Therapeutic Potential of ⁴⁷Sc in Comparison to ¹⁷⁷Lu and ⁹⁰Y: Preclinical Investigations

Abstract

Targeted radionuclide therapy with ¹⁷⁷Lu- and ⁹⁰Y-labeled radioconjugates is a clinically-established treatment modality for metastasized cancer. ⁴⁷Sc is a therapeutic radionuclide that decays with a half-life of 3.35 days and emits medium-energy β⁻-particles. In this study, ⁴⁷Sc was investigated, in combination with a DOTA-folate conjugate, and compared to the therapeutic properties of ¹⁷⁷Lu-folate and ⁹⁰Y-folate, respectively. In vitro, ⁴⁷Sc-folate demonstrated effective reduction of folate receptor-positive ovarian tumor cell viability similar to ¹⁷⁷Lu-folate, but ⁹⁰Y-folate was more potent at equal activities due to the higher energy of emitted β⁻-particles. Comparable tumor growth inhibition was observed in mice that obtained the same estimated absorbed tumor dose (~21 Gy) when treated with ⁴⁷Sc-folate (12.5 MBq), ¹⁷⁷Lu-folate (10 MBq), and ⁹⁰Y-folate (5 MBq), respectively. The treatment resulted in increased median survival of 39, 43, and 41 days, respectively, as compared to 26 days in untreated controls. There were no statistically significant differences among the therapeutic effects observed in treated groups. Histological assessment revealed no severe side effects two weeks after application of the radiofolates, even at double the activity used for therapy. Based on the decay properties and our results, ⁴⁷Sc is likely to be comparable to ¹⁷⁷Lu when employed for targeted radionuclide therapy. It may, therefore, have potential for clinical translation and be of particular interest in tandem with ⁴⁴Sc or ⁴³Sc as a diagnostic match, enabling the realization of radiotheragnostics in future.

1. Introduction

Presently, the standard treatment of cancer involves surgery, external beam radiation therapy, and chemotherapy. Depending on the tumor type, more specific treatments, including hormonal therapy or immunotherapy, are also viable options [1]. Most attention is currently drawn to the approach of personalized medicine, where the cancer treatment chosen is dependent on patient- and tumor-specific characteristics [2]. Among these is also the concept of tumor targeting to deliver highly toxic agents, specifically to cancer cells. This option requires the design of tumor-specific ligands (tumor targeting agents) which do not—or only at insignificant levels—accumulate in healthy normal tissue. The goal of this concept is to maximize the therapeutic efficacy of the cytotoxic agent and, at the same time, reduce undesired off-target toxicity.

Tumor-targeted radionuclide therapy has emerged as a promising approach for the treatment of disseminated cancer [3]. β⁻-particle-emitting radionuclides are most often used for this purpose. The tissue range of β⁻-particles covers hundreds of cell diameters, hence, β⁻-radiation affects not only cells expressing the target structure, but also neighboring tumor tissue. This phenomenon is known as the “cross-fire effect” and enables the treatment of medium-sized and larger lesions, even if the target is heterogeneously expressed [4]. Currently, ¹⁷⁷Lu (T_1/2 = 6.65 days; Eβ⁻_av = 134 keV) is the most commonly employed β⁻-particle-emitting radiometal for targeted radionuclide therapy using peptide- or small-molecule-based targeting agents (

Table 1. Decay properties ¹ of the therapeutic radionuclides, ⁴⁷Sc, ¹⁷⁷Lu, and ⁹⁰Y.

Radionuclide	Half-Life	Eβ− Average	Eγ (Intensity) 2	Availability
47Sc	3.35 d	162 keV	159 keV (68%)	produced as part of this research project
177Lu	6.65 d	134 keV	113 keV (6%) 208 keV (10%)	commercially available
90Y	2.67 d	934 keV	none	commercially available

¹https://www.nndc.bnl.gov/nudat2/; ² γ-Lines relevant for SPECT imaging.

) [5].

⁹⁰Y is another radiometal applied clinically, which decays with a half-life of 2.67 days by the emission of β⁻-particles of a relatively high energy (Eβ⁻_av = 934 keV), resulting in a maximum tissue range of 10 mm (Table 1). It is, therefore, suitable for the treatment of larger metastases and has been extensively utilized with somatostatin analogs for the treatment of neuroendocrine tumors [6,7]. On the other hand, these properties also contribute to undesired side effects, particularly hematological and kidney toxicities when using ⁹⁰Y-labeled compounds [8,9]. ¹⁷⁷Lu-based therapy outperformed the use of ⁹⁰Y-labeled compounds in terms of minimizing side effects [10]. The medium-energy β⁻-particles emitted by ¹⁷⁷Lu affect non-targeted tissues to a lesser extent due to its shorter tissue range (maximum tissue range: ~1.5 mm) [11]. A further advantage of ¹⁷⁷Lu over ⁹⁰Y is the co-emission of γ-radiation (Table 1), which can be used for the monitoring of tissue distribution and for pre-therapeutic dosimetry using SPECT.

⁴⁷Sc is a novel radionuclide with decay properties comparable to those of ¹⁷⁷Lu, as it emits medium-energy β⁻-particles (Eβ⁻_av = 162 keV) and γ-radiation (Eγ = 159 keV) suitable for SPECT imaging (Table 1) [12,13]. ⁴⁷Sc decays with a half-life of 3.35 days, which may be suitable for small-molecular-weight ligands, considering their rapid pharmacokinetics. The therapeutic potential of ⁴⁷Sc has been shown previously in a pilot preclinical study using ⁴⁷Sc-labeled folate [14]. It can be used towards a theragnostic approach in tandem with ⁴⁴Sc or ⁴³Sc as a diagnostic match. ⁴⁴Sc (T_1/2 = 4.04 h [15]; Eβ⁺_av = 632 keV) and ⁴³Sc (T_1/2 = 3.89 h; Eβ⁺_av = 476 keV) are radionuclides suitable for PET imaging, due to their favorable physical properties, which makes them interesting for clinical translation [16,17,18]. Preliminary data from clinical applications are available with generator- and cyclotron-produced ⁴⁴Sc, where it was used in combination with prostate specific membrane antigen (PSMA) ligands and somatostatin analogues, respectively [19,20,21].

The goal of this study was to investigate the potential of ⁴⁷Sc for targeted radionuclide therapy and to compare its therapeutic properties with those of ¹⁷⁷Lu and ⁹⁰Y. For this purpose, we used a DOTA-functionalized folate conjugate for folate receptor (FR)-targeted radionuclide therapy of ovarian cancer, which is known to express the folate receptor with the highest frequency (~90% of cases) [22]. The folate conjugate was labeled with ⁴⁷Sc, ¹⁷⁷Lu, and ⁹⁰Y, respectively, and the radiometal complexes were investigated extensively in vitro and in vivo. In this study, ovarian IGROV-1 tumor cells were used instead of the commonly-employed cervical KB tumor cells, because they overexpress the FR at levels that reflect the clinical situation more accurately than is the case for KB cells [23]. The therapeutic effects of various activities of ⁴⁷Sc-, ¹⁷⁷Lu-, and ⁹⁰Y-labeled folate conjugates were investigated in vitro. Biodistribution studies were carried out in IGROV-1 tumor-bearing mice using ¹⁷⁷Lu-folate, in order to calculate the mean absorbed dose to the tumors and kidneys of mice when treated with ⁴⁷Sc-, ¹⁷⁷Lu-, and ⁹⁰Y-folate. Subsequently, the therapy experiment was performed using activities which resulted in equipotent tumor doses of ⁴⁷Sc-, ¹⁷⁷Lu-, and ⁹⁰Y-folate.

2. Materials and Methods

2.1. Radionuclides

⁴⁷Sc was produced using the ⁴⁶Ca(n,γ)⁴⁷Ca→⁴⁷Sc nuclear reaction, as previously reported [12,14]. ⁴⁶Ca targets were irradiated in the high thermal neutron flux reactor at the Institut Laue Langevin (ILL; Grenoble, France) or at the BR2 reactor at SCK.CEN (Mol, Belgium) and ⁴⁷Sc separated from the target material at the Paul Scherrer Institute (PSI; Villigen-PSI, Switzerland). In the course of these studies, the production and separation processes of ⁴⁷Sc were optimized with the use of ⁴⁶CaO targets instead of ⁴⁶Ca(NO₃)₂ targets. ⁴⁶CaO targets (⁴⁶Ca, 5% enrichment, Trace Sciences, Wilmington, DE, USA) were irradiated and the separation method was adapted to enable target dissolution and efficient separation of ⁴⁷Sc from the target material and environmental impurities (Supplementary Materials, Figure S1). After separation, ⁴⁷Sc was obtained as ⁴⁷ScCl₃ in HCl (0.05–0.1 M).

No-carrier-added ¹⁷⁷Lu (¹⁷⁷LuCl₃/HCl 0.04 M) was kindly provided by Isotope Technologies Garching (ITG GmbH, Garching, Germany) and ⁹⁰Y (⁹⁰YCl₃/HCl 0.05 M) was obtained from Medeo AG (Schöftland, Switzerland).

2.2. Preparation of ⁴⁷Sc-Folate, ¹⁷⁷Lu-Folate, and ⁹⁰Y-Folate

The DOTA-functionalized folate conjugate (referred to as cm10), was previously developed at the Center for Radiopharmaceutical Sciences at PSI [24]. The radiolabeling of the folate conjugate with ⁴⁷Sc, ¹⁷⁷Lu, and ⁹⁰Y was carried out at pH 4.5 using HCl (0.05 M, pH 1) and Na-acetate (0.5 M, pH 8), as previously reported for ¹⁷⁷Lu-folate [24]. The reaction mixture was incubated for 10–15 min at 95 °C, followed by quality control using HPLC (Supplementary Materials). The folate radioconjugates were obtained at high radiochemical purity (>97%) and used without additional purification. Sodium diethylenetriaminepentaacetate (Na-DTPA) was added to the injection solution for complexation of potential traces of unreacted radionuclide.

2.3. Tumor Cell Culture and Internalization Experiments

IGROV-1 cells (human ovarian carcinoma cell line) were gifted from Dr. Gerrit Jansen (Department of Rheumatology, Free University Medical Center, Amsterdam, The Netherlands). These cells were cultured in folate-deficient RPMI medium (FFRPMI, Cell Culture Technologies GmbH, Gravesano, Switzerland) supplemented with 10% fetal calf serum (FCS), l-glutamine and antibiotics. Routine cell culture was performed twice a week using trypsin-EDTA (0.25%, Gibco) for detachment of the cells.

Tumor cell uptake and internalization studies were performed using FR-positive IGROV-1 cells as previously reported [25]. The radiofolates were prepared at a molar activity of 10 MBq/nmol and added to IGROV-1 tumor cells in the presence and absence of folic acid, followed by incubation of the cells for 4 h. A detailed description of the procedure is given in the Supplementary Materials. The total uptake and the internalized fraction were calculated as a percentage of total added activity. Each experiment was performed in triplicate.

2.4. Cell Viability Assay

⁴⁷Sc-, ¹⁷⁷Lu- and ⁹⁰Y-folate, prepared at a molar activity of 10 MBq/nmol, were evaluated with regard to their effect on cell viability using a 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay as previously reported [25]. IGROV-1 tumor cells were seeded in 96-well plates at 2.5 × 10³ cells per well in a volume of 200 μL FFRPMI medium (with supplements). The well plates were incubated at 37 °C and 5% CO₂ overnight to allow cell adhesion and growth. The following day, supernatants were removed and either only FFRPMI medium (200 μL, without supplements) or medium containing ⁴⁷Sc-, ¹⁷⁷Lu-, or ⁹⁰Y-folate, at variable concentrations (1.0–40 MBq/mL), was added to each well. The supernatants were removed after 4 h incubation at 37 °C and cells were washed with PBS (200 µL). FFRPMI medium with supplements was added to each well. After an incubation period of 6 days, the cell viability was determined using the MTT reagent and a microplate reader (Victor X3, Perkin Elmer, Waltham, MA, USA) to determine the absorbance of the violet color at 560 nm, as previously reported [26]. To quantify cell viability, the percentage of the absorbance of treated cell samples was compared to the absorbance of untreated control cell samples (= 100% viability). Graphs were prepared using GraphPad Prism software (version 7.0) (GraphPad Software, Inc., San Diego, CA, USA).

2.5. Animal Experiments

All applicable international, national, and/or institutional guidelines for the care and use of animals were followed. In particular, all animal experiments were carried out according to the guidelines of Swiss Regulations for Animal Welfare. The preclinical studies have been ethically approved by the Cantonal Committee of Animal Experimentation and permitted by the responsible cantonal authorities (N° 75545, November 2013 and N° 79692, January 2017). Female CD1 nude mice aged 5 weeks were purchased from Charles River Laboratories (Sulzfeld, Germany). All animals were fed with a folate-deficient rodent diet (ssniff Spezialdiäten GmbH, Sulzfeld, Germany), starting one week prior to the injection of the radiofolates.

2.6. Biodistribution Studies

Biodistribution studies were performed in triplicate two weeks after IGROV-1 tumor cell inoculation (5 × 10⁶ tumor cells in 100 µL of PBS), as previously reported for other tumor mouse models [24]. ¹⁷⁷Lu-folate (3 MBq, 1 nmol/mouse) was injected in a volume of 100 μL into a lateral tail vein. The animals were sacrificed at various time points between 1 h and 10 days after administration of the radioconjugate. Selected tissues and organs were collected, weighed, and the activity was determined using a γ-counter (Perkin Elmer, Wallac Wizard 1480). The results were listed as a percentage of the injected radioactivity per gram of tissue mass (% IA/g), using counts of a defined volume of the original injection solution measured at the same time.

2.7. Dosimetric Calculations

Dosimetric calculations were performed for all three radiofolates using the experimentally-determined tissue distribution data of ¹⁷⁷Lu-folate, under the assumption that the radiofolates would distribute equally irrespective of whether ⁴⁷Sc, ¹⁷⁷Lu, or ⁹⁰Y was coordinated. The data of the ¹⁷⁷Lu-folate biodistribution were converted to non-decay-corrected values using the half-lives of ⁴⁷Sc, ¹⁷⁷Lu, and ⁹⁰Y, respectively. The mean specific absorbed dose (Gy/MBq) to the tumor xenografts and kidneys was calculated by multiplication of the time-integrated activity concentration with the emitted electron energy per decay, the absorbed fraction of the emitted electron energy, and a factor of 1.6 × 10⁻⁷ for converting the unit into Gy/MBq [27]. The absorbed fractions for the tumors (40−50 mg) at the time of injection and the kidneys (~125 mg) were assessed by Monte Carlo simulations using PENELOPE-2014 [28]. Uniform activity distribution within the kidneys and the tumors was assumed for the simulations. The average electron energy emitted per decay for ¹⁷⁷Lu was 147 keV, for ⁴⁷Sc 162 keV, and for ⁹⁰Y 934 keV, according to the National Nuclear Data Center, Brookhaven National Laboratory USA (https://www.nndc.bnl.gov/mird/, accessed in 2017). The contributions of the emitted photons to the absorbed doses were neglected [29].

2.8. Tumor Therapy Studies

The mice were subcutaneously inoculated with 5 × 10⁶ IGROV-1 tumor cells (in 100 µL PBS) into the right shoulder region 4 days prior to the injection of the radiofolates. Each group consisted of 6 animals. Control animals were injected intravenously with saline. The mice received 12.5 MBq (1 nmol) ⁴⁷Sc-folate, 10 MBq (1 nmol) ¹⁷⁷Lu-folate, and 5 MBq (1 nmol) ⁹⁰Y-folate, respectively. Tumor volumes and body weights of mice were monitored three times a week throughout the study. Endpoint criteria were defined as weight loss of >15% of the initial body weight (at Day 0), tumor volume >1000 mm³, ulceration or bleeding of the tumor xenograft or abnormal behavior indicating pain or unease of the animal. The tumor volume (V) was determined with the equation [V = 0.5 × (L × W²)], where L is the longest axis and W is the perpendicular axis to L, as previously reported [30]. The results of the tumor growth were presented as the average of absolute tumor volumes of each group. The body weight was presented as a relative body weight (RBW), calculated according to the equation RBW = W_x/W₀; with W_x = body weight at Day x and W₀ = body weight at Day 0. The relative tumor volume (RTV) was calculated according to the equation RTV = V_x/V₀; with V_x = tumor volume at Day x and V₀ = tumor volume at Day 0. The therapeutic efficacy was expressed as the percentage of tumor growth inhibition (%TGI), calculated according to the formula: %TGI = 100 − (RTV_T/RTV_C × 100); where RTV_T is the mean relative tumor volume of treated mice and RTV_C is the mean relative tumor volume of untreated control mice determined at Day 18, when the first control mouse was euthanized. Tumor growth delay indices (TGDI) were calculated according to the formula: TGDI = TGD_T/TGD_C. This was defined as the mean tumor growth delay ratio of treated (TGD_T) and untreated animals (TGD_C) which was required to increase the RTV 2- and 5-fold, as previously reported [25,31]. The data sets were analyzed for significance using a one-way ANOVA with Tukey’s multiple comparison post-test using GraphPad Prism software (version 7). A p-value of <0.05 was considered as statistically significant.

2.9. Determination of Blood Plasma Parameters

Blood was sampled from the retrobulbar vein under isoflurane inhalation anesthesia prior to euthanasia of the mice when they had reached an endpoint criterion. Blood samples were centrifuged (4 °C, 20 min, 1600 rpm) and the supernatant (blood plasma) was analyzed using a Fuji Dri-Chem 4000i analyzer (Polymed Medical Center AG, Glattbrugg, Switzerland) measuring creatinine (CRE), blood urea nitrogen (BUN), alkaline phosphatase (ALP), and total bilirubin (TBIL). Data were analyzed for significance using a one-way ANOVA test (GraphPad Prism software, version 7.0). A p-value of <0.05 was considered as statistically significant.

2.10. Histopathological Investigations

Pathological evaluation of mice without tumors was conducted two weeks after injection of ⁴⁷Sc-folate, ¹⁷⁷Lu-folate, and ⁹⁰Y-folate, respectively, at two different activity levels. A full macroscopic examination was performed in each animal and the kidneys, bone marrow (sternum and femur), and spleen were sampled for histological assessment. The tissues were fixed in 4% neutral-buffered formalin and routinely embedded in paraffin wax. Before paraffin embedding, sternum and femur tissues were decalcified at room temperature in an ethylene- diamine-tetraacetic acid (EDTA)-citrate solution for 2 days. Sections of 3–5 µm thickness were prepared on glass slides and routinely stained with hematoxylin eosin (HE). Histological lesions were semi-quantitatively scored by a veterinary pathologist in a blind manner using a severity grading scheme that ranged from 0 to 5. This indicated no lesions (score 0), minimal (score 1), mild (score 2), moderate (score 3), moderate to severe (score 4) and severe damage (score 5). Renal microscopic changes of the glomerular, tubular, and interstitial compartments were scored separately and the sum of the three values used to obtain a cumulative score representing radiation nephropathy damage, as described previously [32]. Histological evaluation of the spleen was conducted to determine radiation injury in the lymphoid cells of the white pulp using a score of 0 (no lymphoid depletion) to a score of 5 (severe lymphoid depletion), as well as the extra-medullary hematopoiesis (EMH) of the red pulp scoring from a score of 0 (no EMH observed) to a score of 5 (large numbers of EMH precursors observed). The hematopoietic function was also evaluated in the bone marrow by estimating the overall cellularity using a score of 0 (no reduction in cellularity) to a score of 5 (severe reduction in cellularity) and the proportion of the different cellular lineages (including granulocytic precursors, erythroid cells and megakaryocytes).

3. Results

3.1. Radiofolate Preparation and Tumor Cell Internalization Studies

The ⁴⁷Sc was obtained in dilute hydrochloric acid in sufficient quality and quantity for the required radiolabeling and subsequent preclinical study (Supplementary Materials, Figure S1). For the in vitro studies reported herein, the radiolabeling of the folate conjugate with ⁴⁷Sc, ¹⁷⁷Lu, and ⁹⁰Y was performed at molar activities up to 12.5 MBq/nmol, 20 MBq/nmol, and 10 MBq/nmol, respectively, yielding the product with a chemical purity >97% (Supplementary Materials, Figure S2).

Cell uptake and internalization were studied using IGROV-1 tumor cells in culture. The total uptake (surface-exposed FR-bound and internalized fraction) of all radiofolates after 4 h incubation was in the range 17−21% of added activity, while the internalized fraction was 7−12%. The uptake in IGROV-1 tumor cells which were incubated with excess folic acid was <0.1%, indicating the FR-dependent uptake of the radiofolates (Figure 1).

3.2. Cell Viability Assay

The viability of the IGROV-1 tumor cells was reduced after treatment with the folate conjugate labeled with each radionuclide. A concentration of 5 MBq/mL reduced the viability to 80% and 67% in the case of ⁴⁷Sc-folate and ¹⁷⁷Lu-folate, respectively (Figure 2). Treatment of the tumor cells with the same activity concentrations of ⁹⁰Y-folate reduced the viability of the cells to 26%. A radiofolate concentration of 20 MBq/mL reduced the viability of cells to ~35% and ~34% in the case of ⁴⁷Sc-folate and ¹⁷⁷Lu-folate, respectively, and caused almost complete cell killing with <1% viable cells when using ⁹⁰Y-folate.

3.3. Biodistribution and Dosimetry

The mean absorbed dose to the tumor and kidneys was calculated by taking the physical properties of the radionuclide into account, along with the size of the tissue, and the uptake, residence, and clearance of the radioactivity over time. For this purpose, biodistribution studies were performed in IGROV-1 tumor-bearing mice using ¹⁷⁷Lu-folate as a surrogate for all radiofolates, based on the assumption that the tissue distribution profile of all radiofolates would be equal—as previously shown for ⁴⁴Sc/⁴⁷Sc-labeled tumor targeting ligands and their ¹⁷⁷Lu-labeled counterparts [14,33]. The biodistribution data of ¹⁷⁷Lu-folate in IGROV-1 tumor-bearing mice are available in the Supplementary Materials, Tables S1 and S2. The mean tumor absorbed doses of ⁴⁷Sc-folate, ¹⁷⁷Lu-folate, and ⁹⁰Y-folate were calculated as 1.7 Gy/MBq, 2.1 Gy/MBq, and 4.3 Gy/MBq, respectively, and calculations for the mean absorbed kidney doses revealed 1.8 Gy/MBq, 2.3 Gy/MBq, and 4.4 Gy/MBq, respectively. In order to obtain a mean absorbed tumor dose of ~21 Gy, mice had to be treated with 12.5 MBq, 10 MBq, and 5 MBq of ⁴⁷Sc-folate, ¹⁷⁷Lu-folate, and ⁹⁰Y-folate, respectively (Supplementary Materials, Table S3). This quantity of injected activity resulted in a mean absorbed kidney dose of ~22−23 Gy.

3.4. Therapy Experiments

The therapeutic effect of ⁴⁷Sc-folate, ¹⁷⁷Lu-folate, and ⁹⁰Y-folate was assessed in IGROV-1 tumor-bearing mice over a period of eight weeks. The quantity of applied radiofolate was adapted for each radionuclide, in order to obtain comparable mean absorbed doses to tumor xenografts and kidneys.

In all treated groups, the average tumor growth was delayed in comparison to untreated control mice (Figure 3a; Supplementary Materials, Figure S3). The tumor growth inhibition (TGI) in treated mice was all in the same range of 67% to 76%, irrespective of whether the mice received ⁴⁷Sc-folate, ¹⁷⁷Lu-folate, or ⁹⁰Y-folate, respectively (p > 0.05). As an additional measure of the therapeutic effect, tumor growth delay indices (TGDI) were calculated. The TGDI₂, defined as the mean tumor growth delay ratio of treated and untreated mice required to double the relative tumor volume (RTV), was in the range 1.5–1.7. The TGDI calculated for the tumor volumes to increase 5-fold was highest in mice that received ⁹⁰Y-folate (1.6), followed by ¹⁷⁷Lu-folate (1.5), and ⁴⁷Sc-folate (1.3). These values revealed a significantly increased TGDI₂ and TGDI₅ for the treated groups as compared to untreated controls (p < 0.05); however, no statistically significant difference was found among the radiofolate-treated groups (p > 0.05) (

Table 2. Analysis of the therapy experiment.

Radioligand	Activity [MBq]	Tumor D [Gy]	Kidney Dose [Gy]	Euthanasia of the First Mouse	Median Survival [days]	TGI [%]	TGDI2	TGDI5
Saline	-	-		Day 18	26	-	1.0 ± 0.3	1.0 ± 0.2
47Sc-folate	12.5	21.3	22.5	Day 32	39	69 ± 17 1	1.5 ± 0.3 1	1.3 ± 0.2 1
177Lu-folate	10	20.8	23.0	Day 29	43	67 ± 28 1	1.5 ± 0.5 1	1.5 ± 0.3 1
90Y-folate	5	21.5	22.0	Day 36	41	76 ± 15 1	1.7 ± 0.2 1	1.6 ± 0.1 1

¹ values of treated mice were significantly different from values of untreated controls (p < 0.05), but no statistically significant difference was shown among the treated groups (p > 0.05).

).

The survival curves reflected the results of tumor growth, as the endpoint criterion was always due to oversized tumors (Figure 3b). The median survival of mice in each group ranged from 39 to 43 days in treated animals, which was significantly enhanced as compared to 26 days median survival in control mice. The average relative body weight (RBW) was stable in all groups, indicating that the treatment did not have a negative influence on the overall condition of the mice (Figure 3c). There was one case in the ¹⁷⁷Lu-folate-treated group that experienced more than 10% body weight loss, which can be ascribed to the fast tumor growth which already required euthanasia at Day 29 after treatment (Supplementary Materials, Figure S4).

On the day of euthanasia, creatinine (CRE) levels were below the detection limit in all cases. Blood urea nitrogen (BUN), alkaline phosphatase (ALP), and total bilirubin (TBIL) levels were comparable among treated mice and untreated controls, irrespective of the radiofolate that was applied (Supplementary Materials, Table S4).

3.5. Histopathological Assessment

The assessment of the kidneys, bone marrow, and spleen was performed two weeks after injection of the radiofolates, applied at the same activities as in the therapy study, as well as at the doubled quantity of the respective radiofolate (Supplementary Materials, Tables S5 and S6). The histological examination of the renal tissue did not show evidence of radiation nephropathy in any of the groups treated with low activity. This was also the case for the majority of animals treated with high activity (Supplementary Materials, Figure S5). Only two animals treated with 25 MBq ⁴⁷Sc-folate exhibited occasional glomeruli with mild reduction of the mesangial capillaries; however, no other alterations in the tubular and interstitial compartments were detected, and the overall renal cortex did not show signs of parenchymal shrinkage.

Control animals and mice injected with lower activity showed adequate hematopoietic cellularity in the bone marrow, with normal proportions of erythroid cells, granulocytic precursors, and megakaryocytes (Figure 4a). The hematopoietic cellularity was also unremarkable in all the animals treated with high activity (Figure 4b,c), apart from two animals that received higher activity of ⁹⁰Y-folate (10 MBq), which showed mild reduction of the bone marrow cellularity accompanied by adipocyte replacement (Figure 4d). Hematopoietic cell activity was also evaluated in the red pulp of the spleen by determining the presence of extra-medullary hematopoiesis (EMH). The overall cellularity of EMH varied between animals and groups; however, mice treated with 10 MBq ⁹⁰Y-folate tended to have lower values compared to the other groups. On the other hand, none of the animals showed evidence of lymphocyte damage in the white pulp of the spleen (data not shown).

4. Discussion

⁴⁷Sc has been previously proposed as a potential therapeutic match to ⁴⁴Sc, a PET nuclide with great potential for clinical translation [14,19,20,33]. Preclinical studies to investigate ⁴⁷Sc for therapeutic application have been scarce [14], however, and a clinical application does not yet exist due to the challenging production routes of this novel radionuclide [12,13,34,35,36]. In the present study, the aim was to investigate ⁴⁷Sc in a preclinical setting and compare its effect with ¹⁷⁷Lu and ⁹⁰Y, two established therapeutic nuclides which have been extensively employed in clinics over the last two decades.

The production of ⁴⁷Sc was performed via the ⁴⁶Ca(n,γ)⁴⁷Ca→⁴⁷Sc nuclear reaction as previously reported [12]; however, the ⁴⁶CaO used in this work provides better chemical stability during irradiation, which allows the irradiation of higher Ca-target masses. With the transfer into another chemical form, no additional impurities were introduced, hence, identical post-irradiation γ-spectra were obtained, as previously reported [12].

As expected, similar results were found in the uptake and internalization of ⁴⁷Sc-folate, ¹⁷⁷Lu-folate, and ⁹⁰Y-folates in ovarian IGROV-1 cancer cells. The data were in line with previously-reported results, where ⁴⁷Sc-folate was investigated with KB tumor cells [14]. Cell viability experiments, performed with increasing quantities of ⁴⁷Sc-folate, ¹⁷⁷Lu-folate, and ⁹⁰Y-folate, resulted in different effects, dependent on the decay characteristics of the radionuclide employed. The most prominent decrease in viability was found when using ⁹⁰Y-folate, which was in line with the fact that ⁹⁰Y emits β⁻-particles of the highest energy. ⁴⁷Sc-folate and ¹⁷⁷Lu-folate showed similar effects on IGROV-1 tumor cells.

Dose calculations revealed that to obtain the same mean absorbed dose to the tumor and kidneys, slightly higher quantities of ⁴⁷Sc-folate than of ¹⁷⁷Lu-folate were necessary due to the different half-lives of ⁴⁷Sc and ¹⁷⁷Lu. The high β⁻-energy of ⁹⁰Y was responsible for an increased dose deposition, which required only half of the activity of ⁹⁰Y-folate to obtain the same mean absorbed dose as for ⁴⁷Sc-folate and ¹⁷⁷Lu-folate. Treatment of IGROV-1 tumor-bearing mice with these adapted activities had similar effects irrespective of whether ⁴⁷Sc-folate, ¹⁷⁷Lu-folate, or ⁹⁰Y-folate was used. Consistent with these findings, the median survival time of treated mice was comparable in each group. These results were also comparable to those obtained previously in KB tumor-bearing mice, where 10 MBq ⁴⁷Sc-folate resulted in an equal median survival time of 38.5 days [14] as for the IGROV-1 tumor mouse model when treated with 12.5 MBq ⁴⁷Sc-folate (median survival: 39 days).

Due to the higher energy of ⁹⁰Y and, as a consequence, an increased tissue range of the emitted β⁻-particles in comparison to the β⁻-particles emitted by ¹⁷⁷Lu, ⁹⁰Y-folate may have the disadvantage of more side effects to kidneys as observed with somatostatin analogues (⁹⁰Y-DOTATOC/⁹⁰Y-DOTATATE) in clinical studies [8,9,37]. Based on the more similar mean β⁻-energies of ⁴⁷Sc and ¹⁷⁷Lu, kidney damage would most likely be observed at the same dose levels for ⁴⁷Sc- and ¹⁷⁷Lu-labeled tumor targeting agents. In this study, only sporadic and very mild glomerular changes were observed, and none of the mice experienced severe renal side effects that would have been detectable by histopathological investigations. It has to be critically acknowledged, however, that radionephrotoxicity is not commonly observed at short time points after injection but rather, several months after radionuclide therapy [38]. It is, therefore, very likely that the higher activity levels of the applied radiofolates, resulting in a kidney dose of 42–44 Gy, would manifest as signs of radiotoxicity six to eight months after treatment.

Severe side effects were not observed in other tissues and organs. Only mice treated with 10 MBq ⁹⁰Y-folate showed mildly reduced extra-medullary hematopoiesis in the spleen, and two animals also exhibited reduced cellularity in the bone marrow, which could be indicative of irradiation-related damage. The amount of extra-medullary hematopoiesis varied between animals and groups, which means that these changes could represent a transient condition rather than a direct radiotoxic event. Since later time points were not investigated in this study, it remains unknown whether hematopoietic changes would recover or, instead, progress to a more severe stage.

As expected, the application of ⁴⁷Sc-folate revealed similar anti-tumor effects as that for treatment using its ¹⁷⁷Lu- and ⁹⁰Y-labeled counterparts when the applied activity was adjusted to obtain the same mean absorbed tumor dose. In this mouse model, differences among the single therapy groups were not determined, indicating that ⁴⁷Sc may be used under the same medical circumstances as ¹⁷⁷Lu and ⁹⁰Y, respectively. The relatively short half-life of ⁴⁷Sc makes its use particularly attractive in combination with fast-clearing tumor targeting ligands, including peptides and small molecules.

It has to be critically acknowledged that a potential advantage of ⁴⁷Sc over ¹⁷⁷Lu remains questionable, particularly due to the higher intensity of γ-emission (⁴⁷Sc: Eγ = 159 keV; I = 68%) which may unnecessarily contribute to the patient dose. On the other hand, this characteristic may be advantageous in cases where an exact dosimetry is required, as recently claimed for ⁶⁷Cu (oral communication), which has very similar decay characteristics to ⁴⁷Sc with regard to half-life, β⁻-particle emission, and intensity of γ-ray emission. Should it be the case, ⁴⁷Sc would probably be preferred over ⁶⁷Cu, due to its easier coordination chemistry and, hence, the opportunity of using it with any tumor-targeting agent that comprises a DOTA chelator.

5. Conclusions

The in vitro and in vivo evaluation of ⁴⁷Sc indicated comparable efficacy in tumor therapy to that of ¹⁷⁷Lu and ⁹⁰Y, when applied at activities that result in the same mean absorbed tumor doses. These findings confirm the accuracy of the dose estimation to tumors, which was performed for ⁴⁷Sc-folate, ¹⁷⁷Lu-folate, and ⁹⁰Y-folate, respectively. Overall, the characteristics of ⁴⁷Sc were comparable to those of ¹⁷⁷Lu and, hence, these two radionuclides may be interchangeable. It has to be critically acknowledged, however, that the production of ⁴⁷Sc is costly and, hence, the radionuclide is not currently available in large quantities that would enable clinical translation.

⁴⁷Sc could be advantageous in combination with molecules of fast pharmacokinetics, where a shorter half-life would be desirable, and in situations where an intense γ-emission was of interest. Certainly, ⁴⁷Sc could play a pivotal role for radiotheragnostic application in tandem with ⁴³Sc/⁴⁴Sc, enabling the preparation of chemically-identical radiopharmaceuticals for PET imaging and therapeutic applications. This will be feasible as soon as the production of the Sc radioisotope family has been fully established to make them readily available in large quantities, at high quality, and at a reasonable cost.