Abstract
In recent years, new α-particle–, β−-particle–, and Auger electron–emitting radiometals—such as 67Cu, 47Sc, 166Ho, 161Tb, 149Tb, 212Pb/212Bi, 225Ac, and 213Bi—have been produced and evaluated (pre)clinically for therapeutic purposes. In this short review article, the most important routes of production of these radiometals are critically discussed, as are examples of their application in preclinical and clinical studies.
Radionuclide tumor therapy has been used successfully for the treatment of disseminated disease (1). However, clinical application of therapeutic radionuclides is often driven by the availability of radionuclides rather than appropriate decay characteristics for a given disease indication. 90Y and 177Lu are routinely used for targeted radionuclide therapy (1). 90Y decays with a shorter half-life (t1/2) than 177Lu.
The emission of high-energy β−-particles has been used for radionuclide therapy in combination with peptides (e.g., 90Y-DOTATOC) and antibodies (e.g., ibritumomab tiuxetan [Zevalin; Spectrum Pharmaceuticals, Inc.]) (1). 177Lu emits low-energy β−-particles as well as γ-radiation, useful for dosimetry (Table 1) (1). It has been used for peptide receptor–targeted radionuclide therapy (2) and in combination with small molecules, such as prostate-specific membrane antigen (PSMA) ligands (3). Because of the relatively long range of β−-particles in tissue, tumor cells that are not directly targeted may also be affected. This so-called “crossfire” effect renders β−-emitters more suited for the therapy of larger metastases (4).
In contrast to the low linear energy transfer (LET; ∼0.2 keV/mm) of β−-particles, α-particles have a high LET (80–100 keV/μm) but a shorter tissue range (4). The promising potential of high-LET radionuclides was demonstrated recently using 223RaCl2 (Xofigo; Bayer), which revived interest in α-emitters (5). Auger electrons induce multiple ionizations (LET, 4–26 keV/μm) in the immediate vicinity of the decay site and, thus, are promising for the treatment of single cancer cells (4).
In the introduction of new radionuclides for therapy, several properties should be taken into consideration. For the treatment of disseminated disease, the emission of high-LET particles is of particular interest; the coemission of radiation (for PET or SPECT) that can be imaged or the existence of a readily available diagnostic match is clearly an advantage. Moreover, the chemical properties of the radiometal should allow stable coordination using standard chelators. The half-life of the radiometal should match the clinical indications and logistics. The method of production needs to be safe and affordable to allow continuous worldwide supply of the radionuclide at a high quality and in a sufficient quantity.
Here we discuss the therapeutic radionuclides 47Sc, 67Cu, 149Tb, 161Tb, 166Ho, 212Pb/212Bi, 225Ac, and 213Bi (Table 1), selected in accordance with the aforementioned criteria.
COPPER-67
67Cu is a low-energy β−-emitter with γ-ray emission, useful for SPECT (Table 1). Studies using reactors and accelerators to produce 67Cu have been performed (6). The 68Zn(p,2p)67Cu route of production (proton energy, >70 MeV) appears to be most attractive (7); however, besides the desired 67Cu, large quantities of 64Cu (t1/2, 12.7 h) and 67Ga (t1/2, 78.3 h) are coproduced. The described separation methods required sizable columns containing chelating resin for primary separation to recover trace amounts of 67Cu from gram quantities of Zn target material (6,7). At present, there is a shortage of 67Cu supply, mainly because of the lack of cyclotrons producing high-energy protons.
Chelation of copper nuclides with a variety of macrocycles, including 1,4,7,10-tetraazacyclododecane-1,4,7,10-tetraacetic acid (DOTA), 1,4,8,11-tetraazacyclotetradecane-N,N′,N″,N‴-tetraacetic acid (TETA), 1,4,7-triazacyclononane-N-glutaric acid-N′,N″-diacetic acid (NODAGA), and 4-[(1,4,8,11-tetraazacyclotetradec-1-yl)methyl]benzoic acid (CPTA), as well as cross-bridged macrocycles (e.g., 1,4,8,11-tetraazabicyclo[6.6.2]hexadecane-4,11-diacetic acid [CB-TE2A]) and cage-type chelates (e.g., 1,8-diamino-3,6,10,13,16,19-hexaazabicyclo[6,6,6]-eicosane [DiAmSar]) (8,9), has been reported. Generally, the chelation of copper with tetraaza-macrocycles results in low tumor-to-background ratios because of the transchelation of copper to proteins accumulating in the liver (8). However, Zimmermann et al. (10) and Grünberg et al. (11) reported the successful use of DOTA and CPTA chelators in preclinical studies performed with a 67Cu-labeled anti–L1-CAM antibody and F(ab′)2 fragments.
In a pilot study, 67Cu-labeled anti-MUC1 mucin antibody C595 was evaluated for targeting bladder cancer after intravesical administration into cystectomy specimens (12). A study was performed in non-Hodgkin lymphoma patients with 67Cu-labeled TETA–Lym-1 antibody at diagnostic quantities followed by 4 therapy cycles (13). This radioimmunotherapy was revealed to be safe and effective (14,15). In 6 patients with colorectal tumors, 67Cu-labeled CPTA-mAB35 used for imaging purposes before surgery revealed high tumor uptake and favorable tumor-to-blood ratios relative to the results obtained with 131I-labeled antibody (16). These promising clinical results and the relatively easy access to 64Cu as a diagnostic match warrant further efforts to fully assess the therapeutic potential of 67Cu.
SCANDIUM-47
47Sc has decay characteristics similar to those of 67Cu and also has with 44Sc its theranostic match for PET (Table 1) (17). 47Sc can be produced with neutrons via the 47Ti(n,p)47Sc (18,19) or 46Ca(n,γ)47Ca → 47Sc (19,20) nuclear reaction. The production of 47Ca via the 48Ca(γ,n)47Ca nuclear reaction with electron linear accelerators was also evaluated (21). The production of 47Sc via 48Ti(p,2n)47Sc was attempted, but too much of the long-lived 46Sc was coproduced (19). The calcium route is straightforward because easy separation and ingrowing product allow for repeated separations (generator principle) (20).
With regard to complex formation and stability, it was revealed that DOTA forms stable scandium complexes (22). Recently, more specific chelators for scandium, such as monophosphorus acid DOTA analogs and AAZTA, were developed (23,24). The therapeutic potential of 47Sc was demonstrated with a DOTA-folate conjugate; reduced tumor growth and increased average survival time in mice were observed (17). Moreover, high-quality SPECT images were obtained in mice that received 47Sc-DOTA-folate (17). On the basis of these data, 47Sc seems to have promise for clinical application.
HOLMIUM-166
166Ho, a high-energy β−-emitter similar to 188Re (Eβ− average, 763 keV; t1/2, 17.0 h), coemits γ-rays useful for SPECT (Table 1). 166Ho is most frequently produced via the 165Ho(n,γ)166Ho route in combination with poly–lactic acid microspheres for intraarterial radioembolization in patients with liver metastases (25). For this purpose, natHo is incorporated into microspheres and activated with neutrons (26). 166Ho-microspheres administered intraarterially accumulated about 6-fold more in the tumor than in normal liver tissue in rats (27). Toxicity studies revealed no clinically relevant side effects in pigs (28). A phase 1 trial of intraarterial radioembolization using 166Ho-labeled poly–lactic acid microspheres was designed for the treatment of patients with liver cancer (25). 166Ho-based microspheres are considered to be superior to 90Y-based glass or resin microspheres because of the low cost of production of 166Ho and the possibility of SPECT imaging preceding therapy (26,29). Moreover, because holmium is highly paramagnetic, it can be visualized using MRI (25).
Clinical studies were performed to determine the safety and efficacy of treatment of hepatocellular carcinoma with a percutaneously administered 166Ho-chitosan complex (30). The 166Ho-chitosan complex therapy was efficient in terms of response and survival, and toxicity was acceptable, especially in patients with smaller tumors (30).
166Ho can also be produced via the 164Dy(2n,γ)166Dy → 166Ho nuclear reaction, yielding a carrier-free product useful for labeling of biomolecules (31). Among several investigated chelators, DOTA was found to be favorable and the kinetics of distribution of 166Ho-DOTA were similar to those of 177Lu-DOTA (32). The clinical safety and efficacy of 166Ho-labeled macrocyclic tetraphosphonate-1,4,7,10-tetraazacyclododecane-1,4,7,10-tetramethylene-phosphonic acid (DOTMP) were assessed for skeleton-targeted radiotherapy of breast cancer–related bone metastases (33). The median overall survival time was 39.9 mo, and 2 patients remained progression-free for more than 6 y (33). A multicenter dose escalation study demonstrated the potential of 166Ho-DOTMP for the treatment of multiple myeloma (34).
TERBIUM-161
161Tb has chemical and physical characteristics similar to those of 177Lu but coemits a substantial number of Auger electrons in addition to β−-particles (35,36). 161Tb can be produced at a high specific activity by irradiation of enriched 160Gd in a reactor with a high neutron flux via the 160Gd(n,γ)161Gd→161Tb nuclear reaction (35). The separation of 161Tb from the gadolinium target material was performed by cation-exchange chromatography with α-hydroxyisobutyric acid followed by concentration of the 161Tb solution (37), similar to the production of 177Lu. Müller et al. were the first to use 161Tb in a preclinical therapy setting, with tumor-bearing mice (37). An improved therapeutic effect of 161Tb compared with that of 177Lu was described in 2 independent preclinical studies (38–40). The coemitted Auger electrons appeared to be favorable for therapeutic purposes, but the clinical superiority of 161Tb over 177Lu remains to be assessed when a steady supply of 161Tb can be guaranteed.
TERBIUM-149
149Tb was first proposed for targeted α-therapy by Allen and Blagojevic (41). Unlike most α-emitters, it decays predominantly by the emission of a single low-energy α-particle, positrons, and γ-radiation (Fig. 1A) (42). Beyer et al. chose the Nd(12C,5n)149Dy→149Tb production route at the European Organization for Nuclear Research (42). Later, 149Tb was produced by proton-induced spallation of a tantalum target followed by an online isotope separation process (37,42,43). In this scenario, 149Tb had to be separated from isobars and pseudoisobars with a mass of 149 using cation-exchange chromatography (37,42,43).
The first preclinical therapy study with 149Tb was performed in a mouse model of leukemia with 49Tb-labeled cyclohexane-DTPA–functionalized rituximab (43). This therapy resulted in the long-term survival of mice without evidence of recurrence 120 d after treatment, whereas untreated mice or mice treated with a high dose of rituximab reached endpoint criteria after 37 or 43 d, respectively (43). Müller et al. reported the potential of 149Tb-DOTA-folate to treat KB tumor–bearing mice (37,44). The same group also investigated the positron emission of 149Tb, potentially allowing PET simultaneously with α-therapy (Fig. 1B) (45).
LEAD-212 (AND BISMUTH 212)
The β−-emitter 212Pb has been proposed as an in vivo α-emitter generator because of its α-emitting daughter nuclide, 212Bi (Fig. 2). 203Pb was recently described as a diagnostic match that can be readily produced from natTl via the 203Tl(p,n)203Pb nuclear reaction (46). The supply of 212Pb is based on the availability of 228Th extracted from spent nuclear fuel. Because 228Th-based generators were affected by radiolytic damage, a generator based on 224Ra (t1/2, 3.66 d) was developed using 228Th (47). The elution of 212Pb from the 224Ra/212Pb generator requires several steps, including separation from the daughter nuclide, 212Bi (48).
212Pb can be coordinated using a DOTA chelator or S-2-(4-isothiocyanatobenzyl)-1,4,7,10-tetraaza-1,4,7,10-tetra(2-carbamoylmethyl)cyclododecane (TCMC), a macrocyclic chelator specifically developed for the complexation of lead (48). 212Pb-labeled antibodies were investigated in several studies, including human epidermal growth factor 2 (HER2)–targeting trastuzumab in mouse models of peritoneal cancer or orthotopic models of prostate cancer (48). Two clinical studies performed in patients with HER2-positive cancer to determine the safety, distribution, and pharmacokinetics of 212Pb-TCMC-trastuzumab and to determine dosimetry in a dose-escalation study indicated good tolerance and no evidence of radiation-related toxicity (49,50).
ACTINIUM-225
The decay of 225Ac results in 6 daughters (221Fr, 217At, 213Bi, 213Po, 209Pb, and 209Bi) with several α- and β−-disintegrations (Fig. 3) (51). 213Bi has been investigated extensively for α-therapy in (pre)clinical studies (52). 225Ac can be obtained in limited quantities (∼37 GBq/y) by radiochemical separation from a 229Th source (53,54). It can also be produced via the 232Th(p,2p6n)225Ac (55,56), 226Ra(p,2n)225Ac (57), and 226Ra(γ,n)225Ra → 225Ac nuclear reactions (58). The separation of 225Ac from the 226Ra target material was performed using lanthanide extraction resin chromatography (53). For the separation of 225Ac from its target material, a cumbersome 5-column system was reported (54). Because any target material used is radioactive (and long-lived), additional safety precautions must be taken into consideration during production.
The use of 225Ac may be challenging because of its decay chain and the fact that the first disintegration leads to the destruction of the metal complex (because of the high recoil energy) and the subsequent mobilization of the daughter radionuclides—a feature shared by all α-particle–emitting radionuclides (59–61).
DOTA chelators have been used to conjugate 225Ac to antibodies and small molecules (62). The effectiveness of 225Ac-labeled antibodies (such as anti-CD33, anti-CD18, anti-HER2/neu, and anti-PSMA) was demonstrated in vitro against leukemia and lymphoma as well as breast, ovarian, and prostate cancer cells (63). The treatment of prostate tumor–bearing mice with the PSMA-specific antibody 225Ac-J591 improved survival time over that of mice receiving unlabeled J591 (63). Other preclinical studies demonstrated the successful treatment of AR42J tumor–bearing mice with 225Ac-DOTATOC (64) and dose-dependent antitumor effects after treatment with 225Ac-labeled trastuzumab in an intraperitoneal ovarian tumor mouse model (65).
A phase 1 study with the anti-CD33 antibody 225Ac-lintuzumab (HuM195) was performed in patients with relapsed/refractory acute myeloid leukemia (66). Across all dose levels, antileukemic activity was evident. No acute toxicity other than liver function abnormalities occurred (66). A multicenter phase 1 trial with 225Ac-lintuzumab in combination with low-dose cytarabine resulted in a remarkable response represented by bone marrow blast reduction in more than 50% of treated patients (67).
Recently, Kratochwil et al. reported on first-in-human PSMA-targeted α-therapy of metastatic prostate cancer with 225Ac-labeled PSMA-617 (Fig. 4) (68). Strong antitumor activity and good tolerability were observed when 225Ac-PSMA-617 was applied in 3 cycles at bimonthly intervals (69). The 6-mo follow-up revealed a better response in patients treated with 225Ac-PSMA-617 than in patients treated with 177Lu-PSMA-617; however, frequent and more severe xerostomia occurred (70). The impressive clinical results in highly challenging clinical situations demonstrated the effectiveness of the α-emitter (68).
BISMUTH-213
213Bi is available from a 225Ac/213Bi generator (Table 1; Fig. 2). Successful α-therapy with 213Bi has been demonstrated in many preclinical studies and several clinical trials (52). Clinically, 213Bi-labeled substance P was used for local administration in patients with located gliomas (71). High retention of the activity at the target site and radiation-induced necrosis of tumors were observed, without relevant acute local or systemic toxicity (71). The application of 213Bi-DOTATOC resulted in long-lasting antitumor responses in all treated patients (72). Despite the short physical half-life, even systemic application of 213Bi-lintuzumab in 31 patients was effective and resulted in remissions in patients with acute myeloid leukemia (73).
CONCLUSION
Here we discussed several radionuclides with a high therapeutic potential because of their decay properties. For most of them, availability is limited and this factor affects cost negatively. Depending on demand from clinics, the cost can decrease if production takes place in the private sector. Future endeavors of physicists, radiochemists, and radiopharmacists should focus on the development of reliable and efficient production methods. The various physical decay properties, such as variable half-lives and the emission of high- and low-LET particles, could allow the application of the most appropriate radionuclide for a given cancerous disease. Given these features, systemic radionuclide therapy fits the concept of personalized medicine perfectly and, thus, is among the most modern therapeutic strategies to be pursued.
DISCLOSURE
No potential conflict of interest relevant to this article was reported.
- © 2017 by the Society of Nuclear Medicine and Molecular Imaging.
REFERENCES
- Received for publication February 7, 2017.
- Accepted for publication March 13, 2017.