Trans-Cyclooctene Tag with Improved Properties for Tumor

Jul 31, 2014 - CC49–TCO(2) was administered to healthy mice to determine its blood clearance and the in vivo stability of the TCO. Next, pretargetin...
40 downloads 7 Views 1MB Size
Download PDF
Article pubs. acs.

Trans-Cyclooctene Tag with Improved Properties for Tumor Pretargeting with the Diels−Alder Reaction Raffaella Rossin,†,‡ Sander M. J. van Duijnhoven,† Tilman Lap̈ pchen,‡ Sandra M. van den Bosch,‡ and Marc S. Robillard*,† †

Tagworks Pharmaceuticals, High Tech Campus 11, 5656 AE Eindhoven, The Netherlands Department of Minimally Invasive Healthcare, Philips Research, 5656 AE Eindhoven, The Netherlands

‡

S Supporting Information *

ABSTRACT: Radioimmunotherapy (RIT) of solid tumors is hampered by low tumor-to-nontumor (T/NT) ratios of the radiolabeled monoclonal antibodies resulting in low tumor doses in patients. Pretargeting technologies can improve the effectiveness of RIT in cancer therapy by increasing this ratio. We showed that a pretargeting strategy employing in vivo chemistry in combination with clearing agents, proceeds efficiently in tumor-bearing mice resulting in high T/NT ratios. A dosimetry study indicated that the chemical pretargeting technology, which centered on the bioorthogonal Diels−Alder click reaction between a radiolabeled tetrazine probe and a trans-cyclooctene-oxymethylbenzamide-tagged CC49 antibody (CC49−TCO(1)), can match the performance of clinically validated high-affinity biological pretargeting approaches in mice (Rossin et al. J Nucl Med. 2013, 54, 1989−1995). Nevertheless, the increased protein surface hydrophobicity of CC49−TCO(1) led to a relatively rapid blood clearance and concomitant reduced tumor uptake compared to native CC49 antibody. Here, we present the in vivo evaluation of a TCO-oxymethylacetamide-tagged CC49 antibody (CC49−TCO(2)), which is highly reactive toward tetrazines and less hydrophobic than CC49−TCO(1). CC49−TCO(2) was administered to healthy mice to determine its blood clearance and the in vivo stability of the TCO. Next, pretargeting biodistribution and SPECT studies with CC49−TCO(2), tetrazine-functionalized clearing agent, and radiolabeled tetrazine were carried out in nude mice bearing colon carcinoma xenografts (LS174T). CC49−TCO(2) had an increased circulation half-life, a 1.5-fold higher tumor uptake, and a 2.6-fold improved in vivo TCO stability compared to the more hydrophobic TCO-benzamide−CC49. As a consequence, and despite the 2-fold lower reactivity of CC49−TCO(2) toward tetrazines compared with CC49−TCO(1), administration of radiolabeled tetrazine afforded a significantly increased tumor accumulation and improved T/NT ratios in mice pretargeted with CC49−TCO(2). In conclusion, the TCO-acetamide derivative represents a large improvement in in vivo Diels−Alder pretargeting, possibly enabling application in larger animals and eventually humans. KEYWORDS: Diels−Alder, pretargeting, trans-cyclooctene, tetrazine, 177Lu, blood clearance

■

INTRODUCTION

ward tagging of mAbs with minimal perturbation and are less likely to give rise to immunogenicity (Figure 1). We and others demonstrated that the bioorthogonal inverse-electron-demand Diels−Alder reaction between strained trans-cyclooctene (TCO) and electron-deficient tetrazines can be applied in pretargeted radioimmunoimaging (Figure 1),7−9 whereas other bioorthogonal chemical reactions turned out to be less promising in this respect.10−12 Since then, the Diels−Alder pretargeting system has been improved. We demonstrated that axially linked TCO tags are more reactive than their equatorial isomers and that their in

For most tumors, the use of radioimmunotherapy (RIT) is hampered by low tumor-to-blood ratios of the radiolabeled monoclonal antibodies (mAbs), resulting in low tumor doses in patients.1,2 Pretargeted RIT has the potential to overcome this problem. It is based on the tumor-binding of a tagged mAb and the subsequent binding of a small fast-clearing radiolabeled molecule to the prelocalized mAb, resulting in superior tumorto-nontumor (T/NT) ratios.3 Current clinically validated pretargeting systems use noncovalent biological interactions (streptavidin−biotin, antibody−hapten) for the recruitment of the radiolabeled probe to the tumor-bound mAb,4 but these systems can give rise to immunogenicity5 and/or involve extensive reengineering and perturbation of the parent mAb. Bioorthogonal chemical functionalities6 may be an alternative to biological pretargeting components: they enable straightfor© 2014 American Chemical Society

Received: Revised: Accepted: Published: 3090

April 15, 2014 July 21, 2014 July 31, 2014 July 31, 2014 dx.doi.org/10.1021/mp500275a | Mol. Pharmaceutics 2014, 11, 3090−3096

Molecular Pharmaceutics

Article

TCO tagged CC49 (CC49−TCO(2); Figure 2),13 comprising a less hydrophobic oxymethylacetamide linker moiety (Supporting Information, Figure 1), which was expected to result in a reduced protein surface hydrophobicity. Although CC49− TCO(2) is 2-fold less reactive, k2 = 13.5 ± 0.1 × 104 M−1 s−1, toward tetrazine than CC49−TCO(1),13 the lower reactivity of CC49−TCO(2) may contribute to an increased in vivo stability of the TCO tag. We therefore hypothesized that the tumor uptake of radiolabeled tetrazine would be increased for animals pretargeted with CC49−TCO(2), as a result of increased tumor uptake of CC49−TCO(2) and improved tag stability. In this paper, we report the in vivo application of CC49−TCO(2) and compare the results with the previously reported data for CC49−TCO(1).

Figure 1. Tumor pretargeting with the inverse-electron-demand Diels−Alder reaction involving a trans-cyclooctene-modified mAb and a radiolabeled tetrazine.

vivo stability can be improved by reducing the linker length between mAb and tag.13 We furthermore optimized the pretargeting protocol by administering a tetrazine-functionalized clearing agent, which enables rapid reaction with and removal of freely circulating TCO-tagged antibodies from blood prior to injection of the radiolabeled probe.14 The clearing agent, an albumin scaffold comprising multiple tetrazine and galactose residues, and its reaction products with TCO-tagged mAbs are rapidly cleared from blood by the liver via the galactose residues. Combined, this led to a doubling of the tetrazine tumor uptake and a 125-fold improvement of the tumor-to-blood ratio at 3 h post injection, resulting in a predicted 8-fold higher total tumor dose compared with nonpretargeted RIT.14 In these studies, we employed the TAG-72 targeting mAb CC49 functionalized with TCO tags via hydrophobic oxymethylbenzamide linkages (CC49−TCO(1), Figure 2). This construct had a reduced circulation time and tumor uptake compared with unmodified CC49, presumably due to the increased protein surface hydrophobicity. We expected that the use of a more hydrophilic tag would reduce the modest perturbation of the properties of the parent mAb, maintaining its slow blood clearance and high tumor uptake.15,16 Recently, we developed an axially linked

■

MATERIALS AND METHODS General. All reagents and solvents were obtained from commercial sources (Sigma-Aldrich) and used without further purification unless stated otherwise. [111In]Indium chloride, [177Lu]lutetium chloride, and sodium [125I]iodide solutions were purchased from PerkinElmer. Water was distilled and deionized (18 MΩcm) by means of a Milli-Q water filtration system (Millipore). The labeling buffers were treated with Chelex-100 resin (BioRad Laboratories) overnight, filtered through 0.22 μm, and stored at 4 °C. The Bolton−Hunter reagent (N-succinimidyl-3-[4-hydroxyphenyl]propionate, SHPP), gelcode blue protein staining solutions, and Zeba desalting spin columns (40 kDa MW cutoff, 0.5−2 mL) were purchased from Pierce Protein Research (Thermo Fisher Scientific). Amicon Ultra-15 centrifugal filter units (30 and 50 kDa MW cutoff) were purchased from Millipore. Mouse serum was purchased from Innovative Research. The 111In- and 177Lu-labeling yields for tetrazine 3 were determined by radio-TLC, using ITLC-SG strips (Varian Inc.) eluted with 200 mM EDTA in saline and imaged on a phosphor imager (FLA-7000, Fujifilm). In these conditions, free 111In and 177 Lu migrate with Rf = 0.9, while 111In/177Lu-tetrazine remains at the origin. The radiochemical purity of the 111In- and 177Lulabeled tetrazine 3 was determined by radio-HPLC on an Agilent 1100 system equipped with a Gabi radioactive detector (Raytest). The samples were loaded on an Agilent Eclipse XDB-C18 column (4.6 mm × 150 mm, 5 μm), which was eluted at 1 mL/min with a linear gradient of acetonitrile in water containing 0.1% TFA (2 min at 10% acetonitrile followed by an increase to 45% acetonitrile in 11 min). The UV wavelength was preset at 254 nm. The 125I-mAb labeling yields were determined with radio-TLC, using ITLC-SG strips eluted with a 1:1 methanol/ethyl acetate mixture and imaged on a phosphor imager. In these conditions, free [125I]iodide and 125ISHPP migrate with Rf = 0.5−0.9, while 125I-mAbs remain at the origin. The radiochemical purity of 125I-CC49−TCO(2) and 125 I-CC49 were determined by size exclusion chromatography (SEC) and SDS-PAGE. SEC was carried out on an Agilent 1200 system equipped with a Gabi radioactive detector. The samples were loaded on a BioSep-SEC-S 3000 column (300 mm × 7.8 mm, 5 μm particles, Phenomenex) and eluted with PBS pH 7.4 at 1 mL/min. The UV wavelength was preset at 260 and 280 nm. IEF analysis and SDS-PAGE was performed on a Phastgel system using IEF-3−9 gels and 7.5% PAGE homogeneous gels (GE Healthcare Life Sciences), respectively. The IEF calibration solution (broad pI, pH 3−10) was purchased from GE Healthcare, and the protein MW standard

Figure 2. Pretargeting components: schematic representation of CC49-conjugated TCO-oxymethylbenzamide (CC49−TCO(1)), CC49-conjugated TCO-oxymethylacetamide (CC49−TCO(2)), DOTA−tetrazine (3), and galactose-albumin−tetrazine (4). 3091

dx.doi.org/10.1021/mp500275a | Mol. Pharmaceutics 2014, 11, 3090−3096

Molecular Pharmaceutics

Article

the probe concentration (%ID/g) in blood at t = 0. The blood half-lives for the individual mice were used for statistical analysis between CC49, CC49−TCO(1),14 and CC49−TCO(2). In Vivo Stability of CC49−TCO(2). Three tumor-free mice were intravenously administered 125I-CC49−TCO(2) (average of eight TCOs/mAb, 300 μg/100 μL per mouse, ca. 0.8 MBq). At selected time points up to 4 days postinjection, blood samples (50−60 μL) were withdrawn from the vena saphena and transferred into vials containing heparin (5 μL). The blood samples were diluted to 100 μL with PBS and reacted with an excess of carrier-added 177Lu-tetrazine 3, which was radiolabeled shortly before each experimental time point at exactly 0.2 MBq/μg specific activity. The reaction mixtures were incubated for 20 min at 37 °C and then centrifuged for 5 min at 400g to separate the blood cells. Subsequently, 30 μL of the supernatants were eluted through a Zeba desalting spin column (0.5 mL, 40 kDa MW cutoff). Mixtures containing 177Lu-3 in serum/PBS were used to determine the breakthrough of nonmAb-bound 177Lu from the Zeba columns (in triplicate). The activity contained in the eluates was then measured in a gamma counter using a dual-isotope protocol. The 125I-activity was decay corrected to the injection time, and the 177Lu activity was corrected for 177Lu-3 breakthrough. The TCO in vivo stability was obtained by fitting the 177Lu/125I ratios plotted vs the time postinjection normalized to 100% at t = 0. The TCO half-lives for the individual mice were used for statistical analysis between CC49−TCO(1)13 and CC49−TCO(2). Pretargeting Biodistribution Studies. Four tumorbearing mice were injected with 125I-labeled CC49−TCO(2) (average of eight TCOs/mAb, 100 μg/100 μL per mouse; ca. 0.4 MBq), followed by two clearing agent doses (4) 30 and 48 h post-mAb injection (160 μg/100 μL per injection per mouse). Two hours after the last clearing agent injection, the mice were injected with 6.7 nmol 177Lu-3 (8.52 μg/80 μL per mouse containing 100 μg of gentisic acid; ca. 0.5 MBq) and were euthanized 3 h post-tetrazine injection, after which organs and tissues of interest were harvested and counted for radioactivity. Pretargeting SPECT/CT Studies. Two mice bearing LS174T xenografts were injected with CC49−TCO(2) (average of eight TCOs/mAb, 100 μg/100 μL per mouse), two doses of compound 4 (160 μg/100 μL per injection per mouse), and 111In-3 (6.7 nmol/80 μL per mouse containing 100 μg of gentisic acid, ca. 42 MBq) following the pretargeting protocol. Approximately 90 min post-tetrazine injection the mice were anesthetized with isoflurane and imaged on a dedicated small animal SPECT/CT system (NanoSPECT/CT, Bioscan) equipped with four detector heads and converging 9pinhole collimators (pinhole diameter: 1.4 mm). The CT scan (180 projections; 1000 ms per projection; 45 kV peak tube voltage; 177 mA tube current; 35 mm field of view) was followed by the SPECT scan (24 projections; 120 s per projection; photopeaks for 111In set to 171 keV (15% FW) and 245 keV (20% FW)). Three hours post-tetrazine injection, the mice were euthanized by anesthesia overdose and a high resolution SPECT/CT scan was performed (360 projections and 2000 ms per projection for CT; 36 projections and 300 s per projection for SPECT). The SPECT images were reconstructed using HiSPECTNG (SciVis GMBH) to an isotropic voxel size of 300 μm. The CT images were reconstructed using InVivoScope (Bioscan) to an isotropic voxel size of 200 μm. The post-mortem SPECT data were

solution (Precision Plus dual color standard) was purchased from BioRad. The radioactivity distribution on TLC plates and IEF/SDS-PAGE gels was evaluated with a phosphor imager (FLA-7000, Fujifilm) with the AIDA software (Raytest). Syntheses of Pretargeting Components. The synthesis of CC49−TCO(2), DOTA−tetrazine (3), and galactosealbumin−tetrazine (4), and CC49 production have been described elsewhere.7,13,14 Radiochemistry. The radiolabeling of DOTA−tetrazine (3) with 177Lu or 111In have been reported elsewhere.13 CC49 and CC49−TCO(2) were radio-iodinated with the Bolton− Hunter method because earlier experiments indicated that TCO can interfere to a certain extent with the Iodogen labeling method. The specific activity of 177Lu-3 used for biodistribution and that of 111In-3 used for SPECT/CT imaging was 0.07−0.15 MBq/nmol and ca. 6 MBq/nmol, respectively. The specific activity of 125I-CC49 and 125I-CC49−TCO(2) for animal experiments was adjusted to 2−4 kBq/μg by adding nonradioactive CC49 or CC49−TCO(2), respectively. The immunoreactivity of 125I-CC49−TCO(2) and 125I-CC49 were assessed by reacting 1 μg of mAb with 20 equiv of bovine submaxillary mucin (BSM) in a 1% bovine serum albumin solution in PBS for 30 min at 37 °C, followed by SEC analysis. Animal Experiments. All animal experiments were performed according to the principles of laboratory animal care (NIH publication 85-23, revised 1985) and the Dutch national law “Wet op de Dierproeven” (Stb 1985, 336). The in vivo experiments were performed in tumor-free or tumorbearing nude female Balb/C mice (20−25 g body weight, Charles River Laboratories). The human colon cancer cell line LS174T was obtained from the ATCC and maintained in Eagle’s Minimal Essential Medium (Sigma) supplemented with 10% heat inactivated fetal calf serum (Gibco), penicillin (100 U/mL), streptomycin (100 μg/mL), and 2 mM Glutamax. Mice were inoculated subcutaneously with 5 × 106 cells in 100 μL of sterile PBS and were used 7−10 days after tumor inoculation, when the tumors reached a size of approximately 70−200 mm3. At the end of each experiment, the mice were anesthetized and euthanized by cervical dislocation. Blood was withdrawn by heart puncture, and selected organs and tissues were harvested and blotted dry. All samples were weighed and then combined with 1 mL of PBS. The sample radioactivity was counted in a gamma counter (Wizard 1480, PerkinElmer) along with standards to determine the percentage injected dose per gram (%ID/g) and the percentage injected dose per organ (%ID/organ). The tissues from single-isotope experiments were measured using 10−80, 10−380, and 150−500 keV energy windows for 125I, 177Lu, and 111In, respectively. The tissues from dual-isotope experiments were measured using dual-isotope protocol (10−80 and 155−380 keV energy windows for 125 I and 177Lu, respectively) with crosscontamination correction. CC49 and CC49−TCO(2) Blood Clearance. Two groups of three tumor-free mice were injected intravenously with 125ICC49 and 125I-CC49−TCO(2) (average of eight TCOs/mAb, 100 μg/100 μL per mouse, ca. 0.4 MBq). The mice were serially bled at 5 or 60 min, 3 and 6 h, 1, 2, and 3 days, and euthanized 4 days post-mAb injection. Blood samples were counted for radioactivity. Blood clearance data was fitted with a two-phase exponential decay function, and the area under the curve (AUC) was determined using GraphPad Prism. Blood half-lives were calculated per group and per individual mouse using formula t1/2,AUC = ln(2) · AUC/C0, with C0 representing 3092

dx.doi.org/10.1021/mp500275a | Mol. Pharmaceutics 2014, 11, 3090−3096

Molecular Pharmaceutics

Article

quantitatively evaluated by drawing volume of interests (VOI) in tumor, kidneys, liver, and thigh muscle using InVivoScope. Data Analysis. The data are presented as mean %ID/g or % ID/organ ± one standard deviation (SD). Curve fitting, calculation of AUC, and two-tailed unpaired t tests were performed with GraphPad Prism version 4.1. The difference between two data points was considered statistically significant when p < 0.05.

■

RESULTS Antibody Functionalization and Radiochemistry. The TAG-72 targeting mAb CC49 was reacted with the axial isomer of TCO-oxymethylacetamide-NHS,13 which afforded an average of eight TCO groups per mAb. Radioiodination of CC49 and CC49−TCO(2) was carried out with the Bolton− Hunter reagent in PBS. The procedure afforded a 65−70% labeling yield and, after size exclusion purification, the radiolabeled mAbs had a greater than 98% radiochemical purity, as evidenced by radio-TLC and SDS-PAGE analysis (Supporting Information, Figures 2 and 3). TCO conjugation and 125I-labeling did not affect CC49 immunoreactivity, as confirmed in a binding assay with bovine submaxillary mucin (BSM) (Supporting Information, Figure 4). The DOTA− tetrazine 3 was labeled with 177Lu or 111In for biodistribution experiments and SPECT/CT imaging, respectively, giving >99% labeling yield and >95% radiochemical purity.13 Blood Clearance and in Vivo Stability. Next, the blood kinetics were determined for 125I-CC49 and 125I-CC49− TCO(2) (Figure 3). We observed a 1.6-fold increased blood

Figure 4. Normalized in vivo degradation of CC49−TCO(1) and CC49−TCO(2) in tumor-free mice (fitted with linear regression). Data are mean ± SD (n = 3). Data for CC49−TCO(1) were originally published in ref 13.

TCO(2), our investigations continued with pretargeting biodistribution studies of 125I-CC49−TCO(2) and 177Lutetrazine-3 in combination with two clearing agent doses (4) in LS174T tumor-bearing mice (Table 1). The two doses of clearing agent efficiently cleared the 125I-CC49−TCO(2) from blood to 0.31 ± 0.13%ID/g. We observed a high uptake of 125ICC49−TCO(2) in tumor, which was significantly higher than that found for 125I-CC49−TCO(1).14 Also liver, spleen, and kidney tissue showed a significantly higher accumulation of 125ICC49−TCO(2) compared with 125I-CC49−TCO(1) (p < 0.05). No significant 125I activity was found in stomachs and thyroids, indicating that 125I-CC49−TCO(2) was stable toward in vivo dehalogenation (Table 1). Following 177Lu-tetrazine-3 injection, a selective and effective on-tumor reaction (38% yield, based on TCO) was observed between 125I-CC49−TCO(2) and 177Lu-3, yielding a significantly higher tumor uptake of 177Lu-3 for mice pretargeted with 125 I-CC49−TCO(2) (Table 1) compared with the previously reported pretargeting results with 125I-CC49−TCO(1).14 In all other examined tissues, there was no significant difference between the uptake of 177Lu-3 pretargeted with 125I-CC49− TCO(2) or 125I-CC49−TCO(1). Accordingly, improved 177Lu3 T/NT ratios were observed for mice pretargeted with 125ICC49−TCO(2) compared with mice pretargeted with 125ICC49−TCO(1), varying from 1.1-fold for bladder and bone, 1.3-fold for kidney, 1.4-fold for liver, to 1.6-fold for spleen (Table 2). Furthermore, 177Lu-3 cleared rapidly from blood and all nonreacted probe was eliminated via the urine within 3 h. Corresponding SPECT/CT imaging experiments of mice injected with CC49−TCO(2), followed by two clearing steps and injection of 111In-tetrazine-3, confirmed the high tetrazine uptake in tumor and low retention in nontarget organs (Figure 5).

Figure 3. Blood clearance of 100 μg of 125I-CC49, 100 μg of 125ICC49−TCO(1) (9 equiv TCO-oxymethylbenzamide/mAb), and 100 μg of 125I-CC49−TCO(2) (8 equiv TCO-oxymethylacetamide/mAb) in healthy mice (fitted with a two-phase exponential decay function). Data are mean %ID/g ± SD (n = 3). Half-lives t1/2,AUC are 25.9 h for 125 I-CC49, 14.1 h for 125I-CC49−TCO(1), and 22.0 h for 125I-CC49− TCO(2). Blood clearance data for 125I-CC49−TCO(1) were originally published in ref 14.

■

DISCUSSION Pretargeted RIT is a promising approach for the delivery of a therapeutic radiation dose to solid tumors while sparing normal tissues.1 We and others7−9 have employed the bio-orthogonal inverse-electron-demand Diels−Alder cycloaddition between trans-cyclooctene (TCO) and tetrazines as a new strategy for tumor pretargeting. In a recent paper, we improved the in vivo stability and reactivity of the TCO tag and demonstrated that the reaction constant is up to 10-fold higher for axially linked TCO tags compared with their equatorial isomers.13 Within the tested series, the mAb−TCO conjugate CC49−TCO(1) (Figure 1), containing an oxymethylbenzamide-linker between

clearance half-life of 22.0 h for 125I-CC49−TCO(2) compared with the 14.1 h observed earlier for 125I-CC49−TCO(1) (p < 0.05).14 Native 125I-CC49 showed a 1.2-fold longer blood halflife (25.9 h) compared with CC49−TCO(2) (p = 0.30). Assessment of the in vivo TCO stability of 125I-CC49−TCO(2) in circulation in mice revealed a half-life of 10.3 days, a 2.6-fold slower deactivation than the half-life of 3.9 days found earlier for 125I-CC49−TCO(1) (Figure 4) (p < 0.01).13 Pretargeting Protocol: Biodistribution and SPECT/CT Imaging. Encouraged by the significant improvements in blood half-life and in vivo TCO stability for 125I-CC49− 3093

dx.doi.org/10.1021/mp500275a | Mol. Pharmaceutics 2014, 11, 3090−3096

Molecular Pharmaceutics

Article

Table 1. Dual-Isotope Biodistribution of Tumor-Bearing Micea

125

I-CC49−TCO(1) and

177

Lu-3, and

125

I-CC49−TCO(2) and

125

177

I (%ID/g)

125

organ tumor blood heart lung liver spleen kidney bladder muscle bone brain

125

I-CC49−TCO(1) 22.22 0.19 0.50 0.68 0.68 0.42 0.31 0.62 0.21 0.20 0.01 125

± ± ± ± ± ± ± ± ± ± ±

2.58 0.04 0.08 0.14 0.35 0.14 0.04 0.22 0.12 0.04 0.00

32.88 ± 0.31 ± 0.62 ± 0.99 ± 2.35 ± 0.67 ± 0.47 ± 1.09 ± 0.22 ± 0.29 ± 0.02 ± 125 I (%ID/organ)

I-CC49−TCO(1)

stomach small intestine large intestine thyroid

0.05 0.31 0.22 0.30

± ± ± ±

I-CC49−TCO(2)

0.00 0.05 0.04 0.19

CC49−TCO(1) pretargeted

4.35* 0.13 0.22 0.24 0.82* 0.07* 0.10* 0.32 0.08 0.08 0.00

± ± ± ± ± ± ± ± ± ± ±

I-CC49−TCO(2) ± ± ± ±

0.01 0.08 0.31 0.00

CC49−TCO(2) pretargeted

1.09 0.01 0.01 0.06 0.01 0.01 0.19 0.22 0.01 0.01 0.00

9.25 0.03 0.05 0.24 0.19 0.09 1.50 0.61 0.03 0.08 0.01

CC49−TCO(1) pretargeted

0.03 0.09 0.22 0.23

± ± ± ±

Lu-3 in LS174T

Lu (%ID/g)

Lu-3

177

125

0.15 0.42 0.61 0.69

6.13 0.03 0.04 0.20 0.18 0.09 1.23 0.54 0.02 0.05 0.01

177

177

0.01 0.02 0.12 0.00

177

± ± ± ± ± ± ± ± ± ± ±

177

Lu-3

2.16* 0.00 0.01 0.02 0.02 0.01 0.24 0.22 0.00 0.04 0.00

Lu (%ID/organ)

Lu-3

CC49−TCO(2) pretargeted 0.02 0.11 0.48 0.01

± ± ± ±

177

Lu-3

0.00 0.03 0.31 0.00

a The mice were injected with 125I-CC49−TCO(1) or 125I-CC49−TCO(2), two doses (30 and 48 h post-mAb injection) of galactose-albumin− tetrazine (4), 177Lu-tetrazine 3 at 50 h post-mAb injection, and euthanized 3 h later. Data are mean %ID/g ± SD (n = 4; * p < 0.05). Biodistribution data for 125I-CC49−TCO(1) and CC49−TCO(1) pretargeted 177Lu-3 were originally published in ref 14.

Table 2. Tumor-to-Nontumor (T/NT) Ratios for 177LuTetrazine 3 in Mice Bearing LS174T Colon Carcinoma Xenograftsa T/NT ratios blood heart lung liver spleen kidney bladder muscle bone brain

CC49−TCO(1) pretargeted 177 Lu-3 254 143 34 35 68 5 15 246 125 1089

± ± ± ± ± ± ± ± ± ±

59 33 14 4 8 1 12 20 26 184

CC49−TCO(2) pretargeted 177 Lu-3 304 187 39 50 108 6 16 328 142 1675

± ± ± ± ± ± ± ± ± ±

81 20 8 10* 22* 2 3 96 60 243

a The mice were injected with 125I-CC49−TCO(1) or 125I-CC49− TCO(2), two doses (30 and 48 h post-mAb injection) of galactosealbumin−tetrazine (4), 177Lu-tetrazine 3 at 50 h post-mAb injection, and euthanized 3 h later. Data are mean ± SD (n = 4, * p < 0.05). Data for CC49−TCO(1) pretargeted 177Lu-3 were originally published in ref 14.

mAb and TCO, was found to be the most reactive CC49conjugated TCO toward 177Lu-3 (k2 = 27.3 × 104 M−1 s−1). We furthermore developed a clearing agent capable of eliminating circulating CC49−TCO, prior to tetrazine probe injection, yielding significantly improved tetrazine T/NT ratios.14 We here set out to further improve the tumor uptake of the tetrazine probe by increasing the tumor uptake of the tagged antibody and by increasing the stability of the antibody-bound tag. Linker Improvement. We recognized that the relatively high protein surface hydrophobicity of the mAb−TCO conjugate CC49−TCO(1) may be affecting the pretargeting efficacy. It has been demonstrated that the linkers between antibody and drug molecules in antibody−drug conjugates

Figure 5. SPECT/CT image (maximum intensity projection) of a live mouse bearing a LS174T xenograft pretargeted with CC49−TCO(2), 3 h after 111In-3 injection. The image shows high radioactivity uptake in the tumor and low uptake in nontarget organs. Most of the activity is present in the urinary bladder. A post-mortem SPECT/CT image (high resolution) is available in the Supporting Information, Figure 5.

(ADCs) play a critical role in pharmacokinetics and biodistribution.15 ADCs with relatively hydrophobic linker− drug combinations clear faster from circulation than more 3094

dx.doi.org/10.1021/mp500275a | Mol. Pharmaceutics 2014, 11, 3090−3096

Molecular Pharmaceutics

Article

pretargeted with CC49−TCO(2) compared with CC49− TCO(1). These data indicate that the increased CC49− TCO(2) tumor accumulation and in vivo TCO stability outweighed its lower reactivity. Calculation of the absolute amounts of TCO and tetrazine present in tissues, based on the %ID/g values of 125I and 177Lu, revealed an on-tumor reaction yield (based on TCO) of 38% (corrected for TCO degradation in vivo). This is slightly lower than the previously reported 46% on-tumor reaction yield for CC49−TCO(1) and 177Lutetrazine-3,14 which is possibly due to the lower reactivity of CC49−TCO(2). As in our previous studies, we demonstrated that the radiolabeled tetrazine itself is not retained in the mouse, except in TCO-containing tissues and in kidneys and bladder due to excretion. As expected, the increased 125ICC49−TCO(2) uptake in liver did not result in increased 177 Lu-tetrazine-3 accumulation in this tissue due to rapid internalization of the antibody-clearing agent conjugate, precluding the reaction. The fast and selective reaction of the tetrazine probe with the CC49−TCO(2)-pretargeted tumors, and the resulting excellent T/NT ratios was also demonstrated in a corresponding SPECT/CT imaging study in live mice using 111In-tetrazine-3 (Figure 5). Previous dosimetry calculations suggested that tetrazine-3 can deliver a therapeutic dose to mice bearing tumors pretargeted with CC49−TCO(1).14 We estimated that less than 21 Gy would be delivered to the tumor with 177Lu-CC49, whereas the pretargeting approach would deliver 171 Gy, exceeding 80−100 Gy, commonly considered the threshold for a therapeutic effect for radioimmunotherapy of solid tumors. Therefore, we expect that preadministration of CC49−TCO(2) instead of CC49− TCO(1) will further improve pretargeted RIT with less radiation to normal tissues. An increased dose may increase the percentage of complete responses, while a retained dose will reduce the radiation burden to healthy tissues. In addition, optimized pharmacokinetics and tumor uptake is essential to maximize reaction yields upon clinical translation, where the tumor-bound antibody concentrations are lower. The observed 1.2-fold difference in blood clearance half-life for CC49 and CC49−TCO(2) may eventually be further reduced by employing more hydrophilic linkers,15 increasing tumor accumulation and potentially CC49−TCO stability. Finally, switching from nonspecific lysine conjugation to a site-specific conjugation strategy, thereby eliminating conjugate heterogeneity, is expected to further improve the targeting, pharmacokinetics, and stability of the antibody conjugate.19,20

hydrophilic moieties, leading to a lower tumor exposure and uptake.15,16 Likewise, we observed a 4-fold lower tumor uptake of CC49−TCO(1) compared with a CC49 antibody functionalized with hydrophilic DOTA chelates.14 We therefore hypothesized that the pretargeting system can be further improved by inclusion of a mAb−TCO conjugate with a less hydrophobic linker, such as an oxymethylacetamide moiety (Supporting Information, Figure 1).13 While this compound, here named CC49−TCO(2), is 2-fold less reactive toward 177 Lu-3 (k2 = 13.5 × 104 M−1 s−1) compared with CC49− TCO(1), we reasoned that this could be outweighed by an increased tumor uptake of a more hydrophilic CC49−TCO, leading to increased tumor uptake of tetrazine probe. TCO Reactivity versus TCO Stability. Earlier experiments showed that a ∼10-fold difference in reactivity of equatorial/ axial-linked TCO-tagged CC49 conjugate is associated with a ∼2-fold inversed difference in in vivo stability of the TCO tag.13 Assuming an inverse−linear correlation between stability and reactivity, the TCO tag in CC49−TCO(2) is expected to be ca. 1.6-fold more stable than in CC49−TCO(1), which may further contribute to an increased tumor uptake of the tetrazine probe. When we studied the in vivo stability of CC49− TCO(2), we were surprised to find a 2.6-fold longer TCO deactivation half-life for CC49−TCO(2) compared with CC49−TCO(1) (p < 0.01). In addition to a difference in reactivity, a reduced interaction between CC49−TCO(2) and serum proteins may have contributed to this unexpected and much improved in vivo TCO half-life.13 We previously reported that the interaction between CC49−TCO with albumin-bound copper can result in isomerization of the reactive transcyclooctene tag to its unreactive cis-cyclooctene (CCO) analogue, and that this can be significantly impeded by increasing the steric hindrance around the TCO tag.13 In that respect, the shorter acetamide linker compared with the benzamide linker may have further increased the steric hindrance around the TCO tag, improving its in vivo stability. In addition, as albumin has binding pockets for hydrophobic compounds and has also been shown to bind to hydrophobic cyclooctyne tags,11,17,18 the reduced hydrophobicity of the acetamide linker may have contributed to a reduced albumin interaction and corresponding TCO to CCO conversion. Blood Kinetics. The blood clearance half-life of 125I-CC49− TCO(2) was 1.6-fold higher than for 125I-CC49−TCO(1) (p < 0.05), which is in agreement with other studies that show slower blood clearance for hydrophilic antibody-conjugates compared with hydrophobic antibody-conjugates. Much to our satisfaction, native 125I-CC49 showed only a 1.2-fold longer blood clearance half-life compared with 125I-CC49−TCO(2) (p = 0.30), indicating that 125I-CC49−TCO(2) is approaching the pharmacokinetics of native 125I-CC49. In Vivo Biodistribution. A dual-isotope pretargeting biodistribution study in LS174T tumor-bearing mice, employing 125I-CC49−TCO(2), clearing agent, and 177Lu-tetrazine-3 showed a significant increase in mAb tumor targeting for CC49−TCO(2) compared with CC49−TCO(1). Also, liver (3.5-fold), spleen (1.6-fold), and kidney (1.5-fold) showed significantly higher levels of CC49−TCO(2) 5 h after the last clearing agent injection. The higher liver uptake is most likely a result of the higher mAb concentration in blood at the time of clearing agent administration and the subsequent active liver uptake of the mAb-clearing agent conjugate. Intravenous administration of 177Lu-tetrazine-3 yielded a satisfying 1.5-fold increase in tumor uptake and improved T/NT ratios for mice

■

CONCLUSION Using a less hydrophobic TCO as pretargeting tag afforded a CC49−TCO conjugate with increased in vivo tag stability, a longer blood clearance half-life, and improved tumor accumulation. This resulted in an increased tumor uptake of the small and fast clearing radiolabeled tetrazine probe and increased T/NT ratios. This work demonstrates that bioorthogonal pretargeting based on the inverse-electrondemand Diels−Alder reaction can be further improved by optimizing the pharmacokinetics of the tagged antibody and by increasing the stability of the tag, even at the expense of some reactivity. We expect that the markedly reduced perturbation of the tagged mAb afforded by the oxymethylacetamide-linked TCO and the extended in vivo tag stability will also lead to companion imaging applications for biologics. 3095

dx.doi.org/10.1021/mp500275a | Mol. Pharmaceutics 2014, 11, 3090−3096

Molecular Pharmaceutics

■

Article

azide probes for in vivo Staudinger ligation in a pretargeted imaging and therapy approach. Bioconjugate Chem. 2011, 22 (10), 2072−2081. (13) Rossin, R.; van den Bosch, S. M.; ten Hoeve, W.; Carvelli, M.; Versteegen, R. M.; Lub, J.; Robillard, M. S. Highly reactive transcyclooctene tags with improved stability for Diels−Alder chemistry in living systems. Bioconjugate Chem. 2013, 24 (7), 1210−1217. (14) Rossin, R.; Läppchen, T.; van den Bosch, S. M.; Laforest, R.; Robillard, M. S. Diels−Alder reaction for tumor pretargeting: in vivo chemistry can boost tumor radiation dose compared with directly labeled antibody. J. Nucl. Med. 2013, 54 (11), 1989−1995. (15) Zhao, R. Y.; Wilhelm, S. D.; Audette, C.; Jones, G.; Leece, B. A.; Lazar, A. C.; Goldmacher, V. S.; Singh, R.; Kovtun, Y.; Widdison, W. C.; Lambert, J. M.; Chari, R. V. J. Synthesis and evaluation of hydrophilic linkers for antibody−maytansinoid conjugates. J. Med. Chem. 2011, 54 (10), 3606−3623. (16) Hamlett, K. J.; Senter, P. D.; Chace, D. F.; Sun, M. M. C.; Lenox, J.; Cerveny, C. G.; Kissler, K. M.; Bernhardt, S. X.; Kopcha, A. K.; Zabinski, R. F.; Meyer, D. L.; Francisco, J. A. Effects of drug loading on the antitumor activity of a monoclonal antibody drug conjugate. Clin. Cancer Res. 2004, 10 (20), 7063−7070. (17) Chang, P. V.; Prescher, J. A.; Sletten, E. M.; Baskin, J. M.; Miller, I. A.; Agard, N. J.; Lo, A.; Bertozzi, C. R. Copper-free click chemistry in living animals. Proc. Natl. Acad. Sci. U. S. A. 2010, 107 (5), 1821− 1826. (18) van Geel, R.; Pruijn, G. J. M.; van Delft, F. L.; Boelens, W. C. Preventing thiol-yne addition improves the specificity of strainpromoted azide−alkyne cycloaddition. Bioconjugate Chem. 2012, 23 (3), 392−398. (19) Junutula, J. R.; Raab, H.; Clark, S.; Bhakta, S.; Leipold, D. D.; Weir, S.; Chen, Y.; Simpson, M.; Ping Tsai, S.; Dennis, M. S.; Lu, Y.; Gloria Meng, Y.; Ng, C.; Yang, J.; Lee, C. C.; Duenas, E.; Gorrell, J.; Katta, V.; Kim, A.; McDorman, K.; Flagella, K.; Venook, R.; Ross, S.; Spencer, S. D.; Lee Wong, W.; Lowman, H. B.; Vandlen, R.; Sliwkowski, M. X.; Scheller, R. H.; Polakis, P.; Mallet, W. Site-specific conjugation of a cytotoxic drug to an antibody improves the therapeutic index. Nature Biotechnol. 2008, 26 (8), 925−932. (20) Boswell, C. A.; Mundo, E. E.; Zhang, C.; Bumbaca, D.; Valle, N. R.; Kozak, K. R.; Fourie, A.; Chuh, J.; Koppada, N.; Saad, O.; Gill, H.; Shen, B.; Rubinfeld, B.; Tibbitts, J.; Kaur, S.; Theil, F.; Fielder, P. J.; Khawli, L. A.; Lin, K. Impact of drug conjugation on pharmacokinetics and tissue distribution of anti-STEAP1 antibody-drug-conjugates in rats. Bioconjugate Chem. 2011, 22 (10), 1994−2004.

ASSOCIATED CONTENT

S Supporting Information *

Log P calculations for TCO(1) and TCO(2) derivatives, CC49−TCO(2) radiolabeling data, CC49 and CC49−TCO(2) immunoreactivity, and post-mortem SPECT/CT images of a pretargeting study with CC49−TCO(2). This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*Phone: +31402748264. Fax: + 31402744906. E-mail: marc. [email protected]. Notes

The authors declare the following competing financial interest(s): Employee and shareholder of Tagworks.

■

ACKNOWLEDGMENTS We thank Dr. Ebo Bos and Dr. Frans Kaspersen for insightful discussions and Dr. Iris Verel, Monique Berben, Caren van Kammen, Carlijn van Helvert, and Melanie Blonk for support with in vivo experiments. This research was supported by NanoNextNl (The Netherlands).

■

REFERENCES

(1) Boswell, C. A.; Brechbiel, M. W. Development of radioimmunotherapeutic and diagnostic antibodies: an inside-out view. Nucl. Med. Biol. 2007, 34 (7), 757−778. (2) Song, H.; Sgouros, G. Radioimmunotherapy of solid tumors: searching for the right target. Curr. Drug Delivery 2011, 8 (1), 26−44. (3) Meredith, R. F.; Buchsbaum, D. J. Pretargeted radioimmunotherapy. Int. J. Radiat. Oncol. Biol. Phys. 2006, 66 (2), S57−S59. (4) Sharkey, R. M.; Goldenberg, D. M. Pretargeted radioimmunotherapy of cancer. In Monoclonal Antibody and PeptideTargeted Radiotherapy of Cancer; Reilly, R. M., Ed.; John Wiley & Sons: Hoboken, NJ, 2010; pp 241−287. (5) Goldenberg, D. M.; Chang, C. H.; Sharkey, R. M.; Rossi, E. A.; Karacay, H.; McBride, W.; Hansen, H. J.; Chatal, J.; Barbet, J. Radioimmunotherapy: is avidin−biotin pretargeting the preferred choice among pretargeting methods? Eur. J. Nucl. Med. Mol. Imaging 2003, 30 (5), 777−780. (6) Sletten, E. M.; Bertozzi, C. R. Bioorthogonal Chemistry: Fishing for Selectivity in a Sea of Functionality. Angew. Chem. Int. Ed. 2009, 48 (38), 6974−6998. (7) Rossin, R.; Renart Verkerk, P.; van den Bosch, S. M.; Vulders, R. C. M.; Verel, I.; Lub, J.; Robillard, M. S. In vivo chemistry for pretargeted tumor imaging in live mice. Angew. Chem. Int. Ed. 2010, 49 (19), 3375−3378. (8) Devaraj, N. K.; Thurber, G. M.; Keliher, E. J.; Marinelli, B.; Weissleder, R. Reactive polymer enables efficient in vivo bioorthogonal chemistry. Proc. Natl. Acad. Sci. U. S. A. 2012, 109 (13), 4762−4767. (9) Zeglis, B. M.; Sevak, K. K.; Reiner, T.; Mohindra, P.; Carlin, S. D.; Zanzonico, P.; Weissleder, R.; Lewis, J. S. A pretargeted PET imaging strategy based on bioorthogonal Diels−Alder click chemistry. J. Nucl. Med. 2013, 54 (8), 1389−1396. (10) Neves, A. A.; Stöckmann, H.; Harmston, R. R.; Pryor, H. J.; Alam, I. S.; Ireland-Zecchini, H.; Lewis, D. Y.; Lyons, S. K.; Leeper, F. J.; Brindle, K. M. Imaging sialylated tumor cell glycans in vivo. FASEB J. 2011, 25 (8), 2528−2537. (11) Van den Bosch, S. M.; Rossin, R.; Renart Verkerk, P.; ten Hoeve, W.; Janssen, H. M.; Lub, J.; Robillard, M. S. Evaluation of strained alkynes for Cu-free click reaction in live mice. Nucl. Med. Biol. 2013, 40 (3), 415−423. (12) Vugts, D. J.; Vervoort, A.; Stigter-van Walsum, M.; Visser, G. W. M.; Robillard, M. S.; Versteegen, R. M.; Vulders, R. C. M.; Herscheid, J. D. M. van Dongen GAMS. Synthesis of phosphine and antibody3096

dx.doi.org/10.1021/mp500275a | Mol. Pharmaceutics 2014, 11, 3090−3096