Enhancement of 211At Uptake via the Sodium Iodide Symporter by the Addition of Ascorbic Acid in Targeted α-Therapy of Thyroid Cancer

211At is an α-emitter that has similar chemical properties to iodine and is used in targeted α-therapy. In the present study, we added ascorbic acid (AA) to 211At solution to increase the radiochemical purity of astatide and evaluated its efficacy against differentiated thyroid cancer, which is characterized by the expression of sodium/iodide symporter (NIS). Methods: Crude 211At solution (AA(−)) and 211At solution treated with AA (AA(+)) were prepared. Uptake by the thyroid was compared between the 2 solutions in normal male Wistar rats (n = 6). Cellular uptake in K1-NIS cells was analyzed under the AA(+) and AA(−) conditions. AA(+) was injected at 3 doses into K1-NIS xenograft mice: 1 MBq (n = 6), 0.4 MBq (n = 6), and 0.1 MBq (n = 6), and vehicle was injected into control mice (n = 6). The treatment effects were compared among the 4 groups. Results: Uptake by the thyroid was significantly enhanced in rats injected with the AA(+) as compared with those injected with AA(−). Cellular uptake analysis showed significantly increased uptake of 211At by the K1-NIS cells under the AA(+) condition as compared with the AA(−) condition. In the mouse xenograft model, the K1-NIS tumors showed significant accumulation of 211At at 3 and 24 h after administration (22.5 ± 10.4 and 12.9 ± 6.8 percentage injected dose, respectively). Tumor growth was immediately inhibited in a dose-dependent manner after administration of 211At. In the survival analysis, the 211At groups (0.1, 0.4, and 1 MBq) showed significantly better survival than the control group. Conclusion: Uptake of 211At was enhanced in differentiated thyroid cancer cells as well as the normal thyroid using 211At solution treated with AA. The method also showed dose-dependent efficacy against the K1-NIS xenografts, suggesting its potential applicability to targeted α-therapy.

Radi oactive iodine has long been used clinically for patients with differentiated thyroid cancer (1)(2)(3). 131 I is used for the ablation of thyroid remnants or treatment of metastatic thyroid cancer (1). However, some patients with multiple metastases are refractory to repetitive 131 I treatment, despite the targeted regions showing sufficient iodine uptake (4,5). According to the criteria described in the 2015 American Thyroid Association guidelines, radioactive iodine-refractory cancer includes metastatic disease that progresses despite showing substantial uptake of radioactive iodine (6). In such patients, b-particle therapy using 131 I is inadequate and another strategy is needed using a more effective radionuclide targeting the sodium/iodide symporter (NIS).
Astatine ( 211 At) is receiving increasing attention as an a-emitter for targeted radionuclide therapy (7)(8)(9). 211 At is a halogen element with similar chemical properties to iodine (10). a-particles emitted from 211 At with a branching ratio of 41.8% (5.98 MeV) have higher linear energy transfer than b-particles from 131 I (0.97 MeV) and exert a better therapeutic effect by inducing DNA double-strand breaks and free radical formation (11). Petrich et al. reported that 211 At accumulated in the normal thyroid and in tumor xenografts of a genetically modified NIS-expressing papillary thyroid cancer cell line (12). They also showed that 211 At was effective against thyroid cancer xenografts (12). Meanwhile, the oxidative states of 211 At solutions may vary and the chemical form and dynamics in the body of 211 At have not yet been clearly elucidated, because there is no stable isotope of 211 At (13,14). It is presumed that a plenitude of astatine species, such as At 1 , AtO 2 , At(OH) 2 2 , AtO 2 2 , AtO(OH) 2 2 , and AtO 1 , as well as At 2 , are present in basic or acidic aqueous solutions under an oxidative or reductive condition. In many radiopharmaceutical solutions, ascorbic acid (AA) is frequently added as a stabilizing agent since it is vitamin C and can safely be administered in humans.
There is a possibility of stabilizing the oxidative states of 211 At in its solutions by adding an appropriate reducing agent. In the present study, we prepared 211 At solutions focusing on the radiochemical purity of astatide (or astatide ion) and evaluated its distribution in normal thyroid tissue via transport through NIS, as well as its efficacy against differentiated thyroid cancer in a tumor xenograft model.

MATERIALS AND METHODS
Preparation of 211 At Solutions 211 At was produced by the 209 Bi(a,2n) 211 At reaction, followed by separation and purification by a dry distillation method. 211 At was dissolved in 100 mL of distilled water. The 211 At-crude solution at a final concentration of 10 MBq/mL was mixed with AA as a reducing agent at a final concentration of 1.2 w/v% and sodium bicarbonate as a pH adjuster at a final concentration of 2.1 w/v% at pH 8.0 and allowed to stand for 1 h at ambient temperature. Separately, the 211 At-crude solution was also mixed with the other reducing agents, namely cysteine, glutathione, sodium sulfite, or ferrous sulfate, at a final concentration of 1 w/v% under the same conditions. The resultant mixtures of 211 At solutions were analyzed by thin-layer chromatography (TLC) using Typhoon 7000 (GE Healthcare). The reagents were purchased from Nacalai Tesque.
In Vitro Cellular Uptake and Survival Analysis K1 cells (human papillary thyroid carcinoma) were provided by the European Collection of Authenticated Cell Cultures. K1 cells were maintained in culture medium, D-MEM:Ham F12:MCDB 105 (2:1:1) supplemented with 2 mM glutamine and 10% heat-inactivated fetal bovine serum. K1-NIS cells were obtained by transfection using the human SLC5A5 (NIS) gene clone (OriGene). K1 cells and K1-NIS cells were seeded onto a 24-well plate (1 · 10 5 /well) and cultured for 2 d. After treatment of 211 At, the cells were washed twice with phosphate-buffered saline(2) and the radioactivity was measured using a 2480 Wizard 2 g-counter (Perkin Elmer).
K1-NIS cells were seeded onto a 96-well plate (2 · 10 4 /well) and cultured for 2 d. 211 At and 131 I were serially diluted with the culture medium (100 mL/well). After 48 h of incubation, we measured cell viability using Cell Counting Kit 8 (Dojindo).

Preparation of Animals
Normal male Wistar rats and male ICR and SCID mice were purchased from Japan SLC Inc. and Charles River Japan, Inc., respectively. Animals were housed under a 12-h light/12-h dark cycle and given free access to food and water. The Wistar rats were fed with a lowiodine diet during the 2 wk before the experiment to make the thyroid condition uniform, in accordance with the method used in a previous study (15). Tumor xenograft models were established by subcutaneous injection of K1-NIS cells (1-2 · 10 7 cells) suspended in 0.2 mL of culture medium and Matrigel (1:1; BD Biosciences) into the SCID mice.
All animal experiments complied with the guidelines of the Institute of Experimental Animal Sciences. The protocol was approved by the Animal Care and Use Committee of the Osaka University Graduate School of Medicine. The criteria for euthanasia were as follows: when the animals had intolerable suffering, when a significant decrease in activity or a marked decrease in food and water intake was observed, or when the end of the observation period was reached (up to 84 d). Euthanasia was through deep anesthesia by isoflurane inhalation.

Administration of 211 At Solutions and Imaging Analysis
Two groups of normal Wistar rats (n 5 6; 12 wk old; body weight, 295. Planar and SPECT images were acquired with a g-camera system (E-cam; Siemens) with a low-energy all-purpose collimator (16). The energy window was set at 79 keV 6 20% targeting the x-rays emitted from the daughter nuclide of 211 Po (17). The radioactivity in the major organs was measured with a g-counter after euthanasia and dissection at 24 h. Regions of interest were placed using AMIDE software (version 1.0.4). Radioactivity levels in the major organs were measured with a g-counter after euthanasia and dissection at 24 h. Uptake was normalized by the injected dose (MBq) and body weight (g). The equivalent dose (Gy) in the dosimetry of 211 At was estimated according to a previous report (18). Tumoral uptake was estimated from the planar images at 3 and 24 h after injection, and the area under the curve after 24 h was assumed to decrease with physical decay.

Histologic Analysis
After the animals were sacrificed by euthanasia, the tumor, thyroid, and stomach were resected. The specimens were fixed overnight with 4% paraformaldehyde and cryoprotected in 30% sucrose in phosphate-buffered saline. Frozen sections of the samples were then incubated with NIS-antibody (Anti-SLC5A5, Rabbit-Poly; Atlas Antibodies). Immunohistochemistry was performed using the EnVision1 system-HRP Labeled Polymer Anti-Rabbit (K4003) (DAKO Corp.). For evaluation of toxicity, the thyroid and stomach were resected and frozen sections were stained with hematoxylin and eosin.

Statistical Analysis
Comparisons of the values between 2 groups were performed using an unpaired t test. Statistical analyses were performed using SPSS (version 19.0), and probability values of less than 0.05 were considered to denote statistical significance. Survival analysis was performed using the Kaplan-Meier method, and the log-rank test with Holm correction was used for the group comparison.

RESULTS
TLC analysis depicted that the crude 211 At solution was a mixture of at least 3 chemical species of 211 At, as shown in Figure 1A. Although AA(1) was composed of a single chemical species of 211 At (Rf, 0.79) with a high radiochemical yield (95.7%), this species showed a similar TLC profile to that of 123 I-NaI (I 2 ; Rf, 0.85). Figure 1B shows the TLC profiles of the 211 At solutions after addition of various reducing agents. Cysteine and glutathione also provided 211 At-asatide ions of high radiochemical purity, but the purity was less than that under the AA(1) condition. On the other hand, sodium sulfite and ferrous sulfate had no significant effect on the TLC profiles of the crude 211 At solutions under these conditions. Among the various reducing agents, addition of AA provided a 211 At solution of the highest radiochemical purity, and it was stable for 24 h. We used AA as the reducing agent for further experiments.
Whole-body distributions of 211 At in the normal rats are shown in Figure 2. High uptake was observed in the thyroid, stomach, and bladder in both planar and SPECT images. Uptake by the thyroid was significantly better after administration of AA(1) than after administration of AA(2). Measurement using the g-counter also showed high accumulation and significantly enhanced thyroid uptake after injection of AA(1) as compared with AA(2) ( Table 1).
Cellular uptake analysis showed high uptake of 211 At in the K1-NIS cells but almost no uptake in the K1 cells (Fig. 3A), suggesting that 211 At is transported into differentiated thyroid cancer cells through NIS. Uptake in the K1-NIS cells was significantly increased under the AA(1) condition as compared with the AA(2) condition (Fig. 3B).
The mouse tumor xenograft model revealed high accumulation in the K1-NIS tumor at 3 and 24 h after administration (Fig. 4A). Planar images revealed that the activity concentration of 211 At was 22.5 6 10.4 percentage injected dose at 3 h and 12.9 6 6.8 percentage injected dose at 24 h in the K1-NIS xenograft. The equivalent dose in the tumor was estimated to be 9.7 6 7.0 Gy. Measurement of the radioactivity in the K1-NIS xenografts (n 5 3) at 24 h after the injection of 211 At is shown in Table 2.
The change in tumor size after administration of 211 At is shown in Figure 5A. Tumor growth was inhibited immediately after administration of 211 At, and the treatment effect was dose-dependent. In the 1-MBq group, tumor growth was suppressed until approximately 40 d after the injection and regrowth was relatively slow. With regard to body weight, there was a slight drop in the 1-MBq group as compared with the other groups (Fig. 5B). However, this effect was transient, with the body weight restored within 2 wk and remaining relatively stable thereafter. For the survival analysis, the 211 At administration groups (0.1, 0.4, and 1 MBq) showed significantly better survival rates than the control group (Fig. 5C).
Expression of NIS was confirmed in the K1-NIS tumor xenografts, thyroid, and stomach wall by immunohistochemical staining (Fig. 6).

DISCUSSION
In the present study, we demonstrated that the radiochemical purity of 211 At was increased by treatment with AA. Uptake of AAtreated 211 At was enhanced in the NIS-expressing cells in in vitro studies and in the normal thyroid after intravenous administration. AA(1) showed a dose-dependent treatment effect in the K1-NIS xenograft models. 211 At is a very promising a-emitter as it can be produced in an accelerator and has similar chemical properties to iodine (10). However, one of the major problems of astatine is that there is no stable isotope and the basic chemistry is poorly understood (18). In previous publications on 211 At, the oxidation states of aqueous solutions of astatine, such as At 2 , At(0), HOAt, AtO 3 2 , and H 5 AtO 6 , vary (19). In the present study, we found that the oxidation states of 211 At converged into a single component, which is assumed to be astatide ion (At 2 ) according to the TLC analysis using 123 I and 211 At. 211 At-astatide ions are labile in aqueous solutions and easily oxidized to higher oxidation states (18)(19)(20). Several reducing agents, including AA, were useful for the preparation of 211 At-astatide ions in aqueous solution and for stabilization. TLC analyses of the solutions are desired, since the quality of the solutions vary.
In a previous study in which the distribution of astatine was evaluated, high uptake was observed in the thyroid, stomach, lungs, and spleen (10,18). Astatine is actively transported across the plasma membrane into the cytoplasm via NIS (12). The thyroid and stomach also showed a high accumulation of iodine, reflecting expression of NIS (18). These results were compatible with the whole-body distribution in the normal rats of our study, suggesting that NIS is the key symporter in the uptake process of astatine.
As shown in Figure 6, stomach cells also showed NIS expression. However, no significant increase of 211 At uptake was observed in the stomach after administration of AA(1) as compared with AA(2). Spetz et al. reported the difference in the whole-body distribution between 211 At and 125 I/ 131 I (18). In their study, decreased uptake of 211 At by the thyroid and increased uptake in the stomach were observed as compared with 125 I/ 131 I. Their study suggested that the transport mechanism of 211 At via NIS might differ between the thyroid and the stomach. There are some differences in NIS expression by glycosylation modification, dimerization, and transcriptional regulation (including epigenetic regulation) (21). It is possible that these factors or some other unknown mechanisms might affect the transport function of NIS for astatide, explaining the difference in uptake between the stomach and the thyroid.
For K1-NIS tumor, we performed a cellular uptake analysis to compare 211 At uptake between AA(1) and AA(2), in which uptake was normalized by the number of cells. Uptake of 211 At was significantly increased under the AA(1) condition as compared with the AA(2) condition in K1-NIS cells (Fig. 3B), as well as in the thyroid glands of normal rats (Fig. 2). Meanwhile, we need to consider the ethical 3R rules (replacement, reduction, and refinement) in animal experiments. Therefore, we thought it would be precise and ethically reasonable to compare by simultaneous cellular uptake analysis. Petrich et al. evaluated the treatment effect of 211 At in K1-NIS xenograft mice (12). In their study, a total activity of 2.5 MBq was intraperitoneally administered in 3 fractions within 16 d. Although they stated that 211 At 2 was used in their experiments, there was no description of any detailed analysis of the 211 At solutions used in their study. In our study, we evaluated the oxidation states of 211 At in solution in greater  detail and proved that appropriate preparation of the astatine solution using reducing agents, such as AA, is important for maximizing the treatment effect of 211 At in the form of astatide. They also showed that fractionated 211 At therapy resulted in complete remission of the K1-NIS xenografts and also stated that a single administration caused stable shrinkage for up to 8 wk but was followed by tumor regrowth. We observed tumor shrinkage for up to 41 d after the administration of 211 At (1 MBq), followed thereafter by regrowth, similar to the findings of the aforementioned study. They also reported that administration of a total of 2.5 MBq of 211 At caused moderate damage in the thyroid by showing decreased uptake in 99m TcO 4 scintigraphy and atrophy in the histologic analysis. We showed that a smaller dose (0.1 MBq) of 211 At is also effective for suppression of tumor growth, although the regrowth was faster than after administration of the higher dose (0.4 MBq or 1 MBq). These results suggest that a more fractionated administration can be performed during targeted a-therapy using 211 At in the clinical setting to reduce the side effects in normal organs. Fractionated administration of an a-emitter has been successfully adapted in 223 Ra therapy for bone metastasis in castrationresistant prostate cancer (22). Therefore, targeted a-therapy with 211 At would be clinically feasible, and the short half-life of 211 At (7.2 h) is more ideal for preventing potential side effects after repeated administration over a short period. Besides differentiated thyroid cancer, there are other cancers that show expression of NIS, such as breast cancer and gastric cancer (23,24). In patients with these cancers, 211 At therapy can be one of the treatment options, especially for those with multiple metastases, in combination with appropriate iodine blocking of the normal thyroid (25). Recently, other groups in Japan have successfully demonstrated the excellent treatment effect of 211 At-labeled antibody or compounds for peritoneal dissemination from gastric cancer and pheochromocytoma (8,9). Appropriate targeting is our next challenge, with careful evaluation of dehalogenation of the 211 At compounds (13).
In the present study, the structure of the follicle was disrupted in the thyroid tissue at 3, 7, and 15 d after the administration of AA(1), whereas no significant difference was observed in the structure of gastric mucosa (Supplemental Fig. 1; supplemental materials are available at http://jnm.snmjournals.org). It was suggested that 211 At could ablate the thyroid tissues, but no significant effect was observed in the stomach. There are some publications about the chronic toxicity of 211 At. Cobb et al. reported that spleen, lymph nodes, bone marrow, gonads, thyroid glands, salivary glands, and stomach were affected after administration of 211 At in mice (61 kBq/g of body weight) (26). However, most of them showed a transient change at 14 d and returned to normal levels or resulted in minimal damage at 56 d after administration, except for thyroid glands and gonads. The normal thyroid glands in thyroid cancer patients are usually removed or ablated in the clinical setting and thus would not be a problem. For the gonads, treatment depends on the tolerability of the patients. Petrich et al. reported that thyroid atrophy was observed in all mice and that inflammation of the lungs and stomach or bowel were observed in some mice during 1 y of follow-up after a total administration of 2.5 MBq of 211 At (12). It was suggested that no critical finding was observed with the administered dose below 1 MBq.
This study had several limitations. First, we did not evaluate the side effects of 211 At, other than the thyroid and stomach. Detailed  analysis of the side effects, especially in relation to the dose dependency, is the next important subject for study before clinical application. Second, we did not perform fractionated administration of 211 At in this study. This is also another important prerequisite to clinical application, as well as for evaluation of the side effects. Third, we preliminarily performed the in vitro cellular survival assay using K1-NIS cells and confirmed dosedependent decreased cell viability in AA(1) compared with 131 I (Supplemental Fig. 2). Comparison of the effectiveness of 211 At and 131 I is essential to prove that 211 At therapy is more beneficial for patients who are refractory to repetitive 131 I treatment with preserved uptake ability of NIS.

CONCLUSION
This study revealed that an increase in the radiochemical purity of astatide in 211 At solution by addition of AA was associated with significantly enhanced uptake of 211 At by both normal thyroid tissue and differentiated thyroid cancer cells. The treatment effect of 211 At solution in the K1-NIS xenograft model was dose-dependent and associated with prolonged survival, suggesting the potential applicability of targeted a-therapy for the treatment of advanced differentiated thyroid cancer.

DISCLOSURE
This study was funded by KAKENHI (C) (15K09955) from the Ministry of Education, Culture, Sports, Science, and Technology (MEXT) and by the QiSS program of the OPERA from the Japan Science and Technology Agency (JST). No other potential conflict of interest relevant to this article was reported.