Nuclear Medicine and Biology xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Nuclear Medicine and Biology journal homepage: www.elsevier.com/locate/nucmedbio
Cyclotron production of high purity 44 CaCO3 targets
44m,44
Sc with deuterons from
C. Alliot a,b,⁎, R. Kerdjoudj c, N. Michel a,c, F. Haddad a,c, S. Huclier-Markai a,c a b c
GIP Arronax, 1 rue Arronax, BP 10112, 44817 Saint-Herblain, France Inserm U892, Centre de Recherche en Cancérologie Nantes - Angers, Institut de Biologie, 9 quai Moncousu, 44035 Nantes cedex 01, France Subatech Laboratory, UMR 6457, Ecole des Mines de Nantes, IN2P3/CNRS, Université de Nantes, 4 rue Alfred Kastler, 44300 Nantes, France
a r t i c l e
i n f o
Article history: Received 10 December 2014 Received in revised form 3 March 2015 Accepted 5 March 2015 Available online xxxx Keywords: 44m Sc/44Sc In-vivo generator Purification process Isotope production
a b s t r a c t Introduction: Due to its longer half-life, 44Sc (T1/2 = 3.97 h) as a positron emitter can be an interesting alternative to 68Ga (T1/2 = 67.71 min). It has been already proposed as a PET radionuclide for scouting bone disease and is already available as a 44Ti/44Sc generator. 44Sc has an isomeric state, 44mSc (T1/2 = 58.6 h), which can be coproduced with 44Sc and that has been proved to be considered as an in-vivo PET generator 44mSc/ 44Sc. This work presents the production route of 44mSc/44Sc generator from 44Ca(d,2n), its extraction/purification process and the evaluation of its performances. Methods: Irradiation was performed in a low activity target station using a deuteron beam of 16 MeV, which favors the number of 44mSc atoms produced simultaneously to 44Sc. Typical irradiation conditions were 60 min at 0.2 μA producing 44 MBq of 44Sc with a 44Sc/44mSc activity ratio of 50 at end of irradiation. Separations of the radionuclides were performed by means of cation exchange chromatography using a DGA® resin (Triskem). Then, the developed process was applied with bigger targets, and could be used for preclinical studies. Results: The extraction/purification process leads to a radionucleidic purity higher than 99.99% ( 43Sc, 46Sc, 48 Sc b DL). 44mSc/44Sc labeling towards DOTA moiety was performed in order to get an evaluation of the specific activities that could be reached with regard to all metallic impurities from the resulting source. Reaction parameters of radiolabeling were optimized, reaching yields over 95%, and leading to a specific activity of about 10–20 MBq/nmol for DOTA. A recycling process for the enriched 44Ca target was developed and optimized. Conclusion: The quality of the final batch with regard to radionucleidic purity, specific activity and metal impurities allowed a right away use for further radiopharmaceutical evaluation. This radionucleidic pair of 44mSc/44Sc offers a quite interesting PET radionuclide for being further evaluated as an in-vivo generator. © 2015 Published by Elsevier Inc.
1. Introduction Current radiotracers for cancer diseases have limitations: imaging must be performed on the same day as tracer production; their halflives (2 h or less) may not be compatible for dosimetry assessment. Most probes may be more efficient if combined with new radionuclides. Recently, the scandium chemistry has revealed a growing interest with an increasing number of papers available on scandium: from 44Ti/ 44Sc generator [1–4], from neutron irradiated Ti [5,6], cyclotron produced 44m Sc/ 44Sc [7–9], natSc [10], 46Sc [11,12] or 47Sc [13–15]. Among them, the most promising is 44Sc with a half-life (T1/2) of 3.97 h and mean positron energy of 0.6 MeV ideal for PET cameras. 44Sc offers a better T1/2 in comparison to 68Ga (T1/2 = 68 min). Many different ways have been investigated to produce 44Sc and 44m Sc. They can be formed using radiation mechanism on 45Sc, by spallation on natFe or natCu. Scandium-44 can also be obtained through the decay of titanium-44 using the so-called 44Ti/44Sc generator, but in ⁎ Corresponding author. E-mail address: [email protected] (C. Alliot).
this case no 44mSc can be obtained. To produce simultaneously 44Sc and 44mSc, medical cyclotrons that currently supply 18F to hospitals can be used via 44Ca(p,n)44,44mSc route. 44 Sc is developed for a new nuclear medical imaging technique using the ability of the 44Sc to emit in 99.9% of the cases a photon with energy of 1.15 MeV and a characteristic emission time of few picoseconds. When calculating radiation doses, the specific emission characteristics of 44Sc make it a unique candidate promising for 3γ-coincidence imaging with potentially improved local resolution, from which the very first results obtained on a small-dimension prototype let foresee a very promising perspective [16]. The long-lived 44mSc isomer (T1/2 = 58.6 h) is of interest, as its longer half-life will allow imaging at later times for more accurate assessment of distribution and absorbed doses. 44mSc decays by internal transition (98.8%) to the ground state ( 44gSc) with an associated small secondary emission (Auger emission 2.74%). The main photon emission (271 keV) will induce some recoil of the nuclei. Due to the large mass difference, this recoil energy is estimated to be only 0.89 eV. Since 44m Sc and 44Sc have the same chemical nature, no change in oxidation state is expected, in contrast to what has been already observed for
http://dx.doi.org/10.1016/j.nucmedbio.2015.03.002 0969-8051/© 2015 Published by Elsevier Inc.
Please cite this article as: Alliot C., et al, Cyclotron production of high purity 44m,44Sc with deuterons from 44CaCO3 targets, Nucl Med Biol (2015), http://dx.doi.org/10.1016/j.nucmedbio.2015.03.002
2
C. Alliot et al. / Nuclear Medicine and Biology xxx (2015) xxx–xxx
140
Nd/140Pr [17,18] and 103Pd/103mRh [19], leading to the release of the isotope from the molecule or radiolysis effects coming by the secondary emission. Consequently 44m Sc/ 44 Sc can be used as an in-vivo generator [9]. In addition, 47Sc (T1/2 = 3.35 d) with its moderate energy β− emissions (162 keV) may be useful for targeted radiotherapy, and could also be used for single photon emission computed tomography (SPECT) imaging. Its photon emission of 159 keV (68%) is ideal for imaging on SPECT cameras. 47Sc has already been investigated for its potential interest for β-RIT [20,21]. Regard to 89Zr (T1/2 = 3.3 d), a radionuclide with longer half-life, is highly preferable for its potential use in immuno-PET. The main drawback is that Zr coordination chemistry requires specific ligands, different from those used with therapeutic radioisotopes. 44/44mSc radionuclides of high specific activity are very attractive for the development of various PET imaging probes to improve the diagnosis of various solid cancers, while enabling further consideration of 47Sc to be used together with 44Sc as a matched pair, for the development of “theranostic” probes (a single molecule labeled with two different radionuclides of the same chemical element) [22,23]. Scandium, with + III as the most common oxidation state, can be easily complexed by the most frequently utilized polyaminocarboxylate chelators. The chemical differences between Sc and Ti or Ca are a major advantage for the separation of no-carrier added 44,44m,47Sc and is expected to be much simpler than the separation of neighboring lanthanides, where the chemical differences are very small. Many authors have previously developed processes to purify scandium from calcium [24–29]. Most of them used solvent extraction. Moreover Kalyanaraman and Khopkar [29] evidenced that scandium can be selectively extracted with mesityl oxide as its thiocyanate complex and then recovers in HCl solution. Nonetheless, calcium seemed to be a significant interfering for quantities higher than 25 mg. These authors evidenced that this quantity is sufficient to cause 2% error in the recovery of scandium. For higher quantities of calcium, the loss of scandium can be important. Vibhute and Khopkar [25] and Karve and Khopkar [28] purified scandium as citrate and ascorbato complex, respectively, using aliquat 336S as extractant. In this case, scandium was obtained in the final solution with some organic acid that could become competing agent as far as radiolabelling was concerned. Radhakrishnan and Owens [27] proposed another route for scandium purification, by liquid–liquid extraction, using tri-n-butyl phosphate as solvent. But in this study the authors evidenced that the selectivity coefficient of scandium as regard to calcium increased when nitric acid concentration increased, while the distribution coefficient of scandium decreased. This meant that, despite the huge amount of calcium used, quantitative recovery of scandium was not easy. Moreover no data were available concerning the other metals which could be present at the end of irradiation. Such viable separation uses precipitation and filtration [30]. This approach takes advantage of the insolubility of Sc(OH)3 either as a precipitate or coprecipitate, and is similar to a technique developed for purifying yttrium from strontium [31,32]. Due to these difficulties, a chromatographic process for purification was thus studied. Rane and Bhatki [26] reported a method using cation exchange resin (DOWEX 50x8) to prepare 45Ca from natural scandium with high resulting specific activity. But this type of resin is not really specific with regard to transition metal. Thus, purification of scandium from transition metal using DOWEX 50 × 8, seems difficult. As the mostly used complexing agent for metals, i.e. DOTA, is not a specific ligand, the presence of others metals requires higher quantities of DOTA for radiolabelling. In order to minimize the metallic impurities, DGA® resin, specific to lanthanides and already used to purify scandium [15], was employed Moreover, for further labeling, recovery of the 44mSc/44Sc from calcium needs to remove bulk alkaline earth and to give the final product in a small volume. Small amounts of residual calcium do not interfere with DOTA chelation [33], but it could be important to minimize this quantity in case of in-vivo injections.
The developed chemical process aimed at recovering scandium while minimizing cationic impurities (i.e. Ca, metals), leading to high specific activities. For this, DOTA radiolabelling was performed in order to evaluate the specific activity with regard to all metallic impurities that could be present in the final source. The extraction and purification process was developed and optimized, together with a recycling process of the initial enriched material, for cost effectiveness purpose. 2. Materials and methods 2.1. Chemicals and reagents Nitric and hydrochloric acid were received as ultrapure solutions (SCP Science). All dilutions were made in Ultrapure water (Millipore, 18.2 MΩ.cm). The commercially available solid phase extraction resin DGA (N,N,N,N-tetra-n-octyldiglycolamide) provided by Triskem® was first eluted with NaOH and rinsed with pure water. Then, it was eluted with 20 mL of HCl 0.1 mol.L−1 to remove all potential metal impurities. Single and multi-elements standards (about 10 ppm SCP Science) were used to determine the partition coefficients. Commercially available 1,4,7,10-tetraazacyclododecane-1,4,7,10-tetraacetic acid (DOTA, Macrocyclics Inc.) was used as received. 2.2. Cyclotron targetry and irradiations To limit contaminants production ( 46Sc, 47Sc, 48Sc) due to the use of natural calcium [7], the irradiation targets are made of 96.9% enriched 44 CaCO3 purchase from Chemgas (density equal to 2.71 g/cm3). In the first time, each target was made of 100 mg of 44CaCO3 pressed under 5 tons to form a 10 mm diameter pellet. The obtained thickness was about 470 μm. To prevent the target contamination during irradiation, the target was placed between two kapton foils. These targets were used to develop the purification process of 44Sc. In the second time, the developed process was checked with larger targets which would be used for preclinical studies. These targets were made of 500 mg of 44 CaCO3 (20 mm diameter, 0.6 mm thickness). On ARRONAX facility, 44Sc was produced using an irradiation station named Nice3 devoted to the low activity productions (few mCi maximum). With this device, our target was placed in air, 6.6 cm away from a 75 μm Kapton foil ensuring the sealing at the end of beamline (Fig. 1). The targets were cooled by air during the irradiation. 2.3. Partition coefficients determinations The distribution coefficient (KD) quantifies the partition of an element between the aqueous solution and the solid phase and is defined as follows: KD ¼
C solid ðmol=g Þ C solution ðmol=LÞ
ð1Þ
where Csolid is the concentration of elements sorbed onto the resin (in g/g of dry resin), and Csolution the total aqueous solution concentration of elements (in mol/L) which remains in solution after equilibration. As the distribution coefficients were calculated as a function of dried resin mass, the percentage of humidity was determined by placing 5 samples of each pre-conditioned resin in oven at 105°C. Moreover as single and multi-elements standards are commercially available in dilute nitric acid. To avoid potential effects of nitrate ions on partition behavior, aliquots of standard solutions were transferred to a pre-cleaned with ultrapure HNO3 2% Teflon beaker. The solutions were evaporated to dryness. After complete evaporation, the solid residues were dissolved in HNO3 or HCl, and the process was repeated three times to ensure complete removal of initial anions. Then pre-conditioned DGA® resin was weighted and dispersed in the studied aqueous solution. The suspensions were stirred for two hours that has been shown
Please cite this article as: Alliot C., et al, Cyclotron production of high purity 44m,44Sc with deuterons from 44CaCO3 targets, Nucl Med Biol (2015), http://dx.doi.org/10.1016/j.nucmedbio.2015.03.002
C. Alliot et al. / Nuclear Medicine and Biology xxx (2015) xxx–xxx
3
Fig. 1. Schematic view of the irradiation station and the associated deuteron energy.
to be sufficient to reach equilibrium. The aqueous phases were collected by filtration and analyzed. A blank solution (without resin) was prepared with each batch of equilibration experiment and analyzed simultaneously with the samples, to determine the initial concentrations of each element. The acidic solutions were collected and were diluted in HNO3 1% prior to analysis. Measurements were performed on a Thermofischer ICAP 6500 DUO ICP-AES. Representative wavelengths for each element were selected, with preference given to higher relative emissions and absence of interferences. Knowing the initial (Cinit) and equilibrium (Ceq) aqueous concentrations of each element, Eq. (1) can be rearranged as followed:
KD ¼
C init −C eq =m C eq =V
ð2Þ
where m (in g) is the weight of dried resin and V (in mL) is the total volume of aqueous solution. Equilibration experiments were replicated using freshly prepared multi-element standard solutions. The final distribution coefficients were calculated based on the arithmetic averages of replicate analysis. The resin concentration (in g/mL) was chosen for each batch to minimize the global uncertainty of partition coefficient. The reproducibility of logarithm of KD was better than 5% for all elements.
to dryness and recovered in concentrated ultrapure hydrochloric acid (9 mol.L −1). The solution was then loaded on a pre-conditioned AG1 × 8 column to retain all metallic impurities (Cu, Co, Fe…). The column was rinsed with 10 mL of HCl 9 mol.L −1 to recover enriched 44Ca. The eluted solution was evaporated to dryness and recovered in a mixture of bicarbonate buffer 0.1 mol.L −1 (pH = 10.33)/methanol. The solution was then filtered through 0.22 μm filters (Cellulose Acetate, Millipore) to discard supernatant, and the residue was dried in an oven at 105°C to remove water and methanol. The obtained solid of calcium carbonate was suitable for manufacturing targets for next irradiations. 2.6. Labeling DOTA with 44m,44Sc To 450 μL of solution of DOTA (i.e. 10 nmol, Macrocyclics Inc.) were added 50 μL (i.e. 2 nmol) of 44Sc and mixed in a 2 mL screw-Cap Wheaton V-bottom vial. The solution was placed in a boiling water bath at 90°C for 30 min and then cooled till room temperature was reached. To test the radiolabelling yield, a radio-TLC was performed by spotting 2 μL onto a TLC Flex Plate (silica gel 60A, F-254, 200 μm, Selecto Scientific) and eluted with a developing solution of 0.04 mol.L − 1 aqueous NH4OAc/Methanol, 50/50 (v/v). The activity distribution on the plates was assessed with a Packard Cyclone Phosphor-Plate imaging system (Perkin Elmer). 3. Results and discussion
2.4. Radiochemical separations 3.1. Cyclotron targetry and irradiations Firstly, the purification process was developed using an artificial target made from natural calcium (100 mg of natural calcium carbonate). A multi-elements standard solution was prepared from single and multielements standard solutions and then added to the artificial target after dissolution. The obtained solution was loaded onto the DGA® resin (200 mg of pre-conditioned dried resin). The recovery of all elements (Al, Ca, K, Fe, Co, Cr, Cu, Mn, Ni, Sc, Zn) as a function of acid nature, concentration and elution volume was determined to define the elution profiles of all elements. The aim was to recover scandium selectively and quantitatively. Then, the purification process was applied to irradiated 44CaCO3 targets. The targets were removed from the target holder and dropped into 4 mol.L −1 of HCl solution. The aqueous solution was loaded onto the DGA® column. At the end of the purification process, the scandium solution was evaporated to dryness and recovered in a small volume of HCl 0.1 mol.L −1. The final solutions were measured by a gamma-detector (ORTEC) with a high resolution HPGe detector, previously calibrated with multi-gamma source from CERCA LEA, to determine the radionucleidic purity and by ICP-AES to determine the concentration of potential major impurities. 2.5. Enriched calcium recycling As DGA® resin is specific to lanthanides, a first solution with metals and calcium was recovered from purification process. It was evaporated
A typical irradiation realized on Nice3 device corresponds to a period of 1 h and intensity of 200 nA. After irradiation, the calcium pellet was recovered and directly dissolved in HCl 4 mol.L −1. An aliquot of this solution was counted by gamma spectrometry to determine the activities of all γ emitters. Under these conditions, the determined activity (44 ± 3 MBq) was compared with the theoretical production (48 MBq) calculated from TALYS code. Even if the results obtained with TALYS code were not better than those obtained with others codes (MCNPX, PHITS), the shape fits rather well the shape of experimental values [34]. This production yield was in the same order of magnitude than the one obtained with protons (57 MBq) for the optimal energy (12 MeV). Moreover, concerning the 44mSc production, the experimental 44mSc/44Sc activity ratio was 2.21 ± 0.13% which corresponded to a cross section ratio equal to 33%, as regard to half-life ratio. This ratio, in good agreement with TALYS calculation (about 2.34%), was significantly higher than the one obtained with 12 MeV protons beam. The cross section ratio for 12 MeV protons was equal to 7.77% (Fig. 2). Consequently, the deuterons route seems to be a promising way to produce non-carrier added 44Sc, while optimizing the 44mSc/44Sc ratio. Other production routes allow a better 44mSc/44Sc ratio by bombarding a 45Sc target [35–37], but the final product is carrier-added, which is a major drawback if one wants to use it for further radiopharmaceutical use. Concerning the by-products, produced by deuterons route at this energy, no 43Sc was detected which is normal since the energy threshold is 17 MeV, and about 0.7 MBq 42K and 0.6 kBq of 43K were determined.
Please cite this article as: Alliot C., et al, Cyclotron production of high purity 44m,44Sc with deuterons from 44CaCO3 targets, Nucl Med Biol (2015), http://dx.doi.org/10.1016/j.nucmedbio.2015.03.002
4
C. Alliot et al. / Nuclear Medicine and Biology xxx (2015) xxx–xxx
Fig. 2. 44mSc/44Sc ratio as a function of the projectile energy.
The potassium impurities can be easily discarded by the purification process. High specific activity of the final batch could be thus expected. 3.2. Partition coefficients Fig. 3 shows the partition coefficient in logarithmic scale on DGA® resin as a function of HCl. A high KD value (higher than 10) at a specific acid concentration means the ion is preferentially retained on the resin, while a low KD indicates the release of the ions to the mobile phase. Although partial separation could be potentially achieved at KD N 1,
large discrepancies in distribution coefficients were required for quantitative separation of analytes from matrix elements in natural samples. Previously, Horwitz et al. [38] and Pourmand and Dauphas [39] studied some of the properties of DGA® extraction chromatographic resins in slurry-packed gravity columns to explore the potential application of these resins. Even if almost all elements were studied by these authors, the chromatographic behavior of scandium on DGA® was not studied [38,39]. To develop a purification process for 44Sc produced from enriched calcium targets, this element plays a key role. KD values for scandium were determined, and partition coefficients for different potential impurities coming from target holder (Figs. 3 and 4) were monitored simultaneously. The partitioning behavior of most studied elements on resin is somehow comparable between our study and published works [38,39]. However, there are some differences. Pourmand et al. and Horwitz et al. reported that Ca(II) was slightly adsorbed for HCl concentration higher than 2 mol.L − 1 (log KD = 1). This was not observed in this work, as Ca(II) remained in mobile phase up to 3 mol.L −1. Concerning Ni(II), Pourmand et al. [39] reported that the cation remained in aqueous solutions whichever HCl concentration was concerned, while it had a similar behavior than Co(II) and Cu(II) in the present study. These discrepancies in element behaviors could not be explained by the different conditions used by all the authors, since all standards were converted to HCl, and all partition coefficients were determined for dried resin). The adsorption of potassium was equally studied, and no significant adsorption was observed in HCl media. The partitioning behaviors of zinc, iron and scandium are relatively similar. In addition, HCl is not the appropriate medium to separate these elements. Aiming to discard these elements, their distribution coefficients were determined in HNO3 media (Fig. 4). Sc(III) was strongly adsorbed whatever HNO3 concentration, while Zn(II) and Fe(III) remained in the aqueous phase. The behavior of zinc was comparable to the one determined by Pourmand et al. [39]. For Fe(III), Horwitz et al. [38] and Pourmand et al. [39] evidenced strong adsorption whatever medium and acid concentration. For these authors, Fe(III) could not be discarded from Sc(III) in HCl or HNO3 media. The difference between their results and ours is a critical point for scandium purification. Indeed, iron impurities can compete with scandium for radiolabelling. 3.3. Radiochemical separations In the following section, the DGA® distribution coefficients were thus used for developing a simple chromatographic method of extraction for the separation of Sc from Ca and all associated metallic impurities contained in 44,44mSc produced source. The recovery of elements, which are not adsorbed and totally recovered at the end of the rinsing step (mainly K, Cu, Co, and Ni), is not shown. The final extraction scheme for Ca, Fe, Zn and Sc was optimized based on the partition
Fig. 3. Distribution coefficients of elements on DGA resin as a function of HCl concentration.
Fig. 4. Distribution coefficients of elements on DGA resin as a function of HNO3 concentration.
Please cite this article as: Alliot C., et al, Cyclotron production of high purity 44m,44Sc with deuterons from 44CaCO3 targets, Nucl Med Biol (2015), http://dx.doi.org/10.1016/j.nucmedbio.2015.03.002
C. Alliot et al. / Nuclear Medicine and Biology xxx (2015) xxx–xxx
coefficient of elements from Figs. 3 to 4 and replicates of elution experiments. The acid concentration of the loading solution on DGA® was fixed at 4 mol.L −1 HCl. Matrix elements, including alkali and transition metals such as Co, Cu, Ni were removed from the resin during the load and subsequent rinse in 4 mol.L −1 HCl. Although more than 90% of calcium could be successfully eluted in 20 mL of 4 mol.L−1 HCl (Fig. 5), significant tailing was observed. This loss of enriched material is not acceptable for cost effectiveness. Nonetheless, as it is necessary to discard other metallic elements before recovering scandium, the remaining 10% will be recovered simultaneously with these elements to optimize purification time. Following this first step of elution, 12 mL of 1 mol.L −1 HNO3 were necessary to elute iron and zinc (Fig. 5). This second step allowed discarding all potential impurities as a contrary of Krajewski et al. [40]. 10 mL of 0.1 mol.L −1 HCl was then necessary to elute quantitatively scandium from the DGA® resin. The eluted solution is then evaporated to dryness and recovered in a small volume of 0.1 mol.L−1 HCl. 44 Sc was separated from the irradiated calcium target using the procedure reported previously. The efficiency of our separation method is 88 ± 3%, which is consistent with the previously reported procedures using DGA® resin [41,42]. Radionucleidic analysis through gamma spectrometry showed that the final solutions contained only 44,44mSc. No other significant radionuclides were evidenced even after a cooling time corresponding to several half-lives of scandium. ICP-AES analyses of the eluates gave a value under the detection limit for stable scandium resulting in a specific activity higher than 50 MBq/μmol. This value was significantly lower than the one obtained by Hoehr et al. [41] (1400 MBq/μmol), but higher values are expected by increasing produced activities. Moreover the calcium content in the purified samples was under detection limit even if this last one was not the most significant competitive element for complexation by DOTA. The other major metallic contaminants were Fe and Al. The total concentration of both elements was 1.14 ± 0.66 ppm, which was lower than the one determined by Severin and Hoehr [7,41]. As DOTA ligands used for radiolabelling is not really specific complexing agents, the total concentration of metallic impurities is a key factor for radiolabelling yield and the total concentration of DOTA. Metal contamination will not increase significantly when increasing produced activity. It was observed experimentally that increasing the activity from 100 MBq/mL to 375 MBq/mL kept the concentration of metallic impurities of approx. 1 ppm for the same final volume of the solution. By contrast the specific activity will increase almost linearly with the produced activity. This assumption was confirmed with larger targets developed for preclinical studies. By increasing the targets size (five times more enriched calcium), the final 44Sc quality was improved with produced activity. Based on the results obtained in this work, optimized irradiation parameters to be used at Arronax for up scaling production are defined as followed: incident deuteron energy of 20 MeV on the pellet; output
5
energy of 10 MeV, a production yield at EOB of 469 MBq/μAh is determined. This configuration corresponds to a target thickness of 800 μm. In this energy range, the activity ratio of 44mSc/ 44Sc is expected to be 0.0248. For a typical irradiation of 4 hours and 50 μA/h, we can expect at the end of irradiation about 2500 mCi and 60 mCi of 44Sc and 44m Sc respectively. 3.4. Enriched calcium recycling Due to the high cost of enriched material, a recycling process of the target was developed. The 4 mol.L−1 HCl and 1 mol.L−1 solutions obtained from the extraction process were mixed and evaporated to dryness. With the mixture bicarbonate/methanol, the kinetic of solvent evaporation was increased, and the solubility of calcium carbonate was lowered [43]. The recovery yield of enriched calcium was 90 ± 2%. The suspension was kept for further process recycling to minimize calcium losses. The recycling targets were irradiated, and no significant difference of production yield was observed. 3.5. Radiolabelling As DOTA is not a specific complexing agent, all cationic metals (for example Fe, Zn) can be labeled by this chelating molecule. Radiolabeling yields from our resulting batch of 44Sc were studied for metal-to-ligand ratio (and not for scandium-to-ligand ratio) ranging from 1/1 to 1/50 for DOTA. The most suitable conditions have been identified as a metal-toligand ratio of 1:3 in acetate buffer at pH = 4, and heating the mixture at 70°C for 20 minutes provided the best labeling yields N95%. These values were in agreement with previous studies performed on 46 Sc-DOTA [12,41], 44Sc-DOTATOC and 44Sc-DOTATATE with eluted from 44Ti generator [1,44]. All studies have evidenced that the specific activity was not the major parameter for radiolabelling by DOTA, but was of major importance for further use with biological vectors. Consequently, the total metal concentration has to be as low as possible to minimize the quantity of DOTA and later on of biological probes. In conditions defined previously for large scale production, the reachable specific activity of 44Sc will be higher than 1 GBq/μmol of DOTA four hours after the end of irradiation (time necessary for chemical purification and radiolabelling). 3.6. Conclusion No carrier added 44Sc production was developed in order to make available 44,44mSc batches with high specific activities. The obtained production yields were about 220 MBq/μA.h for small targets which was in agreement with theoretical calculations. For optimizing the simultaneous production of 44mSc, low energy nuclear reactions with deuterons at 16MeV were chosen, leading to higher cross section and lower amount of co-produced radioisotopes in comparison to proton. A method using DGA® resin has been proposed to separate scandium from carbonate calcium targets. The process allows to remove metallic impurities like Fe, Al, Zn leading to a high specific activity suitable for efficient radiolabelling. This production/purification process with a global recovery yield of about 90% could be easily transferred to a medium-sized cyclotron using deuterons or a biomedical cyclotron using protons. Acknowledgments
Fig. 5. Elution curves for artificial target of 44Ca irradiated with deuteron on a DGA® resin (50–100μm particle size).
The ARRONAX cyclotron is a project promoted by the Regional Council of Pays de la Loire financed by local authorities, the French government and the European Union. This work has been, in part, supported by a grant from the French National Agency for Research called "Investissements d'Avenir", Equipex Arronax-Plus no ANR-11EQPX-0004 and Labex no ANR-11-LABX-0018-01.
Please cite this article as: Alliot C., et al, Cyclotron production of high purity 44m,44Sc with deuterons from 44CaCO3 targets, Nucl Med Biol (2015), http://dx.doi.org/10.1016/j.nucmedbio.2015.03.002
6
C. Alliot et al. / Nuclear Medicine and Biology xxx (2015) xxx–xxx
References [1] Pruszyski M, Majkowska-Pilip A, Loktionova N, Eppard E, Roesch F. Radiolabeling of DOTATOC with the long-lived positron emitter 44Sc. Appl Radiat Isot 2012;70:974–9. [2] Pruszyski M, Loktionova N, Filosofov D, Roesch F. Post-elution processing of 44Ti/44Sc generator derived 44Sc for clinical application. Appl Radiat Isot 2010;68:1636–41. [3] Mausner L, Kolsky K, Joshi V, Srivastava S. Radionuclide development at BNL for nuclear medicine therapy. Appl Radiat Isot 1998;49:285–94. [4] Filosofov D, Loktionova N, Roesch F. A 44Ti/44Sc radionuclide generator for potential application of 44Sc-based PET radiopharmaceuticals. Radiochim Acta 2010;98:149–56. [5] Pietrelli L, Mausner L, Kolsky K. Separation of carrier free 47Sc from titanium targets. J Radioanal Nucl Chem 1992;157:335–45. [6] Bokhari TH, Mushtaq A, Khan IU. Separation of no-carrier-added radioactive scandium from neutron irradiated titanium. J Radioanal Nucl Chem 2010;283:389–93. [7] Severin G, Engle J, Valdovinos H, Barnhart T, Nickles R. Cyclotron produced 44gSc from natural calcium. Appl Radiat Isot 2012;70:1526–30. [8] Müller C, Bunka M, Reber J, Fischer C, Zhernosekov K, Türler A, et al. Promises of cyclotron-produced 44Sc as a diagnostic match for trivalent β–emitters: in-vitro and in-vivo study of a 44Sc-DOTA-folate conjugate. J Nucl Med 2013;54:2168–74. [9] Huclier-Markai S, Kerdjoudj R, Alliot C, Bonraisin A, Michel N, Haddad F, et al. Optimization of reaction conditions for the radiolabeling of DOTA and DOTA-peptide with 44m/44Sc and experimental evidence of the feasibility of an in-vivo PET generator. Nucl Med Biol 2014;41:e36–43. [10] Koumarianou E, Pawlak D, Korsak A, Mikolajczak R. Comparison of receptor affinity of natSc-DOTATATE versus natGa-DOTATATE. Nucl Med Rev 2011;14:85–9. [11] Majkowska-Pilip A, Bilewicz A. Macrocyclic complexes of scandium radionuclides as precursors for diagnostic and therapeutic radiopharmaceuticals. J Inorg Biochem 2011;105:313–20. [12] Huclier-Markai S, Sabatie A, Ribet S, Kubicek V, Paris M, Vidaud C, et al. Chemical and biological evaluation of scandium(III)-polyaminocarboxylate complexes as potential PET agents and radiopharmaceuticals. Radiochim Acta 2011;99:653–62. [13] Polosak M, Kasperek A, Krajewski S, Bilewicz A. Stability of 47Sc-complexes with acyclic polyamino-polycarboxylate ligands. J Radioanal Nucl Chem 2013;295:1867–72. [14] Müller C, Bunka M, Haller S, Köster U, Groehn V, Bernhardt P, et al. Promising prospects for 44Sc/47Sc based theragnostics: application of 47Sc for radionuclide tumor therapy in mice. J Nucl Med 2014;55:1658–64. [15] Cydzik I, Krajewski S, Abbas K, Simonell F, Bulgheroni A, Kasperek A, et al. Labelling of DOTATATE with cyclotron produced 44Sc and 43Sc. Q J Nucl Med Mol Imaging 2012;56:33–4. [16] Grignon C, Barbet J, Bardies M, Carlier T, Chatal J, Couturier O. Nuclear medical imaging using β+/γ coincidences from 44Sc radionuclide with liquid xenon as detection medium. Nucl Instrum Meth A 2007;571:142–53. [17] Zhernosekov K, Filosofov D, Qaim SM, Rösch F. A 140Nd/140Pr radionuclide generator based on physico-chemical transitions in 140Pr complexes after electron capture decay of 140Nd-DOTA. Radiochim Acta 2007;95:319–27. [18] Rösch F, Brockmann J, Lebedev NA, Qaim SM. Production and radiochemical separation of the Auger electron emitter 140Nd. Acta Oncol 2000;39:727–30. [19] Van Rooyen J, Szucs Z, Zeevaar J. A possible in-vivo generator 103Pd/103mRh recoil considerations. Appl Radiat Isot 2008;66:1346–9. [20] Kolsky K, Joshi V, Mausner L, Srivastava S. Radiochemical purification of no-carrieradded scandium-47 for radioimmunotherapy. Appl Radiat Isot 1998;49:1541–9. [21] Mausner LF, Joshi L, Kolsky K, Meinken G, Mease R, Sweet M, et al. Evaluation of chelating agents for radioimmunotherapy with scandium-47. J Nucl Med 1995; 36(Suppl. 5):104 in Proceedings of the meeting of the society of Nuclear Medicine. Minneapolis, United States.
[22] Herzog H, Rösch F, Stöcklin G, Lueders C, Qaim SM, Feinendegen LE. Measurement of pharmacokinetics of yttrium-86 radiopharmaceuticals with PET and radiation dose calculation of analogous yttrium-90 radiotherapeutics. J Nucl Med 1993;34:2222–6. [23] Baum R, Kulkarni H. Theranostics: from molecular imaging using Ga-68 labeled tracers and PET/CT to personalized radionuclide therapy - the bad berka experience. Theranostics 2012;2:437–47. [24] Zhang P, You S, Zhang L, Feng S, Hou S. A solvent extraction process for the preparation of ultrahigh purity scandium oxide. Hydrometallurgy 1997;47:47–56. [25] Vibhute C, Khopkar S. Separation of scandium(III) as citrate complex by extraction with aliquat 336S. Indian J Chem 1985;24A:444–6. [26] Rane A, Bhatki K. Rapid radiochemical separations of strontium-90/yttrium-90 and calcium-45/scandium-46 on a cation exchange resin. Anal Chem 1966;38:1598–601. [27] Radhakrishnan K, Owens T. Separation of scandium and calcium by liquid-liquid extraction using tri-n-butyl phosphate as solvent. J Chem Eng Data 1972;17:478–82. [28] Karve M, Khopkar S. Separation of scandium(III) as ascorbato complex by extraction with aliquat 336S. Bull Chem Soc Jpn 1991;64:655–8. [29] Kalyanaraman S, Khopkar S. Solvent extraction separation of scandium with 4-methyl-3-pentene-2-one as its thiocyanate complex. Anal Chem 1977;49: 1192–994. [30] Duval J, Kurbatov M. Separation of carrier-free scandium from a calcium target. J Am Chem Soc 1953;75:2246–8. [31] Rösch F, Qaim SM, Stöcklin G. Production of the positron emitting radioisotope 86Y for nuclear medical application. Appl Radiat Isot 1993;44:677–81. [32] Avila-Rodriguez M, Nye J, Nickles R. Production and separation of non-carrier-added 86 Y from enriched 86Sr targets. Appl Radiat Isot 2008;66:9–13. [33] Abbas K, Cydzik I, Simonell F, Krajewski S, Kasperek A, Bilewicz A. Cyclotron production of 44Sc - new radionuclide for PET technique. J Labelled Compd Rad 2011;54(Suppl. S1):S53. [34] Blideanu V, Garcia M, Joyer P, Lopez D, Mayoral A, Ogando F, et al. Deuteron cross section evaluation for safety and radioprotection calculations of IFMIF/EVEDA accelerator prototype. J Nucl Mat 2011;417:1271–4. [35] Luo J, Liu R, Jiang L, Liu Z, Sun G, Ge S. Cross-sections of 45Sc(n, 2n)44m, gSc reaction from the reaction threshold to 20 MeV. Radiochim Acta 2013;101:607–12. [36] Bostan M, Qaim SM. Excitation functions of threshold reactions on 45Sc and 55Mn induced by 6 to 13 MeV neutrons. Phys Rev C 1994;49:266–71. [37] Bailey S. Yield ratios for the isomeric pair Sc44m,44 formed in (α, αn) and (α, n) reactions. Phys Rev C 1961;123:579–82. [38] Horwitz E, McAlister D, Bond A, Barrans R. Novel extraction of chromatographic resins based on tetraalkyldiglycolamides: characterization and potential applications. Solvent Extr Ion Exch 2005;23:319–44. [39] Pourmand A, Dauphas N. Distribution coefficients of 60 elements on TODGA resin: application to Ca, Lu, Hf, U and Th isotope geochemistry. Talenta 2010; 81:741–53. [40] Krajewski S, Bilewicz A, Cydzik I, Abbas K, Bulgheroni A, Simonelli F, et al. Cyclotron production of 44Sc for clinical application. Radiochim Acta 2013;101:333–8. [41] Hoehr C, Oehlke E, Benard F, Jaeil Lee C, Hou X, Badesso B, et al. 44gSc production using a water target on a 13 MeV cyclotron. Nucl Med Biol 2014;41:401–6. [42] Bunka M, Zhernosekov K, Eichholzer Y, Schibli R, Turler A. Cyclotron production of scandium-44. Labor für Radio und Umweltchemie der Universität Bern und des Paul Scherrer Institut; 2011. [43] Kan A, Fu G, Tomson M. Effect of methanol on carbonate equilibrium and calcite solubility in a gas/methanol/water/salt mixed system. Langmuir 2002;18: 9713–25. [44] Loktionova N, Filosofov D, Pruszynski M, Roesch F. Design and performance of a novel 5 mCi 44Ti/44Sc radionuclide generator. J Nucl Med 2010;51(Suppl. 2):590.
Please cite this article as: Alliot C., et al, Cyclotron production of high purity 44m,44Sc with deuterons from 44CaCO3 targets, Nucl Med Biol (2015), http://dx.doi.org/10.1016/j.nucmedbio.2015.03.002
Contents lists available at ScienceDirect
Nuclear Medicine and Biology journal homepage: www.elsevier.com/locate/nucmedbio
Cyclotron production of high purity 44 CaCO3 targets
44m,44
Sc with deuterons from
C. Alliot a,b,⁎, R. Kerdjoudj c, N. Michel a,c, F. Haddad a,c, S. Huclier-Markai a,c a b c
GIP Arronax, 1 rue Arronax, BP 10112, 44817 Saint-Herblain, France Inserm U892, Centre de Recherche en Cancérologie Nantes - Angers, Institut de Biologie, 9 quai Moncousu, 44035 Nantes cedex 01, France Subatech Laboratory, UMR 6457, Ecole des Mines de Nantes, IN2P3/CNRS, Université de Nantes, 4 rue Alfred Kastler, 44300 Nantes, France
a r t i c l e
i n f o
Article history: Received 10 December 2014 Received in revised form 3 March 2015 Accepted 5 March 2015 Available online xxxx Keywords: 44m Sc/44Sc In-vivo generator Purification process Isotope production
a b s t r a c t Introduction: Due to its longer half-life, 44Sc (T1/2 = 3.97 h) as a positron emitter can be an interesting alternative to 68Ga (T1/2 = 67.71 min). It has been already proposed as a PET radionuclide for scouting bone disease and is already available as a 44Ti/44Sc generator. 44Sc has an isomeric state, 44mSc (T1/2 = 58.6 h), which can be coproduced with 44Sc and that has been proved to be considered as an in-vivo PET generator 44mSc/ 44Sc. This work presents the production route of 44mSc/44Sc generator from 44Ca(d,2n), its extraction/purification process and the evaluation of its performances. Methods: Irradiation was performed in a low activity target station using a deuteron beam of 16 MeV, which favors the number of 44mSc atoms produced simultaneously to 44Sc. Typical irradiation conditions were 60 min at 0.2 μA producing 44 MBq of 44Sc with a 44Sc/44mSc activity ratio of 50 at end of irradiation. Separations of the radionuclides were performed by means of cation exchange chromatography using a DGA® resin (Triskem). Then, the developed process was applied with bigger targets, and could be used for preclinical studies. Results: The extraction/purification process leads to a radionucleidic purity higher than 99.99% ( 43Sc, 46Sc, 48 Sc b DL). 44mSc/44Sc labeling towards DOTA moiety was performed in order to get an evaluation of the specific activities that could be reached with regard to all metallic impurities from the resulting source. Reaction parameters of radiolabeling were optimized, reaching yields over 95%, and leading to a specific activity of about 10–20 MBq/nmol for DOTA. A recycling process for the enriched 44Ca target was developed and optimized. Conclusion: The quality of the final batch with regard to radionucleidic purity, specific activity and metal impurities allowed a right away use for further radiopharmaceutical evaluation. This radionucleidic pair of 44mSc/44Sc offers a quite interesting PET radionuclide for being further evaluated as an in-vivo generator. © 2015 Published by Elsevier Inc.
1. Introduction Current radiotracers for cancer diseases have limitations: imaging must be performed on the same day as tracer production; their halflives (2 h or less) may not be compatible for dosimetry assessment. Most probes may be more efficient if combined with new radionuclides. Recently, the scandium chemistry has revealed a growing interest with an increasing number of papers available on scandium: from 44Ti/ 44Sc generator [1–4], from neutron irradiated Ti [5,6], cyclotron produced 44m Sc/ 44Sc [7–9], natSc [10], 46Sc [11,12] or 47Sc [13–15]. Among them, the most promising is 44Sc with a half-life (T1/2) of 3.97 h and mean positron energy of 0.6 MeV ideal for PET cameras. 44Sc offers a better T1/2 in comparison to 68Ga (T1/2 = 68 min). Many different ways have been investigated to produce 44Sc and 44m Sc. They can be formed using radiation mechanism on 45Sc, by spallation on natFe or natCu. Scandium-44 can also be obtained through the decay of titanium-44 using the so-called 44Ti/44Sc generator, but in ⁎ Corresponding author. E-mail address: [email protected] (C. Alliot).
this case no 44mSc can be obtained. To produce simultaneously 44Sc and 44mSc, medical cyclotrons that currently supply 18F to hospitals can be used via 44Ca(p,n)44,44mSc route. 44 Sc is developed for a new nuclear medical imaging technique using the ability of the 44Sc to emit in 99.9% of the cases a photon with energy of 1.15 MeV and a characteristic emission time of few picoseconds. When calculating radiation doses, the specific emission characteristics of 44Sc make it a unique candidate promising for 3γ-coincidence imaging with potentially improved local resolution, from which the very first results obtained on a small-dimension prototype let foresee a very promising perspective [16]. The long-lived 44mSc isomer (T1/2 = 58.6 h) is of interest, as its longer half-life will allow imaging at later times for more accurate assessment of distribution and absorbed doses. 44mSc decays by internal transition (98.8%) to the ground state ( 44gSc) with an associated small secondary emission (Auger emission 2.74%). The main photon emission (271 keV) will induce some recoil of the nuclei. Due to the large mass difference, this recoil energy is estimated to be only 0.89 eV. Since 44m Sc and 44Sc have the same chemical nature, no change in oxidation state is expected, in contrast to what has been already observed for
http://dx.doi.org/10.1016/j.nucmedbio.2015.03.002 0969-8051/© 2015 Published by Elsevier Inc.
Please cite this article as: Alliot C., et al, Cyclotron production of high purity 44m,44Sc with deuterons from 44CaCO3 targets, Nucl Med Biol (2015), http://dx.doi.org/10.1016/j.nucmedbio.2015.03.002
2
C. Alliot et al. / Nuclear Medicine and Biology xxx (2015) xxx–xxx
140
Nd/140Pr [17,18] and 103Pd/103mRh [19], leading to the release of the isotope from the molecule or radiolysis effects coming by the secondary emission. Consequently 44m Sc/ 44 Sc can be used as an in-vivo generator [9]. In addition, 47Sc (T1/2 = 3.35 d) with its moderate energy β− emissions (162 keV) may be useful for targeted radiotherapy, and could also be used for single photon emission computed tomography (SPECT) imaging. Its photon emission of 159 keV (68%) is ideal for imaging on SPECT cameras. 47Sc has already been investigated for its potential interest for β-RIT [20,21]. Regard to 89Zr (T1/2 = 3.3 d), a radionuclide with longer half-life, is highly preferable for its potential use in immuno-PET. The main drawback is that Zr coordination chemistry requires specific ligands, different from those used with therapeutic radioisotopes. 44/44mSc radionuclides of high specific activity are very attractive for the development of various PET imaging probes to improve the diagnosis of various solid cancers, while enabling further consideration of 47Sc to be used together with 44Sc as a matched pair, for the development of “theranostic” probes (a single molecule labeled with two different radionuclides of the same chemical element) [22,23]. Scandium, with + III as the most common oxidation state, can be easily complexed by the most frequently utilized polyaminocarboxylate chelators. The chemical differences between Sc and Ti or Ca are a major advantage for the separation of no-carrier added 44,44m,47Sc and is expected to be much simpler than the separation of neighboring lanthanides, where the chemical differences are very small. Many authors have previously developed processes to purify scandium from calcium [24–29]. Most of them used solvent extraction. Moreover Kalyanaraman and Khopkar [29] evidenced that scandium can be selectively extracted with mesityl oxide as its thiocyanate complex and then recovers in HCl solution. Nonetheless, calcium seemed to be a significant interfering for quantities higher than 25 mg. These authors evidenced that this quantity is sufficient to cause 2% error in the recovery of scandium. For higher quantities of calcium, the loss of scandium can be important. Vibhute and Khopkar [25] and Karve and Khopkar [28] purified scandium as citrate and ascorbato complex, respectively, using aliquat 336S as extractant. In this case, scandium was obtained in the final solution with some organic acid that could become competing agent as far as radiolabelling was concerned. Radhakrishnan and Owens [27] proposed another route for scandium purification, by liquid–liquid extraction, using tri-n-butyl phosphate as solvent. But in this study the authors evidenced that the selectivity coefficient of scandium as regard to calcium increased when nitric acid concentration increased, while the distribution coefficient of scandium decreased. This meant that, despite the huge amount of calcium used, quantitative recovery of scandium was not easy. Moreover no data were available concerning the other metals which could be present at the end of irradiation. Such viable separation uses precipitation and filtration [30]. This approach takes advantage of the insolubility of Sc(OH)3 either as a precipitate or coprecipitate, and is similar to a technique developed for purifying yttrium from strontium [31,32]. Due to these difficulties, a chromatographic process for purification was thus studied. Rane and Bhatki [26] reported a method using cation exchange resin (DOWEX 50x8) to prepare 45Ca from natural scandium with high resulting specific activity. But this type of resin is not really specific with regard to transition metal. Thus, purification of scandium from transition metal using DOWEX 50 × 8, seems difficult. As the mostly used complexing agent for metals, i.e. DOTA, is not a specific ligand, the presence of others metals requires higher quantities of DOTA for radiolabelling. In order to minimize the metallic impurities, DGA® resin, specific to lanthanides and already used to purify scandium [15], was employed Moreover, for further labeling, recovery of the 44mSc/44Sc from calcium needs to remove bulk alkaline earth and to give the final product in a small volume. Small amounts of residual calcium do not interfere with DOTA chelation [33], but it could be important to minimize this quantity in case of in-vivo injections.
The developed chemical process aimed at recovering scandium while minimizing cationic impurities (i.e. Ca, metals), leading to high specific activities. For this, DOTA radiolabelling was performed in order to evaluate the specific activity with regard to all metallic impurities that could be present in the final source. The extraction and purification process was developed and optimized, together with a recycling process of the initial enriched material, for cost effectiveness purpose. 2. Materials and methods 2.1. Chemicals and reagents Nitric and hydrochloric acid were received as ultrapure solutions (SCP Science). All dilutions were made in Ultrapure water (Millipore, 18.2 MΩ.cm). The commercially available solid phase extraction resin DGA (N,N,N,N-tetra-n-octyldiglycolamide) provided by Triskem® was first eluted with NaOH and rinsed with pure water. Then, it was eluted with 20 mL of HCl 0.1 mol.L−1 to remove all potential metal impurities. Single and multi-elements standards (about 10 ppm SCP Science) were used to determine the partition coefficients. Commercially available 1,4,7,10-tetraazacyclododecane-1,4,7,10-tetraacetic acid (DOTA, Macrocyclics Inc.) was used as received. 2.2. Cyclotron targetry and irradiations To limit contaminants production ( 46Sc, 47Sc, 48Sc) due to the use of natural calcium [7], the irradiation targets are made of 96.9% enriched 44 CaCO3 purchase from Chemgas (density equal to 2.71 g/cm3). In the first time, each target was made of 100 mg of 44CaCO3 pressed under 5 tons to form a 10 mm diameter pellet. The obtained thickness was about 470 μm. To prevent the target contamination during irradiation, the target was placed between two kapton foils. These targets were used to develop the purification process of 44Sc. In the second time, the developed process was checked with larger targets which would be used for preclinical studies. These targets were made of 500 mg of 44 CaCO3 (20 mm diameter, 0.6 mm thickness). On ARRONAX facility, 44Sc was produced using an irradiation station named Nice3 devoted to the low activity productions (few mCi maximum). With this device, our target was placed in air, 6.6 cm away from a 75 μm Kapton foil ensuring the sealing at the end of beamline (Fig. 1). The targets were cooled by air during the irradiation. 2.3. Partition coefficients determinations The distribution coefficient (KD) quantifies the partition of an element between the aqueous solution and the solid phase and is defined as follows: KD ¼
C solid ðmol=g Þ C solution ðmol=LÞ
ð1Þ
where Csolid is the concentration of elements sorbed onto the resin (in g/g of dry resin), and Csolution the total aqueous solution concentration of elements (in mol/L) which remains in solution after equilibration. As the distribution coefficients were calculated as a function of dried resin mass, the percentage of humidity was determined by placing 5 samples of each pre-conditioned resin in oven at 105°C. Moreover as single and multi-elements standards are commercially available in dilute nitric acid. To avoid potential effects of nitrate ions on partition behavior, aliquots of standard solutions were transferred to a pre-cleaned with ultrapure HNO3 2% Teflon beaker. The solutions were evaporated to dryness. After complete evaporation, the solid residues were dissolved in HNO3 or HCl, and the process was repeated three times to ensure complete removal of initial anions. Then pre-conditioned DGA® resin was weighted and dispersed in the studied aqueous solution. The suspensions were stirred for two hours that has been shown
Please cite this article as: Alliot C., et al, Cyclotron production of high purity 44m,44Sc with deuterons from 44CaCO3 targets, Nucl Med Biol (2015), http://dx.doi.org/10.1016/j.nucmedbio.2015.03.002
C. Alliot et al. / Nuclear Medicine and Biology xxx (2015) xxx–xxx
3
Fig. 1. Schematic view of the irradiation station and the associated deuteron energy.
to be sufficient to reach equilibrium. The aqueous phases were collected by filtration and analyzed. A blank solution (without resin) was prepared with each batch of equilibration experiment and analyzed simultaneously with the samples, to determine the initial concentrations of each element. The acidic solutions were collected and were diluted in HNO3 1% prior to analysis. Measurements were performed on a Thermofischer ICAP 6500 DUO ICP-AES. Representative wavelengths for each element were selected, with preference given to higher relative emissions and absence of interferences. Knowing the initial (Cinit) and equilibrium (Ceq) aqueous concentrations of each element, Eq. (1) can be rearranged as followed:
KD ¼
C init −C eq =m C eq =V
ð2Þ
where m (in g) is the weight of dried resin and V (in mL) is the total volume of aqueous solution. Equilibration experiments were replicated using freshly prepared multi-element standard solutions. The final distribution coefficients were calculated based on the arithmetic averages of replicate analysis. The resin concentration (in g/mL) was chosen for each batch to minimize the global uncertainty of partition coefficient. The reproducibility of logarithm of KD was better than 5% for all elements.
to dryness and recovered in concentrated ultrapure hydrochloric acid (9 mol.L −1). The solution was then loaded on a pre-conditioned AG1 × 8 column to retain all metallic impurities (Cu, Co, Fe…). The column was rinsed with 10 mL of HCl 9 mol.L −1 to recover enriched 44Ca. The eluted solution was evaporated to dryness and recovered in a mixture of bicarbonate buffer 0.1 mol.L −1 (pH = 10.33)/methanol. The solution was then filtered through 0.22 μm filters (Cellulose Acetate, Millipore) to discard supernatant, and the residue was dried in an oven at 105°C to remove water and methanol. The obtained solid of calcium carbonate was suitable for manufacturing targets for next irradiations. 2.6. Labeling DOTA with 44m,44Sc To 450 μL of solution of DOTA (i.e. 10 nmol, Macrocyclics Inc.) were added 50 μL (i.e. 2 nmol) of 44Sc and mixed in a 2 mL screw-Cap Wheaton V-bottom vial. The solution was placed in a boiling water bath at 90°C for 30 min and then cooled till room temperature was reached. To test the radiolabelling yield, a radio-TLC was performed by spotting 2 μL onto a TLC Flex Plate (silica gel 60A, F-254, 200 μm, Selecto Scientific) and eluted with a developing solution of 0.04 mol.L − 1 aqueous NH4OAc/Methanol, 50/50 (v/v). The activity distribution on the plates was assessed with a Packard Cyclone Phosphor-Plate imaging system (Perkin Elmer). 3. Results and discussion
2.4. Radiochemical separations 3.1. Cyclotron targetry and irradiations Firstly, the purification process was developed using an artificial target made from natural calcium (100 mg of natural calcium carbonate). A multi-elements standard solution was prepared from single and multielements standard solutions and then added to the artificial target after dissolution. The obtained solution was loaded onto the DGA® resin (200 mg of pre-conditioned dried resin). The recovery of all elements (Al, Ca, K, Fe, Co, Cr, Cu, Mn, Ni, Sc, Zn) as a function of acid nature, concentration and elution volume was determined to define the elution profiles of all elements. The aim was to recover scandium selectively and quantitatively. Then, the purification process was applied to irradiated 44CaCO3 targets. The targets were removed from the target holder and dropped into 4 mol.L −1 of HCl solution. The aqueous solution was loaded onto the DGA® column. At the end of the purification process, the scandium solution was evaporated to dryness and recovered in a small volume of HCl 0.1 mol.L −1. The final solutions were measured by a gamma-detector (ORTEC) with a high resolution HPGe detector, previously calibrated with multi-gamma source from CERCA LEA, to determine the radionucleidic purity and by ICP-AES to determine the concentration of potential major impurities. 2.5. Enriched calcium recycling As DGA® resin is specific to lanthanides, a first solution with metals and calcium was recovered from purification process. It was evaporated
A typical irradiation realized on Nice3 device corresponds to a period of 1 h and intensity of 200 nA. After irradiation, the calcium pellet was recovered and directly dissolved in HCl 4 mol.L −1. An aliquot of this solution was counted by gamma spectrometry to determine the activities of all γ emitters. Under these conditions, the determined activity (44 ± 3 MBq) was compared with the theoretical production (48 MBq) calculated from TALYS code. Even if the results obtained with TALYS code were not better than those obtained with others codes (MCNPX, PHITS), the shape fits rather well the shape of experimental values [34]. This production yield was in the same order of magnitude than the one obtained with protons (57 MBq) for the optimal energy (12 MeV). Moreover, concerning the 44mSc production, the experimental 44mSc/44Sc activity ratio was 2.21 ± 0.13% which corresponded to a cross section ratio equal to 33%, as regard to half-life ratio. This ratio, in good agreement with TALYS calculation (about 2.34%), was significantly higher than the one obtained with 12 MeV protons beam. The cross section ratio for 12 MeV protons was equal to 7.77% (Fig. 2). Consequently, the deuterons route seems to be a promising way to produce non-carrier added 44Sc, while optimizing the 44mSc/44Sc ratio. Other production routes allow a better 44mSc/44Sc ratio by bombarding a 45Sc target [35–37], but the final product is carrier-added, which is a major drawback if one wants to use it for further radiopharmaceutical use. Concerning the by-products, produced by deuterons route at this energy, no 43Sc was detected which is normal since the energy threshold is 17 MeV, and about 0.7 MBq 42K and 0.6 kBq of 43K were determined.
Please cite this article as: Alliot C., et al, Cyclotron production of high purity 44m,44Sc with deuterons from 44CaCO3 targets, Nucl Med Biol (2015), http://dx.doi.org/10.1016/j.nucmedbio.2015.03.002
4
C. Alliot et al. / Nuclear Medicine and Biology xxx (2015) xxx–xxx
Fig. 2. 44mSc/44Sc ratio as a function of the projectile energy.
The potassium impurities can be easily discarded by the purification process. High specific activity of the final batch could be thus expected. 3.2. Partition coefficients Fig. 3 shows the partition coefficient in logarithmic scale on DGA® resin as a function of HCl. A high KD value (higher than 10) at a specific acid concentration means the ion is preferentially retained on the resin, while a low KD indicates the release of the ions to the mobile phase. Although partial separation could be potentially achieved at KD N 1,
large discrepancies in distribution coefficients were required for quantitative separation of analytes from matrix elements in natural samples. Previously, Horwitz et al. [38] and Pourmand and Dauphas [39] studied some of the properties of DGA® extraction chromatographic resins in slurry-packed gravity columns to explore the potential application of these resins. Even if almost all elements were studied by these authors, the chromatographic behavior of scandium on DGA® was not studied [38,39]. To develop a purification process for 44Sc produced from enriched calcium targets, this element plays a key role. KD values for scandium were determined, and partition coefficients for different potential impurities coming from target holder (Figs. 3 and 4) were monitored simultaneously. The partitioning behavior of most studied elements on resin is somehow comparable between our study and published works [38,39]. However, there are some differences. Pourmand et al. and Horwitz et al. reported that Ca(II) was slightly adsorbed for HCl concentration higher than 2 mol.L − 1 (log KD = 1). This was not observed in this work, as Ca(II) remained in mobile phase up to 3 mol.L −1. Concerning Ni(II), Pourmand et al. [39] reported that the cation remained in aqueous solutions whichever HCl concentration was concerned, while it had a similar behavior than Co(II) and Cu(II) in the present study. These discrepancies in element behaviors could not be explained by the different conditions used by all the authors, since all standards were converted to HCl, and all partition coefficients were determined for dried resin). The adsorption of potassium was equally studied, and no significant adsorption was observed in HCl media. The partitioning behaviors of zinc, iron and scandium are relatively similar. In addition, HCl is not the appropriate medium to separate these elements. Aiming to discard these elements, their distribution coefficients were determined in HNO3 media (Fig. 4). Sc(III) was strongly adsorbed whatever HNO3 concentration, while Zn(II) and Fe(III) remained in the aqueous phase. The behavior of zinc was comparable to the one determined by Pourmand et al. [39]. For Fe(III), Horwitz et al. [38] and Pourmand et al. [39] evidenced strong adsorption whatever medium and acid concentration. For these authors, Fe(III) could not be discarded from Sc(III) in HCl or HNO3 media. The difference between their results and ours is a critical point for scandium purification. Indeed, iron impurities can compete with scandium for radiolabelling. 3.3. Radiochemical separations In the following section, the DGA® distribution coefficients were thus used for developing a simple chromatographic method of extraction for the separation of Sc from Ca and all associated metallic impurities contained in 44,44mSc produced source. The recovery of elements, which are not adsorbed and totally recovered at the end of the rinsing step (mainly K, Cu, Co, and Ni), is not shown. The final extraction scheme for Ca, Fe, Zn and Sc was optimized based on the partition
Fig. 3. Distribution coefficients of elements on DGA resin as a function of HCl concentration.
Fig. 4. Distribution coefficients of elements on DGA resin as a function of HNO3 concentration.
Please cite this article as: Alliot C., et al, Cyclotron production of high purity 44m,44Sc with deuterons from 44CaCO3 targets, Nucl Med Biol (2015), http://dx.doi.org/10.1016/j.nucmedbio.2015.03.002
C. Alliot et al. / Nuclear Medicine and Biology xxx (2015) xxx–xxx
coefficient of elements from Figs. 3 to 4 and replicates of elution experiments. The acid concentration of the loading solution on DGA® was fixed at 4 mol.L −1 HCl. Matrix elements, including alkali and transition metals such as Co, Cu, Ni were removed from the resin during the load and subsequent rinse in 4 mol.L −1 HCl. Although more than 90% of calcium could be successfully eluted in 20 mL of 4 mol.L−1 HCl (Fig. 5), significant tailing was observed. This loss of enriched material is not acceptable for cost effectiveness. Nonetheless, as it is necessary to discard other metallic elements before recovering scandium, the remaining 10% will be recovered simultaneously with these elements to optimize purification time. Following this first step of elution, 12 mL of 1 mol.L −1 HNO3 were necessary to elute iron and zinc (Fig. 5). This second step allowed discarding all potential impurities as a contrary of Krajewski et al. [40]. 10 mL of 0.1 mol.L −1 HCl was then necessary to elute quantitatively scandium from the DGA® resin. The eluted solution is then evaporated to dryness and recovered in a small volume of 0.1 mol.L−1 HCl. 44 Sc was separated from the irradiated calcium target using the procedure reported previously. The efficiency of our separation method is 88 ± 3%, which is consistent with the previously reported procedures using DGA® resin [41,42]. Radionucleidic analysis through gamma spectrometry showed that the final solutions contained only 44,44mSc. No other significant radionuclides were evidenced even after a cooling time corresponding to several half-lives of scandium. ICP-AES analyses of the eluates gave a value under the detection limit for stable scandium resulting in a specific activity higher than 50 MBq/μmol. This value was significantly lower than the one obtained by Hoehr et al. [41] (1400 MBq/μmol), but higher values are expected by increasing produced activities. Moreover the calcium content in the purified samples was under detection limit even if this last one was not the most significant competitive element for complexation by DOTA. The other major metallic contaminants were Fe and Al. The total concentration of both elements was 1.14 ± 0.66 ppm, which was lower than the one determined by Severin and Hoehr [7,41]. As DOTA ligands used for radiolabelling is not really specific complexing agents, the total concentration of metallic impurities is a key factor for radiolabelling yield and the total concentration of DOTA. Metal contamination will not increase significantly when increasing produced activity. It was observed experimentally that increasing the activity from 100 MBq/mL to 375 MBq/mL kept the concentration of metallic impurities of approx. 1 ppm for the same final volume of the solution. By contrast the specific activity will increase almost linearly with the produced activity. This assumption was confirmed with larger targets developed for preclinical studies. By increasing the targets size (five times more enriched calcium), the final 44Sc quality was improved with produced activity. Based on the results obtained in this work, optimized irradiation parameters to be used at Arronax for up scaling production are defined as followed: incident deuteron energy of 20 MeV on the pellet; output
5
energy of 10 MeV, a production yield at EOB of 469 MBq/μAh is determined. This configuration corresponds to a target thickness of 800 μm. In this energy range, the activity ratio of 44mSc/ 44Sc is expected to be 0.0248. For a typical irradiation of 4 hours and 50 μA/h, we can expect at the end of irradiation about 2500 mCi and 60 mCi of 44Sc and 44m Sc respectively. 3.4. Enriched calcium recycling Due to the high cost of enriched material, a recycling process of the target was developed. The 4 mol.L−1 HCl and 1 mol.L−1 solutions obtained from the extraction process were mixed and evaporated to dryness. With the mixture bicarbonate/methanol, the kinetic of solvent evaporation was increased, and the solubility of calcium carbonate was lowered [43]. The recovery yield of enriched calcium was 90 ± 2%. The suspension was kept for further process recycling to minimize calcium losses. The recycling targets were irradiated, and no significant difference of production yield was observed. 3.5. Radiolabelling As DOTA is not a specific complexing agent, all cationic metals (for example Fe, Zn) can be labeled by this chelating molecule. Radiolabeling yields from our resulting batch of 44Sc were studied for metal-to-ligand ratio (and not for scandium-to-ligand ratio) ranging from 1/1 to 1/50 for DOTA. The most suitable conditions have been identified as a metal-toligand ratio of 1:3 in acetate buffer at pH = 4, and heating the mixture at 70°C for 20 minutes provided the best labeling yields N95%. These values were in agreement with previous studies performed on 46 Sc-DOTA [12,41], 44Sc-DOTATOC and 44Sc-DOTATATE with eluted from 44Ti generator [1,44]. All studies have evidenced that the specific activity was not the major parameter for radiolabelling by DOTA, but was of major importance for further use with biological vectors. Consequently, the total metal concentration has to be as low as possible to minimize the quantity of DOTA and later on of biological probes. In conditions defined previously for large scale production, the reachable specific activity of 44Sc will be higher than 1 GBq/μmol of DOTA four hours after the end of irradiation (time necessary for chemical purification and radiolabelling). 3.6. Conclusion No carrier added 44Sc production was developed in order to make available 44,44mSc batches with high specific activities. The obtained production yields were about 220 MBq/μA.h for small targets which was in agreement with theoretical calculations. For optimizing the simultaneous production of 44mSc, low energy nuclear reactions with deuterons at 16MeV were chosen, leading to higher cross section and lower amount of co-produced radioisotopes in comparison to proton. A method using DGA® resin has been proposed to separate scandium from carbonate calcium targets. The process allows to remove metallic impurities like Fe, Al, Zn leading to a high specific activity suitable for efficient radiolabelling. This production/purification process with a global recovery yield of about 90% could be easily transferred to a medium-sized cyclotron using deuterons or a biomedical cyclotron using protons. Acknowledgments
Fig. 5. Elution curves for artificial target of 44Ca irradiated with deuteron on a DGA® resin (50–100μm particle size).
The ARRONAX cyclotron is a project promoted by the Regional Council of Pays de la Loire financed by local authorities, the French government and the European Union. This work has been, in part, supported by a grant from the French National Agency for Research called "Investissements d'Avenir", Equipex Arronax-Plus no ANR-11EQPX-0004 and Labex no ANR-11-LABX-0018-01.
Please cite this article as: Alliot C., et al, Cyclotron production of high purity 44m,44Sc with deuterons from 44CaCO3 targets, Nucl Med Biol (2015), http://dx.doi.org/10.1016/j.nucmedbio.2015.03.002
6
C. Alliot et al. / Nuclear Medicine and Biology xxx (2015) xxx–xxx
References [1] Pruszyski M, Majkowska-Pilip A, Loktionova N, Eppard E, Roesch F. Radiolabeling of DOTATOC with the long-lived positron emitter 44Sc. Appl Radiat Isot 2012;70:974–9. [2] Pruszyski M, Loktionova N, Filosofov D, Roesch F. Post-elution processing of 44Ti/44Sc generator derived 44Sc for clinical application. Appl Radiat Isot 2010;68:1636–41. [3] Mausner L, Kolsky K, Joshi V, Srivastava S. Radionuclide development at BNL for nuclear medicine therapy. Appl Radiat Isot 1998;49:285–94. [4] Filosofov D, Loktionova N, Roesch F. A 44Ti/44Sc radionuclide generator for potential application of 44Sc-based PET radiopharmaceuticals. Radiochim Acta 2010;98:149–56. [5] Pietrelli L, Mausner L, Kolsky K. Separation of carrier free 47Sc from titanium targets. J Radioanal Nucl Chem 1992;157:335–45. [6] Bokhari TH, Mushtaq A, Khan IU. Separation of no-carrier-added radioactive scandium from neutron irradiated titanium. J Radioanal Nucl Chem 2010;283:389–93. [7] Severin G, Engle J, Valdovinos H, Barnhart T, Nickles R. Cyclotron produced 44gSc from natural calcium. Appl Radiat Isot 2012;70:1526–30. [8] Müller C, Bunka M, Reber J, Fischer C, Zhernosekov K, Türler A, et al. Promises of cyclotron-produced 44Sc as a diagnostic match for trivalent β–emitters: in-vitro and in-vivo study of a 44Sc-DOTA-folate conjugate. J Nucl Med 2013;54:2168–74. [9] Huclier-Markai S, Kerdjoudj R, Alliot C, Bonraisin A, Michel N, Haddad F, et al. Optimization of reaction conditions for the radiolabeling of DOTA and DOTA-peptide with 44m/44Sc and experimental evidence of the feasibility of an in-vivo PET generator. Nucl Med Biol 2014;41:e36–43. [10] Koumarianou E, Pawlak D, Korsak A, Mikolajczak R. Comparison of receptor affinity of natSc-DOTATATE versus natGa-DOTATATE. Nucl Med Rev 2011;14:85–9. [11] Majkowska-Pilip A, Bilewicz A. Macrocyclic complexes of scandium radionuclides as precursors for diagnostic and therapeutic radiopharmaceuticals. J Inorg Biochem 2011;105:313–20. [12] Huclier-Markai S, Sabatie A, Ribet S, Kubicek V, Paris M, Vidaud C, et al. Chemical and biological evaluation of scandium(III)-polyaminocarboxylate complexes as potential PET agents and radiopharmaceuticals. Radiochim Acta 2011;99:653–62. [13] Polosak M, Kasperek A, Krajewski S, Bilewicz A. Stability of 47Sc-complexes with acyclic polyamino-polycarboxylate ligands. J Radioanal Nucl Chem 2013;295:1867–72. [14] Müller C, Bunka M, Haller S, Köster U, Groehn V, Bernhardt P, et al. Promising prospects for 44Sc/47Sc based theragnostics: application of 47Sc for radionuclide tumor therapy in mice. J Nucl Med 2014;55:1658–64. [15] Cydzik I, Krajewski S, Abbas K, Simonell F, Bulgheroni A, Kasperek A, et al. Labelling of DOTATATE with cyclotron produced 44Sc and 43Sc. Q J Nucl Med Mol Imaging 2012;56:33–4. [16] Grignon C, Barbet J, Bardies M, Carlier T, Chatal J, Couturier O. Nuclear medical imaging using β+/γ coincidences from 44Sc radionuclide with liquid xenon as detection medium. Nucl Instrum Meth A 2007;571:142–53. [17] Zhernosekov K, Filosofov D, Qaim SM, Rösch F. A 140Nd/140Pr radionuclide generator based on physico-chemical transitions in 140Pr complexes after electron capture decay of 140Nd-DOTA. Radiochim Acta 2007;95:319–27. [18] Rösch F, Brockmann J, Lebedev NA, Qaim SM. Production and radiochemical separation of the Auger electron emitter 140Nd. Acta Oncol 2000;39:727–30. [19] Van Rooyen J, Szucs Z, Zeevaar J. A possible in-vivo generator 103Pd/103mRh recoil considerations. Appl Radiat Isot 2008;66:1346–9. [20] Kolsky K, Joshi V, Mausner L, Srivastava S. Radiochemical purification of no-carrieradded scandium-47 for radioimmunotherapy. Appl Radiat Isot 1998;49:1541–9. [21] Mausner LF, Joshi L, Kolsky K, Meinken G, Mease R, Sweet M, et al. Evaluation of chelating agents for radioimmunotherapy with scandium-47. J Nucl Med 1995; 36(Suppl. 5):104 in Proceedings of the meeting of the society of Nuclear Medicine. Minneapolis, United States.
[22] Herzog H, Rösch F, Stöcklin G, Lueders C, Qaim SM, Feinendegen LE. Measurement of pharmacokinetics of yttrium-86 radiopharmaceuticals with PET and radiation dose calculation of analogous yttrium-90 radiotherapeutics. J Nucl Med 1993;34:2222–6. [23] Baum R, Kulkarni H. Theranostics: from molecular imaging using Ga-68 labeled tracers and PET/CT to personalized radionuclide therapy - the bad berka experience. Theranostics 2012;2:437–47. [24] Zhang P, You S, Zhang L, Feng S, Hou S. A solvent extraction process for the preparation of ultrahigh purity scandium oxide. Hydrometallurgy 1997;47:47–56. [25] Vibhute C, Khopkar S. Separation of scandium(III) as citrate complex by extraction with aliquat 336S. Indian J Chem 1985;24A:444–6. [26] Rane A, Bhatki K. Rapid radiochemical separations of strontium-90/yttrium-90 and calcium-45/scandium-46 on a cation exchange resin. Anal Chem 1966;38:1598–601. [27] Radhakrishnan K, Owens T. Separation of scandium and calcium by liquid-liquid extraction using tri-n-butyl phosphate as solvent. J Chem Eng Data 1972;17:478–82. [28] Karve M, Khopkar S. Separation of scandium(III) as ascorbato complex by extraction with aliquat 336S. Bull Chem Soc Jpn 1991;64:655–8. [29] Kalyanaraman S, Khopkar S. Solvent extraction separation of scandium with 4-methyl-3-pentene-2-one as its thiocyanate complex. Anal Chem 1977;49: 1192–994. [30] Duval J, Kurbatov M. Separation of carrier-free scandium from a calcium target. J Am Chem Soc 1953;75:2246–8. [31] Rösch F, Qaim SM, Stöcklin G. Production of the positron emitting radioisotope 86Y for nuclear medical application. Appl Radiat Isot 1993;44:677–81. [32] Avila-Rodriguez M, Nye J, Nickles R. Production and separation of non-carrier-added 86 Y from enriched 86Sr targets. Appl Radiat Isot 2008;66:9–13. [33] Abbas K, Cydzik I, Simonell F, Krajewski S, Kasperek A, Bilewicz A. Cyclotron production of 44Sc - new radionuclide for PET technique. J Labelled Compd Rad 2011;54(Suppl. S1):S53. [34] Blideanu V, Garcia M, Joyer P, Lopez D, Mayoral A, Ogando F, et al. Deuteron cross section evaluation for safety and radioprotection calculations of IFMIF/EVEDA accelerator prototype. J Nucl Mat 2011;417:1271–4. [35] Luo J, Liu R, Jiang L, Liu Z, Sun G, Ge S. Cross-sections of 45Sc(n, 2n)44m, gSc reaction from the reaction threshold to 20 MeV. Radiochim Acta 2013;101:607–12. [36] Bostan M, Qaim SM. Excitation functions of threshold reactions on 45Sc and 55Mn induced by 6 to 13 MeV neutrons. Phys Rev C 1994;49:266–71. [37] Bailey S. Yield ratios for the isomeric pair Sc44m,44 formed in (α, αn) and (α, n) reactions. Phys Rev C 1961;123:579–82. [38] Horwitz E, McAlister D, Bond A, Barrans R. Novel extraction of chromatographic resins based on tetraalkyldiglycolamides: characterization and potential applications. Solvent Extr Ion Exch 2005;23:319–44. [39] Pourmand A, Dauphas N. Distribution coefficients of 60 elements on TODGA resin: application to Ca, Lu, Hf, U and Th isotope geochemistry. Talenta 2010; 81:741–53. [40] Krajewski S, Bilewicz A, Cydzik I, Abbas K, Bulgheroni A, Simonelli F, et al. Cyclotron production of 44Sc for clinical application. Radiochim Acta 2013;101:333–8. [41] Hoehr C, Oehlke E, Benard F, Jaeil Lee C, Hou X, Badesso B, et al. 44gSc production using a water target on a 13 MeV cyclotron. Nucl Med Biol 2014;41:401–6. [42] Bunka M, Zhernosekov K, Eichholzer Y, Schibli R, Turler A. Cyclotron production of scandium-44. Labor für Radio und Umweltchemie der Universität Bern und des Paul Scherrer Institut; 2011. [43] Kan A, Fu G, Tomson M. Effect of methanol on carbonate equilibrium and calcite solubility in a gas/methanol/water/salt mixed system. Langmuir 2002;18: 9713–25. [44] Loktionova N, Filosofov D, Pruszynski M, Roesch F. Design and performance of a novel 5 mCi 44Ti/44Sc radionuclide generator. J Nucl Med 2010;51(Suppl. 2):590.
Please cite this article as: Alliot C., et al, Cyclotron production of high purity 44m,44Sc with deuterons from 44CaCO3 targets, Nucl Med Biol (2015), http://dx.doi.org/10.1016/j.nucmedbio.2015.03.002
