Evaluation of [64Cu]Cu-NOTA-PEG7-H-Tz for Pretargeted Imaging in LS174T Xenografts—Comparison to [111In]In-DOTA-PEG11-BisPy-Tz

Pretargeted nuclear imaging for the diagnosis of various cancers is an emerging and fast developing field. The tetrazine ligation is currently considered the most promising reaction in this respect. Monoclonal antibodies are often the preferred choice as pretargeting vector due to their outstanding targeting properties. In this work, we evaluated the performance of [64Cu]Cu-NOTA-PEG7-H-Tz using a setup we previously used for [111In]In-DOTA-PEG11-BisPy-Tz, thereby allowing for comparison of the performance of these two promising pretargeting imaging agents. The evaluation included a comparison of the physicochemical properties of the compounds and their performance in an ex vivo blocking assay. Finally, [64Cu]Cu-NOTA-PEG7-H-Tz was evaluated in a pretargeted imaging study and compared to [111In]In-DOTA-PEG11-BisPy-Tz. Despite minor differences, this study indicated that both evaluated tetrazines are equally suited for pretargeted imaging.


Introduction
Pretargeted imaging has attracted increased interest over the last decades and is an emerging and fast developing field within oncology [1]. Until recently, pretargeted strategies were mainly based on the noncovalent interactions between the binding of biotin to (strept)avidin, duplex formation of two complementary strands of oligonucleotides or via the use of bispecific antibodies [2]. More recently, the focus has shifted to covalent bond forming methods [3][4][5]. This requires reactions that are not only bioorthogonal, but also exceptionally fast for in vivo applications. Several different methods exist, but the tetrazine ligation between a trans-cyclooctene (TCO) and a tetrazine (Tz) is currently considered the most promising reaction for this purpose [6]. The reaction is initiated via an inverse-electron-demand Diels-Alder (IEDDA) reaction and undergoes subsequently a retro-Diels-Alder reaction, with the expulsion of nitrogen gas. As such, the tetrazine ligation is irreversible [7]. Usually, monoclonal antibodies (mAbs), but also other nanomedicines, are used as pretargeting vectors. They possess exceptional targeting abilities [8]. Typically, TCOs are conjugated to the pretargeting vector and the tetrazine is used as the imaging probe [1,9].
[ 111 In]In-DOTA-PEG 11 -BisPy-Tz (2), was the first radiolabeled Tz to be applied for this purpose, using TCO conjugated CC49 (CC49-TCO), an mAb that targets tumor associated glycoprotein 72 (TAG-72), as the pretargeting vector [4]. The TAG-72 antigen has been shown to have limited internalization when CC49 binds, making it ideal for pretargeting approaches [10]. Using this setup, [ 111 In]In-DOTA-PEG 11 -BisPy-Tz (2) showed high tumor uptake, as well as high tumor-to-background ratios, in mice bearing human colon carcinoma xenografts of LS174T cancer cells that overexpress TAG72. Many groups have continuously worked to improve pretargeting strategies based on the tetrazine ligation. For example, we have recently used the same experimental setup to compare the performance of [ 111 In]In-DOTA-PEG 11 -BisPy-Tz (2) to a 68 Ga-labeled variant of the tracer [11]. In general, it is usually difficult to compare the performance of different tetrazine-based molecular imaging probes, because factors such as animal strain, tumor models, pretargeting vectors, and evaluation time points are varied [12][13][14][15][16][17][18][19][20]. This is also the case for [ 64 Cu]Cu-NOTA-PEG 7 -H-Tz (1), another radiolabeled tetrazine, which is reported to be highly effective for pretargeted imaging, however, using a different TCO-conjugated pretargeting vector and mouse tumor model [3]. In order to allow for direct comparison of tetrazine-based imaging agents, there is a need to harmonize the experimental setup.
In this study, we evaluated [ 64 Cu]Cu-NOTA-PEG 7 -H-Tz (1) in mice bearing LS174T tumor xenografts, using CC49-TCO as a pretargeting vector. The same setup that was previously used for the evaluation of [ 111 In]In-DOTA-PEG 11 -BisPy-Tz (2) by us and others [3,4,11], and allows us to compare the in vivo performance of [ 64 Cu]Cu-NOTA-PEG 7 -H-Tz to previous results [11]. Figure 1 displays both tetrazines used in this study. There are some differences between positron emission tomography (PET) used for imaging of [ 64 Cu]Cu-NOTA-PEG 7 -H-Tz (1), and single-photon emission computed tomography (SPECT) used for imaging of [ 111 In]In-DOTA-PEG 11 -BisPy-Tz, for example with regard to the resolution and sensitivity [21]. In general, these differences are less pronounced for preclinical systems, and will as such not be included in this comparison. However, it could affect the performance if translated to clinical settings, and should therefore also be taken into consideration [22].  (2), was the first radiolabeled Tz to be applied for this purpose, using TCO conjugated CC49 (CC49-TCO), an mAb that targets tumor associated glycoprotein 72 (TAG-72), as the pretargeting vector [4]. The TAG-72 antigen has been shown to have limited internalization when CC49 binds, making it ideal for pretargeting approaches [10]. Using this setup, [ 111 In]In-DOTA-PEG11-BisPy-Tz (2) showed high tumor uptake, as well as high tumor-to-background ratios, in mice bearing human colon carcinoma xenografts of LS174T cancer cells that overexpress TAG72. Many groups have continuously worked to improve pretargeting strategies based on the tetrazine ligation. For example, we have recently used the same experimental setup to compare the performance of [ 111 In]In-DOTA-PEG11-BisPy-Tz (2) to a 68 Ga-labeled variant of the tracer [11]. In general, it is usually difficult to compare the performance of different tetrazine-based molecular imaging probes, because factors such as animal strain, tumor models, pretargeting vectors, and evaluation time points are varied [12][13][14][15][16][17][18][19][20]. This is also the case for [ 64 Cu]Cu-NOTA-PEG7-H-Tz (1), another radiolabeled tetrazine, which is reported to be highly effective for pretargeted imaging, however, using a different TCO-conjugated pretargeting vector and mouse tumor model [3]. In order to allow for direct comparison of tetrazinebased imaging agents, there is a need to harmonize the experimental setup.
In this study, we evaluated [ 64 Cu]Cu-NOTA-PEG7-H-Tz (1) in mice bearing LS174T tumor xenografts, using CC49-TCO as a pretargeting vector. The same setup that was previously used for the evaluation of [ 111 In]In-DOTA-PEG11-BisPy-Tz (2) by us and others [3,4,11], and allows us to compare the in vivo performance of [ 64 Cu]Cu-NOTA-PEG7-H-Tz to previous results [11].  [21]. In general, these differences are less pronounced for preclinical systems, and will as such not be included in this comparison. However, it could affect the performance if translated to clinical settings, and should therefore also be taken into consideration [22].

Comparison of Physicochemical Properties and In Vitro Stability
The structural design of [ 64 Cu]Cu-NOTA-PEG7-H-Tz (1) and [ 111 In]In-DOTA-PEG11-BisPy-Tz (2) follows the same principle. Both tetrazines consist of a reactive tetrazine moiety and a polar chelator, connected via a polyethylene glycol (PEG) spacer. However, the overall charge of the metal complex is -1 for the copper (Cu II )-NOTA complex (1) and 0 for the indium (In III )-DOTA complex (2). The lipophilicity of the compounds is also fairly distinct. Whereas the [ 64 [3,23]. Moreover, the H-Tz in 1 displays a slightly lower reactivity than the BisPy-Tz in 2 (Table 1). We recently showed that higher polarity and higher reactivity of tetrazines are beneficial to achieve higher target-to-background ratios [24]. As such, the [ 111 In]In-DOTA-PEG11-BisPy-Tz (2) may have better imaging properties. However, only trends and not absolute relationships were identified within our study, and target-to-background ratios were almost

Comparison of Physicochemical Properties and In Vitro Stability
The structural design of [ 64 Cu]Cu-NOTA-PEG 7 -H-Tz (1) and [ 111 In]In-DOTA-PEG 11 -BisPy-Tz (2) follows the same principle. Both tetrazines consist of a reactive tetrazine moiety and a polar chelator, connected via a polyethylene glycol (PEG) spacer. However, the overall charge of the metal complex is −1 for the copper (Cu II )-NOTA complex (1) and 0 for the indium (In III )-DOTA complex (2). The lipophilicity of the compounds is also fairly distinct. Whereas the [ 64 Cu]Cu-NOTA-PEG 7 -H-Tz (1) possesses a logD 7.4 of approximately −2.44, the [ 111 In]In-DOTA-PEG 11 -BisPy-Tz (2) displays a logD 7.4 of ca. −4.51 [3,23]. Moreover, the H-Tz in 1 displays a slightly lower reactivity than the BisPy-Tz in 2 (Table 1). We recently showed that higher polarity and higher reactivity of tetrazines are beneficial to achieve higher target-to-background ratios [24]. As such, the [ 111 In]In-DOTA-PEG 11 -BisPy-Tz (2) may have better imaging properties. However, only trends and not absolute relationships were identified within our study, and target-to-background ratios were almost identical at a cLogD 7.4 of approximately <−3 and a rate constant > 200 M −1 s −1 almost identical [24]. Both tetrazines studied herein have values in these ranges. Therefore, we cannot predict whether the faster reaction kinetics and the lower lipophilicity of [ 111 In]In-DOTA-PEG 11 -BisPy-Tz (2) may, in fact, turn out to be of greater benefit for pretargeted imaging. Both probes revealed similar in vitro stabilities, in PBS, as well as in serum [3,4]. After 2 h at 37 • C, approximately 85% of the compounds were intact. Table 1 lists these relevant physicochemical properties, the reactivity, and the in vitro stability data of both compounds.  [23], c data taken from reference [3], d data taken from reference [24], and e data taken from reference [4].

Ex Vivo Blocking Study
NOTA-PEG 7 -H-Tz was first evaluated in an in-house-developed ex vivo blocking assay [24]. For direct comparison, DOTA-PEG 11 -BisPy-Tz was also included in the test. The principle behind this assay is based on traditional receptor competition studies. In short, 72 h prior to the injection of the Tz, CC49-TCO (100 µg,~7 TCOs/mAb, 3.9 nmol TCO) was administered into BALB/c nude mice, bearing LS174T colon carcinoma xenografts. Unlabeled Tz (10 equivalents with respect to the injected quantity of TCO) were administered 1 h before administration of [ 111 In]In-DOTA-PEG 11 -BisPy-Tz (2) (1 equivalent with respect to injected TCO amount). Animals were euthanized after 22 h, tissues dissected, and the ex vivo biodistribution of [ 111 In]In-DOTA-PEG 11 -BisPy-Tz (2) determined using a gamma counter. This allowed for quantification of the blocking effect of the unlabeled tetrazines. Both compounds were able to block the tumor uptake by >96% ( Figure 2). Minor differences in uptake of [ 111 In]In-DOTA-PEG 11 -BisPy-Tz (2) was found between animals pretreated with NOTA-PEG 7 -H-Tz and DOTA-PEG 11 -BisPy-Tz in heart, spleen, and kidney tissue.

Pretargeted Imaging
Pretargeted imaging was performed using a similar setup to the ex vivo blocking assay. CC49-TCO was administered 72 h prior to the administration of the radiolabeled tetrazine and PET imaging of [ 64 Cu]Cu-NOTA-PEG7-H-Tz (1) was carried out 2 and 22 h after the radioligand administration. The tumors were clearly visible at both time points and the background uptake was generally low ( Figure 3B). Image analysis showed that the tumor uptake increased from 3.2 ± 0.3 percentage of injected dose per gram (% ID/g, mean  SEM) 2 h post injection (p.i.) to 7.7 ± 0.2% ID/g 22 h p.i. (Figure 3A,D). The corresponding tumor-to-muscle (T/M) ratios were 6.4 and 19.3, respectively. No specific tumor uptake was detected in control animals that were injected with [ 64 Cu]Cu-NOTA-PEG7-H-Tz (1) without any prior administration of CC49-TCO.
The uptake in tumor tissue as well as in other organs of [ 64 Cu]Cu-NOTA-PEG7-H-Tz (1) was slightly lower than those previously found when evaluating [ 111 In]In-DOTA-PEG11-BisPy-Tz (2), using the same setup ( Figure 3C,D) [11]. Significant amounts of activity of 2.9 ± 0.5% ID/g were also found in the heart (surrogate for the blood activity) after 2 h, which slightly decreased over the next 20 h to 2.3 ± 0.1% ID/g, resulting in tumor-toblood (T/B) ratios of 1.1 and 3.3, respectively ( Figure 3D). These values are significantly lower than those observed for [ 111 In]In-DOTA-PEG11-BisPy-Tz (2), and indicate that [ 64 Cu]Cu-NOTA-PEG7-H-Tz (1) is excreted at a faster rate [11]. This also results in higher tumor-to-background ratios ( Figure 3D). However, free Tz is excreted within the first hour and as such, the observed blood uptake cannot be explained by free circulating Tz. A likely explanation for the unexpected high blood uptake over 22 h is that a fraction of the radiolabeled Tz ligates to residual CC49-TCO still circulating within the blood stream. This can also explain the observed increase in tumor uptake over time, as the ligation adduct in the circulation gradually accumulates at the target side over time. This phenomenon has also previously been described for [ 111 In]In-DOTA-PEG11-BisPy-Tz (2) [5].
In conclusion, we evaluated the performance of [ 64 Cu]Cu-NOTA-PEG7-H-Tz (1) in vivo using a setup previously used for [ 111 In]In-DOTA-PEG11-BisPy-Tz (2) and were therefore able to directly compare the performance of the two imaging agents. Both radiolabeled tetrazines performed equally well in an ex vivo blocking assay. Similarly, the pretargeted imaging study revealed only minor differences in performance between the two imaging agents. Although the tumor uptake of [ 64 Cu]Cu-NOTA-PEG7-H-Tz (1) was lower than that of [ 111 In]In-DOTA-PEG11-BisPy-Tz (2), the former cleared at a higher rate,

Pretargeted Imaging
Pretargeted imaging was performed using a similar setup to the ex vivo blocking assay. CC49-TCO was administered 72 h prior to the administration of the radiolabeled tetrazine and PET imaging of [ 64 Cu]Cu-NOTA-PEG 7 -H-Tz (1) was carried out 2 and 22 h after the radioligand administration. The tumors were clearly visible at both time points and the background uptake was generally low ( Figure 3B). Image analysis showed that the tumor uptake increased from 3.2 ± 0.3 percentage of injected dose per gram (% ID/g, mean ± SEM) 2 h post injection (p.i.) to 7.7 ± 0.2% ID/g 22 h p.i. (Figure 3A,D). The corresponding tumor-to-muscle (T/M) ratios were 6.4 and 19.3, respectively. No specific tumor uptake was detected in control animals that were injected with [ 64 Cu]Cu-NOTA-PEG 7 -H-Tz (1) without any prior administration of CC49-TCO.
The uptake in tumor tissue as well as in other organs of [ 64 Cu]Cu-NOTA-PEG 7 -H-Tz (1) was slightly lower than those previously found when evaluating [ 111 In]In-DOTA-PEG 11 -BisPy-Tz (2), using the same setup ( Figure 3C,D) [11]. Significant amounts of activity of 2.9 ± 0.5% ID/g were also found in the heart (surrogate for the blood activity) after 2 h, which slightly decreased over the next 20 h to 2.3 ± 0.1% ID/g, resulting in tumor-to-blood (T/B) ratios of 1.1 and 3.3, respectively ( Figure 3D). These values are significantly lower than those observed for [ 111 In]In-DOTA-PEG 11 -BisPy-Tz (2), and indicate that [ 64 Cu]Cu-NOTA-PEG 7 -H-Tz (1) is excreted at a faster rate [11]. This also results in higher tumorto-background ratios ( Figure 3D). However, free Tz is excreted within the first hour and as such, the observed blood uptake cannot be explained by free circulating Tz. A likely explanation for the unexpected high blood uptake over 22 h is that a fraction of the radiolabeled Tz ligates to residual CC49-TCO still circulating within the blood stream. This can also explain the observed increase in tumor uptake over time, as the ligation adduct in the circulation gradually accumulates at the target side over time. This phenomenon has also previously been described for [ 111 In]In-DOTA-PEG 11 -BisPy-Tz (2) [5].
In conclusion, we evaluated the performance of [ 64 Cu]Cu-NOTA-PEG 7 -H-Tz (1) in vivo using a setup previously used for [ 111 In]In-DOTA-PEG 11 -BisPy-Tz (2) and were therefore able to directly compare the performance of the two imaging agents. Both radiolabeled tetrazines performed equally well in an ex vivo blocking assay. Similarly, the pretargeted imaging study revealed only minor differences in performance between the two imaging agents. Although the tumor uptake of [ 64 Cu]Cu-NOTA-PEG 7 -H-Tz (1) was lower than that of [ 111 In]In-DOTA-PEG 11 -BisPy-Tz (2), the former cleared at a higher rate, resulting in higher T/B and T/M ratios. For both radiolabeled tetrazines, however, the tumors were easily detected on the images, as early as 2 h p.i. Based on this, we believe that [ 64 Cu]Cu-NOTA-PEG 7 -H-Tz (1) and [ 111 In]In-DOTA-PEG 11 -BisPy-Tz (2) are equally suited for a pretargeted approach.   (2) previously obtained are also included (n =3). 20 Data are shown as mean ± standard error of mean (SEM). * Image-derived uptake in heart from SPECT and PET images used as a surrogate for blood.   (2) previously obtained are also included (n =3). 20 Data are shown as mean ± standard error of mean (SEM). * Image-derived uptake in heart from SPECT and PET images used as a surrogate for blood.

Organic Chemistry
All reactions involving dry solvents or sensitive agents were performed under a nitrogen atmosphere and glasswares were dried prior to use. Commercially available chemicals were used without further purification. Solvents were dried prior to use with an SG water solvent purification system (Pure Process Technology, Nashua, NH, USA) or dried by standard procedures, and reactions were monitored by analytical thin-layer chromatography (TLC, Merck silica gel 60 F 254 aluminum sheets). Flash chromatography was carried out using Merck silica gel 60Å (35-70 µm) (Sigma-Aldrich, Darmstadt, Germany). The 1 H-NMR spectra were recorded on a 400 MHz Avance III or 600 MHz Avance III HD, and 13 C NMR spectra on a 101 MHz Avance III or 151 MHz Avance III HD (Bruker, Bremen, Germany). Analytical HPLC was performed using an UltiMate HPLC system consisting of an LPG-3400A pump (1 mL/min), a WPS-3000SL autosampler, and a 3000 Diode Array Detector installed with a Gemini-NX C18 (250 × 4.60 mm, 3 µm) column. Solvent A: H 2 O + 0.1% TFA; Solvent B: MeCN-H 2 O 9:1 + 0.1% TFA. For HPLC control, data collection, and data handling, Chromeleon software v. 6.80 was used. Preparative HPLC was carried out on an Ultimate HPLC system with an LPG-3200BX pump, a Rheodyne 9721i injector, a 10 mL loop, an MWD-3000SD detector (200, 210, 225 and 254 nm) (ThermoScientific, Loughborough, UK), and a Gemini-NX C18 (250 × 21.2 mm, 5 µm) column for preparative purifications, or a Gemini-NX C18 (250 × 10.00 mm, 5 µm) column for semipreparative purifications. Solvent A: H 2 O + 0.1% TFA; Solvent B: MeCN-H 2 O 9:1 + 0.1% TFA. For HPLC control, data collection, and data handling, Chromeleon software v. 6.80 was used. Chiral preparative HPLC was performed using the same instrumentation as mentioned above. UPLC-MS spectra were recorded using an Acquity UPLC H-Class series solvent delivery system equipped with an autoinjector coupled to an Acquity QDa and TUV detectors installed with an Acquity UPLC ® BECH C18 (50 × 2.1 mm, 1.7 µm) column (Waters, Eschborn, Germany). Solvent A: 5% aq MeCN + 0.1% HCO 2 H; Solvent B: MeCN + 0.1% HCO 2 H. Usually, gradient ratios from A:B 1:0 to 1:1 (5 min) were performed depending on the polarity of the compounds. For data collection and data handling, MassLynx software (Waters, Eschborn, Germany) was used. Optical rotations were determined in a thermostated cuvette on an Anton Paar MCP300 Modular Circular Polarimeter. Compounds were dried under high vacuum or freeze dried using a CoolSafe Freeze Dryer (ScanVac, Lillerød, Denmark). The purity of compounds submitted for pharmacological characterization was determined by 1 H-NMR and HPLC to be > 95%, unless otherwise noted.

Pretargeted Imaging
Tumor-bearing mice were administered intravenously 100 µg of CC49-TCO 3 days prior to the imaging experiment. At the imaging experiment, mice were injected with [ 64 Cu]Cu-NOTA-PEG 7 -H-Tz (1) (10 MBq/100 µL, 3.2 nmol) via the tail vein. Two hours after the injection, the animals were moved to a small animal PET/CT scanner (Inveon ® , Siemens Medical Solutions, Malvern, PA, USA), and a PET acquisition (energy window of 350-650 KeV and a time resolution of 6 ns) was performed, followed by a CT scan (360 projections, 65 kV, 500 µA and 400 ms). This procedure was repeated at 22 h p.i. Sinograms from PET scans were reconstructed using a three-dimensional maximum a posteriori algorithm with scatter correction and CT-based correction for attenuation. PET and CT images were coregistered and analyzed using Inveon Research Workplace (Siemens). The mean % ID/g in different tissues was extracted by manually creating regions of interest (ROI) on fused PET/CT images. GraphPad Prism 9 (GraphPad Software, San Diego, CA, USA) was used for analyzing and plotting data.

Data Availability Statement:
The data presented in this study are available on request from the corresponding author.