Cyclotron production of 43Sc and 44gSc from enriched 42CaO, 43CaO, and 44CaO targets

1 Introduction

An increasingly popular approach to cancer therapy is combining chemically similar pairs of radionuclides as radiolabels of diagnostic and therapeutic drugs (“theranostics”). Positron emission tomography (PET) of ⁶⁸Ga- (t_1/2 = 67.71 min; β⁺_Eavg = 829.5 keV; β⁺_intensity = 88.91%) labeled somatostatin analogs and prostate-specific membrane antigen targeting agents has been used to monitor response to ¹⁷⁷Lu- (t_1/2 = 6.644 days; β⁻_EAvg = 133.6 keV; β^-_intensity = 100.0%) based treatment (Weineisen, et al., 2015; Grubmüller, et al., 2018; Heinzel, et al., 2019), to select potential patients for therapy (Gains, et al., 2011), and to detect lesions not found on ¹⁷⁷Lu single photon emission tomography (SPECT) (Sainz-Esteban, et al., 2012). However, the high average positron energy and short half-life of ⁶⁸Ga limit its clinical application. Because ⁶⁸Ga’s half-life is too short for off-site transport, clinics require an on-site cyclotron or a⁶⁸Ge/⁶⁸Ga generator to provide patient doses.

Radioisotopes of scandium are promising alternatives to ⁶⁸Ga. There are three clinically interesting radioisotopes of scandium: ⁴³Sc (t_1/2 = 3.891 h; β⁺_Eavg = 476 keV; β⁺_intensity = 88.1%), ^44gSc (t_1/2 = 3.97 h; β⁺_Eavg = 632 keV; β⁺_intensity = 94.27%), and ⁴⁷Sc (t_1/2 = 3.349 days; β⁻_Eavg = 162 keV; β^-_intensity = 100.0%) (National Nuclear Data Center, Brookhaven National, 2022). Both ⁴³Sc and ^44gSc emit positrons with lower mean energies than ⁶⁸Ga, have high positron branching ratios, and are chemically identical to the therapeutic radionuclide ⁴⁷Sc. Additionally, the ionic radius and coordination behavior of scandium are better matched to lutetium than gallium (Robert, 1976); its positron-emitting radioisotopes are viable diagnostic analogues of ¹⁷⁷Lu in radiopharmaceuticals. Proton bombardment of natural metallic calcium targets produces ^44gSc (Severin, et al., 2012), but long-lived (t_1/2 = 44 h–84 days) radionuclidic impurities (⁴⁶Sc, ⁴⁷Sc, and ⁴⁸Sc) are co-produced due to the many stable isotopes of calcium (⁴⁰Ca, 96.941%; ⁴²Ca, 0.647%; ⁴³Ca, 0.135%; ⁴⁴Ca, 2.086%; ⁴⁶Ca, 0.004%; ⁴⁸Ca, and 0.187%). Producing ⁴³Sc and ^44gSc from isotopically enriched calcium minimizes the presence of long-lived radioisotopes of scandium. This work investigates the following reactions on isotopically enriched calcium targets: ⁴²Ca(d,n)⁴³Sc, ⁴³Ca(p,n)⁴³Sc, ⁴³Ca(d,n)^44gSc, ⁴⁴Ca(p,n)^44gSc, ⁴⁴Ca(p,2n)⁴³Sc, and ⁴⁴Ca(d,2n)^44gSc.

There are several methods for separating ^43/44gSc from calcium targets. Following target dissolution, extraction chromatography resins such as UTEVA^® (Valdovinos, et al., 2015), N,N,N′,N′-tetra-n-octyldiglycolamide (DGA) (Alliot, et al., 2015), and hydroxamate (Severin, et al., 2012) isolate ^43/44gSc from the calcium target material. Additionally, adding ion exchange resins such as Dowex^® 50WX8 in combination with extraction chromatography resin such as DGA further removes trace metal impurities (Rane et al., 1966; Müller, et al., 2013; van der Meulen, et al., 2015).

Irradiating titanium targets with protons also produces ^43/44gSc (Loveless, et al., 2021; Şekerci̇ et al., 2023). Like the calcium targets, ^natTi targets are feasible for producing ^43/44gSc, but production of high radionuclidic purity ^43/44gSc requires enriched titanium (⁴⁶Ti and ⁴⁷Ti) targets due to reactions on other stable isotopes of titanium (⁴⁶Ti, 8.25%; ⁴⁷Ti, 7.44%; ⁴⁸Ti, 73.72%; ⁴⁹Ti, 5.41%; ⁵⁰Ti, 5.18%). The ⁴⁶Ti(p,α)⁴³Sc and ⁴⁷Ti(p,α)^44gSc reaction routes have lower cross sections than proton-induced ^43/44gSc reactions on calcium target material, but ⁴⁶Ti and ⁴⁷Ti are cheaper than the enriched calcium material, and may also handle higher currents (Chernysheva, et al., 2021). Domnanich et al., 2017a, reported a 90–120 μAh, 15.1 ± 1.9 MeV proton irradiation of 10 mg of 97.0% ± 0.02% isotopically enriched ⁴⁶TiO₂ targets produced 1.17–1.22 MBq/μAh ⁴³Sc with 98.2% ± 0.3% radionuclidic purity. Similarly, a 183 µAh, 12.0 ± 2.3 MeV proton irradiation of 10 mg of 57.9% ± 1.8% isotopically enriched ⁴³CaCO₃ (12.36% ⁴⁴CaCO₃) produced 2.62 MBq/μAh ⁴³Sc with 66.2% ± 1.5% radionuclidic purity (33.3% ± 1.5% ^44gSc) (Domnanich, et al., 2017b). Production of the 33.3% ± 1.5% ^44gSc was the result of the (p,n) reaction on the 12.36% ⁴⁴CaCO₃ present in the target material. The two routes achieved similar radionuclidic purities with the ⁴⁶TiO₂(p,α)⁴³Sc reaction producing 99.7% ± 0.5% ^43/44gSc and the ⁴³CaCO₃(p,n)⁴³Sc reaction producing 99% ± 2% ^43/44gSc at end of bombardment.

For sites without accelerators, a generator of ^44gSc’s long-lived parent ⁴⁴Ti (t_1/2 = 60.0 year) (National Nuclear Data Center, Brookhaven National, 2022) has been developed (Gajecki and Marino., 2023; under review; Schumann et al., 2007; Loktionova, et al., 2010; Filosofov, et al., 2010; Pruszyński and Loktionova, 2010; Radchenko, et al., 2016; Radchenko, et al., 2017; Rane, et al., 2015; Kerdjoudj, et al., 2016; Müller, et al., 2018; Roesch, 2012)¹. ⁴⁴Ti’s long half-life, the potential for high efficiency elutions (>97%) (Filosofov, et al., 2010), and the lack of ^44mSc make this system appealing. For a⁴⁴Ti/^44gSc generator to be clinically viable, the generator must have low ⁴⁴Ti breakthrough, ^44gSc elutions suitable for labeling (low volume, high activity concentration), and consistent elution results. Several published generator designs use AG 1-X8 anion exchange resins which result in large elution volumes and ⁴⁴Ti breakthrough requiring reverse elution modes (Filosofov, et al., 2010; Pruszyński and Loktionova, 2010; Roesch, 2012; Kerdjoudj, et al., 2016). More recent designs use ZR hydroxamate resin which may increase generator lifetimes due to the increased number of elutions achieved before ⁴⁴Ti breakthrough is observed (Gajecki and Marino, 2023). A ⁴⁷Ca/⁴⁷Sc generator system for the therapeutic radioisotope of scandium is also in development (Mausner, et al., 1998; Rane, et al., 2015; Starovoitova, et al., 2015; Misiak, et al., 2017; Carzaniga et al., 2019; Pawlak, et al., 2019). However, ⁴⁷Ca production requires enriched ⁴⁸Ca which is expensive and not widely available.

It is also possible to produce ⁴⁷Sc through proton bombardment of titanium targets [^natTi(p,x)⁴³,^44g,⁴⁷Sc, ⁴⁸Ti(p,2n)⁴⁷Sc, ⁵⁰Ti(p,α)⁴⁷Sc]. These reactions suffer from co-production of ⁴⁶Sc and ⁴⁸Sc at high energies (>30–35 MeV) so lower proton energies are ideal (Chernysheva, et al., 2021). There are also two reaction routes to ⁴⁷Sc from calcium target material: ⁴⁸Ca(p,2n)⁴⁷Sc and ⁴⁴Ca(α,p)⁴⁷Sc (Minegishi, et al., 2016; Sitarz, et al., 2018; Carzaniga et al., 2019; Wibowo et al., 2021; Aikawa, et al., 2022). However, neither ⁴⁸Ca target material nor alpha accelerators are widely accessible, and these reactions routes will likely not have widespread use. Other alternative production paths to produce ⁴⁷Sc include proton bombardment of natural vanadium targets [^natV(p,x)⁴⁷Sc] (Jafari, et al., 2019; Barbaro, et al., 2021; De Nardo, et al., 2021), photonuclear reactions via electron linear accelerators [⁴⁸Ti(γ,p)⁴⁷Sc and ⁵¹V(γ,α)⁴⁷Sc] (Kazakov, et al., 2012; Rotsch, et al., 2018; Loveless, et al., 2019; Aliev, et al., 2020; Snow, et al., 2021) and reactor production with fast [⁴⁶Ca(n,γ)⁴⁷Ca → ⁴⁷Sc] or thermal [⁴⁷Ti(n,p)⁴⁷Sc] neutrons (Domnanich, et al., 2017b; Jalilian, et al., 2021).

Once radiochemically separated from the target material, ^44gSc has been successfully complexed and exhibited high in-vivo stability with the commonly used chelator 1,4,7,10-tetraazacyclododecane-1,4,7,10-tetraacetic acid (DOTA) (Huclier-Markai, et al., 2014). As an alternative to DOTA, novel chelators designed and optimized for scandium include the heptadendate AAZTA ligands, the nonmacrocyclic H₄pypa chelator, and the triaza-macrocycle-based picolinate-functionalized chelator H₃mpatcn that can be labeled with scandium at room temperature (Li, et al., 2020a; Vaughn, et al., 2020a; Li, et al., 2020b; Vaughn, et al., 2020b; Fersing, et al., 2022). In vivo and in vitro studies have investigated ^44gSc labeled with DOTA-based peptides (Hernandez, et al., 2014; van der Meulen, et al., 2015; Eppard, et al., 2017; Singh, et al., 2017) and conjugated AAZTA-PSMA peptides (Sinnes, et al., 2020; Ghiani, et al., 2021). Additionally, ^44gSc-labeled antibody fragments detected 4T1 murine breast cancer tumors and U87MG human glioblastoma tumors in murine models (Chakravarty, et al., 2014; Li, et al., 2020b).

Preclinical and clinical PET/CT studies have compared the image quality of ^44gSc (Bunka, et al., 2016; Ferguson, et al., 2019; Rosar, et al., 2020; Lima, et al., 2021) and, less frequently, ⁴³Sc (Lima, et al., 2021), to conventional PET radionuclides. These studies have shown that both ⁴³Sc and ^44gSc have favorable contrast and image resolution compared to ⁶⁸Ga. ⁴³Sc’s, and ^44gSc’s relatively longer half-lives enable transportation to nearby clinics, increasing their accessibility.

Currently, enriched calcium is only available as CaCO₃, a suboptimal material for cyclotron irradiation due to CaCO₃’s low thermal conductivity and decarboxylation (and subsequent release of CO₂ at high temperatures) in beam. Additionally, while previous works have investigated cross section measurements of various proton- and deuteron-induced reactions on calcium, there is a lack of comprehensive studies comparing the various radionuclidic purities and production yields for small cyclotron-produced ⁴³Sc and ^44gSc. The goal of this work is to investigate and optimize the possible production routes of both ⁴³Sc and ^44gSc using enriched calcium targets irradiated on a small cyclotron capable of accelerating 16 MeV protons and 8 MeV deuterons and compare their contrast on clinical PET/CT scanners.

2 Materials and methods

2.1 Chemicals

Enriched ⁴²CaCO₃, ⁴³CaCO₃, and ⁴⁴CaCO₃ materials were obtained from the Department of Energy National Isotope Development Center with isotopic purities of 93.58%, 83.93%, and 98.89%, respectively (Table 1). Solutions were prepared with optima grade HCl and HNO₃ purchased from Fisher Chemical and 18 MΩ·cm⁻¹ water. N,N,N′,N′-tetrakis-2-ethylhexyldiglycolamide (branched DGA) resin was purchased from Eichrom. DOTA chelator and DOTA-(Tyr³)-octreotate (DOTA-TATE) were purchased from Macrocyclics.

TABLE 1

TABLE 1. Isotopic composition of enriched CaCO₃ material received from the DOE National Isotope Development Center.

2.2 Target fabrication and cyclotron irradiation

Enriched CaCO₃ targets have been reported to withstand proton currents up to 50 μA when kept relatively thin (0.1–0.2 mm) and of low mass (<10 mg) (van der Meulen, et al., 2015). However, this material is very hygroscopic and readily absorbs water from the atmosphere. Additionally, when sufficiently heated in beam, CaCO₃ decarboxylates, forming CaO, offgassing CO₂, potentially damaging the target, and reducing accelerator beam transmission. A transition from preclinical to clinical scale productions suggests need for larger target masses and/or higher beam currents to increase yield. To compensate for this increased power deposition, we used CaO as an alternative target material. CaCO₃ was decarboxylated to CaO in a vertical tube furnace (Thermansys) heated in an inert atmosphere to 900°C–950°C. A graphite crucible containing approximately 150–300 mg of CaCO₃ was lowered into a flat-bottomed quartz tube (the graphite crucible facilitated removal of the powder) and a compression fitting with an argon gas line was connected to the quartz tube to prevent degradation of the graphite crucible during heating. The crucible and quartz tube were placed in the preheated furnace under argon gas flow for 60–90 min. To confirm complete thermal decomposition of the CaCO₃ material, the mass of the crucible and powder was measured before and after heating, and the decomposition of the graphite crucible was accounted for. Following the decarboxylation procedure, the CaO was quickly pressed into a 9.5 mm diameter niobium crucible with a hydraulic press (250 kg/cm²) and covered with a thin niobium cover foil to confine the target material and degrade the incident proton/deuteron energy (Figure 1). For the ⁴⁴Ca(p,n)^44gSc reaction, two incident energies were used. A 50 µm thick foil degraded the incident 16 MeV protons to 15.2 MeV and a 150 μm thick niobium cover foil degraded the incident energy to 13.6 MeV (calculated using SRIM) (Ziegler and Ziegler, 2010). Additionally, the 50 µm thick foil degraded the incident 8 MeV deuterons to 5.8 MeV. These energies were chosen to minimally degrade the 16 MeV proton beam (incident energy of 15.2 MeV) and 8 MeV deuteron beam (incident energy of 5.8 MeV) and to cover the peak of the cross-section curve (incident energy of 13.6 MeV). Figure 2 shows experimentally measured cross sections along with the energy range investigated. The target was then loaded on the University of Wisconsin, Madison’s GE PETtrace cyclotron or placed under vacuum in a furnace preheated to >100°C to keep the target from absorbing moisture. For the CaCO₃ targets, 75–150 mg CaCO₃ was first heated in a polypropylene centrifuge tube at 90°C for 30 min to ensure that the material was dry and then pressed directly into the niobium crucible. All targets were irradiated for 15°min to 1 h with 5–20 μA protons or 3–10 μA deuterons with water jet cooling on the back of the target.

FIGURE 1

FIGURE 1. Niobium crucible containing ∼60 mg pressed CaO powder (left) and covered with 50 μm niobium cover foil (right).

FIGURE 2

FIGURE 2. Cross sections for the ⁴²Ca(d,n)⁴³Sc, ⁴³Ca(d,n)^44gSc (top) (Carzaniga, et al., 2019; de Waal, et al., 1971a; Koning et al., 2012; de Waal, et al., 1971), ⁴³Ca(p,n)⁴³Sc, and ⁴⁴Ca(p,n)^44gSc (bottom) reactions. The energy range investigated is highlighted in gray and a fit of the experimental data is shown in blue.

To compare current tolerances, pressed powder ^natCaCO₃ and ^natCaO targets of equal thickness were irradiated with 13.6 MeV protons. Cyclotron vacuum pressure was monitored during irradiation, and peak and stabilized pressures were recorded as the current was increased in 1 μA steps.

2.3 HPGe spectrometry

Targets were assayed via High Purity Germanium (HPGe) gamma spectrometry with an aluminum-windowed detector (Ortec) coupled to a Canberra Model C1519 research amplifier and multi-channel analyzer (full-width half-maximum at 1,333 keV = 1.8 keV) to quantify radionuclide activities 10–30 min after irradiation. ²⁴¹Am, ¹³³Ba, ¹⁵²Eu, ¹³⁷Cs, and ⁶⁰Co check sources traceable to NIST were used to energy and efficiency calibrate the detector. Full targets were measured 200–400 cm from the detector face to quantify ^43/44g/44mSc activity, and longer-lived radioisotopes of scandium (⁴⁶Sc, ⁴⁷Sc, and ⁴⁸Sc) were quantified through 250 μL aliquots of the dissolved target (5 mL) measured overnight approximately 20 cm from the detector face. To determine end of bombardment (EoB) yields of long-lived scandium radioisotopes, the activity of ⁴³Sc or ^44gSc quantified in the aliquot provided a correction factor to obtain ^46,47,48Sc full target activity.

2.4 Radiochemical isolation of ^43/44gSc

Targets were dissolved in 5 mL 9 M HCl heated at 80°C with magnetic stir bar agitation for 30 min. Radiochemical separation of ^43/44gSc from ^42/43/44CaO target material used single-column extraction chromatography with branched DGA (132 ± 67 mg, 5.5 mm column diameter) resin pre-conditioned with 3 mL 0.01 M HCl followed by 5 mL 9 M HCl (Figure 3). The 5 mL dissolved target solution was loaded on to the conditioned resin using a peristaltic pump (1.6 mL/min) and collected for target recycling along with a 10 mL rinse of 6 M HCl. To remove trace metals and reduce the acidity in the column, a 12 mL 1 M HNO₃ rinse was performed and ^43/44gSc was subsequently eluted with four 500 μL fractions of 0.01 M HCl (Alliot, et al., 2015; Aluicio-Sarduy, et al., 2018). Activities were assayed in a Capintec CRC-15R dose calibrator (⁴³Sc setting #825; ^44gSc setting #938). Capintec dose calibrator settings were taken from the user’s manual and confirmed via cross-calibration with HPGe spectroscopy.

FIGURE 3

FIGURE 3. Ca/Sc separation strategy utilizing branched DGA resin.

2.5 Recovery of enriched calcium

The enriched calcium target material was recovered through precipitation from the 5 mL load and 10 mL rinse solutions. The calcium was precipitated as CaCO₃ through careful drop-wise addition of saturated (NH₄)₂CO₃ solution (approximately 30 mL), allowing 10–30 s for the solution to settle and CO₂ to be released between drops of (NH₄)₂CO₃. The precipitate was then centrifuged, decanted, and dried under argon at 120°C. The dried CaCO₃ could then be decarboxylated to CaO for future irradiations following the same procedure described above.

2.6 Apparent molar activity measurements and trace metals analysis

Trace metals analysis on the eluted ^43/44gSc solutions was performed by microwave plasma-atomic emission spectrometry (MP-AES) using an Agilent Technologies 4,200 spectrometer. Calibration standards containing 10 ppb to 100 ppm of Fe, Zn, Cu, and Ca were prepared in 0.01 M HCl solutions from 1,000 mg/L standards purchased from Sigma-Aldrich.

The apparent molar activity of the eluted ^43/44gSc solutions was determined through titration by reaction with varying concentrations of DOTA chelator. Polypropylene centrifuge tubes containing increasing concentrations of DOTA dissolved in 18 MΩ water (100 μL, 0–100 μg/mL) were prepared and small aliquots (10–15 μL, 3.7 ± 0.04 MBq) of ^43/44gSc eluate were added to each vial. Additionally, 2–5 μL of 1.5 M NaOAc (pH = 4.5) buffer were added to each vial to reach a final pH between 4–4.5. The solutions were then vortexed and heated on a shaker plate at 80°C and 300°RPM for 30 min. The percent ^43/44gSc complexed with DOTA was determined through thin layer chromatography (TLC) using aluminum backed silica gel plates (TLC Silica gel 60 F₂₅₄, Sigma-Aldrich) in 50 mM ethylenediaminetetraacetic acid (EDTA). Plates were imaged with storage phosphor autoradiography (Packard Cyclone). Under these conditions, the radiolabeled ^43/44gSc-DOTA stayed at the origin (retention factor, R_f = 0) while the free ^43/44gSc travelled with the mobile front (R_f = 1). The apparent molar activity (MBq/µmol) of ^43/44gSc with DOTA was found by dividing the activity of ^43/44gSc in each vial by twice the amount of DOTA required to complex 50% of the activity.

Radiolabeling experiments were also performed with the DOTA-based peptide DOTA-TATE following the same radiolabeling conditions listed above for DOTA. Briefly, 100 μL ^43/44gSc (37 ± 5 MBq) was vortexed with 2–3 μL DOTA-TATE (4.3–5.7°nmol, 0.2 mCi/nmol; 7.14°MBq/nmol) and 2–5 μL 1.5 M NaOAc pH 4.5 buffer and heated to 80°C while shaking at 300°RPM for 15–30 min. Percent complexation was determined through radio-TLC and high-performance liquid chromatography (HPLC) using a C18 column (4.6 × 250 mm, Dionex 5 μm 120 Å; ThermoFisher) and a water (0.1% trifluoroacetic acid, TFA) to acetonitrile gradient (5% MeCN, 0–3.5°min; 5%–35% MeCN, 3.5–13°min; 35%–90% MeCN, 13–15°min; 90% MeCN, 15–19°min; 90%–5% MeCN, 19–20°min; 5% MeCN, 20–25 min) flowing at 1 mL/min.

2.7 Clinical phantom scans

In this study, ⁴³Sc, ^44gSc, and a mixture of ⁴³Sc and ^44gSc (^43/44gSc) were compared to other positron-emitting radionuclides (¹⁸F, ⁶⁸Ga, and ⁶⁴Cu) in a clinical Derenzo phantom (acrylic) purchased from Phantech (Madison, WI). The University of Wisconsin, Madison’s GE PETtrace cyclotron was used to produce ⁴³Sc [⁴²CaO(d,n)⁴³Sc], ^44gSc [⁴⁴CaO(p,n)^44gSc], ¹⁸F [H₂¹⁸O(p,n)¹⁸F], and ⁶⁴Cu [⁶⁴Ni(p,n)⁶⁴Cu]. The mixture of ^43/44gSc (88% ⁴³Sc, 12% ^44gSc by activity at end of chemistry) was obtained through 13.6 MeV proton bombardment of 83.93% isotopically enriched ⁴³CaO targets containing 5.06% ⁴⁴Ca. ⁶⁸Ga was eluted from a GalliaPharm ⁶⁸Ge/⁶⁸Ga generator (5 mL, 0.1°M HCl, Eckert and Ziegler, Inc.) at the University of Wisconsin, Madison radiopharmaceutical production facility. Table 2 lists the physical decay characteristics of the radionuclides used in this study and the effective half-life, positron intensity, and mean positron energy for the ^43/44gSc mixture.

TABLE 2

TABLE 2. Physical decay characteristics and effective half-life and positron intensity of the radionuclides used in this study (National Nuclear Data Center, Brookhaven National, 2022).

The Derenzo phantom consisted of 6 sections of rods with diameters of 3.5, 4.0, 4.5, 5.0, 5.5, and 6.0°mm, and was filled with activities shown in Table 3. Filled activities were increased for low positron intensity radionuclides (⁶⁴Cu) and short-lived radionuclides (⁶⁸Ga and ¹⁸F) to keep scan times consistent (under 15 min). The concentrated activity solutions were prepared by first mixing 250 mL 18 MΩ water, 200 μL Tween^® 80 (Sigma-Aldrich), 25 μL 6 M HCl (final pH 4.5), and then adding the radioactivity (2.74–19.97 MBq; 0.075–2.16 mCi). Tween^® was used to prevent bubbles from forming in the linear Derenzo phantom and the acid was added to prevent the radiometals from sticking to the walls of the phantom. Additionally, 2–3 drops of food coloring were added to facilitate detection of air bubbles when filling the small rods of the Derenzo phantom. Radioactivity was drawn into 1 mL syringes and measured in a Capintec CRC-15R dose calibrator before and after adding to the prepared solutions.

TABLE 3

TABLE 3. Filled radioactivity derived from dose calibrator and HPGe recorded radioactivities and measured volumes.

The filled phantom was scanned on a clinical GE Discovery 710 and GE Discovery MI PET/CT (GE Healthcare, Waukesha, WI) scanner. The GE Discovery 710 has 13,824 lutetium-yttrium oxyorthosilicate (LYSO) scintillator crystals (4.2 × 6.3 × 25 mm³) optically connected to photomultiplier tubes without a light-guide (Yoon et al., 2015). The axial field of view is 15.7 cm, and the transaxial field of view is 70 cm. The energy window is 425–650 keV with a coincidence window of 4.9 ns. The GE Discovery MI has 19584 LYSO scintillator crystals (3.95 × 5.3 × 25 mm³) connected to arrays of silicon photomultiplier (SiPM) detectors through light-guides (Hsu et al., 2017). The axial field of view is 20 cm, and the transaxial field of view is 70 cm. The energy and coincidence windows are identical to the GE Discovery 710 at 425–650 keV and 4.9 ns, respectively. Ten million coincidences were collected regardless of which radionuclide the phantom was filled with. Data were reconstructed using an OSEM-TOF reconstruction algorithm (VPFX; without point spread function) with a 256 × 256 matrix size and 16 subsets, 8 iterations on the GE Discovery MI and 24 subsets, 5 iterations on the GE Discovery 710.

Contrast was compared across the six radionuclides. To determine contrast, a transverse slice in the center of the phantom was selected, and a line profile was drawn from the center of one outermost rod to the center of the opposing outermost rod. The maximum and minimum values were then used to calculate the contrast:

$C_{rod}=\frac{R_{\mathit{max}}-R_{\mathit{min}}}{R_{\mathit{max}}+R_{\mathit{min}}}$

where $R_{\mathit{max}}$ is the maximum value and $R_{\mathit{min}}$ is the minimum value from the line profile.

3 Results

3.1 Target fabrication and cyclotron irradiation

Pressed powder ^natCaCO₃ and ^natCaO targets of approximately equal thickness (0.63 ± 0.02 mm; 158 ± 11 mg) were fabricated and irradiated with 13.6 MeV protons. The time required to reach 20 μA was 61 min for the ^natCaCO₃ target and 23 min for the ^natCaO target. The ^natCaCO₃ targets generated more gas load on the cyclotron’s vacuum system, slowing their progression to 20 μA. Following irradiation, the ^natCaO target sustained far less damage and did not suffer material losses while the ^natCaCO₃ was fractured during irradiation (Figure 4).

FIGURE 4

FIGURE 4. Peak pressure on the target as the current was increased (left). Pictures of the ^natCaO (middle) and ^natCaCO₃ (right) targets immediately following 0–20 µA proton irradiation.

For deuteron production yield measurements (n = 3 per route), ^42/43CaO targets that were thick to the 5.8 MeV deuteron beam were fabricated (0.25 ± 0.01 mm; 61 ± 3 mg) (Figure 2). For proton production yield measurements (n = 3 per route), ^43/44CaO targets that were 4.7 MeV thick to the 13.6 MeV beam (0.62 ± 0.02 mm; 149 ± 2 mg) were fabricated to capture the flat region of the cross section curve for the ⁴³Ca(p,n)⁴³Sc and ⁴⁴Ca(p,n)^44gSc reactions (Figure 2) (Carzaniga, et al., 2017). Energy loss through the CaO targets was calculated using the Bethe-Bloch formula.

The ⁴⁴Ca(p,2n)⁴³Sc reaction has an energy threshold close to the maximum 16 MeV protons that the GE PETtrace can accelerate, so the thinner 50 μm niobium foil was used to minimally degrade the proton energy to 15.2 MeV (targets were kept the same mass and were consequently only 4.3 MeV thick for this reaction).

Targets were irradiated with either 5 μA protons or 3 μA deuterons with a total charge between 0.5 and 1.6 µAh. Production yields for the ⁴²Ca(d,n)⁴³Sc, ⁴³Ca(p,n)⁴³Sc, ⁴³Ca(d,n)^44gSc, ⁴⁴Ca(p,n)^44gSc, and ⁴⁴Ca(p,2n)⁴³Sc reactions are shown in Table 4. No ⁴³Sc from the ⁴⁴Ca(p,2n)⁴³Sc was observed, indicating that 15.2 MeV is too close to the energy threshold for the reaction to produce significant quantities of ⁴³Sc. This result agrees with previously reported data at 15.3 MeV (Carzaniga, et al., 2017).

TABLE 4

TABLE 4. Observed production yields for the 5 routes investigated along with target masses and energy thickness (Carzaniga, et al., 2017; Carzaniga, et al., 2019)^* (de Waal, et al., 1971)⁺.

Targets were also evaluated for radionuclidic purity via HPGe spectrometry. Figure 5 shows HPGe-measured yield decay-corrected to EoB (mCi/μAh) for each radioisotope of scandium, and the radionuclidic purity as a percentage of total activity decay-corrected to EoB is listed in Table 4 (n = 3 per reaction route).

FIGURE 5

FIGURE 5. Activity of each scandium radioisotope that was present in the full target at EoB for each of the 5 reaction routes that were investigated.

The GE PETtrace beam spot on target is larger than the 9.5 mm target diameter, so the control system’s measured integrated charge on target (µA·h) overestimates integrated fluence on the ^42/43/44Ca targets. To account for this, the niobium cover foil was used to determine the percentage of the beam that intercepted the target. After irradiation, a beam profile was obtained through autoradiography, and an ROI with a width of 9.5 mm was drawn and centered on the beam profile. A larger >20 mm diameter ROI was also drawn encompassing the entire beam spot. Additionally, four other 9.5 mm ROIs with an offset of up to 1.5 mm from center were drawn around the central region to determine the variance in beam intersection that may be caused from the beam drifting from the center. These values were independently measured for the proton and deuteron beams and were also evaluated before and after a cyclotron ion source puller electrode replacement that occurred during the measurement of these production yields. The measured beam intersections and the max difference between the offset regions are listed in Table 5 with a correction to the measured production yields listed in Table 4.

TABLE 5

TABLE 5. Measured beam intercepts for the proton and deuteron beams.

Figure 6 shows the predicted physical yields versus target thickness for a 13.6 MeV proton (left) and 5.8 MeV deuteron beam (right). Dashed lines are predicted yields taken from fits to published cross section values. The fitted cross sections used for this calculation are shown in blue in Figure 2. Experimentally measured, beam-spot corrected yields are additionally shown in Figure 6. Experimentally measured yields are similar to, but slightly lower than, predicted yields except in the case of the ⁴³Ca(d,n)^44gSc reaction where measured yields were higher than predicted. Predicted yields were based on cross-sections reported by Carzaniga (Carzaniga, et al., 2017; Carzaniga, et al., 2019) except for the ⁴³Ca(d,n)^44gSc reaction where the only experimental data available was reported in (de Waal et al., 1971a). For the other deuteron-induced reaction reported in (de Waal et al., 1971b) [⁴²Ca(d,n)⁴³Sc], the cross sections were lower than those reported by Carzaniga et al. (2019) for the same energy range. The discrepancy between cross sections reported by Carzaniga and de Waal are likely the cause of the difference in trends between the experimentally measured and predicted yields observed in this work.

FIGURE 6

FIGURE 6. Measured production yields compared to literature reported cross sections fitted to TENDL_19 predicted curves for 13.6 MeV incident protons (left) and 5.8 MeV incident deuterons (right) (de Waal, et al., 1971b; Levkovski, 1991; Koning et al., 2012; Krajewski, et al., 2013; Carzaniga, et al., 2017; Carzaniga, et al., 2019).

Figure 3 illustrates the separation strategy for isolating the Sc⁺³ ions from the Ca⁺² ions. The low affinity for Ca⁺² ions on the branched DGA resin permitted efficient recycling from the 5 mL load and 10 mL 6 M HCl rinse stages of the separation (Alliot, et al., 2015). Other +2 metals (such as Cu⁺²) are also rinsed from the resin in this step. The 1 M HNO₃ wash removes metals such as iron and zinc due to the low affinity for these ions in 1 M HNO₃ and minimal loss of Sc⁺³ was observed as seen in the elution profile (Figure 7) (Alliot, et al., 2015). Elution with 2 mL of 0.01 M HCl effectively removes 92.8% ± 4.0% of the loaded scandium. However, there is large variation in the activity that is in each 0.01 M HCl elution fraction (n = 10), perhaps due to variations in volumes of liquid being used to elute the activity (residual liquid may be left in tubing or on walls in parts of the system). Following a 1 h, 5 μA, 13.6 MeV proton irradiation of a 150 mg ⁴⁴CaO target, this chemical isolation procedure isolates 1.4 ± 0.3 GBq (38 ± 9 mCi) of ^44gSc with 99.5% ± 0.05% radionuclidic purity in 2 mL 0.01 M HCl 2 h after EoB.

FIGURE 7

FIGURE 7. Elution profile for the Ca⁺²/Sc⁺³ radiochemical separation used in this work. Activity distribution in each of the 4 eluted fractions varies with each experiment, but the overall eluted fraction remains consistent.

3.2 Recovery of enriched calcium

^42/43/44Ca was recovered in the form of ^42/43/44CaCO₃ through slow addition of (NH₄)₂CO₃ and converted back to ^42/43/44CaO through decarboxylation in a vertical tube furnace (900°C for 60–90 min) for future irradiations. The enriched ^42/43/44CaCO₃ target material was recovered with an efficiency of 95% ± 2% (n = 20). If argon gas is passed over the graphite crucible during the decarboxylation procedure, the crucible surface remains smooth and losses from transferring and heating are <1%.

3.3 Apparent molar activity measurements and trace metals analysis

The quality of the final ^43/44gSc eluent was determined through titrimetric reactions with the chelator DOTA and the ability to label with the DOTA-based peptide DOTA-TATE. Molar activity of ^43/44gSc produced through a 30–60 min, 13.6 MeV proton energy and 5 μA current irradiation of either ⁴³CaO or ⁴⁴CaO targets was 48 ± 11 GBq/μmol (1.3 ± 0.3 Ci/μmol) (n = 3) at end of chemistry. Additionally, 37 ± 5 MBq (1 ± 0.14 mCi) aliquots of ^43/44gSc were labeled with DOTA-TATE at a molar activity of 7.14 MBq/nmol (0.2 mCi/nmol) and assayed using HPLC. Radio-HPLC chromatograms (Figure 8) show near quantitative labeling (>98%). Free ^43/44gSc eluted with the solvent front (4 min) and ^43/44gSc-DOTA-TATE eluted at 15.6 min.

FIGURE 8

FIGURE 8. Radio-HPLC chromatogram of ^44gSc-DOTA-TATE.

Amounts of Ca and commonly seen metals such as Fe, Zn, and Cu that may affect Sc-labelling were measured through MP-AES and are listed in Table 6 (n = 5). The efficient removal of metals such as Fe, Zn, and Cu in the radiochemical isolation strategy resulted in the high apparent molar activity observed with Sc and DOTA. The mass of Ca that was co-loaded on to the column and seen in the eluent resulted in minimal enriched material loss (<1 mg).

TABLE 6

TABLE 6. Trace metals analysis of an eluted fraction of ^43/44gSc using MP-AES.

3.4 Clinical phantom scans

The contrast, C_rod, was compared between the 6 radionuclides on the two PET/CT scanners. Figure 9 shows a cross-sectional image of the reconstructed Derenzo phantom on the GE Discovery 710 (similar images were seen on the GE Discovery MI). Qualitative differences can be seen between shorter-range positron-emitters such as ⁶⁴Cu and ¹⁸F compared to the higher-range positron-emitter ⁶⁸Ga. All six sections of rods are easily distinguishable for ⁶⁴Cu and ¹⁸F while the lower diameter rods (3.5 and 4 mm) are significantly blurred in the ⁶⁸Ga image. Radioisotopes of scandium demonstrate image quality between the low and high positron range radionuclides where the smallest diameter rods are indistinguishable, but resolution improves with the 4 mm diameter rods. The calculated contrast values are shown in Figure 9 as well.

FIGURE 9

FIGURE 9. Reconstructed PET scans of the Derenzo phantom filled with each radionuclide on the GE Discovery 710 (top). Comparison of contrast ($C_{rod}$) for each radionuclide investigated on both scanners (bottom).

On both scanners, ⁶⁴Cu and ¹⁸F demonstrated the greatest contrast across the varying rod diameters while ⁶⁸Ga performed the worst across all rod diameters which is consistent with what we would expect from the positron energies listed in Table 2. Between ⁴³Sc, ^44gSc, and ^43/44gSc, there is no significant difference in contrast, suggesting minimal benefit in contrast from ⁴³Sc’s softer positron energy using these clinical PET/CT systems.

4 Discussion

The goal of this study was to compare production yields, radionuclidic purity, and PET image quality between ⁴³Sc and ^44gSc. Among the six reactions that produce ⁴³Sc and ^44gSc with protons and deuterons on small cyclotrons, the ⁴⁴Ca(p,2n)⁴³Sc and ⁴⁴Ca(d,2n)^44gSc reactions are only possible above 15.2 MeV protons and 5.8 MeV deuterons, eliminating some commercial machines from consideration. The proton-induced reactions investigated here [⁴³Ca(p,n)⁴³Sc and ⁴⁴Ca(p,n)^44gSc] had significantly higher production yields (by approximately a factor of 10) than the deuteron-induced reactions [⁴²Ca(d,n)⁴³Sc and ⁴³Ca(d,n)^44gSc] but were performed with higher mass targets (approximately 150 mg versus 60 mg CaO) and higher incident energies (13.6 or 15.2 MeV protons versus 5.8 MeV deuterons). The peak of the cross-section curves for the deuteron-induced reactions were not fully captured with the energies and target thickness used here (Figure 2, gray area). Accelerators capable of accelerating higher energy deuterons would be able to produce higher yields through these routes. Thick target yields for the proton-induced reactions were not investigated in this work due to the large masses (∼250 mg CaO) required to degrade the proton energy in the target from 13.6 MeV to 4 MeV where the cross section approaches zero (Carzaniga, et al., 2017).

Due to the relatively low isotopic purity of available ⁴³Ca (83.93%), an EoB radionuclidic purity of only 87.81% ± 0.03% ⁴³Sc was achievable through proton bombardment of ⁴³Ca. The main radioisotopic contaminants from this reaction are ^44gSc at 12.16% ± 0.03% total activity and ⁴⁸Sc (t_1/2 = 43.71 h; β^-_mean = 220.5 keV; β⁺_intensity = 100%) at 0.02% ± 0.01% total activity at EoB. The production of ^44gSc results from a (p,n) reaction on the 5.06% ⁴⁴Ca present in the enriched ⁴³Ca target material and could be minimized through use of higher purity ⁴³Ca target material. With 100% isotopically pure ⁴³Ca, ^44gSc would be removed and ⁴³Ca(p,n)⁴³Sc measured yields would increase to 288.6 MBq/μAh (7.38 mCi/μAh), or 557.2 MBq/μAh (15.06 mCi/μAh) when corrected for beam intercept. Considering the alternative deuteron-induced reaction, ⁴²Ca(d,n)⁴³Sc, the radionuclidic purity at EoB was 99.36% ± 0.05% ⁴³Sc at EoB with no co-production of ^44gSc’s metastable state ^44mSc. However, this route will require longer irradiation times and a cyclotron capable of accelerating deuterons to produce the same EoB activity as the proton-induced route.

For producing ^44gSc, the ⁴⁴Ca(p,n)^44gSc route results both in higher production yields and greater radionuclidic purity than the ⁴³Ca(d,n)^44gSc route with the conditions investigated in this work. This is partially due to the low purity of the ⁴³Ca and the smaller mass of the target. Cyclotrons capable of accelerating higher energy deuterons (>8 MeV) along with higher purity and thicker ⁴³Ca targets would increase the production yields compared to those reported here. However, the higher predicted cross sections for the ⁴⁴Ca(p,n)^44gSc reaction compared to ⁴³Ca(d,n)^44gSc (Figure 2) suggest that the highest yields of ^44gSc would still be through proton bombardment of ⁴⁴Ca. The presence of longer-lived radioisotopes of scandium such as ⁴⁶Sc are also comparable in both reaction routes in the target thickness and energy range investigated herein: At EoB, there was 0.007% ± 0.003% ⁴⁶Sc for ⁴⁴Ca(p,n)^44gSc and 0.013% ± 0.003% ⁴⁶Sc for ⁴³Ca(d,n)^44gSc.

When considering translation to clinical studies, long-lived radioisotopes of scandium (⁴⁶Sc, ⁴⁷Sc, and ⁴⁸Sc) should be minimized to avoid patient dose and handling of radioactive waste (De Nardo, et al., 2021). The threshold for ⁶⁸Ge (t_1/2 = 270.8 days) impurity set by the European Pharmacopoeia in ⁶⁸Ge/⁶⁸Ga generator elutions is 0.001% (Council of Europe, 2013, 2013). The presence of ⁴⁶Sc (t_1/2 = 83.79 days), ⁴⁷Sc (t_1/2 = 3.3492 days), and ⁴⁸Sc (t_1/2 = 43.71 h) for the reaction routes at the irradiation parameters investigated herein are lower than the 0.001% ⁶⁸Ge limit at EoB with the exception of ⁴³Ca(p,n)⁴³Sc which results in 0.0012% ⁴⁸Sc at the end of bombardment.

Radiochemical isolation with branched DGA resin was procedurally simple and achieved efficient target recycling and high radiochemical yields [95% ± 2% enriched calcium recovery (n = 20) and 92% ± 4% ^43/44gSc activity in 2 mL 0.01 M HCl (n = 10)]. From the calcium mass that was measured via MP-AES, the Ca/Sc separation factor is on the order of ∼10⁵. Quantitative labeling with DOTA-TATE and high apparent molar activities with DOTA were additionally achieved with this separation strategy and minimal trace metals were observed through MP-AES analysis.

Because PET resolution is impacted by positron range (energy), ⁴³Sc may have better PET resolution than ^44gSc. Additionally, it has been suggested that ^44gSc’s higher energy gamma (1,157 keV; 99.9%) as opposed to ⁴³Sc’s lower energy gamma (373 keV; 22.5%) may contribute negatively to image quality due to the potential to scatter within a PET scanner’s 511 keV window (Carzaniga, et al., 2017) and negatively to patient dosimetry. In this study we observed no significant difference in the contrast of the phantoms filled with ⁴³Sc, ^44gSc, or the mixture ^43/44gSc.

5 Conclusion

This work demonstrated that isotopically enriched CaO targets can be used to produce high yield and high radionuclidic purity ⁴³Sc and ^44gSc through proton and deuteron bombardment. While several reaction routes produce either ⁴³Sc or ^44gSc, laboratory capabilities, circumstances, and budgets are more likely to dictate which reaction route is chosen. When comparing PET image quality, ⁴³Sc and ^44gSc contrast were comparable, and in some cases favorable compared to other conventional PET radionuclides. In addition, no significant difference in contrast was observed due to the lower energy of ⁴³Sc’s positron compared to ^44gSc. These results indicate that, while ^44gSc’s higher energy gamma emission and co-produced metastable state (^44mSc) may have dosimetric consequences for in-human studies in the future, with regards to yields, radionuclidic purity, and image contrast, ^44gSc has superior or comparable characteristics compared to ⁴³Sc.

Data availability statement

The raw data supporting the conclusion of this article will be made available by the authors, without undue reservation.

Author contributions

KB: Methodology, experimental investigation, analysis, experimental design, original draft writing, review and editing EA-S: Methodology, analysis, review and editing TB: Experimental investigation, analysis, review and editing SH: Analysis, review and editing AO: Analysis, review and editing KB: Analysis, review and editing JB: Resources, methodology, review PE: Methodology, analysis, review and editing TB: Resources, review and editing AP: Experimental design, resources, methodology, analysis, review and editing JE: Experimental design, methodology, analysis, review and editing.

Funding

Research reported in this publication was supported by the National Cancer Institute of the National Institutes of Health under Award Number T32CA009206. This work was supported by an award from the Department of Energy Isotope Program AO gratefully acknowledges support from NIH F31 Ruth L Kirschstein Predoctoral Individual National Research Service Award F31CA239617.

Conflict of interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Footnotes

¹Kretowicz, M. N. (2023). Promising Scandium-44 radionuclides for nuclear medicine from a ⁴⁴Ti/^44gSc generator.

References

Aikawa, M., Hanada, Y., Ichinkhorloo, D., Haba, H., Takács, S., Ditrói, F., et al. (2022). Production cross sections of ⁴⁷Sc via alpha-particle-induced reactions on natural calcium up to 29 MeV. Nucl. Instrum. Methods Phys. Res. Sect. B Beam Interact. Mater. Atoms 515, 1–6. doi:10.1016/j.nimb.2022.01.008

CrossRef Full Text | Google Scholar

Aliev, R. A., Belyshev, S. S., Furkina, E. B., Khankin, V. V., Kuznetsov, A. A., Dzhilavyan, L. Z., et al. (2020). Photonuclear production of medically relevant radionuclide ⁴⁷Sc. J. Radioanalytical Nucl. Chem. 326, 1099–1106. doi:10.1007/s10967-020-07400-5

CrossRef Full Text | Google Scholar

Alliot, C., Kerdjoudj, R., Michel, N., Haddad, F., and Huclier-Markai, S. (2015). Cyclotron production of high purity ^44m,44Sc with deuterons from ⁴⁴CaCO₃ targets. Nucl. Med. Biol. 42 (5), 524–529. doi:10.1016/j.nucmedbio.2015.03.002

PubMed Abstract | CrossRef Full Text | Google Scholar

Aluicio-Sarduy, E., Hernandez, R., Valdovinos, H. F., Kutyreff, C. J., Ellison, P. A., Barnhart, T. E., et al. (2018). Simplified and automatable radiochemical separation strategy for the production of radiopharmaceutical quality ⁸⁶Y using single column extraction chromatography. Appl. Radiat. Isotopes 142, 28–31. doi:10.1016/j.apradiso.2018.09.016

PubMed Abstract | CrossRef Full Text | Google Scholar

Barbaro, F., Canton, L., Carante, M. P., Colombi, A., De Dominicis, L., Fontana, A., et al. (2021). New results on proton-induced reactions on vanadium for ⁴⁷Sc production and the impact of level densities on theoretical cross sections. Phys. Rev. C 104, 044619. doi:10.1103/physrevc.104.044619

CrossRef Full Text | Google Scholar

Bunka, M., Müller, C., Vermeulen, C., Haller, S., Türler, A., Schibli, R., et al. (2016). Imaging quality of ⁴⁴Sc in comparison with five other PET radionuclides using Derenzo phantoms and preclinical PET. Appl. Radiat. Isotopes 110, 129–133. doi:10.1016/j.apradiso.2016.01.006

PubMed Abstract | CrossRef Full Text | Google Scholar

Carzaniga, T. S., Auger, M., Braccini, S., Bunka, M., Ereditato, A., Nesteruk, K. P., et al. (2017). Measurement of ⁴³Sc and ⁴⁴Sc production cross-section with an 18 MeV medical PET cyclotron. Appl. Radiat. Isotopes 129, 96–102. doi:10.1016/j.apradiso.2017.08.013

PubMed Abstract | CrossRef Full Text | Google Scholar

Carzaniga, T. S., and Braccini, S. (2019). Cross-section measurement of ^44mSc,⁴⁷Sc, ⁴⁸Sc and ⁴⁷Ca for an optimized ⁴⁷Sc production with an 18 MeV medical PET cyclotron. Appl. Radiat. Isotopes 143, 18–23. doi:10.1016/j.apradiso.2018.10.015

PubMed Abstract | CrossRef Full Text | Google Scholar

Carzaniga, T. S., van der Meulen, N. P., Hasler, R., Kottler, C., Peier, P., Türler, A., et al. (2019). Measurement of the ⁴³Sc production cross-section with a deuteron beam. Appl. Radiat. Isotopes 145, 205–208. doi:10.1016/j.apradiso.2018.12.031

PubMed Abstract | CrossRef Full Text | Google Scholar

Chakravarty, R., Goel, S., Valdovinos, H. F., Hernandez, R., Hong, H., Nickles, R. J., et al. (2014). Matching the decay half-life with the biological half-life: ImmunoPET imaging with ⁴⁴Sc-labeled cetuximab fab fragment. Bioconjugate Chem. 25 (12), 2197–2204. doi:10.1021/bc500415x

PubMed Abstract | CrossRef Full Text | Google Scholar

Chernysheva, M., Loveless, S. C., Brossard, T., Becker, K., Cingoranelli, S., Aluicio-Sarduy, E., et al. (2021). Accelerator production of scandium radioisotopes: Sc-43, Sc-44, and Sc-47. Curr. Radiopharm. 14 (4), 359–373. doi:10.2174/1874471014999210112205535

PubMed Abstract | CrossRef Full Text | Google Scholar

Council of Europe (2013). Gallium (68Ga) edotreotide injection. European Directorate for the Quality of Medicines.

Google Scholar

De Nardo, L., Mou, L., Furlanetto, D., Rosato, A., Esposito, J., Meléndez-Alafort, L., et al. 2021. Preliminary dosimetric analysis of DOTA-folate radiopharmaceutical radiolabelled with 47Sc produced through natV(p,x)47Sc cyclotron irradiation. Phys. Med. Biol., 66(2).025003, doi:10.1088/1361-6560/abc811

PubMed Abstract | CrossRef Full Text | Google Scholar

de Waal, T. J., Peisach, M., and Pretorius, R. (1971b). Activation cross sections for proton-induced reactions on calcium isotopes up to 5·6 MeV. J. Inorg. Nucl. Chem. 33 (9), 2783–2789. doi:10.1016/0022-1902(71)80038-6

CrossRef Full Text | Google Scholar

de Waal, T. J., Peisach, M., and Pretorius, R. (1971a). Activation gross sections for deuteron-induced reactions on calcium isotopes up to 5.5 MeV. Radiochim. Acta 15, 123–127. doi:10.1524/ract.1971.15.3.123

CrossRef Full Text | Google Scholar

Domnanich, K. A., Eichler, R., Müller, C., Jordi, S., Yakusheva, V., Braccini, S., et al. (2017a). Production and separation of ⁴³Sc for radiopharmaceutical purposes. EJNMMi 2 (1), 14. doi:10.1186/s41181-017-0033-9

PubMed Abstract | CrossRef Full Text | Google Scholar

Domnanich, K. A., Müller, C., Benešová, M., Dressler, R., Haller, S., Köster, U., et al. (2017b). ⁴⁷Sc as useful β–-emitter for the radiotheragnostic paradigm: A comparative study of feasible production routes. EJNMMI Radiopharm. Chem. 2 (5), 5. doi:10.1186/s41181-017-0024-x

PubMed Abstract | CrossRef Full Text | Google Scholar

Eppard, E., de la Fuente, A., Benešová, M., Khawar, A., Bundschuh, R. A., Gärtner, F. C., et al. (2017). Clinical translation and first in-human use of [⁴⁴Sc]Sc-PSMA-617 for PET imaging of metastasized castrate-resistant prostate cancer. Theranostics 7 (18), 4359–4369. doi:10.7150/thno.20586

PubMed Abstract | CrossRef Full Text | Google Scholar

Ferguson, S., Jans, H. S., Wuest, M., Riauka, T., and Wuest, F. (2019). Comparison of scandium-44 g with other PET radionuclides in pre-clinical PET phantom imaging. EJNMMI Phys. 6 (23), 23. doi:10.1186/s40658-019-0260-0

PubMed Abstract | CrossRef Full Text | Google Scholar

Fersing, C., Masurier, N., Rubira, L., Deshayes, E., and Lisowski, V. (2022). AAZTA-derived chelators for the design of innovative radiopharmaceuticals with theranostic applications. Pharmaceuticals 15 (2), 234. doi:10.3390/ph15020234

PubMed Abstract | CrossRef Full Text | Google Scholar

Filosofov, D. V., Loktionova, N. S., and Rösch, F. (2010). A ⁴⁴Ti/⁴⁴Sc radionuclide generator for potential application of ⁴⁴Sc-based PET-radiopharmaceuticals. rca-Radiochimica Acta 98 (3), 149–156. doi:10.1524/ract.2010.1701

CrossRef Full Text | Google Scholar

Gains, J. E., Bomanji, J. B., Fersht, N. L., Sullivan, T., D'Souza, D., Sullivan, K. P., et al. (2011). ¹⁷⁷Lu-DOTATATE molecular radiotherapy for childhood neuroblastoma. J. Nucl. Med. 52 (7), 1041–1047. doi:10.2967/jnumed.110.085100

PubMed Abstract | CrossRef Full Text | Google Scholar

Gajecki, L., Marino, C., Cutler, C. S., and Sanders, V. A. (2023). Evaluation of hydroxamate-based resins towards a more clinically viable 44Ti/44Sc radionuclide generator. Appl. Radiat. Isotopes 192, 110588. doi:10.1016/j.apradiso.2022.110588

CrossRef Full Text | Google Scholar

Ghiani, S., Hawala, I., Szikra, D., Trencsényi, G., Baranyai, Z., Nagy, G., et al. (2021). Synthesis, radiolabeling, and pre-clinical evaluation of [⁴⁴Sc]Sc-AAZTA conjugate PSMA inhibitor, a new tracer for high-efficiency imaging of prostate cancer. Eur. J. Nucl. Med. Mol. Imaging 48, 2351–2362. doi:10.1007/s00259-020-05130-0

PubMed Abstract | CrossRef Full Text | Google Scholar

Grubmüller, B., Senn, D., Kramer, G., Baltzer, P., D’Andrea, D., Grubmüller, K. H., et al. (2018). Response assessment using ⁶⁸Ga-PSMA ligand PET in patients undergoing ¹⁷⁷Lu-PSMA radioligand therapy for metastatic castration-resistant prostate cancer. Eur. J. Nucl. Med. Mol. Imaging 46, 1063–1072. doi:10.1007/s00259-018-4236-4

PubMed Abstract | CrossRef Full Text | Google Scholar

Heinzel, A., Boghos, D., Mottaghy, F. M., Gaertner, F., Essler, M., von Mallek, D., et al. (2019). ⁶⁸Ga-PSMA PET/CT for monitoring response to ¹⁷⁷Lu-PSMA-617 radioligand therapy in patients with metastatic castration-resistant prostate cancer. Eur. J. Nucl. Med. Mol. Imaging 46, 1054–1062. doi:10.1007/s00259-019-4258-6

PubMed Abstract | CrossRef Full Text | Google Scholar

Hernandez, R., Valdovinos, H. F., Yang, Y., Chakravarty, R., Hong, H., Barnhart, T. E., et al. (2014). ⁴⁴Sc: An attractive isotope for peptide-based PET imaging. Mol. Pharm. 11 (8), 2954–2961. doi:10.1021/mp500343j

PubMed Abstract | CrossRef Full Text | Google Scholar

Hsu, D. F., Ilan, E., Peterson, W. T., Uribe, J., Lubberink, M., and Levin, C. S. (2017). Studies of a next-generation silicon-photomultiplier–based time-of-flight PET/CT system. J. Nucl. Med. 58 (9), 1511–1518. doi:10.2967/jnumed.117.189514

PubMed Abstract | CrossRef Full Text | Google Scholar

Huclier-Markai, S., Kerdjoudj, R., Alliot, C., Bonraisin, A., Michel, N., Haddad, F., et al. (2014). 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. 41, e36–e43. doi:10.1016/j.nucmedbio.2013.11.004

PubMed Abstract | CrossRef Full Text | Google Scholar

Jafari, A., Aboudzadeh, M. R., Azizakram, H., Sadeghi, M., Alirezapour, B., Rajabifar, S., et al. (2019). Investigations of proton and deuteron induced nuclear reactions on natural and enriched Titanium, Calcium and Vanadium targets, with special reference to the production of ⁴⁷Sc. Appl. Radiat. Isotopes 152, 145–155. doi:10.1016/j.apradiso.2019.07.007

PubMed Abstract | CrossRef Full Text | Google Scholar

Jalilian, A. R., et al. (2021). IAEA activities on ⁶⁷Cu, ¹⁸⁶Re, ⁴⁷Sc theranostic radionuclides and radiopharmaceuticals. Curr. Radiopharm. 14 (4), 306–314. doi:10.2174/18744729mtewimjka0

PubMed Abstract | CrossRef Full Text | Google Scholar

Kazakov, A. G., Ekatova, T. Y., and Babenya, J. S. (2012). Photonuclear production of medical radiometals: A review of experimental studies. J. Radioanalytical Nucl. Chem. 328, 493–505. doi:10.1007/s10967-021-07683-2

CrossRef Full Text | Google Scholar

Kerdjoudj, R., Pniok, M., Alliot, C., Kubíček, V., Havlíčková, J., Rösch, F., et al. (2016). Scandium(III) complexes of monophosphorus acid DOTA analogues: A thermodynamic and radiolabelling study with ⁴⁴Sc from cyclotron and from a ⁴⁴Ti/⁴⁴Sc generator. Dalton Trans. 45, 1398–1409. doi:10.1039/c5dt04084a

PubMed Abstract | CrossRef Full Text | Google Scholar

Koning, A. J., and Rochman, D. (2012). Modern nuclear data evaluation with the TALYS code system. Nucl. Data Sheets 113 (12), 2841–2934. doi:10.1016/j.nds.2012.11.002

CrossRef Full Text | Google Scholar

Krajewski, S., Cydzik, I., Abbas, K., Bulgheroni, A., Simonelli, F., Holzwarth, U., et al. (2013). Cyclotron production of ⁴⁴Sc for clinical application. Radiochim. Acta 101, 333–338. doi:10.1524/ract.2013.2032

CrossRef Full Text | Google Scholar

Levkovski, V. N. (1991). Cross sections of medium mass nuclide activation (A= 40-100) by medium energy protons and alpha-particles (E= 10-50 MeV). Tallinn: Inter-Vesi.

Google Scholar

Li, L., Jaraquemada-Peláez, M. d. G., Aluicio-Sarduy, E., Xiaozhu, W., et al. (2020a). Coordination chemistry of [Y(pypa)]− and comparison immuno-PET imaging of [⁴⁴Sc]Sc- and [⁸⁶Y]Y-pypa-phenyl-TRC105. R. Soc. Chem. 49, 5547–5562. doi:10.1039/D0DT00437E

PubMed Abstract | CrossRef Full Text | Google Scholar

Li, L., Jaraquemada-Peláez, M. d. G., Aluicio-Sarduy, E., Wang, X., Jiang, D., Sakheie, M., et al. (2020b). [^nat/44Sc(pypa)]−: Thermodynamic stability, radiolabeling, and biodistribution of a prostate-specific-membrane-antigen-targeting conjugate. Inorg. Chem. 59 (3), 1985–1995. doi:10.1021/acs.inorgchem.9b03347

PubMed Abstract | CrossRef Full Text | Google Scholar

Lima, T. V. M., Gnesin, S., Strobel, K., Pérez, M. d. S., Roos, J. E., Müller, C., et al. (2021). Fifty shades of scandium: Comparative study of PET capabilities using Sc-43 and Sc-44 with respect to conventional clinical radionuclides. Diagnostics 11 (10), 1826. doi:10.3390/diagnostics11101826

PubMed Abstract | CrossRef Full Text | Google Scholar

Loktionova, N., Dmitry, F., Marek, P., Frank, R., et al. (2010). Design and performance of a novel 5 mCi ⁴⁴Ti/⁴⁴Sc-radionuclide generator. J. Nucl. Med. 51, 1467.

Google Scholar

Loveless, C. S., Blanco, J. R., Diehl, G. L., Elbahrawi, R. T., Carzaniga, T. S., Braccini, S., et al. (2021). Cyclotron production and separation of scandium radionuclides from natural titanium metal and titanium dioxide targets. J. Nucl. Med. 62 (1), 131–136. doi:10.2967/jnumed.120.242941

PubMed Abstract | CrossRef Full Text | Google Scholar

Loveless, C. S., Radford, L. L., Ferran, S. J., Queern, S. L., Shepherd, M. R., and Lapi, S. E. (2019). Photonuclear production, chemistry, and in vitro evaluation of the theranostic radionuclide ⁴⁷Sc. EJNMMI Res. 9 (42), 42. doi:10.1186/s13550-019-0515-8

PubMed Abstract | CrossRef Full Text | Google Scholar

Mausner, L. F., Kolsky, K., Joshi, V., and Srivastava, S. (1998). Radionuclide development at BNL for nuclear medicine therapy. Appl. Radiat. Isotopes 49 (4), 285–294. doi:10.1016/s0969-8043(97)00040-7

PubMed Abstract | CrossRef Full Text | Google Scholar

Minegishi, K., Kotaro, N., Masami, F., Hisashi, S., Tomoyuki, O., Ming-Rong, Z., et al. 2016. Production of scandium-43 and -47 from a powdery calcium oxide target via the nat/44Ca(α,x)-channel. Appl. Radiat. Isotopes, Volume 116, pp. 8–12.doi:10.1016/j.apradiso.2016.07.017

PubMed Abstract | CrossRef Full Text | Google Scholar

Misiak, R., Walczak, R., Wąs, B., Bartyzel, M., Mietelski, J. W., and Bilewicz, A. (2017). ⁴⁷Sc production development by cyclotron irradiation of ⁴⁸Ca. J. Radioanalytical Nucl. Chem. 313, 429–434. doi:10.1007/s10967-017-5321-z

PubMed Abstract | CrossRef Full Text | Google Scholar

Müller, C., Bunka, M., Reber, J., Fischer, C., Zhernosekov, K., Türler, A., et al. (2013). 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. 5 (12), 2168–2174. doi:10.2967/jnumed.113.123810

PubMed Abstract | CrossRef Full Text | Google Scholar

Müller, C., Domnanich, K. A., Umbricht, C. A., and van der Meulen, N. P. (2018). Scandium and terbium radionuclides for radiotheranostics: Current state of development towards clinical application. Br. J. Radiology 91 (1091), 20180074. doi:10.1259/bjr.20180074

CrossRef Full Text | Google Scholar

National Nuclear Data Center, Brookhaven National , 2022. National nuclear data center, brookhaven national. Available at: https://www.nndc.bnl.gov/nudat3/[Accessed 05 April 2023].

Google Scholar

Pawlak, D., Wojdowska, W., Parus, L. J., Cieszykowska, I., Zoltowska, M., Garnuszek, P., et al. (2019). Comparison of separation methods for ⁴⁷Ca/⁴⁷Sc radionuclide generator. Appl. Radiat. Isotopes 151, 140–144. doi:10.1016/j.apradiso.2019.05.020

PubMed Abstract | CrossRef Full Text | Google Scholar

Pruszyński, M., Loktionova, N. S., Filosofov, D., and Rösch, F. (2010). Post-elution processing of ⁴⁴Ti/⁴⁴Sc generator-derived ⁴⁴Sc for clinical application. Appl. Radiat. Isotopes 68 (9), 1636–1641. doi:10.1016/j.apradiso.2010.04.003

CrossRef Full Text | Google Scholar

Radchenko, V., Engle, J. W., Medvedev, D. G., Maassen, J. M., Naranjo, C. M., Unc, G. A., et al. (2017). Proton-induced production and radiochemical isolation of ⁴⁴Ti from scandium metal targets for ⁴⁴Ti/⁴⁴Sc generator development. Nucl. Med. Biol. 50, 25–32. doi:10.1016/j.nucmedbio.2017.03.006

PubMed Abstract | CrossRef Full Text | Google Scholar

Radchenko, V., Meyer, C., Engle, J., Naranjo, C., Unc, G., Mastren, T., et al. (2016). Separation of ⁴⁴Ti from proton irradiated scandium by using solid-phase extraction chromatography and design of ⁴⁴Ti/⁴⁴Sc generator system. J. Chromatogr. A 1477, 39–46. doi:10.1016/j.chroma.2016.11.047

PubMed Abstract | CrossRef Full Text | Google Scholar

Rane, A. T., and Bhatki, K. S. (1966). Rapid radiochemical separations of strontium-90-yttrium-90 and calcium-45-scandium-46 on a cation exchange resin. Anal. Chem. 38 (11), 1598–1601. doi:10.1021/ac60243a039

CrossRef Full Text | Google Scholar

Rane, S., Harris, J. T., and Starovoitova, V. N. (2015). ⁴⁷Ca production for ⁴⁷Ca/⁴⁷Sc generator system using electron linacs. Appl. Radiat. Isotopes 97, 188–192. doi:10.1016/j.apradiso.2014.12.020

PubMed Abstract | CrossRef Full Text | Google Scholar

Robert, S. D. (1976). Revised effective ionic radii and systematic studies of interatomic distances in halides and chalcogenides. Acta Crystallogr. Sect. A Cryst. Phys. Diffr. Theor. general Crystallogr. 32 (5), 751–767. doi:10.1107/s0567739476001551

CrossRef Full Text | Google Scholar

Roesch, F. (2012). Scandium-44: Benefits of a long-lived PET radionuclide available from the ⁴⁴Ti/⁴⁴Sc generator system. Curr. Radiopharm. 5 (3), 187–201. doi:10.2174/1874471011205030187

PubMed Abstract | CrossRef Full Text | Google Scholar

Rosar, F., Buchholz, H. G., Michels, S., Hoffmann, M. A., Piel, M., Waldmann, C. M., et al. (2020). Image quality analysis of ⁴⁴Sc on two preclinical PET scanners: A comparison to ⁶⁸Ga. EJNMMI Phys. 7 (16), 16. doi:10.1186/s40658-020-0286-3

PubMed Abstract | CrossRef Full Text | Google Scholar

Rotsch, D. A., Brown, M. A., Nolen, J. A., Brossard, T., Henning, W. F., Chemerisov, S. D., et al. (2018). Electron linear accelerator production and purification of scandium-47 from titanium dioxide targets. Appl. Radiat. Isotopes 131, 77–82. doi:10.1016/j.apradiso.2017.11.007

PubMed Abstract | CrossRef Full Text | Google Scholar

Sainz-Esteban, A., Prasad, V., Schuchardt, C., Zachert, C., Carril, J. M., and Baum, R. P. (2012). Comparison of sequential planar ¹⁷⁷Lu-DOTA-TATE dosimetry scans with ⁶⁸Ga-DOTA-TATE PET/CT images in patients with metastasized neuroendocrine tumours undergoing peptide receptor radionuclide therapy. Eur. J. Nucl. Med. Mol. Imaging 39, 501–511. doi:10.1007/s00259-011-2003-x

PubMed Abstract | CrossRef Full Text | Google Scholar

Schumann, D., Horn, S., and Neuhausen, J. (2007). Design of a ⁴⁴Ti/⁴⁴Sc generator system. Annual Report. Villigen, Switzerland: Paul Scherrer Institut & Universitaet Bern, 41.

Google Scholar

Şekerci̇, M., Özdoğan, H., and Kaplan, A. (2023). A study on the cross-section data of 43,44m,46,47Sc isotopes via (d,x) reactions on natural abundance targets under the effects of deuteron optical models. Appl. Radiat. Isotopes.94, 110714, doi:10.1016/j.apradiso.2023.110714

CrossRef Full Text | Google Scholar

Severin, G. W., Engle, J., Valdovinos, H., Barnhart, T., and Nickles, R. (2012). Cyclotron produced ^44gSc from natural calcium. Appl. Radiat. Isotopes 70 (8), 1526–1530. doi:10.1016/j.apradiso.2012.04.030

PubMed Abstract | CrossRef Full Text | Google Scholar

Singh, A., van der Meulen, N. P., Müller, C., Klette, I., Kulkarni, H. R., Türler, A., et al. (2017). First-in-Human PET/CT imaging of metastatic neuroendocrine neoplasms with cyclotron-produced ⁴⁴Sc-dotatoc: A proof-of-concept study. Cancer Biotherapy Radiopharmeceuticals 32 (4), 124–132. doi:10.1089/cbr.2016.2173

PubMed Abstract | CrossRef Full Text | Google Scholar

Sinnes, J.-P., Bauder-Wüst, U., Schäfer, M., Sung Moon, E., et al. (2020). ⁶⁸Ga, ⁴⁴Sc and ¹⁷⁷Lu-labeled AAZTA5-PSMA-617: Synthesis, radiolabeling, stability and cell binding compared to DOTA-PSMA-617 analogues. EJNMMI Radiopharm. Chem. 5 (28), 28. doi:10.1186/s41181-020-00107-8

PubMed Abstract | CrossRef Full Text | Google Scholar

Sitarz, M., Szkliniarz, K., Jastrzębski, J., Choiński, J., Guertin, A., Haddad, F., et al. (2018). Production of Sc medical radioisotopes with proton and deuteron beams. Appl. Radiat. Isotopes 142, 104–112. doi:10.1016/j.apradiso.2018.09.025

PubMed Abstract | CrossRef Full Text | Google Scholar

Snow, M. S., Foley, A., Ward, J. L., Kinlaw, M. T., Stoner, J., and Carney, K. P. (2021). High purity ⁴⁷Sc production using high-energy photons and natural vanadium targets. Appl. Radiat. Isotopes 178, 109934. doi:10.1016/j.apradiso.2021.109934

CrossRef Full Text | Google Scholar

Starovoitova, V. N., Cole, P. L., and Grimm, T. L. (2015). Accelerator-based photoproduction of promising beta-emitters ⁶⁷Cu and ⁴⁷Sc. J. Radioanalytical Nucl. Chem. 305, 127–132. doi:10.1007/s10967-015-4039-z

CrossRef Full Text | Google Scholar

Valdovinos, H. F., Hernandez, R., Barnhart, T., Graves, S., Cai, W., and Nickles, R. (2015). Separation of cyclotron-produced ⁴⁴Sc from a natural calcium target using a dipentyl pentylphosphonate functionalized extraction resin. Appl. Radiat. Isotopes 95, 23–29. doi:10.1016/j.apradiso.2014.09.020

PubMed Abstract | CrossRef Full Text | Google Scholar

van der Meulen, N. P., Bunka, M., Domnanich, K. A., Müller, C., Haller, S., Vermeulen, C., et al. (2015). Cyclotron production of ⁴⁴Sc: From bench to bedside. Nucl. Med. Biol. 42 (9), 745–751. doi:10.1016/j.nucmedbio.2015.05.005

PubMed Abstract | CrossRef Full Text | Google Scholar

Vaughn, B. A., Aluicio-Sarduy, E., Devaraj, J., Olson, A. P., Jonathan, E., Boros, E., et al. (2020a). Chelation with a twist: A bifunctional chelator to enable room temperature radiolabeling and targeted PET imaging with scandium-44. R. Soc. Chem. 11, 333–342. doi:10.1039/C9SC04655K

PubMed Abstract | CrossRef Full Text | Google Scholar

Vaughn, B. A., Koller, A. J., Chen, Z., Ahn, S. H., Loveless, C. S., Cingoranelli, S. J., et al. (2020b). Homologous structural, chemical, and biological behavior of Sc and Lu complexes of the picaga bifunctional chelator: Toward development of matched theranostic pairs for radiopharmaceutical applications. Bioconjugate Chem. 32 (7), 1232–1241. doi:10.1021/acs.bioconjchem.0c00574

PubMed Abstract | CrossRef Full Text | Google Scholar

Weineisen, M., Schottelius, M., Simecek, J., Baum, R. P., Yildiz, A., Beykan, S., et al. (2015). ⁶⁸Ga- and ¹⁷⁷Lu-labeled PSMA I&T: Optimization of a PSMA-targeted theranostic concept and first proof-of-concept human studies. J. Nucl. Med. 56 (8), 1169–1176. doi:10.2967/jnumed.115.158550

PubMed Abstract | CrossRef Full Text | Google Scholar

Wibowo, F. A., and Kambali, I. (2021). Calculated cyclotron-based scandium-47 production yields via ⁴⁸Ca(p, 2n)⁴⁷Sc nuclear reaction. J. Phys. Conf. Ser. 1816 (1), 012099. doi:10.1088/1742-6596/1816/1/012099

CrossRef Full Text | Google Scholar

Yoon, H. J., Jeong, Y. J., Son, H. J., Kang, D. Y., Hyun, K. Y., and Lee, M. K. (2015). Optimization of the spatial resolution for the GE Discovery PET/CT 710 by using NEMA NU 2-2007 standards. J. Korean Phys. Soc. 66, 287–294. doi:10.3938/jkps.66.287

CrossRef Full Text | Google Scholar

Ziegler, J. F., Ziegler, F., and Biersack, J. (2010). Srim – the stopping and range of ions in matter (2010). Res. Sect. B Beam Interact. Mater. Atoms 268, 1818–1823. doi:10.1016/j.nimb.2010.02.091

CrossRef Full Text | Google Scholar