WO2018148367A1 - Azide scavenging resin - Google Patents

Abstract

A TentaGelc™-based azide resin can rapidly remove unreacted alkyne precursor from [18F]fluoroethylazide radiolabeling reactions leading to dramatically increased specific activities.

Description

AZIDE SCAVENGING RESIN

CROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This application claims priority to U.S. Provisional Patent Application No. 62/456,169, filed on February 8, 2017, in the United States Patent and Trademark Office, and all the benefits accruing therefrom under 35 U.S.C. § 1 19, the content of which is incorporated herein in its entirety by reference.

FIELD OF THE INVENTION

[0002] This invention relates to chemical labeling procedures, and more specifically to an azide resin-based system, use of which results in compounds with increased specific activities.

BACKGROUND

[0003] The copper catalyzed azide-alkyne cycloaddition reaction is a versatile example of "click chemistry" owing mainly to its high yield, chemical orthogonality, water compatibility, and rapid kinetics (Kolb et al. (2001 ) Angew. Chem. In. Ed. 40:2004-2021 ). Click chemistry has found extensive use in the field of fluorine-18 academic radiochemistry where fast reaction kinetics, high yields, and mild reaction conditions are highly desirable. One popular application of fluorine-18 click chemistry involves the radiosynthesis of the prosthetic group [¹⁸F]fluoroethylazide (FEA) followed by its distillation and subsequent reaction with an alkyne precursor (mainly small molecules or peptides).

[0004] While this reaction is rapid and highly efficient, separation of the unreacted precursor from the desired fluorine-18-labeled product can be very difficult through conventional methods such as chromatography, mainly due to the small size difference resulting from conjugation of the fluoroethylazide moiety. As the size of the precursor increases, this separation becomes progressively more difficult, if not impossible on the time scale of fluorine-18 radiochemistry. This poses a challenge for the radiosynthesis of fluorine-18-labeled macromolecules (e.g., peptides and small proteins) with high apparent specific activity for in vivo receptor imaging.

The specific activity for a radiopharmaceutical is a measurement of radioactivity per mass of cold pharmaceutical. For fluorine-18-labeled radiopharmaceuticals, this often becomes a measurement of any co-eluting cold impurity, particularly the precursor compound, which often competes with the radiopharmaceutical for binding to the biological target. This value is also referred to as apparent specific activity, usually expressed in GBq/μΜ, and it is particularly relevant when the imaging target is a saturatable system, such as a cell surface receptor. This metric takes on additional importance when the radiopharmaceutical is synthesized by prosthetic group radiochemistry (e.g., fluoroethylazide), where the difference in structure between precursor and labeled species is subtle and the prosthetic group lies outside the binding site of the target. Low apparent specific activities, therefore, can result in poor tracer uptake and low signal-to-noise in tissues that express the target of interest.

[0005] Conventional methods to increase the apparent specific activity involve either 1 ) the downscaling of the precursor used in the radiosynthesis which may adversely affect reaction yield and kinetics, or 2) time consuming HPLC purification to reduce the precursor concentration in the final formulation. The latter is the preferred method in fluorine-18 radiochemistry, where substantial quantities of precursor are often needed to drive labeling quickly to completion. As discussed above, HPLC-based purifications often fail as the molecular weight of the precursor increases. In addition to adding significant time to the radiosynthesis,

chromatography also introduces additives and solvents (e.g., acetonitrile, TFA) which must be removed in subsequent steps and specifically accounted for during product quality control and Good Manufacturing Practice (GMP) production.

[0006] Thus, what is needed is effective compositions and methods for removing unreacted precursors from [¹⁸F]fluoroethylazide radiolabeling reactions and for obtaining labeled products with increased specific activities. SUMMARY

[0007] An inexpensive, azide-functionalized resin has been developed to rapidly remove unreacted alkyne precursor following the fluoroethylazide labeling reaction and integrated it into a fully automated radiosynthesis platform.

[¹⁸F]fluoroethylazide labeling of 4 different alkynes ranging from < 300 Da to > 1700 Da has been carried out in which 98% of the unreacted alkyne was removed in less than 20 minutes at room temperature to afford the final radiotracers in > 99% radiochemical purity with specific activities up to > 200 GBq/μΜ. This methodology has been applied to label a novel cyclic peptide (previously shown to bind the Her2 receptor with high affinity) and demonstrated tumor-specific uptake and low nonspecific background by PET/CT. This methodology is automated, rapid, mild, and general allowing peptide-based fluorine-18 radiotracers to be obtained with clinically-relevant specific activities without chromatographic separation within currently standard total synthesis times.

[0008] It has been discovered that a TentaGel™-based azide resin can rapidly remove unreacted alkyne precursor from [¹⁸F]fluoroethylazide radiolabeling reactions leading to dramatically increased specific activities. This discovery has been exploited to produce efficient resins useful with a series of model compounds with increasing molecular weight, from a small molecule, /V-Propargylphthalimide, to SUPR4, a cyclic nonapeptide with mid-nanomolar affinity for the Her2 receptor recently obtained by directed evolution of cyclic peptide libraries containing unnatural amino acids.

[0009] In one aspect, the present disclosure provides a azide-functionalized solid support for removing the unreacted alkynes from an azide-alkyne click radiolabeling reaction for ¹⁸F radioradiotracers. In one embodiment, the solid support is a polymer resin. In a specific embodiment, the solid support is a TentaGel™ resin. In another embodiment, the azide-alkyne click radiolabeling reaction is a

[¹⁸F]fluoroethylazide radiolabeling reaction. In yet another embodiment, the alkynes can be strained alkynes or terminal alkynes. In an additional embodiment, one or more azidoacetic acid groups are conjugated with each amine in the underivatized resin. Multiplying the equivalents of azide on the polymer resin results in improved stripping efficiency of the conjugated resin. In certain embodiments, the

radiolabeling reaction can be manual or automated. Specifically, an automated radiolabeling reaction can be performed on commercially available platforms such as, but not limited to, GE Tracerlab and NanoTek automated devices. These devices utilize a stripping resin module, where the composition of the present invention can be incorporated into a cartridge within the automated hands-free workflow system. Thus, the invention relates to an industrial process for radiolabeling incorporating an azide-functionalized resin, where the process includes a solid or semi-solid support for removing unreacted alkynes from an azide-alkyne click radiolabeling reaction for ¹⁸F radioradiotracers.

[0010] In another aspect, the present disclosure provides a method of producing an azide-functionalized solid support, wherein the solid support can be used to remove the unreacted alkynes from an azide-alkyne click radiolabeling reaction for ¹⁸F radioradiotracers. In a specific aspect, the present disclosure provides a method of improving the specific activity of radiolabel compound, wherein an azide- functionalized solid support is used to remove unreacted alkyne precursors from a [¹⁸F]fluoro radiolabeling reaction.

[0011] In one embodiment, the present disclosure provides a method of enhancing the specific activity of [¹⁸F]fluoroethylazide -labeled biologies comprising affibodies, fibronectin-based affinity agents, single-chain variable fragment (ScFv), antigen-binding (Fab) fragment, and antibodies. In another embodiment, the proposed method in the disclosure can be applied to the removal of alkynes from process scale reactions, which can be useful in the petrochemical industry where eliminating alkynes from a feedstock or process line is desirable.

[0012] In one embodiment, the disclosed method can be used to enrich a click reaction product. This reaction product may be difficult to purify or isolate from the unreacted precursors when the reaction is conducted using standard azide- functionalized chemicals without a solid support. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] The foregoing and other objects of the present disclosure, the various features thereof, as well as the disclosure itself may be more fully understood from the following description, when read together with the accompanying drawings in which:

[0014] Figure 1 A is a general schematic representation of the synthesis of an azide-functionalized solid support according to one aspect of the present disclosure;

[0015] Figure 1 B is a general schematic representation of scavenging unreacted alkynes using the azide-functionalized solid support according to one aspect of the present disclosure;

[0016] Figure 2A is a schematic representation of the synthesis of an azide resin according to the disclosure;

[0017] Figure 2B is a schematic representation of [¹⁸F] fluoroethylazide labeling and scavenging during radiosynthesis of compound [¹⁸F]1 ;

[0018] Figure 3 is a graphic representation of the stability of an azide resin according to the disclosure at 4 °C;

[0019] Figure 4 is a set of schematic representations of the [¹⁸F]-labeled compounds synthesized: [¹⁸F]1 , [¹⁸F]2, [¹⁸F]3, and [¹⁸F]4 and [¹⁸F]-SUPR4;

[0020] Figure 5 is a set of radio-HPLC scans of [¹⁸F]1 -/; [¹⁸F]1 -UV₂₂₀, 1 (alklyine precursor-UV₂2o, and [¹⁸F] UV220 using gradient A;

[0021] Figure 6 is a set of radio-HPLC scans of [¹⁸F ]2- [¹⁸F]2-UV₂₂₀ 2 (alklyine precursor- UV220, and [¹⁹F]-2 UV220 using gradient B; [0022] Figure 7 is a set of radio-HPLC scans of [¹⁸F]3-y^", [¹⁸F]3- UV₂₂₀ and 3 (alklyine precursor- UV220, using gradient C;

[0023] Figure 8A is a set of radio-HPLC scans of [¹⁸F]-SUPR4-y^", [¹⁸F]-SUPR- UV220, [¹⁹F]-SUPR-UV₂₂₀, using gradient D;

[0024] Figure 8B is a set of radio-HPLC scans of [¹⁸F]-SUPR4-y^", [¹⁸F]SUPR- UV220, [¹⁹F]-SUPR- UV220, using gradient E;

[0025] Figure 9A is a set of radio-HPLC scans of the synthesis of [¹⁸F]-SUPR;

[0026] Figure 9B is a graphic representation of cell lines with varying levels of Her2 expression (BT474 and SKOV3 (shCTRL);

[0027] Figure 9C is a graphic representation of in vitro binding of SKOV

(shHER2], and MDA-231 [¹⁸F]-SUPR4 in BT474 cells pre-blocked with [¹⁹F]- SUPR4, Cy5-SUPR4, and DOTA-SUPR4;

[0028] Figure 10A is a set of representations of PET/CT scans of axial (Ax), coronal (Cor), and Sagittal (Sag) slices of mice with a Her2-positive SKOV3 subcutaneous tumor which had been treated with [¹⁸F]-SUPR4;

[0029] Figure 10B is a set of PET/CT scans of mouse slices with a SKOV3 (R) or MDA-MB-231 (L) tumor labeled with [¹⁸F]-SUPR4;

[0030] Figure 10C is a graphic representation of the Standard Uptake Values (SUVmean) for each tumor and contralateral muscle obtained from PET/CT images and used to calculate the tumormuscle ratio of [¹⁸F]-SUPR4 uptake in the SKOV3 and MD-MB-231 mouse models; and

[0031] Figure 10D is a set of radio-HPLC scans of mouse plasma collected 1 hour post-injection of [¹⁸F]-SUPR4. DESCRIPTION

[0032] The disclosures of these patents, patent applications, and publications in their entireties are hereby incorporated by reference into this application in order to more fully describe the state of the art as known to those skilled therein as of the date of the invention described and claimed herein. The instant disclosure will govern in the instance that there is any inconsistency between the patents, patent applications, and publications and this disclosure.

[0033] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this disclosure belongs. The initial definition provided for a group or term herein applies to that group or term throughout the present specification individually or as part of another group, unless otherwise indicated.

[0034] An azide-derivatized solid support acts as an efficient scavenger of unreacted alkynes in the click labeling reaction and eliminate the need for HPLC purification. As used herein, a "solid" support is a general term that is intended to include semi-solid and gel forms, as well as colloids. Despite differences in chemical composition and molecular weight, the azide resin removes > 98% of the unreacted alkyne precursors in under 20 minutes at room temperature. The final ¹⁸F-labeled products were determined to have specific activities between 12 GBq/μΜ to 222 GBq/μΜ and radiochemical purities > 99%. All radiosyntheses were carried out on the GE TracerLab™ automatic synthesis module and did not require the use of manual manipulation or HPLC purification. To demonstrate the utility of generating radiotracers with high apparent specific activities using this method, we carried out PET/CT imaging using ¹⁸F-labeled SUPR4 to visualize Her2 expression in mouse models of breast cancer. This is a general technique to enhance the specific activity of large molecule (> 500 Da) ¹⁸F-radiotracers on automated platforms under mild conditions with minimal extension of synthesis time.

[0035] A general representation of the synthetic route to an azide-functionalized solid support is shown in Fig. 1 A. The azide functional groups are used to modify the solid support surface, which enable the solid support to efficiently react with alkyne molecules and act as an alkyne scanvenger. This scanvenging process is shown in Fig. 1 B, which removes unreacted alkynes in a ¹⁸F-labeling reaction. Specific examples according to the present disclosure are presented in the following.

[0036] The PEGylated TentaGel™ resin was selected as the solid support because of its biocompatibility as well as its hydrophilicity and ease of chemical modification. Reaction of amine-derivatized TentaGel™ (TentaGel™-NH₂) with 2- azidoacetic acid (Fig. 2A) resulted in a highly derivatized resin with up to 96% of the reactive amines capped. The resin was found to be highly stable and can be stored up to 6 months at 4 °C in DMF with only 3% loss of reactivity (Fig. 3).

[0037] To test the utility of the resin for pre-clinical and clinical applications, the [¹⁸F]fluoroethyazide method (Glaser ef a/. (2007) Bioconiugate Chem. 18:989-993) was adopted for implementation on the GE TracerLab™ automated radiosynthesis platform. In this approach, aqueous [¹⁸F]fluoride is trapped on an anion-exchange resin and eluted into a reaction vessel containing tosylethylazide where [¹⁸F]FEA is formed by displacement of the tosyl group in the presence of Kryptofix. The resulting [¹⁸F]FEA is distilled into a secondary reaction vessel where labeling of the alkyne precursor is carried out in the presence of copper. While many

radiosyntheses involve the automated production of [¹⁸F]FEA, the click step is often carried out manually. The entire radiosynthesis of [¹⁸F]FEA and subsequent click step was implemented on the automated module GE TracerLab™ where the scavenging unit can be incorporated as part of the purification work-flow in between the click labeling step and the final formulation unit. [¹⁸F]fluoride was trapped into a QMA cartridge, released and dried in the presence of K₂C0₃ and Kryptofix 2,2,2. Production of the prosthetic group [¹⁸F]FEA from tosylethylazide occurred in the TracerLab™ main reactor in 15 minutes at 80 °C, followed by distillation of

[¹⁸F]FEA under flow of nitrogen into a receiving vial containing the alkyne precursor (830 nM) and a mixture of catalysts and ligands (Fig. 2B). Yields for the production of the [¹⁸F]FEA on TracerLab™ were up to 35% non-decay corrected (mean = 29.5% ± 2.9%, n=7). After proceeding for 20 minutes at room temperature, the click reaction was transferred to the azide resin slurry (80 mg) which was pre- activated with the same reagents immediately prior to the start of synthesis. The stripping reaction was allowed to proceed at room temperature with nitrogen agitation followed by dilution into 0.085% NaH₂P0₄, purification on a C18 cartridge, and elution with PBS:EtOH for injection. The total synthesis time is approximately 80 minutes.

[0038] A commercially available alkyne-containing phthalimide was used to demonstrate the feasibility of the present methodology on the GE TracerLab™ automated synthesis platform. The ¹⁸F-labeled product [¹⁸F]1 (Fig. 4) was readily obtained in 24% decay-corrected yield (dcy) in 80 minutes. Analysis of the formulated product by radio-HPLC showed > 99% removal of the unreacted alkyne (Fig. 5) and a radiochemical purity > 99%. The specific activity of the final product was 12 GBq/μΜ with respect to the starting material (Table 1 ).

TABLE 1

Summary of Resin Stripping Efficiency and Radiochemical Results

Percentages are reported as average ± standard deviation. *n=3 repeats.

[0039] Analysis of the azide resin following synthesis showed approximately 10% of the eluted activity was non-specifically trapped indicating relatively low nonspecific binding.

[0040] Having demonstrated the methodology's feasibility using the phthalimide model compound, a similar radiosynthesis was carried out with an N-terminal capped dipeptide alkyne to generate the labeled product 2 with 25% dcy and a specific activity > 200 GBq/μΜ (Fig. 6). This result was particularly encouraging given the presence of an unprotected carboxylic acid in close proximity to the alkyne group which appears to lower the click labeling efficiency by a factor of two relative to compound 1 but has no appreciable effect on the stripping efficiency. This is likely due to the stoichiometric excess of azide groups present on the resin (> 30-fold molar excess) and potentially the enhanced reactivity of polymer-bound azides in the copper catalyzed cycloaddition. [0041] While compounds 1 and 2 can be resolved from their unreacted alkyne precursors by HPLC, larger peptides frequently present a much more daunting chromatographic challenge. To determine if this methodology could be applied to large peptides with molecular weights > 1 ,000 Da, the following was done. The model compound 3 was chosen due to its high molecular weight and diverse side- chain functionalities including the presence of two carboxylic acids and a

carboxamide in close proximity to the reactive alkyne. As seen in Table 1 , the click labeling reaction was quite poor (< 1 %) but the resin stripping efficiency was > 99% by analytical HPLC (Fig. 7).

[0042] Having established that this methodology could enable the synthesis of large peptide radiotracers with high specific activity, the technique was applied to the radiolabeling of SUPR4, a cyclic peptide containing unnatural amino acids (N- methyl norvaline) previously selected for binding to the Her2 receptor by mRNA display. SUPR4 was shown to have mid-nanomolar affinity for the Her2 receptor, rapid renal clearance from circulation, and high Her2-dependent tumor uptake by near-infrared (NIR) optical imaging. Previous radiochemical experiments showed that SUPR4 could be efficiently labeled using the fluoroethyl azide technique, but could not be resolved from the starting material by HPLC (Figs. 8A - 8B) . Given the modest affinity and rapid clearance of this peptide, it was critical to increase the specific activity in order to obtain measureable tumor uptake above background by PET/CT imaging. Using the azide resin methodology, a 40% labeling efficiency of SUPR4 was obtained on the TracerLab™ and > 98% removal of precursor yielding a 99% radiochemically pure product with a specific activity of up to 40 GBq/μΜ (Fig. 9A).

[0043] The ¹⁸F-labeled SUPR4 showed Her2-dependent binding in cell-based assays (Fig. 9B), which was efficiently blocked by both the "cold" version as well as Cy5- and DOTA-conjugated versions of the same peptide (Fig. 9C). These data suggest that the pendant functionality at the C-terminus may play a role in binding of SUPR4 to the Her2 receptor.

[0044] PET/CT imaging of subcutaneous SKOV3 (Her2-positive) and MDA-MB-

231 (Her2-negative) mouse models showed Her2-dependent uptake of the ¹⁸F-

SUPR4 radiotracer with relatively low background uptake in non-tumor tissues

(Figs. 10A - 10C). In addition to uptake in the tumor, significant signal was observed in the gallbladder and kidneys consistent with rapid renal clearance with minor hepatobiliary excretion. Post-imaging analysis of tumormuscle (T:M) uptake ratios showed a modest but statistically significant increase in SKOV3 tumors relative to MDA-MB-231 tumors which showed the expected T:M of 1 indicating no specific probe uptake in the absence of the Her2 receptor (Fig. 10C). Serum analysis by radio-HPLC at 1 hour post-injection showed > 98% of the radiotracer was intact by comparison to an uninjected sample of the radiotracer, again in agreement with previous in vitro serum stability analysis (Fig. 10D).

[0045] In summary, a simple, inexpensive, and automated radiochemical methodology has been created and validated for dramatically enhancing the specific activity of peptides and small molecules by integrating well-established [¹⁸F]fluoroethylazide labeling protocols with a scavenging azide resin. In contrast Previous efforts to enhance specific activity in click-based radiosyntheses have relied on either post-labeling derivatization to increase chromatographic separation of the derivatized precursor or sequestration of unreacted strained alkyne precursors using a conceptually similar azide resin. The present method does not require a chromatography step and represents a significant improvement in both synthesis time and downstream quality control. In contrast to the second strategy, our method has been fully automated on the TracerLab™ platform and is suitable for the processing of multi-Ci levels of input activity. This method also has the advantage of generality: all four alkyne precursors in this study were removed with near quantitative efficiency despite significant differences in size and chemical composition. The conjugation of a propargylglycine-modified precursor peptide (compounds 3 and 4) and the low-molecular weight fluoroethylazide synthon results in only a modest change in the structure of the labeled peptide, which minimizes the probability of altered affinity, selectivity, or biodistribution. This is in contrast to the strain-promoted click reaction which uses large hydrophobic ring systems that may adversely affect the affinity, solubility, and/or biodistribution of the resulting ligand.

[0046] Reference will now be made to specific examples illustrating the disclosure. It is to be understood that the examples are provided to illustrate exemplary embodiments and that no limitation to the scope of the disclosure is intended thereby. EXAMPLES

EXAMPLE 1

Synthetic Chemistry

1 . Synthesis of Azide Scavenging Resin

[0047] TentaGel™ S-NH₂ (1 g, 0.45 μΜ NH₂/g, Sigma) was swelled for 1 hr in DMF followed by extensive washing in DMF. 2-azidoacetic acid (100 μΙ_, 1 .3 mM) and HBTU (500 mg, 1 .3 mM) were dissolved in DMF (7 mL) and added to the washed resin along with /V,/V-Diisopropylethylamine (235 μί, 1 .4 mM). The reaction was allowed to proceed for 1 hr at room temperature (RT) after which the resin was filtered and the reaction was carried out again in the same conditions. After the second coupling, the resin was washed with DMF and DCM and dried under vacuum. A semi-quantitative Kaiser test (Sarin et al. (1981 ) Anal. Biochem, 1 17(1 ): 147-157) was performed on three independently synthesized batches of resin and the azide loading was found to be between 96% and 86% (mean = 91 %) based on residual reactive amino groups. For comparison, the loading efficiency of the same resin exhaustively acylated with acetic anhydride was found to be 93%.

2. Synthesis of 2-Azidoethyl Tosylate

[0048] 2-Azidoethyl tosylate was synthesized according to the protocol described by Demco and Sharpless (Demco et al. (2001 ) J. Org. Chem. 66:7945-7950) with minor modifications. 2-bromoethanol (2.5 g, 20 mM) was refluxed with NaN₃ (1 .56 g, 24 mM) in H₂0 (6 mL) overnight. The reaction was cooled down to RT, dried over MgS0₄ and extracted twice with DCM. The organic layer was dried again over MgS0₄ and filtered to give crude 2-azidoethanol in DCM. To this, p-toluenesulfonyl chloride (3.9 g, 20 mM) and triethylamine (4 mL, 28 mM) were added. The reaction was stirred 90 min under nitrogen at RT then quenched with glycine (300 mg) for 2 hrs at RT. The organic layer was washed twice with 1 M NaOH, dried over MgS0₄, and concentrated by rotary evaporation. The crude product was purified by flash chromatography on silica gel (EtOAc:Hexanes, 3:7) to afford the title compound as a pale yellow oil (892 mg, 19% yield). Analysis by ESI+ (Expected [M+H]⁺ =

242.05. Observed [M+H]⁺ = 241 .85).

3. Synthesis of 2-Fluoroethylazide (FEA)

[0049] 2-Fluoroethylazide (FEA) was synthesized according to the protocol described by Glaser et al (Glaser et al. (2007) Bioconiuqate Chem. 18:989-993). 2- Fluoroethyltosylate (128 mg, 586 μΜ) was dissolved in anhydrous DMF (10 ml_) and reacted with NaN₃ (1 13 mg, 1 .76 mM) for 48 hr under nitrogen. The residual solid was filtered off and 2-fluoroethylazide stored under nitrogen and used directly in the next step with no additional purification. Analysis by ESI+ (Expected [M+H]⁺ = 90.0. Observed [M+H]⁺ = 90.7). Warning - this compound is potentially explosive and should not be concentrated to dryness.

4. Synthesis of 2-(2-(1 -(2-Fluoroethyl)-1 H-1 ,2,3-triazol-4-yl)ethyl)isoindoline-1 ,3- dione:

[0050] Commercially available 2-fluoroethanol (250 mg, 3.9 mM) was dissolved in dry DCM (7.8 ml_) and MsCI (362 μΙ_, 4.68 mM) and triethylamine (822 μΙ_, 5.85 mM) were added at 0 °C. After 1 hr the reaction was quenched with saturated

NaHC0₃, phases were separated, aqueous phases were extracted twice with DCM and the combined organic phases were dried over MgS0₄. After filtration of the salts and evaporation under vacuum, 2-fluoroethyl methanesulfonate was obtained as yellow oil (593.3 mg, 100%) and used in the next step without further

purification. 2-Fluoroethyl methanesulfonate was dissolved in DMF (7.8 ml_) and

NaN₃ (507 mg, 7.8 mM) was added. The mixture was heated at 60 °C overnight, solids filtrated off and the 2-fluoroethylazide solution in DMF (200 μΙ_, 0.1 mM) added to a mixture of A/-(3-butynyl)phthalimide (10 mg, 50 μΜ), CuS0₄ (50 μΙ_, 35 mg/mL water solution) and sodium ascorbate (50 μΙ_, 174 mg/mL PBS solution).

The mixture was stirred overnight at RT, solids were filtrated off, and residue was concentrated under vacuum. The mixture was purified by chromatography on silica gel (EtOAc/hexanes: 1 , 1 ) to afford the title compound 1 as yellow solid (7 mg,

45%). Analysis by ESI+ (Expected [M+H]⁺ = 289.1 . Observed [M+H]⁺ = 288.68). (1 ): ¹ H NMR (500 MHz, Chloroform-d) δ 7.83 (dd, J = 5.4, 3.0 Hz, 2H; Ar), 7.71 (dd, J = 5.5, 3.0 Hz, 2H), 7.54 (s, 1 H; triazole), 4.75 (ddd, J = 46.8, 5.2, 4.1 Hz, 2H; CH₂F), 4.63 (dt, J = 27.0, 4.6 Hz, 2H; CH₂CH₂F), 4.03 (t, J = 7.2 Hz, 2H;

CH₂CH₂C=), 3.16 (t, J = 7.2 Hz, 2H; CH₂CH₂C=).

5. 3-(2-(2-(2-Methoxyphenyl)acetamido)-3-methylbutanamido)-4-oxo-4-((1 - phenylbut-3-yn-1 -yl)amino)butanoic acid

[0051] Commercially synthesized 4-(tert-butoxy)-2-(2-(2-(2- methoxyphenyl)acetamido)-3-methylbutanamido)-4-oxobutanoic acid (GenScript,

Piscataway, NJ) (30 mg, 69 μΜ) was dissolved in dry DCM (700 μΙ_) with HBTU (52 mg, 138 μΜ). DIEA (24 μΙ_, 138 μΜ) was added and allowed to react at 25 °C for 5 min. Next, 1 -phenylbut-3-yn-1 -amine hydrochloride (19.5 mg, 104 μΜ) was added and reacted at 25 °C with end-over-end rotation. After 6 hr, reaction was mixed with

10 ml_ ethyl acetate and the organic phase washed twice with 1 N HCI, twice with saturated NaHC0₃ and twice with brine. The organic phase was dried over MgS0₄.

After filtration of the salts and evaporation under vacuum, the ferf-butyl protecting group was removed by treatment with of a 95:5 mixture of TFA: H₂0 (367.5 μΙ_).

After one hr, the reaction was neutralized with 10 N NaOH (400 μΙ_). Precipitate was dissolved in a 1 : 1 mixture of CH₃CN:H₂0 (400 μΙ_ ) and purified by reverse phase HPLC (Luna® 5 μιη C18(2), LC Column 250 x 21 .2 mm; Phenomenex,

Torrance, CA) using gradient elution (20-55% Buffer B over 10 min, 55-65% Buffer

B over 15 min; Buffer A: dH₂0 with 0.1 % (v/v) TFA, Buffer B: CH₃CN + 0.1 % (v/v)

TFA).. After lyophilization, the title compound was obtained as a white solid (8 mg,

23%). Analysis by ESI+ (Expected [M+H]⁺ = 508.24. Observed [M+H]⁺ = 508.15).

[0052] ¹ H NMR (500 MHz, DMSO, d6 - (CD₃)₂SO) 512.29 (s, 1 H), 8.44 (d, J=8.4

Hz, 1 H), 8.24 (d, J=8.85 Hz, 1 H), 8.04 (d, J=6.94Hz, 1 H), 7.25 (m, 6H), 7.16 (m,

1 H), 6.92 (dd, J=8.3, 1 .0Hz, 1 H), 6.84 (td, J=7.4, 1 .7 Hz, 1 H), 4.89 (q, J=7.7Hz,

1 H), 4.67 (ddd, J=9.7, 8.3, 4.1 Hz, 1 H), 3.89 (t, J=7.3, 4.9Hz, 1 H), 3.7 (s, 3H), 3.56

(d, J=15.08Hz, 1 H), 3.45 (d, J=15.08Hz, 1 H), 2.76 (m, 2H), 2.46 (m, 3H), 1 .88 (m,

1 H), 0.88 (d, J=6.69, 3H), 0.821 (d, J=6.74 Hz, 3H). ¹³C NMR (500 MHz, DMSO, d6 - (CD₃)₂SO) 5172.17, 171 .70, 171 .40, 170.16, 157.58, 141 .76, 131 .09,

128.57,128.57, 128.39, 127.57, 127.34, 127.34, 124.87, 120.56, 1 1 1 .01 , 81 .83,

72.98, 59.81 , 55.73, 52.21 , 49.88, 36.90, 36.75, 30.23, 25.85, 19.37,19.29 6. Synthesis of 4-((2-(1 -(2-Fluoroethyl)-1 H-1 ,2, 3-triazol-4-yl)-1 - phenylethyl)amino)-

3-(2-(2-(2-methoxyphenyl)acetamido)-3-methylbutanamido)-4-oxobutanoic acid

(2)

[0053] Commercially synthesized (GenScript, Piscataway, NJ) 4-(tert-butoxy)-2- (2-(2-(2-methoxyphenyl)acetamido)-3-methylbutanamido)-4-oxobutanoic acid (30 mg, 69 μΜ) was dissolved in dry DCM (700 μΙ_) with HBTU (52 mg, 138 μΜ). DIEA (24 μΙ_, 138 μΜ) was added and the mixture stirred at 25 °C for 5 min. Next, 1 - phenylbut-3-yn-1 -amine hydrochloride (19.5 mg, 104 μΜ) was added and the mixture stirred at 25 °C with end-over-end rotation. After 6 hr, reaction was mixed with ethyl acetate (10 mL) and the organic phase washed twice with 1 N HCI, twice with saturated NaHC0₃ and twice with brine. The organic phase was dried over MgS0₄. After filtration of the salts and evaporation under vacuum, the remaining amber oil was dissolved in a 1 : 1 mixture of CH₃CN:DMF (300 μΙ_). To this were added CuS0₄ (8 mg) and L-ascorbic acid (15 mg) in of H₂0 (500 μΙ_), followed by 100 mM TBTA (2 μΙ_) and then 0.2 M 2-fluoroethylazide (500 μΙ_). The reaction was stirred overnight at 25 °C with end-over-end rotation. Insoluble material was removed by filtration and the clarified solution purified by reverse phase HPLC (Luna® 5 μιη C18(2), LC Column 250 x 21 .2 mm; Phenomenex, Torrance, CA) using gradient elution (20% to 55% Buffer B over 10 min, 55% to 65% Buffer B over 15 min; Buffer A: dH₂0 with 0.1 % (v/v) TFA, Buffer B: CH₃CN + 0.1 % (v/v) TFA). The major peaks were combined and the fe -butyl protecting group was removed by treatment with a 95:5 mixture of TFA:H₂0 (367.5 μί). After one hr, the reaction was neutralized by the addition of 10N NaOH (400 μί). Precipitate was dissolved in a 1 : 1 mixture of CH₃CN:H₂0 (400 μΙ_) and purified by reverse phase HPLC as described above to afford the title compound as a white solid (24.4 mg, 54%).

Analysis by ESI+ (Expected [M+H]⁺ = 597.28. Observed [M+H]⁺ = 597.52).

7. Peptide Synthesis [0054] Peptides were synthesized using an automated peptide synthesizer (Biotage Alstra). Rink Amide MBHA (364 mg, 0.55 μΜ/g, P3 Bio) was swelled in DMF for 10 min, heated to 65 °C followed by washing with DMF.

Fluorenylmethyloxycarbonyl chloride (FMOC) deprotections were performed using two successive reactions at 50 °C for 3 min in 20% 4-Methylpiperidine in DMF. Amino acid couplings were carried out at 65 °C for 7 min using 5 equivalents of amino acid delivered at 0.4 M, 5 equivalents of 1 -[Bis(dimethylamino)methylene]- 1 H-1 ,2,3-triazolo[4,5-b]pyridinium 3-oxide hexafluorophosphate (HATU) delivered at 0.5 M and 10 equivalents 4-methylmorpholine (NMM) delivered at 2 M. Following completion of the amino acid sequence on resin, the N-terminus was capped with 15 equivalents of glutaric anhydride delivered at 0.4 M and 30 equivalents NMM delivered at 2 M for 7 min at 65 °C. The C-terminal MTT-protected Lysine was deprotected by washing with DCM followed by 30 min of continuous washing with 5% TFA and 1 % Triisopropylsilane (TIS) in DCM followed by neutralization of the acid with successive washes of DCM, DMF, 10% NMM in DMF and DMF. Two sequential reactions were performed to complete cyclization at 65 °C for 20 min using 10 equivalents of (7-Azabenzotriazol-1 -yloxy)tripyrrolidinophosphonium hexafluorophosphate (PyAOP) delivered at 0.5 M and 10 equivalents NMM delivered at 2 M. The peptide was cleaved from resin over the course of 3 hr using 95:2.5:2.5 TFA:TIS:H₂0. Cleaved peptide was ether precipitated, dried and filtered with Phenex-PTFE 15 mm syringe filter (Phenomenex). The crude product was purified on a Vydac C-18 reverse phase HPLC column using gradient elution (20% Buffer B for 5 min, 20% to 70% Buffer B over 35 min; Buffer A: dH₂0 with 0.1 % (v/v) TFA, Buffer B: CH₃CN + 0.1 % (v/v) TFA). Lyophilized peptide was

reconstituted in DMSO and quantitated by absorbance at 280 nm.

8. Synthesis of Compound 3

[0055] A 120 mM solution of alkyne precursor in DMSO (5 μΐ, 600 nM) was added to a solution of 180 mM 2-fluoroethylazide in DMF (100 μΙ_, 18 μΜ, 30 equivalents) along with a 100 mM TBTA in DMF (8 μΙ_, 720 nM), CuS0₄/ascorbic acid solution in dH₂0 (100 μΙ_, 64 mM CuS0₄, 340 mM Ascorbic acid), and CH₃CN (140 μί). The reaction was allowed to proceed overnight at RT and purified by preparative reverse-phase HPLC using gradient elution (5% Buffer B to 60% Buffer B over 30 min; Buffer A: dH₂0 with 0.1 % (v/v) TFA, Buffer B: CH₃CN + 0.1 % (v/v) TFA) to yield product 3 after lyophilization (0.18 mg, 17% yield). Analysis by ESI+ (Expected [M+H]⁺ = 1763.84. Observed [M+H]⁺ = 1764.41 ).

9. Synthesis of SUPR4 (4)

A 16.6 mM solution of alkyne precursor in DMSO (36 μ ΐ, 600 nM) was added to a solution of 180 mM fluoroethylazide in DMF (64 μί, 11.5 μΜ, 19 equivalents) along with 100 mM TBTA in DMF (8 μΙ_, 720 nM), CuS0₄/ascorbic acid solution in dH₂0 (100 μΙ_, 64 mM CuS0₄, 340 mM Ascorbic acid), dH₂0 (20 μΙ_), and CH₃CN (224 μί). The reaction was allowed to proceed overnight at RT and purified by preparative reverse-phase HPLC using gradient elution (5% Buffer B to 60% Buffer B over 30 min; Buffer A: dH₂0 with 0.1 % (v/v) TFA, Buffer B: CH₃CN + 0.1 % (v/v) TFA) to yield product 4 after lyophilization (0.48 mg, 60% yield). Analysis by ESI+ (Expected [M+H]⁺ = 1344.69. Observed [M+H]⁺ = 1344.41 ).

10. Synthesis of SUPR4-Cy5

[0056] SUPR4-Alkyne (3 mg) dissolved in DMSO (250 μΙ_), Cul (2.3 mg, μΜ, 7 equivalents) and Ascorbic acid (7.2 mg, 22 equivalents) dissolved in DMF, 100 mM TBTA (20 μΙ_, 1 .1 equivalents) dissolved in DMF:dH₂0 (3: 1 ) and N, N- Diisopropylethylamine (DIPEA; 8 μΙ_, 25 equivalents) were added to Cyanine5- Azide (1 .7 mg, 1 .1 equivalents) in a total volume of 1 ml_. The reaction was end- over-end rotated for 4 hr at RT. The reaction was purified on a Vydac C-18 reverse phase HPLC column using gradient elution (25% Buffer B for 5 min, 25% to 75% Buffer B over 35 min; Buffer A: dH₂0 with 0.1 % (v/v) TFA, Buffer B: ACN + 0.1 % (v/v) TFA). Dried peptide was reconstituted in DMSO and quantitated by

absorbance at 648 nm. Analysis by ESI+ (Expected [M]⁺ = 1820.0. Observed [M]⁺ = 1820.3).

1 1 . Synthesis of SUPR4-DOTA

[0057] SUPR4-Cys (3 mg) dissolved in DMSO (250 μΙ_) and 1 ,4,7,10- Tetraazacyclododecane-1 ,4,7-tris-acetic acid-10-maleimidoethylacetamide (3.9 mg, 3 equivalents) dissolved in DMF (250 μΙ_) were added to Phosphate Buffer (500 μΙ_, pH 7.5) and end-over-end rotated at RT for 2+ hr. The reaction was purified on a Vydac C-18 reverse phase HPLC column using gradient elution (25% Buffer B for 5 min, 25% to 75% Buffer B over 35 min; Buffer A: dH₂0 with 0.1 % (v/v) TFA, Buffer B: ACN + 0.1 % (v/v) TFA). Dried peptide was reconstituted in DMSO and quantitated by absorbance at 280 nm. Analysis by ESI+ (Expected [M+H]⁺ = 1789.87. Observed [M+H]⁺ = 1790.60).

EXAMPLE 2

Radiochemistry

[0058] Radiosyntheses were performed on TracerLab™ FX (General Electric Healthcare, Munster, Germany) automatic module. [¹⁸F]Fluoride was obtained as an aqueous solution from the MD Anderson Cyclotron Radiochemical Facility (CRF). [¹⁸F]Fluoride was adsorbed on an ion exchange cartridge (Pre-conditioned Sep-PAK® Light QMA Cartridge, ABX GmbH, Radeberg, Germany). [¹⁸F]Fluoride was flushed into the reaction vial with a potassium carbonate and Kryptofix 2.2.2. water/CH₃CN solution (700 μΙ_; 52.8 mg K₂C0₃, 240.1 mg K₂₂₂ , 4 mL water, 16 mL CH₃CN). The solution was dried under vacuum and under nitrogen flow at 60 °C for 2 min. 500 μί dry CH₃CN was added and then the mixture was azeotropically dried at 120 °C for additional 3 min.

[0059] Synthesis of [¹⁸F]2-fluoroethylazide was carried out by adding 2- azidoethyltosylate precursor (5 mg) in CH₃CN (0.5 m) to the dried [¹⁸F]fluoride and stirring at 80 °C for 15 min. Distillation of the volatile [¹⁸F]2-fluoroethylazide was performed under N₂ flow for 2.5 min at 60 °C into a receiving vial containing CuS0₄ (50 μί; 35 mg/mL in water), sodium ascorbate (50 μί; 174 mg/mL in PBS), TBTA (13 μί; 100 mg/mL in DMF) and piperidine (13 μί; 20% in DMF) and alkyne precursor (830 nM in 25 μί DMF). Click radiolabeling of the alkyne precursor was performed at RT for 20 min.

[0060] The mixture was transferred into a plastic solid phase scavenging reactor containing the azide resin (800 μΙ_; 80 mg/mL) pre-swollen in DMF, washed with a CuS0₄ and sodium ascorbate mixture (200 μί), and loaded with CuS0₄ (50 μί; 35 mg/mL in water), sodium ascorbate (50 μί; 174 mg/mL in PBS), TBTA (13 μί;

100 mg/mL in DMF) and piperidine (13 μί; 20% in DMF). The resulting slurry was agitated for 20 min at RT after which the resin was filtered off and washed with DMF (200 μί). The combined filtrate was transferred into a quenching vial containing 0.085% (v/v) NaH₂P0₄ in water (15 mL) before being loaded on to a light C18 cartridge (Sep-PAK® Light, Waters, Milford, USA). The cartridge was washed with 6 mL of water, dried under nitrogen and eluted with 1 mL of ethanol (compound [¹⁸F]1 ) or 1 :1 ethanol/PBS (compounds [¹⁸F]2 - [¹⁸F]4). The overall synthesis time was approximately 80 min. Activity was determined by dose calibrator and a sample was taken for quality control (QC).

[0061] QC was performed by analytical radio-HPLC on a C18 column (Econosil

C18, 10 μπη, 250 mm, 4.6 mm), a water (0.1 % (v/v) TFA) and CH₃CN (0.1 % (v/v)

TFA) gradient (5% B→ 60% B in 30 min) with a flow of 1 mL/min. The identity of the radiolabeled compound was confirmed by co-elution of the original cold standard (synthesis described above). Prior to QC, a standard curve was constructed on the same column using a series of starting material concentrations to determine the relationship between peak area (UV detection) and precursor concentration. This standard curve was then used to determine the amount of unreacted starting material in the final formulated product. To assess the scavenging efficiency of the resin as single unit, the filtrate was analyzed by analytical HPLC immediately after filtration (no SepPak purification) to assess the remaining concentration of alkyne precursor as described above.

EXAMPLE 3

Biological Studies

[0062] All cell lines were maintained at 37 °C with 5% carbon dioxide and 95% humidified air. BT474 cells were grown in DMEM/F-12 (50/50) with 3.6 μg/mL human insulin (Sigma), 10% fetal bovine serum (FBS) and 1 %

penicillin/streptomycin (PS). MDA-MB-231 cells were grown in RPMI 1640 with 10% FBS and 1 % PS. SKOV3 shRNA control (shCTRL) and shRNA Her2 knockdown (shHER2) cell lines were generated by the University of Texas MD Anderson Cancer Center shRNA & ORFeome Core Facility using Dharmacon GIPZ lentiviral shRNAs (Genecode V3LHS/635339) according to manufacturer protocols. SKOV3 shCTRL and SKOV3 shHER2 were maintained using McCoy's 5A media supplemented with 10% FBS, 1 % PS and 2-3 μg/mL puromycin.

1 . [¹⁸F]SUPR4 Cell Binding: Cell Panel Experiment

[0063] BT474, SKOV3 (shCTRL), SKOV3 (shHER2), and MDA-MB-231 cells were seeded into 12 well plates (200,000 cell per well) and allowed to incubate for 24 hr. Wells were aspirated and fresh media containing [¹⁸F]SUPR4 (20 μθϊ) was added to each well for 30 min at 37 °C. After the 30 min incubation, cells were washed 3x with fresh media and Ix with PBS. Cells were then trypsinized and quadruplicate (n=4) samples of each cell line were counted for activity using a Packard Cobra Quantum Gamma Counter. 10 μί of each sample was used for cell counting on a BIO-RAD TC20 automated cell counter for normalization. Data was analyzed using unpaired t-tests on GraphPad Prism 6 (** p<0.01 ; *** p<0.001 , **** p<0.0001 ).

2. [¹⁸F]SUPR4 Cell Binding: Blocking Experiment [0064] 50,000 BT474 cells were seeded into 24 well plates. After 48 hr, media was removed and replaced with fresh serum-free media for 1 hr. Well for blocking were aspirated and incubated with 20 μΜ SUPR4, SUPR4-Cy5 or SUPR4-DOTA for 15 min prior to 15 μθί of [¹⁸F]SUPR4 being added to each well for 30 min. Each condition was done in quadruplicate (n=4). After incubation, cells were washed 3x with media and 1x with PBS. 1 M NaOH + 0.5% sodium dodecyl sulfate (SDS) was then added to each well. After 5 min the solution was removed and counted for activity on a Packard Cobra Quantum Gamma Counter. Data was analyzed using unpaired t-tests on GraphPad Prism 6 (**** p<0.0001 ).

3. [¹⁸F]SUPR4 PET/CT and Data Analysis

[0065] Animal procedures were carried out with approval of MD Anderson Cancer Center's Institutional Animal Care and Use Committee. Female athymic nude mice were subcutaneously injected with 1 .5 x 10⁶ SKOV3 or MDA-MB-231 cells near the shoulder. After 8 wk to 10 wk, mice were intravenously injected with 100 μθϊ of [¹⁸F]-SUPR4 via the tail vein. PET/CT images were acquired 1 hr post injection on a Bruker Albira PET/CT/SPECT Preclinical Imaging System and iteratively reconstructed using an iterative MLEM algorithm. PET/CT data was analyzed and quantitated using PMOD v3.505 software. Volumes of interest were drawn manually on CT images and used to calculate SUV, %ID/g, and tumor-to- muscle ratios on corresponding registered PET images. Statistical analysis was performed using unpaired t-tests in Graphpad Prism 6. Representative images were generated using the Siemens Inveon Research Workplaces v4.2 software package.

EQUIVALENTS

[0066] Those skilled in the art will recognize, or be able to ascertain, using no more than routine experimentation, numerous equivalents to the specific embodiments described specifically herein. Such equivalents are intended to be encompassed in the scope of the following claims.

Claims

1 . A composition comprising: a solid support having azide-functionalized reactive groups.

2. The composition of claim 1 , wherein the solid support is a polymer resin.

3. The composition of claim 2, wherein the polymer resin includes an

amine-derivatized TentaGel™ resin.

4. The composition of claim 1 , which is represented by the following

structure:

5. The composition of claim 1 , used in an industrial process comprising an azide-alkyne click reaction.

6. The composition of claim 1 , used in an industrial process comprising an

[¹⁸F]fluoroethylazide radiolabeling reaction, wherein the radiolabeling reaction is manual or automated.

7. The composition of claim 1 , used in an industrial process comprising an azide-alkyne click reaction, wherein an alkyne used in the click reaction is a strained alkyne or a terminal alkyne.

8. The composition of claim 1 , used in an industrial process comprising an azide-alkyne click reaction, wherein the solid support is used in process scale reactions to remove unreacted alkynes from the click reaction mixture.

9. The composition of claim 1 , wherein a plurality of azidoacetic acid groups are conjugated with an amine group on the solid support.

10. The composition of claim 1 , used in an industrial process comprising an azide-alkyne click reaction, whereby the composition is used to enrich a clicked reaction product in a separation step, wherein the reaction product recovery yields are increased using solid support.

1 1 . A method of producing an azide-functionalized solid support composition, comprising conjugating an azido-compound with a solid support, thereby conjugating azidoacetic acid groups with amine groups on the resin, thereby producing an azide-functionalized solid support.

12. The method of claim 1 1 , wherein the azido-compound is 2-azidoacetic acid.

13. The method of claim 1 1 , wherein the solid support includes an amine- derivatized TetaGel™ resin.

14. The method of claim 1 1 , wherein a plurality of azidoacetic acid groups are conjugated with an amine group in the resin.

15. A method comprising: using an azide-functionalized solid support to

remove unreacted alkyne precursors from a [¹⁸F]fluoro radiolabeling reaction, wherein the specific activity of a resulting radiolabeled compound is increased.

16. The method of claim 15, wherein the [¹⁸F]fluoro radiolabeling reaction is a

[¹⁸F]fluoroethylazide radiolabeling reaction, which reaction can be manual or automated.

17. The method of claim 15, wherein the radiolabeled compound is a biologic compound comprising, an antibody, an antibody fragment, a bispecific antibody, an affibody, a fibronectin-based affinity agent, a single-chain variable fragment (ScFv), an antigen-binding (Fab) fragment or a polypeptide.

The method of claim 17 wherein the yield and the specific activity of th biologic is increased.