Search of ligands suitable for 212Pb/212Bi in vivo generators

The short half-life of 212Bi and 213Bi limits the application of these radionuclides in α radionuclide therapy. The labeling of biomolecules with 212Pb (mother nuclide of 212Bi) instead of 212Bi or 213Bi has the advantage of obtaining a conjugate with a half-life of 10.6 h, compared with of 60 min for 212Bi or 46 min for 213Bi. Previous attempts to prepare a potential in vivo generator with 212Pb complexed by the DOTA chelator failed, because about 36 % of Bi was reported to escape as a result of the radioactive decay \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$^{{212}}{\text{Pb}}{\mathop{\longrightarrow}\limits^{\beta ^{ - }}}{^{212}{\text{Bi}}}$$\end{document}. Herein, we report studies on the stability of the 212Pb complexes with eight selected polydentate ligands, which demonstrate high affinity for 3+ metal cations. From the ligand studied DOTP and BAPTA show a sufficient 212Pb labeling yields but only 212Pb–DOTP complex is stable in isotonic solution of sodium chloride making this way radioactivity level of released 212Bi is below the limit of detection. It should be emphasized that the DOTP complex is stable only in the case when the concentration of free DOTP exceeds 10−4 M.


Introduction
In the field of targeted radiotherapy, the selection of radionuclide depends on the type of the treated disease. Solid tumors are generally treated with high and medium energy b --emitters such as 90 Y, 188 Re and 131 I, because their b --particles have a tissue range of several millimeters. The effective tissue range of b --particles is not optimal for treatment of tumors forming small clusters of cells and for treatment of single cancer cells and micrometastases. Treatment of these neoplastic diseases could be more effective with a-emitters, which combine short range with high linear energy transfer, combination that results in the relatively high biological effect and cytotoxicity [1]. Owing to this, a-particles are able to make lethal double strand breaks in DNA. When the double stranded DNA molecules breaks, there is very little chance to repair such damage. Humm and Cobb [2] reported that to attain single cell kill probability of 99.99 % tens of thousands of b-decays at the cell membrane are required, whereas in the case of a-emitters only few a-decays at the cell membrane are sufficient to kill malignant cells. Due to high radiotoxicity of a-particles, high degree of accuracy is required to deliver the radiation to the target cells without targeting normal cells. From the medical point of view, a-particles can be used either for treatment of cancer micro-metastasis, or to destroy tumor margins after surgical resection. Another potential application is in treating cancers such as lymphoma and leukemia, which are present as free-floating tumor cells in the circulation system [3]. Till now, only few clinical studies with 213 Bi and 211 At labeled peptides and monoclonal antibodies have demonstrated the potential of alpha particle emitting isotopes in radionuclide therapy [4,5].
There are only few a-particle emitting radionuclides, which have properties suitable for developing therapeutic Centre for Radiochemistry and Nuclear Chemistry, Institute of Nuclear Chemistry and Technology, Dorodna 16 03-195, Warszawa, Poland e-mail: a.bilewicz@ichtj.waw.pl radiopharmaceuticals: generator-obtained 212 Bi (t 1/2 = 60 min), 213 Bi (t 1/2 = 46 min), 226 Th (t 1/2 = 30 min), 225 Ac (t 1/2 = 10 days), 227 Th (t 1/2 = 18.7 days), as well as the cyclotron-produced 211 At (t 1/2 = 7.2 h). The available aemitters have serious shortcomings, because in the case of 225 Ac and 227 Th the designed ligand must form chemically stable complexes with both parent and decay radionuclides. 225 Ac decays directly to 221 Fr (alkali metal), which has a halflife of 4.9 min and escapes from 225 Ac-radiobioconjugate. Similar situation appears in the case of 227 Th, where the decay product, the gaseous 219 Rn, easily liberates itself from 227 Thradioconjugate. Application of 211 At is limited, because astatine as the heaviest halogen forms weak bond with a carbon atom in the biomolecule. Therefore, 211 At-bioconjugates are unstable under physiological conditions.
In the case of 212 Bi, 213 Bi and 226 Th short half-life often limits the application of these nuclides to situations when the tumor cells are rapidly accessible to the targeting agent. However, the short half-life of 212 Bi could be effectively lengthened by chelation of the parent 212 Pb radionuclide (t 1/2 = 10.6 h) to a biomolecule [6]. In comparison with direct use of 212 Bi, radiopharmaceuticals based on 212 Pb would have much broader applicability, because the half-life of 212 Pb corresponds better with the pharmacokinetics of various biomolecules. Moreover, the 212 Pb-212 Bi in vivo generator delivers the dose per unit of administered activity ten times greater than that in the case of 212 Bi alone or of the 213 Bi a-emitter [7]. Thus, the required activity of the radiopharmaceutical preparation would be greatly reduced, and making this way generation and administration of the a-emitting radiopharmaceutical much easier.
It is very important that 212 Bi formed in the b --decay of 212 Pb remains bound to the carrier. This is because free bismuth localizes in the kidneys, prohibiting this way the use of structures that are not effective in stabilizing 212 Bi in vivo [8]. In theory, the decay of 212 Pb should not generate a problem with retention of 212 Bi. The calculated recoil energy of the Bi nucleus is only about 0.5 eV. This is not sufficient to break a chemical bond, which requires about 10 eV. However, over 30 % of the c-rays emitted when 212 Pb decays are internally converted during the decay time. The resulting cascade of conversion electrons brings 212 Bi to highly ionized states such as Bi 5? and Bi 7? , hence the energy required to neutralize the charge is sufficient to break chemical bonds [9]. The potential use of 212 Pb as an in vivo generator has been studied in earlier works [8,10,11]. Previous attempts to prepare a potential in vivo generator with 212 Pb complexed by the DOTA chelator [11] failed, because about 36 % of Bi was reported to escape as a result of the radioactive decay 212 Pb À! b À 212 Bi. Because the free highly energetic radiobismuth escapes from the complex during the decay, toxicity emerges when unchelated 212 Bi accumulates in various organs, mainly in kidneys. Formation of kinetically inert Bi 3? -DOTA complexes is very slow, therefore liberated 212 Bi very poorly reassociates with DOTA.
In this paper we report the formation and stability studies of 212 Pb complexes with various polydentate ligands exhibited faster than DOTA kinetics of complex formation.

Lead-212
The 1 MBq of 212 Pb (t 1/2 = 60 min) was obtained from 232 U as one of the decay products. Separation of 212 Pb from 232 U and other decay products was performed in a two-step procedure. In the first step, 224 Ra was eluted by 0.1 M HNO 3 from HDEHP-Teflon column loaded with 232 U. In the second step 212 Pb was separated from 224 Ra on cation exchange resin Dowex 50 9 8 by elution with 1.0 M HCl. The effluent was acidified with HNO 3 , evaporated and the residue was dissolved in 0.01 M HNO 3 .

Measurements
The radioactivity was measured by c-spectrometer using the HPGe detector (Canberra) with associated electronics (resolution 2.09 keV for 1,332 keV 60 Co line, efficiency ca. 30 %), coupled to the multichannel analyzer TUKAN (The Andrzej Soltan Institute for Nuclear Studies, Świerk, Poland).

Synthesis of radiolabeled complexes
The experimental conditions for labeling, such as the metal-to-ligand molar ratio, pH, time of reaction and temperature, were optimized to achieve a high complexation efficiency. The 212 Pb complexes with the studied ligands were synthesized by mixing 50 ll of non-carrieradded 212 Pb in 0.01 M HNO 3 with 5 ll of either 10 -1 or 10 -2 M solution of the respective ligand. The volume of solution was adjusted to 500 ll by adding 0.01 M CH 3 COONH 4 solution, and pH were settled at pH 6 or 7 using 2 M NaOH. Complexes with acyclic ligands were prepared at room temperature in 2 h.

Determination of labeling efficiency and assay
The determination of labeling efficiency was achieved in accordance with the modified procedure proposed by Mirzadeh et al. [9] by isolation of uncomplexed cations by the use of chelating Chelex 100 resin in a small column (d = 3 mm, h = 10 mm). In preliminary experiments we found that when solution containing nca 212 Pb and 212 Bi was loaded on the column all activity remained on the column, even after elution with 0.1 M NH 4 NO 3 . In next step the 212 Pb and 212 Bi radionuclides were quantitatively eluted with 2 ml of 5 M HCl. We assumed that under the same conditions the negatively charged complexes of Pb 2? and Bi 3? would be eluted from the column by 0.1 M NH 4 NO 3 . This separation procedure was tested on Pb and Bi complexes formed by 0.01 M DOTA and DTPA ligands, and we found that these complexes were completely eluted by 2 ml of 0.1 M NH 4 NO 3 .

Assay of 212 Bi after decay of 212 Pb-L complexes
The complexes were prepared as described above. Concentration of the synthesized complexes was decreased using isotonic solution of sodium chloride (0.9 % NaCl solution), in order to obtain 0.5 ml samples. Solutions were incubated for 4 h to attain 212 Pb-212 Bi radioactive equilibrium and then in order to separate complexes from the uncomplexed cations the solution was loaded on the column filled with Chelex 100 resin (3 9 10 mm). To achieve the separation the column was washed with 2 ml of 0.1 M NH 4 NO 3 solution which eluted the complexes. The retained uncomplexed 212 Pb and 212 Bi cations were next eluted with 2 ml of 5 M HCl. The activities of the eluted fractions were measured over 5 h time period.

Results and discussion
The labeling of biomolecules with 212 Pb instead of 212 Bi or 213 Bi has the advantage of obtaining a conjugate with a halflife of 10 h, instead of 60 min for 212 Bi or 46 min for 213 Bi.
Therefore, when 212 Pb labeled conjugate is used, the delivered dose is much greater per unit of administered activity than in the case of 212/213 Bi conjugates [7]. As noted in [12] a dose of 10 mCi of 212 Pb was equally effective as a 500 mCi injected dose of 213 Bi. However, as reported by Mirzadeh et al. [9] and Miao et al. [13] approximately one-third of the radioactivity escaped from the DOTA chelator due to ionization associated with the decay of 212 Pb to 212 Bi. In the case of radiobioconjugate Fu-Min Su et al. [14] found that 212 Pb-DOTA-biotin was initially stable, but 30 % of 212 Bi activity was released from the DOTA-biotin in 4 h. This result is in agreement with that reported by Mirzadeh et al. [9] who found that 36 % of 212 Bi activity was released from 212 Pb-DOTA in the decay.
Redistribution was not a concern for 212 Pb internalized in tumor cells, since diffusion of metal ions across the cell membrane would be very slow. However, loss of 212 Bi from circulating 212 Pb-bioconjugate could allow 212 Bi to redistribute and irradiate normal organs.
In the previous studies, DOTA and its N,N,N,N-tetraamide analog were used for binding 212 Pb to biomolecules [11]. In our opinion, because formation of kinetically inert Bi 3? -DOTA complex is very slow, the released 212 Bi from the 212 Pb-DOTA complex very poorly reassociates with DOTA. In our studies, we examined selected acyclic and cyclic polyaminopolicarboxylate ligands, which form complexes with bismuth cations more rapidly than does DOTA. The ligands demonstrating high affinity for 3? metal cations like Fe 3? and lanthanides were selected for our studies. The structure of the ligands is presented in the Fig. 1.
From the studied ligands DOTP and BAPTA are the only two, which can be taken into consideration for designing new applicable radioconjugates, because they demonstrate sufficient labeling yields Table 1. The high yield of labeling can be achieved only in the case, when the ligand concentration exceeds 10 -4 M. The remaining ligands form complexes with 212 Pb with too low efficiency. Therefore, only the 212 Pb-DOTP and 212 Pb-BAPTA complexes were selected for studying stability in isotonic solution of sodium chloride (0.9 % NaCl).
As shown in Table 2 the 212 Pb-DOTP complex is stable in isotonic solution of sodium chloride, because at DOTP concentration of 10 -4 M only very small amount of 212 Pb escapes into solution. The radioactivity level of released 212 Bi is under the limit of detection. Comparison of our results with those on 212 Pb-DOTA, described by Mirzadeh et al. [9], shows that DOTA forms with 212 Pb kinetically inert complexes. Unfortunately, 212 Bi the decay product of 212 Pb, released to solution very poorly reassociates with DOTA. On the contrary, DOTP forms with 212 Pb more labile complexes, for which the escaped 212 Bi easily reassociates with the ligand. It should be emphasized that 212 Pb-DOTP is stable only in the case when concentration of the free ligand exceeds 10 -4 M.
The results obtained show that DOTP could be used as a ligand in designing 212 Pb/ 212 Bi in vivo generators, but only Search of ligands suitable for 212 Pb/ 212 Bi 207 in the case when high specific activity of the radiopharmaceutical is not required, as it happens in palliation therapy of bone metastasis.  The activity of the 212 Pb-DOTP solution was 2.6 9 10 4 cpm and that of 212 Pb-BAPTA 2.5 9 10 4 cpm