Sadler et al. EJNMMI Radiopharmacy EJNMMI Radiopharmacy and Chemistry (2022) 7:21 https://doi.org/10.1186/s41181-022-00173-0 and Chemistry REVIEW Open Access Cutting edge rare earth radiometals: prospects for cancer theranostics Alexander W. E. Sadler1, Leena Hogan2, Benjamin Fraser2 and Louis M. Rendina1* *Correspondence: louis.rendina@sydney.edu.au Abstract 1 School of Chemistry, The Background: With recent advances in novel approaches to cancer therapy and University of Sydney, Sydney, imaging, the application of theranostic techniques in personalised medicine has NSW 2006, Australia 2 emerged as a very promising avenue of research inquiry in recent years. Interest has ANSTO Life Sciences, Australian Nuclear Science and Technology been directed towards the theranostic potential of Rare Earth radiometals due to their Organisation (ANSTO), Kirrawee, closely related chemical properties which allow for their facile and interchangeable NSW 2232, Australia incorporation into identical bifunctional chelators or targeting biomolecules for use in a diverse range of cancer imaging and therapeutic applications without additional modification, i.e. a “one-size-fits-all” approach. This review will focus on recent progress and innovations in the area of Rare Earth radionuclides for theranostic applications by providing a detailed snapshot of their current state of production by means of nuclear reactions, subsequent promising theranostic capabilities in the clinic, as well as a discussion of factors that have impacted upon their progress through the theranostic drug development pipeline. Main body: In light of this interest, a great deal of research has also been focussed towards certain under-utilised Rare Earth radionuclides with diverse and favourable decay characteristics which span the broad spectrum of most cancer imaging and therapeutic applications, with potential nuclides suitable for α-therapy (149Tb), β−- therapy (47Sc, 161Tb, 166Ho, 153Sm, 169Er, 149Pm, 143Pr, 170Tm), Auger electron (AE) therapy (161Tb, 135La, 165Er), positron emission tomography (43Sc, 44Sc, 149Tb, 152Tb, 132La, 133La), and single photon emission computed tomography (47Sc, 155Tb, 152Tb, 161Tb, 166Ho, 153Sm, 149Pm, 170Tm). For a number of the aforementioned radionuclides, their progres- sion from ‘bench to bedside’ has been hamstrung by lack of availability due to produc- tion and purification methods requiring further optimisation. Conclusions: In order to exploit the potential of these radionuclides, reliable and eco- nomical production and purification methods that provide the desired radionuclides in high yield and purity are required. With more reactors around the world being decom- missioned in future, solutions to radionuclide production issues will likely be found in a greater focus on linear accelerator and cyclotron infrastructure and production meth- ods, as well as mass separation methods. Recent progress towards the optimisation of these and other radionuclide production and purification methods has increased the feasibility of utilising Rare Earth radiometals in both preclinical and clinical settings, thereby placing them at the forefront of radiometals research for cancer theranostics. Keywords: Rare earth, Lanthanoid, Theranostics, Cancer, Imaging, Therapy, Radionuclide production © The Author(s) 2022. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the mate- rial. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licens es/ by/4. 0/. Sadler et al. EJNMMI Radiopharmacy and Chemistry (2022) 7:21 Page 2 of 55 Introduction With advances in approaches to cancer therapy and imaging, the application of thera- nostic techniques for more “personalised” approaches to patient treatment has emerged as a very promising avenue of research inquiry (Marin et al. 2020; DeNardo and DeNa- rdo 2012). The theranostic protocol in nuclear medicine typically involves either the selection and administration of a radionuclide pair of which one is for imaging (e.g. 68Ga or 18F for PET imaging) and the other for therapy (e.g. 177Lu for β−-therapy) (Filippi et al. 2020), ideally of the same element to minimise pharmacokinetic/biological behaviour differences in vivo (Türler 2019); or the utilisation of a single radionuclide that exhibits favourable decay characteristics that permit dual functionality as both an imaging and therapeutic agent (Türler 2019). As theranostic models of cancer treatment gain momen- tum both in the research laboratory and in clinical application, the need for a reliable, economical and high purity/yield source of appropriate radionuclides is required. Con- sequently, it is often the economic/production/availability factors that bolster the popu- larity of certain radionuclides in the minds of researchers and sustains inquiry into their clinical usefulness, rather than their specific decay characteristics that determine their clinical suitability. However, as clinical demand for certain isotopic decay characteristics increases, the focus shifts to improving the production process and economic utility of certain radionuclides that show significant promise for clinical application. In recent years, interest has been focussed around the theranostic potential of Rare Earth radionuclides. Of the lanthanoids, they exhibit similar chemical properties due to the filling of the 4f electron orbital across the period leaving the 6s electrons exposed (Elliott 2013; Cotton 2006). These 4f electrons are largely shielded from chemical inter- action due to the filled 5s and 5p orbitals, and this gives rise to similar properties includ- ing similar stable oxidation states (generally +3), hard Lewis acid behaviour, affinity for hard donor sites for complexation (O and N) and largely electrostatic bonding (Elliott 2013; Cotton 2006; Moeller et al. 1965; Peacock 2016). These similar chemical properties make them of interest to researchers due to the potential for different lanthanoid radio- nuclides to be easily and reliably incorporated into the same (or similar) bifunctional chelators or targeting biomolecules for cancer treatment. This significantly simplifies the syntheses of such compounds by employing a “one-size-fits-all” approach through allow- ing the same ligand scaffold to coordinate different radiolanthanoids in an interchange- able manner depending on the intended imaging or therapeutic purposes (Cotton 2006; Moeller et al. 1965; Cutler et al. 2000). The gold-standard of Rare Earth metal complexa- tion currently is 1,4,7,10-tetraazacyclododecane-1,4,7,10-tetraacetic acid (DOTA) which is easily incorporated into targeting molecule structures and forms very stable com- plexes with Rare Earth metals as shown by its use in radiotherapeutics such as Lutetium- 177-DOTA-Octreotate ([177Lu]Lu-DOTA-TATE) or Lutetium-177-DOTA-Octreotide ([177Lu]Lu-DOTA-TOC) as well as for chelation of Gd for MRI contrast agents (Cutler et al. 2000; Graf et al. 2020; Staanum et al. 2021; Strosberg et al. 2017; Clough et al. 2019; Yang et al. 2008; Yokoyama and Shiraishi 2020) (see Figs. 1 and 2). Due to the fact that complementary diagnostic and therapeutic radionuclides are frequently incorporated into different targeting molecules/vectors, they can exhibit different biodistribution and pharmacokinetics in the body. Similar chemical characteristics that allow for stable complexation without structural variation to the chelator or targeting molecule have the S adler et al. EJNMMI Radiopharmacy and Chemistry (2022) 7:21 Page 3 of 55 O O O O O O O O N N N N O N Gd O Gd O O O OH N N N N N N OH O Gd O O O O O O O O O O HO Fig. 1 Gadolinium contrast agents (left to right) Gd-DOTA, Gadolinium(III)-tris(carboxymethyl)-1,4,7,10-tet raazacyclododecane-butriol (Gd-BT-DO3A) and Gadolinium(III)-diethylenetriaminepentaacetic acid (Gd-DTPA) O O O O O O N N N N O 177 OLu 177Lu O O O O HO N N HO N N O H N O O HN H N N O H O N OH O N O H S S NH O NHS H S O N OH H O N OH NH H NH N O N N H H N O N N H O H HO H O HO O OH O OH H2N H2N [177Lu]Lu-DOTA-TATE [177Lu]Lu-DOTA-TOC Fig. 2 Molecular structures of [177Lu]Lu-DOTA-TATE and [177Lu]Lu-DOTA-TOC potential to address (at least in part) issues surrounding differences in biodistribution between imaging radionuclides and therapeutic radionuclides after administration. This is imperative for effective theranostic treatment regimens because the imaging radio- nuclide and the therapy radionuclide must ideally exhibit near identical biodistribution, selective tumour uptake and bioelimination tendencies in order for tandem theranostic imaging and therapy to work in a complementary manner. Identical bifunctional chela- tors/targeting biomolecules for complexation of a range of radionuclides (each of which display similar chemical properties in of themselves) significantly reduce the variability in these factors after administration. In addition to advantages in similar chemical properties, synthesis and clinical admin- istration, certain Rare Earth radionuclides have garnered attention due to their radio- active decay characteristics showing significant potential for theranostic applications (Cutler et al. 2000; Mishiro et al. 2019). These applications tend to fit into three catego- ries: complementary radionuclides for already established imaging or therapy agents; novel radionuclides that have been hamstrung by lack of availability or reliable produc- tion facilities/methods to date, or as potential improvements over existing therapies for particularly intractable cases of cancer malignancy. While aspects regarding developments in radiolanthanoid chelation have been dis- cussed in conjunction with production, radiochemistry and application of popular radi- olanthanoids currently progressing towards or already in clinical use (Cutler et al. 2000; Mishiro et al. 2019; Amoroso et al. 2017; Kostelnik and Orvig 2019; Notni and Wester 2018), the state of production and potential application of certain cutting edge radiolan- thanoids has not been reviewed in depth. This review with focus on recent progress and Sadler et al. EJNMMI Radiopharmacy and Chemistry (2022) 7:21 Page 4 of 55 innovations in the area of underutilised lanthanoid radionuclides of interest for thera- nostic applications by providing a snapshot of the state of their production via nuclear reactions, subsequent promising indications of their theranostic capabilities in the clinic, as well as factors that have impacted upon their progress through the development pipe- line (see Table 1). Notably, Y, Gd and Lu have been discussed at length in other reviews with regards to their production and implementation in theranostic settings, and as such will not be detailed  in this review. The reader is encouraged to consult the work of Kostelnik and Orvig (Kostelnik and Orvig 2019), as well as Robertson and Rendina (Robertson and Rendina 2021) for further information on these Rare Earth radiometals. Scandium: 43Sc, 44Sc, 47Sc In recent years, focus on the theranostic concept of chemically identical and matched radionuclide pairs has increased due to the desire to minimise differences in the chemi- cal behaviour between compounds administered to patients, while also simplifying the synthesis of such compounds (Türler 2019; Notni and Wester 2018). Having two radio- nuclides of the same element incorporated into the same bifunctional chelator molecule (one for diagnostics and one for therapy) ensures that differences in kinetic properties and in vivo biological behaviour are minimal due to identical chemical behaviour (Türler 2019). This allows for the evaluation of therapeutic efficacy using in vivo imaging with the reasonable assumption that biological and chemical behaviour in the body will be more or less identical (Türler 2019; Notni and Wester 2018). Scandium has been prof- fered as one such element that has suitable radionuclides for diagnostic imaging and therapy, with three theranostically relevant radionuclides that have warranted study for their clinical application: 43Sc, 44Sc and 47Sc (Vaughn et al. 2021). 43Sc and 44Sc are both suitable for Positron Emission Tomography (PET) imaging due to their respective posi- tron emissions, while 47Sc has potential as a β−-emitter for therapeutic applications that can also be observed using Single Photon Emission Computed Tomography (SPECT) imaging due to its γ-ray emission characteristics (Türler 2019; Snow et  al. 2021). PET imaging with 44Sc (t1/2 = 4.04 h) or 43Sc (t1/2 = 3.89 h) has been proposed as potentially advantageous over the current use of 68Ga due to both radionuclides exhibiting half-lives almost 4 times longer than Ga, which would enable longer acquisition times for PET imaging, better image quality while also enabling the production and transportation of Sc radionuclides over longer distances to more remote facilities requiring radionuclides for PET imaging (Kostelnik and Orvig 2019; Mikolajczak et  al. 2021). These radionu- clides are also good diagnostic matches for other lanthanoids already in use such as 177Lu and 90Y due to similarities in their binding and chemical properties (Kostelnik and Orvig 2019). In addition, their chemical nature allows them to be easily chelated by DOTA and DOTA-like analogues and form stable complexes with decreased propensity for demet- allation in the body, while also maintaining consistency between the bifunctional che- lators/targeting molecule structures used in subsequent therapeutic steps (Kostelnik and Orvig 2019). 47Sc has advantageous simultaneous imaging capabilities (SPECT) in conjunction with its β−-therapy applications, which would allow for its individual use in certain circumstances, but also allows its ideal pairing with diagnostically relevant Sc S adler et al. EJNMMI Radiopharmacy and Chemistry (2022) 7:21 Page 5 of 55 Table 1 Decay characteristics and applications for certain Rare Earth radionuclides Radionuclide Half- Decay Average Maximum Nuclear reaction(s) for Applications life (h) characteristics decay γ energy radionuclide production to nuclear particle (keV) medicine energy (intensity (keV)a %)a 43Sc 3.89 β+ (88%) 419 373 (22%) 40Ca(α,n)43Ti → 43Sc PET EC (12%) 43Ca(p,n)43Sc 46Ti(p,α)43Sc 44Sc 4.04 β+ (94%) 632 1157 (100%) 45Sc(p,2n)44Ti → 44Sc PET EC (6%) 44Ca(p,n)44Sc 44Ca(d,2n)44Sc 44Ca(α,3np)44Sc 47Sc 80.4 β− (100%) 162 159 (68%) 48Ca(p,2n)47Sc β−-therapy, 44Ca(α,p)47Sc SPECT 46Ca(n,γ)47Ca → 47Sc 47Ti(n,p)47Sc 48Ti(γ,p)47Sc 51 V(γ,α)47Sc 48Ca(γ,n)47Ca → 47Sc 149Tb 4.1 α (17%) 3970 (α) 165 (26%) Ta-foil spallation α-therapy, PET β+ (7%) 728 (β+) 352 (29%) 152Gd(p,4n)149Tb EC (76%) 389 (18%) 141Pr(12C,4n)149Tb 817 (12%) 142Nd(12C,5n)149Tb 853 (16%) 152Tb 17.5 β+ (20%) 1142 344 (64%) Ta-foil spallation PET EC (80%) 152Gd(p,n)152Tb 155Tb 127.7 EC (100%) 87 (32%) Ta-foil spallation SPECT 105 (25%) 155Gd(p,n)155Tb 159Tb(p,5n)155Dy → 155Tb 161Tb 166.8 β− (100%) 154 (β−) 26 (23%) 160Gd(n,γ)161Gd → 161Tb β−-therapy, 12.4 e − per 3.75 per 49 (17%) Auger electron decay AE 75 (10%) therapy, SPECT 132La 4.6 β+ (42%) 1290 132Ba(p,n)132La PET EC (58%) 134Ba(p,3n)132La 133La 3.9 β+ (7%) 461 278 (2%) 134Ba(p,2n)133La PET EC (93%) 290 (1%) 135Ba(p,3n)133La 302 (2%) 135La 18.9 EC (> 99%) < 4 per AE 481 (2%) 135Ba(p,n)135La Auger electron 10.6 e− per 136Ba(p,2n)135La therapy decay 137Ba(p,3n)135La 166Ho 26.8 β− (100%) 665 80.6 (6%) 165Ho(n,γ)166Ho β−-therapy, 1379 (1%) 164Dy(n,γ)165Dy(n,γ)16 SPECT 6Dy → 166Ho 153Sm 46.3 β− (100%) 224 103 (29%) 152Sm(n,γ)153Sm β−-therapy, SPECT 165Er 10.4 EC (100%) 5.3 166Er(p,2n)165Tm → 165Er Auger electron (65.6%) 165Ho(p,n)165Er therapy 38.4 165Ho(d,2n)165Er (4.8%) 169Er 225.6 β− (100%) 100 168Er(n,γ)169Er β−-therapy 149Pm 53.0 β− (100%) 363 286 (3%) 148Nd(n,γ)149Nd → 149Pm β−-therapy 148Nd(p,γ)149Pm 150Nd(p,2n)149Nd → 149Pm 148Nd(d,n)149Pm 150Nd(d,3n)149Pm 150Nd(γ,n)149Nd → 149Pm 143Pr 326.4 β− (100%) 315 141Pr(n,γ)142Pr(n,γ)143Pr β−-therapy 142Ce(n,γ)143Ce → 143Pr 170Tm 3086.4 β− (100%) 317 84 (3%) 169Tm(n,γ)170Tm β−-therapy, SPECT Sadler et al. EJNMMI Radiopharmacy and Chemistry (2022) 7:21 Page 6 of 55 Table 1 (continued) a Values calculated from the IAEA Live Chart of Nuclides, Nuclear Structure and Nuclear Decay Data; https:// www- nds.i aea. org/r elnsd/v char thtml/ VChart HTML. html (accessed 16 May 2022) radionuclides in fulfilment of the chemically identical theranostic concept (Müller et al. 2014a). Bulwarks to the implementation of Sc radionuclides in matched-pair theranostic applications have arisen in the form of production on the scale and radionuclidic/radio- chemical purity required for their seamless translation into the clinic. The production methods of each radionuclide and their limitations are discussed in the context of their theranostic potential. 44Sc 44Sc has proceeded the most with regards to adequate production and purity, with a number of production methods being utilised with varying degrees of success. Initial forays into its production were by means of a 44Ti/44Sc generator initially utilising the reliable but indirect 45Sc(p,2n)44Ti → 44Sc nuclear reaction with high proton flux due to long-lived 44Ti (Filosofov et  al. 2010; Hassan et  al. 2018). The first of these generators culminated in the elution of ~ 180  MBq of high purity 44Sc in 20  mL of eluate, with a separation factor of 2 × 106 (Filosofov et al. 2010). This method of production also incor- porated a post-process washing step to reduce the volume of the eluate and remove contaminants such as oxalate anions (Filosofov et  al. 2010; Rösch and Baum 2011; Pruszyński et al. 2010). The 44Sc obtained from this process was used to label DOTA- TOC and was subsequently used for the first time in patients (Roesch and Scandium-44, 2012). More recently, generator produced 44Sc was used for the first in-human PET imaging of patients with metastasised castration resistant prostate cancer via radiola- belling a prostate-specific membrane antigen (PSMA) targeting ligand (Vipivotide tet- raxetan—PSMA-617) (Eppard et al. 2017). Despite the excellent separation of 44Sc from the 44Ti target, the lack of by-production of 44mSc impurities and the ready use of the resulting radionuclide in clinical studies (Filosofov et al. 2010; Roesch and Scandium-44, 2012; Eppard et  al. 2017), comparative studies of indirect vs. direct production meth- ods for 44Sc have noted that the post-elution “reverse” purification processes required to limit 44Ti breakthrough are challenging and have consequently limited the applicability of this production method outside situations where long 44Sc transportation times will be incurred (Hassan et al. 2018). Higher radionuclidic purity is also reported when more direct production methods are used, typically employing cyclotrons (Hassan et al. 2018). Alternative methods of production have also been achieved using cyclotrons and both natural and enriched Ca targets (Severin et al. 2012; Müller et al. 2013). The natural Ca target route typically yields of up to ~ 650 MBq of 44Sc after target irradiation of 1 h, but multiple impurities were noted after the final product was isolated, namely 44mSc (the production of which is avoided when utilising the indirect 44Ti generator approach), 47Sc and 48Sc (Severin et al. 2012). The impurities from this production method pose issues for translation into clinical settings due to their long half-lives (44mSc: t = 58.6 h, 471/2 Sc: t1/2 = 80.4 h, 48Sc: t 1/2 = 43.7 h). To avoid such impurities, enriched 44Ca targets have typ- ically been used (Müller et al. 2013; Krajewski et al. 2013). These routes use proton, deu- teron or α-particle irradiation and yield 44Sc in adequate yield and radionuclidic purity S adler et al. EJNMMI Radiopharmacy and Chemistry (2022) 7:21 Page 7 of 55 via 44Ca(p,n)44Sc, 44Ca(d,2n)44Sc or 44Ca(α,3np)44Sc nuclear reactions (Hassan et  al. 2018). Respectively, activities of 1900 MBq/50 μA using enriched [44Ca]CaCO3 powder and 11 MeV protons for 60–90 min (Meulen et al. 2015), and 44 MBq/0.2 μA.h using the same target irradiated with 16.4  MeV deuterons and a  60 min irradiation time (Alliot et al. 2015) have been reported as the most promising for high radionuclidic purity and yield of 44Sc, favoured over using other enriched Ca targets such as 40Ca, 42Ca and 43Ca. As a result, production and separation processes (based on extraction chromatogra- phy) for such nuclear reactions have been finely tuned to produce high radionuclidic purity 44Sc in quantities of up to ~ 2 GBq using optimised proton beams with energies of approximately 11  MeV, and progress in this area of cyclotron-produced 44Sc has increased its attractiveness for preclinical and clinical studies due to greater radionuclide availability (Alliot et al. 2015). 43Sc Interest in 43Sc as an alternative to 44Sc has been invigorated due to the absence of co- emitting high-energy γ-rays compared to 44Sc, which pose clinical issues regarding patient radiation dose burdens (Domnanich et al. 2017a; Carzaniga et al. 2017). Similar to the desired production route for 44Sc, 43Sc has been produced primarily using natu- ral and enriched calcium targets in a cyclotron and irradiation by α-particles (Walczak et al. 2015; Szkliniarz et al. 2015, 2016; Minegishi et al. 2016). Typical nuclear reactions employed have been direct: 40Ca(α,p)43Sc; and indirect: 40Ca(α,n)43Ti → 43Sc. High radionuclidic purities (> 99%) and low impurities (< 1.5 × 10–5%) have been consistently reported when using enriched 40Ca targets and irradiation periods of 1–2 h (Walczak et  al. 2015; Szkliniarz et  al. 2015, 2016; Minegishi et  al. 2016). Irradiation of natural calcium targets with α-particles produced 43Sc with activities of ~ 100 MBq/µAh, with the capability of producing radionuclidically pure 43Sc at activities on the order of sev- eral GBq (Walczak et  al. 2015). Optimised separation using UTEVA resin resulted in chemically pure 43Sc suitable for PET applications (Walczak et al. 2015). However, the α-particle energies required for these production routes range between ~ 24–28  MeV, and the lack of availability of cyclotrons with sufficient high-current α-particle beams for such applications has been an impediment to their widespread use (Walczak et al. 2015; Szkliniarz et  al. 2016). Another proposed route involves employing enriched 42Ca tar- gets and deuteron irradiation using the 42Ca(d,n)43Sc nuclear reaction, however a similar availability issue is encountered regarding sufficient deuteron beamlines (Walczak et al. 2015; Szkliniarz et al. 2016). Comparison of production routes using enriched 43Ca and 46Ti irradiated with pro- tons have also been investigated, using the 43Ca(p,n)43Sc and 46Ti(p,α)43Sc nuclear reac- tion respectively (Domnanich et al. 2017a; Carzaniga et al. 2017). Both routes make use of moderate proton beamline energies ~ 10–15 MeV which are readily available at most commercially operable biomedical cyclotrons (Domnanich et al. 2017a; Carzaniga et al. 2017). Irradiation of enriched 46Ti targets has resulted in activity yields of ~ 60–225 MBq after ~ 7 h proton bombardment, levels which are reported to be readily attainable using the above proton beamline energies, and with radionuclidic purity of > 98% 43Sc (only 1.5% 44Sc and < 0.08% all other Sc isotopes) (Domnanich et al. 2017a). On the other hand, the enriched 43Ca target route produced the desired 43Sc at higher activity yields than Sadler et al. EJNMMI Radiopharmacy and Chemistry (2022) 7:21 Page 8 of 55 the 46Ti target, but with the trade-off that radionuclidic purity was lower and depend- ent on the enrichment level of the target material (Carzaniga et  al. 2017). Using 58% enriched 43Ca targets, irradiation caused the simultaneous production of both 43Sc and 44Sc, resulting in a ratio of ~ 67% 43Sc and ~ 33% 44Sc with a 43Sc activity yield of ~ 250– 480 MBq/µAh, while another investigation reported final activity yields of ~ 130 MBq, with near identical 43Sc/44Sc ratios in the final product, and noted that using lower impinging proton beam energies allowed for an increase the final purity at the cost of overall yield (Carzaniga et  al. 2017). Higher yields reported when using enriched cal- cium targets have been partially attributed to the higher nuclear reaction cross-section of 43Ca compared to 46Ti (~ 250–275 mb and ~ 40 mb respectively) (Domnanich et  al. 2017a; Carzaniga et al. 2017). The weighing up of the relevant advantages and drawbacks of each of these production routes has shown that the 43Ca route has greater applicabil- ity for routine production of larger yields of 43Sc for radiopharmaceutical applications, since target material recycling and comparatively less complicated target preparation and separation procedures are able to mitigate the cost of enriched 43Ca (Domnanich et al. 2017a; Carzaniga et al. 2017). 47Sc A number of nuclear reactions using a variety of facilities have been investigated for the production of 47Sc. Cyclotrons, nuclear  reactors and linear accelerators have been used in this regard, utilising protons, neutrons and α-particles for target material bom- bardment (Minegishi et  al. 2016; Misiak et  al. 2017; Domnanich et  al. 2017b; Kolsky et  al. 1998; Bartoś et  al. 2012). Cyclotron production of 47Sc has been studied via the 48Ca(p,2n)47Sc nuclear reaction with an optimum proton energy range of 24–17  MeV and an irradiation time of 5 h using a natural C aCO3 target (Misiak et al. 2017). Separa- tion of 47Sc from the target was achieved using Chelex-100 and UTEVA resins as well as filtration using a 0.2 µm filter. Optimum recovery of Sc isotopes (96%) was achieved using a 0.2 µm filter in 15 min, whereas the extraction resins exhibited lesser recoveries (85% and 79% Sc recovery respectively) over 30 min (Misiak et al. 2017). Activity yields reported using the natural Ca target were ~ 144 kBq/µAh at the optimum proton beam energy, but with only 87% radionuclidic purity (Misiak et al. 2017). These results are par- tially attributable to the low percentage of 48Ca present in natural Ca (0.187% natural abundance). Enriched 48Ca targets have been entertained as a solution to the low activity yields reported, but the prohibitive cost of 48Ca enrichment has stalled the study of this route, despite the potential to produce 47Sc at activity levels on the order of GBq (Misiak et al. 2017). An alternative method of production using a cyclotron involves an enriched 44CaO target (97% 44Ca) and α-particle irradiation via the 44Ca(α,p)47Sc nuclear reaction and using a precipitation method with a 0.22 µm sterile filter (1.5 h separation) (Minegi- shi et al. 2016). Even with a highly enriched target material, the activity yield 1.5 h after the end of bombardment (EOB) was ~ 780 kBq/µAh and radionuclidic purity post-sepa- ration was lower than that reported when using a 48Ca target (22.9% 47Sc vs. 73.5% 43Sc at EOB, but rising gradually to a maximum 85% at ~ 35.6 h after EOB as dominant 43Sc decayed) (Minegishi et al. 2016). These were attributed to the relatively long half-life of 47Sc and low production yields (Minegishi et al. 2016; Misiak et al. 2017). Sadler et al. EJNMMI Radiopharmacy and Chemistry (2022) 7:21 Page 9 of 55 High thermal neutron flux reactor production routes have been used in conjunction with enriched 46Ca targets using the indirect 46Ca(n,γ)47Ca → 47Sc nuclear reaction to produce 47Sc using a “pseudo-generator-like” approach similar to that employed in the production of 44Sc (Müller et  al. 2014a; Domnanich et  al. 2017b). This approach has shown promise, with one study reporting up to 2  GBq of 47Sc being feasibly isolated, while optimisation of the post-production separation and isolation procedures, along with repeated separation of in-grown 47Sc from the target material over the following days, allowed for 1.5 GBq to be isolated with a radionuclidic purity exceeding 99.99% in a ~ 700 µL fraction (Domnanich et al. 2017b). This investigation reported the first relia- ble and reproducible production of 47Sc of sufficiently high yield and radionuclidic purity for clinical applications by way of thermal neutron bombardment of enriched 46Ca tar- gets. This development serves to potentially mitigate the high cost of the enriched 46Ca target material through the implementation of efficient separation and target material recovery methods, and highlights the potential for 47Sc to be produced in the necessary quantities and with the necessary purity for radiopharmaceutical purposes in the future (Domnanich et al. 2017b). Comparison of this enriched 46Ca method with fast neutron bombardment of an enriched 47Ti target was also reported in the same study (Dom- nanich et al. 2017b). Lower radionuclidic purities were reported using the 47Ti(n,p)47Sc nuclear reaction, along with significant long-lived 46Sc impurities ranging from 3.8 to 11.5% which were dependent on the irradiation period and neutron energy (Domnanich et  al. 2017b; Bokhari et  al. 2010; Zerkin and Pritychenko 2018). These detracting fac- tors were deemed unfeasible for the production of 47Sc with sufficient purity and yield for radiopharmaceutical purposes, hereby bolstering the attractiveness of the enriched 46Ca route (Müller et al. 2014a; Domnanich et al. 2017b). Other studies have supported these conclusions (Kolsky et al. 1998; Bartoś et al. 2012). A similar investigation using an enriched 48Ti target and proton irradiation via the 48Ti(p,2p)47Sc nuclear reaction encountered analogous issues of 46Sc impurities (Srivastava 2012). Alternative inquiries into 47Sc have been conducted by employing linear accelerators using both natural and enriched Ti as well as Ca targets with γ-ray irradiation. Investiga- tion of enriched 48Ti targets using the 48Ti(γ,p)47Sc nuclear reaction has been reported, with comparison to 51  V(γ,α)47Sc (Yagi and Kondo 1977). The former was initially deemed superior to the latter despite the near 100% natural abundance of 51V due to unavoidable co-production of 46Sc and 48Sc, as well as low yields of the desired radionu- clide when using the 51V target (Yagi and Kondo 1977). More recently, due to the issue of high prices for enriched Ca and Ti starting materi- als, the 51 V(γ,α)47Sc reaction has been revisited using high-energy photons from elec- tron linear accelerators has allowed for the production of carrier-free 47Sc at purity levels of > 99.998% and 98.8% using 20 MeV and 38 MeV bremsstrahlung energies, respectively (Snow et  al. 2021). With appropriate method development, it has been theorised that the required therapeutic dose quantities of 3700 MBq of 47Sc could be feasibly produced using this method, with the 20  MeV energy requirements being attainable at nuclear medicine facilities for potential “dose-on-demand” (Snow et al. 2021). However, the 48Ti(γ,p)47Sc nuclear reaction is still marred by the same impurities of 46Sc and 48Sc at percentages of 1.3 and 10.3% of the 47Sc activity at the end of irradiation (EOI), with a number of other Sc isotopes also present (Yagi and Kondo 1977). Purity Sadler et al. EJNMMI Radiopharmacy and Chemistry (2022) 7:21 Page 10 of 55 and yields with low to negligible 46Sc and 48Sc contaminants could only be improved with irradiation energies above 40 MeV with highly enriched [48Ti]TiO2 targets. Irradiation of said target with 30, 40, 45 and 60 MeV bremsstrahlung energies showed best results were achieved at 60 MeV, with 46Sc and 48Sc impurities at 0.57 and 0.26% respectively 12 h post-EOI (Yagi and Kondo 1977). It has been reported elsewhere more recently that maximising 47Sc production using this reaction is possible with 22 MeV irradiation ener- gies, while also minimising the co-production of other Sc isotope impurities (Mamtimin et  al. 2015). Also more recently, natural T iO2 targets irradiated with Bremsstrahlung photons were able to yield 47Sc at 4.25 MBq/g.kW.hr and 6.92 MBq/g.kW.hr using 35 and 40 MeV beam energies, with the prospect of producing up to ~ 2.96 GBq and ~ 4.81 GBq of 47Sc using the same beam energies with a 5 g target (Rotsch et al. 2018). The indirect 48Ca(γ,n)47Ca → 47Sc nuclear reaction has also been investigated using electron linear accelerators, with results showing the plausibility of high specific activ- ity 47Sc being able to be produced via this method (Rane et al. 2015; Starovoitova et al. 2015). The optimum electron beam energy for 47Ca production was calculated as 40 MeV in one study, with the size of the target reported as playing an integral role in the result- ing specific activity of 47Sc produced (Rane et al. 2015). Smaller target sizes were said to result in higher average photon flux through the target, hereby increasing the specific activity produced (Rane et al. 2015). However, it was highlighted in another study that accurate and reliable photonuclear cross sections are necessary for ensuring optimum electron beam parameters as well as correctly predicting yields of desired radionuclides and minimising the co-production of radionuclide contaminants (Starovoitova et  al. 2015). Terbium: 149Tb, 152Tb, 155Tb and 161Tb Terbium has been dubbed the ’Swiss army knife’ of nuclear medicine due to the wide range of radioisotopic tools possible from this element. The medically relevant radio- nuclides 149Tb, 152Tb, 155Tb and 161Tb cover all the bases: alpha, beta, and Auger emis- sions suitable for therapeutic applications (149Tb,  161Tb), and positron and gamma emissions for diagnosis (152Tb, 155Tb) making terbium an ideal candidate for theranostic applications. 149Tb Alpha radionuclides have garnered significant interest in nuclear medicine due to their ability to deliver a high radiation dose to tumours. This property has allowed for the treatment of a variety of diseases including neuroendocrine tumours (Navalkissoor and Grossman 2019) and, most successfully, prostate cancer with bone metastases (Skel- ton et al. 2020). The alpha emitting radionuclides most used in these treatments (225Ac, 223Ra) have alpha emitting daughter products. 149Tb is unique in that it has only has a single alpha emission in its decay pathway decreasing the potential impacts of off-target radiation dose to patients. Due to this property, it has attracted research interest for its potential application in targeted α settings, notably in folate receptor targeted α-therapy (Müller et  al. 2012, 2014b), as well as certain radioimmunotherapy (RIT) applica- tions (Beyer et  al. 2004a). Other notable properties of 149Tb include a relatively short half-life (T1/2 = 4.1  h), low α-energy (3.97  MeV, Iα = 16.7%), and a positron emission S adler et al. EJNMMI Radiopharmacy and Chemistry (2022) 7:21 Page 11 of 55 ( Eβ+,mean = 730 keV, Iβ+ = 7.1%) that may allow, via the use of quatitative PET imaging, the acquisition of patient dose distribution data during therapeautic administration of the radionuclide (Müller et al. 2016). 149Tb can be produced via three main pathways (1) proton induced spallation, (2) heavyion induced nuclear reactions (e.g. 12C), and (3) light ion induced (e.g. proton or 3He) nuclear reactions (Beyer et al. 2002). The proton induced spallation of Ta targets has been used for many decades to pro- duce a range of radionuclides. Relatively large amounts of 149Tb are able to be pro- duced by the nuclear reaction Ta(p,x)149Tb, however on-line mass separation processes are required to yield a product of high radionuclidic purity. This method utilises thick Ta targets to compensate for the overall lower neutron capture cross sections that are characteristic for radiolanthanoid production using this technique, and has been calcu- lated to potentially produce up to 19 GBq/μA with a 100 g/cm2 target and proton beam energies on the order of 100 μA (Beyer et al. 2002). 149Tb produced via this method has recently been used in the preclinical analysis of [149Tb]Tb-DOTA-1-NaI3-octreotide ([149Tb]Tb-DOTA-NOC) and [ 149Tb]Tb-PSMA-617 (see Fig.  3), as suitable [ 149Tb] TbCl3 formulations were readily obtained for radiolabelling after on-line mass separa- tion (Müller et al. 2016; Umbricht et al. 2019). Radiochemical purity (> 98%) and specific activity (5 MBq/nmol) of [ 149Tb]Tb-DOTA-NOC have been achieved, while [ 149Tb]Tb- PSMA-617 was prepared at > 98% radiochemical purity at 6 MBq/nmol levels, both for- mulations of which were suitable for preclinical studies on AR42J tumour-bearing mice (Müller et al. 2016) and PSMA-positive PC-3 PIP tumour-bearing mice (Umbricht et al. 2019), respectively. Of the lightion induced production methods, the bombardment of 151Eu with 70 MeV 3He can yield 150 MBq/μA of 149Tb during an 8 h irradiation which is appropriate for therapeutic applications (Moiseeva et  al. 2020). The radiochemical purification of 149Tb from the Eu target is relatively straightforward however the limited number of high-intensity 3He beams worldwide significantly limits potential production via this method. Alternatively, proton bombardment of 152Gd can yield 2600  MBq/uAh 149Tb via the 152Gd(p,4n)149Tb nuclear reaction, however the lack of availability of suffi- ciently enriched 152Gd target material (only 0.2% natural abundance) has restricted the O O O N H O N N 149Tb NH O N N O O NH O HO O O O O O N N H H OH OH Fig. 3 Molecular structure of [149Tb]Tb-PSMA-617 Sadler et al. EJNMMI Radiopharmacy and Chemistry (2022) 7:21 Page 12 of 55 feasibility of this approach, as more refined target enrichment is necessary for worth- while production of 149Tb via this method (Beyer et al. 2002). Heavy ion induced 149Tb production can also be achieved via direct 141Pr(12C,4n)149Tb and indirect 142Nd(12C,5n)149Dy → 149Tb routes. Quantities of 149Tb suitable for anti- body radiolabelling have been produced using these methods, after the removal of impu- rities and bulk target material via cation-exchange chromatography on an Aminex A5 column (60 mm, 3 μm, ~ 13 μm particle size) with α-hydroxyisobutyric acid (α-HIBA) as eluent and subsequent reconstitution in HCl for the final formulation of [149Tb]TbCl3 (Sarkar et al. 1999). 155Tb 155Tb undergoes radioactive decay exclusively by electron capture (EC) and has two gamma emissions at 87 keV (32%) and 105 keV (25%) that are suitable for SPECT imag- ing. The longer half-life (5.32 d) makes it particularly suitable for delivery vectors with longer biological half-lives such as antibodies and large proteins. The 155Tb radionuclide as a SPECT agent has theranostic application as a companion dosimetry and treatment planning tool for 149/161Tb labelled radiotherapeutics. Preclinical evaluation of [ 155Tb] Tb-DOTA-TOC in mice bearing neuroendocrine tumours has been undertaken (Müller et al. 2014c). 155Tb has been produced by numerous methods including α-irradiation of Eu targets (Levin et al. 1977), via photonuclear reactions using an EA-25 electron accel- erator (Levin et  al. 1981), and by proton and deuterium bombardment of Gd targets on 11 to 22 MeV cyclotrons (Dmitriev et al. 1989; Favaretto et al. 2021). Notably, 155Tb has been produced at the CERN-MEDICIS facility via spallation of a high purity Ta-foil target with 1.4 GeV protons, after which online mass separation of m/z 155 was used to collect approximately 20  MBq of the desired 155Tb (along with other isobars) onto a Zn-plated Au foil. Ion-exchange and extraction chromatography were subsequently used to isolate 155Tb from isobaric impurities, which gave a final 155Tb formulation with radionuclidic purity > 99.9% that was deemed suitable for pre-clinical use (Webster et al. 2019). 152Tb and matched pairs with 149/155/161Tb 152Tb is a positron emitting radionuclide with a relatively long half-life (T1/2 = 17.5  h) and relatively high-energy positron emissions ( Eβ+mean = 2795  keV, 13.9%) and has potential as a SPECT radionuclide due to multiple gamma emissions. The high positron emission energy, however, leads to lower spatial resolution compared to established PET radionuclides such as 18F and 11C. These high-energy positron emissions in combination with an array of gamma emissions raises some concerns for radiation burden for patients and workers. 152Tb is used as a tool for dosimetry and treatment monitoring for 149/161Tb radiotherapeutics and being a PET radiolanthanoid it also has potential application as a companion PET diagnostic/treatment planning tool for 153Sm and 165Dy radiotherapeu- tics. 152Tb has been used in combination with 149/155/161Tb labelled DOTA-conjugates targeting the folate receptor (Chopra 2004), in radioimmunoconjugates for targeted α-therapy for malignant melanoma (Rizvi et al. 2000), and in the first-in-human appli- cation for radiotherapy of neuroendocrine tumours as [152Tb]Tb-DOTA-TOC (Baum et  al. 2017). 152Tb is produced by heavy ion reactions and proton-induced spallation S adler et al. EJNMMI Radiopharmacy and Chemistry (2022) 7:21 Page 13 of 55 of Ta targets followed by isotopic separation (Allen et al. 2001), but the production of 152Tb in quantities useful for clinical applications remains a significant challenge for light charged particle and heavy ion (HI) activation (Naskar and Lahiri 2021). Lanthanum: 132La, 133La and 135La The trio of lanthanum radionuclides 132La, 133La and 135La present an interesting pros- pect for the fulfilment of the ideal theranostic concept of chemically-identical match- ing of diagnostic and therapeutic radionuclides (Aluicio-Sarduy et  al. 2020). 132La ( t1/2 = 4.6 h) is a positron emitter that has garnered interest as a PET imaging radionu- clide and has seen applications as a surrogate imaging agent to probe the in vivo biodis- tribution of 225Ac due to similarities in their chemical properties (Aluicio-Sarduy et al. 2021). Likewise, the positron emitter 133La has also emerged as a PET imaging candi- date with the potential to improve the PET imaging quality of metastases and smaller tumours. This is due to the lower maximum positron emission energies exhibited by 133La, which result in greater spatial resolution in PET imaging compared to using other radionuclides of interest such as 68 Ga and 44Sc (Nelson et al. 2020). Its longer half-life (t1/2 = 3.9 h) than the commonly-used 18F and 11C PET agents also presents 133La as a longer-lived alternative for the PET imaging/monitoring of longer biological processes (Nelson et al. 2020). Their therapeutic counterpart, 135La ( t1/2 = 18.9 h), has also attracted research interest due to its Auger electron emissions with high linear energy transfer (LET) that provide an attractive alternative to traditional β−-therapy due to their shorter tissue range and higher number of ionisation events per unit distance travelled being ideally suited for a more targeted approach (Aluicio-Sarduy et al. 2020). Their reduced tissue range and increased propensity for double strand breaks in DNA also makes them ideal for the treatment of smaller tumours and metastases, as they cause more local- ised damage and reduce collateral damage to healthy tissue (Aluicio-Sarduy et al. 2020; Fonslet et al. 2017). Their potential as theranostic pairs has been investigated through their complexation by 10-((6-carboxypyridin-2-yl)methyl)-1,4,7,10-tetra-azacyclodo- decane-1,4,7-triacetic acid (DO3Apic) and N,N’-bis[(6-carboxy-2-pyridil)methyl]-4,13- diaza-18-crown-6 (macropa) chelators and incorporation into PMSA-targeting agent 2-[3-(1,3-dicarboxypropyl)ureido]pentanedioic acid (DUPA) (see Figs. 4 and 5) for the in vivo imaging of PSMA-expressing xenografts in mice (Aluicio-Sarduy et al. 2020). Production of La radionuclides has been achieved using natural Ba targets with cyclo- trons (Fonslet et al. 2017). The production of 135La via the proton irradiation of natural Ba targets was achieved using 16.5 MeV proton beam energies and irradiation times of 235–280  min (Fonslet et  al. 2017). This production method resulted in the formation of a number of La isotopes, however the short-lived nature of 134La ( t1/2 = 6.5  min), 136La (t1/2 = 9.9 min), and slightly longer-lived 132La (t1/2 = 4.5 h) and 133La (t1/2 = 3.9 h) allowed for the targets to be left to decay for 12–24 h to increase the overall activity of 135La and minimise contamination with other isotopes. Purification consisted of dissolu- tion in aqueous HCl, heating, pH adjustment and CM cation-exchange resin with 0.1 M HCl for separation and elution of purified 135La (Fonslet et al. 2017). Average produc- tion amounts had activity yields of ~ 407  MBq/µAh with radionuclidic purities of 98% at 20 h after irradiation. This method employed chemical separations capable of recov- ering > 98% of 135La produced with an effective molar activity of ~ 70  GBq/µmol in its Sadler et al. EJNMMI Radiopharmacy and Chemistry (2022) 7:21 Page 14 of 55 O OH HO O O NH O O NH N N O H HN O HO O N 13XLa N O N O O N O Fig. 4 Molecular structure of [13XLa]La-DO3Apic-DUPA, where X = 2, 5 O OH O O HO O O O S O O NH N O O N N H NH N NH H N 13XLa N HO O O O Fig. 5 Molecular structure of [13XLa]La-macropa-DUPA, where X = 2, 5 final formulation (Fonslet et al. 2017). As such, the high activity, radionuclidic purity and yield of the 135La isolated were sufficient for the intended clinical applications, and the use of medical cyclotrons already in operation also makes this production route favour- able for the production of clinically relevant amounts of 135La in the future (Fonslet et al. 2017). However, this study noted a relatively modest separation factor of ~ 102, but that the high labelling effective specific activity obtained rendered the scrupulous separation of 135La unnecessary (Fonslet et al. 2017). It was also noted that enriched 135Ba targets would increase overall production yield and radionuclidic purity. The use of enriched Ba targets in the form of [ 135Ba]BaCO3 (94.9% 135Ba enrichment) in conjunction with cyclotrons has also been investigated for the production of no- carrier-added (n.c.a) 135La (Mansel and Franke 2015). This study reported radiochemi- cal yields of ~ 83% and activity yields of ~ 43 MBq using 18 MeV beamline energies and separation using La selective resins, however these results were not suitable for clini- cal applications and are not of the commercially viable scale necessary for such applica- tions (Mansel and Franke 2015). As reported elsewhere, for production routes involving enriched 135Ba to become more viable, significant target development is required to permit the irradiation of an enriched barium oxide or salt, as well as improvements to allow a reduction in 135Ba2+ recycling after the separation process (Fonslet et al. 2017). In addition, separation procedures for the radiochemical isolation of La radionuclides have been few, and those employed so far have resulted in moderate chemical purity S adler et al. EJNMMI Radiopharmacy and Chemistry (2022) 7:21 Page 15 of 55 and separation factors (Fonslet et al. 2017; Mansel and Franke 2015; Aluicio-Sarduy et al. 2019). This has resulted in focus on the production of 135La as a matched radionuclide pair with both 132La and 133La from natural Ba targets for a more seamless incorporation of the theranostic concept into current bifunctional chelators and targeting vectors for further applications. Cyclotrons have also been used for the production of 132La from natural Ba targets via the 132Ba(p,n)132La chemical reaction, in a similar manner to the production of 135La, for the production of the theranostically relevant 132La/135La radionuclide pair (Aluicio- Sarduy et al. 2021, 2019). However, it has been reported that current production routes to 132La using natural Ba targets have been hindered by current cyclotron production methods requiring long irradiation times and the relatively low natural abundance of the requisite 132Ba isotope (0.1% natural abundance) (Nelson et al. 2020). These production route issues, in conjunction with potentially unfavourable decay characteristics such as the high-energy positron emission of 132La resulting in reduced spatial resolution for PET imaging (particularly small tumours), and the high abundance of γ-ray emis- sions which could result in dose tolerance issues in patients and transportation/handling problems, have led to the investigation of 133La as a less production-intensive and more easily handled/administered alternative (Nelson et al. 2020). The 133La/135La theranostic pair were efficiently produced in a more recent work via natural Ba targets with pro- ton bombardment in medical cyclotrons, in an analogous way to the previously reported production of the 132La/135La pair (Nelson et al. 2020; Aluicio-Sarduy et al. 2019). This study made use of a new type of natural Ba metal target which was completely encap- sulated in an Ag disc covered with Al foil, with assembly being simple and components reusable (Nelson et  al. 2020). Targets were irradiated with 22  MeV proton beams for 25–200 min before separation of the desired radionuclide pair using a modified method similar to that used previously for the separation of the 132La/135La radionuclide pair (Nelson et al. 2020). Formation of 131La and 132La were observed in small amounts, and 134La and 136La were observed in significant amounts at EOB, but soon decayed owing to their short half-lives (6.45 and 9.87 min respectively) (Nelson et al. 2020). The result- ing radionuclide pair was recovered consistently at > 88% after the automated separation (~ 35 min), and the activity yields were ~ 231 MBq 133La and ~ 166 MBq 135La at EOB, with saturated yields of ~ 161  MBq/µA 133La and ~ 561  MBq/µA 135La. These values were an order of magnitude higher than those reported for the 132La/135La pair (Nelson et  al. 2020; Aluicio-Sarduy et  al. 2019). Radionuclidic purity and yield were high, and the 132Ba(p,n)132La nuclear reaction was largely avoided due to low isotopic abundance of 132Ba in the target material and a low reaction cross-section over the energy range used. The radionuclide pair was produced with sufficient activity yield and radionuclidic purity for radiolabelling applications with DOTA and macropa resulting high incorpora- tion reported for both chelators (Nelson et al. 2020). This method enabled the produc- tion and isolation of clinically relevant activities of 133La/135La using low cyclotron beam energies with shorter irradiation times than those needed for 132La/135La, and without the added expense of isotopically enriched Ba (Nelson et al. 2020). It was also noted that enrichment of the Ba target to remove the 132Ba isotope would improve the purity even more due to the effective removal of the 132La impurity arising from the 132Ba(p,n)132La reaction. This would also remove any 131La arising from the decay of 132La, leaving only Sadler et al. EJNMMI Radiopharmacy and Chemistry (2022) 7:21 Page 16 of 55 the desired radionuclide pair after a decay period of 3 h. Adopting such targets would also lower the cyclotron energy required for the production of radionuclidically pure 133La/135La (Nelson et al. 2020). Holmium: 166Ho 166Ho has seen a wide range of applications in cancer therapy over the last few decades (Shi et  al. 2017). A wide array of therapeutic applications have been investigated with pre-clinical and clinical trials and studies being reported for radiopharmaceutical ther- apies for bone marrow cancer (Bayouth et  al. 1995a, b; Giralt et  al. 2003; Christofori- dou et  al. 2017), metastatic bone pain palliation (Voorde et  al. 2019; Bahrami-Samani et al. 2010), brain cancer (Huh et al. 2005; Ha et al. 2013), liver cancer (Cho et al. 2005; Kim et al. 2006), and prostate cancer (Seong et al. 2005; Kwak et al. 2005), among oth- ers. This radionuclide has gained attention for its potential theranostic applications due to its favourable decay characteristics that allow for its use as both a therapeutic agent and an imaging agent (Tan et al. 2020). 166Ho has a physical half-life of 26.8 h and emits β−-particles [ Eβmax = 1.854 MeV (50.0%) and 1.774 MeV (48.7%)] suitable for β−- therapy (Voorde et al. 2019), while also producing γ-emissions (80.6 keV, 6.2%) that can be used for γ-scintigraphy or SPECT without an excessive dose burden to the patient or radiation damage during transportation, handling, storage or administration since the γ-emissions are not high-energy. In addition, the highly paramagnetic nature of 166Ho (4f11 with 3 unpaired electrons) has led to its use as a magnetic resonance imaging (MRI) contrast agent (Shi et al. 2017; Tan et al. 2020; Vente et al. 2007). These proper- ties allow for the unification of both aspects of the theranostic concept in one radionu- clide, which has advanced 166Ho as a promising option over the more readily-researched theranostically-matched radionuclide pair approach, due to the alleviation of the need for investigation into theranostic counterparts that satisfy the requirements of similar chemical properties, production availability, chelation kinetics, biodistribution, pharma- cokinetics and toxicity concerns, among others. Furthermore, 166Ho has the additional apparent advantage of a comparably less complicated nuclear reaction target enrichment process, due to the availability of the requisite parent isotope, 165Ho, in 100% natural abundance (Shi et al. 2017; Bahrami-Samani et al. 2010). This has led to 166Ho produc- tion using the 165Ho(n,γ)166Ho nuclear reaction, and has allowed for (in combination with a thermal neutron capture cross-section of 64.7 b) relatively high specific activity yields of 166Ho (2–5  GBq/mg) to be produced (Mishiro et  al. 2019; Iaea-Tecdoc 2003; Zolghadri et al. 2013; Yousefnia et al. 2014). This method typically utilises neutron beam energies ~ 4 × 1013 n/cm2/s and irradiation times of from 20 to 60 h (Iaea-Tecdoc 2003; Zolghadri et al. 2013; Yousefnia et al. 2014). It has however been reported elsewhere that issues arise with this method due to the resulting 166Ho being carrier-added and hence not suitable for certain radiolabelling applications, as well as the long-lived 166mHo impu- rity (t1/2 = 1200 y) (Voorde et al. 2019). Moreover, despite the thermal neutron cross-sec- tion of this nuclear reaction resulting in high activity levels of the desired radionuclide, specific activity levels are lower than desired at saturation yields due to only a small proportion of the target material undergoing neutron capture and converting to 166Ho (Voorde et al. 2019). As a result, the 166Ho produced at these modest specific activities cannot be used for the purposes of radiolabelling target molecules (Voorde et al. 2019). S adler et al. EJNMMI Radiopharmacy and Chemistry (2022) 7:21 Page 17 of 55 Moreover, radiopharmaceuticals based on 166Ho have often encountered issues with dissemination to the relevant medical facilities that require them due to the relatively short half-life of 166Ho (t1/2 = 26.8 h) (Voorde et al. 2019). The production sites for 166Ho, namely nuclear reactors, are typically located further away from most medical facili- ties, and consequently the 166Ho is only able to be transported to facilities within a small radius of the production site due to its disadvantageous half-life (Voorde et  al. 2019). This method of production resulting in carrier-added 166Ho with lower-than-required specific activities encounters the issues of requiring direct access to a nuclear reactor for radionuclide production and transportation. In addition, the long irradiation times required by this method (typically ~ 60  h) are not ideal. This has led to other produc- tion methods being investigated, namely a 166Ho radionuclide ‘generator’ approach using 164Dy as target material for no-carrier-added 166Ho (Vosoughi et al. 2017a, b). Alternative methods of this nature use 164Dy as the target material in conjunction with a double neutron capture reaction 164Dy(n,γ)165Dy(n,γ)166Dy → 166Ho (Dadachova et al. 1994; Lahiri et al. 2004). This reaction proceeds via an intermediate 165Dy isotope (t1/2 = 2.33 h) with a neutron capture cross-section of 3900 barns (Voorde et al. 2019). The generator-like approach allows for an accumulation of 166Ho to accrue over time due to the significantly longer half-life of 166Dy compared to 166Ho, and selective elu- tion of the desired radionuclide is facile (Vosoughi et al. 2017a, b), typically utilising a metal-free HPLC system with cation-exchange columns and pH-adjusted α-HIBA as the eluent (Dadachova et al. 1994; Lahiri et al. 2004). This allows for 166Ho to be gen- erated on-demand without access to a nuclear reactor and alleviates issues that would otherwise arise regarding distribution from remote production sites. The α-HIBA acts as a complexing agent, and owing to the smaller ionic radius and larger charge density of the 166Ho due to lanthanoid contraction, a more thermodynamically stable complex is formed compared to the analogous 166Dy-complex, and the 166Ho-complex is removed from the column first (Voorde et al. 2019; Dadachova et al. 1994). Having been separated from the 166Dy-complex, the 166Ho is demetallated from the α-HIBA chelator using acidic chloride solutions and a subsequent cation-exchange separation from α-HIBA gives 166Ho in a carrier-free formulation in solution with a radiochemical yield of > 95% and very low breakthrough of 166Dy (< 0.1%) (Dadachova et  al. 1994). This separation method had a separation factor of ~ 103 between 166Ho and 166Dy which was achieved in under 2 h (Dadachova et al. 1994). Similar results were reported in another study, but with essentially no 166Dy breakthrough, which was attributed to the use of an Aminex A7 column compared to the Dowex 50 exchangers which proved ineffective at separat- ing the 166Ho from the target material (Lahiri et al. 2004). Other chromatographic paradigms have been employed for 166Ho separation from 166Dy post-production with high radionuclidic purities being reported (Vosoughi et al. 2017b). Extraction chromatography with Eichrom LN2 resin (containing 2-ethylhex- ylphosphonic acid mono-2-ethylhexyl ester, HEH[EHP]) as the stationary phase extract- ant was used with an eluent of 1.5 M nitric acid at a temperature of 25  °C (Vosoughi et al. 2017b). This study incorporated a pre-washing phase using 0.1 M nitric acid for the removal of impurities before subsequent extraction (Vosoughi et al. 2017b). In a simi- lar exploitation of chemical properties using α-HIBA in previous investigations, compl- exation of 166Ho over 166Dy is achieved due to a more thermodynamically stable chelate Sadler et al. EJNMMI Radiopharmacy and Chemistry (2022) 7:21 Page 18 of 55 being formed due to holmium’s smaller ionic radius and consequently higher charge density. A flow rate of 1.5 mL/min was used to give quantitative separation in 1.5 h with a resultant no-carrier-added formulation of [166Ho]HoCl3 being isolated with radionu- clidic purity of > 99.9%, a high separation yield of 76% and a radiochemical purity of > 99% (Vosoughi et al. 2017b). The 166Ho isolated from this production route was of high spe- cific activity and suitable for radiolabelling purposes and the production yield was suf- ficient for clinical applications (Vosoughi et  al. 2017b). Another methodology using a similar Eichrom Ln SPS resin (containing di-(2-ethylhexyl)phosphoric acid HDEHP) as the stationary phase extractant has also been used to significant effect for the same separation purposes (Monroy-Guzman et al. 2015; Monroy-Guzman and Jaime 2015a). This method of separation involved a multi-step process: Irradiation of the nitrate salt of 164Dy, formation of the parent/daughter radionuclide pair 166Dy/166Ho via dissolution of the salt target in 0.15 M nitric acid, adsorption onto the Eichrom Ln SPS resin-loaded chromatographic column, desorption and elution of 166Dy with 1.5 M nitric acid, then desorption of 166Ho with 3 M nitric acid followed by precipitation of the 166Ho(OH)3 salt through addition of NaOH to the 166Ho eluate, then finally re-dissolution in 0.1 M HCl to afford the desired [ 166Ho]HoCl3 formulation (Monroy-Guzman et  al. 2015). Radio- nuclidic purities of > 99.9% were attained using this method with the added advantage of a relatively short separation time of 20–25  min, both of which are ideal for subse- quent radiolabelling and clinical applications (Monroy-Guzman et al. 2015). The 166Ho produced via this method is carrier-free, and was separated with 100% efficiency from the target material and parent isotope (Monroy-Guzman et al. 2015; Monroy-Guzman and Jaime 2015a). Electrophoresis or ion-chromatography have also been suggested as a means of sepa- ration of the 166Dy/166Ho pair to afford no-carrier-added 166Ho, but has not resulted in a formulation of the desired radionuclide with the appropriate activity yield or radio- nuclidic purity necessary for radiolabelling or further biomedical applications, as only partial separation being feasible (Dadachova et al. 1995). The target material used was Dy2O3 and the same nuclear reaction as the previous methods was employed, however the separation utilised HDEHP or tri-butyl phosphate (TBP) as the stationary chromato- graphic phase with elution using mobile phases of nitric acid that were relatively higher in concentration than other methods (3–12 M H NO3) (Dadachova et al. 1995). In the same study, electrophoresis with α-HIBA as the chelating agent for 166Ho was investi- gated, again with only partial separation being possible (Dadachova et al. 1995). Samarium: 153Sm Samarium-153 is an attractive example of a β−-emitter with appropriate γ-emissions (energy = 103  keV, 28%) suitable for imaging in conjunction with therapy (Tan et  al. 2020). Typical β−-particles emitted by this radionuclide have average energies of 0.23 MeV, with three main emission energies of 808 (18%), 705 (50%) and 635 (32%) keV respectively (Tan et al. 2020; Bombardieri et al. 2018). The combination of β−-particle emission and γ-emission in the one radionuclide has garnered attention for theranostic applications due to appropriate tissue penetration of medium-energy β−-particles (aver- age range of 0.5  mm, and an effective range of 2–3  mm) for targeted treatment, with concurrent γ-imaging capabilities by means of SPECT or γ-scintigraphy (Voorde et al. S adler et al. EJNMMI Radiopharmacy and Chemistry (2022) 7:21 Page 19 of 55 2019; Bombardieri et al. 2018; Sun et al. 1996). With a half-life appropriate for radionu- clide therapy ( t1/2 = 46.3 h), it has gained considerable attention due to its widespread use in the clinic as a bone pain palliation agent in patients with painful bone metasta- ses arising as complications from various cancers (Das and Banerjee 2017; Pillai 2010; Anderson and Nuñez 2007; Jong et al. 2016; Fricker 2006). Incorporation of 153Sm into the ethylenediaminetetra(methylene phosphonate) (EDTMP) ligand (see Fig.  6) has proved especially efficacious for such applications, due to the targeted nature of its selec- tive uptake into the bone matrix, particularly the new bone matrix formed by osteo- blasts, and was approved by the US FDA in 1997 (Pillai 2010; Anderson and Nuñez 2007; Jong et al. 2016; Fricker 2006; Goeckeler et al. 1987; Lattimer et al. 1990; Atkins 1998). 153Sm complexes for use in treating bone metastases have the advantage of being capa- ble of both imaging the affected bone sites by means of a radionuclide bone scan, and providing efficacious β−-therapy for these cancer metastases, hereby fulfilling the thera- nostic concept of joint diagnostics and therapy ( Quadramet®. 2017; Sartor et al. 2004; Morris et al. 2009). As such, with bone pain metastases being fairly common complica- tions in patients with a variety of different types of cancer (> 50% of cancer patients), 153Sm has found application in the treatment of a wide range of bone metastases aris- ing as complications from a number of different types of cancer (Voorde et  al. 2019). Most significantly, 80% of breast, prostate and lung cancer patients experience painful bone metastases, with reductions in bone pain being essential to patient recovery, for which the administration of [ 153Sm]Sm-EDTMP has been reported as greatly efficacious (Voorde et al. 2019; Eary et al. 1993; Silberstein 2005; Kolesnikov-Gauthier et al. 2018). Reduction of bone-metastasis-related pain after administration of [153Sm]Sm-EDTMP (37 MBq/kg standard dose) has been shown to occur in 55–70% of patients in certain studies (Eary et al. 1993; Silberstein 2005; El-Amm and Aragon-Ching 2016). A recent study has also shown that [153Sm]Sm-EDTMP was equally safe and effective as a bone pain palliation agent when compared to its 177Lu-labelled analogue (Taheri et al. 2018). Both radiopharmaceuticals were compared in a double-blind study for their efficacy and safety when administered for the amelioration of commonly encountered symptoms of bone metastases in cancer patients. No difference was observed between groups regard- ing pain alleviation, with both groups experiencing significantly reduced pain from week 2 onwards and lasting for 12 weeks post-treatment (Taheri et al. 2018). Efficacious bone pain palliation results utilising 153Sm have been echoed by the observation that 67% of patients with painful bone metastases have displayed overall therapeutic bone stabilisa- tion/regression after only one dose of [153Sm]Sm-EDTMP (Elzahry et al. 2018). Focus on the use of 153Sm for such bone-metastasis applications has also been due to advantages it has over other radionuclides used for bone palliation like 89Sr, such as shorter physical half-life resulting in lower dose burdens to patients, lower β−-particle O O P N N P - - O 153O Sm O O P O O P - - O O O O Fig. 6 Molecular Structure of [153Sm]Sm-EDTMP Sadler et al. EJNMMI Radiopharmacy and Chemistry (2022) 7:21 Page 20 of 55 energy, more efficient delivery of radiation, fast bioelimination from the body and lower myelotoxicity (Tan et al. 2020; Pillai 2010; Serafini et al. 1998; Turner et al. 1989). Fur- thermore, the low probability of a γ-emission from 89Sr (909 keV, 9.555 × 10–5%) (Schima 1998) has led clinicians and researchers to resort to bremsstrahlung imaging or com- plementary radiotracer matches in order to image or monitor the administration of 89Sr (Qaim et al. 2018), which has led to 153Sm garnering attention as an alternative due to its γ-emissions being of significantly higher probability and suitable for SPECT and γ-scintigraphy (Tan et al. 2020). To this end, the γ-emission from 153Sm is ideally suited for high spatial resolution imaging with low signal-to-noise ratios which can be taken at different stages during treatment, with the added advantage of allowing for evaluation of radioactive leakage to other organs after each bone palliation treatment (Production 2021). Imaging of 153Sm for the purposes of monitoring bone palliation therapy progres- sion, radiopharmaceutical biodistribution or dosimetry measurements is performed using either a gamma camera or a SPECT/CT scanner equipped with high resolution, low-energy collimators (Tan et al. 2020). Images attained from 153Sm bone scans have been compared with bone scan images that used the more common [99mTc]Tc-meth- ylenediphosphonate (99mTc-MDP, see Fig. 7), and the results have been comparable in image quality and utility (Anderson and Nuñez 2007; Tripathi et al. 2006; Ramachandran et al. 2011). Comparable performance outcomes of this nature, in conjunction with its combined therapeutic and imaging efficacy, have served to reinforce the suitability and utility of 153Sm as a theranostic cancer agent. Researchers and clinicians have also investigated the potential for expanding the remit of 153Sm radiopharmaceuticals to include the treatment of other types of cancer, due to its suitable decay characteristics for other cancer treatment applications, afford- ability over other radionuclides and less complicated mode of production (Tan et  al. 2020). Theranostic treatment regimens for liver tumours have been proposed using 153Sm-labelled microparticles for transarterial radioembolization with post-procedural imaging, and an experimental study concluded that the 153Sm microparticles exhibited superior characteristics and performance compared to analogous 90Y microspheres (Hashikin et al. 2015). While the results are encouraging, dosimetry studies are impera- tive for ascertaining the correct amount of 153Sm activity required to achieve equivalent tumour doses and commensurate results compared to the readily used and commer- cially available 90Y microspheres (Tan et al. 2020). Progress was made in this regard, as a Monte Carlo simulation study was conducted to ascertain organ dose levels from hepatic radioembolization procedures using different radionuclides (Hashikin et al. 2016). The study reported that tumour dose equivalents for an activity of 1.82  GBq 90Y could be achieved with an estimated 4.44 GBq of 153Sm (~ 2.4 times the activity) (Hashikin et al. 2016). In light of these results, more investigation into specific dosimetry requirements, OH HO O P OH O P O 99mTc O O O P OH P O O OH HO Fig. 7 Molecular structure of bone scan agent [99mTc]Tc-methylenediphosphonate S adler et al. EJNMMI Radiopharmacy and Chemistry (2022) 7:21 Page 21 of 55 in vivo studies of pharmacokinetics, biodistribution and stability are needed for progress to be made in this regard (Tan et al. 2020). Moreover, the potential for 153Sm to be incorporated into treatment for spinal cancer and metastases has been explored (Donanzam et al. 2013; Montaño and Campos 2019). Added advantages such as cost-effectiveness, ease of production, theranostic proper- ties and availability have led 153Sm to be regarded as a potential radionuclide for these applications, with the most useful characteristic being its ability to be imaged as well as provide the appropriate β−-therapy. 153Sm was investigated as a potential neutron- activated radionuclide for incorporation into calcium phosphate for a biomaterial-based treatment for metastases of the spine by means of a radioactive bone cement applied to the vertebrae, hereby combining vertebroplasty and radiotherapy (Donanzam et  al. 2013). The biomaterial was a calcium phosphate bone cement which was synthesised with stable Sm using the sol–gel route for calcium phosphate synthesis. The resulting bone cement (containing stable Sm) underwent neutron activation, which produced a radioactive 153Sm-infused bone cement that was found to have activities of 153Sm on the order of ~ 33 MBq/mg, which is promising for further applications in radiovertebro- plasty (Donanzam et al. 2013). However, as with other approaches to radioactive bone cement (Tan et al. 2020; Hirsch et al. 2008), further clinical research and data must be collected before this relatively novel approach to spinal cancer and metastasis treat- ment is deemed appropriate for widespread clinical application. A similar approach was employed in another study with polymethylmethacrylate (PMMA) and hydroxyapatite (HA) as the bone cement, with results being similarly promising, however it was noted that further study regarding radiotoxicity, cytotoxicity, dosimetry, radiobiology and per- formance in the clinic must be conducted for further progress to be made (Montaño and Campos 2019). Production of 153Sm has typically been achieved in carrier-added form by means of neutron irradiation of both natural and enriched S m2O3 targets in a nuclear reactor, with the purity of the resulting 153Sm depending significantly on the enrichment of the target material (Voorde et al. 2019). These methods exploit the 152Sm(n,γ)153Sm nuclear reac- tion which has a relatively high neutron capture cross section of 206 barns (Tan et al. 2020). Only medium neutron fluxes are required due to the relatively high neutron cap- ture cross-section of the reaction (Das and Pillai 2013). Disadvantages encountered dur- ing the production process for 153Sm have included its relatively short half-life, making production and transportation logistics challenging; the carrier-added form of the prod- uct radionuclide; the long irradiation times required for sufficient activities and yields to be produced; and the propensity for long-lived radioactive impurities to be produced from side nuclear reactions during bombardment (Voorde et al. 2019). As can be expected, the use of natural Sm targets (3.1% 144Sm, 15% 147Sm, 11.2% 148Sm, 13.8% 149Sm, 7.4% 150Sm, 26.7% 152Sm, and 22.8% 154Sm) results in a product with sig- nificant impurities, the most significant of which are the long-lived 145Sm (t1/2 = 340 d), 151Sm (t1/2 = 88.8 y) and 155Eu (t1/2 = 4.76 y), and is therefore generally deemed not ideal for the production of high radiopurity and specific activity 153Sm (Islami-Rad et  al. 2011; Chakravarty et  al. 2018; Kalef-Ezra et  al. 2015; Ramamoorthy et  al. 2002). Despite this, progress has been made in the area of post-purification after irradiation of natural Sm2O3 targets by means of separating the desired 153Sm from longer-lived Eu Sadler et al. EJNMMI Radiopharmacy and Chemistry (2022) 7:21 Page 22 of 55 contaminants using ion reduction, ion exchange and solvent extraction methods (Islami- Rad et al. 2011). Of these methods, the ion exchange approach has allowed for the iso- lation of 153Sm from such impurities with recovery yields of > 66% and purities > 99.8% (Islami-Rad et al. 2011). Separation has also been achieved using electrochemical means, with 153Sm recovery yields > 85% and acceptable radiopurities suitable for some clinical applications, exemplified by the incorporation of the purified 153Sm obtained into [ 153Sm] Sm-EDTMP formulations with radiolabelling yields of > 98% and radiochemical purities of > 99% (Chakravarty et al. 2018). From natural Sm targets, activities of ~ 32 GBq after electrochemical purification of a ~ 37  GBq activity batch have been obtained with the 154Eu and 156Eu contaminants not being detected, implying high radiopurity suitable for [153Sm]Sm-EDTMP radiolabelling (Chakravarty et al. 2018). Studies comparing the suit- ability of 153Sm produced by either natural or enriched Sm targets have generally con- curred that enriched [ 152Sm]Sm2O3 targets result in higher overall activities and specific activities, however it has been noted that enriched [ 152Sm]Sm2O3 is considerably more expensive (Chakravarty et al. 2018; Ramamoorthy et al. 2002). This has prompted some researchers to focus on natural Sm targets and purification methods over purchasing enriched Sm targets (Islami-Rad et al. 2011). Additionally, it has also been noted in cer- tain studies that, for the bone pain palliation purposes for which 153Sm is typically used, sufficient specific activities can be achieved using natural Sm targets (Pillai 2010; Das and Pillai 2013; Ramamoorthy et al. 2002). For the requirements of targeted radionuclide therapy and further clinical applications, higher specific activities are needed particularly for radiolabelling purposes such as incorporation of 153Sm into antibodies, peptides or other targeting vectors/bifunctional chelators (Das and Pillai 2013). Obtaining sufficient specific activities for the aforemen- tioned purposes generally requires enriched 152Sm targets and long irradiation times on the order of days to ensure a high yield of the required radionuclide (Voorde et al. 2019). By using an enriched [152Sm]Sm2O3 target with neutron irradiation at 2–5 × 1013 n/cm2/s for a period of 3 to 7 days, one study reported specific activities of 44 GBq/mg 153Sm, which was determined to be approximately 4 times higher than the specific activi- ties obtained using natural S m2O3 targets (Ramamoorthy et al. 2002). These results are echoed elsewhere with approximately 4 times higher activities of 153Sm being produced using enriched targets compared to natural targets, using 10 g target sizes and similar neutron irradiation beam energies and irradiation times (Chakravarty et al. 2018). Forays into the subsequent purification of 153Sm produced from isotopically enriched targets has mirrored those of natural Sm targets, as some impurities arising from the nuclear reaction of the enriched target are common to the natural target, namely the long-lived 154Eu contaminant (t1/2 = 8.6 y). Importantly, the use of enriched [152Sm] Sm2O3 markedly reduces the side-production of 155Eu and 156Eu, but the presence of the 154Eu contaminant in the 153Sm product is very similar to the level observed using natu- ral Sm2O3 (Ramamoorthy et al. 2002). This is due to the 153Eu(n,γ)154Eu side reaction that has a neutron capture cross section higher than that of the desired 152Sm(n,γ)153Sm reaction (312 vs. 206 barns) (Voorde et al. 2019). Such an impurity has the potential to pose issues in the clinic, as patients can only be administered 0.093  kBq of 154Eu per MBq of 153Sm during treatment (Kalef-Ezra et al. 2015; Bourgeois et al. 2011). As a con- sequence of this regulation, production routes for 153Sm using either natural or enriched S adler et al. EJNMMI Radiopharmacy and Chemistry (2022) 7:21 Page 23 of 55 targets must include measures to limit the ingrowth of 154Eu by either tuning their irra- diation parameters to minimise the cross section of the 153Eu(n,γ)154Eu nuclear reaction and maximise the production yield of 153Sm, or incorporating a post-irradiation purifica- tion process (Voorde et al. 2019). Due to Eu impurities being a significant issue for 153Sm production from both natural and enriched targets, purification measures for 153Sm from both types of target are similar in their conception. Removal of these impurities from the desired product have been investigated using ion-exchange chromatography, electrochemical separation, and solvent extraction. Solvent extraction and ion-chroma- tographic methods of purification are more conventional, while electrochemical separa- tion has gained traction recently (Voorde et al. 2019; Chakravarty et al. 2018). The same study that reported success with purification of 153Sm from natural targets using elec- trochemical separation also compared the same method of purification with enriched targets after irradiation (Chakravarty et al. 2018). The method involves electro-amalga- mation, during which the E u3+ impurities are reduced to their divalent state by means of a mercury pool cathode in an electrolytic cell. The separation occurs by way of dis- solution of the Sm2O3 target (natural or enriched) in 0.1  M HCl, then transfer of the dissolved target to the separation solution of 0.15  M lithium citrate (which assists in the prevention of hydroxide precipitation) in the electrolytic cell. The electrolytic pro- cess, using a constant applied voltage of 6 V, reduces the E u3+ after which transfer of the reduced Eu2+ ions to the mercury cathode occurs quickly, leaving behind an electrolyte solution of 153Sm after the mercury cathode is drained and filtered off (Chakravarty et al. 2018). This method has displayed its feasibility and applicability for purification of both natural and enriched targets, with > 85% purified 153Sm recovery for both target types in addition to high radiolabelling yields (> 98%) and radiopurities (Chakravarty et al. 2018). The loss of 10–15% 153Sm from the recovery was attributed to the reduction of S m3+ to its less stable divalent form S m2+ and can be amalgamated into the mercury cathode in a similar way to the Eu impurities (Chakravarty et al. 2018). This procedure was found to be suitable for large-scale 153Sm purification and the resultant 153Sm was deemed appropriate for radiolabelling purposes (Chakravarty et al. 2018). Other methods involv- ing solvent extraction or ion exchange chromatography have been employed to exploit the differences in coordination behaviour of Eu and Sm (Islami-Rad et al. 2011; Jelinek et al. 2008, 2007; Schwantes et al. 2008; Peppard et al. 1962). An ion exchange chroma- tography method, developed in a similar way to that of the previously discussed 166Ho purification via cation-exchange HPLC, involves preferential complexation of the Eu impurities with α-HIBA and elution prior to Sm recovery. The method consists of target dissolution in 1 M HCl and subsequent separation on a Dowex-50 W cation-exchange resin with 0.2  M α-HIBA pH-adjusted to 4.8 as the mobile phase with a flow rate of 1 mL/cm2/min (Islami-Rad et al. 2011). In a similar manner to the separation of 166Ho and 166Dy, the smaller ionic radius and higher charge density resulting from the lanth- anoid contraction leads to a preferential coordination of Eu by α-HIBA due to higher thermodynamic stability and a lower affinity for the stationary phase, and consequently the Eu elutes first, followed by the selective elution of the purified 153Sm. Recoveries of 153Sm were not as high as those reported for electrochemical separation (> 66% vs. > 85% respectively), but radiopurity was > 99.8% (Islami-Rad et  al. 2011). Improvements in recovery yields and purities of the isolated 153Sm have been achieved by the inclusion Sadler et al. EJNMMI Radiopharmacy and Chemistry (2022) 7:21 Page 24 of 55 of a Eu3+ reduction step prior to solvent extraction or ion exchange chromatographic separation. Reducing the contaminant E u3+ to its divalent state allows for the leverag- ing of the significantly different chemical properties and separation/extraction behav- iour exhibited by Eu2+ in comparison to Sm3+. The additional electron present in the 4f orbital of Eu2+ (4f7) results in an overall decrease in charge density which arises from the increase in the atom’s ionic radius, and hence imparts different chelation and separation kinetics compared to S m3+. Certain ion exchange chromatography and solvent extrac- tion methods have incorporated a prior reduction step into procedures that use HDEHP, as the extractant, which allows for selective chelation of Sm3+ ions while leaving behind the Eu2+ impurities (Jelinek et al. 2008, 2007; Schwantes et al. 2008; Peppard et al. 1962). The selective reduction of E u2+ prior to separation results in a much lower affinity for the chosen extractant resin, and thus the E u2+ is eluted first, with concentrated HCl being used subsequently for desorption and final elution of desired Sm (Jelinek et  al. 2008, 2007). However, it has been noted that back-oxidation of E u2+ to E u3+ caused by dissolved oxygen in the system or by photooxidation is an issue with such approaches. A recent study has also investigated the potential of utilising a quaternary ammonium ionic liquid, Aliquat 336 nitrate, for the selective extraction and purification of high purity medical 153Sm (Voorde et al. 2018). The study used stable isotopes of Sm and Eu, with the view that the results obtained regarding the feasibility of the method would be indicative of the results to be expected when using their radioactive counterparts. The procedure involved an aqueous feed solution of both Sm and Eu (each 1 g/L) with either a high nitrate or chloride concentration (using inert salts of N O −3 or C l− respectively) which was mixed with a large excess of Zn granules to selectively reduce the Eu3+ to Eu2+. The aqueous phase was then added to the organic phase containing Aliquat 336 nitrate and both were purged with nitrogen to prevent dissolved or atmospheric oxy- gen from back-oxidising the Eu2+. The phases were then mixed to facilitate extraction, centrifuged to allow swift phase disentanglement, and the aqueous phase was separated and analysed by inductively coupled plasma optical emission spectrometry (ICP-OES) for Sm and Eu concentrations and separation factors were calculated (Voorde et  al. 2018). A subsequent back-extraction of Sm from the ionic liquid phase by diluting the aqueous phase with water gave the final purified product (Voorde et al. 2018). Results from a battery of experiments with varied mixing times and temperatures concluded that S m3+ was efficiently extracted into the Aliquat 336 nitrate phase much more readily than Eu2+ when nitrate aqueous media are used, leaving behind the Eu2+ in the aque- ous phase where it remains sufficiently stable (Voorde et al. 2018). This is due to the fact that Ln3+ ions have been shown to form stable complexes with bidentate nitrate ligands, while Ln2+ ions (or any other M2+ ions) are unable to do so (Vander Hoogerstraete and Binnemans 2014; Larsson and Binnemans 2015, 2017). High nitrate salt concentrations in the aqueous phase allowed for the S m3+ ions to be salted out into the ionic liquid more readily, due to changes in ion hydration and activity that allowed for bidentate binding of Sm3+ by the nitrate anions, as evidenced by poorer separation factors when high chloride concentrations were used in the aqueous feed solution due to C l− ions having higher hydration energies (Voorde et  al. 2019). Furthermore, it was also found that both Eu2+ and Zn2+ from the chemical reduction could be simultaneously removed from the desired Sm if a high nitrate aqueous phase was used rather than chloride, due S adler et al. EJNMMI Radiopharmacy and Chemistry (2022) 7:21 Page 25 of 55 to the similar extraction behaviours of Eu2+ and Zn2+ (Voorde et al. 2018). High separa- tion factors were achieved by this method in a short time frame. While results for this approach appear promising, further method optimisation is required before its applica- bility can be successfully translated to large scale 153Sm production for clinical purposes. The study also investigated the efficacy of adding a size selective extraction agent (0.05  M dicyclohexano-18-crown-6, DCH18C6) to the ionic liquid phase for selective chelation of E u2+ in a similar manner to methods used for Sr2+ chelation. This modified approach resulted in poorer separation factors regardless of the anion used for high salt concentration in the aqueous feed solution (Voorde et al. 2018). Furthermore, it has been suggested that improvements could be made to refine the procedure (Voorde et  al. 2019, 2018). The introduction of Z n2+ contaminants from the chemical reduction process could be avoided by using electrolysis for reduction of Eu3+, and cation exchange chromatography could be used for removal of remaining salts and isolation of Sm after back extraction, after which the purified Sm solution could be reduced to dryness and the Sm redissolved in an appropriate radiolabelling solution (Voorde et al. 2019). Precipitation via hydrolysis of Sm in an alkaline medium followed by redissolution has also been proffered as possible alternative (Voorde et  al. 2019). These approaches still require further exploration to determine their efficacy. While focus on more conventional post-irradiation separation and purification of 153Sm has been significant, investigation into new ways of increasing the specific activ- ity has also gained traction. In a recent novel approach, a proof-of-concept method was demonstrated for the production of very high specific activity 153Sm by using neutron bombardment in a high flux reactor in tandem with off-line mass separation (Voorde et al. 2021). This method involved irradiating highly enriched [ 152Sm]Sm(NO3)3 targets converted from [ 152Sm]Sm2O3 (98.7%) with neutron beam energies of 2.0–2.5 × 1014 n/cm2/s, and subsequent mass separation with laser resonance enhanced ionisation to dramatically increase the specific activity of the product radionuclide (Voorde et al. 2021). Mass separation efficiencies achieved were 4.5% on average with a maximum of 12.7%, and the 153Sm underwent radiochemical processing and post-purification using DGA extraction chromatography, ion exchange chromatography with α-HIBA and extraction chromatography with LN3 extraction resin to give the final [ 153Sm]SmCl3 formulation that was deemed suitable for radiolabelling. Specific activities as high as 1.87 TBq/mg were achieved at the time of mass separation collection after radio- chemical processing. Near-quantitative yields were achieved for the radiolabelling of S-2-(4-isothiocyanatobenzyl)-1,4,7,10-tetraazacyclododecane tetraacetic acid (p-SCN- Bn-DOTA, see Fig. 8) and the method demonstrated the potential for TBq activites of clinical-grade 153Sm to be produced for radiolabelling purposes (Voorde et  al. 2021). This approach is quite new, and while results are promising, further optimisation of the mass separation protocol to increase the separation efficiency and greater availability of high flux neutron reactors are required before widespread application of this method is feasible. Sadler et al. EJNMMI Radiopharmacy and Chemistry (2022) 7:21 Page 26 of 55 O O O N N O 153Sm O N N NCS O O O Fig. 8 Molecular structure of [153Sm]Sm-p-SCN-Bn-DOTA Erbium: 169Er, 165Er 169Er With the widespread clinical application of 177Lu as a higher-energy β−-emitter for the treatment of various tumours, including neuroendocrine tumours (Strosberg et al. 2017), and prostate cancer (Das et al. 2016), interest has also been directed towards the search for alternative β−-emitters of medium or lower-energy and shorter tissue pene- tration lengths. Radionuclides exhibiting these characteristics have been suggested to be better suited to targeting smaller or more localised metastases that have not adequately responded to higher-energy β−-therapy using radionuclides such as 177Lu. Shorter tissue penetration lengths and lower-energy β−-emissions result in greater tumour-to-normal- tissue radiation dose absorption, far less collateral damage to healthy tissue surrounding the tumour site while also reducing the radiation dose burden to patients, which allow for a more personalised approach to cancer treatment based on individual patient cir- cumstances and a tailored approach to cancer therapy which can greatly improve patient prognoses (Uusijärvi et al. 2006; Talip et al. 2021). As such, focus on the production of radionuclides that fit these criteria has been directed towards 169Er as a potential alter- native to 177Lu due to its “softer” decay characteristics, particularly regarding bone pain palliation applications as 177Lu can cause bone marrow suppression (Talip et  al. 2021; Kratochwil et al. 2019; Farahati et al. 2017; Bouchet et al. 2000). 169Er fulfils the afore- mentioned characteristics, with an acceptable half-life (t1/2 = 9.4  d) for transportation requirements and therapeutic applications, maximum β−-emission energies of approxi- mately 350  keV and mean tissue penetration ranges of 0.2–0.3  mm, and with essen- tially no γ-emissions, establishing its advantages over higher-energy β−-emitters with accompanying γ-emissions which can increase patient dose burden and dosimetry issues (Uusijärvi et al. 2006). To date, 169Er has been primarily used for radiation synovectomy of smaller joints such as those in the fingers (Knut 2015; Karavida and Notopoulos 2010), while it has also been considered for applications in bone pain palliation of skeletal metastases arising from lung, breast or prostate cancer (Bouchet et al. 2000). The seem- ingly limited range of applications for this radionuclide has been hamstrung by insuf- ficient production methods that can only provide 169Er in carrier-added form and with low specific activity. Despite only low specific activities being able to be achieved with current production methods, these specific activities have been deemed appropriate for radiation synovectomy, where 169Er is used in a citrate colloid form or in hydroxyapa- tite (HA) form, and lower specific activities are suitable for therapeutic use (Farahati S adler et al. EJNMMI Radiopharmacy and Chemistry (2022) 7:21 Page 27 of 55 et al. 2017; Chakraborty et al. 2014). However, the attractive radionuclide characteris- tics exhibited by 169Er have led to suggestions of targeted RIT and radionuclide therapy applications through incorporation of 169Er into tumour receptor, antigen-targeting or bifunctional chelator targeting vectors, with the potential for use as a theranostic pair with 68 Ga or, more significantly, 44Sc due to its similar chelation kinetics and ability to form stable complexes with bifunctional chelators already well-established in the litera- ture (Knapp and Dash 2016; Formento-Cavaier et al. 2020). These radionuclide applica- tions, however, require much higher specific activities of 169Er to be produced on a much larger scale for radiolabelling, pre-clinical and clinical use (Knapp and Dash 2016). Production of 169Er has been conventionally achieved by means of neutron irradiation of both natural and highly enriched Er targets, with the level of activity being closely related to the enrichment of the target. The natural composition of E r2O3 consists of 162Er (0.14%), 164Er (1.61%), 166Er (33.6%), 167Er (22.95%), 168Er (26.8%) and 170Er (14.9%), and with a low neutron capture cross-section of approximately 2.3 barns the prospec- tive yield from the desired 168Er(n,γ)169Er nuclear reaction is relatively poor, resulting in a low specific activity carrier-added product that is not suitable for radiolabelling purposes for which it has been suggested (Knapp and Dash 2016; Mughabghab and Mughabghab 2018). Production of carrier-added 169Er using natural targets has largely been superseded in favour of highly enriched [168Er]Er2O3 targets, due to the fact that 163Er, 165Er, 171Er and 171Tm impurities are co-produced in significant amounts when using natural targets and the low percentage of the desired 168Er target isotope (approxi- mately one quarter of the natural abundance) results in low specific activity of 169Er (Knapp and Dash 2016; Chakravarty et  al. 2014). Highly enriched [168Er]Er2O3 targets have allowed for a reduction in certain isotopic impurities being produced, however the low neutron capture cross section of the desired nuclear reaction still poses issues for production methods. In order to circumvent issues surrounding the low neutron capture cross section, long irradiation times of several weeks are required to achieve sufficient activity, and post-irradiation processing and purification methods have been employed to improve the specific activity of the 169Er product (Voorde et  al. 2019; Chakravarty et al. 2014). Progress towards higher specific activity formulations of 169Er suitable for pre-clinical and clinical purposes has been achieved with a focus on post-purification regimens due to the carrier-added nature of the conventional 168Er(n,γ)169Er neutron bombardment production method, with promising results indicated by the use of electrochemical sepa- ration and purification (Chakravarty et al. 2014). Issues noted with the use of enriched [168Er]Er2O3 targets and neutron bombardment have been raised regarding the co-pro- duction of 169Yb as an impurity, due to the presence of trace amounts of Yb in the target and the relatively high neutron capture cross section of the 168Yb(n,γ)169Yb nuclear reac- tion (2300 barns) (Chakravarty et al. 2014). Consequently, the removal of this impurity is paramount for the purity of the product, as well as minimising the prospect of dosimetry and dose burden issues due to the long half-life (t1/2 = 32 d) and γ-emissions character- istic of 169Yb decay (Chakravarty et al. 2014). Separation of 169Yb from the desired 169Er was investigated using electro-amalgamation by means of selective reduction of Y b3+ to Yb2+ and preferential electrochemical deposition onto a mercury pool cathode, similar to that reported for the separation of Eu impurities from 153Sm (Chakravarty et al. 2014). Sadler et al. EJNMMI Radiopharmacy and Chemistry (2022) 7:21 Page 28 of 55 The requisite 169Er was produced in a nuclear reactor by means of the 168Er(n,γ)169Er nuclear reaction using highly enriched [ 168Er]Er2O3 target material (98.2% 168Er) and a thermal neutron flux of approximately 8 × 1013 n/cm2/s for an irradiation period of 3 weeks, culminating in an activity level of ~ 3.7 GBq of 169Er with ~ 30 MBq of 169Yb and a small amount of 171Tm as impurities at the end of bombardment. The irradiated target is then cooled and dissolved in 1  M HCl, followed by evaporation and reconstitution in de-ionised water, after which the 169Er/169Yb mixture is dissolved in 0.15 M lithium citrate buffer [to prevent the formation of Yb(OH)3] and pH-adjusted to 6–7 with 0.1 M NH4OH and stable YbCl3 is added as a source of carrier ions to achieve the minimum concentration of Yb3+ (0.6–0.8 mM) required to increase the amalgamation kinetics and ensure complete removal of 169Yb (Chakravarty et al. 2014). An applied potential of min- imum 8 V (current = 500 mA) for a minimum of 15 min was found to be sufficient for complete removal of 169Yb, and the electro-amalgamation process was performed twice on the electrolyte solution. The first electro-amalgamation was shown to reduce the Yb impurities to < 0.1% with < 5% loss of 169Er activity, while repeating the process ensured quantitative removal of 169Yb, confirmed by the absence of γ-photopeaks correspond- ing to 169Yb in the HPGe γ-spectra of the purified 169Er solution. This method resulted in the production of radiopure 169Er at ~ 3.7  GBq levels with a yield of > 95% deemed suitable for radiolabelling of hydroxyapatite (HA) for radiation synovectomy and 1,4,7,10-tetraazacyclododecane-1,4,7,10-tetramethylenephosphonic acid (DOTMP, see Fig. 9) for bone pain palliation purposes, with radiolabelling yields of 98.7% and 99.1% respectively and with radiochemical purity > 99% (Chakravarty et al. 2014). Such applica- tions, however, allow for the use of lower specific activity formulations of 169Er, for which the results reported were deemed appropriate. This method highlights the potential for greater specific activity 169Er to be produced free from isotopic impurities, however, fur- ther improvements in specific activity are required for the translation of this method of 169Er production and purification into theranostic applications in receptor-targeted radionuclide therapy or RIT. Motivation to improve the specific activities attainable from the 168Er(n,γ)169Er nuclear reaction and subsequent processing has led to directed efforts to circumvent the issue of the very poor neutron capture cross-section. Investigations into methods of increas- ing specific activity prior to radiochemical separation and purification have proved promising, with substantial increases in specific activity reported using mass separation procedures after neutron irradiation of enriched [168Er]Er2O3 targets (Talip et al. 2021; O O O P O - O P N N O 169Er O N N - P O O P O O - O Fig. 9 Molecular Structure of [169Er]Er-DOTMP S adler et al. EJNMMI Radiopharmacy and Chemistry (2022) 7:21 Page 29 of 55 Formento-Cavaier et al. 2020). Due to the issue of the carrier-added form of 169Er that is produced by the neutron bombardment approach, subsequent separation of 169Er from the 168Er target has proved challenging due to their chemically identical nature. Whereas conventional chemical separation techniques permit separation of impuri- ties such as 169Yb and 171Tm due to minute differences in chelation chemistry, kinet- ics or electrochemical reduction properties, other separation techniques such as mass separation have been identified as promising avenues of inquiry to selectively separate target radionuclei from their carriers (Talip et al. 2021; Formento-Cavaier et al. 2020). The principle of mass separation has been employed using both electromagnetic mass separation using surface ionisation, and resonant laser ionisation (Talip et al. 2021; For- mento-Cavaier et  al. 2020). One reported study used offline electromagnetic isotope mass separation at the CERN-MEDICIS facility after high flux neutron irradiation of an enriched [168Er]Er2O3 target (~ 7 mg) for 6.5 days at a neutron beam energy of 1.3 × 1015 n/cm2/s, resulting in a calculated specific activity of 3.7 GBq/mg at the end of bombard- ment. After transfer of the irradiated sample to the CERN-MEDICIS facility, the calcu- lated specific activity at the time of separation was 1.3 GBq/mg (Formento-Cavaier et al. 2020). The sample was dissolved in 1 M H NO3 and heated to dryness to deposit the Er sample as its nitrate salt on a graphite holder in a Re boat. The holder in positioned in the mass separator and slowly heated under vacuum to 2000–2100 °C to release the Er ions which are then mass separated and the m/z 169 ion beam is selectively collected by means of implantation into a Zn-coated Au foil (Formento-Cavaier et al. 2020). The mass separation procedure improved the ratio of target 168Er to desired 169Er by a fac- tor of approximately 200, as the ~ 2000:1 168Er:169Er ratio was decreased to ~ 10:1 with a consequent improvement in specific activity from ~ 1.3 GBq/mg to ~ 240 GBq/mg. This proof-of-principle experiment resulted in the production of a usable dose of ~ 17 MBq of 169Er from ~ 9 GBq present at the time of mass separation, and this study represents the first production of very high specific activity 169Er (Formento-Cavaier et al. 2020). Despite the improvements in specific activity exemplified by this method, the mass sepa- ration efficiency was only ~ 0.2% which is unsuitable for large-scale production purposes and requires improvement. It was posited that reduction of residual 168Er in the final product could be achieved by optimising the mass separator slit position to minimise the tail of the m/z 168 peak overlapping with the desired m/z 169 peak. Isotopic impurities of isobaric 169Yb were observed when the implanted foil was analysed at the end of bom- bardment, at ~ 0.2% of the atoms implanted, corresponding to an activity of ~ 10  kBq. This study did not proceed with post-purification processes, which would allow for the removal of this impurity. It was also noted that a higher ionisation efficiency would be expected using laser resonant ionisation instead of surface ionisation, and a publica- tion detailing results from a stable Er experiment investigating this ionisation method is forthcoming (Formento-Cavaier et al. 2020). Improvements to this method have shown significant promise regarding the prospect of high specific activity 169Er being feasibly produced on those scales necessary for preclinical and clinical applications in receptor- targeted radionuclide therapy and theranostics, however separation efficiency remains the largest bulwark to its implementation (Formento-Cavaier et al. 2020). Building on this proof of concept, further progress has been made regarding the use of mass separation techniques for the production of high specific-activity 169Er in Sadler et al. EJNMMI Radiopharmacy and Chemistry (2022) 7:21 Page 30 of 55 combination with radiochemical separation procedures by employing resonant laser ionisation mass separation and chromatographic techniques for purification (Talip et al. 2021). This particular study used the newly established MEDICIS’ Laser Ion Source Assembly (MELISSA) for the offline mass separation of 169Er from target 168Er. A series of enriched [ 168Er]Er2O3 targets (7.9–14.2 mg) ware irradiated with high flux neutrons of ~ 1.1 × 1015 n/cm2/s for a period of 7 days, after which they were transferred for mass separation at CERN-MEDICIS. The irradiated samples were transferred to a Ta target container, connected to a Re ion source and heated to 2200 °C for a two-stage laser reso- nance ionisation using two Ti:sapphire lasers each set to an optimised wavelength for selective 169Er ionisation (laser 1: 24943.95  cm−1, laser 2: 24337.32  cm−1) (Talip et  al. 2021; Studer et al. 2016; Lassen et al. 2017). The 169Er ion beam (m/z 169) arising from the resonant laser assisted ionisation was selectively mass separated using a magnetic sector field and deposited onto a Zn-coated Au foil catcher, and seven mass-separated samples were produced for subsequent radiochemical separation. The mass-separated samples were dissolved along with their Zn coated Au foil catchers in 6 M nitric acid and the resultant solution was separated on a DGA resin column by rinsing with 6 M nitric acid to selectively elute the majority of the Zn, followed by elution with 0.05 M HCl of the desired 169Er along with other lanthanoid contaminants (169Yb and 171Tm) that were co-produced during the neutron bombardment stage (Talip et al. 2021). This eluate was then loaded onto a Sykam macroporous cation exchange resin column and 169Yb was separated from 169Er using 0.06–0.08  M α-HIBA as the complexing agent before a third separation on LN3 resin to remove the α-HIBA and 0.02  M HCl was used to remove residual Zn. 169Er was subsequently eluted in 1 mL 2 M HCl and passed through a final separation on TK200 resin to ensure total removal of Zn from the final formulation which was evaporated and reconstituted in a 250 μL 0.05 M HCl solution (Talip et al. 2021). This method resulted in the production of 169Er activities that were four times higher than those achieved using the surface ionisation method reported pre- viously, and the separation efficiency achieved was 0.5% when resonant laser ionisation was used (c.f. ~ 0.2% for surface ionisation) (Talip et  al. 2021; Formento-Cavaier et  al. 2020). Activity measurements of the samples ranged from 4.70 to 73.2 MBq and radio- nuclidic purities of > 99.9% were confirmed upon analysis of four of the seven samples collected. 168/169 isotope ratios were measured on the same four samples with a range of 1.60–14.62 with increases in the ratio being attributed to the tail of the mass 168 ion beam. This presented a substantial improvement over the surface ionisation method that reported 168Er:169Er ratios of ~ 10:1 (Talip et al. 2021; Formento-Cavaier et al. 2020). The activities and purities obtained were deemed suitable for the radiolabelling of PSMA-617 for subsequent use in a preliminary in vitro cell viability study, and radio- chemical and chemical purities of 169Er were evaluated via radiolabelling of PSMA-617. [169Er]Er-PSMA-617 was prepared at 10 MBq/nmol concentration with a radiochemi- cal purity > 98% and compared to the clinically-established [177Lu]Lu-PSMA-617 in a tumour cell viability assay using PC-3 PIP tumour cells, where results indicated reduced tumour cell viability of ~ 89% and ~ 69% at 5  MBq/mL and 10  MBq/mL respectively. These results were overshadowed by the greater reduction in cell viability displayed by 177Lu at the same concentration, as well as at lower concentrations of 1–2.5 MBq/mL, however it was noted that the assay could only be performed once due to the limited S adler et al. EJNMMI Radiopharmacy and Chemistry (2022) 7:21 Page 31 of 55 availability of the 169Er isolated from the method. Moreover, it was noted that the rela- tively low molar activity of the prepared radioligand likely affected tumour receptor saturation, and that further preclinical results using higher molar activity preparations of [169Er]Er-PSMA-617 would be needed to properly ascertain its therapeutic potential, though the initial results are promising (Talip et al. 2021). In order to obtain higher activ- ities necessary for more extensive preclinical studies, efficiencies of the resonant laser ionisation mass separation method need to be significantly improved from 0.5% to at least 20%, commensurate with efficiency results reported for other mass separated radi- olanthanoids (Talip et al. 2021; Studer et al. 2016; Kieck et al. 2019). Additionally, mass separator optimisation for high neutron flux irradiation protocols and isotope-selective laser ionisation schemes are potential developments to achieve higher 169Er activities and mass separation efficiencies, and longer irradiation times with minimal activity loss during separation, purification and transportation have been noted as potential ways to achieve higher 169Er yields on larger production scales (Talip et al. 2021). 165Er Research efforts surrounding radiolanthanoid applications in theranostics have also focussed on potential radionuclides suitable for more targeted therapies of smaller metastases, tumours and disseminated cancers that require shorter particle path lengths and more localised energy deposition to reduce collateral damage to healthy tissues. As such, Auger electron emitters have received attention due to their favourable decay characteristics of short particle path lengths (1 nm up to ~ 5 µm) and high linear energy transfer (LET) which result in the emitted particles depositing their energy in the range of 4–26 keV/µm, suitable for highly localised targeting of important cell structures such as the nucleus or mitochondria (Kassis 2008, 2004). Auger electrons (AEs) are emitted as the result of decay by means of electron capture or internal conversion, and multiple AEs can be emitted during one nuclear decay event (Buchegger et al. 2006). These AEs are capable of damaging DNA by way of direct interaction causing double strand breaks (Martin et al. 1988), or via the formation of radicals from the hydrating water molecules surrounding the DNA macromolecule (Nikjoo et al. 1996). Consequently, Auger emit- ters must be highly targeted and lie in very close proximity to the important cellular structures for effective treatment (Kassis 2004; Ramogida and Orvig 2013; Bousis et al. 2010). One such radionuclide that has attracted attention as a potential Auger emitter for radionuclide therapy and theranostic applications is 165Er, due to its favourable half life ( t1/2 = 10.4 h) and electron capture decay mechanism resulting in AE emissions of 5.3  keV (65.6%) and 38.4  keV (4.8%), with accompanying low-energy X-rays (average energy 48.8 keV) (Talip et al. 2021; Gracheva et al. 2020). In addition, the absence of any γ-emission during 165Er nuclear decay corresponds to a lower dose burden for patients (Gracheva et al. 2020). Interest in this radiolanthanoid has also been bolstered by its AE emissions having similar energies to those of 125I (t1/2 = 59.4 d), which has been stud- ied extensively for its DNA cytotoxicity due to its AE emission properties (Gracheva et al. 2020; Yasui et al. 2001). As with 169Er, 165Er also has the potential to be paired with companion diagnostic radionuclides such as 68 Ga or, in particular, 44Sc for theranostic applications due to similarities in complexation kinetics afforded by the similar chemi- cal properties of the Rare Earth metals (Cotton 2006; Moeller et  al. 1965). Moreover, Sadler et al. EJNMMI Radiopharmacy and Chemistry (2022) 7:21 Page 32 of 55 the need for pre-clinical studies to elucidate any additional therapeutic effect of Auger therapy in combination with β−-therapy through administration of both 169Er and 165Er in different ratios has been highlighted as an interesting avenue of inquiry (Talip et al. 2021). There have been a number of nuclear reaction routes reported for the production of 165Er, typically using protons, deuterons or neutrons as the bombardment particles with either natural or enriched Er targets, or natural Ho targets (100% natural abundance of 165Ho) (Sadeghi et al. 2010; Zandi et al. 2013; Beyer et al. 2004b; Tárkányi et al. 2007, 2008a, b, c, 2009; Hermanne et al. 2013). The use of natural Er targets utilises the gen- erator approach to produce ingrown 165Er after the β−-decay of 165Tm produced via either the natEr(p,xn)165Tm → 165Er or the natEr(d,xn)165Tm → 165Er nuclear reactions, giving the resultant 165Er in no-carrier-added form after separation from the target material (Tárkányi et  al. 2007; Vaudon et  al. 2018). These charged particle bombard- ment approaches on natural Er have resulted in the production of reasonably long-lived 167Tm (t 1681/2 = 9.25 d) and Tm (t1/2 = 93.1 d) impurities, and the natural abundance of the main contributing 166Er isotope being only ~ 33.5% has led to enriched 166Er targets being preferred (Vaudon et al. 2018). Furthermore recently, alpha particle bombardment has been investigated via the natEr(α,x)165Tm → 165Er route, but was deemed not practi- cal when compared to proton and deuteron bombardment due to low yields of 165Tm (Aliev et al. 2021; Kormazeva et al. 2021). Enriched 166Er targets for the 166Er(p,2n)165Er nuclear reaction are expensive and have therefore led to shifts in research focus to 165Ho targets paired with proton and deuteron bombardment, with subsequent chemical sep- aration (Tárkányi et  al. 2008b, c). These production methods have been favoured due to the fact that they exhibit lower target material costs due to the 100% natural abun- dance of 165Ho, and the feasibility of the 165Ho(p,n)165Er and 165Ho(d,2n)165Er nuclear reactions which can be achieved at commercial cyclotrons already in operation using proton and deuteron beams of lower energies (< 20 MeV) (Zandi et al. 2013; Tárkányi et al. 2008b, c; Vaudon et al. 2018). Of these two nuclear reactions, the deuteron reac- tion has been reported to produce larger amounts of 165Er activity at low-energy cyclo- trons, but with the trade-off of higher amounts of 166Ho as an impurity, with 165Er/166Ho ratios of ~ 400/1 c.f. 8/1 for proton and deuteron irradiation respectively (Tárkányi et al. 2008c; Vaudon et al. 2018). A noteworthy advantage of lower-energy irradiation beams is that only 165Er is produced when low-energy protons are used, however lower-energy deuteron beams can result in stable 166Er being produced on the 165Ho targets in addi- tion to 165Er which gives a no-carrier-added but not carrier-free product (Tárkányi et al. 2009). In contrast, at higher energies (> 30  MeV), natEr(p,xn)165Tm → 165Er is the pre- ferred reaction (Tárkányi et al. 2008a, 2009). Further challenges arise in the separation of 165Er from the 165Ho target, with separation factors for the Er/Ho pair being among the lowest for lanthanoid separations (~ 1.5), and greater improvements are thereby required for further practical applications of 165Er to be attainable. Certain separations of 165Er and 165Ho have centred around LN2 resin and nitric acid as the eluent (similar to previous methodologies for radiolanthanoid separations), however it has been reported that the reduced adsorption capacity of LN2 resin limits the amount of 165Er that can be successfully separated (Vaudon et al. 2018). Further investigation into optimisation of post-purification has occurred more recently, with the production of 165Er by means S adler et al. EJNMMI Radiopharmacy and Chemistry (2022) 7:21 Page 33 of 55 of the proton irradiation route at 8.6–16.2 MeV beam energies on 165Ho2O3 targets at low-energy cyclotrons (Gracheva et  al. 2020). In this study, an irradiation and separa- tion protocol for larger amounts of 165Ho target material (up to 200 mg 165Ho2O3) was developed for the isolation of 165Er suitable for radiolabelling and in vivo applications. The maximum cross section of ~ 180 mb for the 165Ho(p,n)165Er nuclear reaction was achieved at a beam energy of 10.3 MeV, and cross section measurements over the range of 6.7–18.2 MeV were in good agreement with those previously reported (Tárkányi et al. 2008b). Activities of 165Er up to 1.6 GBq were produced via irradiation of a 10 μA tar- get pellet at the optimum beam energy of ~ 13.4 MeV (determined from shorter irradia- tions and yield measurements) for 8–10 h, and the co-production of 163Er was avoided by ensuring that an impinging energy lower than the 165Ho(p,3n)163Er threshold energy [17 MeV—TALYS-based evaluated nuclear data library 2017 (TENDL-2017)] was used. Subsequent separation of 165Er from the target material employed a two-column method of cation-exchange chromatography on Sykam resin using 0.08 M α-HIBA followed by extraction chromatography on a LN3 column with 0.1 M HCl for final formulation con- centration. A hot cell was used in the separation module to allow for higher activities of 165Er to be used, and 165Er was separated from the target Ho and co-produced Tm impurities in 7 h and eluted in 500 µL 0.1 M HCl in its final formulation. Five separations were performed and the radionuclidic purity of the [ 165Er]ErCl3 product was ascertained by measuring the characteristic 47.6 and 46.7 keV X-ray energies of 165Er using a HPGe detector (> 99.9% 165Er with no other detectable radionuclide impurities) (Gracheva et  al. 2020). ICP-OES was used on two of the purifications to calculate the natural Er content of the product, with natEr/165Er ratios measured to be 60/1 and 51/1 respec- tively, indicating that the product was not carrier free and consequently not suitable for DOTA-NOC radiolabelling. The presence of natEr in the final formulation was attributed to impurities in the Ho target material, as the presence of certain Tm impurities (namely 166Tm, 167Tm and 168Tm) formed from (p,n) reactions on natEr isotopes also suggested this. To rectify this issue, separation of natural Er impurities from the 165Ho2O3 target prior to irradiation was required, and a Ho target recycling procedure was developed which involved increasing the α-HIBA concentration to 0.11 M and eluting the 165Ho2O3 target separately after 165Er elution. Nitric acid (1.0 M) was added to the Ho eluate and the resultant solution was separated on another column using AG MP-50 macroporous cation exchange resin. The 165Ho was eluted as its nitrate salt in 40 mL 7.0 nitric acid, evaporated to dryness, and finally converted to its oxide by heating at high tempera- ture (> 600  °C) (Gracheva et  al. 2020). This recycling methodrecovered 845  mg of Ho target material, and subsequent irradiations of the recycled target material with post- purification to evaluate the efficacy of target recycling in removing target contaminants prior to irradiation are forthcoming. Irradiation of recycled Ho targets free from natural Er impurities holds promise for higher specific activity no-carrier-added formulations of 165Er to be produced, as future DOTA-NOC radiolabelling studies using 165Er from recycled targets are on the horizon (Gracheva et al. 2020). While this production and purification process holds promise for 165Er to be produced for preclinical applications, the ~ 10 h purification time results in a significant loss in activity in the final formulation due to the half life of 165Er (t1/2 = 10.4 h), which poses issues for large scale production Sadler et al. EJNMMI Radiopharmacy and Chemistry (2022) 7:21 Page 34 of 55 and clinical applications which require further method development and process opti- misation (Gracheva et al. 2020). Despite the previous intimation that enriched 166Er targets are prohibitively expensive, work has progressed on evaluating the feasibility of both the 166Er(p,2n)165Tm → 165Er and the 166Er(d,3n)165Tm → 165Er cyclotron production methods (Sadeghi et  al. 2010; Zandi et  al. 2013). Calculations have been performed using the ALICE/ASH and EMPIRE-3.1 codes for nuclear reactions to evaluate the excitation functions, yields and target thicknesses that are optimum for the four routes to 165Er, namely 166Er(p,2n)165Tm → 165Er, natEr(p,x)165Tm → 165Er, 165Ho(p,n)165Er and 165Ho(d,2n)165Er, and these have been compared against experimental data obtained from other investiga- tions (Zandi et al. 2013). From these data, it was concluded that the proton irradiation of 166Er for the generator production of 165Er from 165Tm afforded the highest production yield of 165Er in no-carrier-added form (Zandi et al. 2013). The 166Er(p,2n)165Tm → 165Er reaction, for optimum yield of 165Er, requires proton beam energies of 16–23 MeV with a maximum nuclear reaction cross section at 21 MeV, which is readily achievable at low- energy cyclotrons (18 MeV), but yield maximisation would require higher beam energies not available at such facilities (Zandi et al. 2013). For the analogous deuteron reaction on 166Er, a maximum cross Sect. (1523.9 mb) was calculated for deuteron beam energies of 25 MeV with an optimum range of 22–29 MeV, however the co-production of 166Tm (t1/2 = 7.7 h), 164Tm (t1/2 = 2 min) and 163Tm ( t1/2 = 1.81 h) impurities posed purification issues. Furthermore, the required deuteron beam energies are higher than those avail- able at 18 MeV commercial cyclotrons, which necessitates the use of more specialised facilities which are scarce (Sadeghi et al. 2010). The potential for the production of 165Er using an on-line mass separation approach has also been suggested, with a proof-of-principle experiment performed at the ISAC Facility at TRIUMF and the resulting 165Er used for the successful radiolabelling of DOTA (Fiaccabrino et al. 2021). This method implanted a mass 165 ion beam from a Ta foil target into an Al implantation target, and the 165Er and 165Tm activity was retrieved via repeated evaporation and rinsing of the implantation target surface layer with 0.1 M HCl (Fiaccabrino et al. 2021). A post-purification procedure based on LN resin and HCl was used, however complete separation of 165Er from 165Tm was not achieved and the fraction with the highest percentage of 165Er activity (~ 17%) and lowest level of 165Tm impurity was selected for radiolabelling of macropa and DOTA. Macropa radiolabelling was unsuccessful (< 1%) whereas quantitative labelling of DOTA was achieved (> 99%). Despite these results, further improvements in purification are necessary before more substantial radiolabelling and preclinical studies can be conducted. Further work at the ISAC facility regarding yield measurements, radiolabelling and preclinical studies for a number of medically relevant radiolanthanoids, including 165Er, have been planned in the future (Fiaccabrino et al. 2021). Promethium: 149Pm Of the radiolanthanoids currently in use in medical applications today, only three can be readily produced in no-carrier-added form from a different lanthanoid target material (149Pm, 166Ho and 177Lu) (Hu et al. 2002). 149Pm has been considered a promising addi- tion to the library of theranostic radionuclides due to its favourable decay characteristics S adler et al. EJNMMI Radiopharmacy and Chemistry (2022) 7:21 Page 35 of 55 O O O NH2 O N N M O O O O O O O H H H H N N N N N N N N N N N NH H H H H H 2 O O O O O HN O NH S N Fig. 10 Molecular structure of lanthanoid DO3A-amide-βAla-BBN complexes, where M = 149Pm, 153Sm, 177Lu O O O N N O O S M O N NH N N N H O HNO O O O Fig. 11 Molecular structure of lanthanoid DOTA-biotin complexes, where M = 149Pm, 166Ho, 177Lu and reasonable physical half-life (t1/2 = 2.21 d) (Hu et  al. 2002). This radiolanthanoid exhibits appropriate β−-decay characteristics (1.07 MeV (95.9%)) with an accompanying γ-emission (286 keV (2.8%)) which is appropriate for concurrent imaging in theranostic settings as well as for dosimetry and biodistribution evaluation, while also being of low enough energy to avoid dose burden complications in potential clinical applications (Hu et al. 2002; Nieschmidt et al. 1965; Bunney et al. 1960). In addition, the ability for this radionuclide to be produced in no-carrier-added form with high specific activity permits its use in RIT and receptor-targeted radionuclide therapy applications which necessi- tate high specific activities and radiopurity for successful radiolabelling, with preclinical studies reporting promising results for future applications (Hu et al. 2002; Lewis et al. 2004; Mohsin et  al. 2006, 2011). Preclinical work for receptor-targeted radiotherapy applications has seen 149Pm incorporated into DOTA bombesin analogues for in  vivo comparison with [153Sm]Sm- and [177Lu]Lu-DO3A-amide-βAla-BBN complexes (see Fig.  10), with close to identical behaviour being observed across the three complexes with regards to biodistribution (Hu et al. 2002). Pretargeted RIT potential has also been investigated through the preparation of [ 149Pm]Pm-, [ 166Ho]Ho- and [ 177Lu]Lu-DOTA- biotin complexes (see Fig. 11) and pretargeting them to LS174T colorectal tumours in nude mice with the CC49 scFvSA antibody fusion protein, with tumour uptake and bio- distribution of the 149Pm complex performing commensurately with those of the 166Ho and 177Lu complexes, with favourable tumour targeting, body clearance and urinary excretion being reported (Lewis et al. 2004). Further preclinical work in LS174T human colon carcinoma xenograft nude mice models with N-hydroxysulfosuccinimidyl DOTA (DOTA-OSSu) and methoxy-DOTA (MeO-DOTA) complexes of 149Pm conjugated to the CC49 anti-TAG-72 monoclonal antibody has also reported high tumour uptake and favourable biodistribution properties for RIT applications, with MeO-DOTA affording higher conjugate stability and tumour uptake as well as lower kidney retention compared Sadler et al. EJNMMI Radiopharmacy and Chemistry (2022) 7:21 Page 36 of 55 to DOTA-OSSu (Mohsin et al. 2006). Both pretargeted and conventional RIT methods for LS174T colon tumours in nude mice models have been compared for their efficacy using [149Pm]Pm-, [ 166Ho]Ho- and [ 177Lu]Lu-DOTA-biotin and CC49 compounds, with [149Pm]Pm-DOTA-biotin and [ 149Pm]Pm-CC49 compounds both considerably increas- ing median survival over saline controls, significantly inhibiting tumour growth and exhibiting high tumour selectivity with little to no off-target tissue toxicity (Mohsin et al. 2011). In comparison to 166Ho and 177Lu, the 149Pm complexes outperformed the analogous 166Ho complexes in median survival, median time of progression and tumour doubling time, whereas 177Lu was deemed more efficacious than 149Pm (Mohsin et  al. 2011). As such, preclinical studies have reported encouraging results for the application of 149Pm in RIT, and have established 149Pm as a potential alternative to the well-estab- lished 90Y for therapies where shorter tissue penetration ranges, imageable γ-emissions or softer β−-energies are required (Uusijärvi et al. 2006; Mohsin et al. 2011). Radiolabel- ling and preclinical investigation into the in vitro stability, in vivo stability and biodis- tribution of aminocarboxylate and octreotide complexes of no-carrier-added 149Pm in comparison with the analogous 166Ho and 177Lu complexes has also been conducted (Li et al. 2003). Production of 149Pm in no-carrier-added form is readily achieved by means of neu- tron bombardment of enriched 148Nd targets via the 148Nd(n,γ)149Nd nuclear reaction (σth = 2.5 barn) with subsequent β−-decay of 149Nd ( t1/2 = 1.73  h) to 149Pm, potentially allowing for theoretical specific activities of 15  GBq/mg (Uusijärvi et  al. 2006; Nie- schmidt et al. 1965; Bunney et al. 1960; Abrarov and Aminova 1975). The use of highly enriched 148Nd targets ensures the avoidance of coproduction of the reasonably long- lived 147Nd (t1/2 = 11 d) impurity due to the comparable thermal neutron capture cross- section of the 146Nd(n,γ)147Nd nuclear reaction (σth = 1.4 barn). The desired 149Pm is readily obtained via chemical separation from the Nd target after irradiation using extraction chromatography based on HDEHP resin (40% by weight) loaded onto Amber- chrom CG-71 inert polymeric absorbent (60% by weight) as the stationary phase, and HNO3 as the eluent (Monroy-Guzman et al. 2015; Ketring et al. 2003; Monroy-Guzman and Jaime 2015b). In one study, the Nd target was loaded onto the HDEHP/Amber- chrom column after being dissolved in 0.15  M HCl, and the Nd was eluted first due to its slightly lower charge density and ionic radius using 0.5 M H NO3 after which the desired 149Pm was eluted using 5 M HNO3 (Ketring et al. 2003). The 149Pm fractions are then evaporated and reconstituted in a suitable solution for further applications (usually HCl) with high radiopurity. More recently, similar chromatographic methods have been employed for the separation of micro–macro component systems typically encountered when producing no-carrier-added and/or carrier free radioanthanoids (Monroy-Guz- man et al. 2015; Monroy-Guzman and Jaime 2015b). HDEHP resin on an Ambercrhom CG-71 support were again used, however irradiation targets were dissolved in 0.15 M HNO3 for column loading and distribution coefficients and separation factors were calculated in order to optimise the separation of the desired radiolanthanoid from the bulk target material (Monroy-Guzman et al. 2015; Monroy-Guzman and Jaime 2015b). Specifically for the separation of 149Pm from Nd, optimum separation was achieved when 0.18 M H NO3 was used for elution of Nd and 149Pm was stripped from the col- umn with 1.5 M H NO3, resulting in a separation efficiency of 98.4%. The 149Pm was then S adler et al. EJNMMI Radiopharmacy and Chemistry (2022) 7:21 Page 37 of 55 precipitated out from the H NO3 solution as 149Pm(OH)3 through addition of NaOH to the eluate, and the hydroxide was subsequently redissolved in 0.1 M HCl to give [149Pm] PmCl3 in its final carrier free formulation with a radiopurity of > 99.9% (Monroy-Guz- man et al. 2015; Monroy-Guzman and Jaime 2015b). Other irradiation methods have been investigated on natural Nd targets and cross- sections have been measured for the 148Nd(p,γ)149Pm, 150Nd(p,2n)149Nd → 149Pm, 148Nd(d,n)149Pm, and 150Nd(d,3n)149Pm nuclear reactions (Yang et  al. 2015; Tárkányi et  al. 2017; Lebeda et  al. 2014, 2012; Tárkányi et  al. 2014). Measured cross-sections of these reactions are considerably lower than the well-established 148Nd(n,γ)149Nd → 149Pm nuclear reaction, and so have not seen routine use for 149Pm production. Moreover, investigation into the production of 149Pm by photonuclear means has also been undertaken by way of the 150Nd(γ,n)149Nd → 149Pm nuclear reac- tion with multi-stage extraction chromatography techniques for separation (Dikiy et al. 2015). This method used bremsstrahlung energies of up to 12.5  MeV and natural Nd targets. While high specific activities of 149Pm are reportedly producible via this method on a daily basis (0.5 Ci 149Pm per day), the cross-section for the reaction is only 220 mb (c.f. 2.5 barns for 148Nd(n,γ)149Nd → 149Pm) and 147Pm impurities are also produced via the 148Nd(γ,n)147Nd → 147Pm side nuclear reaction which has a comparable cross-section of ~ 200 barns (Dikiy et al. 2015). It has been noted that enriched 150Nd targets are able to increase the daily yield to 10 Ci per day, however the requirement of a linear accel- erator with electron energies of 36 MeV and currents of 260 μA has limited the overall feasibility of this irradiation method for larger scale 149Pm production (Dikiy et al. 2015). Praseodymium: 143Pr Another radiolanthanoid that has been identified for its theranostic potential is the moderate-energy β−-emitter 143Pr (Mishiro et  al. 2019). As a pure β−-emitter (Eβ(max) = 0.93 MeV, Eβ(mean) = 0.315 MeV/decay) with a relatively long half life ( t1/2 = 13.6 d) and the additional advantage of no-carrier-added formulations being obtainable, 143Pr has been proposed for use in radiolabelled antibodies for RIT (Uusijärvi et  al. 2006; Vimalnath et  al. 2005; Humm 1986). Selection of a radionuclide for RIT is contingent upon appropriately matching the physical half-life of the radionuclide to the biological half-life and pharmacokinetic properties of the antibody to which the radionuclide will be labelled (Humm 1986). Due to the fact that antibodies may to take up to several days to accumulate and localise at the therapy site and display slow or delayed clearance from the rest of the body (i.e. other organs or the blood), radionuclides for such therapy must have sufficiently long half-lives as not to decay too rapidly before maximum tumour uptake or localisation of the radiolabelled antibody is achieved. Moreover, the photon abundance of the candidate radionuclide is imperative, since the slower pharmacoki- netic properties of antibodies and longer radionuclide half-lives necessitate longer radio- immunoconjugate exposure times for RIT patients to ensure optimal tumour uptake and body clearance. Consequently, radionuclides with higher photon abundances will result in a higher absorbed dose to the entire body of the patient as the radioimmunoconjugate is metabolised and localised over a longer period of time, whereas lower photon abun- dances alleviate this dose burden to normal body tissue. 143Pr exhibits a reasonably long half-life in conjunction with its medium-energy β−-emission and low photon abundance Sadler et al. EJNMMI Radiopharmacy and Chemistry (2022) 7:21 Page 38 of 55 (p/e < 0.2) which have been deemed to be suitable decay characteristics for potential RIT applications. Furthermore, inquiry into the tumour-to-normal-tissue distribution ratios (TNDs) of potential therapeutic radionuclides has demonstrated that 143Pr exhibits small tumour TND values similar to those of the well-established RIT nuclide 131I, and higher TND values for larger tumours compared to 131I due to lower photon emission which reduces extraneous damage to healthy tissue during uptake and localisation (Uusijärvi et al. 2006). Production approaches for 143Pr have centred around two nuclear reaction schemes (Knapp and Dash 2016; Vimalnath et  al. 2005). The less desirable production route involves neutron irradiation of 141Pr target material (100% natural abundance) by means of the 141Pr(n,γ)142Pr nuclear reaction (σth = 11.49 barns) to produce the inter- mediate 142Pr isotope, which then undergoes a second neutron capture event via the 142Pr(n,γ)143Pr nuclear reaction, thereby exploiting the higher thermal neutron capture cross-section of the intermediate 142Pr (σth = 20.03 barns) (Knapp and Dash 2016). This method gives the desired 143Pr in carrier-added form and typically results in a significant 142Pr impurity being present, the amount of which is highly dependent upon the irradia- tion time and the cooling period post-irradiation. Additionally, the compounding of the relatively low thermal neutron capture cross sections for both nuclear reactions culmi- nates in low production yields for this method. The second production method involves the use of Ce targets and neutron bom- bardment to obtain carrier-free 143Pr by means of the indirect 142Ce(n,γ)143Ce → 143Pr nuclear reaction (σth = 0.95 barns) followed by chromatographic separation of 143Pr from the bulk cerium target (Vimalnath et al. 2005; Kubota 1976; Peppard et al. 1957). Both natural and enriched 142Ce targets have been used for this method, and as with other radionuclide production routes, the use of natural 142Ce targets has resulted in the by-production of impurities (Vimalnath et al. 2005). Natural cerium contains 136Ce (0.185%), 138Ce (0.251%), 140Ce (88.45%) and 142Ce (11.114%), all of which have the potential to undergo neutron bombardment and form a number of impurities in various ratios that depend upon the length of bombardment, neutron flux and cooling period. The most significant impurities formed are 137Ce, 139Ce and 141Ce, with 141Ce (t1/2 = 32.5 d) being a particularly significant impurity due to its longer half-life compared to the desired 143Pr, the high natural abundance of its parent nuclide (140Ce – 88.45% natural abundance) and the comparable neutron capture cross section of the parent nuclide (σ = 0.57 barns) (Vimalnath et al. 2005). 141Ce is also a radiolanthanoid deemed suitable for potential therapeutic purposes, and so the concurrent production of two therapeu- tic radionuclides necessarily increases the emitted radioactivity level of the target mate- rial which can pose handling issues prior to chromatographic separation, and so 142Ce enriched targets are required to minimise the potential radiation dose burden (Knapp and Dash 2016). Separation of 143Pr from irradiated Ce targets has been investigated by means of precipitation followed by alumina column separation or filtration (Vimalnath et  al. 2005), cation exchange chromatography (Kubota 1976), as well as by extraction with dioctyl phosphoric acid (Peppard et al. 1957). More recently, production of no-carrier-added 143Pr from natural Ce(NH4)4(SO4)4 and CeO2 target materials has been reported using neutron beam energies of ~ 1 × 1013 n/ S adler et al. EJNMMI Radiopharmacy and Chemistry (2022) 7:21 Page 39 of 55 cm2/s and an irradiation period of 7 days (Vimalnath et al. 2005). At EOB, 7.4 MBq/mg and 2.3 MBq/mg activities of 143Ce were produced from the CeO2 and Ce(NH4)4(SO4)4 targets respectively, with 139Ce and 141Ce impurities present in non-negligible amounts. After bombardment, the targets were either dissolved in concentrated H NO3 (Ce(NH4)4(SO4)4 target) or concentrated HNO3 with 30% H2O2 (CeO2 target) and were stood for a cooling period of 5  days to allow in-growth of 143Pr from decaying 143Ce. Following this, 1 M NaBrO3 was added to each of the irradiated target solutions with heating at 80  °C for 10  min before the addition of saturated HIO3 and cooling on an ice bath to precipitate out the cerium target as Ce(IO3)4. Two final separation protocols were used to obtain 143Pr, using either an alumina column or filtration. For column sepa- ration, the target solution with precipitate was loaded onto an alumina stationary phase column and 30 mL 1% H IO3 solution was used as the eluent, followed by a subsequent column washing with 20 mL double distilled water after which the collected 143Pr eluate was evaporated and reconstituted as [143Pr]PrCl3 in 0.1 M HCl in its final formulation (Vimalnath et  al. 2005). For filtration after precipitation, the target solution with pre- cipitate was filtered through a G-2 sintered funnel and the 143Pr filtrate was evaporated and reconstituted in 0.1 M HCl as [ 143Pr]PrCl3 (Vimalnath et al. 2005). These separation methods culminated in the isolation of 143Pr activities of 0.9  MBq/mg of CeO2 target and 0.3 MBq/mg of Ce(NH4)4(SO4)4 target, with high resolution gamma spectrometry indicating that potential Ce isotope contamination in the final [ 143Pr]PrCl3 solutions was lower than the detection limit and hence would not affect the final specific activity or radiopurity. The absence of Ce in any appreciable amount was corroborated by negative phosphomolybdic acid, benzidine and H 2O2/NH3 spot tests. The resultant no-carrier- added 143Pr was of high radiopurity and suitable for hydroxyapatite radiolabelling (99.5% radiolabelling efficiency) (Vimalnath et  al. 2005). Despite the no-carrier-added and radiopure nature of the product, issues such as the harsh column conditions, the large amount of Ce target material to separate from the desired 143Pr and the volume of eluent necessitated a separation time of 3–4 h for the column approach, which in addition to the time required for evaporation and reconstitution resulted in protracted purification times which are not desirable for large scale radionuclide production. Furthermore, the Ce(IO3)4 precipitation approach resulted in excess NaBrO3 and NaBr being present in the final [143Pr]PrCl3 formulation, which could reportedly be removed by repeating the evaporation of the residue, however reconstitution with HCl still affords a final formu- lation with significant NaCl, NaI and NaBr chemical impurities which are not ideal for further radiolabelling applications and improvements to the radiochemical separation protocol are required (Vimalnath et al. 2005). Additional improvements to this approach include the use of enriched 142Ce target materials to reduce the co-production of isotopic impurities and improve overall activity yield, and increasing the irradiating neutron flux to mitigate the relatively low neutron capture cross-section of the 142Ce(n,γ)143Ce reac- tion and the long irradiation times required for sufficiently high activity of 143Ce to be produced. Optimising these conditions could allow for the production of up to 4–6 GBq of no-carrier-added 143Pr from each batch (Vimalnath et al. 2005). Separation of target Ce from 143Pr has also been achieved via cation-exchange chro- matographic methods (Kubota 1976). Such methods exploit the inherent differences in charge density and ionic radius of C e4+ compared to trivalent 143Pr to separate the Sadler et al. EJNMMI Radiopharmacy and Chemistry (2022) 7:21 Page 40 of 55 desired radionuclide from the target material. CeO2 is chosen as an ideal target mate- rial due to the Ce being tetravalent and thus more easily separable from 143Pr3+ due to a smaller ionic radius and higher charge density which result in different complexation kinetics and interactions with eluents, extraction resins or stationary phases used in cation-exchange separation. In one study, due to reported issues with the dissolution of CeO2 in HNO3 or HCl media, natural CeO2 was formed from the pyrolysis of natural high purity C e2(CO3)3 in order to improve the solubility of the oxide in H NO3/H2O2 solutions (Kubota 1976). The prepared CeO2 target was subjected to a 7.6 × 1012 n/cm2/s neutron flux for 77 h and left for a cooling period of 1 month, after which it was readily dissolved in concentrated HNO3 with 30% H2O2 and the solution evaporated to dryness and reconstituted in aqueous  HCl. The reconstituted residue was then loaded onto a Diaion SK-1 cation-exchange column which was washed with distilled water. Elution was performed with 0.25 M citrate solution at a flow rate of 27 mL/hr and separation of 143Pr from target Ce and impurities was monitored by γ-spectrometry (Kubota 1976). A sec- ond cation-exchange separation on a smaller column was performed with the 143Pr-con- taining fractions using HCl as the eluent, and the final 143Pr-containing fractions were combined, evaporated and reconstituted in HCl. The elution profile for the separation showed the presence of overlapping 147Nd and 141Ce impurities, each respectively aris- ing from either impurities in the natural target material or the high natural abundance of parent 140Ce. These contaminants could be avoided by using an enriched 142CeO2 tar- get material. Despite overlapping impurities, the intervening fractions free of impurities contained > 99% of the total 143Pr activity (1.03 mCi) with a radiopurity of > 99.99% con- firmed by γ-ray spectra and β−-activity monitoring over 2 months (Kubota 1976). Other separation methods for the Ce/Pr lanthanoid pair have also exploited the differ- ing chemistry exhibited by Ce in its tetravalent state, namely via extraction with dioctyl phosphoric acid (Peppard et  al. 1957), or more recently by using liquid–liquid extrac- tion methodologies with ionic liquids (Gras et  al. 2017). Similar to other ionic liquid extraction techniques for lanthanoid separation mentioned earlier, ionic liquid extrac- tion and separation of Ce from other lanthanoids (particularly Pr) relies on the selec- tive oxidation of C e3+ to C e4+ while other lanthanoids remain in their trivalent state. In one investigation (Gras et  al. 2017), Ce was separated from lanthanoids of similar mass (La, Pr and Nd) using two different ionic liquids: trihexyltetradecylphosphonium bis(trifluoromethansulfonyl)imide ([P66614][NTf2]) and 1-methyl-1-butylpyrrolidinium bis(trifluoromethanesulfonyl)imide ([C1C4Pyrr][NTf2]). Initial studies indicated that [C C Pyrr][NTf ] exhibited better Ce4+1 4 2 extraction efficiency by a factor of two, and thus was selected as the ionic liquid extractant for subsequent separations. A mixture of tri- valent lanthanoid salts (La, Ce, Pr and Nd) was treated with oxygen under alkaline con- ditions (2 M NaOH, 30 °C, 3 h) to selectively oxidise C e3+ to Ce4+ (87.1% Ce3+ oxidised to Ce4+) and cause the precipitation of Ce(OH)4, while the other unoxidized lanthanoids precipitated out as their trivalent hydroxide salts. The precipitated salts were filtered, washed and subsequently dissolved in HNO3 and the aqueous lanthanoid salt mixture formed was mixed with the ionic liquid ([C1C4Pyrr][NTf2]), stirred for 20 min and cen- trifuged at 6000 rpm for 10 min to facilitate the selective extraction of tetravalent Ce into the ionic liquid phase (Gras et al. 2017). The two phases were then separated and the Ce4+ stripped from the ionic liquid using dilute H NO3. The lanthanoids in their trivalent S adler et al. EJNMMI Radiopharmacy and Chemistry (2022) 7:21 Page 41 of 55 state are very poorly extracted into the ionic liquid due to their inability to form nega- tively charged complexes with nitrate ions that are soluble in the ionic liquid, whereas the tetravalent Ce is speculated to form a negatively charged polynitratoceriate(IV) com- plex which is readily extracted into the ionic liquid phase (Gras et al. 2017). Distribution coefficients (D) for Ce4+ and separation factors (β) for Ce4+ versus other Ln3+ ions were found to depend significantly upon the H NO3 concentration of the aqueous phase and the concentration of C e4+ in the aqueous phase, with highest D and β values being obtained when higher H NO3 aqueous phase concentrations were used in combination with low Ce4+ concentrations (Gras et al. 2017). In an exemplar protocol for the separa- tion of Ce from other lanthanoids using [C1C4Pyrr][NTf2], experimental parameters of 3.45  M HNO3 and 0.01  M Ce4+ were chosen to optimise the distribution coefficients and separation factors without significantly compromising the overall throughput of the method. Using these conditions, the protocol successfully achieved a Ce4+ distribution coefficient of 74.9 and Ce4+/Ln3+ separation factors on the order of 104. Issues with the overall efficacy of this separation approach arose during the oxidation step, during which only 87.1% of total Ce3+ is oxidised to Ce4+ (Gras et al. 2017). This was attributed to the slow kinetics of the oxidation reaction caused by the low solubility of Ce2(SO4)3 in basic conditions and the imposed 3  h reaction time. Ultrasonication and longer oxidation times are expected to increase the overall oxidation percentage and thus total separation efficiency of the method (Gras et al. 2017). Despite this, 98.8% of the C e4+ produced by the oxidation reaction was extracted and recovered after only one extraction and strip- ping cycle, while repeating the procedure resulted in 99.5% of the C e4+ being recovered (Gras et  al. 2017). Thus, with the requisite improvements to the oxidation reaction to maximise the oxidation percentage, near quantitative extraction and recovery of Ce from a mixture of similar-mass lanthanoids is highly feasible. However, while these sepa- ration results are promising for an ideal solution with similar Ln3+ concentrations, the application of this protocol to the separation of bulk target Ce from small amounts of 143Pr in a macro–micro component system for radionuclide production is yet to be real- ised, and has the potential to encounter extraction efficiency issues due to the observa- tion that distribution coefficients for Ce4+ decrease with increasing Ce4+ concentration. Thulium: 170Tm In addition to 153Sm receiving considerable attention and subsequent approval for bone pain palliation treatment, 170Tm has been proposed as a suitable radionuclide for bone pain palliation due to favourable decay characteristics and an advantageous half-life (t1/2 = 128.6 d). Other radionuclides currently used for bone pain palliation such as 89Sr and 153Sm, while clinically beneficial, either still exhibit certain undesirable decay prop- erties or lack certain desirable properties which have detracted from their overall suit- ability for bone pain palliation applications. The use of 89Sr in the form of [89Sr]SrCl2 has typically encountered issues with dose burden, due to the higher-energy β−-particles emitted by 89Sr causing bone marrow suppression (Vats et al. 2013). Furthermore, the absence of an imageable γ-emission has made dosimetry, biodistribution and pharma- cokinetic studies more challenging since they cannot be conducted concurrently. In the case of 153Sm, while a lower β−-particle energy compared to 89Sr has alleviated issues relating to patient dose burden, logistical issues have arisen due to a shorter half-life Sadler et al. EJNMMI Radiopharmacy and Chemistry (2022) 7:21 Page 42 of 55 which limits the distances the radionuclide can be transported before substantial decay and loss of overall activity has occurred. In response to the identification of certain disadvantages for currently used bone pain palliation radionuclides, 170Tm has gained attention due to its longer physical half-life which increases the transportation and distribution radius and thereby the availability of the radionuclide (Goyal and Antonarakis 2012), while the lower β−-particle energies (βmax = 0.968 MeV) than the widely-used [89Sr]SrCl2 reduce the patient dose burden and consequently alleviate the potential for bone marrow suppression (Vats et al. 2013; Das et al. 2009). Moreover, 170Tm emits an imageable γ-photon (84 keV, 3.26%) which allows for dosimetry, biodistribution and pharmacokinetic data to be collected via SPECT imaging during treatment without any appreciable contribution to the patient dose burden (Das et al. 2009; Polyak et al. 2014). With advantages over both 89Sr and 153Sm, 170Tm has demonstrable potential as an alternative bone pain palliation agent while also displaying the requisite theranostic properties of simultaneous imaging and therapeutic capabilities (Vats et al. 2013; Das et al. 2009; Guerra Liberal et al. 2016; Shirvani-Arani et al. 2013). As a result, 170Tm is an interesting candidate for the application of theranos- tics to bone pain palliation. Another advantage of 170Tm is the method by which it is produced, which provides a cost-effective production and purification pathway to access the target radiolanthanoid at sufficient activity yields using moderate flux nuclear reactors (Vats et  al. 2013; Das et al. 2009; Danon et al. 1998; Neves et al. 2002). This is due to the fact that 170Tm is typically produced by means of the thermal neutron bombardment of 169Tm (100% natu- ral abundance) via the 169Tm(n,γ)170Tm nuclear reaction, which exhibits a relatively high thermal neutron capture cross section (σth = 109 barns) as well as a high resonance inte- gral (σRI = 1548 barn) (Das et al. 2009; Danon et al. 1998; Linden et al. 1974). An illustra- tive protocol for 170Tm production involved the use of powdered 169Tm2O3 targets and a thermal neutron flux of 6 × 1013 n/cm2/s for 60 days, culminating in a 170Tm specific activity of 6.36 GBq/mg. At EOB, the target was dissolved in 1 M HCl with heating to aid dissolution, after which the solution was evaporated to near dryness and reconsti- tuted as the desired chloride salt in double-distilled water (Das et  al. 2009). The final [170Tm]TmCl3 formulation exhibited a radiopurity of 100%, ascertained by the absence of any peaks in the γ-spectrum that did not correspond to 170Tm decay. In addition, the absence of radionuclidic impurities with substantially longer half-lives was confirmed by allowing the 170Tm to decay for 7 half-lives (~ 2.5 years) and monitoring the γ-spectrum which did not reveal any peaks that were not attributable to 170Tm (Das et  al. 2009). This 170Tm formulation was used to successfully radiolabel EDTMP with > 99% radio- chemical purity, and a normal Wistar rat model was used for dosimetric, biodistribution and scintigraphic imaging studies of the resultant [ 170Tm]Tm-EDTMP complex. Results showed significant preferential accumulation in the skeleton while other tissues and organs exhibited low to negligible uptake, positively highlighting the targeting ability of the complex (Das et al. 2009). Notably, when employing neutron bombardment on 169Tm targets for the produc- tion of 170Tm, special attention must be paid to the irradiation parameters to minimise the formation of long-lived 171Tm impurities ( t1/2 = 1.92 y) (Sahiralamkhan et al. 2016). The formation of unwanted 171Tm is the result of a double neutron capture event on the S adler et al. EJNMMI Radiopharmacy and Chemistry (2022) 7:21 Page 43 of 55 target 169Tm via the 169Tm(n,γ)170Tm(n,γ)171Tm nuclear reaction due to the similarly high cross section of the second neutron capture on 170Tm (σ = 92 barns c.f. σ = 109 barns). Consequently, when irradiated under medium to high neutron flux conditions, appre- ciable amounts of 171Tm are formed which will remain as contaminants long after the desired 170Tm has decayed due to its much longer half-life (Sahiralamkhan et al. 2016). This poses significant clinical translation issues for formulations of 170Tm due to the contaminant 171Tm exhibiting vastly different β−-decay and γ-emission properties, with a maximum β−-particle energy of only 97 keV (approximately 10% of 170Tm β−-particle energy) and a very low abundance γ-emission energy (66.7 keV, 0.16%) (Sahiralamkhan et  al. 2016; Kajan et  al. 2018; Tishchenko et  al. 2015). Furthermore, due to its identi- cal chemical nature, 171Tm cannot be separated from the desired 170Tm using traditional chemical means such as chromatography or extraction methodologies and consequently the specific activity and radiopurity of the final formulation are adversely affected. At higher neutron fluxes and longer irradiation periods, a greater degree of transmutation of 170Tm to 171Tm is observed, and as a result, 171Tm impurities are formed at signifi- cantly higher levels which impact upon the radiopurity and dilute the specific activity of the final formulation. Consequently, higher neutron fluxes on the order of 1014 and longer irradiation periods of > 120  days are not deemed ideal for preparation of clini- cally applicable 170Tm (Sahiralamkhan et al. 2016). Optimum irradiation conditions that achieve sufficient specific activity of 170Tm while essentially eliminating the by-produc- tion of 171Tm could be theoretically achieved using a neutron flux of 5 × 1013 n/cm2/s for a 15 day irradiation period, which would result in ~ 1.42 GBq/mg 170Tm with a radiopu- rity of > 99.9% (Sahiralamkhan et al. 2016). Such conditions were employed in the pro- duction of 170Tm for EDTMP radiolabelling, with ~ 1.52 GBq/mg 170Tm being produced at > 99.9% radiopurity suitable for bone pain metastatic applications. Subsequent radiola- belling of EDTMP was achieved at > 98% yield, equivalent to ~ 32.3 GBq/mM for [ 170Tm] Tm-EDTMP (Sahiralamkhan et  al. 2016). This formulation was deemed suitable for a preliminary clinical study on a human male (73 y) with skeletal metastases arising from primary prostate carcinoma. 185 MBq of [170Tm]Tm-EDTMP was administered and the patient was imaged 24 h after administration whereby significant localised preferential uptake in metastatic skeletal lesions was observed, comparable to that of a [ 99mTc]Tc- MDP bone scan (740 MBq, imaged 4 h after administration) (Sahiralamkhan et al. 2016). Compared to earlier production of 170Tm, which used a slightly higher neutron flux and a longer irradiation period (Das et al. 2009), ~ 4 times lower overall specific activity was obtained with the optimised parameters, however the final formulation was still deemed suitable for clinical applications in bone pain palliation (for which lower specific activ- ity formulations are appropriate) and the shorter irradiation time has the potential to improve overall production and transportation logistics (Sahiralamkhan et  al. 2016). Importantly however, the irradiation parameters employed by the earlier study afforded a [170Tm]TmCl3 product with radiopurity of practically 100%, hereby demonstrating that slightly higher neutron fluxes with longer irradiation times are capable of affording a product with higher specific activity than those obtained using the optimised param- eters without necessarily compromising the final radiopurity (Das et al. 2009). The feasi- bility of this approach is also corroborated by the theoretical calculations reported in the more recent investigation (Sahiralamkhan et al. 2016). Sadler et al. EJNMMI Radiopharmacy and Chemistry (2022) 7:21 Page 44 of 55 Further preclinical work towards the implementation of 170Tm in bone pain palliation applications has been undertaken by radiolabelling a series of both cyclic and acyclic polyaminophosphonic acid-based chelators, namely EDTMP, diethylenetriaminepenta- methylenephosphonic acid (DTPMP), triethylenetetraminehexamethylenephosphonic acid (TTHMP), 1,4,7,10-tetraazacyclododecane-1,4,7,10-tetramethylenephosphonic acid (DOTMP) and 1,4,8,11-tetraazacyclotetradecane-1–4-8–11-tetramethylenephos- phonic acid (CTMP) (see Fig. 12) (Vats et al. 2013; Shirvani-Arani et al. 2013; Shirmardi et al. 2020; Das et al. 2017). Dosimtery, biodistribution, pharmacokinetic, scinitigraphic and SPECT imaging studies of these radiolabelled compounds have given a prepon- derance of promising preclinical data indicating the significant potential for 170Tm as a theranostic probe for bone pain palliation arising from various cancers metastasising to the skeleton (Vats et al. 2013; Shirvani-Arani et al. 2013; Shirmardi et al. 2020; Das et al. 2017). In one study, 170Tm was prepared using a 169Tm2(NO3)3 target and a neutron flux of 3–4 × 1013 n/cm2/s for 5 days, resulting in a final formulation with adequate spe- cific activity and radiopurity (> 99.99%) suitable for subsequent radiolabelling EDTMP with a yield of > 99% and high in  vitro stability (> 95% after two months incubation at RT). Biodistribution studies and SPECT imaging of the resulting complex in Wistar rats showed high uptake in the skeleton (70% of administered dose), low levels of uptake for other major organs, fast clearance from the blood and renal excretion (Shirvani-Arani et al. 2013). Other polyaminophosphonic acid chelators have been evaluated in another study to compare potential alternatives to the more widely used EDTMP (Vats et  al. 2013). A 169Tm2O3 target was irradiated with a 7 × 1013n/cm2/s neutron flux for 60 days to produce 6.44 TBq/g of [ 170Tm]TmCl3 with radiopurity of 99.62%. After radiochemi- cal experiments to ascertain the optimum complexation conditions for each ligand, they were radiolabelled with 170Tm resulting in radiochemical purities of > 98%, except for CTMP for which approximately 97% complexation was achieved. High in  vitro O - O O O O O P - P - - O O O O P O O O O N N P O O PN P N N P O 170Tm - O 170Tm - O O O O N N N N - 170Tm P O O P O P P P - - O O - O O - O O O O O O P O O O - O 170Tm-EDTMP 170Tm-DOTMP 170Tm-DTPMP - O O -O O OP P O O O P - O P N N - N N O O O 170Tm O - - -O - 170 -O O Tm O N N P N N O O P P O O P O P P O - O - O O O - O O O 170Tm-CTMP 170Tm-TTHMP Fig. 12 Structures of 170Tm-labelled phosphonate chelators S adler et al. EJNMMI Radiopharmacy and Chemistry (2022) 7:21 Page 45 of 55 stability was also observed for all complexes. Biodistribution of each complex was inves- tigated using a Wistar rat model, in which each complex showed negligible uptake in all vital organs and other tissues and displayed significant selective uptake in the bone matrix, which was confirmed by scinitigraphy. Fast renal clearance was also observed for each metal complex (Vats et al. 2013). Of the four complexes prepared, the best per- forming compound displaying the greatest potential for further investigation and devel- opment for clinical applications was [170Tm]Tm-DOTMP, owing to the lower amount of ligand required to achieve > 98% radiolabelling yield over a large pH range of 2—10, and higher thermodynamic stability of the complex (Vats et al. 2013). More recently, however, [ 170Tm]Tm-EDTMP has attracted more considerable research attention. Further preclinical studies and preliminary clinical investigations have afforded results that are indicative of the significant future potential of [170Tm]Tm- EDTMP as a theranostic agent in bone pain palliation (Shirmardi et al. 2020; Das et al. 2017). One illustrative example is the efficacy evaluation of a freeze-dried EDTMP kit for the in-house preparation of [ 170Tm]Tm-EDTMP which has provided encouraging results (Das et al. 2017). Conclusions As the frontiers of theranostics continue to be explored, it is highly probable that Rare Earth radionuclides will play an increased role  in future. Further research into opti- mising production methods for radionuclides that are already in use, as well as in the development pipeline, will facilitate more streamlined translation to clinical applica- tions, however with more nuclear reactors being retired in the next decade and research focus shifting towards linear accelerator and cyclotron production methods, Rare Earth radionuclides that can be reliably and cost-effectively produced using linear accelera- tors and/or cyclotrons will most likely see greater application in the clinic. In particular, those Rare Earth radionuclides that can be produced at commercial or biomedical cyclo- trons already in operation (e.g. certain isotopes of Sc, Tb, La and Er) will be favoured due to the added logistical convenience of minimal transportation or on-site produc- tion. With interest in α-emitters increasing, 149Tb will likely play a larger role in thera- nostics in alpha settings due to its accompanying γ-emission for SPECT, however more facilities with 149Tb production capabilities, most notably high intensity 3He beamlines and mass separation capabilities for proton spallation, will be required in order to pro- duce the desired quantities of 149Tb at reasonable cost for future clinical work. Likewise, theranostically matched Tb radionuclides for 149Tb will likely gain traction, however a greater emphasis on linear accelerator production methods for Tb isotopes will likely be required, as well as greater focus on mass separation methods to achieve the requi- site radionuclidic purities when using proton spallation on Ta-foil for Tb isotope pro- duction. Rare Earth Auger emitters possess the interchangeability in chelator structures that would be advantageous for clinical evaluation of theranostic protocols, however for radionuclides such as 165Er further isotope production optimisation is necessary. As such, for the production of radionuclides such as 153Sm and 165Er where target mate- rial separation and/or impurity separation proves more challenging when using more conventional approaches, mass separation methods (both on-line and off-line) will likely play a more prominent role in future purification protocols for such radionuclides, due Sadler et al. EJNMMI Radiopharmacy and Chemistry (2022) 7:21 Page 46 of 55 to promising indications that much higher radionuclidic purities are potentially attain- able. It must, however, also be noted that the separation efficiencies of current mass separation methods require further optimisation in order to achieve these desired out- comes in the future. 166Ho has also been presented as a versatile radionuclide in many cancer settings, and as Rare Earth bifunctional chelators are explored further, the remit of this radionuclide will likely be expanded further. In addition, as radionuclide produc- tion methods and purification improve the overall activity yield and purity of desired Rare Earth radionuclides, the versatility of interchangeability in established ligand struc- tures (e.g. DOTA, EDTMP and related structures) will expand the library of available radionuclides for therapy (α, β, γ) and imaging (MRI, PET, SPECT) to include more Rare Earth radiosiotopes, which will allow for a more personalised theranostic approach to cancer treatment in years to come. While the outlook for future innovation in the field of Rare Earth radionuclides for theranostic applications is positive, the retirement of aging reactors around the world necessitates a shift in research focus to linear accelerator and cyclotron production methods to maintain pace with the growing demand for radionu- clides for nuclear medicine applications. Consequently, an increase in linear accelerator and cyclotron infrastructure will be required to meet future production demands. Abbreviations AE Auger electron CERN-MEDICS M edical isotopes collected from ISOLDE CTMP 1 ,4,8,11-Tetraazacyclotetradecane-1-4-8-11-tetramethylenephosphonic acid DCH18C6 Dicyclohexano-18-crown-6 DO3Apic 10-((6-Carboxypyridin-2-yl)methyl)-1,4,7,10-tetra-azacyclododecane-1,4,7-triacetic acid DOTA 1,4,7,10-Tetraazacyclododecane-1,4,7,10-tetraacetic acid DOTA-OSSu N-Hydroxysulfosuccinimidyl DOTA DOTA-NOC D OTA-1-NaI3-octreotide DOTA-TATE DOTA-octreotate DOTA-TOC DOTA-octreotide DOTMP 1,4,7,10-Tetraazacyclododecane-1,4,7,10-tetramethylenephosphonic acid DTPMP Diethylenetriaminepentamethylenephosphonic acid DUPA 2-[3-(1,3-Dicarboxypropyl)ureido]pentanedioic acid EDTMP Ethylenediaminetetra(methylene phosphonate) EOB E nd of bombardment EOI End of irradiation Gd-BT-DO3A Gadolinium(III)-tris(carboxymethyl)-1,4,7,10-tetraazacyclododecane-butriol Gd-DTPA Gadolinium(III)-diethylenetriaminepentaacetic acid HA H ydroxyapatite HDEHP Di-(2-ethylhexyl)phosphoric acid HEH[EHP] 2 -Ethylhexylphosphonic acid mono-2-ethylhexyl ester α-HIBA α -Hydroxyisobutyric acid HPGe High purity germanium ICP-OES Inductively coupled plasma optical emission spectrometry ISAC Isotope Separator and Accelerator ISOLDE Isotope seperator on line device LET L inear energy transfer macropa N ,N’-Bis[(6-carboxy-2-pyridil)methyl]-4,13-diaza-18-crown-6 MDP Methylenediphosphonate MELISSA M EDICIS’ Laser Ion Source Assembly MeO-DOTA Methoxy-DOTA MRI Magnetic resonance imaging PET Positron emission tomography PMMA Polymethylmethacrylate PSMA Prostate-specific membrane antigen PSMA-617 Vipivotide tetraxetan RIT Radioimmunotherapy p-SCN-Bn-DOTA S-2-(4-isothiocyanatobenzyl)-1,4,7,10-tetraazacyclododecanetetraacetic acid SPECT Single photon emission computed tomography TBP Tri-butylphosphate TENDL TALYS-based evaluated nuclear data library Sadler et al. EJNMMI Radiopharmacy and Chemistry (2022) 7:21 Page 47 of 55 TND Tumour-to-normal-tissue distribution ratio TRIUMF T ri-University Meson Facility TTHMP Triethylenetetraaminehexamethylenephosphonic acid Acknowledgements Not applicable. Author contributions AWES and LMR contributed to the conception, design, and proof-reading of the manuscript. Material preparations were performed by AWES, LH and BF. All authors read and approved the final manuscript. Funding We thank the Australian Research Council (ARC) for funding (ARC Discovery Program Grant ID: DP190103461). Availability of data and materials Not applicable. Declarations Ethics approval and consent to participate Not applicable. Consent for publication Not applicable. Competing interests The authors declare no competing interests. Received: 25 May 2022 Accepted: 22 July 2022 References Abrarov OA, Aminova MM. Production of a promethium-149 radioactive isotope (carrier-free). Dokl Akad Nauk UzSSR. 1975;11:17–8. Aliev RA, Khomenko IA, Kormazeva ES. Separation of 167Tm, 165Er and 169Yb from erbium targets irradiated by 60 MeV alpha particles. J Radioanal Nucl Chem. 2021;329:983–9. https:// doi. org/ 10. 1007/ s10967- 021- 07865-y. Allen BJ, Goozee G, Sarkar S, Beyer G, Morel C, Byrne AP. Production of terbium-152 by heavy ion reactions and proton induced spallation. Appl Radiat Isot. 2001;54:53–8. https:// doi. org/ 10. 1016/ s0969- 8043(00) 00164-0. Alliot C, Kerdjoudj R, Michel N, Haddad F, Huclier-Markai S. Cyclotron production of high purity (44m,44)Sc with deuter- ons from (44)CaCO3 targets. Nucl Med Biol. 2015;42(6):524–9. Aluicio-Sarduy E, Hernandez R, Olson AP, Barnhart TE, Cai W, Ellison PA, et al. Production and in vivo PET/CT imaging of the theranostic pair 132/135La. Sci Rep. 2019;9:10658. https:// doi. org/ 10. 1038/ s41598- 019- 47137-0. Aluicio-Sarduy E, Barnhart TE, Weichert J, Hernandez R, Engle JW. Cyclotron-produced (132)La as a PET imaging surrogate for therapeutic (225)Ac. J Nucl Med. 2021;62:1012–5. https:// doi. org/1 0. 2967/ jnumed. 120. 255794. AluicioSarduy E, Thiele NA, Martin KE, Vaughn BA, Devaraj J, Olson AP, Barnhart TE, Wilson JJ, Boros E, Engle JW. Establish- ing radiolanthanum chemistry for targeted nuclear medicine applications. Chem A Eur J. 2020;26(6):1238–42. https:// doi. org/ 10.1 002/ chem. 20190 5202. Amoroso AJ, Fallis IA, Pope SJA. Chelating agents for radiolanthanides: applications to imaging and therapy. Coord Chem Rev. 2017;340:198–219. https:// doi. org/ 10. 1016/j. ccr. 2017. 01. 010. Anderson P, Nuñez R. Samarium lexidronam (153Sm-EDTMP): skeletal radiation for osteoblastic bone metastases and osteosarcoma. Expert Rev Anticancer Ther. 2007;7:1517–27. https:// doi. org/ 10. 1586/ 14737 140.7. 11. 1517. Atkins HL. Overview of nuclides for bone pain palliation. Appl Radiat Isot. 1998;49:277–83. https:// doi. org/ 10. 1016/ S0969- 8043(97) 00039-0. Bahrami-Samani A, Bagheri R, Jalilian AR, Shirvani-ARAni S, Ghannadi-Maragheh M, Shamsaee M. Production, quality control and pharmacokinetic studies of 166Ho-EDTMP for therapeutic applications. Sci Pharm. 2010;78:423–34. https:// doi. org/ 10. 3797/ sciph arm. 1004- 21. Bartoś B, Majkowska A, Krajewski S, Bilewicz A. New separation method of no-carrier-added 47Sc from titanium targets. Radiochim Acta. 2012;100:457–62. https:// doi. org/ 10. 1524/ ract. 2012. 1938. Baum RP, Singh A, Benešová M, Vermeulen C, Gnesin S, Köster U, et al. Clinical evaluation of the radiolanthanide ter- bium-152: first-in-human PET/CT with 152Tb-DOTATOC. Dalton Trans. 2017;46:14638–46. https:// doi. org/ 10.1 039/ C7DT0 1936J. Bayouth JE, Macey DJ, Kasi LP, Garlich JR, McMillan K, Dimopoulos MA, et al. Pharmacokinetics, dosimetry and toxicity of holmium-166-DOTMP for bone marrow ablation in multiple myeloma. J Nucl Med. 1995a;36:730–7. Bayouth JE, Macey DJ, Boyer AL, Champlin RE. Radiation dose distribution within the bone marrow of patients receiving holmium-166-labeled-phosphonate for marrow ablation. Med Phys. 1995b;22:743–53. https:// doi. org/ 10. 1118/1. 597491. Beyer GJ, Čomor JJ, Daković M, Soloviev D, Tamburella C, Hagebø E, et al. Production routes of the alpha emitting 149Tb for medical application. Radiochim Acta. 2002;90:247–52. https:// doi. org/ 10. 1524/ ract. 2002. 90.5_ 2002. 247. Sadler et al. EJNMMI Radiopharmacy and Chemistry (2022) 7:21 Page 48 of 55 Beyer GJ, Miederer M, Vranješ-Đurić S, Čomor JJ, Künzi G, Hartley O, et al. Targeted alpha therapy in vivo: direct evidence for single cancer cell kill using 149Tb-rituximab. Eur J Nucl Med Mol Imaging. 2004a;31:547–54. https:// doi. org/ 10. 1007/s 00259- 003-1 413-9. Beyer GJ, Zeisler SK, Becker DW. The Auger-electron emitter 165Er: excitation function of the 165Ho(p, n)165Er process. Radiochim Acta. 2004b;92:219–22. Bokhari TH, Mushtaq A, Khan IU. Separation of no-carrier-added radioactive scandium from neutron irradiated titanium. J Radioanal Nucl Chem. 2010;283:389–93. https:// doi.o rg/ 10. 1007/ s10967-0 09- 0370-6. Bombardieri E, Seregni E, Evangelista L, Chiesa C, Chiti A. Clinical applications of nuclear medicine targeted therapy. Springer; 2018. Bouchet LG, Bolch WE, Goddu SM, Howell RW, Rao DV. Considerations in the selection of radiopharmaceuticals for pallia- tion of bone pain from metastatic osseous lesions. J Nucl Med. 2000;41:682–7. Bourgeois M, Isnard H, Gourgiotis A, Stadelmann G, Gautier C, Mialle S, et al. Sm isotope composition and Sm/Eu ratio determination in an irradiated 153Eu sample by ion exchange chromatography-quadrupole inductively coupled plasma mass spectrometry combined with double spike isotope dilution technique. J Anal at Spectrom. 2011;26:1660–6. https:// doi.o rg/ 10. 1039/ C1JA1 0070J. Bousis C, Emfietzoglou D, Hadjidoukas P, Nikjoo H. Monte Carlo single-cell dosimetry of Auger-electron emitting radionu- clides. Phys Med Biol. 2010;55:2555–72. https:// doi.o rg/ 10. 1088/ 0031- 9155/ 55/9/ 009. Buchegger F, Perillo-Adamer F, Dupertuis YM, Bischof DA. Auger radiation targeted into DNA: a therapy perspective. Eur J Nucl Med Mol Imaging. 2006;33:1352–63. https:// doi. org/ 10.1 007/s 00259- 006-0 187-2. Bunney LR, Abriam JO, Scadden EM. Half-lives of 149Pm and 151Pm. J Inorg Nucl Chem. 1960;12:228–33. https:// doi. org/ 10. 1016/ 0022- 1902(60)8 0365-X. Carzaniga TS, Auger M, Braccini S, Bunka M, Ereditato A, Nesteruk KP, et al. Measurement of 43Sc and 44Sc production cross-section with an 18MeV medical PET cyclotron. Appl Radiat Isot. 2017;129:96–102. https:// doi. org/1 0. 1016/j. aprad iso.2 017. 08.0 13. Chakraborty S, Das T, Chirayil V, Lohar SP, Sarma HD. Erbium-169 labeled hydroxyapatite particulates for use in radiation synovectomy of digital joints – a preliminary investigation. Radiochim Acta. 2014;102:443–50. https://d oi. org/1 0. 1515/r act-2 013- 2166. Chakravarty R, Chakraborty S, Chirayil V, Dash A. Reactor production and electrochemical purification of 169Er: a potential step forward for its utilization in in vivo therapeutic applications. Nucl Med Biol. 2014;41:163–70. https://d oi. org/ 10. 1016/j. nucme dbio. 2013. 11. 009. Chakravarty R, Chakraborty S, Khan MS, Ram R, Sarma HD, Dash A. An electrochemical approach for removal of radio- nuclidic contaminants of Eu from 153Sm for effective use in metastatic bone pain palliation. Nucl Med Biol. 2018;58:8–19. https://d oi. org/1 0. 1016/j. nucme dbio. 2017.1 1. 010. Cho BC, Kim EH, Choi HJ, Kim JH, Roh JK, Chung HC, et al. A pilot study of trans-arterial injection of 166Holmium-Chitosan complex for treatment of small hepatocellular carcinoma. Yonsei Med J. 2005;46:799–805. Chopra A. [(149/152/155/161)Tb]-Labeled DOTA-folate conjugated to an albumin-binding entity. In: Molecular imaging and contrast agent database (MICAD). Bethesda (MD): National Center for Biotechnology Information (US); 2004. Christoforidou AV, Saliba RM, Williams P, Qazilbash M, Roden L, Aleman A, et al. Results of a retrospective single institution analysis of targeted skeletal radiotherapy with (166)Holmium-DOTMP as conditioning regimen for autologous stem cell transplant for patients with multiple myeloma. Impact on transplant outcomes. Biol Blood Marrow Trans- plant. 2007;13:543–9. https:// doi. org/ 10. 1016/j.b bmt. 2006. 12.4 48. Clough TJ, Jiang L, Wong K-L, Long NJ. Ligand design strategies to increase stability of gadolinium-based magnetic reso- nance imaging contrast agents. Nat Commun. 2019;10:1420. https:// doi. org/ 10.1 038/s 41467- 019-0 9342-3. Cotton S. Lanthanide and actinide chemistry. Chichester: John Wiley & Sons Ltd; 2006. Cutler CS, Smith CJ, Ehrhardt GJ, Tyler TT, Jurisson SS, Deutsch E. Current and potential therapeutic uses of lanthanide radioisotopes. Cancer Biother Radiopharm. 2000;15:531–45. https:// doi.o rg/ 10. 1089/ cbr. 2000.1 5. 531. Dadachova E, Mirzadeh S, Lambrecht RM, Hetherington EL, Knapp FF. Separation of carrier-free holmium-166 from neutron-irradiated dysprosium targets. Anal Chem. 1994;66:4272–7. https:// doi. org/ 10.1 021/a c000 95a024. Dadachova E, Mirzadeh S, Lambrecht RM, Hetherington EL, Knapp FF Jr. Separation of carrier-free166Ho from Dy2O3 targets by partition chromatography and electrophoresis. J Radioanal Nucl Chem Lett. 1995;199:115–23. https:// doi. org/1 0. 1007/B F021 62474. Danon Y, Werner CJ, Youk GU, Block RC, Slovacek RE, Francis NC, et al. Neutron total cross-section measurements and resonance parameter analysis of holmium, thulium, and erbium from 0.001 to 20 eV. Nucl Sci Eng. 1998;128:61–9. Das T, Banerjee S. Radiopharmaceuticals for metastatic bone pain palliation: available options in the clinical domain and their comparisons. Clin Exp Metas. 2017;34:1–10. https:// doi. org/ 10. 1007/ s10585- 016- 9831-9. Das T, Pillai MRA. Options to meet the future global demand of radionuclides for radionuclide therapy. Nucl Med Biol. 2013;40:23–32. https:// doi. org/1 0. 1016/j. nucme dbio. 2012. 09.0 07. Das T, Chakraborty S, Sarma HD, Tandon P, Banerjee S, Venkatesh M, et al. 170Tm-EDTMP: a potential cost-effective alternative to 89SrCl2 for bone pain palliation. Nucl Med Biol. 2009;36:561–8. https:// doi.o rg/ 10. 1016/j. nucme dbio. 2009. 02.0 02. Das T, Guleria M, Parab A, Kale C, Shah H, Sarma HD, et al. Clinical translation of 177Lu-labeled PSMA-617: initial experi- ence in prostate cancer patients. Nucl Med Biol. 2016;43:296–302. https:// doi. org/ 10.1 016/j. nucme dbio.2 016.0 2. 002. Das T, Shinto A, Kamaleshwaran KK, Sarma HD, Mohammed SK, Mitra A, et al. Radiochemical studies, pre-clinical investi- gation and preliminary clinical evaluation of 170Tm-EDTMP prepared using in-house freeze-dried EDTMP kit. Appl Radiat Isot. 2017;122:7–13. https:// doi. org/ 10. 1016/j.a pradi so. 2016. 12. 058. DeNardo GL, DeNardo SJ. Concepts, consequences, and implications of theranosis. Semin Nucl Med. 2012;42:147–50. https:// doi.o rg/ 10. 1053/j. semnuc lmed. 2011.1 2. 003. Dikiy NP, Dovbnya AN, Lyashko YV, Medvedeva EP, Medvedev DV, Uvarov VL. Photonuclear production of Pm-149. Voprosy Atomnoj Nauki i Tekhniki. 2015:157–9. S adler et al. EJNMMI Radiopharmacy and Chemistry (2022) 7:21 Page 49 of 55 Dmitriev PP, Molin GA, Dmitrieva ZP. Production of155Tb for nuclear medicine in the reactions155Gd(pn),156Gd(p2n), and155Gd(d2n). Soviet Atomic Energy. 1989;66:470–2. https:// doi. org/ 10. 1007/ BF011 23521. Domnanich KA, Eichler R, Müller C, Jordi S, Yakusheva V, Braccini S, et al. Production and separation of 43Sc for radiophar- maceutical purposes. EJNMMI Radiopharm Chem. 2017a;2:14. https:// doi.o rg/ 10. 1186/ s41181- 017- 0033-9. Domnanich KA, Müller C, Benešová M, Dressler R, Haller S, Köster U, et al. 47Sc as useful β–-emitter for the radiother- agnostic paradigm: a comparative study of feasible production routes. EJNMMI Radiopharm Chem. 2017b;2:5. https:// doi. org/ 10. 1186/ s41181-0 17- 0024-x. Donanzam BA, Campos TPR, Dalmázio I, Valente ES. Synthesis and characterization of calcium phosphate loaded with Ho-166 and Sm-153: a novel biomaterial for treatment of spine metastases. J Mater Sci Mater Med. 2013;24:2873– 80. https://d oi. org/ 10.1 007/ s10856- 013- 5024-0. Eary JF, Collins C, Stabin M, Vernon C, Petersdorf S, Baker M, et al. Samarium-153-EDTMP biodistribution and dosimetry estimation. J Nucl Med. 1993;34:1031–6. El-Amm J, Aragon-Ching JB. Targeting bone metastases in metastatic castration-resistant prostate cancer. Clin Med Insights Oncol. 2016. https:// doi. org/1 0. 4137/ CMO. Ss3075 1. Elliott R. Magnetic properties of rare earth metals. Springer Science & Business Media; 2013. Elzahry M, Diab W, Sinzinger H. The optimal efficacy of a single therapeutic dose of Sm-153 EDTMP in the treatment of painless skeletal metastases. J Clin Exp Radiol. 2018;1:1–7. Eppard E, de la Fuente A, Benešová M, Khawar A, Bundschuh RA, Gärtner FC, et al. Clinical translation and first in-human use of [44Sc]Sc-PSMA-617 for PET imaging of metastasized castrate-resistant prostate cancer. Theranostics. 2017;7:4359–69. https:// doi. org/ 10. 7150/t hno.2 0586. Farahati J, Elliott J, Höppner S, Stein L, Gilman E, Kumm D, et al. Post-radiosynovectomy imaging of Er-169 using scintigra- phy and autoradiography. Clin Case Rep. 2017;5:1048–50. https:// doi. org/1 0. 1002/c cr3. 987. Favaretto C, Talip Z, Borgna F, Grundler PV, Dellepiane G, Sommerhalder A, et al. Cyclotron production and radiochemical purification of terbium-155 for SPECT imaging. EJNMMI Radiopharm Chem. 2021;6:37. https://d oi. org/ 10.1 186/ s41181-0 21- 00153-w. Fiaccabrino DE, Kunz P, Radchenko V. Potential for production of medical radionuclides with on-line isotope separa- tion at the ISAC facility at TRIUMF and particular discussion of the examples of 165Er and 155Tb. Nucl Med Biol. 2021;94–95:81–91. https://d oi. org/1 0. 1016/j. nucme dbio. 2021. 01.0 03. Filippi L, Chiaravalloti A, Schillaci O, Cianni R, Bagni O. Theranostic approaches in nuclear medicine: current status and future prospects. Expert Rev Med Devices. 2020;17:331–43. https:// doi. org/1 0. 1080/ 17434 440.2 020.1 7413 48. Filosofov DV, Loktionova NS, Rösch F. A 44Ti/44Sc radionuclide generator for potential application of 44Sc-based PET- radiopharmaceuticals. Radiochim Acta. 2010;98:149–56. https:// doi. org/ 10. 1524/ ract. 2010. 1701. Fonslet J, Lee B, Tran T, Siragusa M, Jensen M, Kibedi T, et al. 135La as an Auger-electron emitter for targeted internal radiotherapy. Phys Med Biol. 2017. https:// doi. org/1 0. 1088/ 1361- 6560/ aa9b44. Formento-Cavaier R, Köster U, Crepieux B, Gadelshin VM, Haddad F, Stora T, et al. Very high specific activity erbium 169Er production for potential receptor-targeted radiotherapy. Nucl Instrum Methods Phys Res Sect B. 2020;463:468–71. https://d oi. org/ 10. 1016/j. nimb. 2019. 04. 022. Fricker SP. The therapeutic application of lanthanides. Chem Soc Rev. 2006;35:524–33. https:// doi.o rg/ 10. 1039/ B50960 8C. Giralt S, Bensinger W, Goodman M, Podoloff D, Eary J, Wendt R, et al. 166Ho-DOTMP plus melphalan followed by periph- eral blood stem cell transplantation in patients with multiple myeloma: results of two phase 1/2 trials. Blood. 2003;102:2684–91. https:// doi. org/ 10. 1182/ blood- 2002- 10- 3250. Goeckeler WF, Edwards B, Volkert WA, Holmes RA, Simon J, Wilson D. Skeletal localization of samarium-153 chelates: potential therapeutic bone agents. J Nucl Med. 1987;28:495–504. Goyal J, Antonarakis ES. Bone-targeting radiopharmaceuticals for the treatment of prostate cancer with bone metastases. Cancer Lett. 2012;323:135–46. https:// doi. org/ 10. 1016/j.c anlet. 2012.0 4. 001. Gracheva N, Carzaniga TS, Schibli R, Braccini S, van der Meulen NP. 165Er: a new candidate for Auger electron therapy and its possible cyclotron production from natural holmium targets. Appl Radiat Isot. 2020;159:109079. https://d oi. org/ 10.1 016/j. aprad iso. 2020. 109079. Graf J, Ulrich-Frank P, Henning J, Denecke T, Arsenic R, Brenner W, et al. Prognostic significance of somatostatin receptor heterogeneity in progressive neuroendocrine tumor treated with Lu-177 DOTATOC or Lu-177 DOTATATE. Eur J Nucl Med Mol Imaging. 2020;47:881–94. https://d oi. org/ 10.1 007/ s00259- 019- 04439-9. Gras M, Papaiconomou N, Chainet E, Tedjar F, Billard I. Separation of cerium(III) from lanthanum(III), neodymium(III) and praseodymium(III) by oxidation and liquid-liquid extraction using ionic liquids. Sep Purif Technol. 2017;178:169–77. https:// doi. org/ 10. 1016/j. seppur.2 017. 01. 035. Guerra Liberal FDC, Tavares AAS, Tavares JMRS. Palliative treatment of metastatic bone pain with radiopharmaceuticals: a perspective beyond Strontium-89 and Samarium-153. Appl Radiat Isot. 2016;110:87–99. https:// doi. org/ 10. 1016/j. aprad iso. 2016. 01. 003. Ha EJ, Gwak H-S, Rhee CH, Youn SM, Choi C-W, Cheon GJ. Intracavitary radiation therapy for recurrent cystic brain tumors with holmium-166-Chico : a pilot study. J Korean Neurosurg Soc. 2013;54:175–82. https:// doi. org/ 10. 3340/ jkns. 2013. 54.3. 175. Hashikin NA, Yeong CH, Abdullah BJ, Ng KH, Chung LY, Dahalan R, et al. Neutron activated samarium-153 micropar- ticles for transarterial radioembolization of liver tumour with post-procedure imaging capabilities. PLoS ONE. 2015;10:e0138106. https://d oi. org/ 10.1 371/j ourn al. pone. 01381 06. Hashikin NAA, Yeong CH, Guatelli S, Abdullah BJJ, Ng KH, Malaroda A, et al. Organ doses from hepatic radioembolization with 90Y, 153Sm, 166Ho and 177Lu: A Monte Carlo simulation study using Geant4. J Phys Conf Ser. 2016;694:012059. https://d oi. org/ 10. 1088/1 742- 6596/ 694/1/ 012059. Hassan HE, Alabyad M, Mohamed GY. Production of Ti-44 -> Sc-44 generator in comparison with direct routes by cyclo- trons: cross section evaluation using nuclear models codes. Arab J Nucl Sci Appl. 2018;51:57–72. Hermanne A, Adam-Rebeles R, Tarkanyi F, Takacs S, Csikai J, Takacs MP, et al. Deuteron induced reactions on Ho and La: experimental excitation functions and comparison with code results. Nucl Instrum Methods Phys Res Sect B. 2013;311:102–11. https:// doi. org/ 10.1 016/j. nimb.2 013. 06.0 14. Sadler et al. EJNMMI Radiopharmacy and Chemistry (2022) 7:21 Page 50 of 55 Hirsch AE, Medich DC, Rosenstein BS, Martel CB, Hirsch JA. Radioisotopes and vertebral augmentation: dosimetric analy- sis of a novel approach for the treatment of malignant compression fractures. Radiother Oncol. 2008;87:119–26. https:// doi. org/ 10.1 016/j. radonc. 2008. 01.0 10. Hu F, Cutler CS, Hoffman T, Sieckman G, Volkert WA, Jurisson SS. Pm-149 DOTA bombesin analogs for potential radio- therapy: in vivo comparison with Sm-153 and Lu-177 labeled DO3A-amide-βAla-BBN(7–14)NH2. Nucl Med Biol. 2002;29:423–30. https:// doi. org/ 10. 1016/ S0969- 8051(02) 00290-1. Huh R, Park YS, Lee JD, Chung YS, Park YG, Chung SS, et al. Therapeutic effects of Holmium-166 chitosan complex in rat brain tumor model. Yonsei Med J. 2005;46:51–60. https:// doi. org/ 10. 3349/ ymj.2 005.4 6.1. 51. Humm JL. Dosimetric aspects of radiolabeled antibodies for tumor therapy. J Nucl Med. 1986;27:1490–7. IAEA-TECDOC M. 1340. Manual for reactor produced radioisotope. Vienna: International Atomic Energy Agency; 2003. Islami-Rad SZ, Shamsaei M, Gholipour-Peyvandi R, Ghannadi-Maragheh M. Reactor production and purification of 153Sm radioisotope via natSm target irradiation. Radiochemistry. 2011;53:642–5. https:// doi.o rg/ 10. 1134/ S1066 36221 10601 29. Jelinek L, Wei Y, Arai T, Kumagai M. Selective Eu(III) electro-reduction and subsequent separation of Eu(II) from rare Earths(III) via HDEHP impregnated resin. Solvent Extr Ion Exch. 2007;25:503–13. https:// doi. org/ 10. 1080/ 07366 290701 4159 11. Jelinek L, Wei Y, Arai T, Kumagai M. Study on separation of Eu(II) from trivalent rare earths via electro-reduction and ion exchange. J Alloy Compd. 2008;451:341–3. https://d oi. org/1 0. 1016/j. jallc om. 2007. 04.1 39. Jong J-d, Oprea-Lager DE, Hooft L, de Klerk JMH, Bloemendal HJ, Verheul HMW, et al. Radiopharmaceuticals for Palliation of bone pain in patients with castration-resistant prostate cancer metastatic to bone: a systematic review. Eur Urol. 2016;70:416–26. https://d oi. org/ 10. 1016/j. eururo. 2015. 09. 005. Kajan I, Heinitz S, Dressler R, Reichel P, Kivel N, Schumann D. Emission probability of the 66.7 keV \ensuremath{\gamma} transition in the decay of 171Tm. Phys Rev C. 2018;98:055802. https:// doi.o rg/ 10. 1103/ PhysR evC.9 8. 055802. Kalef-Ezra JA, Valakis ST, Pallada S. Samarium-153 EDTMP for metastatic bone pain palliation: the impact of europium impurities. Physica Med. 2015;31:104–7. https:// doi.o rg/ 10. 1016/j. ejmp. 2014. 10. 078. Karavida N, Notopoulos A. Radiation synovectomy: an effective alternative treatment for inflamed small joints. Hippokra- tia. 2010;14:22–7. Kassis AI. The amazing world of auger electrons. Int J Radiat Biol. 2004;80:789–803. https:// doi. org/ 10.1 080/0 9553 00040 00176 63. Kassis AI. Therapeutic radionuclides: biophysical and radiobiologic principles. Semin Nucl Med. 2008;38:358–66. https:// doi. org/ 10. 1053/j.s emnuc lmed.2 008.0 5. 002. Ketring AR, Ehrhardt GJ, Embree MF, Bailey KD, Tyler TT, Gawenis JA, et al. Production and supply of high specific activity radioisotopes for radiotherapy applications. ALASBIMN J. 2003;5:7. Kieck T, Dorrer H, Düllmann CE, Gadelshin V, Schneider F, Wendt K. Highly efficient isotope separation and ion implanta- tion of 163Ho for the ECHo project. Nucl Instrum Methods Phys Res Sect A. 2019;945:162602. https:// doi. org/ 10. 1016/j.n ima.2 019. 162602. Kim JK, Han K-H, Lee JT, Paik YH, Ahn SH, Lee JD, et al. Long-term clinical outcome of phase IIb clinical trial of percutane- ous injection with holmium-166/chitosan complex (milican) for the treatment of small hepatocellular carcinoma. Clin Cancer Res. 2006;12:543–8. https:// doi. org/ 10. 1158/ 1078- 0432.C cr- 05- 1730. Knapp FF, Dash A. Radiopharmaceuticals for therapy. Springer; 2016. Knut L. Radiosynovectomy in the therapeutic management of arthritis. World J Nucl Med. 2015;14:10–5. https:// doi.o rg/ 10. 4103/ 1450- 1147. 150509. Kolesnikov-Gauthier H, Lemoine N, Tresch-Bruneel E, Olivier A, Oudoux A, Penel N. Efficacy and safety of 153Sm-EDTMP as treatment of painful bone metastasis: a large single-center study. Support Care Cancer. 2018;26:751–8. https:// doi. org/1 0. 1007/s 00520- 017-3 885-3. Kolsky KL, Joshi V, Mausner LF, Srivastava SC. Radiochemical purification of no-carrier-added scandium-47 for radioimmu- notherapy. Appl Radiat Isot. 1998;49:1541–9. https:// doi.o rg/ 10. 1016/ S0969- 8043(98) 00016-5. Kormazeva ES, Khomenko IA, Unezhev VN, Aliev RA. Experimental study of α-particle induced reactions on natural erbium for the production of Auger-emitters 167Tm, 165Er and 169Yb. Appl Radiat Isot. 2021;177:109919. https:// doi. org/ 10. 1016/j.a pradi so. 2021. 109919. Kostelnik TI, Orvig C. Radioactive main group and rare earth metals for imaging and therapy. Chem Rev. 2019;119:902–56. https:// doi. org/ 10. 1021/ acs. chemr ev.8 b002 94. Krajewski S, Cydzik I, Abbas K, Bulgheroni A, Simonelli F, Holzwarth U, et al. Cyclotron production of 44Sc for clinical application. Radiochim Acta. 2013;101:333–8. https:// doi. org/ 10. 1524/ ract. 2013.2 032. Kratochwil C, Fendler WP, Eiber M, Baum R, Bozkurt MF, Czernin J, et al. EANM procedure guidelines for radionuclide therapy with 177Lu-labelled PSMA-ligands (177Lu-PSMA-RLT). Eur J Nucl Med Mol Imaging. 2019;46:2536–44. https://d oi. org/ 10. 1007/ s00259- 019- 04485-3. Kubota M. Preparation of high purity praseodymium-143 from neutron irradiated cerium oxide by cation-exchange separation. J Nucl Sci Technol. 1976;13:492–6. https:// doi.o rg/ 10. 1080/ 18811 248.1 976. 97340 62. Kwak C, Hong SK, Seong SK, Ryu JM, Park MS, Lee SE. Effective local control of prostate cancer by intratumoral injection of 166Ho-chitosan complex (DW-166HC) in rats. Eur J Nucl Med Mol Imaging. 2005;32:1400–5. https://d oi. org/1 0. 1007/ s00259- 005- 1892-y. Lahiri S, Volkers KJ, Wierczinski B. Production of 166Ho through 164Dy(n, γ)165Dy(n, γ)166Dy(β−)166Ho and separation of 166Ho. Appl Radiat Isot. 2004;61:1157–61. https:// doi. org/ 10.1 016/j. aprad iso. 2004. 03. 117. Larsson K, Binnemans K. Separation of rare earths by split-anion extraction. Hydrometallurgy. 2015;156:206–14. https:// doi. org/1 0. 1016/j. hydro met. 2015.0 4. 020. Larsson K, Binnemans K. Separation of rare earths by solvent extraction with an undiluted nitrate ionic liquid. J Sustain Metall. 2017;3:73–8. https:// doi. org/ 10. 1007/ s40831- 016- 0074-4. Lassen J, Li R, Raeder S, Zhao X, Dekker T, Heggen H, et al. Current developments with TRIUMF’s titanium-sapphire laser based resonance ionization laser ion source. Hyperfine Interact. 2017;238:33. https://d oi. org/1 0. 1007/ s10751- 017-1 407-9. S adler et al. EJNMMI Radiopharmacy and Chemistry (2022) 7:21 Page 51 of 55 Lattimer JC, Corwin LA, Stapleton J, Volkert WA, Ehrhardt GJ, Ketring AR, et al. Clinical and clinicopathologic response of canine bone tumor patients to treatment with samarium-153-EDTMP. J Nucl Med. 1990;31:1316–25. Lebeda O, Lozza V, Schrock P, Štursa J, Zuber K. Excitation functions of proton-induced reactions on natural Nd in the 10–30 MeV energy range, and production of radionuclides relevant for double-β decay. Phys Rev C. 2012;85:014602. https:// doi.o rg/ 10. 1103/ PhysR evC.8 5. 014602. Lebeda O, Lozza V, Petzoldt J, Štursa J, Zdychová V, Zuber K. Excitation functions of proton-induced reactions on natural Nd and production of radionuclides relevant for double beta decay: completing measurement in 5–35 MeV energy range. Nucl Phys A. 2014;929:129–42. https:// doi. org/ 10.1 016/j. nuclp hysa. 2014. 06.0 10. Levin VI, Tronova IN, Dmitriev PP, Tikhomirova EA, Gromova NP, Gus’kov AF. Preparation of carrier-free terbium-155. Radiokhimiya. 1977;19:388–93. Levin VI, Malinin AB, Tronova IN. Production of radionuclide by photonuclear reactions. I. Production of terbium-155 and thulium-167 using electron accelerator EA-25. Radiochem Radioanal Lett. 1981;49:111–7. Lewis MR, Zhang J, Jia F, Owen NK, Cutler CS, Embree MF, et al. Biological comparison of 149Pm-, 166Ho-, and 177Lu-DOTA-biotin pretargeted by CC49 scFv-streptavidin fusion protein in xenograft-bearing nude mice. Nucl Med Biol. 2004;31:213–23. https:// doi. org/ 10. 1016/j.n ucme dbio. 2003. 08. 004. Li WP, Smith CJ, Cutler CS, Hoffman TJ, Ketring AR, Jurisson SS. Aminocarboxylate complexes and octreotide complexes with no carrier added 177Lu, 166Ho and 149Pm. Nucl Med Biol. 2003;30:241–51. https:// doi. org/ 10. 1016/ s0969- 8051(02)0 0418-3. Mamtimin M, Harmon F, Starovoitova VN. Sc-47 production from titanium targets using electron linacs. Appl Radiat Isot. 2015;102:1–4. https:// doi. org/ 10.1 016/j. aprad iso.2 015. 04.0 12. Mansel A, Franke K. Production of no-carrier-added 135La at an 18 MeV cyclotron and its purification for investigations at a concentration range down to 10–15 mol/L. Radiochim Acta. 2015;103:759–63. https://d oi. org/1 0. 1515/ ract- 2015- 2427. Marin JFG, Nunes RF, Coutinho AM, Zaniboni EC, Costa LB, Barbosa FG, et al. Theranostics in nuclear medicine: emerging and re-emerging integrated imaging and therapies in the era of precision oncology. Radiographics. 2020;40:1715– 40. https:// doi. org/ 10. 1148/ rg. 20202 00021. Martin RF, D’Cunha G, Pardee M, Allen BJ. Induction of double-strand breaks following neutron capture by DNA-bound 157Gd. Int J Radiat Biol. 1988;54:205–8. https:// doi. org/ 10. 1080/ 09553 008814 5516 41. Mikolajczak R, Huclier-Markai S, Alliot C, Haddad F, Szikra D, Forgacs V, et al. Production of scandium radionuclides for theranostic applications: towards standardization of quality requirements. EJNMMI Radiopharmacy and Chemis- try. 2021;6:19. https://d oi. org/1 0. 1186/ s41181- 021- 00131-2. Minegishi K, Nagatsu K, Fukada M, Suzuki H, Ohya T, Zhang M-R. Production of scandium-43 and -47 from a powdery calcium oxide target via the nat/44Ca(α, x)-channel. Appl Radiat Isot. 2016;116:8–12. https:// doi. org/ 10.1 016/j. aprad iso. 2016. 07. 017. Mishiro K, Hanaoka H, Yamaguchi A, Ogawa K. Radiotheranostics with radiolanthanides: design, development strategies, and medical applications. Coord Chem Rev. 2019;383:104–31. https:// doi. org/ 10. 1016/j. ccr. 2018.1 2. 005. Misiak R, Walczak R, Wąs B, Bartyzel M, Mietelski JW, Bilewicz A. 47Sc production development by cyclotron irradiation of 48Ca. J Radioanal Nucl Chem. 2017;313:429–34. https:// doi. org/ 10. 1007/ s10967-0 17- 5321-z. Moeller T, Martin DF, Thompson LC, Ferrús R, Feistel GR, Randall WJ. The coordination chemistry of yttrium and the rare earth metal ions. Chem Rev. 1965;65:1–50. https:// doi. org/ 10. 1021/ cr602 33a001. Mohsin H, Jia F, Sivaguru G, Hudson MJ, Shelton TD, Hoffman TJ, et al. Radiolanthanide-labeled monoclonal antibody CC49 for radioimmunotherapy of cancer: biological comparison of DOTA conjugates and 149Pm, 166Ho, and 177Lu. Bioconjug Chem. 2006;17:485–92. https://d oi. org/ 10. 1021/ bc0502 356. Mohsin H, Jia F, Bryan JN, Sivaguru G, Cutler CS, Ketring AR, et al. Comparison of pretargeted and conventional CC49 radioimmunotherapy using 149Pm, 166Ho, and 177Lu. Bioconjug Chem. 2011;22:2444–52. https:// doi. org/ 10. 1021/ bc200 258x. Moiseeva AN, Aliev RA, Unezhev VN, Zagryadskiy VA, Latushkin ST, Aksenov NV, et al. Cross section measurements of 151Eu(3He,5n) reaction: new opportunities for medical alpha emitter 149Tb production. Sci Rep. 2020;10:508. https:// doi. org/ 10.1 038/ s41598- 020- 57436-6. Monroy-Guzman F, Jaime SE. Separation of micro-macrocomponent systems:149Pm – Nd, 161Tb-Gd,166Ho-Dy and177Lu-Yb by extraction chromatography. J Mex Chem Soc. 2015a;59:143–50. Monroy-Guzman F, Jaime SE. Separation of micro-macrocomponent systems: 149Pm-Nd, 161Tb-Gd, 166Ho-Dy and 177Lu-Yb by extraction chromatography. J Mex Chem Soc. 2015b;59:143–50. Monroy-Guzman F, Barreiro F, Salinas E, Trevino A. Radiolanthanides device production. World J Nucl Sci Technol. 2015;5:111–9. https:// doi. org/ 10.4 236/ wjnst.2 015. 52011. Montaño CJ, de Campos TPR. Radioactive cement OF PMMA AND HAP-Sm-153, Ho-166, OR RE-188 for bone metastasis treatment. Acta Ortop Bras. 2019;27:64–8. https://d oi. org/1 0. 1590/ 1413-7 8522 01927 01190 288. Morris MJ, Pandit-Taskar N, Carrasquillo J, Divgi CR, Slovin S, Kelly WK, et al. Phase I study of samarium-153 lexidronam with docetaxel in castration-resistant metastatic prostate cancer. J Clin Oncol. 2009;27:2436–42. https:// doi. org/ 10. 1200/ JCO.2 008. 20.4 164. Mughabghab SF. Chapter 1 - thermal cross sections. In: Mughabghab SF, editor. Atlas of neutron resonances. 6th ed. Amsterdam: Elsevier; 2018. p. 1–19. Müller C, Zhernosekov K, Köster U, Johnston K, Dorrer H, Hohn A, et al. A unique matched quadruplet of terbium radio- isotopes for PET and SPECT and for#±- and Ô -radionuclide therapy: an in vivo proof-of-concept study with a new receptor-targeted folate derivative. J Nucl Med. 2012;53:1951–9. Müller C, Bunka M, Reber J, Fischer C, Zhernosekov K, Türler A, et al. Promises of cyclotron-produced 44Sc as a diag- nostic match for trivalent β—emitters: in vitro and in vivo study of a 44Sc-DOTA-folate conjugate. J Nucl Med. 2013;54:2168–74. https:// doi. org/ 10.2 967/ jnumed. 113. 123810. Müller C, Bunka M, Haller S, Köster U, Groehn V, Bernhardt P, et al. Promising prospects for 44Sc-47Sc-based theragnostics: application of 47Sc for radionuclide tumor therapy in mice. J Nucl Med. 2014a;55:1658–64. https://d oi. org/1 0. 2967/j numed. 114.1 41614. Sadler et al. EJNMMI Radiopharmacy and Chemistry (2022) 7:21 Page 52 of 55 Müller C, Reber J, Haller S, Dorrer H, Köster U, Johnston K, et al. Folate receptor targeted alpha-therapy using terbium-149. Pharmaceuticals. 2014b;7:353–65. Müller C, Fischer E, Behe M, Köster U, Dorrer H, Reber J, et al. Future prospects for SPECT imaging using the radi- olanthanide terbium-155 — production and preclinical evaluation in tumor-bearing mice. Nucl Med Biol. 2014c;41:e58–65. https://d oi. org/ 10. 1016/j. nucme dbio. 2013. 11. 002. Müller C, Vermeulen C, Köster U, Johnston K, Türler A, Schibli R, et al. Alpha-PET with terbium-149: evidence and perspec- tives for radiotheragnostics. EJNMMI Radiopharm Chem. 2016;1:5. https:// doi. org/1 0. 1186/s 41181- 016-0 008-2. Naskar N, Lahiri S. Theranostic terbium radioisotopes: challenges in production for clinical application. Front Med. 2021. https://d oi. org/ 10. 3389/ fmed.2 021.6 75014. Navalkissoor S, Grossman A. Targeted alpha particle therapy for neuroendocrine tumours: the next generation of peptide receptor radionuclide therapy. Neuroendocrinology. 2019;108:256–64. https:// doi.o rg/ 10. 1159/ 00049 4760. Nelson BJB, Wilson J, Andersson JD, Wuest F. High yield cyclotron production of a novel 133/135La theranostic pair for nuclear medicine. Sci Rep. 2020;10:22203. https:// doi.o rg/ 10. 1038/ s41598- 020- 79198-x. Neves M, Kling A, Lambrecht RM. Radionuclide production for therapeutic radiopharmaceuticals. Appl Radiat Isot. 2002;57:657–64. https:// doi. org/ 10. 1016/ S0969- 8043(02) 00180-X. Nieschmidt EB, Potnis VR, Ellsworth LD, Mandeville CE. Nuclear states of 149Pm. Nucl Phys. 1965;72:236–40. https:// doi. org/1 0. 1016/0 029- 5582(65) 90642-5. Nikjoo H, Martin RF, Charlton DE, Terrissol M, Kandaiya S, Lobachevsky P. Modelling of Auger-induced Dna damage by incorporated125i. Acta Oncol. 1996;35:849–56. https:// doi.o rg/ 10. 3109/ 02841 86960 910403 6. Notni J, Wester H-J. Re-thinking the role of radiometal isotopes: towards a future concept for theranostic radiopharma- ceuticals. J Labelled Compd Radiopharm. 2018;61:141–53. https:// doi. org/ 10.1 002/ jlcr. 3582. Peacock AFA. De Novo Designed imaging agents based on lanthanide peptides complexes. In: Peptide, protein and enzyme design. Elsevier; 2016. p. 557–80. https:// doi. org/ 10.1 016/ bs.m ie. 2016. 05. 051. Peppard DF, Mason GW, Moline SW. The use of dioctyl phosphoric acid extraction in the isolation of carrier-free 90Y, 140La, 144Ce, 143Pr, and 144Pr. J Inorg Nucl Chem. 1957;5:141–6. https://d oi. org/1 0. 1016/ 0022-1 902(57) 80055-4. Peppard DF, Horwitz EP, Mason GW. Comparative liquid-liquid extraction behaviour of europium (II) and europium (III). J Inorg Nucl Chem. 1962;24:429–39. https://d oi. org/ 10. 1016/ 0022-1 902(62) 80039-6. Pillai M. Metallic radionuclides and therapeutic radiopharmaceuticals. Poland: Institute of Nuclear Chemistry and Tech- nology Warszawa; 2010. p. 50–86. Polyak A, Das T, Chakraborty S, Kiraly R, Dabasi G, Joba RP, et al. Thulium-170-labeled microparticles for local radiotherapy: preliminary studies. Cancer Biother Radiopharm. 2014;29:330–8. https:// doi. org/ 10. 1089/ cbr.2 014. 1680. IAEA. Production, quality control and clinical applications of radiosynovectomy agents: IAEA; 2021. Pruszyński M, Loktionova NS, Filosofov DV, Rösch F. Post-elution processing of (44)Ti/(44)Sc generator-derived (44)Sc for clinical application. Appl Radiat Isot. 2010;68:1636–41. https://d oi. org/1 0. 1016/j. aprad iso. 2010.0 4. 003. Qaim SM, Scholten B, Neumaier B. New developments in the production of theranostic pairs of radionuclides. J Radioanal Nucl Chem. 2018;318:1493–509. https://d oi. org/ 10. 1007/ s10967- 018- 6238-x. Quadramet®. Quadramet® prescribing information. 2017:1–12. Ramachandran K, Begum B. Comparison of Tc-99m MDP and Sm-153 EDTMP bone scan. Indian J Nucl Med. 2011;26:163– 4. https:// doi. org/ 10.4 103/0 972- 3919. 104005. Ramamoorthy N, Saraswathy P, Das MK, Mehra KS, Ananthakrishnan M. Production logistics and radionuclidic purity aspects of Sm for radionuclide therapy. Nucl Med Commun. 2002;23:83–9. https:// doi. org/1 0. 1097/0 00062 31- 20020 1000- 00013. Ramogida CF, Orvig C. Tumour targeting with radiometals for diagnosis and therapy. Chem Commun. 2013;49:4720–39. https:// doi. org/ 10.1 039/C 3CC4 1554F. Rane S, Harris JT, Starovoitova VN. 47Ca production for 47Ca/47Sc generator system using electron linacs. Appl Radiat Isot. 2015;97:188–92. https:// doi.o rg/ 10. 1016/j. apradi so. 2014. 12. 020. Rizvi ASM, Sarkar S, Goozee G, Allen BJ. Radioimmunoconjugates for targeted α therapy of malignant melanoma. Mela- noma Res. 2000;10:281–9. Robertson AG, Rendina LM. Gadolinium theranostics for the diagnosis and treatment of cancer. Chem Soc Rev. 2021;50:4231–44. https:// doi. org/ 10.1 039/ D0CS0 1075H. Roesch F. Scandium-44: benefits of a long-lived PET radionuclide available from the 44Ti/44Sc generator system. Curr Radiopharm. 2012;5:187–201. https:// doi. org/1 0. 2174/1 8744 71011 20503 0187. Rösch F, Baum RP. Generator-based PET radiopharmaceuticals for molecular imaging of tumours: on the way to THERA- NOSTICS. Dalton Trans. 2011;40:6104–11. https://d oi. org/1 0. 1039/C 0DT01 504K. Rotsch DA, Brown MA, Nolen JA, Brossard T, Henning WF, Chemerisov SD, et al. Electron linear accelerator production and purification of scandium-47 from titanium dioxide targets. Appl Radiat Isot. 2018;131:77–82. https:// doi. org/ 10. 1016/j. aprad iso.2 017. 11.0 07. Sadeghi M, Enferadi M, Tenreiro C. Nuclear model calculations on the production of Auger emitter 165 Er for targeted radionuclide therapy. J Mod Phys. 2010;1:217–25. https://d oi. org/ 10. 4236/ jmp. 2010. 14033. Sahiralamkhan M, Chakravarty R, Chakraborty S, Kamaleshwaran KK, Shinto A, Dash A. Irradiation parameters play a cru- cial role in the (n, γ) production of 170Tm suitable for clinical use in bone pain palliation. J Radioanal Nucl Chem. 2016;307:1105–13. https:// doi.o rg/ 10. 1007/ s10967-0 15- 4323-y. Sarkar S, Allen B, Imam S, Goozee G, Leigh J, Meriaty H. Production and separation of terbium-149,152 for targeted cancer therapy. In: Proceedings of second international conference on isotopes, Sydney; 1999. Sartor O, Reid RH, Hoskin PJ, Quick DP, Ell PJ, Coleman RE, et al. Samarium-153-Lexidronam complex for treatment of painful bone metastases in hormone-refractory prostate cancer. Urology. 2004;63:940–5. https:// doi. org/ 10. 1016/j.u rolo gy.2 004.0 1. 034. Schima FJ. Decay of 89Sr and the emission probability of the 909.12keV gamma-ray transition. Appl Radiat Isot. 1998;49(9–11):1359–61. https:// doi.o rg/ 10. 1016/ S0969- 8043(97) 10074-4. S adler et al. EJNMMI Radiopharmacy and Chemistry (2022) 7:21 Page 53 of 55 Schwantes JM, Sudowe R, Nitsche H, Hoffman DC. Applications of solvent extraction in the high-yield multi-process reduction/separation of Eu from excess Sm. J Radioanal Nucl Chem. 2008;276:543–8. https://d oi. org/ 10. 1007/ s10967- 008- 0539-4. Seong SK, Ryu JM, Shin DH, Bae EJ, Shigematsu A, Hatori Y, et al. Biodistribution and excretion of radioactivity after the administration of 166Ho-chitosan complex (DW-166HC) into the prostate of rat. Eur J Nucl Med Mol Imag- ing. 2005;32:910–7. https:// doi. org/1 0. 1007/s 00259- 005- 1792-1. Serafini AN, Houston SJ, Resche I, Quick DP, Grund FM, Ell PJ, et al. Palliation of pain associated with metastatic bone cancer using samarium-153 lexidronam: a double-blind placebo-controlled clinical trial. J Clin Oncol. 1998;16:1574–81. https:// doi.o rg/ 10. 1200/ JCO.1 998.1 6.4. 1574. Severin GW, Engle JW, Valdovinos HF, Barnhart TE, Nickles RJ. Cyclotron produced 44gSc from natural calcium. Appl Radiat Isot. 2012;70:1526–30. https:// doi. org/ 10.1 016/j. aprad iso. 2012. 04.0 30. Shi Y, Johnsen AM, Di Pasqua AJ. Holmium for use in cancer therapy. Comments Inorg Chem. 2017;37:281–300. https://d oi. org/ 10. 1080/ 02603 594. 2017. 13334 98. Shirmardi SP, Saniei E, Das T, Noorvand M, Erfani M, Bagheri R. Internal dosimetry studies of 170Tm-EDTMP complex, as a bone pain palliation agent, in human tissues based on animal data. Appl Radiat Isot. 2020;166:109396. https:// doi. org/ 10.1 016/j. aprad iso.2 020.1 09396. Shirvani-Arani S, Bahrami-Samani A, Meftahi M, Jalilian AR, Ghannadi-Maragheh M. Production, quality control and biodistribution studies of thulium-170-labeled ethylenediamine (tetramethylene phosphonic acid). Radiochim Acta. 2013;101:37–44. https:// doi. org/ 10.1 524/ ract. 2013. 1999. Silberstein EB. Teletherapy and radiopharmaceutical therapy of painful bone metastases. Semin Nucl Med. 2005;35:152–8. https://d oi. org/ 10. 1053/j. semnu clmed. 2004. 11.0 06. Skelton WPT, Dibenedetto SW, Pang SS, Pan K, Barish JL, Nwosu-Iheme A, et al. A single-center retrospective analysis of the effect of radium-223 (Xofigo) on pancytopenia in patients with metastatic castration-resistant prostate cancer. Cureus. 2020;12:e6806-e. https:// doi. org/ 10. 7759/ cureus. 6806. Snow MS, Foley A, Ward JL, Kinlaw MT, Stoner J, Carney KP. High purity 47Sc production using high-energy photons and natural vanadium targets. Appl Radiat Isot. 2021;178:109934. https:// doi.o rg/ 10. 1016/j. aprad iso. 2021. 109934. Srivastava SC. Paving the Way to personalized medicine: production of some promising theragnostic radionuclides at Brookhaven national laboratory. Semin Nucl Med. 2012;42:151–63. https:// doi. org/1 0. 1053/j. semnu clmed. 2011. 12.0 04. Staanum PF, Frellsen AF, Olesen ML, Iversen P, Arveschoug AK. Practical kidney dosimetry in peptide receptor radio- nuclide therapy using [177Lu]Lu-DOTATOC and [177Lu]Lu-DOTATATE with focus on uncertainty estimates. EJNMMI Phys. 2021;8:78. https:// doi. org/1 0. 1186/s 40658- 021-0 0422-2. Starovoitova VN, Cole PL, Grimm TL. Accelerator-based photoproduction of promising beta-emitters 67Cu and 47Sc. J Radioanal Nucl Chem. 2015;305:127–32. https:// doi. org/ 10.1 007/ s10967- 015- 4039-z. Strosberg J, El-Haddad G, Wolin E, Hendifar A, Yao J, Chasen B, et al. Phase 3 trial of (177)Lu-dotatate for midgut neu- roendocrine tumors. N Engl J Med. 2017;376:125–35. https:// doi. org/ 10.1 056/ NEJMo a1607 427. Studer D, Dyrauf P, Naubereit P, Heinke R, Wendt K. Resonance ionization spectroscopy in dysprosium. Hyperfine Interact. 2016;238:8. https:// doi.o rg/ 10. 1007/ s10751-0 16- 1384-4. Sun JX, Walter B, Sandefer EP, Page RC, Digenis GA, Ryo UY, et al. Explaining variable absorption of a hypolipidemic agent (CGP 43371) in healthy subjects by gamma scintigraphy and pharmacokinetics. J Clin Pharmacol. 1996;36:230–7. https:// doi.o rg/ 10. 1002/j. 1552-4 604.1 996. tb041 92.x. Szkliniarz K, Jastrzebski J, Bilewicz A, Chajduk E, Choinski J, Jakubowski A, et al. Medical radioisotopes produced using the alpha particle beam from the Warsaw heavy ion cyclotron. Acta Phys Pol A. 2015;127:1471–4. https:// doi. org/ 10. 12693/ APhys PolA. 127.1 471. Szkliniarz K, Sitarz M, Walczak R, Jastrzębski J, Bilewicz A, Choiński J, et al. Production of medical Sc radioisotopes with an alpha particle beam. Appl Radiat Isot. 2016;118:182–9. https://d oi. org/ 10. 1016/j.a prad iso. 2016. 07. 001. Taheri M, Azizmohammadi Z, Ansari M, Dadkhah P, Dehghan K, Valizadeh R, et al. 153Sm-EDTMP and 177Lu-EDTMP are equally safe and effective in pain palliation from skeletal metastases. Nuklearmedizin. 2018;57:174–80. Talip Z, Borgna F, Müller C, Ulrich J, Duchemin C, Ramos JP, et al. Production of mass-separated erbium-169 towards the first preclinical in vitro investigations. Front Med. 2021. https://d oi. org/1 0. 3389/ fmed.2 021. 643175. Tan HY, Yeong CH, Wong YH, McKenzie M, Kasbollah A, Md. Shah MN, et al. Neutron-activated theranostic radionu- clides for nuclear medicine. Nucl Med Biol. 2020;90–91:55–68. https://d oi. org/1 0. 1016/j. nucme dbio. 2020. 09. 005. Tárkányi F, Hermanne A, Király B, Takács S, Ditrói F, Baba M, et al. Study of activation cross-sections of deuteron induced reactions on erbium: production of radioisotopes for practical applications. Nucl Instrum Methods Phys Res Sect B. 2007;259:829–35. https:// doi. org/ 10. 1016/j.n imb. 2007. 01.2 87. Tárkányi F, Takács S, Hermanne A, Ditrói F, Király B, Baba M, et al. Study of activation cross sections of proton induced reac- tions on erbium for practical applications. Nucl Instrum Methods Phys Res Sect B. 2008a;266:4872–6. https:// doi. org/ 10. 1016/j. nimb.2 008. 08.0 05. Tárkányi F, Hermanne A, Takács S, Ditrói F, Király B, Kovalev SF, et al. Experimental study of the 165Ho(p, n) nuclear reaction for production of the therapeutic radioisotope 165Er. Nucl Instrum Methods Phys Res Sect B. 2008b;266:3346–52. https://d oi. org/ 10. 1016/j.n imb. 2008. 05.0 05. Tárkányi F, Hermanne A, Takacs S, Ditroi F, Király B, Kovalev S, et al. Experimental study of the 165Ho (d, 2n) and 165Ho (d, p) nuclear reactions up to 20 MeV for production of the therapeutic radioisotopes 165Er and 166gHo. Nucl Instrum Methods Phys Res, Sect B. 2008c;266:3529–34. Tárkányi F, Takács S, Hermanne A, Ditrói F, Király B, Baba M, et al. Investigation of production of the therapeutic radioiso- tope 165Er by proton induced reactions on erbium in comparison with other production routes. Appl Radiat Isot. 2009;67:243–7. https:// doi.o rg/ 10. 1016/j. apradi so. 2008. 10.0 06. Sadler et al. EJNMMI Radiopharmacy and Chemistry (2022) 7:21 Page 54 of 55 Tárkányi F, Takács S, Ditrói F, Hermanne A, Yamazaki H, Baba M, et al. Activation cross-sections of deuteron induced nuclear reactions on neodymium up to 50 MeV. Nucl Instrum Methods Phys Res Sect B. 2014;325:15–26. https:// doi. org/1 0. 1016/j. nimb. 2014. 01. 024. Tárkányi F, Hermanne A, Ditrói F, Takács S. Activation cross sections of proton induced nuclear reactions on neodymium up to 65 MeV. J Radioanal Nucl Chem. 2017;314(2):1425–44. https:// doi. org/ 10.1 007/ s10967- 017- 5498-1. Tishchenko VK, Petriev VM, Skvortsov VG. Radiopharmaceuticals based on polyaminophosphonic acids labeled with α−, β−, and γ-emitting radionuclides (review). Pharm Chem J. 2015;49:425–31. https:// doi. org/ 10. 1007/ s11094- 015- 1299-4. Tripathi M, Singhal T, Chandrasekhar N, Kumar P, Bal C, Jhulka PK, et al. Samarium-153 ethylenediamine tetramethylene phosphonate therapy for bone pain palliation in skeletal metastases. Indian J Cancer. 2006;43:86–92. https:// doi. org/ 10. 4103/ 0019- 509x. 25890. Türler A. Matched pair theranostics. Chimia (aarau). 2019;73:947–9. https:// doi. org/ 10. 2533/ chimia. 2019.9 47. Turner JH, Claringbold PG, Hetherington EL, Sorby P, Martindale AA. A phase I study of samarium-153 ethylenediamine- tetramethylene phosphonate therapy for disseminated skeletal metastases. J Clin Oncol. 1989;7:1926–31. https:// doi. org/ 10. 1200/ JCO.1 989.7. 12. 1926. Umbricht CA, Köster U, Bernhardt P, Gracheva N, Johnston K, Schibli R, et al. Alpha-PET for Prostate cancer: preclinical investigation using 149Tb-PSMA-617. Sci Rep. 2019;9:17800. https:// doi. org/ 10. 1038/ s41598-0 19- 54150-w. Uusijärvi H, Bernhardt P, Rösch F, Maecke HR, Forssell-Aronsson E. Electron- and positron-emitting radiolanthanides for therapy: aspects of dosimetry and production. J Nucl Med. 2006;47:807–14. Van de Voorde M, Van Hecke K, Binnemans K, Cardinaels T. Separation of samarium and europium by solvent extraction with an undiluted quaternary ammonium ionic liquid: towards high-purity medical samarium-153. RSC Adv. 2018;8:20077–86. https:// doi. org/ 10.1 039/C 8RA03 279C. Van de Voorde M, Van Hecke K, Cardinaels T, Binnemans K. Radiochemical processing of nuclear-reactor-produced radiolanthanides for medical applications. Coord Chem Rev. 2019;382:103–25. https:// doi.o rg/ 10. 1016/j. ccr. 2018. 11. 007. Van de Voorde M, Duchemin C, Heinke R, Lambert L, Chevallay E, Schneider T, et al. Production of Sm-153 With very high specific activity for targeted radionuclide therapy. Front Med. 2021. https:// doi.o rg/ 10. 3389/ fmed. 2021. 675221. Van der Linden R, De Corte F, Hoste J. A compilation of infinite dilution resonance integrals, II. J Radioanal Chem. 1974;20:695–706. https:// doi. org/1 0. 1007/B F025 14313. van der Meulen NP, Bunka M, Domnanich KA, Müller C, Haller S, Vermeulen C, et al. Cyclotron production of 44Sc: From bench to bedside. Nucl Med Biol. 2015;42:745–51. https:// doi.o rg/ 10. 1016/j. nucmed bio.2 015. 05.0 05. Vander Hoogerstraete T, Binnemans K. Highly efficient separation of rare earths from nickel and cobalt by solvent extrac- tion with the ionic liquid trihexyl(tetradecyl)phosphonium nitrate: a process relevant to the recycling of rare earths from permanent magnets and nickel metal hydride batteries. Green Chem. 2014;16:1594–606. https:// doi. org/1 0. 1039/ c3gc4 1577e. Vats K, Das T, Sarma HD, Banerjee S, Pillai MR. Radiolabeling, stability studies, and pharmacokinetic evaluation of thulium- 170-labeled acyclic and cyclic polyaminopolyphosphonic acids. Cancer Biother Radiopharm. 2013;28:737–45. https:// doi. org/1 0. 1089/c br. 2013. 1475. Vaudon J, Frealle L, Audiger G, Dutillly E, Gervais M, Sursin E, et al. First steps at the cyclotron of Orléans in the radiochem- istry of radiometals: 52Mn and 165Er. Instruments. 2018;2:15. Vaughn BA, Koller AJ, Chen Z, Ahn SH, Loveless CS, Cingoranelli SJ, et al. Homologous structural, chemical, and biological behavior of Sc and Lu complexes of the picaga bifunctional chelator: toward development of matched theranos- tic pairs for radiopharmaceutical applications. Bioconjug Chem. 2021;32:1232–41. https://d oi. org/1 0. 1021/ acs. bioco njchem. 0c005 74. Vente MA, Hobbelink MG, van Het Schip AD, Zonnenberg BA, Nijsen JF. Radionuclide liver cancer therapies: from concept to current clinical status. Anticancer Agents Med Chem. 2007;7:441–59. https:// doi. org/ 10.2 174/ 18715 20077 81058 569. Vimalnath KV, Das MK, Venkatesh M, Ramamoorthy N. Prospects and problems in the production of 143Pr for radionu- clide therapy applications. Radiochim Acta. 2005;93:419–26. https:// doi. org/ 10. 1524/ ract. 2005.9 3.7. 419. Vosoughi S, Jalilian AR, Shirvani-Arani S, Bahrami-Samani A, Salek N. Preparation of 166Dy/166Ho-chitosan as an in vivo generator for radiosynovectomy. J Radioanal Nucl Chem. 2017a;311:1657–64. https:// doi.o rg/ 10. 1007/ s10967- 016-5 146-1. Vosoughi S, Shirvani-Arani S, Bahrami-Samani A, Salek N, Jalilian AR. Production of no-carrier-added Ho-166 for targeted therapy purposes. Iran J Nucl Med. 2017b;25:15–20. Walczak R, Krajewski S, Szkliniarz K, Sitarz M, Abbas K, Choiński J, et al. Cyclotron production of 43Sc for PET imaging. EJNMMI Phys. 2015;2:33. https:// doi. org/ 10. 1186/ s40658- 015- 0136-x. Webster B, Ivanov P, Russell B, Collins S, Stora T, Ramos JP, et al. Chemical purification of terbium-155 from pseudo- isobaric impurities in a mass separated source produced at CERN. Sci Rep. 2019;9:10884. https:// doi.o rg/ 10. 1038/ s41598- 019- 47463-3. Yagi M, Kondo K. Preparation of carrier-free 47Sc by the 48Ti (γ, p) reaction. Int J Appl Radiat Isot. 1977;28:463–8. https:// doi. org/1 0. 1016/ 0020-7 08X(77) 90178-8. Yang JJ, Yang J, Wei L, Zurkiya O, Yang W, Li S, et al. Rational design of protein-based MRI contrast agents. J Am Chem Soc. 2008;130:9260–7. https:// doi. org/ 10. 1021/ ja800 736h. Yang S-C, Kim K, Song T-Y, Lee Y-O, Kim G. Production cross sections of products in the proton induced reactions on natNd in the energy region up to 45 MeV. Nucl Instrum Methods Phys Res Sect B. 2015;362:142–50. https:// doi. org/ 10. 1016/j. nimb. 2015. 09. 061. Yasui LS, Hughes A, DeSombre ER. Relative biological effectiveness of accumulated 125IdU and 125I-Estrogen decays in estrogen receptor-expressing MCF-7 human breast cancer cells. Radiat Res. 2001;155:328. Yokoyama M, Shiraishi K. Stability evaluation of Gd chelates for macromolecular MRI contrast agents. Magn Reson Mater Phys Biol Med. 2020;33:527–36. https:// doi. org/ 10. 1007/ s10334-0 19- 00805-8. S adler et al. EJNMMI Radiopharmacy and Chemistry (2022) 7:21 Page 55 of 55 Yousefnia H, Zolghadri S, Jalilian AR, Tajik M, Ghannadi-Maragheh M. Preliminary dosimetric evaluation of 166Ho-TTHMP for human based on biodistribution data in rats. Appl Radiat Isot. 2014;94:260–5. https://d oi. org/ 10. 1016/j. aprad iso. 2014. 08. 017. Zandi N, Sadeghi M, Afarideh H. Evaluation of the cyclotron production of 165Er by different reactions. J Radioanal Nucl Chem. 2013;295:923–8. https:// doi. org/ 10. 1007/ s10967- 012-2 116-0. Zerkin VV, Pritychenko B. The experimental nuclear reaction data (EXFOR): extended computer database and web retrieval system. Nucl Instrum Methods Phys Res Sect A. 2018;888:31–43. https:// doi. org/ 10. 1016/j. nima. 2018. 01. 045. Zolghadri S, Jalilian AR, Naseri Z, Yousefnia H, Bahrami-Samani A, Ghannadi-Maragheh M, et al. Production, quality control and biological evaluation of (166)Ho-PDTMP as a possible bone palliation agent. Iran J Basic Med Sci. 2013;16:719–25. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.