Summary

Radionuclides emitting short-range, low-energy Auger electrons are of particular interest for radiotherapy, allowing precise targeting of cancer cells, including micrometastases and circulating tumor cells. Among the Auger electron emitters, ²⁰¹Tl has been identified as one of the most promising radionuclides for targeted radiotherapy. It has one of the highest numbers of Auger and other secondary electrons emitted per decay (average 36.9), a convenient half-life (73 h), a stable daughter product (²⁰¹Hg), and worldwide availability due to its previous extensive widespread use in myocardial imaging. Additionally, it exhibits adequate molar activity and favorable cellular dosimetry. Indeed, our preliminary in vitro studies demonstrated the significant radiotoxic potential of internalized ²⁰¹Tl, leading to increased nuclear DNA damage and clonogenic toxicity compared to clinically evaluated Auger electron-emitters such as ¹¹¹In and ⁶⁷Ga. However, a major hurdle identified for therapeutic applications of ²⁰¹Tl is the lack of effective chelators with which to incorporate it into tumor-selective targeting vehicles.
Efforts have previously been made to chelate ²⁰¹Tl in the 1+ and 3+ oxidation states. Kryptofix derivatives were evaluated as chelators for [²⁰¹Tl]Tl⁺ to selectively deliver it to prostate cancer cells upon conjugation with a PSMA-targeting motif; while they exhibited high radiolabeling yield and in vitro stability in serum, both alone and in the conjugates, incubation with cells resulted in the release of Tl⁺ from the complexes. A range of commonly-used chelators, including DOTA, DTPA, EDTA, and picolinic acid derivatives, were evaluated for binding [²⁰¹Tl]Tl³⁺, but the complexes were found to be unstable, likely due to the easy and quick reduction of Tl³⁺ to the 1+ oxidation state, which results in metal dissociation from the complex.
A substance known for its ability to bind thallium(I) is Prussian blue [PB, ferric hexacyanoferrate(II), (Fe^(III)₄[Fe^(II)(CN)₆]₃·xH₂O)], which is registered by the US Food and Drug Administration (FDA) for treating radioactive/nonradioactive thallium and radioactive cesium poisoning. It is a dark blue pigment that forms a cubic crystalline structure composed of Fe²⁺ and Fe³⁺ linked through coordinated cyanide groups. It exhibits high affinity for binding monovalent thallium ions (Tl⁺) within the interstitial spaces created by the lattice structure. Recently, PB has regained attention from the scientific community, as it can be easily assembled into a diverse range of nanoparticles (PBNPs) of different shapes and sizes with unique and tunable physicochemical properties, leading to wide-ranging applications as ion exchange materials, ion batteries, photomagnets, electrochemical sensors, and biosensors. Moreover, the water dispersibility and biocompatibility of PBNPs have encouraged exploration of their uses in biomedical research. For example, PBNPs have served as drug carriers, contrast agents for magnetic resonance imaging and photoacoustic imaging, nanoenzymes, and sensitizers for photothermal therapy. Ultrasmall PBNPs, coated with aminopolyethylene glycol or glucose-functionalized aminotriethylene glycol to control their biodistribution, have been used for ²⁰¹Tl delivery for diagnostic imaging. They showed in vivo accumulation of ²⁰¹Tl-PBNPs in lungs and liver during the first 3 h after intravenous (i.v) injection, whereas unbound ²⁰¹Tl⁺ at these time points accumulated mainly in kidneys. ²⁰¹Tl-radiolabeled citric acid-coated PBNPs have been investigated as contrast agents for dual single-photon emission computed tomography and magnetic resonance imaging (SPECT/MRI). In another study, different core shapes and coatings (including dextran and a phospholipid bilayer) were shown by SPECT to control the biodistribution of the PBNPs in vivo.Despite these uses of ²⁰¹Tl-PBNPs in preclinical nuclear imaging, their radiotoxic effects in cancer radionuclide therapy remain unexplored.
The authors have previously described the synthesis and physicochemical characterization of two types of PB nanoparticles (PBNPs), coated with citric acid (caPBNPs) and chitosan (chPBNPs). These coatings imparted distinct shapes and opposite surface charges (negative and positive, respectively). Both types of PBNPs proved to efficiently bind ²⁰¹Tl⁺ with high radiochemical stability for at least 72 h under physiological conditions. Furthermore, it was experimentally confirmed that the thallium inclusion mechanism involves thallium ions occupying interstitial spaces within the PB crystal structure, affecting the ionic composition but not the structure of the core or shell of the nanoparticles. In the present study, the authors compared the in vitro accumulation and retention of ²⁰¹Tl within cancer cells, as well as the subcellular localization and radiotoxicity of ²⁰¹Tl, when administered in the form of ²⁰¹Tl-caPBNPs, ²⁰¹Tl-chPBNPs, and unbound ²⁰¹Tl⁺. They also evaluated the ability of unlabeled intracellular PBNPs to sequester ²⁰¹Tl administered to cells in the form of free ²⁰¹Tl⁺ ions. Finally, the in vivo and ex vivo biodistribution and retention of radiolabeled chPBNPs were evaluated after direct injection into subcutaneous tumor xenografts in mice and compared to unbound ²⁰¹Tl⁺.

Results from nanoScan® SPECT/CT

Mice were randomized across 2 groups: the control group, where 4 mice were injected with [²⁰¹Tl]TlCl, and 5 mice injected with ²⁰¹Tl-chPBNPs. SPECT/CT imaging was done for 3 mice in each group, followed by the ex vivo biodistribution study. The remaining mice were used for the ex vivo biodistribution study only. Mice were anesthetized with 2.0% isoflurane (VetTech Solutions Ltd.) at a O₂ flow rate of 1.0 L/min, and 0.4–0.5 MBq of ²⁰¹Tl in the required form in 10 μL volume was injected into the tumors. After 1 h under anesthesia, 3 mice from the group were imaged for 30 min with the SPECT/CT NanoScan 80 W (Mediso Ltd., Budapest, Hungary) with three-mouse hotel settings (see Supporting Information). After SPECT/CT scanning, mice were allowed to recover from anesthesia. All mice were reanesthetized and rescanned by SPECT/CT at 24 and 48 h post injection. When the 48 h scanning procedure was completed, mice were culled, and dissection of organs was performed immediately for the quantification of ex vivo biodistribution of ²⁰¹Tl. VivoQuant v3.5-patch2 software (inviCRO, Massachusetts, USA) was used to view and quantify all reconstructed images. SPECT/CT biodistribution quantification: the regions of interest were manually contoured for tumors and kidneys using CT to mark their boundaries. SPECT/CT images were shown as maximum intensity projection (MIPs) or in transverse sections with the same intensity scale bar for all the groups. Data are expressed as standardized uptake values (SUV) and decay-corrected. For ex vivo biodistribution, organs and tissues were dissected and weighed. Radioactivity in each organ was then measured with the CompuGamma CS1282 gamma counter (over 60 s, energy window: 81–110) and converted to Bq using a [²⁰¹Tl]TlCl calibration curve. Biodistribution data are expressed as percentage of injected activity (IA) (% IA, decay-corrected) or percentage of IA per gram (% IA/g, decay-corrected).
To compare the retention of ²⁰¹Tl-chPBNPs with that of unbound ²⁰¹Tl⁺ in a subcutaneous murine xenograft model, A549 tumor-bearing mice were injected intratumorally with 0.4–0.5 MBq of [²⁰¹Tl]TlCl (control group) or ²⁰¹Tl-chPBNPs. SPECT/CT images acquired after administration showed that in both groups, by 1 h postinjection (p.i.), most activity was contained within the tumors (Figure 5A) with a very small amount of activity in the kidneys (Figure 5B). However, SPECT imaging at later time points revealed slower efflux from tumors in the mice injected with ²⁰¹Tl-chPBNPs, with 2.1 ± 1.3% IA remaining in tumors at 24 h compared to 0.4 ± 0.1% IA for mice injected with [²⁰¹Tl]TlCl (n = 3, p = 0.09, Figures 5B, S10A, S11). At 48 h, images showed 1.4 ± 1.0% IA remaining in tumors in the ²⁰¹Tl-chPBNPs group compared to just 0.3 ± 0.1% IA in the [²⁰¹Tl]TlCl group (n = 3, p = 0.1, Figure S10A). The remaining activity was predominantly detected in kidneys, with smaller amounts distributed across other organs such as small and large intestine (Figures 5B, S10B and S11).
The biodistribution trends observed by SPECT imaging were placed on a quantitative footing by post-mortem ex vivo organ counting. Radioactivity concentrations (% IA/g) in tissues are shown in Figure 5C, confirming that ²⁰¹Tl administered as ²⁰¹Tl-chPBNPs showed retention in tumors was significantly (around 3.6 fold) higher than unbound ²⁰¹Tl⁺ (7.6 ± 7.0% IA/g and 2.1 ± 0.3% IA/g, respectively, p = 0.02; Figure 5D). Outside the tumors, the biodistribution did not differ significantly between the two groups (Figure 5C); ²⁰¹Tl activity was mostly present in kidneys, with an average of 21.5 ± 2.6% IA/g in the control group and 20.9 ± 4.1% IA/g in the ²⁰¹Tl-chPBNPs group (Figure 5E). Activity was also present in urine, indicating predominantly renal excretion in both groups. Figure S12 presents the biodistribution in both groups expressed as % IA. Overall, the in vivo biodistribution experiments showed that after intratumoral injection in mice, radioactivity administered in the form of ²⁰¹Tl-chPBNPs was retained within the tumors significantly longer than unbound ²⁰¹Tl⁺.

• Both types of PBNPs, once internalized, demonstrated the capability to increase cellular uptake and retention of subsequently administered ²⁰¹Tl⁺, presumably by sequestering thallium that enters the cell via the Na⁺/K⁺-ATPase pump. The enhanced uptake of ²⁰¹Tl-chPBNPs led to higher radiotoxicity (whether assessed as clonogenicity or DNA damage) compared to that of ²⁰¹Tl-caPBNPs or unbound ²⁰¹Tl⁺.
• ²⁰¹Tl-chPBNPs exhibited significantly delayed efflux of ²⁰¹Tl from tumors in vivo compared to unbound ²⁰¹Tl⁺, with a distribution of radioactivity released from the tumor resembling that of ²⁰¹Tl⁺ and showing predominantly renal clearance.
• With the current lack of satisfactory chelators for ²⁰¹Tl, PBNPs present an alternative avenue for delivering radioactive thallium to cancer cells for therapeutic purposes, albeit additional work is required to optimize their biodistribution and further enhance specific accumulation and retention in tumors.

Full article on pubs.acs.org