Summary

Investigations on the Ga(III) Complex of EOB-DTPA and Its 68Ga Radiolabeled Analogue

Published: August 17, 2016
doi:

Summary

A procedure for the isolation of EOB-DTPA and subsequent complexation with natural Ga(III) and 68Ga is presented herein, as well as a thorough analysis of all compounds and investigations on labeling efficiency, in vitro stability and the n-octanol/water distribution coefficient of the radiolabeled complex.

Abstract

We demonstrate a method for the isolation of EOB-DTPA (3,6,9-triaza-3,6,9-tris(carboxymethyl)-4-(ethoxybenzyl)-undecanedioic acid) from its Gd(III) complex and protocols for the preparation of its novel non-radioactive, i.e., natural Ga(III) as well as radioactive 68Ga complex. The ligand as well as the Ga(III) complex were characterized by nuclear magnetic resonance (NMR) spectroscopy, mass spectrometry and elemental analysis. 68Ga was obtained by a standard elution method from a 68Ge/68Ga generator. Experiments to evaluate the 68Ga-labeling efficiency of EOB-DTPA at pH 3.8–4.0 were performed. Established analysis techniques radio TLC (thin layer chromatography) and radio HPLC (high performance liquid chromatography) were used to determine the radiochemical purity of the tracer. As a first investigation of the 68Ga tracers' lipophilicity the n-octanol/water distribution coefficient of 68Ga species present in a pH 7.4 solution was determined by an extraction method. In vitro stability measurements of the tracer in various media at physiological pH were performed, revealing different rates of decomposition.

Introduction

Gadoxetic acid, a common name for the Gd(III) complex of the ligand EOB-DTPA1, is a frequently used contrast agent in hepatobiliary magnetic resonance imaging (MRI).2,3 Due to its specific uptake by liver hepatocytes and high percentage of hepatobiliary excretion it enables the localization of focal lesions and hepatic tumors.2-5 However, certain limitations of the MRI technique (e.g., toxicity of the contrast agents, limited applicability in patients with claustrophobia or metal implants) call for an alternative diagnostic tool.

Positron emission tomography (PET) is a molecular imaging method, wherein a small amount of a radioactive substance (tracer) is administered, upon which its distribution in the body is recorded by a PET scanner.6 PET is a dynamic method that allows for high spatial and temporal resolution of images as well as quantification of the results, without having to deal with the side-effects of MRI contrast agents. The informative value of the obtained metabolic information can be further increased by combination with anatomical data received from additional imaging methods, as most commonly achieved by hybrid imaging with computed tomography (CT) in PET/CT scanners.

The chemical structure of a tracer suitable for PET must include a radioactive isotope serving as positron emitter. Positrons have a short life-span since they almost immediately annihilate with electrons of the atom shells of surrounding tissue. By annihilation two 511 keV gamma photons with opposite direction of movement are emitted, which are recorded by the PET scanner.7,8 To form a tracer, PET nuclides may be bound covalently to a molecule, as is the case in 2-deoxy-2-[18F]fluoroglucose (FDG), the most extensively used PET tracer.7 However, a nuclide may also form coordinative bonds to one or several ligands (e.g., [68Ga]-DOTATOC9,10) or be applied as dissolved inorganic salts (e.g., [18F] sodium fluoride11). Altogether, the structure of the tracer is crucial as it determines its biodistribution, metabolism and excretion behavior.

A suitable PET nuclide should combine favorable characteristics like convenient positron energy and availability as well as a half-life adequate for the intended investigation. The 68Ga nuclide has become an essential force in the field of PET over the last two decades.12,13 This is mainly due to its availability through a generator system, which allows on-site labeling independently from the vicinity of a cyclotron. In a generator, the mother nuclide 68Ge is absorbed on a column from which the daughter nuclide 68Ga is eluted and subsequently labeled to a suitable chelator.6,14 Since the 68Ga nuclide exists as a trivalent cation just like Gd(III)10,13, chelating EOB-DTPA with 68Ga instead would yield a complex with the same overall negative charge as gadoxetic acid. Accordingly, that 68Ga tracer might combine a similar characteristic liver specificity with the suitability for PET imaging. Although gadoxetic acid is purchased and administered as disodium salt, in the following context we will refer to it as Gd[EOB-DTPA] and to the non-radioactive Ga(III) complex as Ga[EOB-DTPA], or 68Ga[EOB-DTPA] in case of the radiolabeled component for the sake of convenience.

To evaluate their applicability as tracers for PET, radioactive metal complexes need to be examined extensively in in vitro, in vivo or ex vivo experiments first. To determine the suitability for a respective medical problem, various tracer characteristics like biodistribution behavior and clearance profile, stability, organ specificity and cell or tissue uptake need to be investigated. Due to their non-invasive character, in vitro determinations are often performed prior to in vivo experiments. It is generally acknowledged that DTPA and its derivatives are of limited suitability as chelators for 68Ga due to these complexes lacking kinetic inertness, resulting in comparably fast decomposition when administered in vivo.14-20 This is primarily caused by apo-transferrin acting as a competitor for 68Ga in plasma. Nevertheless, we investigated this new tracer concerning its possible application in hepatobiliary imaging, wherein diagnostic information may be provided within minutes post-injection3,4,21-23, thereby not necessarily requiring long-term tracer stability. For this purpose we isolated EOB-DTPA from gadoxetic acid and initially performed the complexation with natural Ga(III), which exists as mixture of two stable isotopes, 69Ga and 71Ga. The complex thus obtained served as non-radioactive standard for the following chelation of 68Ga. We used established methods and simultaneously evaluated their suitability for determining the 68Galabeling efficiency of EOB-DTPA and to investigate the lipophilicity of the new 68Ga tracer and its stability in different media.

Protocol

1. Preparation of EOB-DTPA and Ga[EOB-DTPA]

Caution: Please consult all relevant material safety data sheets (MSDS) of the used organic solvents, acids and alkalines before use. Perform all steps in a fume hood and use personal protective equipment (safety glasses, gloves, lab coat).

  1. Isolation of EOB-DTPA from gadoxetic acid
    1. Put 3 ml of 0.25 M gadoxetic acid injectable solution into a flask. Add 500 mg (5.6 mmol) of oxalic acid to the stirred solution.
    2. After stirring for 1 hr, filter the suspension through a frit using reduced pressure. Wash the residue three times with 3 ml of water, respectively.
    3. Combine the aqueous filtrates and equip the solution with a pH electrode. Add 12 M hydrochloric acid to the filtrate until the pH is about -0.1.
    4. Remove the solvent in vacuo to yield a colorless residue. Store under inert gas.
    5. Wash the residue thoroughly (at least three times) with ethyl acetate to remove the excess of oxalic acid. Dry the residue in vacuo.
    6. Redissolve the residue in 2 ml of water at room temperature and then cool the solution in an ice bath. Without removing the ice bath, add 0.5 M aqueous sodium hydroxide solution dropwise until the formation of a colorless gluey solid is observed.
    7. Remove the water by decantation. Wash the solid two more times with 1 ml of cold water. Dry the solid in vacuo to yield the first product fraction.
    8. Isolate a second product fraction from the combined fractions of decanted water via column chromatography (silica, methanol/water 4/1).24 Remove the solvent in vacuo.
    9. If the thus obtained solid is not pure white, redissolve it in 1 ml of water, add 10 ml of ethanol and subsequently 10 ml of diethyl ether to precipitate the product. Filter through a frit using reduced pressure and dry in vacuo.
    10. Combine both solid fractions of EOB-DTPA and perform NMR spectroscopic,25 mass spectrometric26 and elemental27 analyses.
  2. Synthesis of Ga[EOB-DTPA]
    CAUTION: Store solid Ga(III) chloride under a dry inert atmosphere, since upon contact with air, moisture or grease decomposition takes place, resulting in corrosive fumes and formation of yellow, brown or black impurities.
    1. Prepare a 0.11 M stock solution by dissolving 1.94 g (11.0 mmol) of Ga(III) chloride in 100 ml of water. Dilute 1 ml of 25% aqueous ammonia solution with 4 ml of water.
    2. Dissolve 80 mg (0.15 mmol) of EOB-DTPA in a flask in 10 ml of water. If necessary, heat the solvent to achieve complete dissolution.
    3. Add 1.4 ml (0.15 mmol) of the Ga(III) chloride stock solution. Equip the flask with a stirrer and pH-electrode. Add diluted aqueous ammonia solution dropwise until the pH of the solution is approximately 4.1. Stir at room temperature for 30 min.
    4. Remove the solvent in vacuo. Place the residue in a flask, equipped with a stillhead with a central and parallel side neck. Equip the central neck with a cooling finger and the side neck with a vacuum pump outlet
    5. Heat the residue under reduced pressure (125 °C, 0.6 mbar). Periodically remove sublimated ammonium chloride (visible as white coating of the glass surface) from the cooling finger and still head, as well as from the upper parts of the flask with a slightly wet cloth. Continue the process until there is no visible formation of new sublimate.
    6. To remove final traces of ammonium chloride wash the residue three times with 0.5 ml of hot methanol, respectively. Dry the colorless residue in vacuo. Perform NMR spectroscopic,25 mass spectrometric26 and elemental27 analyses.

2. General Labeling Procedure

CAUTION: All experiments including direct or indirect contact with radioactive substances must be undertaken by trained personnel only. Please use appropriate shielding equipment. Collect any radioactive waste separately and store and dispose in accordance with valid regulations.

  1. Elution of the generator
    Note: A 40 mCi 68Ge/68Ga generator with the mother nuclide bound as oxide on dodecyl-3,4,5-trihydroxybenzoate silica was used. Elution and purification may be performed manually or, as was the case in this procedure, as a combined automated process using a peristaltic pump and dispenser unit.
    1. Prepare solutions of 5.5 M, 1.0 M and 0.05 M hydrochloric acid. Prepare a solution of 5.0 M sodium chloride containing 25 µl of 5.5 M hydrochloric acid per ml. Prepare a buffer solution of pH 4.6 by combining 4.1 g sodium acetate, 1 ml HCl (30%) and 2.5 ml glacial acetic acid and diluting the mixture with water to 50 ml.
    2. Precondition the PS-H+ cartridge by flushing it slowly with 1 ml of 1.0 M hydrochloric acid and subsequently 5 ml of water.
    3. Elute the silica column of the generator with 4 ml 0.05 M HCl.12 Load the 68Ga eluate onto the PS-H+ cartridge.
    4. Flush the cartridge with 5 ml of water and subsequently dry it with 5 ml of air. Elute the 68Ga from the cartridge with 1 ml 5.0 M acidified sodium chloride solution.28
  2. Labeling of EOB-DTPA with 68Ga
    1. Dissolve 1 mg (1.9 µmol) of EOB-DTPA in 1 ml of water. From this solution take 100 µl (0.19 µmol) and dilute them with 9.9 ml of water to prepare a 19 µM (10 µg/ml) stock solution of EOB-DTPA.
    2. Remove 50 µl (equaling 22-29 MBq) of the solution containing 68Ga and put into a vial. Add 50 µl (0.5 µg) of a 19 mM stock solution of EOB-DTPA and 300 µl of buffer to raise the pH to 4.0. Shake briefly and incubate the solution at room temperature for 5 min. Remove an aliquot of 1-5 µl and put to HPLC or TLC analysis.
    3. Perform radio HPLC analysis on a reversed phase (RP) C18 column.29 Use the following mobile phase: A – water/trifluoroacetic acid (99.9%/0.1%), B – acetonitrile/trifluoroacetic acid (99.9%/0.1%), gradient: 06 min 80% A → 0% A (0.5 ml/min), 610 min 0% A (0.5 ml/min).
    4. Determine the peak intensities of the radio HPLC signals as area under curve. Calculate the labeling yield as radiochemical purity (RCP) of the tracer as follows:
      RCP = AGa-EOB-DTPA/(AGa + AGa-EOB-DTPA) ∙ 100%
      AGa-EOB-DTPA: area under curve of 68Ga[EOB-DTPA]
      AGa: area under curve of free 68Ga

3. Labeling Efficiency

  1. Perform labeling procedures as described in section 2. Use a consistent range of starting activity of 68Ga eluate, e.g., 22-29 MBq (40-140 µl, depending on the freshness of the eluate).
  2. Add the required amount of buffer solution to adjust the pH to 3.8-4.0 (40-190 µl, depending on the volume of 68Ga eluate). Add the required amount of ligand stock solution (10-70 µl of a 19 mM solution).
  3. Add the required amounts of water to adjust the overall volume of each labeling probe to 1.75 ml. Mix thoroughly and let the sample stand for 5 min at room temperature. Perform HPLC analysis as described in section 2 to determine the labeling yield.
  4. Perform labeling procedures with amounts of ligand between 0.1 µg and 0.7 µg in steps of 0.1 µg. Perform experiments in triplicates for each ligand concentration. Calculate the mean yield and standard deviation.

4. In Vitro Stability

  1. General procedure and preparations
    1. Dissolve a tablet of phosphate buffered saline (PBS) in 200 ml of deionized water to prepare a PBS stock solution with a phosphate concentration of 10 mM.
    2. Perform labeling of 22-29 MBq 68Ga with 0.5 µl of EOB-DTPA stock solution, as described in section 2. Depending on the volume of the 68Ga eluate, adjust the amount of buffer, as described in section 3. Withdraw samples of labeling solution containing 6-12 MBq of tracer to perform stability measurements.
    3. Perform radio TLC analysis on 80 mm silica gel coated aluminum plates using 0.1 M aqueous sodium citrate as eluent and analyze the plates with a TLC radioactivity scanner.30 Determine the intensities of the TLC signals as area under curve. Calculate the RCP of the tracer as follows:
      RCP = AGa-EOB-DTPA/(AGa-free + AGa-EOB-DTPA + AGa-colloidal) ∙ 100%
      AGa-EOB-DTPA: area under curve of 68Ga[EOB-DTPA]
      AGa-free: area under curve of free 68Ga
      AGa-colloidal: area under curve of colloidal 68Ga
    4. Calculate RCPt/RCP0 for every time point. Plot the thus standardized RCP vs. time difference since the starting point t = 0 min.
      RCPt = RCP of 68Ga[EOB-DTPA] at time point t.
      RCP0 = RCP of 68Ga[EOB-DTPA] at t = 0 min.
  2. Stability in phosphate buffered saline (A)
    1. To 65 µl of labeling solution add 150 µl of PBS stock solution and 60 µl of sodium hydroxide solution (0.1 M) to raise the pH to 7.4. Mix thoroughly.
    2. Remove an aliquot of 1-5 µl to perform TLC analysis ('starting point'). Immediately store the solution in an incubator at 37 °C and remove aliquots to perform TLC analysis at representative time points over 3 hr.
  3. Stability towards excess of apo-transferrin in PBS (B)
    1. To 120 µl of labeling solution add 50 µl of PBS stock solution and 430 µl of sodium hydroxide solution (0.1 M) to raise the pH to 7.4. Add 40 µl of a solution of apo-transferrin (25 mg/ml). Mix thoroughly.
    2. Remove an aliquot of 1-5 µl to perform TLC analysis ('starting point'). Immediately store the solution in an incubator at 37 °C and remove aliquots to perform TLC analysis at representative time points over 3 hr.
  4. Stability in human serum (C)
    1. To 500 µl of human serum add 25 µl of labeling solution and 45 µl of sodium hydroxide solution (0.1 M) to raise the pH to 7.4. Mix thoroughly.
    2. Remove an aliquot of 1-5 µl to perform TLC analysis ('starting point'). Immediately store the solution in an incubator at 37 °C and remove aliquots to perform TLC analysis at representative time points over 3 hr.

5. Determination of Distribution Coefficient LogD

  1. Perform labeling procedures as described in section 2. To 50 µl of labeling solution add 20 µl of PBS stock solution and 170 µl of sodium hydroxide solution (0.1 M) to raise the pH to 7.4.
  2. Withdraw 200 µl from that solution and put it into a plastic V-vial. Add 200 µl of n-octanol. Close the vial and vortex for 2 min. Then centrifuge the sample at 1,600 x g for 5 min.
  3. Remove triplicates of 40 µl from the n-octanol phase and the aqueous phase each and put them in separate V-vials. Be careful not to mix up the layers.
  4. Measure the activity of each sample in a gamma well counter for 30 sec. For each sample immediately repeat the measurement twice and thereof calculate the mean activity Ᾱt in counts per minute (cpm). List the thus gained Ᾱt,W1, Ᾱt,W2 and Ᾱt,W3 (activities in aqueous samples) and Ᾱt,O1, Ᾱt,O2, Ᾱt,O3 (activities in n-octanol) along with the respective time point t of their determination.
  5. Define the time point of the measurement of the last sample as t0. Determine and list Δt in min by calculating Δt = t-t0. Perform decay correction of Ᾱt, using the following formula:
    0 = Ᾱt · 2(Δt/68 min).
  6. Calculate Ᾱ0,W as the mean of Ᾱ0,W1, Ᾱ0,W2 and Ᾱ0,W3 as well as Ᾱ0,O as the mean of Ᾱ0,O1, Ᾱ0,O2 and Ᾱ0,O3. Calculate logD using the following formula:
    logD = log[(Ᾱ0,O · 40 µg)/(Ᾱ0,W · 33 µg)].
  7. Perform the entire experiment in triplicates and calculate the mean logD along with its standard deviation.

Representative Results

The ligand EOB-DTPA and the non-radioactive Ga(III) complex were analyzed via 1H and 13C{1H} NMR spectroscopy, mass spectrometry and elemental analysis. The results listed in Table 1 and depicted in Figures 1-6 verify the purity of the substances.

Elution of the 68Ge/68Ga generator yielded solutions of 400-600 MBq 68Ga. The described labeling procedure results in the formation of the desired tracer 68Ga[EOB-DTPA], indicated as radio HPLC peak exhibiting a retention time of 2.8 min (Figure 7). Comparison with the retention time of the Ga[EOB-DTPA] standard in the UV-vis detector at 220 nm (2.7 min, Figure 8) confirms successful labeling. Uncoordinated 68Ga is detected as radio peak at 2.1 min (Figure 7). The 68Ga-labeling efficiency of EO-BDTPA was investigated by determining the labeling yield as a function of the ligand concentration via HPLC (Figure 9). The yields were determined in triplicate and standard deviations were calculated.

Depending on the pH and the concentration of anions present in solution, uncoordinated or non-labeled 68Ga may exist in various species, e.g., gallates or insoluble hydroxide.31 The generalized term "free 68Ga"32 is used for all non-labeled species in solution except the hydroxide, which is generally referred to as "colloidal 68Ga". Under the described analysis conditions, free 68Ga moves with the solvent front (Rf = 1.0) on a TLC plate. Colloidal 68Ga cannot be detected via HPLC, while on a TLC plate it appears as activity at the origin (Rf = 0). A representative chromatogram of a TLC plate analyzed with a TLC radioactivity scanner is shown in Figure 10. The tracer exhibits different retention behavior, depending on whether a sample of labeling solution (pH 3.8-4.0, Rf = 0.3) or a sample of physiological pH (Rf = 0.5) was analyzed.

To investigate the stability of the tracer, freshly labeled 68Ga[EOB-DTPA] was added to samples of physiological pH, containing diluted PBS (phosphate concentration 5.5 mM, A), excess of apo-transferrin (1.6 mg/ml in diluted PBS with a phosphate concentration of 0.8 mM, B) and human serum (C), respectively. Over time, the radiochemical purity (RCPt) of tracer in the samples was determined via TLC. The percentage of intact tracer was calculated as the ratio of RCPt at the respective time points and RCP0 at the starting point (Table 2). This was necessary due to the labeling solutions containing tracer of differing RCP0 (93-96%). The thus standardized percentage of intact tracer is depicted as a function of time in Figure 11.

For the determination of logD aqueous samples of tracer in a diluted PBS solution were prepared. The samples were mixed with n-octanol, centrifuged and subsequently aliquots were removed to determine the activity concentration in both phases. Activity values and subsequent calculation of logD thereof are depicted in Table 3. The mean logD value is 3.54 ± 0.08.

Figure 1
Figure 1. 1H-NMR spectrum of EOB-DTPA. The spectrum was recorded in D2O at 400.1 MHz. Please click here to view a larger version of this figure.

Figure 2
Figure 2. 13C{1H}-NMR spectrum of EOB-DTPA. The spectrum was recorded in D2O at 100.6 MHz. Please click here to view a larger version of this figure.

Figure 3
Figure 3. MS of EOB-DTPA (electrospray ionization (ESI), methanol, negative mode). Please click here to view a larger version of this figure.

Figure 4
Figure 4. 1H-NMR spectrum of Ga[EOB-DTPA]. The spectrum was recorded in D2O at 400.1 MHz. Please click here to view a larger version of this figure.

Figure 5
Figure 5. 13C{1H}-NMR spectrum of Ga[EOB-DTPA]. The spectrum was recorded in D2O at 100.6 MHz. Please click here to view a larger version of this figure.

Figure 6
Figure 6. MS of Ga[EOB-DTPA] (ESI, methanol, negative mode), along with a detailed depiction of the isotope pattern of the molecular peak. Please click here to view a larger version of this figure.

Figure 7
Figure 7. Representative HPLC chromatogram of a sample of 68Ga[EOB-DTPA] containing in parts uncoordinated 68Ga, as recorded by the radioactivity detector. Uncoordinated 68Ga exhibits a retention time of 2.1 min, while the tracer is detected at 2.8 min. Please click here to view a larger version of this figure.

Figure 8
Figure 8. Representative HPLC chromatogram of the standard substance Ga[EOB-DTPA], as detected in the UV-vis channel at 220 nm. The retention time of the cold standard is 2.7 min. Please click here to view a larger version of this figure.

Figure 9
Figure 9. Depiction of 68Ga-labeling efficiency of EOB-DTPA. The labeling yield as determined via HPLC is plotted as a function of the concentration of EOB-DTPA (22-29 MBq starting activity, pH 3.8-4.0, 5 min, RT). The standard deviation is depicted by error bars. Please click here to view a larger version of this figure.

Figure 10
Figure 10. Representative TLC chromatogram revealing different 68Ga species. A sample of 68Ga[EOB-DTPA] in diluted PBS (phosphate concentration 5.5 mM, pH = 7.4) was analyzed after 110 minutes of incubation. Exemplary distribution of colloidal 68Ga (Rf = 0), 68Ga[EOB-DTPA] (Rf = 0.5) and free 68Ga (Rf = 1.0) on a 70 mm TLC plate as detected by a TLC radioactivity scanner is presented. Counts are decay corrected. Please click here to view a larger version of this figure.

Figure 11
Figure 11. Stability determinations of 68Ga[EOB-DTPA] in different media. The decay corrected, standardized percentage of intact tracer as determined via TLC, is depicted as a function of time. Please click here to view a larger version of this figure.

Table 1
Table 1. Results of NMR spectroscopic, MS and elemental analyses performed for EOB-DTPA and Ga[EOB-DTPA]. Relative MS peak intensities are given in %, assignment to peaks are given in square brackets. Elemental CHN values were calculated for C23H33N3O11·H2O (EOB-DTPA) and (NH4)0.75H1.25[C23H28GaN3O11]·2H2O (Ga[EOB-DTPA]).

Table 2
Table 2. Stability determination of 68Ga[EOB-DTPA] in different media. The RCP of 68Ga[EOB-DTPA] in media A, B and C was determined via TLC at given time points. The composition of the samples is given as percentages in % of tracer / free 68Ga / colloidal 68Ga. The percentage of intact tracer is standardized as ratio of RCPt/RCP0. RCP0 is the respective RCP of the tracer at t = 0 min.

Table 3
Table 3. Determination of logD. Decay corrected values Ᾱ0,X of three aliquots (x = 1, 2, 3) removed from each phase (W: aqueous, O: n-octanol) of a sample. All activities are given in cpm. LogD is calculated as described in section 5 of the protocol. The experiment was repeated twice.

Discussion

EOB-DTPA is accessible through a multi-step synthesis33 but may just as well be isolated from available contrast agents containing gadoxetic acid. For this purpose, the central Gd(III) ion can be precipitated with an excess of oxalic acid. After removing Gd(III) oxalate and oxalic acid the ligand can be isolated by precipitation in cold water at pH 1.5. However, in order to enhance yields column chromatography of the filtrate can be performed instead or as a follow-up procedure. Either method yields the analytically pure ligand in total yields of 70% (Figures 1-3, Table 1).

We found that in order to isolate Ga[EOB-DTPA] adjusting the pH with ammonia solution is advantageous compared to the use of sodium hydroxide, since the by-product ammonium chloride may be removed from the very hydrophilic residue via sublimation. Under the aforementioned conditions this process takes place slowly. Since non-negligible amounts of chloride were still detectable after five days, the remaining salt was washed out with methanol. Although this work-up procedure results in partial loss of Ga[EOB-DTPA], the product was obtained in analytical purity with an overall yield of 46% (Figures 4-6, Table 1). For the isolation of both EOB-DTPA and its Ga(III) complex, the use of reversed phase chromatography should be considered as an alternative method of purification, especially since decomposition of silica gel is likely when using highly polar solvents.

The labeling process of EOB-DTPA required the use of highly pure solvents, chemicals and metal-free equipment to avoid the presence of competing metal ions, due to 68Ga being present in nanomolar amounts (2 MBq of 68Ga in a 1.75 ml sample equal a nuclide concentration of 0.14 nM). Labeling of EOB-DTPA to 68Ga occurs at pH 3.8-4.0 within five minutes at room temperature. Investigations on the 68Ga-labeling efficiency require determination of the labeling yield while keeping the reaction conditions pH, temperature and reaction time as well as starting activity of 68Ga constant or in a justifiable range. For each data point (i.e., ligand concentration) the experiment should be performed at least three times to provide a reasonable confidence level, since the concentrations of both ligand and 68Ga are very low and the labeling yield therefore sensitive to even slight deviations of the reaction conditions. For example, as the 68Ga eluate ages, aliquots of increasing volume need to be withdrawn to provide a constant starting activity, thereby requiring increasing volumes of buffer. Furthermore, aging of the eluate results in increasing concentrations of the decay product 68Zn, which itself might act as a competitor for 68Ga, thus negatively affecting labeling efficiency.13,34,35 Practically quantitative labeling of 22-29 MBq 68Ga is achieved under the aforementioned conditions with amounts of EOB-DTPA ≥ 0.7 µg (Figure 9), with contents of free 68Ga ≤ 2% and about 5% of colloidal 68Ga present in samples.

While HPLC provided superior baseline separation of free 68Ga and 68Ga[EOB-DTPA], it is not suited to detect colloidal 68Ga. We therefore chose TLC to determine the RCP during stability measurements, wherein the quantification of transferrin or protein-bound 68Ga was required. We found baseline separation acceptable for this purpose (Figure 10); however, the use of size exclusion chromatography or filtration methods15,36 to remove colloidal fractions, followed by HPLC analysis, might be considered as alternatives. The 68Ga complex exhibits a stronger retention on TLC plates (Rf = 0.3) if the sample is withdrawn directly from labeling solution as opposed to samples at physiological pH (Rf = 0.5). We suggest this observation might be explained by different protonation states of the complex.

In vitro stability determinations of 68Ga tracers are usually performed in PBS15,17 or alternative buffer systems mimicking the physiological pH37, as well as in solutions containing apo-transferrin37, which is the main competitor for 68Ga in blood, or in human serum15,17. In our experiments the addition of 0.1 M sodium hydroxide solution to PBS was required to adjust the pH of the samples to 7.4. We could not assert that the phosphate concentration influences the rate of degradation, since stability experiments in solutions of varying phosphate concentration (0.8 mM and 5.5 mM (A)) yielded non-reproducible results. However, we found that a solution B, containing apo-transferrin (1.6 mg/ml, which is within range of the normal plasma content38) and 0.8 mM phosphate (human blood usually exhibits a phosphate level of 0.8–1.5 mM39,40), causes decomposition at a rate comparable to that observable in human serum (C). In solutions AC, after 185 min the content of colloidal 68Ga had increased by about 24%, while the content of free 68Ga had increased by 11% in solution A, 17% in solution B and 27% in solution C (Table 2). The fact that 68Ga formed by tracer decomposition is predominantly present as free 68Ga as opposed to colloidal or protein-bound 68Ga in B and C might be due to transferrin saturation or comparably slow transferrin binding rates.41  The observed overall low stability (Figure 11) of 68Ga[EOB-DTPA] is comparable to tracers featuring similar DTPA derived chelators.15,16,18 Usually, information on the early arterial and venous perfusion phase of the liver are gained by performing MRI scans within the first 3 minutes4,21 after administering Gd[EOB-DTPA], while the hepatocyte presence is detected in the delayed phase 20 minutes3,4,23 up to several hours21,22 after injection. After 20 minutes in human serum 93% of 68Ga[EOB-DTPA] remain intact. Expectedly, the signal-to-noise ratio by that time would have deteriorated due to increasing amounts of 68Ga-transferrin, which is present in plasma and tissue expressing transferrin receptors, as well as free 68Ga gallate.41,42

For predicting a tracers tissue distribution n-octanol/water partition coefficients logP or distribution coefficients logD can be determined as ratio of activity concentrations in the two phases. By definition, the logD parameter does not differentiate between multiple species present in a medium, which makes it suitable for our experiments due to the possibility of different protonation states of the tracer as well as its decomposition in the aqueous phase. To determine logD by extraction the aqueous medium is usually buffered with PBS to mimic blood conditions.17,43-45 For aforementioned reasons we used diluted PBS, exhibiting a phosphate concentration of 0.8 mM and physiological pH. Following extraction with n-octanol and centrifugation, the removal of several aliquots from the same phase allows for inaccuracies caused by pipetting to be reduced. Due to the very low activity concentrations in n-octanol one should be careful to avoid cross-contamination with the aqueous phase and to ensure quantitative transfer into a separate vial. Distribution coefficients determined by this procedure were reproducible, and while they allow for a rough estimation of lipophilicity, a direct comparison to a logP of Gd[EOB-DTPA] is not possible. Due to the specificity of Gd[EOB-DTPA] resulting not primarily from lipophilicity but rather its hepatobiliary uptake additional experiments in living subjects or cells would be necessary to provide more extensive information on the biodistribution as well as stability in vivo of 68Ga[EOB-DTPA]. Altogether, an application as imaging agent for the perfusion and early hepatobiliary phase is imaginable.

Disclosures

The authors have nothing to disclose.

Acknowledgements

The authors have no acknowledgements.

Materials

primovist Bayer 0.25 M
gallium(III) chloride Sigma-Aldrich Co. 450898
water (deionized)  tap water deionizing equipment by Auma-Tec GmbH
hydrochloric acid 12 M VWR 20252.29
sodium hydroxide Polskie Odczynniki Chemiczne S.A. 810925429
oxalic acid Sigma-Aldrich Co. 75688
ethyl acetate Brenntag GmbH 10010447
silica gel Merck KGaA 1.10832.9025 Geduran Si 60 0,063-0,2 mm
TLC silica gel 60 F254 Merck KGaA 1.16834.0001
methanol VWR 20903.55
ethanol Brenntag GmbH 10018366
eiethylether VWR 23807.468 stored over KOH plates
ammonia solution (25 %) VWR 1133.1
pH electrode VWR 662-1657
stirring and heating unit Heidolph 505-20000-00
pump Ilmvac GmbH 322002
frit custom design
NMR spectrometer Bruker Coorporation Ultra Shield 400
mass spectrometer Thermo Fisher Scientific Inc.
elemental analyser Hekatech GmbH Analysentechnik EuroVector EA 3000 CHNS
deuterated water D2O euriso-top D214 99,90 % D
Name Company Catalog Number Comments
Material/Equipment required for labeling procedures
68Ge/68Ga generator ITG Isotope Technologies Garching GmbH A150
pump and dispenser system Scintomics GmbH Variosystem
hydrochloric acid 30 % (suprapur) Merck KGaA 1.00318.1000
water (ultrapur) Merck KGaA 1.01262.1000
sodium chloride (suprapur) Merck KGaA 1.06406.0500
sodium acetate (suprapur) Merck KGaA 1.06264.0050
glacial acetic acid (suprapur) Merck KGaA 1.00066.0250
sodium citrate dihydrate VEB Laborchemie Apolda 10782 >98.5%
PS-H+ Cartridge (S) Macherey-Nagel 731867 Chromafix
apo-Transferrin Sigma-Aldrich Co. T2036
PBS  buffer (tablets) Sigma-Aldrich Co. 79382
human serum Sigma-Aldrich Co. H4522 from human male AB plasma
flasks, columns etc. custom design
pH electrode Knick Elektronische Messgeräte GmbH & Co. KG 765-Set
binary pump (HPLC) Hewlett-Packard G1312A (HP 1100)
UV Vis detector (HPLC) Hewlett-Packard G1315A (HP 1100)
radioactive detector (HPLC) EGRC Berthold
HPLC C-18-PFP column Advanced Chromatography Technologies Ltd. ACE-1110-1503/A100528
HPLC glass vials GTG Glastechnik Graefenroda GmbH 8004-HP-H/i3µ
pipette Eppendorf
plastic vials Sarstedt AG & Co. 6542.007
plastic vials Greiner Bio-One International GmbH 717201
activimeter MED Nuklear-Medizintechnik Dresden GmbH Isomed 2010
tweezers custom design
incubator Heraeus Instruments GmbH 51008815
vortex mixer Fisons Whirlimixer
centrifuge Heraeus Instruments GmbH 75003360
gamma well counter MED Nuklear-Medizintechnik Dresden GmbH Isomed 2100
water for chromatography Merck KGaA 1.15333.2500
acetonitrile for chromatography Merck KGaA 1.00030.2500
trifluoroacetic acid Sigma-Aldrich 91707
TLC radioactivity scanner raytest Isotopenmessgeräte GmbH B00003875 equipped with beta plastic detector

References

  1. Weinmann, H. J., et al. A new lipophilic gadolinium chelate as a tissue-specific contrast medium for MRI. Magn. Reson. Med. 22, 233-237 (1991).
  2. Stroszczynski, C., et al. Aktueller Stand der MRT-Diagnostik mit leberspezifischen Kontrastmitteln. Radiologe. 44, 1185 (2004).
  3. Van Beers, B. E., Pastor, C. M., Hussain, H. K. Primovist, Eovist – what to expect. J. Hepatol. 57, 421-429 (2012).
  4. Zech, C. J., Herrmann, K. A., Reiser, M. F., Schoenberg, S. O. MR Imaging in Patients with Suspected Liver Metastases: Value of Liver-specific Contrast Agent Gd-EOB-DTPA. Magn. Reson. Med. Sci. 6, 43-52 (2007).
  5. Leonhardt, M., et al. Hepatic Uptake of the Magnetic Resonance Imaging Contrast Agent Gd-EOB-DTPA: Role of Human Organic Anion Transporters. Drug Metab. Dispos. 38, 1024-1028 (2010).
  6. Wadas, T. J., Wong, E. H., Weisman, G. R., Anderson, C. Coordinating Radiometals of Copper, Gallium, Indium, Yttrium, and Zirconium for PET and SPECT Imaging of Disease. J. Chem. Rev. 110, 2858-2902 (2010).
  7. Ametamey, S. M., Honer, M., Schubiger, P. A. Molecular Imaging with PET. Chem. Rev. 108, 1501-1516 (2008).
  8. Cutler, C. S., Hennkens, H. M., Sisay, N., Huclier-Markai, S., Jurisson, S. S. Radiometals for Combined Imaging and Therapy. Chem. Rev. 113, 858-883 (2013).
  9. Henze, M., et al. PET Imaging of Somatostatin Receptors Using [68GA]DOTA-D-Phe1-Tyr3-Octreotide: First Results in Patients with Meningiomas. J. Nucl. Med. 42, 1053-1056 (2001).
  10. Hofmann, M., et al. Biokinetics and imaging with the somatostatin receptor PET radioligand 68Ga-DOTATOC: preliminary data. Eur. J. Nucl. Med. 28, 1751-1757 (2001).
  11. Blau, M., Nagler, W., Bender, M. A. Fluorine-18: a new isotope for bone scanning. J. Nucl. Med. 3, 332-334 (1962).
  12. Green, M. A., Welch, M. J. Gallium Radiopharmaceutical Chemistry. Int. J. Radiat. Appl. Instrum. B. 16, 435-448 (1989).
  13. Rösch, F. Past, present and future of 68Ge/68Ga generators. Appl. Radiat. Isot. 76, 24-30 (2013).
  14. Liu, S. The role of coordination chemistry in the development of target-specific radiopharmaceuticals. Chem. Soc. Rev. 33, 445-461 (2004).
  15. Haubner, R., et al. Development of (68)Ga-labelled DTPA galactosyl human serum albumin for liver function imaging. Eur. J. Nucl. Med. Mol. Imaging. 40 (68), 1245-1255 (2013).
  16. Yang, W., Zhang, X., Liu, Y. Asialoglycoprotein Receptor-Targeted Radiopharmaceuticals for Measurement of Liver Function. Curr. Med. Chem. 21, 4-23 (2014).
  17. Chauhan, K., et al. 68Ga based probe for Alzheimer’s disease: synthesis and preclinical evaluation of homodimeric chalcone in β-amyloid imaging. Org. Biomol. Chem. 12, 7328-7337 (2014).
  18. Chakravarty, R., Chakraborty, S., Dash, A., Pillai, M. R. A. Detailed evaluation on the effect of metal ion impurities on complexation of generator eluted 68Ga with different bifunctional chelators. Nucl. Med. Biol. 40, 197-205 (2013).
  19. Clevette, D. J., Orvig, C. Comparison of ligands of differing denticity and basicity for the in vivo chelation of aluminum and gallium. Polyhedron. 9, 151-161 (1990).
  20. Prinsen, K., et al. Development and evaluation of a 68Ga labeled pamoic acid derivative for in vivo visualization of necrosis using positron emission tomography. Bioorg. Med. Chem. 18, 5274-5281 (2010).
  21. Vogl, T. J., et al. Liver tumors: comparison of MR imaging with Gd-EOB-DTPA and Gd-DTPA. Radiology. 200, 59-67 (1996).
  22. Reimer, P., et al. Phase II clinical evaluation of Gd-EOB-DTPA: dose, safety aspects, and pulse sequence. Radiology. , 177-183 (1996).
  23. Ba-Ssalamah, A., et al. MRT der Leber. Radiologe. 44, 1170-1184 (2004).
  24. Scott, R. P. W. . Journal of Chromatography Library. 22A, A137-A160 (1983).
  25. Reichenbaecher, M., Popp, J. . Strukturanalytik organischer und anorganischer Verbindungen. , (2007).
  26. Gross, J. H. . Mass Spectrometry: A Textbook. , (2004).
  27. Ma, T. S., Rittner, R. C. . Modern Organic Elemental Analysis. , (1979).
  28. Mueller, D., et al. Simplified NaCl Based 68Ga Concentration and Labeling Procedure for Rapid Synthesis of 68Ga Radiopharmaceuticals in High Radiochemical Purity. Bioconjugate Chem. 23, 1712-1717 (2012).
  29. Roberts, T. R. Radio-column chromatography. Journal of Chromatography Library. 14, 103-132 (1978).
  30. Roberts, T. R. Radio-thin-layer chromatography. Journal of Chromatography Library. 14, 45-83 (1978).
  31. Green, M. A., Welch, M. J. Gallium radiopharmaceutical chemistry. Nucl. Med. Biol. 16, 435-448 (1989).
  32. Notni, J., Plutnar, J., Wester, H. J. Bone-seeking TRAP conjugates: surprising observations and their implications on the development of gallium-68-labeled bisphosphonates. EJNMMI Res. 2, 13 (2012).
  33. Schmitt-Willich, H., et al. Synthesis and Physicochemical Characterization of a New Gadolinium Chelate: The Liver-Specific Magnetic Resonance Imaging Contrast Agent Gd-EOB-DTPA. Inorg. Chem. 38, 1134-1144 (1999).
  34. Zhernosekov, K., Nikula, T. 68Ga generator for positron emission tomography. , (2012).
  35. Simecek, J., Hermann, P., Wester, H. J., Notni, J. How is 68Ga Labeling of Macrocyclic Chelators Influenced by Metal Ion Contaminants in 68Ge/68Ga Generator Eluates?. ChemMedChem. 8, 95-103 (2013).
  36. Baur, B., et al. Synthesis, Radiolabelling and In Vitro Characterization of the Gallium-68-, Yttrium-90- and Lutetium-177-Labelled PSMA Ligand, CHX-A”-DTPA-DUPA-Pep. Pharmaceuticals (Basel). 7, 517-529 (2014).
  37. Boros, E., et al. RGD conjugates of the H2dedpa scaffold: synthesis, labeling and imaging with 68Ga. Nucl. Med. Biol. 39, 785-794 (2012).
  38. Beck, W. S. . Hematology. , (1998).
  39. Patel, V., Morrissey, J. . Practical and Professional Clinical Skills. , (2001).
  40. Bartke, A., Constanti, A. . Basic Endocrinology. , (1998).
  41. Bernstein, L. R. Mechanisms of Therapeutic Activity for Gallium. Pharmacol. Rev. 50, 665-682 (1998).
  42. Clausen, J., Edeling, C. J., Fogh, J. 67Ga Binding to Human Serum Proteins and Tumor Components. Cancer Res. 34, 1931-1937 (1974).
  43. Dumont, R. A., et al. Novel 64Cu- and 68Ga-Labeled RGD conjugates show improved PET imaging of αvβ3 integrin expression and facile radiosynthesis [Erratum to document cited in CA156:116856. J. Nucl. Med. 52, 1498 (2011).
  44. Pohle, K., et al. 68Ga-NODAGA-RGD is a suitable substitute for 18F-Galacto-RGD and can be produced with high specific activity in a cGMP/GRP compliant automated process. Nucl. Med. Biol. 39, 777-784 (2012).
  45. Notni, J., Pohle, K., Wester, H. J. Be spoilt for choice with radiolabelled RGD peptides: Preclinical evaluation of 68 Ga-TRAP(RGD)3. Nucl. Med. Biol. 40, 33-41 (2013).

Play Video

Cite This Article
Greiser, J., Niksch, T., Weigand, W., Freesmeyer, M. Investigations on the Ga(III) Complex of EOB-DTPA and Its 68Ga Radiolabeled Analogue. J. Vis. Exp. (114), e54334, doi:10.3791/54334 (2016).

View Video