optimal ADC performance.
INTRODUCTION
■
ADCs have attracted significant interest for cancer therapy,
largely based on the approved drugs ADCETRIS (brentuximab
vedotin) and KADCYLA (ado-trastuzumab emtansine) for the
In this current work, we extend ADC technology to include a
highly potent and broadly active PBD dimer17,18 as a drug
payload. Previously,19 we demonstrated that a PBD-based ADC
treatment of CD30 and Her2/neu-positive malignancies,
had pronounced activity in models of acute myelogenous
leukemia. We extend these findings to include other
malignancies and illustrate how the hydrophobic properties of
the PBD dimer require site-specific conjugation to minimize
aggregation caused by more highly loaded ADC species. This
was achieved through a serine-to-cysteine mutation at position
239 of the h1F6 (anti-CD70) antibody heavy chains, facilitating
selective drug attachment. The resulting h1F6239C-PBD
conjugate was well characterized, induced immunologically
specific cell kill in vitro, and was highly active in vivo. This new
respectively.1 Both ADCs employ antimitotic agents as the
drug payload and serve as prototypes for many related agents
that are in various stages of preclinical and clinical develop-
ment. Currently, there are more than 20 ADCs in clinical trials
spanning a variety of solid and hematologic cancer
indications.2,3
In the effort to extend the technology and identify and
optimize key parameters that lead to highly active and well-
tolerated ADCs, numerous drug types,4−7 linker strategies,8,9
antigen targets,2 and antibody formats10 have been investigated.
Factors such as drug potency and mechanism,11 linker
stability,12 drug release mechanism,9,13 conjugate composi-
tion,14 pharmacokinetics,10 antigen expression,15 internaliza-
Received: April 29, 2013
Revised: June 11, 2013
Published: June 12, 2013
© 2013 American Chemical Society
1256
dx.doi.org/10.1021/bc400217g | Bioconjugate Chem. 2013, 24, 1256−1263

Bioconjugate Chemistry
Article
anticancer agent incorporates several advancements in ADC
technology.
stir for 3 h. The mixture was aspirated onto a 2 mm radial
chromatography plate and eluted with 3% methanol in
dichloromethane. The fractions were analyzed by TLC (5%
methanol/dichloromethane) and product-containing fractions
were combined and concentrated to give 67 mg (45%).
Analytical HPLC (0.1% formic acid): tR 11.51 min. LC-MS: tR
12.73 min, m/z (ES+) expected 1089.5 (M+H)+, found 1089.6
m/z (ES-) expected 1087.4 (M-H)−, found 1087.3.
EXPERIMENTAL PROCEDURES
■
General Methods and Materials. Commercially available
anhydrous solvents were used without further purification.
Analytical thin layer chromatography was performed on silica
gel 60 F254 aluminum sheets (EMD Chemicals, Gibbstown,
NJ). Radial chromatography was performed on a Chromato-
tron apparatus (Harris Research, Palo Alto, CA). Analytical
HPLC was performed on a Varian ProStar 210 solvent delivery
system configured with a Varian ProStar 330 PDA detector.
Samples were eluted over a C12 Phenomenex Synergi 2.0 ×
150 mm, 4 μm, 80 Å reversed-phase column. The acidic mobile
phase consisted of acetonitrile and water, both containing 0.1%
formic acid. Compounds were eluted with a linear gradient of
acidic acetonitrile from 5% at 1 min post injection, to 95% at 11
min, followed by isocratic 95% acetonitrile to 15 min (flow rate
= 1.0 mL/min). LC-MS was performed on a ZMD Micromass
mass spectrometer interfaced to an HP Agilent 1100 HPLC
instrument equipped with a C12 Phenomenex Synergi 2.0 ×
150 mm, 4 μm, 80 Å reversed-phase column. The acidic eluent
consisted of a linear gradient of acetonitrile from 5% to 95% in
0.1% aqueous formic acid over 10 min, followed by isocratic
95% acetonitrile for 5 min (flow rate = 0.4 mL/min).
Conjugate Preparation. The CD70 antibody engineered
with a cysteine residue at position 239 of the heavy chain
(h1F6239C) was fully reduced by adding 10 equiv of TCEP and
1 mM EDTA and adjusting the pH to 7.4 with 1 M Tris buffer
(pH 9.0). Following 1 h incubation at 37 °C, the reaction was
cooled to 22 °C and 30 equiv of dehydroascorbic acid were
added. The pH was adjusted to 6.5 with 1 M Tris buffer (pH
3.7) and the oxidation reaction was allowed to proceed for 1 h
at 22 °C. This resulted in reformation of native disulfides, but
left the cysteines at position 239 in the reduced state and
available for conjugation. The pH of the solution was then
raised again to 7.4 by addition of 1 M Tris buffer (pH 9.0). To
maintain solubility of the PBD-linker, the antibody itself was
diluted with propylene glycol to a final concentration of 33%.
The solution of PBD linker (3.0 equiv) in propylene glycol was
added to the antibody solution to effect the conjugation. The
final concentration of propylene glycol in the conjugation
reaction was 50%. The reaction was allowed to proceed for 30
min. The mixture was treated with activated charcoal as a slurry
in propylene glycol and water for 30 min. The activated
charcoal was then removed by filtering. Extent of aggregation
was determined by size exclusion chromatography. Drug-
loading was determined by reducing the ADC with NaCNBH3
and dithiothreitol followed by HPLC analysis on a PLRP
column and integration of the heavy and light chain
components.
In Vitro Assays. The following cells were treated with PBD
dimer and PBD ADCs: 786-O (500 cells/well), Caki-1 (1000
cells/well), UM-RC-3 (1500 cells/well), ACHN (3000 cells/
well), MHH-PREB-1 and Raji (7000 cells/well), WSU-NHL
(7000 cells/well), and KMH2 (2000 cells/well). The tissue
culture media consisted of DMEM, 10% FBS, and 1% sodium
pyruvate, nonessential amino acids, and L-glutamine (LifeTech,
Carlsbad, CA). The assay was conducted in 96-well, clear flat-
bottom, black-sided plates. The cells were plated (150 μL/well)
one day prior to treatment, allowed to adhere in a biological
safety cabinet for 1 h, then placed in a CO2 incubator overnight.
Cells were then treated with PBD dimer or ADC and incubated
at 37 °C for 96 h under a 5% CO2 atmosphere. After 96 h, the
cells were labeled with Cell Titer Glo (Promega, Madison, WI,
100 μL/well) and after shaking for 2 min and incubation for an
additional 25 min in the dark, luminescence was measured on a
plate reader. The IC50 values, determined in triplicate, are
defined here as the concentration that results in a 50%
reduction in cell growth relative to untreated controls.
(S)-2-((S)-2-(6-(2,5-Dioxo-2,5-dihydro-1H-pyrrol-1-yl)-
hexanamido)-3-methylbutanamido)propanoic acid (mc-
val-ala-OH). To a solution of H2N-Valine-Alanine-OH
dipeptide (200 mg, 1.06 mmol) dissolved in anhydrous DMF
(10.6 mL) was added maleimidocaproyl NHS ester (327 mg,
1.06 mmol). Diisopropylethylamine (0.92 mL, 5.3 mmol) was
then added and the reaction mixture was stirred under nitrogen
at ambient temperature for 18 h, at which time TLC and
analytical HPLC revealed consumption of the starting
materials. The reaction was diluted with 0.1 M HCl (100
mL) and the aqueous layer was extracted with ethyl acetate (3
× 100 mL). The combined organic layer was washed with water
and brine, then dried over sodium sulfate, filtered, and
concentrated. The crude product was dissolved in minimal
dichloromethane and purified by radial chromatography on a 2
mm chromatotron plate eluted with dichloromethane/meth-
anol mixtures (95:5 to 90:10 dichloromethane/methanol) to
provide 158 mg (39%) as an white solid. TLC: Rf = 0.26, 10%
1
methanol in dichloromethane. H NMR (400 MHz, CDCl3)
(ppm) 0.95 (d, J = 17 Hz, 3H), 0.98 (d, J = 17 Hz, 3H), 1.30
(m, 2H), 1.40 (d, J = 17 Hz, 3H), 1.61 (m, 4H), 2.06 (m, 1H),
2.25 (dt, J = 4, 19 Hz, 2H), 3.35 (s, 1H), 3.49 (t, J = 17 Hz,
2H), 4.20 (d, J = 18 Hz, 1H), 4.38 (m, 1H), 6.80 (s, 2H).
Analytical HPLC (0.1% formic acid): tR 9.05 min. LC-MS: tR
11.17 min, m/z (ES+) expected 382.2 (M+H)+, found 381.9,
m/z (ES-) expected 380.2 (M-H)−, found 379.9.
6-(2,5-Dioxo-2,5-dihydro-1H-pyrrol-1-yl)-N-((S)-1-(((S)-
1-((4-((S)-7-methoxy-8-(3-(((S)-7-methoxy-2-(4-methoxy-
phenyl)-5-oxo-5,11a-dihydro-1H-benzo[e]pyrrolo[1,2-
a][1,4]diazepin-8-yl)oxy)propoxy)-5-oxo-5,11a-dihydro-
1H-benzo[e]pyrrolo[1,2-a][1,4]diazepin-2-yl)phenyl)-
amino)-1-oxopropan-2-yl)amino)-3-methyl-1-oxobutan-
2-yl)hexanamide (PBD-linker). To a mixture of mc-val-ala-
OH (157 mg, 0.41 mmol) in a 5% methanol/dichloromethane
mixture (3 mL) was added EEDQ (102 mg, 0.41 mmol). The
mixture was stirred for 30 min, followed by cooling to 0 °C and
the addition of solid PBD dimer20 (100 mg, 0.138 mmol). The
ice-bath was removed and the reaction mixture was allowed to
Xenograft Studies. All experiments were conducted in
concordance with the Institutional Animal Care and Use
Committee in a facility fully accredited by the Association for
Assessment and Accreditation of Laboratory Animal Care.
ACHN and Raji lymphoma disseminated models: Female C.B-
17 SCID mice (n = 10, Harlan, Indianapolis, IN) were injected
intraperitoneally (ip) with ACHN cells (106) or intravenously
(iv) Raji lymphoma cells (5 × 106). The h1F6239C-PBD and
hIgG239C-PBD ADCs were dosed ip according to schedule at
0.1 and 0.3 mg/kg. The mice were monitored daily and were
1257
dx.doi.org/10.1021/bc400217g | Bioconjugate Chem. 2013, 24, 1256−1263

Bioconjugate Chemistry
Article
Figure 1. Structures of the PBD dimer, linker, and ADC.
terminated with symptoms of metastatic disease including hind
limb paralysis, cranial swelling, bloated abdomen, scruffy coat,
and hunched posture. Surviving mice were euthanized on day
100. Caki-1 and MHH-PREB-1 models: Caki-1 tumor
fragments or MHH-PREB-1 cells (5 × 106) were implanted
in the right flank of nude female mice or C.B-17 SCID mice,
respectively. Mice were randomized to study groups (n = 5 for
Caki-1, and n = 7 for MHH-PREB-1) on day 10 with each
group averaging around 100 mm3 tumor. The h1F6239C-PBD
and hIgG239C-PBD ADCs were dosed ip according to a q7dx2
schedule at 0.1 and 0.3 mg/kg. Tumor volume as a function of
time was determined using the formula (L × W2)/2. Animals
were euthanized when tumor volume reached 1000 mm3.
Cured animals were euthanized on day 90 post implant.
ADC and Antibody ELISA Methods. Plasma samples
from ADC-treated female BALB/c mice were analyzed for total
h1F6239C (mAb) and total ADC (Drug) using two different
types of ELISA methods. To measure mAb concentration,
microtiter plates were coated with a commercially available
murine mAb (2 μg/mL) that binds to the human kappa light
chain region of the antibody component of the h1F6239C-PBD.
Standards, controls, and samples were then incubated on
coated and blocked microtiter plates in duplicate after a 1/5
dilution into assay diluent (PTB buffer). Bound h1F6239C-PBD
was detected with biotinylated anti-ID-h1F6 mAb followed by
addition of a polymer horseradish peroxidase conjugated to
streptavidin (poly-HRP-SA). To measure PBD drug concen-
tration (Drug), microtiter plates were coated with anti-ID-h1F6
mAb that binds to the variable region of the antibody
component of the h1F6239C-PBD antibody-drug conjugate
(ADC). Standards, controls, and samples were then incubated
on coated and blocked microtiter plates in duplicate after a 1/5
dilution into assay diluent (PTB buffer). Bound h1F6239C-PBD
was detected with a biotinylated anti-PBD mAb that binds to
the PBD-linker portion of the ADC followed by addition of a
polymer horseradish peroxidase conjugated to streptavidin
(poly-HRP-SA). For both the mAb and drug ELISA methods,
3,3′,5,5′-tetramethylbenzidine (TMB) was used for detection
which is converted from colorless TMB to a colored TMB
derivative by horseradish peroxidase. The signal is proportional
to the amount of h1F6239C-PBD bound to the plate as
measured by absorbance at OD450 nm-OD630 nm.
RESULTS
■
Drug-Linker Design and Synthesis. We focused on PBD
dimers as a drug class for ADCs due to their potent
cytotoxicity, broad anticancer activity, synthetic accessibility,
and complementary mechanism to our lead auristatin-based
platform technology.1 PBD dimers cross-link DNA by binding
in the minor groove and reacting with interstrand guanine
residues at GATC or related sequences.21−23 Published data
with monomeric PBDs indicated that the most potent forms
contain C-ring unsaturation and 2-aryl substituents.24 Although
not previously demonstrated, we reasoned that PBD dimers
with a C-2 para-aniline residue would have high potency, and
could be used for peptide linker attachment. The PBD dimer
used here (Figure 1)20 was prepared using established methods
including a Suzuki coupling reaction to install the para-aniline
residue.24
Much ADC research to date has focused on peptide-based
linker strategies, both for auristatin8,11 and other drug
classes.5−7 We have demonstrated that attaching a peptide
linker directly to the drug via an aniline residue affords ADCs
with favorable plasma stability and facile drug release within
tumors.7,11 For the PBDs, we employed the known protease-
cleavable valine-alanine5,6 sequence (Figure 1) that was known
to be stable in circulation, but readily cleaved within target cells.
Coupling of the maleimidocaproyl-valine-alanine acid to the
aniline residue of the PBD dimer using the standard peptide
coupling agent EEDQ allowed us to prepare the PBD-linker in
a single step. The structure of the drug-linker is illustrated in
Figure 1.
Conjugation of PBD-Linker to h1F6 and the h1F6239C
mAbs. Our investigation of the PBD dimers as a drug class for
ADCs began with the h1F6 antibody which targets the CD70
cancer antigen.25 The expression of CD70 on both renal cell
carcinoma (RCC) and non-Hodgkin lymphoma (NHL) (Table
1) is well documented,26 and CD70 has limited normal tissue
expression.16 The antigen internalizes efficiently upon ADC
binding.25 Conjugation of the PBDs to mAbs required the use
1258
dx.doi.org/10.1021/bc400217g | Bioconjugate Chem. 2013, 24, 1256−1263

Bioconjugate Chemistry
Article
Table 1. Cellular CD70 Antigen Expression, Rhodamine
Efflux Ability, and in Vitro Cytotoxic Activity of the PBD
Dimer and h1F6239C-PBD ADC on Lymphoma and Renal
Cell Carcinoma Lines
Cytotoxic Activity IC50
(nM)
Antigens/cell
Rhodamine
PBD
h1F6239C
PBD
-
Cell Line
Raji
(×103)
Efflux
Dimer
Lymphoma
23
28
Negative
Negative
Negative
Unknown
0.4
0.8
MHH-PREB-1
WSU-DLCL2
KMH2
1.0
0.4
Figure 2. Conjugation process for the 239C antibody format. The
engineered antibody, expressed in CHO cells, was isolated as the
cysteine disulfide at position 239. The antibody was fully reduced with
TCEP and partially reoxidized with dehydroascorbic acid. The
resulting free cysteines at position 239 were conjugated to the PBD-
linker to give the PBD ADC with nominally 2 drugs/mAb.
35
0.1
0.6
160
0.15
0.05
Renal Cell Carcinoma
786-O
190
Positive
Positive
Negative
Negative
0.3
0.5
0.2
0.3
0.1
UM-RC-3
ACHN
Caki-1
70
55
0.2
0.2
reoxidized with dehydroascorbic acid. Due to the proximity of
the hinge cysteine residues, the reoxidation step selectively
reformed the native disulfide bonds, but left the two engineered
239-cysteine residues in the reduced form and available for
conjugation. Addition of the PBD-linker as a solution in
propylene glycol resulted in efficient conjugation (1.9 drug/
mAb) and PLRP-MS demonstrated exclusive loading of the
PBD-linker onto the heavy chains of the antibody (Figure 3).
140
0.02
of significant amounts of cosolvents due to low aqueous
solubility of the PBD-linker. We found that propylene glycol
(50%) was an effective cosolvent that gave efficient conversion
without precipitation of the drug-linker during the reaction and
had no adverse effects on the protein.
Various constructs were investigated including 2- and 4-
loaded ADCs on endogenous hinge cysteine residues of h1F6
and a site-specific 2-load ADC on h1F6239C. For the hinge
cysteine conjugates, the monoclonal antibody h1F6 was treated
with tris(2-carboxyethyl)phosphine (TCEP) to partially reduce
the interchain disulfide bonds and provide antibody with an
average of 2 or 4 free thiols/antibody.27 Conjugation of
reduced h1F6 with 2 or 4 thiol groups/mAb with the PBD-
linker resulted in ADCs having 13% and 45% aggregate,
respectively (Table 2), as determined by size-exclusion
Table 2. Comparison of Aggregation and Unconjugated
Antibody Percentages for 2- and 4-Loaded Hinge Cysteine
Conjugates versus Engineered Cysteine Conjugates of the
PBD Linker
Figure 3. Analytical characterization of h1F6239C-PBD. A. PLRP
analysis of reduced ADC showed exclusive loading of PBD-linker on
antibody heavy chain. B. Size exclusion analysis of h1F6239C-PBD
showed minimal aggregation.
h1F6-PBD(4) h1F6-PBD(2) h1F6239C-PBD
% Aggregate
45
13
51
1.6
5.4
% Unconjugated mAb
7.2
Aggregation (1.6%) was significantly lower than that obtained
by conjugating through endogenous hinge cysteines (Table 2).
Thus, the recombinant construct provided superior ADCs
compared to hinge disulfide conjugates in terms of ADC
uniformity and aggregation levels.
chromatography (Supporting Information Figure S1). Hydro-
phobic interaction chromatography of h1F6-PBD (average 2-
load) indicated that approximately 50% of the mixture was
unconjugated h1F6. Based on these results we concluded that
the use of endogenous interchain disulfides for conjugation of
this PBD-linker was not a viable approach for ADC formation.
In order to produce uniform ADCs with low aggregation
levels, we used recombinant technologies to introduce new
mAb cysteines for site-specific drug attachment. Cysteines were
introduced at the 239 position on the mAb heavy chains, since
this is a solvent accessible position28 and was empirically
proven to react with drug-maleimide derivatives. The h1F6239C
construct was expressed in Chinese hamster ovary cells and was
applied to the PBD-linker.
Biological Evaluation of PBD Dimer and h1F6239C
-
PBD. The initial in vitro evaluation of the PBD dimer small
molecule established that the drug was highly potent on a panel
of carcinoma and lymphoma cell lines including RCC (786-O,
UM-RC-3, ACHN, and Caki-1), Hodgkin (KMH2), and NHL
(Raji, MHH-PREB-1, WSU-DLCL2) (Table 1). The RCC
lines, 786-O and UM-RC-3, are known to express the MDR-1
(P-glycoprotein) protein and efflux rhodamine dye,29,30 yet
potent activity was observed with the PBD dimer. The IC50
values ranged from 0.1 to 1.0 nM, consistent with the literature
for this potent class of DNA-reactive agents.17 Incorporation of
the para-aniline residue onto one of the C-2 aryl groups did not
lead to apparent loss in drug potency.
The process for conjugating the PBD-linker to h1F6239C is
illustrated in Figure 2. The h1F6239C mAb, isolated from cell
culture as cysteine disulfides, was treated with excess TCEP.
This fully reduced both the position-239 cysteine disulfides and
all interchain disulfide bonds. The mAb was then partially
The cytotoxic activity of h1F6239C-PBD was determined
using the same panel of CD70-positive RCC and lymphoma
1259
dx.doi.org/10.1021/bc400217g | Bioconjugate Chem. 2013, 24, 1256−1263

Bioconjugate Chemistry
Article
lines. Treatment was for 96 h, and IC50 values ranged between
0.2 and 0.8 nM ADC. These values are comparable to the free
PBD dimer IC50 values and well below the antigen saturation
concentration of 7 nM, or 1 μg/mL ADC. In addition, the
effects were immunologically specific, as a nonbinding control
ADC was ∼1/1000th as potent as h1F6239C-PBD (Figure 4).
antibody that recognized the PBD-linker component. In this
experiment, the data for the ADC were compared to h1F6239C
conjugated to N-ethylmaleimide (NEM). This allowed us to
determine the impact the PBD-linker had on ADC clearance.
Mice were injected with a single dose (0.3 or 1.0 mg/kg) of
h1F6239C-PBD or h1F6239C-NEM, and blood samples were
collected at set time points. We found a similar clearance profile
for both the h1F6239C-NEM and h1F6239C-PBD. For example, at
the 0.3 mg/kg dose, the half-lives were 14.0 and 12.3 days,
respectively. The half-life of conjugated PBD, as determined by
the antidrug ELISA, was 8.0 and 8.5 days when dosed at 1.0 and
0.3 mg/kg, respectively. The difference in the mAb ELISA and
drug ELISA half-life measurements may be due to proteolytic
loss of drug from the ADC, or to a modification of the
conjugated PBD that is not recognized by the detection
antibody.
In vivo activity studies were conducted in disseminated
(ACHN RCC and Raji NHL) and subcutaneous (Caki-1 RCC
and MHH-PREB-1 NHL) mouse xenografts (Figure 6). Each
model was CD70-positive and the cell lines were sensitive to
the h1F6239C-PBD in vitro at sub-nM concentrations (Table 1).
In vivo immunologic specificity was assessed by giving
equivalent doses of the nonbinding control hIgG239C-PBD. In
the ACHN model, a cure rate of 100% (10/10 mice) was
obtained with two weekly doses of 0.1 mg/kg of h1F6239C-PBD.
The effects were immunologically specific, since the nonbinding
control group had no survivors at day 75 (Figure 6A). There
was also a distinct survival advantage with immunological
specificity demonstrated in the Raji lymphoma model (Figure
6B) with single doses of 0.1 and 0.3 mg/kg ADC. Using two
weekly 0.3 mg/kg doses in nude mice bearing Caki-1 tumors
resulted initially in 4/5 complete responses based on the
absence of palpable tumor in these mice, with 3/5 cures at the
end of the experiment (Figure 6C). The specificity controls had
minimal effect. Finally, the MHH-PREB-1 NHL model in
SCID mice was very sensitive to h1F6239C-PBD with 100%
cures (7/7 mice) with two weekly 0.1 mg/kg doses (Figure
6D), whereas the hIgG239C-PBD control at the same dose was
nearly devoid of activity. In all the models, h1F6239C-PBD
showed potent antitumor activities and high degrees of
immunologic specificity at well-tolerated doses.
Figure 4. In vitro immunologic specificity of PBD ADCs on CD70-
positive Caki-1 renal cell carcinoma line. Nonbinding hIgG239C-PBD
showed markedly reduced cytotoxic activity (∼1000-fold) compared
to h1F6239C-PBD which binds the CD70 antigen.
We next turned our attention to evaluating the rodent
toxicology, stability, pharmacokinetic (PK), and in vivo
activities of the h1F6239C-PBD conjugate. A single-dose
toxicology study using BALB/c mice demonstrated that
h1F6239C-PBD was well tolerated at 2.5 mg/kg with no weight
loss relative to untreated mice (data not shown). At a single 5
mg/kg dose, bone marrow toxicity occurred, which may be a
clinically relevant finding.
As an initial evaluation of linker stability, we incubated the
ADC in rat plasma for a period of 6 days and recovered the
ADC using a resin that bound serum IgGs. Since no loss of
drug from the ADC was found by PLRP analysis (Supporting
Information Figure S2), we concluded that the linker was stable
in plasma, and that no maleimide transfer to albumin took place
during the observation period. Drug release could be achieved
by treatment of a PBD ADC with the lysosomal protease
cathepsin B, suggesting that this enzyme may be involved in the
intracellular release of the active drug.
We then evaluated the PK of h1F6239C-PBD in mouse using
two different ELISA methods (Figure 5). The h1F6 component
of the ADC in circulation was assessed using an antihuman IgG
reagent, and conjugated PBD drug was measured using an
DISCUSSION
■
PBD dimers constitute a new drug class and cytotoxic
mechanism for ADCs and complement the current technolo-
Figure 5. Pharmacokinetic measurements of h1F6239C-PBD and h1F6239C-NEM conjugates. A. Measurement of h1F6 antibody (mAb) and PBD
drug (Drug) concentrations in mouse plasma after administration of a single 0.3 mg/kg dose of h1F6239C-PBD or h1F239C-NEM conjugate. B.
Measurements of a single 1.0 mg/kg dose of h1F6239C-PBD or h1F6239C-NEM conjugate.
1260
dx.doi.org/10.1021/bc400217g | Bioconjugate Chem. 2013, 24, 1256−1263

Bioconjugate Chemistry
Article
Figure 6. In vivo antitumor activity in xenograft models of RCC and NHL. A. ACHN, and B. Raji lymphoma disseminated models. Female SCID
mice (n = 10) were injected intraperitoneally with ACHN cells (106) or intravenously Raji lymphoma cells (5 × 106). The h1F6239C-PBD and
hIgG239C-PBD ADCs were dosed ip according to schedule at 0.1 or 0.3 mg/kg (arrows). The mice were monitored daily and were terminated with
symptoms of metastatic disease. C. Caki-1, and D. MHH-PREB-1 subcutaneous models. Caki-1 tumor fragments or MHH-PREB-1 cells (5 × 106)
were implanted in the right flank of female nude mice. Mice were randomized to study groups (n = 5 for Caki-1, and n = 7 for MHH-PREB-1) and
treated when a 100 mm3 tumor was reached. The h1F6239C-PBD and hIgG239C-PBD ADCs were dosed ip according to a q7dx2 schedule at 0.1 or 0.3
mg/kg (arrows). Tumor volume as a function of time was determined using the formula (L × W2)/2.
gies based on auristatin and maytansine antimitotic agents.
They react with double-stranded DNA to form interchain DNA
cross-links21−23 that can elude common DNA repair mecha-
nisms,17 leading to potent cytotoxic activities. The highly
potent PBD dimer used here is less sensitive to the P-
glycoprotein resistance mechanism than other less potent ADC
classes,31 and other PBD analogs.32,33 Both the PBD dimer free
drug and conjugate were highly active against cancer cell lines
that express P-glycoprotein and efflux rhodamine dye (Table
1). Since high P-glycoprotein expression is a significant negative
prognostic indicator for RCC patients34,35 as well as several
other malignancies, a PBD ADC may have applications in
settings where other drug classes are ineffective.
Standard conjugation methods, such as conjugation through
antibody native hinge cysteine or lysine residues, afforded
variably loaded species36,37 which led to aggregation. In the case
of the hydrophobic PBD-linker, conjugation via hinge cysteine
residues failed to provide monomeric ADCs, and as a result,
this method was abandoned. Site-specific conjugation to
engineered cysteine residues, such as at position 239 of the
antibody heavy chain, was vital for preparing PBD ADCs that
were monomeric and well-defined. The site-specific introduc-
tion of cysteine residues for conjugation can offer other
advantages such as high serum stability; however, stability does
vary with mutation site.38 In this example, when incubated in
rat plasma for 6 days, essentially no drug loss from hIgG239C-
PBD was observed. This is comparable to data for one other
engineered cysteine-based ADC, but superior to others.39 The
stability is higher than that shown for endogenous hinge
cysteine conjugates.12 The ADC and the NEM-modified mAb
had the same PK profiles demonstrating a negligible impact on
clearance by PBD-linker conjugation.
The activity of h1F6239C-PBD reflects the potency of the
PBD dimer, and is generally greater than ADCs employing
other drug classes.13,40 In the disseminated ACHN RCC and
subcutaneous MHH-PREB-1 NHL xenografts, all of the mice
were cured with two weekly 0.1 mg/kg doses of ADC. It is
noteworthy that a 0.1 mg/kg dose of h1F6239C-PBD carries 1
μg/kg of PBD dimer drug. By way of comparison, it is common
for other classes to be dosed at 10- to 100-fold higher
levels.11,13,40 The treatment of solid tumors, in which exposure
to the cytotoxic drug may be hampered by ADC penetration,
could benefit with higher potency, as less drug would be
required to have an antitumor effect.
With the approval of ADCETRIS (brentuximab vedotin) and
KADCYLA (ado-trastuzumab emtansine), ADCs are emerging
as effective therapies for treating cancer. The research
presented here, in the context of an anti-CD70 therapy,
represents several advances in ADC technology. These include
a highly cytotoxic PBD dimer with a novel DNA-based
1261
dx.doi.org/10.1021/bc400217g | Bioconjugate Chem. 2013, 24, 1256−1263

Bioconjugate Chemistry
Article
drug loading on the antitumor activity of a monoclonal antibody drug
conjugate. Clin. Cancer. Res. 10, 7063−70.
mechanism, a site-specific, well-defined conjugation method,
stable conjugation chemistry, long circulating half-life, and
broad and pronounced activity in mouse xenografts, including
MDR-positive models, at well-tolerated doses. We expect that
this technology will extend the scope of ADCs to include new
cancer targets. While many important issues remain to be
studied, including a detailed investigation of toxicology and the
impact improved ADC stability has on tolerability and potency,
the data presented here for h1F6239C-PBD indicate that this
new technology warrants further investigation for cancer
therapy.19
(11) Doronina, S. O., Toki, B. E., Torgov, M. Y., Mendelsohn, B. A.,
Cerveny, C. G., Chace, D. F., DeBlanc, R. L., Gearing, R. P., Bovee, T.
D., Siegall, C. B., Francisco, J. A., Wahl, A. F., Meyer, D. L., and Senter,
P. D. (2003) Development of potent monoclonal antibody auristatin
conjugates for cancer therapy. Nat. Biotechnol. 21, 778−84.
(12) Alley, S. C., Benjamin, D. R., Jeffrey, S. C., Okeley, N. M.,
Meyer, D. L., Sanderson, R. J., and Senter, P. D. (2008) Contribution
of linker stability to the activities of anticancer immunoconjugates.
Bioconjugate Chem. 19, 759−65.
(13) Doronina, S. O., Mendelsohn, B. A., Bovee, T. D., Cerveny, C.
G., Alley, S. C., Meyer, D. L., Oflazoglu, E., Toki, B. E., Sanderson, R.
J., Zabinski, R. F., Wahl, A. F., and Senter, P. D. (2006) Enhanced
activity of monomethylauristatin F through monoclonal antibody
delivery: effects of linker technology on efficacy and toxicity.
Bioconjugate Chem. 17, 114−24.
(14) Junutula, J. R., Flagella, K. M., Graham, R. A., Parsons, K. L., Ha,
E., Raab, H., Bhakta, S., Nguyen, T., Dugger, D. L., Li, G., Mai, E.,
Lewis Phillips, G. D., Hiraragi, H., Fuji, R. N., Tibbitts, J., Vandlen, R.,
Spencer, S. D., Scheller, R. H., Polakis, P., and Sliwkowski, M. X.
(2010) Engineered thio-trastuzumab-DM1 conjugate with an
improved therapeutic index to target human epidermal growth factor
receptor 2-positive breast cancer. Clin. Cancer. Res. 16, 4769−78.
(15) Grewal, I. S. (2008) CD70 as a therapeutic target in human
malignancies. Expert Opin. Ther. Targets 12, 341−51.
ASSOCIATED CONTENT
* Supporting Information
Size exclusion chromatograms and PLRP analysis. This material
is available free of charge via the Internet at http://pubs.acs.org.
■
S
AUTHOR INFORMATION
Corresponding Author
*Tel. 425 527-4738. Fax. 425 527-4901. E-mail: sjeffrey@
seagen.com.
■
Notes
The authors declare the following competing financial
interest(s): With the exception of DET, all authors are
employed by either Seattle Genetics, Inc., or Spirogen Ltd.
(16) Boursalian, T. E., McEarchern, J. A., Law, C. L., and Grewal, I. S.
(2009) Targeting CD70 for human therapeutic use. Adv. Exp. Med.
Biol. 647, 108−19.
(17) Hartley, J. A., Hamaguchi, A., Coffils, M., Martin, C. R., Suggitt,
M., Chen, Z., Gregson, S. J., Masterson, L. A., Tiberghien, A. C.,
Hartley, J. M., Pepper, C., Lin, T. T., Fegan, C., Thurston, D. E., and
Howard, P. W. (2010) SG2285, a novel C2-aryl-substituted
pyrrolobenzodiazepine dimer prodrug that cross-links DNA and exerts
highly potent antitumor activity. Cancer Res. 70, 6849−58.
(18) Howard, P. W., Chen, Z., Gregson, S. J., Masterson, L. A.,
Tiberghien, A. C., Cooper, N., Fang, M., Coffils, M. J., Klee, S.,
Hartley, J. A., and Thurston, D. E. (2009) Synthesis of a novel C2/
C2′-aryl-substituted pyrrolo[2,1-c][1,4]benzodiazepine dimer prodrug
with improved water solubility and reduced DNA reaction rate. Bioorg.
Med. Chem. Lett. 19, 6463−6.
(19) Sutherland, M. S., Walter, R. B., Jeffrey, S. C., Burke, P. J., Yu,
C., Stone, I., Ryan, M. C., Sussman, D., Zeng, W., Harrington, K. H.,
Klussman, K., Westendorf, L., Meyer, D., Bernstein, I. D., Senter, P. D.,
Drachman, J. G., and McEarchern, J. A. (In Press) SGN-CD33A: A
novel CD33-targeting antibody-drug conjugate utilizing a pyrroloben-
zodiazepine dimer is active in models of drug-resistant AML.
(20) Howard, P. W., Gregson, S. J., Masterson, L. (2010)
Unsymmetrical pyrrolobenzodiazepine-dimers for treatment of pro-
liferative diseases. WO2010/43880 A1.
(21) Rahman, K. M., Thompson, A. S., James, C. H.,
Narayanaswamy, M., and Thurston, D. E. (2009) The pyrrolobenzo-
diazepine dimer SJG-136 forms sequence-dependent intrastrand DNA
cross-links and monoalkylated adducts in addition to interstrand cross-
links. J. Am. Chem. Soc. 131, 13756−66.
(22) Rahman, K. M., H., V., James, C. H., Howard, P. W., and
Thurston, D. E. (2010) DNA sequence preference and adduct
orientation of pyrrolo[2,1-c][1,4]benzodiazepine antitumor agents.
ACS Med. Chem. Lett. 1, 427−32.
REFERENCES
■
(1) Senter, P. D., and Sievers, E. L. (2012) The discovery and
development of brentuximab vedotin for use in relapsed Hodgkin
lymphoma and systemic anaplastic large cell lymphoma. Nat.
Biotechnol. 30, 631−7.
(2) Senter, P. D. (2009) Potent antibody drug conjugates for cancer
therapy. Curr. Opin. Chem. Biol. 13, 235−44.
(3) Sievers, E. L., and Senter, P. D. (2013) Antibody-drug conjugates
in cancer therapy. Annu. Rev. Med. 64, 15−29.
(4) Burke, P. J., Senter, P. D., Meyer, D. W., Miyamoto, J. B.,
Anderson, M., Toki, B. E., Manikumar, G., Wani, M. C., Kroll, D. J.,
and Jeffrey, S. C. (2009) Design, synthesis, and biological evaluation of
antibody-drug conjugates comprised of potent camptothecin ana-
logues. Bioconjugate Chem. 20, 1242−50.
(5) Burke, P. J., Toki, B. E., Meyer, D. W., Miyamoto, J. B., Kissler, K.
M., Anderson, M., Senter, P. D., and Jeffrey, S. C. (2009) Novel
immunoconjugates comprised of streptonigrin and 17-amino-geldana-
mycin attached via a dipeptide-p-aminobenzyl-amine linker system.
Bioorg. Med. Chem. Lett. 19, 2650−3.
(6) Jeffrey, S. C., Nguyen, M. T., Andreyka, J. B., Meyer, D. L.,
Doronina, S. O., and Senter, P. D. (2006) Dipeptide-based highly
potent doxorubicin antibody conjugates. Bioorg. Med. Chem. Lett. 16,
358−62.
(7) Jeffrey, S. C., Torgov, M. Y., Andreyka, J. B., Boddington, L.,
Cerveny, C. G., Denny, W. A., Gordon, K. A., Gustin, D., Haugen, J.,
Kline, T., Nguyen, M. T., and Senter, P. D. (2005) Design, synthesis,
and in vitro evaluation of dipeptide-based antibody minor groove
binder conjugates. J. Med. Chem. 48, 1344−58.
(8) Doronina, S. O., Bovee, T. D., Meyer, D. W., Miyamoto, J. B.,
Anderson, M. E., Morris-Tilden, C. A., and Senter, P. D. (2008) Novel
peptide linkers for highly potent antibody-auristatin conjugate.
Bioconjugate Chem. 19, 1960−3.
(9) Jeffrey, S. C., Andreyka, J. B., Bernhardt, S. X., Kissler, K. M.,
Kline, T., Lenox, J. S., Moser, R. F., Nguyen, M. T., Okeley, N. M.,
Stone, I. J., Zhang, X., and Senter, P. D. (2006) Development and
properties of beta-glucuronide linkers for monoclonal antibody-drug
conjugates. Bioconjugate Chem. 17, 831−40.
(10) Hamblett, K. J., Senter, P. D., Chace, D. F., Sun, M. M., Lenox,
J., Cerveny, C. G., Kissler, K. M., Bernhardt, S. X., Kopcha, A. K.,
Zabinski, R. F., Meyer, D. L., and Francisco, J. A. (2004) Effects of
(23) Rahman, K. M., James, C. H., and Thurston, D. E. (2011) Effect
of base sequence on the DNA cross-linking properties of
pyrrolobenzodiazepine (PBD) dimers. Nucleic Acids Res. 39, 5800−12.
(24) Antonow, D., Kaliszczak, M., Kang, G. D., Coffils, M.,
Tiberghien, A. C., Cooper, N., Barata, T., Heidelberger, S., James, C.
H., Zloh, M., Jenkins, T. C., Reszka, A. P., Neidle, S., Guichard, S. M.,
Jodrell, D. I., Hartley, J. A., Howard, P. W., and Thurston, D. E. (2010)
Structure-activity relationships of monomeric C2-aryl pyrrolo[2,1-
c][1,4]benzodiazepine (PBD) antitumor agents. J. Med. Chem. 53,
2927−41.
1262
dx.doi.org/10.1021/bc400217g | Bioconjugate Chem. 2013, 24, 1256−1263

Bioconjugate Chemistry
Article
(25) Law, C. L., Gordon, K. A., Toki, B. E., Yamane, A. K., Hering,
M. A., Cerveny, C. G., Petroziello, J. M., Ryan, M. C., Smith, L., Simon,
R., Sauter, G., Oflazoglu, E., Doronina, S. O., Meyer, D. L., Francisco,
J. A., Carter, P., Senter, P. D., Copland, J. A., Wood, C. G., and Wahl,
A. F. (2006) Lymphocyte activation antigen CD70 expressed by renal
cell carcinoma is a potential therapeutic target for anti-CD70 antibody-
drug conjugates. Cancer Res. 66, 2328−37.
(26) McEarchern, J. A., Smith, L. M., McDonagh, C. F., Klussman, K.,
Gordon, K. A., Morris-Tilden, C. A., Duniho, S., Ryan, M., Boursalian,
T. E., Carter, P. J., Grewal, I. S., and Law, C. L. (2008) Preclinical
characterization of SGN-70, a humanized antibody directed against
CD70. Clin. Cancer. Res. 14, 7763−72.
S. D., Lee Wong, W., Lowman, H. B., Vandlen, R., Sliwkowski, M. X.,
Scheller, R. H., Polakis, P., and Mallet, W. (2008) Site-specific
conjugation of a cytotoxic drug to an antibody improves the
therapeutic index. Nat. Biotechnol. 26, 925−32.
(40) Tassone, P., Gozzini, A., Goldmacher, V., Shammas, M. A.,
Whiteman, K. R., Carrasco, D. R., Li, C., Allam, C. K., Venuta, S.,
Anderson, K. C., and Munshi, N. C. (2004) In vitro and in vivo activity
of the maytansinoid immunoconjugate huN901-N2′-deacetyl-N2′-(3-
mercapto-1-oxopropyl)-maytansine against CD56+ multiple myeloma
cells. Cancer Res. 64, 4629−36.
(27) Sun, M. M., Beam, K. S., Cerveny, C. G., Hamblett, K. J.,
Blackmore, R. S., Torgov, M. Y., Handley, F. G., Ihle, N. C., Senter, P.
D., and Alley, S. C. (2005) Reduction-alkylation strategies for the
modification of specific monoclonal antibody disulfides. Bioconjugate
Chem. 16, 1282−90.
(28) Fraczkiewicz, R., and Braun, W. (1998) Exact and efficient
analytical calculation of the accessible surface areas and their gradients
for macromolecules. J. Comput. Chem. 19, 319−33.
(29) Lamy, T., Drenou, B., Grulois, I., Fardel, O., Jacquelinet, C.,
Goasguen, J., Dauriac, C., Amiot, L., Bernard, M., and Fauchet, R.
(1995) Multi-drug resistance (MDR) activity in acute leukemia
determined by rhodamine 123 efflux assay. Leukemia 9, 1549−55.
(30) Kawabata, M., Kobayashi, H., Mori, S., Sekiguchi, S., and
Takemura, Y. (1997) Flow cytometric analysis of P-glycoprotein
function by rhodamine 123 dye-efflux assay in human leukemia cells.
Rinsho Byori 45, 891−8.
(31) Jawad, M., Seedhouse, C., Mony, U., Grundy, M., Russell, N. H.,
and Pallis, M. (2010) Analysis of factors that affect in vitro
chemosensitivity of leukaemic stem and progenitor cells to
gemtuzumab ozogamicin (Mylotarg) in acute myeloid leukaemia.
Leukemia 24, 74−80.
(32) Aird, R. E., Thomson, M., Macpherson, J. S., Thurston, D. E.,
Jodrell, D. I., and Guichard, S. M. (2008) ABCB1 genetic
polymorphism influences the pharmacology of the new pyrrolobenzo-
diazepine derivative SJG-136. Pharmacogenomics J. 8, 289−96.
(33) Guichard, S. M., Macpherson, J. S., Thurston, D. E., and Jodrell,
D. I. (2005) Influence of P-glycoprotein expression on in vitro
cytotoxicity and in vivo antitumour activity of the novel
pyrrolobenzodiazepine dimer SJG-136. Eur. J. Cancer 41, 1811−8.
(34) Yu, D. S., Chang, S. Y., and Ma, C. P. (1998) The expression of
MDR-1-related GP-170 and its correlation with anthracycline
resistance in renal cell carcinoma cell lines and multidrug-resistant
sublines. Br. J. Urol. 82, 544−7.
(35) Walsh, N., Larkin, A., Kennedy, S., Connolly, L., Ballot, J., Ooi,
W., Gullo, G., Crown, J., Clynes, M., and O’Driscoll, L. (2009)
Expression of multidrug resistance markers ABCB1 (MDR-1/P-gp)
and ABCC1 (MRP-1) in renal cell carcinoma. BMC Urol. 9, 6.
(36) Lyon, R. P., Meyer, D. L., Setter, J. R., and Senter, P. D. (2012)
Conjugation of anticancer drugs through endogenous monoclonal
antibody cysteine residues. Methods Enzymol. 502, 123−38.
(37) Fleming, M. S., Zhang, W., Lambert, J. M., and Amphlett, G.
(2005) A reversed-phase high-performance liquid chromatography
method for analysis of monoclonal antibody-maytansinoid immuno-
conjugates. Anal. Biochem. 340, 272−8.
(38) Shen, B. Q., Xu, K., Liu, L., Raab, H., Bhakta, S., Kenrick, M.,
Parsons-Reponte, K. L., Tien, J., Yu, S. F., Mai, E., Li, D., Tibbitts, J.,
Baudys, J., Saad, O. M., Scales, S. J., McDonald, P. J., Hass, P. E.,
Eigenbrot, C., Nguyen, T., Solis, W. A., Fuji, R. N., Flagella, K. M.,
Patel, D., Spencer, S. D., Khawli, L. A., Ebens, A., Wong, W. L.,
Vandlen, R., Kaur, S., Sliwkowski, M. X., Scheller, R. H., Polakis, P.,
and Junutula, J. R. (2012) Conjugation site modulates the in vivo
stability and therapeutic activity of antibody-drug conjugates. Nat.
Biotechnol. 30, 184−9.
(39) Junutula, J. R., Raab, H., Clark, S., Bhakta, S., Leipold, D. D.,
Weir, S., Chen, Y., Simpson, M., Tsai, S. P., Dennis, M. S., Lu, Y.,
Meng, Y. G., Ng, C., Yang, J., Lee, C. C., Duenas, E., Gorrell, J., Katta,
V., Kim, A., McDorman, K., Flagella, K., Venook, R., Ross, S., Spencer,
1263
dx.doi.org/10.1021/bc400217g | Bioconjugate Chem. 2013, 24, 1256−1263