18F-LABELED NOVEL TETRAZINES IMAGING PROBES FOR PRETARGETING IN PET IMAGING

Abstract

Novel .sup.18F-labelled tetrazines are provided which are highly reactive to be effective in vivo, suitable for pretargeted positron emission tomography (PET) and accessible in radiochemical yields (RCYs) which allow access to .sup.18F-labeled tetrazines for clinical applications. The .sup.18F-labelled tetrazines are developed using a Cu-mediated click indirect labelling approach. Only a subset of compounds appeared to be suitable for clinical pretargeted imaging strategies, and a particular compound which includes the use of an .sup.18F-labelled azide synthon having an azide structure with glucose as the linker and a triazole moiety within the linker, appears to be highly suited for clinical pretargeted imaging purposes.

Claims

1. A .sup.18F-labelled tetrazine formed by reacting a tetrazine with a .sup.18F-labelled azide synthon, in which said tetrazine is a compound according to any of structures 1-6: ##STR00039## ##STR00040## and said .sup.18F-labelled azide synthon is a compound according to any of structures [.sup.18F]a, [.sup.18F]b, [.sup.18F]c: ##STR00041##

2. The .sup.18F-labelled tetrazine according to claim 1, wherein the tetrazine is a compound according to structure 1, 3 or 4 and the .sup.18F-labelled azide synthon is a compound according to structure [.sup.18F]c, preferably the tetrazine is a compound according to structure 1 and the .sup.18F-labelled azide synthon is a compound according to structure [.sup.18F]c.

3. The .sup.18F-labelled tetrazine according to claim 1, wherein said reacting of tetrazine with a .sup.18F-labelled azide synthon occurs by Cu-catalyzed azide-alkyne [3+2] cycloaddition (CuAAC).

4. Method of producing an image of an organ in an animal or human, comprising forming in the animal or human a product resulting from the biorthogonal ligation between i) a .sup.18F-labelled tetrazine wherein said reacting of tetrazine with a .sup.18F-labelled azide synthon occurs by Cu-catalyzed azide-alkyne [3+2] cycloaddition (CuAAC) and ii) monoclonal antibodies (mAbs) modified with trans-cyclooctene (TCO).

5. The method according to claim 4, wherein the method is pretargeted immune-positron emission tomography (PET) imaging, by which said monoclonal antibodies (mAbs) modified with trans-cyclooctene (TCO) are provided to accumulate in a target tissue such as a tumor tissue prior to conducting said biorthogonal ligation with a .sup.18F-labelled tetrazine.

6. The method according to claim 4, wherein the .sup.18F-labelled tetrazine is formed by reacting a tetrazine with a .sup.18F-labelled azide synthon, in which said tetrazine is a compound according to any of structures: 1, 3 or 4, and said .sup.18F-labelled azide synthon is a compound according to structure [.sup.18F]c, preferably the tetrazine is a compound according to structure 1 and the .sup.18F-labelled azide synthon is a compound according to structure [.sup.18F]c.

7. Use of a .sup.18F-labelled tetrazine according to claim 1 for imaging the accumulation of tumors.

Description

BRIEF DESCRIPTION OF THE FIGURES

[0021] FIG. 1 illustrates the dual-click strategy for the 18F-labelling of mAbs-TCO (CC49-TCO) in murine colon carcinoma.

[0022] FIG. 2 is a schematic illustration of the blocking assay.

[0023] FIG. 3 shows the results of the blocking assay in terms of the blocking effect on the normalized tumor uptake.

[0024] FIG. 4 shows structures for tetrazines 7-12.

[0025] FIG. 5 shows a semi-preparative HPLC chromatogram for [18F]c ((Rt=1600 sec) on a Luna 5μ C18(2) (100 Å 250×10 mm) column.

[0026] FIG. 6 shows analytical HPLC chromatogram for [18F]c (Rt=1600 sec) on a Luna 5μ C18(2) (100 Å 250×10 mm) column.

DETAILED DESCRIPTION

[0027] The present invention provides a suitable indirect labeling approach, by which the combination of different building blocks would allow access to a library of .sup.18F-labeled Tz-derivatives with high structural diversity, including reactive structures that have previously proven difficult to access. The following detailed description explains the design and synthesis of such a library, as well as its evaluation in a newly developed assay for pretargeting, followed by radiolabeling and later evaluation in pretargeted PET imaging studies.

[0028] For the design and synthesis of the tetrazine library, applicants set out to find an appropriate reaction that would allow for fast and efficient incorporation of fluorine-18 under mild conditions. A reaction that possesses these features is the Cu-catalyzed azide-alkyne [3+2] cycloaddition (CuAAC). The CuAAC has shown great feasibility as an indirect .sup.18F-labeling approach for several PET tracers and several synthons functionalized with either an alkyne or azide moiety have been reported.21-25 As such, this reaction was considered to be suitable for the sensitive Tz-scaffold. It was decided to modify the Tz-scaffold with an alkyne and use a .sup.18F-labeled azido-functionalized synthon. In order to cover a wide spectrum of structural diversity, and thereby be able to study the influence of lipophilicity, metabolic stability and reaction kinetics in the ligation with TCO, six different Tz-scaffolds (1-6) were designed (Table 1). For the azides, a group of three synthons ([.sup.18F]a-c) was selected (Table 1). Overall, the combination of six Tz-building blocks with three different .sup.18F-labeled azides offered the possibility to obtain 18 Tz-derivatives with diverse physicochemical properties (Table 1).

TABLE-US-00001 TABLE 1 Overview of the Tz-library accessible via CuAAC between different building blocks. Building Estimated block Struc- Rate Combi- ture constant.sup.a nation R X Y Linker-F (M.sup.−1 s.sup.31 ) 1a 2- Pyri- dyl N [00006] [00007] 230 1b 2- Pyri- dyl N [00008] [00009] 230 1c 2- Pyri- dyl N [00010] [00011] 230 2a 2- Pyri- dyl N [00012] [00013] 230 2b 2- Pyri- dyl N [00014] [00015] 230 2c 2- Pyri- dyl N [00016] [00017] 230 3a H C — [00018] 204 3b H C — [00019] 204 3c H C — [00020] 204 4a H C [00021] [00022] 72 4b H C [00023] [00024] 72 4c H C [00025] [00026] 72 5a 2- Pyri- dyl C [00027] [00028] 13 5b 2- Pyri- dyl C [00029] [00030] 13 5c 2- Pyri- dyl C [00031] [00032] 13 6a Me- thyl C [00033] [00034] 1.4 6b Me- thyl C [00035] [00036] 1.4 6c Me- thyl C [00037] [00038] 1.4 Notes: Second order rate constants are estimated from measurements of the alkyne-Tz building block with TCO at 25° C. in 1,4-dioxane (see supporting information for details)

[0029] FIG. 1 shows a dual-click strategy for the .sup.18F-labeling of CC49-TCO (mAbs modified with TCO) in murine colon carcinoma. The CuAAC was performed between each tetrazine building block and a .sup.18F-labeled azide synthon (click 1), respectively. The .sup.18F-labeled click product was injected into tumor bearing mice, where it reacted with CC49-TCO that has accumulated at the tumor-site (click 2).

[0030] Pretargeted blocking studies: in order to accelerate the screening throughput of potential secondary imaging agents, an assay for pretargeting was established, in which non-radioactive Tz-derivatives were used. This assay was inspired by standard receptor blocking experiments and based on the pretargeted imaging approach reported by Rossin et al. (Angew Chem Int Edit 2010, 49 (19), 3375-3378). In this approach, an .sup.111In-labeled Tz ([.sup.111]In-DOTA-Tz) was used in pair with CC49-TCO, a non-internalizing mAb, which targets the tumor-associated glycoprotein 72 (TAG72) (Bioconjug Chem 2013, 24 (7), 1210-7.) The set-up for the assay is illustrated in FIG. 2. BALB/c mice bearing LS174T colon carcinoma xenografts were injected intravenously (i.v.) with CC49-TCO (100 μg, 0.67 nmol, ˜10 TCOs/mAb) 72 h prior to i.v injection of a non-radioactive Tz (39 nmol/100 μL, 10 equiv with regard to the molar amount of injected [.sup.111In]In-DOTA-Tz). Two hours after administration of the non-radioactive Tz, the .sup.111In-labeled Tz (˜13 MBq/100 μL, 3.9 nmol) was injected. The animals were euthanized after 22 h and an ex vivo biodistribution study was carried out to assess the potential decrease in tumor uptake of [.sup.111In]In-DOTA-Tz, which could arise as a result from a blocking effect of the TCO-moieties on the mAb by non-radioactive Tz-derivatives. Control experiments, in which mice were injected only with [.sup.111In]In-DOTA-Tz were performed to determine the tumor uptake corresponding to 100%. As a positive control for blocking, the non-radioactive precursor of [.sup.111In]In-DOTA-Tz was used (DOTA-Tz), which blocked ≥98% of the [.sup.111In]In-DOTA-Tz uptake in the tumor.

[0031] FIG. 2 shows the basic methodology of the pretargeted blocking studies.

[0032] FIG. 3 depicts results from blocking assay. The blocking effect on the normalized tumor uptake of [.sup.111In]In-DOTA-Tz was measured after mice have been pre-administered with non-radiolabeled Tz. Data represent mean±S.E.M. from n=3 in each group. In the bars of the figure, the lower the bar, the better the result. It is seen that the best compound is 1c (tetrazine structure 1 and azide structure [.sup.18F]c). While 3c and 4c perform relatively well too, but are outperformed by 1c. The azide structure [.sup.18F]c, having a glucose as the linker and a triazole moiety within the linker (see also N3-linker in FIG. 1), appears to be the only one showing suitable performance. The DOTA-Tz is used as a positive control. In addition to the library presented in this invention, we also compared the performance of tetrazines, which are closely resembling the structures (7-10) in WO 2012/012612. None of these structures showed a good blocking effect. FIG. 4 shows the structures Tz 7-10, the results of which also are included in Table 1.

[0033] In conclusion, it is surprising that in vivo tetrazine ligation was not solely dependent on the speed kinetics of the respective tetrazine (see table 1 and FIG. 3). In fact, the linker displayed a key moiety to carry out the tetrazine ligation in vivo and the outcome varied (see FIG. 3). Best results were obtained using a glucose and a triazole moiety within the linker. However and surprisingly, compound 1c performed by far best in the pretargeting assay and reacted with the TCO-moieties at the tumor site in vivo most efficiently. Structural similar cmpds such as 3c performed worse in the blocking experiment.

General Information

[0034] Radiochemistry was performed at two different institutes:

[0035] Department of Clinical Physiology, Nuclear Medicine & PET, Rigshospitalet, Denmark: [18F]Fluoride was produced via the (p,n)-reaction in a cyclotron (60 mikroA CTI Siemens or 40 mikroA Scanditronix) by irradiating [18O]H2O with a 11 MeV (CTI siemens) or 16 MeV (Scanditronix) proton beam. All QMA anion exchange cartridges (Sep-Pak Accell Plus QMA Plus Light, chloride form, Waters) and C18 cartridges (Sep-Pak C18 Plus Short types) were washed with EtOH (20 mL) and water (20 mL) and dried with air before use. Automated syntheses were performed on a Scansys Laboratorieteknik synthesis module housed in a hot cell. Analytical HPLC was performed on a Dionex system connected to a P680A pump, a UVD 170U detector and a Scansys radiodetector. The system was controlled by Chromeleon 6.8 software. Semi-preparative HPLC was performed on the built-in HPLC system in the synthesis module and the flow rate was set to 3 mL/min at all times. Radio-TLC was carried out on same plates as described for the organic chemistry. The fraction of radioactivity on the plates was measured with an instant imager from Packard and analyzed by Optiquant software.

General Procedure for the Preparation of Anhydrous [18F]Fluoride for Radiolabeling (Scansys Module):

[0036] Irradiated [18O]water containing [18F]F— was passed through an anion exchange resin cartridge (Sep-Pak Accell Plus QMA Plus Light, chloride form). [18F]Fluoride trapped on the QMA was then eluted with 1 mL of a Kryptofix222/K2CO3 solution (4,7,13,16,21,24-hexaoxa-1,10-diazabicyclo[8.8.8]hexacosane (330 mg), K2CO3 (100 mg) and water (0.8 mL) in MeOH (19.2 mL)) into a 4 mL glass vial. The resulting mixture was then gently concentrated to dryness at 90-110° C. via azeotropic drying using 2× MeCN (1 mL) and a stream of helium. The procedure took 25-30 min and yielded in the ready to react [18F]]FK-K222 complex.

(2R,3R,4S,5S,6S)-2-Azido-6-([18F]fluoromethyl)tetrahydro-2H-pyran-3,4,5-triol ([.SUP.18.F]c)

[0037] [.sup.18F]Fluoride was dried according to the general procedures described above. To the dried residue containing [18F]F—, precursor 22 (11 mg, 21 μmol) in dry MeCN (500 μL) was added. The mixture was heated at 100° C. for 7 min, thereafter cooled with air for 5 min, before it was diluted with DMSO:water (1:1, 2 mL) for purification by semi-preparative HPLC. The latter was performed on a Luna 5μ C18(2) (100 Å 250×10 mm) column using MeCN (40%) in aqueous phosphate buffer (10 mM, pH 6) as eluent. The collected HPLC fraction was trapped on a Sep-Pak C18 plus cartridge, which was washed with water (5 mL), followed by aqueous 2 M NaOH (1 mL). On-cartridge deprotection was commenced for 60 seconds and the product was eluted with water (2 mL) into a vial containing AcOH (150 μL) to afford [18F]c in 19% RCY d.c. to the starting amount of activity, a RCP of ≥99%.

[0038] The following CuAAC was performed as following: an aqueous solution of CuSO4.5H2O (16 μL, 100 mg/mL) was mixed with an aqueous solution of sodium ascorbate (16 μL, 300 mg/mL), when the color of the mixture turned yellow a solution of BPDS (80 μL, 50 mg/mL) in water was added. The resulting blue/green mixture was added to a solution of the Tz precursor (˜1 mg) in DMF (100 μL). This mixture was then added to the isolated [18F]c. The mixture was stirred at 120° C. for 1-5 min.

[0039] FIG. 5 shows the preparative HPLC chromatogram and FIG. 6 shows an analytical HPLC diagram identifying [18F]c.

[0040] Applicants have thus developed a small library of .sup.18F-labeled Tz-derivatives, which was accessible by combining different .sup.18F-labeled azide synthons with alkyne-functionalized Tz-derivatives via the CuAAC. With this approach Tz-derivatives with high structural diversity. Surprisingly, the in vivo tetrazine ligation was not only dependent on the speed kinetics of the respective tetrazine (see table 1 and FIG. 3). In fact, the linker displayed a key moiety to carry out the tetrazine ligation in vivo and the outcome varied (see FIG. 3). Best results were obtained using a glucose as the linker and a triazole moiety within the linker (N3-linker in FIG. 1). However and surprisingly, compound 1c performed by far best in the pretargeting assay and reacted with the TCO-moieties at the tumor site in vivo most efficiently.